200708302007 IRP.pdf

Avista Corp.
1411 East Mission PO Box 3727
Spokane, Washington 99220-3727
Telephone [phone]
Toll Free 800-727-9170
~~~'V'STA.
Corp.
~:CEIVED
ZUlli .:10
n '
I: :; I
August 29, 2007
Jean D. Jewell, Secretary
Idaho Public Utilities Commission
Statehouse Mail
W. 472 Washington Street
Boise, Idaho 83720 AVU-L-67-0b
Dear Ms. Jewell:
RE:A vista Utilities 2007 Electric Integrated Resource Plan
Per IPUC's Integrated Resource Plan Requirements outlined in Case No.1500-165
Order No. 22299, Case No.GNR-93-, Order No. 24729 and Case No.GNR-93-
Order No. 25260 , Avista Corporation d/b/a! Avista Utilities , hereby submits for filing an
original, an electronic copy and 7 copies of its 2007 Electric Integrated Resource Plan.
The Company submits the IRP to public utility commissions in Idaho and Washington
every two years as required by state regulation. A vista regards the IRP as a methodology
for identifying and evaluating various resource options and as a process by which
establish a plan of action for resource decisions.
The 2007 Plan is notable for the following:
The Company is currently long on energy and capacity until 2011 , this position
will be extended to 2017 for energy and 2015 for capacity with the addition ofthe
Lancaster Generation Facility in 2010;
. Avista s Preferred Resource Strategy Model (PRiSM) developed an efficient
frontier that balances both portfolio risk and cost considerations;
The Preferred Resource Strategy (PRS) includes 350 MW of CCCT, 300 MW of
wind, 35 MW of other renewables, and 87 MW of conservation between 2007 and
2017;
Conservation acquisition is approximately 25 percent higher than in the 2005 IRP;
The PRS no longer includes coal-fired generation; fixed price natural gas takes its
place;
Greenhouse gas emission costs are included in the Base Case;
Washington State s Energy Independence Act (1-937) requirements up through
2016 will be met primarily with plant upgrades; and
Paper use and printing costs have been reduced by putting supporting documents
on our web site at www.avistautilities.com/resources/plans/electric.asp.
Please direct any questions regarding this report to Clint Kalich at [phone].
Sincerely,
Linda Gervais
Regulatory Compliance, State and Federal Regulation
Mr. Rick Sterling
VU -O7 -
www.avistautilities.com ~'iI'ST A'
C"J
,,")
t:J
:;::0.
f'i'J
~;;;
t~.-:
PHOTO CREDITS
- Avista s investment in transmission infrastructure crosses the wheat fields of Washington state s Palo use
region. Photo by Hugh Imhof Avista.
Three key components ofAvista s renewable energy and DSM plans include the Noxon Rapids
Hydro Facility on the Clark Fork River in Montana, education about energy efficient compact
flourescent bulbs, and including power generated at the Stateline Wind Farm on the Southeast border
of Washington and Oregon.
SPECIAL THANKS TO OUR TALENTED VENDORS FROM
THE SPOKANE AREA WHO PRODUCED THIS IRP:
Ross Printing Company
Thinking Cap Design
Printed on recycled paper.
. ;
TABLE OF CONTENTS
Executive Summary
Introduction and Stakeholder Involvement 1-
Loads and Resources
Demand Side Management
Environmental Issues
Transmission Planning
Modeling Approach
Market Modeling Results
Preferred Resource Strategy
Action Items
SAFE HARBOR STATEMENT
This document contains forward-looking statements. Such statements
are subject to a variety of risks, uncertainties and other factors, most
of which are beyond the companys control, and many of which
could have a significant impact on the company s operations, results
of operations and financial condition, and could cause actual results to
differ materially from those anticipated.
For a further discussion of these factors and other important factors
please refer to our reports fIled with the Securities and Exchange
Commission which are available on our website at www.avistacorp.
com. The company undertakes no obligation to update any forward-
looking statement or statements to reflect events or circumstances that
occur after the date on which such statement is made or to reflect the
occurrence of unanticipated events.
TABLE OF TABLES
Table 1:
Table 2:
Table 3:
Table 1.1:
Table 1.
Table 2.
Table 2.
Table 2.
Table 2.4:
Table 2.5:
Table 2.
Table 2.
Table 2.
Table 3.
Table 3.
Table 3.
Table 3.4:
Table 5.
Table 6.
Table 6.
Table 6.
Table 6.4:
Table 6.
Table 6.
Table 6.
Table 6.
Table 6.
Table 6.10:
Table 6.11:
Table 6.12:
Table 6.13:
Table 6.14:
Table 6.15:
Table 6.16:
Table 6.17:
Table 6.18:
Table 6.19:
Table 6.20:
Net Position Forecast
2007 Preferred Resource Strategy Selections (Nameplate MW)
Net Position Forecast with Lancaster
TAC Participants
TAC Meeting Dates and Agenda Items
Global Insights National Forecast Assumptions
Company-Owned Hydro Resources
Company-Owned Thermal Resources
Mid-Columbia Contract Summary
Significant Contractual Rights and Obligations
Capacity L&R Versus Sustained Capacity
Loads & Resources Capacity Forecast (MW)
Loads & Resources Energy Forecast (aMW)
Current Energy Efficiency Programs
Proposed New Energy Efficiency Program
Current Avista Energy Efficiency Programs (kWh)
Recent Hydro Efficiency Upgrade Studies
Estimated Integration Costs Inside Avista s System ($Millions)
AURORAxmp Pools and Zones
Seasonal Natural Gas Price Factors
Natural Gas Basin Prices as % of Henry Hub
New RPS Resources Added to Existing System (aMW)
Annual Average Peak Load Growth
Annual Average Energy Load Growth
Coefficient of Variation of Forward Sumas Natural Gas Prices
Selected Zone s Load Correlations to Eastern Washington Gan-June)
Selected Zone s Load Correlations to Eastern Washington Guly-Dec)
Selected Zone s Load Coefficient of Variation Gan-Jun
Selected Zone s Load Coefficient of Variation Guly-Dec %)
Simulated Average Annual Wind Capacity Factors
Probability Matrix of Carbon "Taxes
Real 2007 Levelized Costs for 2013 CCCT (Full Availability)
Real 2007 Levelized Costs for 2013 SCCT (Full Availability)
Coal Plant Technology Characteristics and Assumed Costs
Regional Coal Transmission Capital Costs
Real 2007 Levelized Costs for 2013 NW Coal Plants (Full Availability $/MWh)
Wind Location Capacity Factors (excludes losses)
Wind Integration Costs
TABLE OF TABLES (continued)
Table 6.21:
Table 6.22:
Table 6.23:
Table 7.
Table 7.
Table 7.
Table 7.4:
Table 7.
Table 7.
Table 7.
Table 7.
Table 7.
Table 7.10:
Table 7.11:
Table 7.12:
Table 7.13:
Table 7.14:
Table 7.15:
Table 7.16:
Table 7.17:
Table 7.18:
Table 7.19:
Table 8.
Table 8.2:
Table 8.
Table 8.4:
Table 8.
Table 8.
Table 8.
Table 8.
Table 8.
Table 8.10:
Table 8.11:
Table 8.12:
Table 8.13:
Table 8.14:
Real 2007 Levelized Costs for 2013 Wind Plants (Full Availability)
Real 2007 Levelized Costs for 2013 Alberta Oil Sands Project (Full Availability)
Real 2007 Levelized Costs for 2008 Other Resources (Full Availability)
Base Case Key Assumptions
Cumulative Western Interconnect Resource Additions (nameplate MW)
Oregon Washington and Northern Idaho Cumulative Resource Selection (MW)
Unconstrained Carbon Future Cumulative Resource Selection (MW)
CSA Carbon Charge Future, Cumulative Resource Selection (MW)
Comparative Levelized Mid-Columbia Prices and Risk (Real 2007 Dollars)
Comparative Levelized Mid-Columbia Prices and Risk (Nominal 2007 Dollars)
Multiple Regression Coefficient Results
Constant Gas Growth Scenario, Cumulative Resource Selection (MW)
High Natural Gas Price Scenario: Cumulative Resource Selection (MW)
Low Natural Gas Price Scenario: Cumulative Resource Selection (MW)
Western Interconnect Average Demand (aGW)
High Load Escalation Scenario: Cumulative Resource Selection (MW)
High Load Escalation Scenario: % Change Cumulative Resources
(%)
Low Load Escalation Scenario: Cumulative Resource Selection (MW)
Low Load Escalation Scenario: % Change Cumulative Resources
(%)
Nuclear Plants Scenario: Cumulative Resource Selection (MW)
Electric Car Scenario Costs ($Billions)
Future and Scenario Market Price Comparisons ($/MWh)
Resource Options Available to Avista for the 2005 and 2007 IRP, first 10 years
2007 IRP Preferred Resource Strategy Selection (Nameplate MW)
2005 IRP Preferred Resource Strategy Selection (Nameplate MW)
Loads & Resources Energy Forecast with PRS (aMW)
Loads & Resource Capacity Forecast with PRS (MW)
Company Resource Capital Requirements ($Millions)
Impacts to Wind & Green Tag Selection (2008-2017)
Impact to Wind Selection with Idaho RPS (MW)
2008-17 Resources for Each Portfolio (Capability MW)
Capital Cost Sensitivities ($2007/kW)
Wind Capacity Selected for 25% Risk Reduction (MW)
Resource Selection Comparison (MW)
Loads & Resources Energy Forecast with PRS (aMW)
Loads & Resource Capacity Forecast with PRS (MW)
TABLE OF FIGURES
Figure 1:
Figure 2:
Figure 3:
Figure 4:
Figure 5:
Figure 6:
Figure 7:
Figure 8:
Figure 9:
Figure 10:
Figure 11:
Figure 12:
Figure 2.
Figure 2.
Figure 2.
Figure 2.4:
Figure 2.
Figure 2.
Figure 2.
Figure 2.
Figure 2.
Figure 2.10:
Figure 2.11:
Figure 2.12:
Figure 2.13:
Figure 2.14:
Figure 2.15:
Figure 2.16:
Figure 2.17:
Figure 3.
Figure 3.
Figure 3.
Figure 3.4:
Figure 3.5:
Figure 4.
Figure 4.
Figure 4.
Figure 4.4:
Load Resource Balance--Capacity (MW)
Load Resource Balance--Energy (aMW)
Efficient Frontier and Traditional Resource Portfolios
Base Case Stochastic Mid-Columbia Prices ($/MWh)
Annual Average Sumas Natural Gas Price Results from 300 Iterations ($/Dth)
Use per Customer
Avista s Retail Sales Forecast
111
VII
Vlli
Cumulative Efficiency Acquisitions
The 2007 Preferred Resource Strategy (aMW)
Amount of Renewable Energy Forecasted to Meet RPS (aMW)
937 Qualifying and Non-Qualifying Avista Renewables (aMW)
Efficient Frontier With and Without Fixed Price Gas Contracts Option
Carbon Footprint (Tons per MWh)
Loads & Resources Capacity Forecast with Lancaster (MW)
Avista s Service Territory
Population Change for Spokane, Kootenai and Bonner Counties (Thousands)
Total Population for Spokane, Kootenai and Bonner Counties (Thousands)
Three-County Population Age 65 and Over (Thousands)
Three-County Job Change (Thousands)
County Non-Farm Jobs (Thousands)
Avista Annual Average Customer Forecast (Thousands)
Household Size Index (% of2007 Household Size)
Annual Net Native Load (aMW)
Calendar Year Peak Demand (MW)
Comparison of Summer and Winter Peak Demand (MW)
Electric Load Forecast Scenarios (aMW)
Avista s Hydroelectric Projects
Capacity Loads and Resources (MW)
Energy Loads and Resources (aMW)
Historical Conservation Acquisition
Year-On- Year Conservation Acquisition (%)
Forecast of Efficiency Acquisition (aMW)
Supply of Evaluated Efficiency Measures
Efficiency Supply Curves Including All Measures
Base Case SO2 Costs ($/ton)
Base Case NOx Costs ($/ton)
Base Case Mercury Costs ($/ounce)
Base Case CO2 Costs ($/ton)
TABLE OF FIGURES (continued)
Figure 5.
Figure 6.
Figure 6.
Figure 6.
Figure 6.4:
Figure 6.
Figure 6.
Figure 6.
Figure 6.
Figure 6.
Figure 6.10:
Figure 6.11:
Figure 6.12:
Figure 6.13:
Figure 6.14:
Figure 6.15:
Figure 6.16:
Figure 6.17:
Figure 6.18:
Figure 7. 1
Figure 7.
Figure 7.
Figure 7.4:
Figure 7.
Figure 7.
Figure 7.
Figure 7.
Figure 7.
Figure 7.10:
Figure 7.11:
Figure 7.12:
Figure 7.13:
Figure 7.14:
Figure 7.15:
Figure 7.16:
Figure 8.
Figure 8.
Figure 8.
Geographic Locations of Proposed Transmission Upgrades
Modeling Process Diagram
NERC Interconnection Map
Henry Hub Natural Gas Forecast ($/Dth)
Daily Natural Gas Prices Shape ($/Dth)
Coal Prices for New Coal Resources ($/ton)
Emission Charges Summary
March 2006 Sumas Natural Gas Contact Price Distribution
2008 Sumas Natural Gas Price (Deterministic & First 30 Draws)
Annual Average of300 Iterations ofSumas Natural Gas Prices ($/Dth)
Hydro Capacity Factor and Statistics for Selected Areas (%)
Water Year Distribution
Distribution of Stochastic Hydro as a Percent of the Mean
August Hourly Wind Generation Distribution
Actual Stateline Generation August 9th through 15th, 2006
Simulated Hourly Columbia Basin Wind Generation for August
Capacity Levels for Northwest Gas-Fired Plants (%)
Real Levelized Costs for Selected Resources at Full Availability ($/MWh)
Real Levelized Costs for Selected Resources with Market Operations ($/MWh)
Oregon Washington and Northern Idaho Resource Positions (GW)
Mid-Columbia Electric Price Forecast ($/MWh)
Western Interconnect Resource Dispatch Contribution
Base Case Stochastic Mid-Columbia Prices ($/MWh)
Volatile Gas Future Stochastic Mid-Columbia Electric Forecast ($/MWh)
Unconstrained Carbon Future Mid-Columbia Electric Price Forecast ($/MWh)
CSA Carbon Charge Future: WI Resource Dispatch Contribution
CSA Carbon Future, Mid-Columbia Electric Price Forecast ($/MWh)
Western Interconnect Total Carbon with Different Futures (Million Tons of CO)
Sumas Gas Price versus Mid-Columbia Electric Prices
Natural Gas Forecasts, Constant Gas Growth versus the Base Case ($/Dth)
Natural Gas Price Forecast Scenarios versus the Base Case ($/Dth)
Western Interconnect Fuel Costs, Nuclear Beginning in 2015 (Nominal $Billions)
Western Interconnect Carbon Emissions (Million Tons ofCO
Impact of Electric Cars on the Western Interconnect (aGW)
Comparison of Total Fuel Costs for the WI in 2017 and 2027 ($Billions)
Amount of Renewable Energy Forecasted to Meet Wash. state RPS (aMW)
Generation Capital Cost Trends (2007 $/kW)
Historical and Future Nameplate Acquisition (MW)
TABLE OF FIGURES (continued)
Figure 8.4:
Figure 8.5:
Figure 8.
Figure 8.
Figure 8.
Figure 8.
Figure 8.10:
Figure 8.11:
Figure 8.12:
Figure 8.13:
Figure 8.14:
Figure 8.15:
Figure 8.16:
Figure 8.17:
Figure 8.18:
Figure 8.19:
Figure 8.20:
Figure 8.21:
Figure 8.22:
Figure 8.23:
Figure 8.24:
Figure 8.25:
Figure 8.26:
Figure 8.27:
Figure 8.28:
Figure 8.29:
Figure 8.30:
Figure 8.31:
Figure 8.32:
Lumpy Resource Acquisition (MW)
Loads & Resources Energy Forecast with PRS (aMW)
Loads & Resource Capacity Forecast with PRS (MW)
Company Resource Mix (% of Energy)
Company Resource Mix (% of Capacity)
Annual Power Supply Expense ($Millions)
Annual Portfolio Volatility
(%)
Forecasted CO2 Tons of Emissions (Thousands)
Forecasted CO2 (Tons/MWh)
Efficient Frontier and Traditional Resource Portfolios
Net Present Value of New Resource and Power Supply Costs by Portfolio (2007 $Millions)- 8-
Volatility (Coefficient ofVariation) of 2017 Power Supply Expenses (%) 8-
2017 Total Power Supply Expenses ($Millions) 8-
Average Annual Power Cost Component Change 2008-2017 (%) 8-
Maximum Annual Cost Change for Power Supply (%) 8-
2008-2017 NPV of Capital Investment (2007 $Millions) 8-
Renewable Resources Included in Each Portfolio (Nameplate MW) 8-
Alternative Resource Planning Criteria (Efficient Frontier Results) 8-
Efficient Frontier With and Without Fixed Price Gas Contract Option 8-
Historical Monthly Gas Prices at Stanfield ($/Dth) 8-
Variable Fuel Costs ofCCCT Plant at Various Gas Hedging Levels ($/MWh) 8-
Portfolio Cost Comparison Versus PRS for Each Market Scenario (%) 8-
Loads & Resources Energy Forecast with Lancaster in PRS (aMW) 8-
Loads & Resources Capacity Forecast with Lancaster in PRS (MW) 8-Efficient Frontier with Lancaster Plant 8-
Efficient Frontier for All Futures
Unconstrained Carbon Future s Efficient Frontier Portfolios
Climate Stewardship Future Efficient Frontier Portfolios
Volatile Gas Future Efficient Frontier Portfolios
LIST OF ACRONYMS AND KEY TERMS
AARG Annual Average Rate of Growth Nominal Discounting Method that Includes
AVA Avista Inflation
aMW Average Megawatts NPCC Northwest Power and Conservation
BPA Bonneville Power Administration Council (formerly Northwest Power
CCCT Combined-Cycle Combustion Turbine Planning Council)
CFL Compact Fluorescent Lamp NPV Net Present Value
Carbon Dioxide NWPP Northwest Power Pool
CSA Climate Stewardship Act (also known as O&M Operations and Maintenance
the McCain-Lieberman Bill)OASIS Open Access Same-Time Information
CVR Controlled Voltage Reduction System
Dth decatherm OSU Oregon State University
Efficiency Personal Computer
EIA Energy Information Administration PGE Portland General Electric
FERC Federal Energy Regulatory PRS Preferred Resource Strategy
Commission PRiSM Preferred Resource Strategy Line
The General Electric Company Programming Model
GHG Greenhouse Gas pSlg Pounds Per Square Inch Gauge
GWh Gigawatt-hour PTC Production Tax Credit
HRSG Heat Recovery Steam Generator PUD Public Utility District
HVAC Heating,Ventilation and Air PURPA Public Utility Regulatory Policies
Conditioning (HV AC)Act of 1978
IDP Idaho Power Company Real Discounting Method that Excludes
IGCC Integrated Gasification Combined Inflation
Cycle RPS Renewable Portfolio Standards
IRP Integrated Resource Plan RTO Regional Transmission Organization
Information Systems SCCT Simple-Cycle Combustion Turbine
kilo-volt TAC Technical Advisory Committee
kilowatt TIG Transmission Improvements Group
kWh kilowatt-hour TRC Total Resource Cost
LIRAP Low Income Rate Assistance Program Triple E External Energy Efficiency Board
Linear Programming VFD Variable Frequency Drive
Mmbtu Million British Thermal Units WECC Western Electricity Coordinating
1 mmbtu = 1 dth of Natural Gas Council
megawatt WNP-Washington Public Power Supply
MWh megawatt-hour System (wpPSS, now Energy
NCEP National Commission for Northwest) - Washington Nuclear
Energy Policy Plant No.
NEB Non-Energy Benefits
2007 IRP KEY MESSAGES
. Resource deficits start in 2014 with loads exceeding
resource capability by 49 MW Deficits are driven by
electricity sales growth averaging 2.3 percent over the
next decade.
. The 2007 Preferred Resource Strategy (FRS) differs
substantially from the 2005 FRS in three main areas:
the removal of coal as a resource, the challenge of
acquiring renewables and the need for natural gas-fired
plants.
. The FRS includes 350 MW of natural gas-fired plants
300 MW of wind, 87 MW of conservation, 38 MW of
hydro plant upgrades and 34 MW of other renewables
by 2017.
. The coal-fired generation forecast in previous plans is
replaced entirely with natural gas-fired resources.
. Conservation acquisition is 25 percent higher
than in the 2005 plan and 85 percent higher than
the 2003 IRP. The company is implementing an
enterprise-wide conservation and energy efficiency
initiative called the "Heritage Project." It builds
the company's long-time commitment to energy
conservation and efficiency, introducing new products
and services to increase customers' energy savings.
. Fewer renewables meet our planned requirements due
to tightening market conditions; renewables legislation
in Washington and Oregon has artificially increased
and accelerated the demand for these resources and
therefore increased their costs. For example, wind
generation costs have increased more than 50 percent
since the 2005 IRP.
. Avista supports national climate change legislation
and is actively participating to ensure cost-effective
solutions for our customers.
. Avista has one of the smallest carbon footprints in
the U.S. because of its renewable energy resources.
According to a Natural Resources Defense Council
study, only seven other major utilities have a smaller
footprint.
. Avista s high percentage of existing renewable hydro
resources, combined with a lack of available cost-
effective renewable resource options, means we must
continue to acquire carbon-emitting generation to
meet future load growth. This increases our total
carbon footprint, but our emissions per MWh
generation fall over time.
. The enactment of new laws imposing emission
performance standards on fossil fueled generation
resources acquired by electric utilities in Washington
Oregon and California narrows our cost-effective
options, at least in the short term, to natural gas-fired
generation.
. The FRS strikes a reasonable balance between keeping
average costs and variation in year-to-year costs low.
. Fixing gas prices does not lower absolute cost, but it
does limit price volatility.
. The power purchase contract for the Lancaster
Generating Plant, previously held by Avista Energy
and transferred to Coral Energy in 2007, will be
available to Avista beginning in 2010. This will provide
approximately 275 MW of natural gas-fired generation
and will be a good resource to serve customer load.
. Action items being developed for the 2009 IRP
include renewable energy and emissions, enhancements
to modeling systems, transmission modeling and
research, and conservation.
. The 2007 IRP was substantially complete when the
company announced the availability of the Lancaster
gas-fired plant to the utility. The Preferred Resource
Strategy, as detailed above, includes 350 MW of
natural gas-fired generation over its first 10 years. The
Lancaster plant is assumed to replace a significant
portion of this component. As the IRP was not
re-run due to the Lancaster addition, in some places
within the 2007 IRP our resource deficiencies and
tabulations are shown with and without the Lancaster
plant.
Executive Summary
EXECUTIVE SUMMARY
Bull River Valley, Montana
Avista s 2007 Integrated Resource Plan (IRP) will guide
utility resource acquisitions over the next two years
and beyond. Besides providing a snapshot of its current
resources and loads, the IRP shows where our resource
portfolio is heading through the Preferred Resource
Strategy (FRS). The FRS is made up of renewable
resources, conservation, efficiency upgrades at existing
facilities and new gas-fired generation. The most
significant change from the 2005 IRP is the exclusion
of coal-fired generation due to changing economics
and recent legislation effectively barring its use.
Conservation acquisition is forecast to rise approximately
25 percent over the 2005 IRP level and by more than 85
percent from the 2003 IRP.
The IRP balances low cost, reliable service and
reasonable future rate volatility. Avista s management and
stakeholders from the Technical Advisory Committee
(TAC) playa key role in directing the IRP process.
TAC members include customers, Commission Staff
consumer advocates, academics, utility peers, government
agencies and interested internal parties. The TAC
provides significant input on modeling, planning
assumptions and the general direction of the planning
process.
RESOURCE NEEDS1
Plant upgrades and conservation acquisition are
inadequate to meet all future load growth. Annual
energy deficits begin in 2011 , with loads exceeding
resource capabilities by 83 aMW Energy deficits rise
to 272 aMW in 2017 and to 513 aMW in 2027. The
company will be short 146 MW of capacity in 2011. In
2017 and 2027, capacity deficits rise to 300 MW and
835 Mw, respectively. Table 1 presents the companys net
position forecast during the first 10 years of the study.
Increasing deficits are a result of2.3 percent energy
and capacity load growth through 2017. Expirations of
certain long-term contracts also add to the deficiencies.
Figures 1 and 2 provide graphical presentations of Avista
load and resource balances. The annual forecasted load
is the summation of our peak forecast plus planning and
operating reserve obligations.
Table 1: Net Position Forecast
1 Energy and Capacity positions exclude the acquisition of Lancaster. The impact of Lancaster on the company s L&R position is detailed
later in this chapter.
Avista Corp 2007 Electric IRP
Executive Summary
000
Figure 1: Load Resource Balance-Capacity (MW)
500
- - - - - - - - - - - - - - - - - - - -
000
500
000
500
............
C")
......
I.C)
......,...
Figure 2: Load Resource Balance-Energy (aMW)
800
600
- -
1,400
200
000
800
600
400
200
- -- -- -- -
Hydro_Coal
c::::::J Gas Dispatch
Total Obligations
............
MODELING AND RESULTS
The company used a multi-step approach to develop
its Preferred Resource Strategy. The process began
by identifying potential new resources to serve future
demand across the Western United States. A Western
Interconnect-wide study was performed to understand
the impact of regional markets. We believe that the
additional efforts to develop this study were necessary
given the significant impact other regions can have on
the Northwest electricity marketplace. Existing resources
were combined with the present transmission grid to
simulate hourly operations from 2008 through 2027.
Net Contracts
Biomass
I!!!!!!IiiiI Gas Peaking Units
C")I.C)
......,.........
Cost-effective new resources and transmission were
added to meet growing loads. Monte Carlo-style analysis
varied hydro, wind, load and gas price data over 300
iterations of potential future conditions. The simulation
results were used to estimate the Mid-Columbia
electricity market. The iterations collectively formed the
Base Case.
Estimated market prices were used to analyze potential
conservation initiatives and available supply-side
resources to meet forecasted company requirements.
Each new resource option was valued against the Mid-
2007 Electric IRP Avlsta Corp
Executive Summary
Figure 3: Efficient Frontier and Traditional Resource Portfolios
2008 to 2017 Total Cost Net Present Value ($Millions)550 1,650 1,750 1,850 1,950 2 050 2 1501,450
220
250
...
0 ~
& :: 180c'liiIV -.s::. ::0-
....
::0-
; 1i 140
u Q.... :ICII tI)
a.. ...
~ ;
100N ~
seer & Green Tags CCCT &
+---:--+--
~enTagS - -- --
:---- --
O%Risk
-------
I--------
,-_
PRSNoFlXedGas Wind 20% -------
'-------
-J-------I 'ConI. to PM Renew abies
Coal Allowed
+ &
cr I --2005 PRS
- - - - - - -:- - - - - - - -- - - - - - - ~ - -
25o/~kc.- -
~ ~ - -.- - - -:- - -+- - - - ~ - - - - -
50% Risk I 100% Risk
.... c... 00 =
75~ i0 v.c --
.....-
11/
'Iii 060'- 0::0-01i
... :IIV tI)45-g;
J!! ~tI) 0a..
No Additions
85 90 95 100 105 110 115
2008 to 2017 Total Cost Percent Change from 75% Cost/25% Risk
120
Columbia market to identify the future value of each
asset to the company, as well as its inherent risk (e.
year-to-year volatility). Future market values and risk
were compared with the capital and fIXed operation
and maintenance (O&M) costs that would be incurred.
Avista s Preferred Resource Strategy Linear Programming
Model (PRiSM) assisted in selecting the FRS. Its
selection was based on forecasted energy and capacity
needs, resource values and limiting power supply expense
variability.
Futures and scenarios test the FRS under alternative
conditions beyond the Base Case and illustrate how
certain resource mixes perform in alternative market
conditions. Futures are stochastic studies using a Monte
Carlo approach to quantitatively assess risk around
expected mean outcome.3 This time-intensive and
multi-variable approach is the most robust method used
for risk assessment. Four futures were modeled for the
2007 IRP: Base Case Volatile Gas, Unconstrained Carbon
and a High Carbon Charges.
A scenario is a deterministic study that changes a
significant underlying assumption to assess the impact
of that change. Scenario results are easier to understand
and require less analytical effort than futures, but they do
not quantitatively assess the variability or risk around the
expected outcome. Seven scenarios were modeled for
the 2007 IRP, including high and low natural gas prices
varying regional load growth and a scenario in which the
Western Interconnect shifted all passenger automobiles to
electricity instead of petroleum fuel.
Two key challenges are addressed when developing
a resource portfolio-cost and risk mitigation. An
efficient frontier finds the optimal level of risk given
a desired level of cost and vice versa. This approach is
similar to finding the optimal mix of risk and return in
a personal investment portfolio. As the expected return
increases, so do risks; but reducing risk decreases overall
investment returns. Choosing the FRS is similar to
the investor s dilemma, but the trade-off is future costs
against future power supply cost variation. Figure 3
presents the changes in costs and risks from the 75/25
cost/risk position on the Efficient Frontier. It also
shows alternative resource portfolios to illustrate generic
resource strategies. The lower horizontal axis displays
the 2008-2017 percentage change in the present value
of existing and future costs. The upper horizontal axis
presents actual present value dollars. The right-hand
3 Stochastic studies use probability distributions (i., means and standard deviations) to forecast future variables.
Avista Corp 2007 Electric IRP Iii
Executive Summary
Figure 4: Base Case Stochastic Mid-Columbia Prices ($/MWh)
200
180 Mean
- Max & Min
. 80% Confidence Interval
---------------------- ----- -
160
140
120
---------------------------- --- -- ----------- -------------- ------ --- ------
100
(J)
(")
oo;t It)
....
vertical axis shows power supply volatility as a single
standard deviation of the average power supply expense.
The left-hand vertical axis shows the percent change in
2017 power supply volatility. Both axes are shown as
percentages of the 75/25 cost/risk mix to illustrate the
relative impacts of moving between resource strategies.
The blue dots represent the Efficient Frontier of various
resource portfolios developed by PRiSM to meet future
resource requirements. The PRS is not on the Efficient
- - - - - - - - - - - - - - - - - - - - -
r--(J)
(")
c:!i It)r--
Frontier curve because resource lumpiness is assumed in
the first 10 years of the study.4 The PRS is based on the
25/75 risk/cost portfolio weighting.
ELECTRICITY AND NATURAL GAS MARKET
FORECASTS
Figure 4 represents Avista s Base Case electricity price
forecast and the range of prices across its Monte Carlo
runs. The selected resource portfolio must provide a
hedge against such price movement.
Figure 5: Annual Average Sumas Natural Gas Price Results from 300 Iterations ($/Dth)
Mean
- Max & Min
- - - - - - - -- - - - - - -- -- - - - - - -- -- -- -- - - - - - - - - - - - - - - - - - - -
. 80% Confidence Interval
....(")....
oo;t It)
....- -
r--(J)
....
c:!i r--
4 Resources enter a utility portfolio in blocks that do not perfectly match load in a given year. For example, it is difficult to cost-effectively
acquire a 35 MW share of a CCCT plant. Instead, resources enter the utility portfolio in larger blocks and manage deficiencies for a period
of years.
2007 Eiectric IRP Avista Corp
Executive Summary
Electricity prices are highly correlated with natural gas
prices. Base Case natural gas prices across the Monte
Carlo simulations at the Sumas trading hub are shown
in Figure 5. Natural gas volatility is similar to electricity
price volatility in Figure 4.
DEMAND SIDE MANAGEMENT ACQUISITION
Figure 6 shows how conservation and energy efficiency
have decreased Avista s energy requirements by nearly
100 aMW since programs began in the late 1970s.
~ 120
r::
800
700
600
500
400
300
200
100
With additional funding recommended by the IRP
and through the Heritage Project, the company expects
accumulated conservation to lower its load growth 87
aMW by 2017. The 2007 IRP conservation acquisition
schedule is approximately 25 percent higher than the
2005 IRP and 85 percent higher than the 2003 IRP.
PREFERRED RESOURCE STRATEGY
The Preferred Resource Strategy is developed after
careful consideration of the information gathered
180
Figure 6: Cumulative Efficiency Acquisitions
120
150
- -- - - - - - - - - - - - - - - - - - - - - - - - - - - -
100
r::
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- -
-- 60
-- 40
ffi
.............................................
Figure 7: The 2007 Preferred Resource Strategy (aMW)
0 CCCT
. Wind
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
0 Conservation
. Other Renewables
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- -- -- --- - ---------
r--.
.....- -..........
C')
5 Actual energy savings total 124 aMW; however, due to expected degradation of historical measures (18-year average measure life),
cumulative savings are lower.
Avista Corp 2007 Electric IRP
Executive Summary
CCCT 280 280 280 350 350 350 350
Coal
Wind 100 100 200 300
Other Renewables
Conservation
Total 327 346 356 541 551 661 772
through the IRP process. The FRS is reviewed by
management and the Technical Advisory Committee.
The 2007 plan relies on conservation, system efficiency
upgrades, renewable resources and gas-fired combined-
cycle combustion turbines (CCCTs). Figure 7 illustrates
the company s Preferred Resource Strategy for the 2007
IRP.
The specific resources contained within the FRS, in
nameplate capability, are shown in Table 2.
The FRS requires between $1.0 and $1.5 billion in
new investments over the next 10 years.6 The 2007
IRP contains lower amounts of wind and other
renewable resources than were included in the 2005 IRP.
Conditions have changed since the 2005 IRP which
have and will impact the cost of renewable resources
relative to traditional thermal alternatives. Recent
legislation promoting renewable resources in Washington
and throughout the West have reduced the amount of
cost-effective renewable resources available to Avista by
increasing and accelerating demand in the short run.
Wind generation costs have increased by more than
100 percent over the past six years and by more than
50 percent since the 2005 IRP. Renewable resources
are being acquired to meet the Washington Energy
Independence Act, Initiative 937 (1-937), passed in
November 2006. This legislation requires larger utilities
in Washington to serve 15 percent of retail load with
renewables by 2020; intermediate targets are 3 percent
in 2012 and 9 percent in 2016. Under 1-937, Avista
must acquire renewable resources regardless of physical
resource balance. We forecast that by 2017 approximately
90 aMW ofI-937-qualifying resources will serve
customers loads, as shown in Figure 8.
160
Figure 8: Amount of Renewable Energy Forecasted to Meet RPS (aMW)
120
. Renewable Need
. Projected Qualified Resources
140
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
100
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - ----- - - - -- - -------- -- -
C\I C\I C\I
C\I
C\I
C')Il)
..-
C\I C\I
Il)
C\I
I'-
C\I
I'-
C\I
..-
C\I
..-
C\I C\I
C')
C\I C\I
The range reflects the possibility that the company might need to invest approximately $0.5 billion to fix the long-term price of its
natural gas (e., purchase of coal gasifier to create pipeline-quality natural gas).
2007 Electric IRP Avista Corp
Executive Summary
Figure 9: 1-937 Qualifying and Non-Qualifying Avista Renewables (aMW)
650
600
550
500
450
400
C\I
,,",,~,"..- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - --
C\I
....
Avista currently serves approximately one-half of
customer requirements with renewable resources (hydro
wind and biomass), and these resources will meet 40
percent of our load obligations in 2017. Unfortunately,
only a small portion of our current renewable resource
portfolio qualifies under 1-937, see Figure 9.
LOWERING VOLATILITY WITH LONG-TERM FIXED
PRICE GAS
Coal-fired generation accounted for a significant portion
of the Avista s FRS mix in both the 2003 and 2005
IRPs. Coal-fired plants provide a hedge against volatile
electricity and natural gas prices because 60 percent
or more of their costs are fIXed through large capital
investments. Variable operating and fuel costs at a coal
plant are modest compared to gas-fired resources. A
resource profile containing coal contributes to stable
power supply expenses.
The cost of operating gas-fired resources, on the
other hand, is highly correlated with the electricity
marketplace. Natural gas prices are volatile. The fIXed
costs of natural gas plants are low relative to their all-
cost, approximately 20 percent, reflecting a low capital
investment. Utility portfolios with large concentrations
of gas-fired generation can suffer from costs that are
C\I
C\I
(")....
C\I
oo;t
....
C\I
....
C\I C\I
r--
C\I
less stable than utilities who rely on other sources of
generation.
Gas-fired plants have not experienced the same rise in
capital costs that coal-fired plants have. In fact, recent
experience by Avista (Coyote Springs 2) and Puget
Sound Energy (Goldendale) indicate that independent
power producers in the Northwest marketplace are
willing to sell their gas-fired plants at prices below the
green field costs assumed in this plan. The enactment of
new laws imposing emission performance standards on
fossil-fueled generation resources acquired by electric
utilities in Washington and California will narrow
baseload technology options, at least in the short-term
to gas-fired generation. This restriction, coupled with
regional load growth and the prospect of additional
greenhouse gas regulations on fossil-fueled generation
resources, particularly coal-fired generation, may
ultimately increase demand for and the cost of gas-fired
plants.
Locking in natural gas costs through a long-term fIXed-
price contract, an investment in a pipeline-quality
coal gasification plant, an investment in gas fields or
through other means makes a gas-fired combined cycle
combustion turbine (CCCT) cost structure behave
Avista Corp vii2007 Electric IRP
Executive Summary
Figure 10: Efficient Frontier With and Without Fixed Price Gas Contracts Option
200
180
......
.l....... \.'..
....;.........~........ .......... j.........
I ...
- - - - - - - - -:- - - - - - - - - ~ - - ,.-- - - - - - - - -:- - - - - - - - - _: - - - - - - - - - ~ - - - - - - - --
I ,
'. - - - - - - - - -: - - - - - - - - - ~ - - - - - - - - . - - - - - -- - - :- - - - - - ~ - - -: - - - - - - - - - ~ - - - - - - - - -
::::~::NO Fixed Priced Gas -
~- - - -_~- +-+- - - - - - -.--- - ---- - - ---
~ 75% Cost 25% Risk
- ~ - - - - - - - - -:- - - - - - - - - -' - - - - - - - - - ~ - - - - - - -
.PRS
....
0 b
& :: 160
I: 'IiII:J
~ 140
1: ~
~ ~ 120
:. '"~ ;
100N ~
90 95 100 105
2008 to 2017 Total Cost Percent Change from PRS
115
financially like a coal-fired resource. Variable costs are
greatly reduced and are much less volatile because a
significant portion of its largest variable component-gas
fuel-is not tied to the natural gas market. In both
high and low gas market conditions the price paid by
customers is the same. In years where natural gas prices
are high, the flXed-cost contract looks very attractive
financially and customers pay less than if the company
relied on shorter-term purchases. On the other hand
years with low natural gas prices make the flXed-cost
contract look financially unattractive compared to a
short-term purchase. Over time, the long-run cost of
operations with flXed-price gas should parallel the cost
of operations where a gas plant is fueled with short-term
gas.
The company tested the benefits of flXed price contracts
with PRiSM and found that the model had a general
preference for flXed price gas because of its ability to
reduce risk. Even with premiums as high as 75 percent
above the forecasted short-term gas price, the PRiSM
model selects this resource option for a portion of the
preferred portfolio. In the Base Case, where a 30 percent
fixed gas price premium is modeled, risk is reduced by
7 A broader discussion of this study is presented in Chapter 8.
8 See www.energycommission.org
9 See www.eia.doe.gov
viii
110
approximately 20 percent, as shown in Figure 10.
empirical study by Avista explains that year-on-year
volatility for a hypothetical CCCT plant could have been
reduced by 50 percent during the years 2002-2006 were
flXed price gas used to fuel the plant.
CARBON EMISSIONS
Carbon emissions are included in the Base Case for the
first time in this IRP cycle. The National Commission
on Energy Policy study, completed in late 2004, provided
the basis for pricing carbon emissions in the Base Case.
To quantify potential risks inherent in a higher carbon
emission cost scenario, the company looked to an Energy
Information Administration study of the McCain-
Lieberman Climate Stewardship Act.9 These two cases
illustrate the potential risk inherent in relying too heavily
on traditional carbon-emitting technologies.
Avista has one of the smallest carbon footprints in the
United States because of its existing renewable energy
resources. Out of the top 100 producers of electric
power in the 2006 Benchmarking Air Emissions study
by the Natural Resources Defense Council, only seven
other utilities have a smaller footprint. However, the
2007 Electric IRP Avista Corp
Executive Summary
Figure 11: Carbon Footprint (Tons per MWh)
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - --------
Thermal Generation
- - - -
_2005 IRP
------- -----------------
........ 2007 IRP
- --------- - - ----- - - - --- -
Total Generation
0.40
- -(j)
company s carbon footprint is forecast to rise over
the IRP timeframe because it would be very difficult
to acquire sufficient amounts of additional cost-
effective renewable resources to meet all future load
growth. Figure 11 forecasts Avistas carbon footprint
for generation and compares it to the 2005 IRP. Our
emissions footprint is approximately 25 percent lower.
LANCASTER
The company announced the sale of its energy marketing
company, Avista Energy, in April 2007. It subsequently
announced that Avista Energy s contract for the Lancaster
Generation Facility output is available to the utility
beginning in 2010. The announcement came after the
company had substantially completed its IRP analysis
and Preferred Resource Strategy. Given that Lancaster
is the same technology and available in the same
timeframe as the 280 MW gas-fired combined cycle
resource identified in the PRS, the resource strategy was
not updated. Instead, an alternative portfolio including
Lancaster is compared to the PRS to illustrate its impacts.
The Lancaster Generation Facility is a 245 MW gas-
fired combined-cycle combustion turbine with an
- -- -"'"
r--
additional 30 MW of duct firing capability. It is a newer
General Electric Frame 7FA that began commercial
service in 2001. Avista controls plant operations
under a tolling arrangement through 2026. Recently
completed preliminary analysis has identified Lancaster
as a potentially cost-effective resource to meet customer
load requirements. The plant is located in Rathdrum
Idaho, in the center of Avista s service territory. It is
significantly lower in cost than a green field plant.
LANCASTER IMPACT ON L&R BALANCES
Lancaster substantially replaces the identified gas-fired
CCCT plant included in the PRS. Table 3 presents the
company s net position with the inclusion of Lancaster.
Figure 12 reflects Lancaster s inclusion in our loads and
resources tabulation.
ACTION ITEMS
Avista s 2007 Action Plan outlines the activities
and studies to be developed and presented in the
2009 Integrated Resource Plan. The Action Plan
was developed with input from Commission Staff
Avista s management team, and the Technical Advisory
Table 3: Net Position Forecast with Lancaster
Avista Corp 2007 Electric IRP
Executive Summary
Figure 12: Loads & Resources Capacity Forecast with Lancaster (MW)
500
2,400
300
200
100
000
900
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
800
700 '
600
500
C\I C\I C\I
......
C\I
......
C\I
(")
C\I
"'"
U')
......
C\I
r--
Committee. The Action Plan is found in Chapter 9.
Categories of action items include renewable energy
and emissions, modeling enhancements, transmission
modeling and research, and conservation.
2007 Electric IRP Avista Corp
Chapter 1- Introduction and Stakeholder Involvement
INTRODUCTION AND STAKEHOLDER INVOLVEMENT
Avista submits a biennial Integrated Resource Plan (IRP)
to the Idaho and Washington public utility commissions.
The 2007 IRP is Avista s 10th plan. It describes the
Preferred Resource Strategy for meeting customers
future requirements while balancing cost and risk.
The company has a statutory obligation to provide
reliable electric service to customers at rates, terms and
conditions that are just, reasonable and sufficient. We
assess resource acquisition strategies and business plans
to meet resource adequacy and renewable portfolio
requirements, and to optimize the value of our current
resource portfolio. Avista uses the IRP as a resource
evaluation tool rather than an acquisition plan. The 2007
IRP focuses on refining our processes for evaluating
resource decisions, requests for proposal and other
acquisition efforts.
IRP PROCESS
Avista actively seeks input from a variety of constituents
including Commission Staff, customers, academics and
other interested parties. The company sponsored five
Technical Advisory Committee (TAC) meetings for
the 2007 IRp, including a two-day meeting in August
2006. The TAC process began on February 24 2006
and ended with a final meeting on April 25, 2007. Over
90 people were invited. Each TAC meeting covered
different aspects of the 2007 IRP planning activities
and solicited contributions and assessments of modeling
assumptions, processes and results. The 2007 IRP marked
the first time that the company provided TAC members
with a draft Preferred Resource Strategy (PRS) in the
middle of the IRP process. The FRS was presented at
the second TAC meeting. It gave TAC participants an
opportunity to understand the potential results of the
IRP modeling process.
STAKEHOLDER PARTICIPATION
The IRP process provides substantial opportunities for
stakeholders to participate in Avista s resource planning
activities. Avista utilizes three different groups of
stakeholders. The main contingent involves stakeholders
with some level of expertise in utility planning, who
provide input concerning the IRP studies, resource data
modeling efforts, and critical review of the modeling
results. This group includes Commission Staff, planners
from other utilities, academics and consultants. The
second group includes parties who are involved with a
critical aspect of the IRP. Examples of members of this
group include environmental advocates and government
agencies. The third group includes delegates from
regional planning efforts, such as the Northwest Power
and Conservation Council and the Western Electricity
Coordinating Council.
PUBLIC PROCESS
The 2007 IRP is a publicly-developed document. All of
the 2007 IRP TAC presentations, along with past IRPs
and TAC presentations, are available for review at www.
avistautilities.com. The entire 2007 IRp, its technical
appendices, and its supporting documents can be
downloaded from this location.
TECHNICAL ADVISORY COMMITTEE
Avista s Integrated Resource Plan benefits from public
input and involvement. The company held six full days
of TAC meetings, which were supplemented with phone
and email contact, to develop this plan. Some of the
topics included in the 2007 TAC series were resource
options, conservation, modeling, fuel price forecasts, load
forecasts, market drivers and
eIIllsslons Issues.
Washington IRP requirements are contained in WAC 480-100-251 Least Cost Planning. Idaho IRP requirements are outlined in Case
No. U-1500-165 Order No. 22299, Case No. GNR-93-, Order No. 24729, and Case No. GNR-93-, Order No. 25260.
Avista Corp 2007 Electric iRP
Chapter 1- Introduction and Stakeholder Involvement
WSUCi of S okane
IPUC
Resource Develo ment Associates
WUTC
WUTC
WUTC
WUTC
Public Counsel
Pu et Sound Ener
WA State Gen Admin
IPUC
Idaho Power
NPCC
CTED
Inland Em
The TAC mailing list includes more than 90 individuals
from 42 different organizations. Avista greatly appreciates
all of the time and effort expended by participants in the
TAC process and we look forward to their continued
involvement in future IRPs. The company would like to
particularly thank the participants listed in Table 1.1 for
their input and involvement.
ISSUE-SPECIFIC PUBLIC INVOLVEMENT ACTIVITIES
In addition to the TAC, Avista sponsors and participates
in other collaborative processes involving public
interests.
External Energy Efficiency ("Triple E'? Board
Since 1995 the Triple E Board has been meeting
biannually to gather and provide guidance on
conservation efforts. The Triple E grew out of the DSM
Issues Group, which was influential in developing the
country s first distribution surcharge for conservation
acquisition.
FERC Hydro Relicensing Clark Fork River Projects
Over 50 stakeholder groups participated in the Clark
Fork hydro-relicensing process beginning in 1993. This
led to the first all-party settlement filed with a FERC
relicensing application and eventual issuance of a 45-
year Federal Energy Regulatory Commission (FERC)
operating license in February 2003. The nationally
recognized Living License concept was a result of
this process. This collaborative process continues
implementing the Living License with stakeholders
participating in various protection, mitigation and
enhancement measures. These measures include the
purchase of over 1 100 acres of wetland and upland
habitat for the bull trout, fish passage programs and
improvements to 19 recreational facilities along the
reservOIr.
FERC Hydro Relicensing Spokane River Projects
Our Spokane River Project license expires in August
2007. Avista s hydro relicensing process for the
Spokane River Projects mimics the Clark Fork process.
Approximately 100 stakeholder groups participate in this
collaborative effort. Draft license applications were filed
with FERC on July 28 2005. FERC recently released a
draft Environmental Impact Statement and held a public
hearing in Spokane on February 8, 2007.
Low Income Rate Assistance Program (LIRAP)
LIRAP is developed through regular meetings with four
1 - 2 2007 Electric IRP Avista Corp
Chapter i-Introduction and Stakeholder Involvement
TAC 1 - February 24 2006
TAC 2 (Day 1) - August 31 , 2006
TAC 2 (Day 2) - September 1 , 2006
TAC 3 - January 10, 2007
TAC 4 - March 28, 2007
TAC 5 - April 25, 2007
community action agencies in the company s Washington
service territory. The program began in 2001 to review
administrative issues and needs. Meetings are held
quarterly.
REGIONAL PLANNING
The Pacific Northwest's generation and transmission
system is operated in a coordinated fashion. Avista
participates in the activities of many organizations
planning efforts. Information from this participation
is used to supplement its integrated resource planning
process. Some of the organizations that Avista
participates in include:
IRP Rules and Regulations
Work Plan Discussion
20051RP and TAC Comments
2007 IRP Topic Discussions: Resource Planning,
Conservation , Analytical Process, and Capacity
Plannin
Review of 2005 Action Plan
IRP Modeling Overview: Emissions, Fuel Price
Forecasting, Modeling Assumptions, Preliminary
Transmission Costs and Paths, Resource Options
and Cost Assumptions, and Futures and Scenarios
2006 Renewables RFP
Future Resource Requirements (L&R)
Review of Futures and Scenarios Market Results
Prelimina Preferred Resource Strate PRS
Preliminary PRS Discussion: Portfolio Selection
Criteria, Futures & Scenarios, PRS Selection Model
and Results
Alternative Ener
Draft PRS Review
Fuel Price Forecast
Clean Coal Presentation
Emissions Update
Load Forecast
Conservation
Market Analysis
Conservation Program Update
Portfolio Selection Criteria
Cost of Service
Transmission Estimates
2007 IRP Draft Outline
Presentation of the 2007 PRS
2007 IRP Action Items
. Western Electricity Coordinating Council
. Northwest Power and Conservation Council
. Northwest Power Pool
. Pacific Northwest Utilities Conference Committee
. ColumbiaGrid
. Northwest Transmission Assessment Committee
. Seems Steering Group - Western Interconnection
. North American Electric Reliability Council
FUTURE PUBLIC INVOLVEMENT
Avista will continue to actively solicit input from
interested parties. Advice will be requested from
members of the Technical Advisory Committee on
Avista Corp 1 - 32007 Electric IRP
Chapter 1- Introduction and Stakeholder Involvement
a wide variety of resource planning issues. We will
continue to work on diversifying TAC membership and
will strive to maintain the TAC meetings as an open
public process.
2007 IRP OUTLINE
The 2007 IRP consists of eight chapters plus an
executive summary and this introduction. A series of
technical appendices supplement this report.
EXECUTIVE SUMMARY
This chapter summarizes the overall results and highlights
key aspects of the 2007 IRP.
CHAPTER 1: INTRODUCTION AND STAKEHOLDER
INVOLVEMENT
This chapter introduces the IRP and provides details
concerning public participation and involvement in the
integrated resource planning process.
CHAPTER 2: LOADS AND RESOURCES
The first half of this chapter covers Avista s load forecast
along with relevant local economic forecasts. The last
half of this chapter describes the company s owned
generating resources, major contractual rights and
obligations, capacity and energy tabulations, and reserve
Issues.
CHAPTER 3: DEMAND SIDE MANAGEMENT
This chapter provides an overview of Avista s energy
efficiency programs, descriptions and analysis of
efficiency measures for the IRP and the selected
programs for the 2007 IRP.
CHAPTER 4: ENVIRONMENTAL ISSUES
This chapter covers emissions issues that were modeled
in the 2007 IRP. The chapter focuses on modeling
efforts and issues surrounding SO ' NO ' Hg and CO
State and federal emissions regulations and policies are
also discussed.
1 -
CHAPTER 5: TRANSMISSION PLANNING
This chapter reviews Avista s distribution and
transmission systems, as well as regional transmission
planning issues. Transmission cost studies used in
modeling efforts are also covered in this chapter.
CHAPTER 6: MODELING APPROACH
This chapter provides the Mid-Columbia and Western
Interconnect market results for the Base Case and
scenario analyses.
CHAPTER 7: MARKET MODELING RESULTS
This chapter covers the results of the Base Case and
scenario analyses for the Western Interconnect and
Mid-Columbia electricity market.
CHAPTER 8: PREFERRED RESOURCE STRATEGY
This chapter provides details about Avista s 2007
Preferred Resource Strategy. It compares the FRS to
a variety of theoretical portfolios under stochastic and
scenario based analyses.
CHAPTER 9: ACTION ITEMS
This chapter reviews the progress made on the 2005 IRP
Action Items and describes the 2007 IRP Action Items.
2007 Electric IRP Avista Corp
Chapter 2- Loads and Resources
LOADS AND RESOURCES
INTRODUCTION & HIGHLIGHTS
Loads and resources represent two key components
the IRP. The first half of this chapter summarizes
customer and load forecasts for our service territory,
including high and low forecasts, load scenarios and an
overview of recent enhancements to our forecasting
models and processes. The second half covers our
resources, including company owned and operated
resources, as well as long-term contracts.
UTILITY LOADS
ECONOMIC CONDITIONS IN THE ELECTRIC SERVICE
TERRITORY
Avista serves a wide area of Eastern Washington and
Northern Idaho. This area is geographically and
economically diverse. Avista serves most of the urbanized
and suburban areas in 24 counties. Figure 2.1 is a map of
the company s electric and natural gas service territory.
Sandpoint, Idaho
CHAPTER HIGHLIGHTS
. Strong economic growth continues throughout the company s service territory.
. Historic conservation acquisitions are included in the load forecast; higher acquisition levels envisioned in this
plan will be in addition to levels included in the forecast.
. Electricity sales growth averages 2.3 percent over the next 10 years (254 aMW) and 2.0 percent over the
entire 20-year forecast.
. Peak loads are expected to grow at 2.4 percent over the next 10 years (400 MW) and 2.1 percent over the
entire 20-year forecast.
. Avista s resource deficits begin in 2011, 2014 with the Lancaster plant.
. Capacity deficiencies drive our resource needs.
Avista Corp 2007 Electric IRP 2 - 1
Chapter 2- Loads and Resources
Figure 2.1: Avista s Service Territory
The economy of the Inland Northwest has transformed
over the past 20 years, from natural resource-based
manufacturing to diversified light manufacturing and
services. Much of the mountainous area of the region is
owned by the Federal government and managed by the
United States Forest Service. Timber harvest reductions
on public lands have closed many local sawmills. Two
pulp and paper plants served by Avista have large forest
land holdings, but they continue to face stiff domestic and
international competition for their products.
Employment expands during expansionary times and
contracts during recessions. Our service territory
experienced large scale unemployment during two
national recessions in the 1980s. Avista s service territory
was mostly bypassed by the 1991/92 national recession
but it was not as fortunate during the 2001 recession.
The effects of recessions and economic growth are
best illustrated by employment for the three principal
Electric Service Area
Natural Gas Service Area
counties in the company s electric service area. Regional
employment data is provided later in this chapter.
Population levels often are more stable than employment
levels during times of economic change; however, total
population often contracts during severe economic
downturns as people leave in search of job opportunities.
Over the past 20 years, only in 1987 did the region
experience a net loss in population. Figure 2.2 details
annual population changes in Bonner, Kootenai and
Spokane counties. Figure 2.3 shows total population in
these three counties.
ECONOMIC, CUSTOMER, AND SALES FORECASTS
People, Jobs and Customers
Avista purchases national and county-level employment
and population forecasts from Global Insight, Inc. Global
Insight is an internationally recognized economic
forecasting consulting firm used by various agencies in
Washington and Idaho. The data encompasses the three
2 - 2 Avista Corp2007 Electric IRP
Chapter 2- Loads and Resources
Figure 2.2: Population Change for Spokane, Kootenai and Bonner Counties (Thousands)
---------------
r--
......g;)......
(Y)
............!;;............
(Y)
............
r--(Y)
......
r--
............
(Y)r--
............
Figure 2.3: Total Population for Spokane, Kootenai and Bonner Counties (Thousands)
000
900
800
. Spokane County
. Kootenai County
IJ Bonner County
------------- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
I II II II I I I I II I I I I II I I I I I I II I I I I I I I
I. . I I. . ..1. . .1.. .1. . . .I~
.. .
I~ I . .1. ~ 1.1. . I .1. . I .
700
600
- --- - - - ------- --- ------------------- - -
500
400
300
200
100
r--
............
(Y)
......
r--
............
(Y)
............
r--(Y)
......
r--(Y)r--
............
Table 2.1: Globallnsi hts National Forecast Assum
Gross Domestic Product
Consumer Price Index
West Texas Crude
Treasu Bonds
Unem 10 ment Rate
5%-
$60-$65
0%-
..:::5.
principal counties which comprise over 80 percent of
our service area economy, namely Spokane County
in Washington and Kootenai and Bonner counties in
Idaho. The national forecast is based on regional forecasts
prepared in March 2006; county-level estimates were
completed inJune 2006.
60-
5%-
The forecast and underlying assumptions used in this
IRP were presented at the third Technical Advisory
Committee meeting for Avista s 2007 Integrated
Resource Plan onJanuary 10,2007. Key forecast
assumptions are shown in Table 2.
Avista Corp 2 - 32007 Electric IRP
Chapter 2- Loads and Resources
Figure 2.4: Three-County Population Age 65 and Over (Thousands)
180
160 . Spokane County
. Kootenai County
0 Bonner County
----------------
140
120
----------------- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
100 - - - - - .
. - - - - - - - - - - - - - - -- -- -- - - - - - - - - - -- -..................
C')
..................
C')
......
Looking forward, the national economy slowed after
recovering from the 2001 recession, setting the stage for
regional economic performance in Avista s service area in
Eastern Washington and Northern Idaho. As shown in
the charts above, population growth has rebounded after
slow growth from 1997 to 2002. Population growth is
expected to continue its recent trend through 2010.
Regional population growth is supported by the
emigration of retirees, representing between 10 and 20
percent of overall population growth. Figure 2.4 presents
the population history and forecasts for individuals 65
years and over in the three-county area. Between 1986
and 2006 this segment grew by compound growth
C')
..................
C')
......
rates of 2.4 percent in Bonner County, 2.0 percent in
Kootenai County and 0.5 percent in Spokane County.
This age group represented 13 percent of the overall
population in 2006. The forecast predicts growth of 2.5
percent, 4.5 percent and 3.5 percent, respectively, pushing
the overall contribution of this age group to 19 percent
in 2027.
Employment growth drives population growth. Figure
5 shows employment trends in the prior two and future
two decades.
Overall non-farm wage and salary employment over the
past 20 years averaged 3.7 percent for Bonner County,
Figure 2.5: Three-County Job Change (Thousands)
------------- - - - - - - - - - - - - - - - - - - - - - - - - -- - ------- -------- - - - - -- - -------- - - - -- ---............
C')
......
C')
........................
C')C')
............
2 - 4 2007 Electric IRP Avista Corp
Chapter 2- Loads and Resources
Figure 2.6: 3-County Non-Farm Jobs (Thousands)
- -- - - -- -- - -.. - .
1.
.. .
1.. . .1. . . . I . . . . . .I .1.
. . .
450
400 . Spokane County
- - -
. Kootenai County
- - - -- -- -- -- -- -- - ---
D Bonner County350
300
250
200
150
100
foo-
............(")............
foo-
............(")
c;;
......
1 percent for Kootenai County and 2.1 percent for
Spokane County. See Figure 2.6. Over the forecast
horizon, growth rates are predicted at 2.6 percent, 3.
percent and 2.6 percent, respectively. As indicated in the
following chart, employment growth is expected to equal
approximately 7 500 new jobs annually.
Customer growth projections follow from baseline
economic forecasts. The company tracks four key
customer classes-residential, commercial, industrial and
street lighting. Residential customer forecasts are driven
by population. Commercial forecasts rely more heavily
on employment and residential growth trends. Industrial
(")
I:;
(")
foo-
......
customer growth is correlated with employment growth.
Street lighting trends with population growth.
Avista forecasts sales by rate schedule. The overall
customer forecast is a compilation of the various
rate schedules of our served states. For example
the residential class forecast is comprised of separate
forecasts prepared for rate schedules 1 , 12 22 and 32 for
Washington and Idaho. See Figure 2.
Avista served 300,928 residential customers, 37 911
commercial customers, 1 388 industrial customers and
425 street lighting customers, or a total of 340,652 retail
550
Figure 2.7: Avista Annual Average Customer Forecast (Thousands)
500
. Street Lights
. Commercial
. Residential
(J Industrial
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
2007-2027 Growth Rate=1.
450
- -
400
---- --- - - - - - --------
1997-2007 Growth Rate=1.
350
- - - - - - - - - - - - - - - - - - - - - - -
300
250
foo-
(")
foo-
......(")
foo-foo-
(")
Avista Corp 2007 Eiectric IRP 2 - 5
Chapter 2- Loads and Resources
electricity customers in 2006. The 2027 forecast predicts
440,789 residential, 53 322 commercial, 1 795 industrial
and 625 street lighting customers for a grand total of
496 532. The 20-year compound growth rate averages
8 percent.
WEATHER, PRICE ELASTICITY, PRICES, CONSERVATION
AND USE PER CUSTOMER
Weather Forecasts
The baseline electricity sales forecast is based on 30-
year normal temperatures for the station at the Spokane
International Airport, as tabulated by the National
Weather Service from 1971 through 2000. Daily values
go back as far as 1890. There are several other weather
stations with historical records in the company s electric
service area; however that data is available over a much
shorter duration. Sales forecasts are prepared using
monthly data, as more granular load information is not
available. The company finds high correlations between
the Spokane International Airport and other weather
stations in its service territory. It uses heating degree
days to measure cold weather and cooling degree days to
measure hot weather in its retail sales forecast.
In response to questions from its Technical Advisory
Committee, the company has prepared a study of the
possible impacts of climate change on its retail load
forecast. Ample evidence of cooling and warming trends
exists in the 115-year record. In recent years the trend
has been one of a warming climate when compared
to the 30-year normal. Recent trends in heating and
cooling degree days for Spokane are roughly equal to
the scientific communitys predictions for this coordinate
on the globe, implying a one-degree warming every
25 years. Extrapolating the trend finds that in 20 years
summer load would be approximately 26 aMw, a 2.
percent, higher than the Base Case. In the winter, loads
would be approximately 40 aMw, or 2 percent, lower.
This change likely would occur gradually, and it appears
that approximately one-third to one-half of this trend is
already captured in our load forecast. The company will
continue to study these data trends in its two-year Action
Plan and report any additional findings in the 2009
Integrated Resource Plan.
Price Elasticity
Price elasticity is a central economic concept of
projecting electricity demand. Price elasticity of demand
is the ratio of the percent change in the quantity
demanded of a good or service to a percentage change
in its price. In other words, elasticity measures the
responsiveness of buyers to changes in electricity prices.
A consumer who is sensitive to price changes has a
relatively elastic demand profile. A customer who is
unresponsive to price changes has a relatively inelastic
demand profile. During the 2000-01 energy crisis
customers showed their sensitivity, or price elasticity,
of demand, reducing their overall electricity usage in
response to price increases.
Cross elasticity of demand, or cross-price elasticity, is the
ratio of the percentage change in the quantity demanded
of one good to a one percent change in the price of
another good. A positive coefficient indicates that
the two products are substitutes; a negative coefficient
indicates they are complementary goods. Substitute
goods are replacements for one another. As the price
of the first good increases relative to the price of the
second good, consumers shift their consumption to the
second good. Complementary goods are used together;
increases in the price of one good result in a decrease in
demand for the second good along with the first. The
principal cross elasticity impact on electricity demand is
the substitutability of natural gas in some applications
including water and space heating.
Income elasticity of demand is the ratio of the percentage
change in the quantity demanded of one good to a 1
percent change in consumer income. Income elasticity
measures the responsiveness of consumer purchases to
2 - 6 Avista Corp2007 Electric IRP
Chapter 2- Loads and Resources
Figure 2.8: Household Size Index (% of 2007 Household Size)
102
101
100
......
C1)C1)
.....- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -......
income changes. Two impacts affect electricity demand.
The first is affordability. As incomes rise, a consumer
ability to pay for goods and services increases. The
second income-related impact is the amount and number
of customers using equipment within their homes and
businesses. Simply stated, as incomes rise consumers
are more likely to purchase more electricity-consuming
equipment, live in larger dwellings and use their electrical
equipment more often.
The correlation between retail electricity prices and the
commodity cost of natural gas has increased in recent
years. We estimate customer class price elasticity in
our computation of electricity and natural gas demand.
Residential customer price elasticity is estimated at
negative 0.15. Commercial customer price elasticity
estimated at negative 0.10. The cross-price elasticity of
natural gas and electricity is estimated to be positive 0.05.
Income elasticity is estimated at positive 0., meaning
electricity is more affordable as incomes rise.
Retail Price Forecast
The retail sales forecast is based on retail prices increasing
an average of3.5 percent annually from 2007 to 2027.
The rate changes are lumpy, rising by 17.5 percent every
five years (five percent above the overall inflation rate).
- - - --- - - - - ----- - - -............
Conservation
It is very difficult to separate the interrelated impacts of
rising electricity and natural gas prices, rising incomes
and conservation programs. We only have data on
total demand and must derive the impacts associated
with consumption changes. The company has offered
conservation programs to its customers since 1978.
The impact of conservation on electrical usage is fully
imbedded in the historical data; therefore, we concluded
that existing conservation levels (5 aMW) are imbedded
in the forecast. Where conservation acquisition decreases
from this level, retail load obligations would increase. As
this IRP forecasts growing conservation acquisition, this
growth reduces retail load obligations.
Use per Customer Projections
Monthly electricity sales and customers by rate schedule
customer class and state from 1997 to 2006 make up the
database used to project usage per customer. Historical
data is weather-normalized to remove the impact of
heating and cooling degree day deviations from expected
normal values, as discussed above. Retail electric price
increase assumptions are applied to price elasticity
estimates to estimate price-induced reductions in
electrical use per customer.
Avista Corp 2007 Electric IRP 2 - 7
Chapter 2- Loads and Resources
The underlying increase in residential use per customer
over the long term is 0.5 percent per year, consistent
with the income elasticity and growth rate per customer.
As shown by Figure 2., the number of persons per
household declines slightly over the next 20 years.
Residential customers tend to be homogeneous relative
to the size of their dwellings. Commercial customers, on
the other hand, are heterogeneous, ranging from small
customers with varying electricity intensity per square
foot of floor space to big box retailers with generally
high intensities. The addition of new large commercial
customers, specifically the largest universities and
hospitals, can greatly skew the average use per average
customer. Customer usage is illustrated in Figure 2.
Estimates for residential usage per customer across all
schedules are relatively smooth. Commercial usage per
customer is forecast to increase for several years, due to
additional buildings either built or anticipated to be built
at several existing very large customers and in particular
at Washington State University campuses in Spokane
and Pullman. For very large customers, we include
expected additions through 2011; after 2011 no additions
are included in the forecast. We will include publicly-
announced long lead time buildings into the forecast
included in future IRPs.
RETAIL ELECTRICITY SALES FORECAST
Between 1997 and 2006 the region was affected by
major economic changes, not the least of which was
a marked increase in retail electricity prices. The
energy crisis of2000-01 included the implementation
of widespread, permanent conservation efforts by our
customers. In 2004, rising retail electricity rates further
reinforced conservation efforts. Several large industrial
facilities served by the company closed permanently
during the 2001-02 economic recession.
The electric retail sales forecast takes a somewhat
conservative approach by assuming closures are
permanent. If these industrial facilities reopen, the
annual electricity retail sales forecast presented in this
plan will be adjusted. Retail electricity consumption
rose 2.3 percent annually from 1997 through 2006. This
increase was despite the combined impacts of higher
prices and decreased electricity demand during the
energy crisis. The forecasted average annual increase in
firm sales over the 2007 to 2027 period is 2.0 percent.
The sales forecast takes a "bottom up" approach
summing forecasts of the number of customers and usage
per customer to produce a retail sales forecast. Individual
forecasts for our largest industrial customers (Schedule
25) include planned or announced production increases
12.
Figure 9: Use per Customer
110
...
100 E
90 "i6
- -------------------------------
80 ~
~ 12.0 --
~ 11.
:;:I
III
w 11.0 -
-- -- - --...
III
~ 10.
...
III
------------------- -----------------------------------
70 ~Residential
Commercial
10.
I"-
......
ffi
......
C")
C\I C\I C\I C\IC\I
......
C")
......
C\I
I"-
C\IC\I
l"-
C\I C\I C\I C\I
C\i
C\I
2 - 8 Avista Corp2007 Electric IRP
Chapter 2- Loads and Resources
16,000
Figure 2.10: Avista s Retail Sales Forecast
000
. Street Lights
. Commercial
. Residential
() Industrial
----------- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - - 1.
000
000
8,000
000
000
000
r--
..-..-
It')r--
or decreases. Lumber and wood products industries
are ramping down from very high production levels
which is consistent with the decline in housing starts at
the national level. The load forecasts for these sectors
were reduced to account for decreased production
levels. Anticipated sales to aerospace and aeronautical
equipment suppliers have increased and local plants have
announced plans to hire more workers and increase their
output.
Actual (i., not weather corrected) retail electricity sales
to Avista customers in 2006 were 8.78 billion kWh.
Heating degree days in 2006 were 93 percent of normal
almost completely offset in terms of energy use by 156
9 C)
CIS
r--
..-..-..-
It')
..-
It')r--
..-..-
percent of normal cooling degree days. The forecast
for 2027 is 13.4 billion kWh, representing a 2.0 percent
compounded increase in retail sales. See Figure 2.10.
Load Forecast
Load forecasts are derived from retail sales. Retail sales
in kilowatt hours are converted into average megawatt
hours using a regression model to ensure monthly load
shapes conform to history. The company s load forecast is
termed its Native Load. Native Load is net of line losses
across the Avista transmission system.
Native Load growth is indicated in Figure 2.11.
Note the significant drop in 2001 during the energy
700
Figure 2.11: Annual Net Native Load (aMW)
600 - - - - - - - - - - - - - - - - - - - - - - - -
500
- -
1,400
- -
300
- -
200
- - - - - - - - - - - - - - - - - - - -
100
------------ - - - - - - - --- - ---
000
900
r--It')r--
..-- -- -- - -
~2005 IRP
_2007 IRP
..-
r--It')It')r--
Avista Corp 2007 Electric IRP 2 - 9
Chapter 2- Loads and Resources
Figure 2.12: Calendar Year Peak Demand (MW)
000
800
600
2,400
200
000
800
600
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - ------ - --------- ------- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - ----- --- - ----------- - ------ - - -- - -- - - --- ---- - - - ----------------- ---- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
1,400
200
000
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
foo-
(j)(j)......(j)(j)(j)......
C')foo-
(j)......
crisis. The loads from 1997 to 2006 are not weather
normalized. The 2005 IRP load forecast is presented for
comparison purposes. Loads are modestly lower in the
2007 IRP compared with the 2005 IRP.
Peak Demand Forecast
The peak demand forecast in each year represents the
most likely value for that year. It does not represent
the extreme peak demand. In statistical terms, the
most likely peak demand has a 50 percent chance of
exceedance in any year. The peak forecast is produced
by running a regression between actual peak demand and
net native load. The peak demand forecast is in Figure
12. Peak loads are expected to grow at 2.4 percent
............
C')f0o-
(j)......
C')foo-
between 2007 and 2017 (400 MW) and 2.1 percent over
the entire 20-year forecast.
Historical data are significantly influenced by extreme
weather data. The comparatively low 1999 peak demand
figure was the result of a warmer-than-average winter
peak day; the peak in 2006 was the result of a below-
average winter peak day. The 1999 and 2006 peak
demand values illustrate why relying on compound
growth rates for the peak demand forecast is an
oversimplification and why the company plans to own or
control enough generation assets and contracts to exceed
expected peak demand.
Figure 2.13: Comparison of Summer and Winter Peak Demand (MW)
200
100
000
900
800
700
600
500
::::' :::::::::::::::::::::::: - _:::: _:::
400
300
200
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -(j)............
C')
......
"tt
......
foo-
......
2 -Avista Corp2007 Electric IRP
Chapter 2- Loads and Resources
Avista has witnessed significant summer load growth
as air conditioning penetration has risen in its service
territory. That said Avista expects to remain a winter-
peaking utility in the foreseeable future. It is possible
that very mild winter weather and extremely hot
summertime temperatures could result in our summer
peak load exceeding our wintertime demand level. This
will be an anomaly. Figure 2.13 illustrates our forecast of
winter and summer peak demands through 2017 and the
expected range of the forecasts at the 80 percent
confidence level. We expect that loads in the summer
and winter of each year have a 10 percent probability of
being higher than shown. Winter peak demand exceeds
summer peak demand in all years; the possibility of a
summer peak being higher than a winter peak in the
same year is possible.
FORECAST SCENARIOS
The discussion so far has concentrated on the Base Case
or most-likely, electricity sales forecast. Forecasting is
inherently uncertain, and alternative electricity growth
scenarios are used to provide insight and guidance for
our resource acquisition plans. At the request of the
Technical Advisory Committee, high and low economic
forecasts were prepared to illustrate how variable our load
forecast might be.
The principal driver of these alternatives is the
standard deviation of annual loads between 1997 and
2006. The average growth rate for the 10-year period
was 2.4 percent, and the standard deviation was 2.
percent. Approximately 75 percent of year- on-year
variation is driven by weather, leaving 25 percent to the
non-weather factors we are interested in evaluating here.
The 80 percent confidence interval (with a 10 percent
chance of exceedance on the high side and a 10 percent
chance of exceedance on the low side) produced a range
of growth for the 20-year period between 0.9 percent
and 3.1 percent. This range is roughly in line with other
Pacific Northwest forecast scenarios.
Avista is not forecasting any changes to its service
territory in these scenarios. Such changes, were they
to occur, would be outside of the scope of this exercise.
Alternative forecasts are presented in Figure 2.14.
Developed specifically for the IRp, these alternative
forecasts should not be confused with other company
or agency forecasts. The scenarios are not boundary
forecasts in that the high forecast should not be
considered the highest possible load trajectory; the low
forecast does not represent the lowest possible forecast.
250
Figure 2.14: Electric Load Forecast Scenarios (aMW)
000
_2007 IRP (Base Case)
~High Case
......... Low Case
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
750
- -- - - - - - - - - - - - - - - - - - -
500
250
------------------- ------
000
- -
750
r--
............
CO)
C\I
r--
C\I
......
C\I
- -------......
C\I
CO)r--CO)C\I C\I r--
Avista Corp 2 -2007 Electric IRP
Chapter 2- Loads and Resources
Figure 2,15: Avista s Hydroelectric Projects
LOADS & RESOURCES
The company relies on a diversified portfolio of
generating assets to meet customer loads. Avista owns
and operates eight hydroelectric projects located on the
Spokane and Clark Fork Rivers. Its thermal assets
include partial ownership of two coal-fired units in
Montana, three natural gas-fired projects within its
service territory, another natural gas-fired project
in Oregon and a biomass plant near Kettle Falls
Washington.
SPOKANE RIVER HYDROELECTRIC PROJECTS
Avista owns and operates six hydroelectric projects on
the Spokane River. FERC licensing for these projects
expires on July 31 2007 (except for Little Falls, which
is state licensed). The company is actively working
with stakeholders on relicensing for the Spokane River
Project. Following is a short description of the Spokane
River projects, including the maximum capacity and
nameplate ratings for each plant. The maximum capacity
of a generating unit is the total amount of electricity that
a particular plant can safely generate. This is often higher
than the nameplate rating because of facility upgrades.
The nameplate, or installed capacity of a plant, is the
plant's capacity as rated by the manufacturer. Figure 2.
is a map of all company-owned hydroelectric projects.
Post Falls
The Post Falls plant, located at its Idaho namesake, began
operation in 1906. Generation was expanded in 1980
with an additional unit. This plant has an 18.0 MW
maximum capability and a 14.8 MW nameplate rating.
Upper Falls
The Upper Falls project began generating in 1922 in
downtown Spokane. This project is comprised of a
single unit with a 10.2 MW maximum capability and
10.0 MW nameplate rating.
Monroe Street
The Monroe Street plant was the company s first
generating unit. It started service in 1890 near what
is now Riverfront Park. Rebuilt in 1992, the single
generating unit now has a 15.0 MW maximum capability
and a 14.8 MW nameplate rating.
2 -2007 Electric IRP Avista Corp
Chapter 2- Loads and Resources
Nine Mile
The Nine Mile project was built by a private developer
in 1908 near Nine Mile Falls, Washington. The company
purchased it in 1925 from the Spokane & Eastern
Railway. Its four units have a 24.4 MW maximum
capability and a 26.4 MW nameplate rating.
Long Lake
The Long Lake project is located above Little Falls
in Eastern Washington. It was the highest spillway
dam with the largest turbines in the world when it
was completed in 1915. The plant was most recently
upgraded with new runners in 1999. The four units in
this project provide 90.4 MW in combined maximum
capability and 70.0 MW nameplate rating.
Little Falls
The Little Falls project was completed in 1910 near Ford
Washington. The four units at this project provide 36.
MW of maximum capability and have a 32.0 MW
nameplate rating.
,;i
CLARK FORK RIVER HYDROELECTRIC PROJECT
The Clark Fork River Project is comprised of
hydroelectric projects in Clark Fork, Idaho, and Noxon
Montana. The plants operate under a FERC license
expiring in 2046.
Cabinet Gorge
The Cabinet Gorge plant started generating power in
1952 with two units. The plant was expanded with
two additional generators in the following year. The
current maximum capability of the plant is 263.2 MW;
it has a nameplate rating of272.2 MW Upgrades at
this project began with the replacement of turbine Unit
1 in 1994. Unit 3 was upgraded in 2001. Unit 2 was
upgraded in 2004. The final unit, Unit 4, received a $6
million turbine upgrade in 2007, increasing its generating
capacity from 55 MW to 64 MW and adding 2.1 aMW
of energy.
Noxon Rapids
The Noxon Rapids project includes four generators
Monroe Street Hydroelectric Facility, Spokane , Washington
Avista Corp 2 -2007 Eiectric IRP
Chapter 2- Loads and Resources
a. ,e'o. m. a. n - wne. I . roo e- SO, uree- s
Project Nameplate Maximum 70-Year
Project River Start Capacity Capability Energy
Name System Location Date (MW)(MW)(aMW)
Monroe Street Spokane Spokane, WA 1890 14.15.13.
Post Falls Spokane Post Falls, 10 1906 14.18.
Nine Mile Spokane Nine Mile Falls, WA 1925 26.4 24.4 16.4
Little Falls Spokane Ford, WA 1910 32.36.22.
LonQ Lake Spokane Ford, WA 1915 70.90.4 52.4
Upper Falls Spokane Spokane, WA 1922 10.10.
Cabinet GorQe Clark Fork Clark Fork, 10 1952 272.263.122.
Noxon Rapids Clark Fork Noxon, MT 1959 466.527.202.
Total All Hydro 905.984.442.
a. I eo o.mla-n - wne.rma.e. so. urea. s
Nameplate Maximum Energy
Project Start Capacity Capability Capability
Name Location Fuel Date (MW)(MW)(aMW)
Colstric 3(15o/J Colstrip, MT Coal 1984 116.114.93.
Colstrip 4 (15%)Colstrip, MT Coal 1986 116.114.93.
Rathdrum Rathdrum, 10 Gas 1995 166.176.135.
Northeast Spokane, WA Gas/Oil 1978 62.66.
Boulder Park Spokane, WA Gas 2002 24.24.23.
Coyote SprinQs 2 Boardman, OR Gas 2003 287.284.250.
Kettle Falls Kettle Falls, WA Wood 1983 46.50.42.
Kettle Falls CT Kettle Falls, WA Gas 2002
Total All Thermal 827.838.653.
T bl 2 2 C
T bl 2 3 C
installed between 1959 and 1960, and a fIfth unit added
in 1977. The current plant configuration has a maximum
capability of 527.0 MW and a nameplate rating of 466.
MW Upgrades to all four units at the Noxon Rapids
facility are scheduled from March 2009 to March 2012.
The upgrades are expected to add 38 MW of capacity
and 6 aMW of energy to the company's resource
portfolio.
Total Hydroelectric Generation
In total, our hydroelectric plants are capable of
generating as much as 984.2 MW Table 2.2 summarizes
the company s hydro projects. This table also includes the
average annual energy output of each facility based on
the 70-year stream flow record.
THERMAL RESOURCES
Avista owns and maintains several thermal assets located
across the Northwest. Each thermal plant is expected to
d H d
dTh
continue to be available through the 20-year duration
of the 2007 IRP. The company s thermal resources
provide dependable low-cost energy to serve base loads
and provide peak load serving capabilities. Table 2.
summarizes the company s thermal projects.
Colstrip
The Colstrip plant, located in Eastern Montana
consists of four coal-fired steam plants owned by a group
of utilities. PPL Global operates the facilities. Avista
owns 15 percent of Units 3 and 4. Unit 3 was completed
in 1984 and Unit 4 was finished in 1986. The company
share of each Colstrip unit has a maximum capability
of 114.6 MW and a nameplate rating of 116.7 MW
Capital improvements to both units were completed in
2006 and 2007 to improve efficiency and reliability and
to increase generation. The upgrades included new high-
pressure steam turbine rotors and conversion from analog
to digital control systems. These capital improvements
2 -2007 Electric IRP Avista Corp
Chapter 2- Loads and Resources
increased the company s share of generation by 4.2 MW
at each unit without any additional fuel consumption.
Rathdrum
Rathdrum is a two-unit, simple-cycle, gas-fired plant
located near Rathdrum, Idaho. The plant entered service
in 1995. It has a maximum capability of 176.0 MW and
a nameplate rating of 166.5 MW
Northeast
The Northeast plant, located in northeast Spokane, is a
two-unit, aero-derivative, simple-cycle plant completed
in 1978. The plant is capable of burning natural gas or
fuel oil, but current air permits prevent the use of fuel oil.
The combined maximum capability of the units is 66.
MW with a nameplate rating of 62.8 MW
Boulder Park
The Boulder Park project was completed in Spokane
Valley in 2002. The site uses six natural gas-fired internal
combustion engines to produce a combined maximum
capability and nameplate rating of24.6 MW
Coyote Springs
Coyote Springs 2 is a natural gas-fired combined cycle
combustion turbine located near Boardman, Oregon.
The plant began service in 2003. The maximum
capability is 264.3 MW and the duct burner provides
the unit with an additional capability of up to 20.4 MW
The nameplate rating is 287.0 MW
Kettle Falls
The Kettle Falls biomass facility was completed in 1983
near Kettle Falls, Washington. The open-loop biomass
steam plant is fueled by waste wood products and has a
maximum capability of50.7 MW Its nameplate rating is
46MW
Kettle Falls CT
The Kettle Falls CT is a natural gas-fired combustion
turbine that began service in 2002. It has a maximum
capability rating of 6.9 MW Exhaust heat from the plant
is routed into the Kettle Falls biomass plant boiler to
increase its efficiency. The plant is capable of running
independently of the biomass steam plant.
POWER PURCHASE AND SALE CONTRACTS
The company utilizes several power supply purchase
and sale arrangements of varying lengths to meet a
portion of its load requirements. This section describes
the contracts in effect during the scope of the 2007
IRP. The contracts provide a number of benefits to the
company, including environmentally low-impact and
low-cost hydro and wind power. An annual summary of
our contracts is contained in Table 2.
Bonneville Power Administration (BPA) Residential
Exchange
The company fIrst entered into settlement agreements
to resolve BPA's Residential Exchange obligation on
October 31 , 2000. Over the first five years of the 10-
year settlement, the company received financial benefits
equivalent to purchasing 90 aMW at BPA's lowest cost-
based rate. The company s benefit level increased to 149
Coyote Springs 2, Boardman, Oregon
Avista Corp 2 -2007 Electric IRP
Chapter 2- Loads and Resources
aMW on October 1 2006. At BPA's option, the 149
aMW may be provided in whole or in part as financial
benefits or as a physical power sale; the IRP assumes the
former based on regional discussions.
On May 3, 2007, the Ninth u.S. Circuit Court of
Appeals issued opinions holding that BPA exceeded its
settlement authority and acted in a manner that was
inconsistent with the Northwest Power Act when it
entered into the settlement agreements. As a result, on
May 21 , 2007 BPA notified Avista that it was suspending
payments.
Bonneville Power Administration WNP-3 Settlement
On September 17, 1985, the company signed settlement
agreements with BPA and Energy Northwest (formerly
the Washington Public Power Supply System or WPPSS),
ending construction delay claims against both parties.
The settlement provides an energy exchange through
June 30 2019, with an agreement to reimburse the
company for certain WPPSS - Washington Nuclear Plant
No.3 (WNP-3) preservation costs and an irrevocable
offer ofWNP-3 capability for acquisition under the
Regional Power Act.
The energy exchange portion of the settlement
contains two basic provisions. The first provision
provides approximately 42 aMW of energy to the
company from BPA through 2019, subject to a contract
minimum of 5.8 million megawatt-hours. The company
is obligated to pay BPA operating and maintenance costs
associated with the energy exchange as determined by a
formula that ranges from $16 to $29 per megawatt-hour
in 1987 dollars.
The second provision provides BPA approximately 33
aMW of return energy at a cost equal to the actual
operating cost of the company s highest-cost resource.
A further discussion of this obligation, and how the
company plans to account for it, is covered under the
Confidence Interval Planning heading of this chapter of
the IRP.
Mid-Columbia Hydroelectric Contracts
During the 1950s and 1960s, various public utility
districts (PUDs) in central Washington developed
hydroelectric projects on the Columbia River. Each
plant was large compared to the loads served by the
PUDs. Long-term contracts were signed with public
municipal and investor-owned utilities throughout the
Northwest to assist with project fmancing and to ensure
a market for the surplus power.
The company entered into long-term contracts for the
output of four of these projects "at cost."The contracts
provide energy, capacity and reserve capabilities. In 2008
they will provide approximately 95 MW of capacity
and 51 aMW of energy. Over the next 20 years, the
Wells and Rocky Reach contracts will expire. While the
company may be able to extend these contracts, it has
no assurance today that extensions will be offered. The
2007 IRP does not include energy or capacity for these
contracts beyond their expiration dates.
The company renewed its contract with Grant PUD
in 2005 for power from the Priest Rapids project. The
contract term will equal the term in the forthcoming
Priest Rapids and Wanapum dam FERC licenses. A
license term of 30 to 50 years is expected. The company
acquired additional displacement power in the Priest
Rapids settlement. Displacement power, through
September 30, 2011 , includes project output available
due to displacement resources being used to serve Grant
PUD's load. A summary of Mid-Columbia contracts is
included in Table 2.4.
Medium- Term Market Purchases
Avista has power purchase contracts for 100 MW
power from 2004 through 2010 from several suppliers.
2 -2007 Electric IRP Avista Corp
Chapter 2- Loads and Resources
I' - o. um. la,o.n ra.umma.
2008 2012 2017
Project Name aMW aMW aMW
Rocky Reach 37.20.
Wells 28.15.28.15.28.15.
Grant County 28.14.63.35.63.32.
Totals 95.50.92.52.92.48,
..
. n ra. c ua.sa.. 10 a. 10.
Capacity Energy
Contract Name Start Date (MW)(aMW)End Date
Grant County Purchase 2005 129.72.TBD
Rockv Reach Purchase 1961 37.19.Oct-2001
Wells Purchase 1967 28.Aua-2018
PGE Capacity Sale 1992 150.Dec-2016
Upriver Dam Purchase 1966 14.10.Dec-2011
WNP-3 Purchase & Sale 1987 82.48.Jun-2019
Medium-Term Purchases 2004 100.100.Dec-2010
PPM Wind Purchase 2004 35.Mar-2011
Total Contract 577,268,
Table 2 4 Mod C I
Nichols Pumping Station
The company provides energy to operate its share of
the Nichols Pumping Station, which supplies water for
the Colstrip plant. The companys share of the Nichols
Pumping Station load is approximately one aMW Avista
is also under contract to provide pumping energy to
other Colstrip owners.
Portland General Electric Firm Capacity Sale
The company contracted to provide Portland General
Electric (PGE) with 150 MW of firm capacity through
December 31 2016. PGE may schedule deliveries up to
its capacity limit during any 10 hours of each weekday.
Within 168 hours PGE returns energy delivered under
the contract.
Stateline Wind Energy Center
The company contracted with PPM Energy in 2004
for 35 MW of nameplate wind capacity from the
Stateline Wind Energy Center located on the Oregon-
Washington border. This 35 MW contract does not
include firming services.
Table 2 5' Si nifica t Co t
1 The PPM wind purchase is shown at its nameplate rating.
Avista Corp
b' C
A summary of all company obligations and rights is
presented in Table 2.
RESERVE MARGINS
Planning reserves accommodate situations when loads
exceed and/or resources are below expectations because
of adverse weather, forced outages, poor water conditions
or other contingencies. There are disagreements within
the industry on adequate reserve margin levels. Many
stem from system differences, such as resource mix
system size and transmission interconnections. For
example, a hydro-based utility generally has a higher
capacity-to-energy ratio than a thermal-based utility.
Reserve margins, on average, increase customer rates
when compared to resource portfolios without reserves.
For example, inexpensive 100 MW peaking resources
overnight costs are around $42 million; this translates
to a $6 million annual expense. Reserve resources have
the physical capability to generate electricity, but high
operating costs limit economic dispatch and the potential
to create revenues to offset capital costs. Some argue
I R' ht dab!'
2007 Electric IRP 2 - 17
Chapter 2- Loads and Resources
that regions with deregulation, or " customer choice
provide strong incentives for industry participants to
underestimate their reserve obligations and lower their
costs at the expense of system reliability.
AVISTA'S PLANNING MARGIN
Avista s planning reserves are not directly based on unit
size or resource type. Planning reserves are set at a level
equal to 10 percent of our one-hour system peak load
plus 90 MW The 90 MW accounts for approximately
60 MW of hydro because of icing on river banks and
30 MW of Colstrip reserves because of coal handling
problems in cold weather situations. This amounts to
roughly a 15 percent planning reserve margin during the
company s peak load hour.
CONFIDENCE INTERVAL PLANNING
Avista uses confidence interval planning to ensure
it has resources adequate to meet customer energy
requirements. Extreme weather conditions can affect
monthly energy obligations by up to 30 percent.
the company lacks generation capability to meet high
load variations, it is exposed to increased short term
market volatility. Analysis of historical data indicates
that an optimal criterion is the use of a 90 percent
confidence interval based on the monthly variability
ofload and hydroelectric generation. This results in
a 10 percent chance of the combined load and hydro
variability exceeding the planning criteria for each
month. In other words, there is a 10 percent chance
that the company would need to purchase energy from
the market in any given month. Avista has considered
Thermals
Hydro
Contracts
larger confidence intervals, but analysis suggests that
the cost of additional resources to cover higher levels of
variability would exceed the potential benefits. Building
to the 99 percent confidence interval could significantly
decrease the frequency of market purchases but would
require approximately 200 MW of additional generation
capability. Additional capital expenditures to support this
level of reliability would put upward pressure on retail
rates.
The 90 percent confidence level varies between 84
aMW and 301 aMW on a monthly basis in 2008, or 166
aMW across the 12-month period. This level is similar
to critical water planning on an annual basis, but is more
precise because it is based on the monthly instead
annual chance of exceedance.
Additional variability is inherent in the WNP-
contract with BPA. The contract includes a return
energy provision that can equal 33 aMW annually. The
contract would be exercised under adverse conditions
such as low hydroelectric generation or high loads, which
the company would also expect to be experiencing.
Requirements under the confidence interval are
increased by 33 aMW to account for the WNP-
obligation through its expiration in 2019.
SUSTAINED PEAKING CAPACITY
Parallel to planning margins is the "gray area" between
energy and capacity planning termed sustained
peaking capacity. Sustained peaking capacity is a
tabulation ofloads and resources over a period exceeding
One Hour to Three Da s or More
Highest Load on Record
2 -Avista Corp2007 Electric IRP
Chapter 2- Loads and Resources
the traditional one-hour definition. It is also a measure
of reliability and recognizes that peak loads do not
stress the system for just one hour. Table 2.6 details the
assumption differences between the company s planning
approach and the sustained capacity approach.
The company has actively participated in the Northwest
Power and Conservation Council's Resource Adequacy
committees over the past few years. Preliminary work
indicates that the Northwest should carry approximately
a 25 percent planning margin in the wintertime and a 17
percent planning margin in the summertime. These
levels are much higher than the 12 to 15 percent
levels recommended in California or for other markets
primarily due to the Northwest's heavier reliance on
hydroelectric generation. Given the various uncertainties
surrounding these higher planning margin levels, and
the fact that they are not yet finalized, the company
plan will not change for this planning cycle. Avista will
continue to participate in this important regional process
and use the results in its future planning when they
become more finalized.
RESOURCE REQUIREMENTS
The differences between loads and resources illustrate
potential needs the company must address through its
future resource acquisition actions. The company plans
to meet both its energy and capacity needs.
CAPACITY TABULATION
The company regularly develops a 20-year service
territory forecast of peak capacity loads and resources.
Peak load is the maximum one-hour obligation
including operating reserves, on the expected average
coldest day in January. Peak resource capability is the
maximum one hour generation capability of company
resources, including net contract contribution, at the
time of the one-hour system peak. This calculation is
performed to ensure that the company has sufficient
resources to meet its load obligations. Avista has surplus
capacity through 2009 without the addition of the
Lancaster plant. Capacity deficits begin in 2010, with
loads exceeding resource capabilities by five MW The
deficits continue to grow as peaking requirements
703 763 815 868 909 019 103 214 2,492
260 266 272 277 281 292 300 311 339
964 029 087 145 190 311 404 525 831
142 154 121 128 084 098 098 070 070
172 172 173 208 128 128
230 230 230 230 230 230 230 230 230
308 308 308 308 308 308 308 308 308
211 211 211 211 211 211 211 211 211
111 123 092 999 939 954 104 996 996
148 146 251 357 300 530 835
24,20.15.19,
275 275 275 275 275 275
148 280 129 255 835
24.20.30,21.16.10.13,19.
Avista Corp 2007 Electric IRP 2 -
Chapter 2- Loads and Resources
000
Figure 2.16: Capacity Loads and Resources (MW)
500
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
000
500
000
500
Hydro_Coal
E:::J Gas Dispatch
I!!!!!iiiI Gas Peaking Units
Net Contracts
Biomass
c:::J Lancaster
Load and Operating Reser\eS
............~'" ......
C')-.t II)
......
r--
Obli ations
Retail Load 125 163 196 230 256 326 379 1 ,450 627
90% Confidence Interval 200 199 196 196 192 192 192 156 156
Total Obli ations 324 362 392 425 448 518 571 606 783
Existin Resources
H dro 540 538 531 528 512 510 509 491 491
Net Contracts 234 234 234 129 107 105 105 106 106
Coal 199 183 188 198 187 187 198 199 186
Biomass
Gas Dis atch 280 295 285 295 280 295 295 280 295
Gas Peakin Units 145 145 141 146 145 146 145 141 145
Total Existing
Resources 446 442 426 342 278 290 299 265 270
Net Positions 121 170 228 272 341 513
Lancaster 254 264 249 264 264 228
Net Positions with
Lancaster 121 288 181 114 513
increase with load growth, and the company s resource
base declines due to the expiration of market purchases
and reductions in power from Mid-Columbia
hydroelectric project contracts. Some year-to-year
variation occurs in the forecast because of maintenance
schedules. With Lancaster included in the planning, our
deficit year moves out to 2014. Table 2.7 summarizes the
forecast.
Avista currently has sufficient capacity resources
primarily because of the relatively large amount of
hydroelectric generation in its resource portfolio.
Hydroelectric resources can provide large amounts
of short-term capacity in relation to the energy they
produce because of storage associated with each project.
Future capacity requirements will be addressed by
acquiring new resources that provide both energy and
capacity, or in the case of intermittent resources like
wind, other resources that provide capacity. Figure 2.
shows this information graphically.
2 - 20 Avista Corp2007 Electric IRP
Chapter 2- Loads and Resources
Figure 2,17: Energy Loads and Resources (aMW)
800
600
400
200
000
800
600
400
200
Hydro_Coal
c::::J Gas Dispatch
I!!!!'IIiiiiI Gas Peaking Units
......
ENERGY TABULATION
Table 2.8 summarizes annual energy loads and resources
for the IRP time horizon. This IRP focuses on meeting
the company s energy requirements to the 90 percent
confidence level. Similar to Table 2., maintenance
schedules affect the output of plants over the IRP
timeframe. Specifically, coal, biomass, gas dispatch and
gas peaking units are affected.
After 2010 new resources are necessary to continue
meeting the 90 percent confidence interval planning
margin criterion. The table shows that the company is
annually in a surplus position through 2010. With the
Lancaster plant, our surplus position moves out to 2016.
Figure 2.17 provides the same information graphically.
Conservation acquisitions are prescriptive, meaning that
customers must take action to lower their energy usage.
Without "programmatic" conservation acquisitions, retail
loads and supply-side resource acquisitions would be
higher. Historically, conservation acquisition levels were
Net Contracts
Biomass
r:=::J Lancaster--Load and Operating Resel\eS
1'0-C')
"'"
included as reductions to retail load. The 2005 IRP
included load that will be met by programmatic
conservation, as an increase to load, and then displays
the conservation resource separately in the table. The
conservation projections shown in Tables 2.7 and 2.8 are
cumulative and illustrate the companys commitment
to continued acquisition of cost-effective conservation.
Activities beyond current levels are discussed in Chapter
3 - Demand Side Management - and are shown as new
resources in later tabulations.
The company expects to experience energy deficits
during some months of all forecast years. As an example
the company anticipates deficits in January and October
of2008 even though the annual position has a 121
aMW surplus. Surplus positions occur in the remaining
months, particularly during spring runoff. The company
balances its montWy positions through short-term
market purchases or sales, exchanges, or other resource
arrangements.
Avista Corp 2 - 212007 Eiectric iRP
Chapter 3- Demand Side Management
DEMAND SIDE MANAGEMENT
A High Efficiency Compact Flourescent Light Bulb
INTRODUCTION
Avista s Demand Side Management (DSM) programs
provide a range of energy efficiency options for
residential, commercial and industrial customers.
They fall into prescriptive and site-specific categories.
Prescriptive programs offer cash incentives for
standardized products such as compact fluorescent
light bulbs and high efficiency appliances. Site-specific
programs provide cash incentives for cost-effective
energy savings measures with a payback greater than
one year. These programs are customized services for
commercial and industrial customers because many
applications need to be tailored to customer premises
and processes. Avista has continuously offered electric
efficiency programs since 1978. Some of Avista s most
notable efficiency achievements include the Energy
Exchanger programs, which converted over 20 000
homes from electric to natural gas for space or water
Figure 3,1: Historical Conservation Acquisition
120180
150
- - - - - - - - - - - - - - - - - - - - - - - - - - - - -
100
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
~ 120
:;;
I'CIrn 60
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
iii::I
------
r--(J)(J)
"tt
(J)(J)(J)(J)(J)(J)
SECTION HIGHLIGHTS
:!!
1:11
:;;
I'CI
CII
::I
::I
(.)----------------
(J)(J)
"tt(J)(J)
"tt(J)(J)(J)(J)
. Avista has assisted its customers in acquiring cost -effective energy efficiency for 30 years.
. Avista has acquired 124 aMW of electric-efficiency in the past three decades; an estimated 96 aMW is currently
online.
. 20 000 customers heat their homes with natural gas today because of the company s fuel-switching programs.
. The company has developed and will maintain the infrastructure necessary to respond quickly in the event another
energy crisis occurs.
. The Heritage Project is re-evaluating our traditional programs, updating economic benchmarks, and revising the
scope to include transmission, distribution and generation facility efficiencies.
Avista Corp 2007 Electric iRP 3 -
Chapter 3- Demand Side Management
heating from 1992-1994; pioneering the country s first
system benefit charge for energy efficiency in 1995;
and the immediate conservation response during the
2001 Western energy crisis, which tripled annual energy
savings at only twice the cost, in half the time, to meet
the customer demand for reducing energy usage during a
period of high prices. The company s programs provide
savings that regularly meet or exceed its regional share
energy efficiency savings as outlined by the Northwest
Power Planning and Conservation Council. Historical
electricity conservation acquisition is illustrated in
Figure 3.
During the 30 years that Avista has actively acquired
electric efficiency resources, a total of 124 aMW of
energy savings has been achieved. We believe that the 96
aMW acquired during the last 18 years is still online and
yielding resource value today.
In this IRP planning cycle, all demand-side management
(DSM) measures and programs have been examined
based on surrogate generation costs. New savings targets
have been established, and the company is planning a
significant ramp-up of energy efficiency activity.
Avista is also expanding the breadth of its efficiency
activities to include demand response initiatives and is
revisiting the potential for transmission and distribution
efficiency measures. These expanded programs are in
development and are not reflected in this IRP, but they
are included as an action item for the 2009 IRP.
THE HERITAGE PROJECT
The companys new demand response initiative is called
the Heritage Project. The Heritage Project focuses on
revamping existing energy efficiency targets by applying
the best practices within the utility industry. This
project continues our legacy of innovation in energy
efficiency efforts and customer education. The goal of
the Heritage Project is to increase the acquisition of
sustainable and cost-effective energy and demand savings
through a comprehensive, state-of-the-art demand
response initiative. The project examines and implements
expanded energy efficiency programs, peak shaving/
shifting programs and other options (e., distribution
system efficiencies).
The Heritage Project focuses on five areas: energy
efficiency, load management, transmission and
distribution efficiencies, analytics and communications.
Each area is supported by analyses and attributes unique
to that function.
1 Cumulative conservation is based upon an 1S-year weighted average measure life.2 NEEA's website, www.nwalliance.org, offers additional details regarding their ventures, governance, proceedings, reports and evaluations.3 It was assumed that historic acquisition would remain flat at the most recent level because there are no reliable 20-year estimates of regional
program acquisition. This assumption is speculative and dependent on the opportunities for regional market transformation during this
period, but it is consistent with the recent history of flat funding of the NEEA organization.
3 -2007 Electric IRP Avista Corp
Chapter 3- Demand Side Management
. -
...se-ne-Icle-roo . ra. m
Start Time Residential and Small Commercial/Industrial/
Commercial/Industrial Institutional
Q1 2007 Fireplace Dampers C&I Quick Hits Program
Q2 2007 Super Efficient Habitat for Humanity Side-Stream Filtration
(HFH) Homes Energy/Heat Recovery Ventilation
Something For Everyone Measures (ERV/HRV)
Demand Control Ventilation (DCV)
Steam Traps
Q3 2007 Geographic Saturation Program Retro-Commissioning Program
Behavioral Proqram
Q4 2007 Regional Natural Gas Market Facilities Model Program (ongoing)
Transformation Program
ENERGY EFFICIENCY
The energy efficiency review evaluated the company
current electric and natural gas efficiency programs
to determine what additional programs can be cost-
effectively acquired in the near-term (2007) and
intermediate term (2008-2010). Avoided costs based on
the 2007 IRp, including factors such as risk and capacity,
were established to determine the cost-effectiveness of
and potential for program expansion. Current delivery
mechanisms and outreach efforts were assessed to
ensure that all customers have knowledge and adequate
opportunities to participate in the company s efficiency
programs. Table 3.1 summarizes the DSM programs.
The company s existing effIciency programs are
thorough, but several additional opportunities were
identified. New programs that are currently under
evaluation are outlined in Table 3.
REQUEST FOR INFORMATION/REQUEST FOR PROPOSALS
In addition to soliciting internal parties and key
stakeholders for concepts to improve the energy
efficiency portfolio, the company also released a broad
request for information (RFI) in 2006 to obtain the
benefit of the opinions outside of our normal range
of contacts. The RFI sought ideas for the company
to cost-effectively enhance its conservation portfolio
through new programs, measures or revisions to existing
Table 3 2' Pro 0
programs. A total of 53 RFI responses were received. An
evaluation of these responses led to two recently released
requests for proposals (RFPs) for electric and natural gas
efficiency programs within the commercial refrigeration
and the residential multi-family housing markets. Four
proposals have been received in response to each of these
RFPs, and the bids are being evaluated.
LOAD MANAGEMENT
Going forward, peak prices are expected to be
significantly higher than prevailing average market prices.
For example, the current AURORAxmp model forecast
shows average highest day prices between two and three
times higher ($80 to $100 per MWh) than average
day prices. In addition, the highest prices will be an
additional two to three times the average of those prices.
This is consistent with recent events in the summer of
2006 where market prices exceeded $200 per MWh.
The company does not anticipate that the summer 2006
event will repeat itself frequently, but it remains to be
seen whether this was an anomaly or an event that will
occur every few years.
With higher peak day prices and additional volatility
likely during super critical peak events, demand
reduction (DR) measures and distributed generation
(DG) has the potential to mitigate cost impacts to
customers and utilities.
Eff '
4 Due to the accelerated nature of the Heritage Project and the simultaneous IRP evaluation, it was not possible to incorporate all of these
measures within the current DSM targets without causing an unnecessary delay in their developement and launch.
Avista Corp 2007 Electric IRP 3 - 3
Chapter 3- Demand Side Management
Load management opportunities are identified that
could be implemented in the near-term (2007) and
the medium term (2008-2010). As with the energy
efficiency examination, an inventory of all potential
load management programs and offerings. The analysis
included a review of trade ally data, industry literature
vendor research and a consultant evaluation. The cost of
new technologies that enable more precise measurement
and control of energy is declining. In order to expedite
implementation of these candidate programs the analysis
was often performed concurrently with the IRP
evaluation, so it was not possible to fully quantify the
impacts of these programs within this IRP cycle. This
quantification has been identified as an action item for
the 2009 IRP.
Five projects, outlined below, have been identified
for immediate implementation with a fiamework
established for future activities. This framework evaluates
infrastructure needs, system and hardware requirements
costs and benefits, and customer acceptance
Residential Demand Response Pilot - This pilot
includes the installation of smart communicating
thermostats at specified locations.
Small Commercial Demand Response Pilot - This
pilot project includes the installation of wireless
dimmable ballasts and/or other technologies in small
commercial premises.
Large Commercial/Industrial Interruptibility
Agreements with larger commerciallindustrial customers
to curtail load during specific events have been successful.
This project would expand and formalize the process
to include prearranged structured agreements. These
agreements could be handled on a buy-back basis in the
near-term and on interruptible rate schedules over the
long-term.
Avista Facilities Demonstration Project - Avista will
test wireless dimming ballasts and other technologies in
our own facilities. Other demand response options will
be considered and tested, as appropriate.
Large Commercial/Industrial Distributed
Generation - In addition to bilateral agreements for
curtailment, the company is examining a distributed
generation program with selected customers in return for
utility-controlled dispatchability.
TRANSMISSION AND DISTRIBUTION
System losses-or lost energy in the form of heat-
naturally occur on utility systems in two ways: first, as the
power is moved over distances and second, by transfers
of electricity through distribution equipment as the
power is "stepped-down" from high-voltage to end-user
voltages. The company s system losses are estimated to be
between 6 percent and 8 percent. Advances in efficient
equipment such as improved transformer technology
may yield system improvements. Design processes, such
as conservation voltage reduction (CVR) and substation
engineering and siting, can also provide energy savings
on the distribution system.
The company s Transmission and Distribution (T&D)
Planning group is examining different ways to
economically reduce system losses. The quantification of
T &D losses and potential loss reductions is in progress.
The cost/benefit relationship will be assessed after the
quantification process has been completed. Several
projects are underway and pilots are under consideration.
Significant time will be required to fully evaluate
the results of the near-term potential projects and to
ascertain potential resource opportunities. It is premature
to incorporate these efforts into the IRP targets, so they
have been identified as an action item for the 2009 IRP.
3 - 4 Avista Corp2007 Electric iRP
Chapter 3- Demand Side Management
ANALYTICS
The identification of the cost-effectiveness of alternative
supply resources and appropriate cost-recovery depends
upon an analytical approach that is technically sound
and transparent. Several departments collaboratively
developed an analytical process to determine overall
resource values of energy and capacity. Resource
valuation for the Heritage Project is based upon seven
categories: five categories are reflected in a total avoided
cost of energy usage and the other two are based upon
system-coincident demand reductions.
Analytical values contributing to an overall resource
value of energy include the avoided cost of energy
and carbon emissions, reduced volatility, reduced
transmission and distribution system losses. Analytical
values contributing to overall avoided costs of system-
coincident capacity include the value of deferring
capital investments for generation and transmission and
distribution. A summary of these calculations has been
provided in the Appendices.
COMMUNICATIONS PLANNING
Communicating the availability of conservation programs
is critical to achieving energy savings. The Heritage
Project is developing a sustained outreach campaign.
This plan is staged for new program roll-outs and is
tailored to select the optimal tool for communicating
each program. This focus includes communications to
all Avista employees, as well as enhanced training for
employees with customer contact.
COOPERATIVE REGIONAL MARKET TRANSFORMATION
PROGRAMS
Avista is a funding and fully participating member
the Northwest Energy Efficiency Alliance (NEEA).
NEEA is funded by investor-owned and public utilities
throughout the Northwest to acquire electric efficiency
measures that are best achieved through market
transformation efforts. These efforts reach beyond
individual service territories and consequently require
regional cooperation to succeed.
NEEA has proven to be a cost-effective component
of regional resource acquisition. Avista has and will
continue to leverage NEEA ventures when cost-effective
enhancements to the programs can be achieved for our
customers.
Attributing regionally acquired resources to individual
utilities is difficult. In order to ensure that resources are
not double-counted at both regional and local levels
NEEA has excluded from their claims all energy for
which local utility rebates have been granted. Therefore
it is correct to sum the local and regional acquisition
to obtain the total impact within the effected markets.
Avista has typically applied our funding share of slightly
less than 4 percent to NEEA's annual claim of energy
saVIngs.
DSM PROGRAM FUNDING
As previously noted, in 1995 the company changed its
approach to cost-recovery ofDSM investments from
the traditional capitalization of the investments to
cost-recovery through a non-bypassable public benefits
surcharge (the DSM tariff rider). The company currently
manages four separate DSM tariff riders for Washington
electric, Idaho electric, Washington natural gas and
Idaho natural gas investments. Based upon the demand
for funds and incoming DSM tariff rider revenues, this
balance can be positive or negative at any particular point
in time.
In 2005 the aggregate DSM tariff rider balance was
returned to zero from a $12.4 million deficit in the
aftermath of the 2001 Western energy crisis. Recent
demand for DSM services has outstripped the incoming
DSM tariff rider revenue. The most recent projection
5 NEEA's website, www.nwalliance.org, offers additional details regarding their ventures, governance, proceedings, reports and evaluations.
Avista Corp 2007 Electric iRP 3 - 5
Chapter 3- Demand Side Management
forecasts a $3.8 million negative balance in the
Washington electric DSM tariff rider at the close of
2007. The Idaho electric DSM balance is projected to be
close to zero at that time.
The company has proposed the capitalization of electric
DSM investments in Washington. The proposal would
continue the current tariff rider mechanism, with the
revenues generated from the tariff rider funding the
revenue requirement of the DSM investments.
Additionally there is a proposal for the recovery oflost
electric margin (or fIXed cost recovery) associated with
the company s DSM efforts. Both of these proposals have
been advanced to provide a more level playing field for
demand and supply-side resource investments.
At present the company is not compensated for the fIXed
costs associated with reductions in load resulting from
electric DSM achievements. The company submitted a
proposal to the Washington Utilities and Transportation
Commission for fIXed cost recovery between general rate
cases.
OVERVIEW OF ELECTRIC-EFFICIENCY IN
THE 2007 IRP
The implementation of the Heritage Project began in
the midst of the 2007 IRP evaluation. Some, but not all
of the Heritage Project initiatives have been incorporated
in this version. The 2009 IRP cycle will fully explore
some of the details and resulting efforts.
CONSISTENCY BETWEEN THE IRP EVALUATION AND
DSM OPERATIONS
For each IRP, the company evaluates energy-efficiency
potential in a manner that can augment the conservation
business planning process and ultimately lead to
appropriate revisions in DSM acquisition operations.
Avista has utilized the IRP process as an opportunity
comprehensively re-evaluate the market. This assessment
evaluates individual technologies (generally prescriptive
programs) where possible and program potential when a
technology approach is infeasible. The evaluation is based
upon an assessment of resource characteristics and the
construction of a conservation supply curve based upon
the levelized total resource cost (TRC) and acquirable
resource potential for each technology. Cost-effective
technologies, compared to the defined avoided cost, are
incorporated into the IRP acquisition target.
The program evaluation is necessary when technologies
in the program cannot be defined to permit their
individual evaluation. This is the case in the company
comprehensive limited income and non-residential
programs.6 The target acquisition for these programs is
based upon modifying the historical baseline for known
or likely changes in the market. This includes but is not
necessarily limited to modifying the baseline for price
elasticity and load growth.
EVALUATION OF EFFICIENCY TECHNOLOGY
OPPORTUNITIES
Avista initiated an internal review of the company
response to the July 24 2006, heat wave and short-term
escalation of regional wholesale electric prices. An
exploration of possible future responses to short-term
price spikes and other longer term approaches to reduce
the impact of market volatility was a key component of
that process. Approximately 140 concepts came out of
a series of meetings attended by a cross-section of the
company.
6 It was assumed that historic acquisition would remain flat at the most recent level because there are no reliable 20-year estimates of
regional program acquisition. This assumption is speculative and dependent on the opportunities for regional market transformation during
this period, but is consistent with the recent history of flat funding of the NEEA organization.
7 The portions of the non-residential market that could be identified and evaluated based upon technology applications were included in
that portion of the study. These components were excluded horn the historical baseline for the remaining non-residential technologies
evaluated under programmatically.
2007 Electric IRP Avista Corp
Chapter 3- Demand Side Management
Avista s DSM analysis staff and Navigant Consulting
performed a six-stage review of this concept list. The
process first evaluated concepts with easily obtained data
and gradually moved toward the more difficult analyses.
Some measures did not rank well enough to warrant
further consideration. The individual phases of the
analytical process follow:
Defining: Refinement and redefinition of the
concept list eliminated duplicative concepts and allowed
an opportunity to develop common definitions for each
concept.
Qualitative ranking:The more clearly defined concepts
from the prior phase were ranked on a qualitative
assessment of feasibility. Opportunities which were
clearly not acquirable by utility intervention were
eliminated from further consideration.
Defining cost characteristics:Those concepts that
were determined to have a reasonable potential for
eventual incorporation into the conservation portfolio
were evaluated on preliminary assessments of cost-
effectiveness. This step required obtaining estimates
incremental customer cost, non-energy benefits, energy
savings and measure life to develop a TRC levelized
cost. Concepts were sorted based upon these cost
characteristics.
Defining resource potential: Acquirable potentials
specific to the Avista service territory were estimated for
the remaining concepts. These acquirable potentials were
the result of an assessment of technical and economic
potential tempered by the realization that utility
intervention cannot successfully address all customer
adoption barriers regardless of the economics. The
acquirable resource potential for some technologies has
been modified, generally upward, as a result of Heritage
Project.
Developing load profiles:This IRP evaluation
the fIrst time that Avista has specifically incorporated
the value of capacity contribution (transmission
distribution and generation) into the overall avoided cost.
Additionally the company is basing the avoided cost of
energy upon a 20-year, 8760-hour avoided cost matrix.
It was necessary to extrapolate the 20-year avoided cost
projection to 40 years given the longevity of some of the
measures. As a consequence of this avoided cost structure
it was necessary to develop an 8760-hour load profIle
for each measure to be evaluated. Navigant Consulting
Group provided 22 residential and non-residential load
profiles for use in this part of the exercise.
Calculating TRC cost-effectiveness: A full TRC
cost-effectiveness evaluation was performed upon the
remaining 39 residential and 36 non-residential
concepts.9 Four concepts were removed from this list due
to questions regarding the viability of the data obtained
in earlier stages or the discovery of previously undetected
fatal flaws to the program. The following section
provides a more detailed evaluation of the review and
acceptance or rejection of these concepts.
A summary list of the concepts reaching the evaluation
stage is included in the Appendices.
EVALUATION OF TRC COST-EFFECTIVENESS FOR
FINALIST CONCEPTS
The construction of the TRC cost for each measure was
based upon the incremental customer cost. Non-energy
benefits were considered, but none of the evaluated
measures had a large enough non-energy benefit to
8 See the Appendices for a list of these load proftles.
9 Three residential and one non-residential concept were subsequently excluded due to concerns over the validity of key resource characteristic
assumptions.
Avista Corp 2007 Electric IRP 3 - 7
Chapter 3- Demand Side Management
materially change the final cost-effectiveness evaluation.
Estimating the TRC values was more difficult. This
required a present value calculation of the avoided energy
and capacity cost over the measure life. The avoided cost
of energy was based upon an application of the measures
8760-hour load profile to the 8760-hour avoided cost
structure. Five energy and two capacity avoided cost
values developed within the Heritage Project Analytical
Roadrnap were applied to the load shapes of each
measure concept.
The valuation of capacity based upon these load
shapes and capacity avoided cost values had never been
incorporated into the evaluation ofDSM opportunities
at Avista. The per kW present values forT&D and
generation capacity estimated in the Analytical Roadmap
were based upon a single fIXed period of time. Escalating
streams of annual values that were consistent with the
values within the Analytical Roadrnap allowed for the
development of capacity values for varying measures lives.
The details of this calculation are contained within the
Appendices.
The consensus of opinion held that, for purposes of
the evaluation ofDSM measures, it was appropriate to
focus upon deferring a summer space-cooling-driven
load. The 71 concepts to be evaluated had significant
differences in their impact upon system coincident load
and these differences were not always apparent based
upon the general pattern of the measure load shape. To
determine the expected impact upon the deemed space
cooling-driven system peak load, the 71 concepts and 23
load shapes (including a flat load option) were
categorized into three groups.
Zero impact Measures that would not have any impact
on a summer space-cooling-driven peak received a zero
valuation regardless of their load profile. This would
include measures such as residential space-heating
efficiencies.
Non-Drivers Measures that were not related to space
cooling but would potentially contribute to system load
during a space cooling-driven peak received a capacity
valuation based upon the average demand of their
specific load profile during eight hour summer peak load
periods.12 These measures include commercial lighting,
residential appliances and so on.
Drivers Those measures that would drive a space
cooling peak received a capacity valuation based upon
the maximum hourly demand identified in their 8760-
hour load profile. This would include measures such
residential and non-residential air conditioning efficiency
measures.
Once the TRC cost and benefit calculations were
completed, a TRC ratio was developed. Even though
this analysis limits the identification of future DSM
acquisition to measures that fully pass the TRC cost-
effectiveness test, the company plans on evaluating all
measures with a benefit-to-cost ratio of 0.75 or higher.
Having identified TRC cost-effective measures it was
necessary to determine the annual acquisition of the
identified potential. Inspection of the results to date
indicated that there was clearly more potential than
identified in the 2005 IRP process (5.4 aMw, excluding
regional acquisition efforts, or 47.5 million first-year
kWh). Thus the acquisition of the potential conservation
requires a ramping-up ofDSM operations, which is
being done through the Heritage Project. A ramp
10 The non-energy benefit, or cost, could have been represented as a TRC cost or benefit as long as the appropriate sign was used in the
evaluation without impacting the ultimate passing or failing of the measure.
11 The specific components of the avoided cost are summarized in the Appendices.
The eight peak hours were 1 p.m. to 8 p., weekdays only, between June 15 and September 15.
2007 Electric IRP Avlsta Corp
Chapter 3- Demand Side Management
Figure 3,2: Year-an-Year Conservation Acquisition (%)
- -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - -......
C')\l)
..................
rate was developed based upon the sales cycle of the
customer decisions and the speed at which programs
could be developed, incorporated into trade ally efforts
and communicated to the customer base. This ramp rate
is represented graphically in Figure 3.2 and outlined in
more detail in the Appendices.
This completed the evaluation of those concepts that
were suitable for review by technology within the IRP.
These results are revisited following the explanation of
the programmatically reviewed elements of the DSM
portfolio.
EVALUATION OF COMPREHENSIVE PROGRAM
ELEMENTS
As a consequence of the all-inclusive nature of Avista
non-residential and limited income portfolio, it was not
feasible to generically evaluate all possible efficiency
measures. Nevertheless it is necessary to develop an
estimate of the potential of these markets in order to
establish a meaningful business planning process. Unique
efficiency measures could not be generically evaluated
as individual technologies. In place of this approach
the company established a historical baseline level of
acquisition and modified it to incorporate the impact of
known or likely changes in the market.
I'-C')
t:!
\l)
The company s limited income portfolio of qualifying
efficiency measures is all-inclusive. It is implemented in
cooperation with community action agencies given wide
latitude in their approaches. Given that no changes were
expected in the ability of the agency infrastructure to
deliver these programs, nor were there any known market
or technology changes that would cause a significant
change in the ability to obtain efficiency resources
from this segment, it was determined that a historical
baseline would be the most appropriate starting point for
estimating future throughput. This historical baseline was
modified for load growth and retail price elasticity based
upon assumptions consistent with the forecasts available
at the time. This resulted in a forecast of limited income
acquisition for incorporation into the final conservation
forecast.
Although some of the measures incorporated into the
site-specific program were specifically evaluated, a large
portion of non-residential acquisition comes from
measures which could not be generically evaluated.
As with the limited income program, the historical
baseline was modified for anticipated load growth and
retail price elasticity to develop a forecast. Unlike the
limited income program, it was necessary to separate
the specifically evaluated measures from the historical
Avista Corp 3 - 92007 Electric IRP
Chapter 3- Demand Side Management
baseline, and then combine the two again as part of the
final expected conservation acquisition.
This process is illustrated in a flowchart in the
Appendices.
COMPILATION OF THE FINAL DSM RESOURCE
ESTIMATES
The following conservation targets were developed
by summing individually evaluated concepts and the
evaluated programs over a 20-year period. The first two
years of those targets are detailed in Table 3.
A graphical representation of the annual conservation
targets for the full 20-year horizon is illustrated in Figure
3. A flat 1.4 aMW estimate of Avista s share of regional
resource acquisition (Avista s pro-rated share of NEEA's
annual savings) is included in the estimate.
A measure-by-measure stacking of the 71 evaluated
concepts, in ascending order oflevelized total resource
cost, leads to a traditional upward-sloping supply curve
for this component of the energy efficiency target, as
illustrated in Figure 3.4. Supply curves for both 2008
and 2009 have been shown to represent the two years
Limited Income Residential
Residential
Prescri tive Non-Residential
Site-S ecific Non-Residential
Total Local Ac uisition
562 956
10,939,762
279,711
39,184 260
52,966,686
594 215
13,674 702
599 639
40,359,787
228,343
Figure 3: Forecast of Efficiency Acquisition (aMW)
.....(")"'"
II)
......
II)
................(")
13 This application of price elasticity is consistent with but not incorporated within forecast assumptions since the efficiency savings quanti-
fied through the company s DSM programs are limited to those which are in excess of the higher of code-minimum or industry standard
practice.
14 In the absence of reliable 20-year estimates of acquisition through regional programs, it was assumed that the historic acquisition would
remain flat during that time at their most recent level. This assumption is speculative and dependent on the opportunities for regional
market transformation during this period but is consistent with the recent history of flat funding of NEEA.
3 -2007 Electric IRP Avista Corp
Chapter 3- Demand Side Management
Figure 3.4: Supply of Evaluated Efficiency Measures
------------------------ --------- --------
~ 1.
'tiI
8 1.
~ 0.
.!! 0.
I!: 0.4
- - --- - - -------- - -------- - - - - - - -- --- - - - - -- - -- - -- - - - -- -- -- - - - - ------------------- - ---- - - - - - - - - - - ----- ---- -- - - - - - - - - - ---- -
Annual GWh acquisition
Figure 3,5: Efficiency Supply Curves Including All Measures
1.4
- 1.
~ 0.
'ii~ 0.
I- 0.4
- - - - - -- - -- - ------- ------ - - - - -- - - - -- - ----- - - - ------- - -- --- -- - - - - - - -- - - - -- --- - ------- - - - - ----- - - --- - - - - - - - ------ - - -- ------- - - - - - - - - -- - -- - - - ----- ----- - - - - - - - -- - - - - - - -- - - - - - - - - - ----- - - - - ------ - ------- ------- - - - --- - - - - - ----- - - -- ----- - - - - - --- - - - - - - -- - - - -- -- - - - ------ -- - --- -- --- - - - - - - - - - ---- --- -- ----
50
Annual GWh acquisition
which will elapse before the next IRP. The rightward
shift of the supply curve over time is a consequence
of the assumptions made in the ramping-up of these
programs.
The rapid sloping of the supply curve tails are the result
of including a few measures that were later determined
to be far more costly than previously anticipated.
These programs, though small, significantly extended the
vertical axis of the supply curve developed for the
efficiency measures.
By adding the target for programmatically-evaluated
energy efficiency efforts to the left portion of the supply
curve, a full assessment of the estimated efficiency targets
can be illustrated. This is shown in Figure 3.
INTEGRATING IRP RESULTS INTO THE
BUSINESS PLANNING PROCESS
The IRP evaluation process provides a high-level
estimate of cost-effective energy efficiency acquisition.
Based upon these results the company can establish a
budget, determine the size and skill sets necessary for
15 These two measures were residential induction cook tops and non-residential demand-controlled ventilation. The measures exceeded a
levelized TRC cost of approximately 80 cents per kWh. Four other measures exceeded levelized TRC costs of25 cents per kWh: non-
residential window films, non-residential light colored roofs, residential smart appliances and non-residential Energy Star office equipment.
Avista Corp 2007 Electric IRP
Chapter 3- Demand Side Management
future conservation operations and identifY general target
markets.
The results of the IRP analysis will establish baseline
goals for the ongoing development of the Heritage
Project's enhancements to Avista s energy efficiency
programs. The near-term planning is summarized by
portfolio in the following sections.
RESIDENTIAL PORTFOLIO
A review of residential concepts and their sensitivity to
key assumptions indicate that more detailed assumptions
based upon actual program plans and target markets may
improve the cost-effectiveness of many concepts that
marginally failed in this analysis. To account for this
marginal failure rate, all concepts with TRC benefit-to-
cost ratios of 0.75 or better will be evaluated as part of
the business planning process. Twenty-seven of the 36
evaluated residential concepts meet this criterion.
Measures that were developed too late for the IRP
evaluation will also be inserted into this re-evaluation
process. One of the recent additions, top-mounted
fireplace dampers, has completed the program planning
and evaluation process and was launched prior to the
completion of this IRP.
LIMITED INCOME RESIDENTIAL PORTFOLIO
Avista has committed to maintaining stable annual
funding and program flexibility for the six community
action agencies delivering limited income energy
efficiency implementation services. The flexibility of
these programs requires periodic updates to program
expectations due to changes in fuel focus and target
measures. The company will also be working to quantify
the future potential impacts of the three-year Northwest
Sustainable Energy for Economic Development project.
NON-RESIDENTIAL PORTFOLIO
Similar to the residential program, it was determined that
there is potential for improvement in evaluated program
concepts to warrant the re-evaluation of any measure
determined to have a TRC cost-to-benefit ratio of 0.
or better. Of the 35 fully evaluated non-residential
concepts, 25 of these meet the TRC criteria. These
programs will be reviewed for target marketing, the
creation of a prescriptive program or for targeting under
the site-specific program.
All electric-efficiency measures qualify for the non-
residential portfolio. The IRP provides account
executives, program managers and end-use engineers
with information regarding potentially cost-effective
target markets, but specific characteristics of customers
facilities override any high-level program prioritization.
UNDERLYING RESOURCE ACQUISITION
COMMITMENT
The IRP evaluation process is both a business planning
process and regulatory requirement. The company uses
this opportunity for comprehensive evaluation as a part
of the management of the company s energy efficiency
portfolio. The acquisition targets provide valuable
information for future budgetary, staffing and resource
planning needs. However, numerical targets do not
displace the companys fundamental obligation to pursue
a resource strategy that best meets the customer needs
under continually changing environments. The targets
established within this IRP planning process may be
modified as necessary to meet these obligations.
SUPPLY SIDE EFFICIENCY
Avista also actively works on improving efficiency of
its generation fleet. The following section highlights
planned and potential hydroelectric efficiency upgrades.
Recent thermal upgrades to the Colstrip plants are
detailed in chapter two.
NOXON RAPIDS
The company plans to upgrade Noxon Rapids units
3 - 12 2007 Electric IRP Avista Corp
Chapter 3- Demand Side Management
. e-ce-. roo Icleo
. .
ra. . e-u'leo
Potential Potential Potential Total
Additional Additional Additional Project
Annual Energy Annual Energy Capacity Capacity
Project (MWh)(aMW)(MW)(MW)
Noxon Rapids 50,808 16.570.
Nine Mile
Rubber Dam 500 26.4
Turbine UpQrades 000 34.4
Upper Falls 63,000 6.4 15.
Little Falls 000 44.
4 beginning in March 2009. The current maximum
capability at Noxon Rapids is 554 MW; however
operating restrictions limit the plant to 532 MW The
upgrades will eliminate the operating restrictions and add
an additional 16 MW to the project, increasing the plant
capability to 570 MW and add 5.8 aMW of energy.
NINE MILE
The company currently uses flashboards at its Nine Mile
plant to increase water storage during the fall and winter
months. The flashboards are released downstream during
spring runoff when the reservoir level must be lowered
to accommodate the increased flow of water. The
flashboards are re-installed every summer. The company
is considering replacing the flashboards with a permanent
pneumatic rubber dam which would automatically
adjust the reservoir level to the flow rate, increasing the
reservoir level when flow is low and decreasing the level
when flows increase. The rubber dam would stabilize
the Nine Mile project as well as eliminate the need to
purchase and reinstall flashboards each year. This project
would increase annual generation by about 6 500 MWh.
T bl 3 4 R t H d
Also two of the four generators at the Nine Mile project
require repair or replacement in the near future. The
company is studying the replacement of these units
in-kind or replacing with larger units to increase the
maximum capacity and maximum flow at the project.
UPPER FALLS
The Upper Falls project, located in downtown Spokane
has one generating unit. The company is currently
studying the advantages of upgrading the turbine runner
and refurbishing other generator components.
LITTLE FALLS
Turbine runners at two of the four generators at Little
Falls have recently been replaced. The company is
studying the benefits of replacing the turbine runners
in the remaining units. Other potential projects include
replacing the step-up transformers and upgrading other
generator components.
A summary of the various hydro efficiency studies is
shown in Table 3.4.
Effi 'St d'
Avista Corp 2007 Eiectric IRP
Chapter 4- Environmental Issues
ENVIRONMENTAL ISSUES
Environmental issues cover a wide variety of topics. To
keep the concepts manageable, this chapter highlights
some of the more important environmental issues
affecting resource planning, the most notable being
thermal plant emissions. The chapter is not intended to
debate the merits or weaknesses of environmental science
or the effects of power generation emissions. Instead
it covers state and federal laws and pending legislation
affecting sulfur dioxide (SO ), nitrogen oxide (NO),
mercury (Hg), and carbon dioxide (CO ) emissions. The
modeling assumptions used for each emission types are
explained. Particular attention is paid to greenhouse
gases (GHG) because their regulatory future is the
most uncertain and has the potential to affect resource
decisions most significantly.
ENVIRONMENTAL CHALLENGES
Emissions present a unique challenge for resource
planning because of continuously evolving scientific
understanding and legislative developments. If
environmental concerns were the only issue faced by
utilities, resource planning would be reduced to choosing
the amount and type of renewable generating technology
to use. However, utility planning is compounded by
requiring cost effectiveness. Each type of generating
resource has distinctive operating characteristics, cost
structures, and environmental challenges. Traditional
generation technologies are well understood. Coal-fired
units have high capital costs, long lead times, and low and
stable fuel costs. Coal plants are difficult to site and are
affected by a host of environmental issues from Hg to
Sheep Grazing Near a Wind Farm in Washington State
CHAPTER HIGHLIGHTS
. The company includes greenhouse gas emissions costs in its Base Case.
. Avista relies on its Climate Change Committee to develop climate change policy and mitigation plans.
. 802, NO ' Hg, and CO2 emissions costs are included in the modeling for the 2007 IRP.
. Avista supports national greenhouse gas legislation that is workable, cost effective, fair, protects the
economy, supports technological innovation and addresses emissions from developing nations.
. Avista is a member of the Clean Energy Group.
Avista Corp 2007 Electric IRP
Chapter 4- Environmental Issues
GHG. Natural gas-fired plants have relatively low capital
costs and more acceptable emission levels but rely on
fuel that has proven to be both high in price and price
volatility. Renewable energy plants, including wind
biomass and solar, have different problems to contend
with. Renewables benefit from potential low or no
fuel costs and low or no emissions, but they are plagued
by capacity problems, wildlife issues, high capital costs
uncertainty regarding production tax credits and an
increasing number of siting issues.
The most uncertain aspect of emissions is future GHG
legislation. There recently has been a tremendous
upsurge in the amount of scientific, public and legislative
attention regarding climate change. There are five
main aspects to consider with climate change: scientific
public, government, legal and fmancial. The scientific
community has shown increasing evidence of human
involvement in global warming, culminating with the
Intergovernmental Panel on Climate Change (IPCC)
Fourth Assessment Report, which was released in
February 2007. This report stated that there is a greater
than 90 percent chance that global warming is the
result of human intervention through greenhouse gas
emissions. The public is becoming increasingly aware of
climate change issues and is pressing for governmental
and corporate action. Legislatively, there are increasing
numbers oflocal, state, regional, and federal GHG
initiatives, renewable portfolio standards and emissions
standards. On the legal front there are issues of state versus
federal jurisdiction, project-specific pressures and attempts
at class action lawsuits. Examples oflegal issues include
the April 2, 2007, u.S. Supreme Court decision that the
Environmental Protection Agency had a duty to regulate
greenhouse gases; the environmentally-pressured decisions
in the leveraged buyout case ofTXU not to build eight
new coal plants; and the climate change lawsuits flied
against utilities, auto makers and oil companies in the
wake of hurricanes along the Gulf Coast. Financially,
there are potential compliance costs, increasing demand
for renewable resources driving up prices and shareholder
pressure regarding climate change issues.
AVISTA'S ENVIRONMENTAL INITIATIVES
AND POLICIES
One of the 2005 IRP action items was to "continue to
monitor emissions legislation and its potential effects on
markets and the company." This action item has received
significant attention throughout the company over the
past two years which resulted in an interdepartmental
meeting on June 8,2006, to cover climate change
topics including: Congress and climate change Avista
GHG inventory, Coyote Springs 2 emissions offsets
emissions assumptions included in the IRP and state
commissions' guidance on climate change. Mter this
meeting, a core group of employees from Environmental
Affairs, Governmental Affairs and Resource Planning
began meeting regularly to discuss current climate
change information and legislative activities affecting
the company. This group also reviewed climate change
policies from other organizations, worked on drafting
Avista s climate change statement and developed
educational pieces.
The core group met with the company s Strategic
Planning Council in March 2007 to discuss current
climate change activities and developments. This
meeting resulted in the appointment of an officer to
spearhead the formalization of Avista s Climate Change
Council (CCC). The CCC has been chartered to be a
clearinghouse on all matters related to climate change.
The CCC:
. anticipates and evaluates strategic needs and
opportunities;
. analyzes the implications of various trends and
proposals;
. develops recommendations on company positions
and action plans; and
. facilitates internal and external communications.
4 - 2 2007 Electric IRP Avista Corp
Chapter 4- Environmental Issues
The core team of the CCC includes members from the
Environmental Affairs, Government Relations, Corporate
Communications, Engineering, Energy Solutions and
Resource Planning departments. Other areas of the
company are invited as needed.
Monthly meetings divide work into immediate and long-
term concerns. Immediate concerns include reviewing
and analyzing state and federal legislation, developing
a corporate climate change policy and responding to
external data requests regarding climate change issues.
Longer term issues involve emissions tracking and
certification, reviewing alternatives and providing
recommendations for GHG reduction goals and
activities, evaluating the merits of joining various
GHG reduction programs, actively participating in the
development of GHG legislation, and benchmarking
climate change policies and activities with other
organizations.
Avista recently joined the Clean Energy Group which
includes Calpine, Entergy, Exelon, Florida Power and
Light, PG&E and Public Service Energy Group. This
group acts collectively to evaluate and support different
GHG legislation such as the Clean Air Planning Act of
2007 sponsored by Tom Carper (D-DE). This legislation
seeks to establish multi-pollutant limits using a market-
based approach to "reducing power plant emissions of
nitrogen oxides, sulfur dioxide, mercury and carbon
dioxide.
AVISTA'S POSITION ON CLIMATE CHANGE LEGISLATION
The company expects federal greenhouse gas legislation
to be enacted within the next two to four years. The
absence of definitive legislation on climate change creates
an uncertain environment as the company develops its
plans for meeting future customer loads. Avista does not
have a preferred form of GHG legislation at this time.
However, the company supports federal legislation that:
. anticipates and evaluates the strategic needs and
opportunities;
. is workable and cost effective;
. is fair;
. is protective of the economy;
. is supportive of technological innovation; and
. is inclusive of emissions from developing nations.
Workable and cost effective legislation would be carefully
crafted to produce actual emission reductions through
a single system, as opposed to competing state, regional
and federal systems. The legislation also needs to be
fair in that it is equitably distributed across all sectors
of the economy based on relative contribution to
GHG emissions. Protecting the economy is of utmost
importance. The legislation cannot be so onerous that it
stalls the economy or fails to have any sort of adjustment
mechanism in case the market solution fails and prices
skyrocket. Supporting a wide variety of technological
innovations should be a key component of any GHG
reduction legislation because innovation can help
maintain costs, as well as provide a potential boost to the
economy through an increased manufacturing base. The
final piece to the legislative solution to climate change
involves developing nations. China will soon overtake
the u.S. as the leading source of GHG emissions.
Legislation should include strategies for working with
other nations directly or through international bodies to
control world-wide emissions.
EMISSIONS CONCERNS FOR RESOURCE
PLANNING
The main emissions concerns for resource planning
involve balancing environmental stewardship and cost
effectiveness, and mitigating the financial impact of
emissions risks. The 2007 IRP focuses on four types of
emissions that are significant to electric generation: SO
NO x' Hg, and CO , Sulfur dioxide is a cause of acid
rain; the Clean Air Act of 1990 capped its emissions at
9 million tons per year starting in 2008. This pollutant
Avista Corp 2007 Electric IRP
Chapter 4- Environmental Issues
is actively regulated through a cap-and-trade program.
Nitrogen oxide is also regulated by the Clean Air Act
of 1990 at 2.0 million tons per year starting in 2008.
Mercury is an emission with planned regulation by the
federal government under a cap-and-trade program.
However, many states are opting out of that program.
Carbon dioxide is a primary greenhouse gas. It is
beginning to be regulated in some states and is the focus
of federal legislation.
EMISSIONS LEGISLATION
There are several themes that emerge from all of the
recently developed climate change legislation. These
include:
. Scientific questions about human contributions to
climate change - is it an anthropogenic or human-
developed phenomenon need to be settled;
. Actions need to be economy-wide, rather than
one or two sectors at a time;
. Technology will be a key component to the
climate change solution. There will most likely
need to be significant investments in carbon
capture and sequestration technology, since coal
likely will continue to be an important part of the
u.s. generation fleet;
. Developing countries should be engaged as
developing nations to expand their economies and
carbon footprints; and
. Long delays in federal legislation increase the
probability of a menagerie of inconsistent
regulatory schemes that may obstruct the efficient
operation of regional or national businesses.
These themes point to national comprehensive GHG
legislation implemented in a timely manner to ensure the
best environmental and fiscal outcomes.
FEDERAL EMISSIONS LEGISLATION
The federal government is currently reviewing at least six
different market-based programs to reduce greenhouse
gas emissions. This is the culmination of many previously
failed attempts at national legislation, the most significant
being the McCain-Lieberman Climate Stewardship Act
submitted to Congress in January 2003 and annually
thereafter. Most legislation relies on a market-based cap-
and-trade system in an attempt to emulate the success of
the national acid rain program. There are many questions
that still need to be resolved before national GHG
legislation can be enacted. These include:
. the allocation of allowances - emissions or
generation-based;
. economy-wide or sector specific;
. offsets;
. incentives for early action;
. economic safety valves;
. up or downstream regulation; and
. cap-and-trade or tax.
There are indications from Congress that federal
legislation will be passed in 2007 , but great uncertainty
still remains over the specifics of the legislation or when
it will be passed into law. The company believes that
some form of market-based GHG legislation is inevitable
and includes it in its Base Case IRP assumptions.
The company introduces CO2 emission charges in
2015. Recent developments in GHG legislation lean
toward an earlier start date, but 2007 IRP modeling
was substantially complete before recent Congressional
activity began. Upon review of the modeling results
the company does not believe that adding charges sooner
would in any way impact its Preferred Resource Strategy.
STATE LEVEL EMISSIONS LEGISLATION
Federal inaction on climate change has spurred many
states to develop their own laws and regulations. Climate
change legislation has taken many forms, including
GHG emissions caps, renewable portfolio standards
(RPS) and mandated efficiency levels. A patchwork
of competing rules and regulations has sprung up for
utilities to follow, making resource planning for utilities
4 - 4 2007 Electric iRP Avista Corp
Chapter 4- Environmental Issues
with multi-jurisdictional responsibilities like Avista more
difficult. Currently there are 23 states and the District
of Columbia with active renewable portfolio standards.
California, Connecticut, North Carolina and Rhode
Island are working on legislation to phase out the use of
incandescent light bulbs.
Some of the more notable state-level GHG initiatives
outside of the Pacific Northwest include the Regional
Greenhouse Gas Initiative (RGGI): an agreement
between 10 Northeastern states (Connecticut, Delaware
Maine, Maryland, Massachusetts, New Hampshire, New
Jersey, New York, Rhode Island, and Vermont) to develop
a cap-aDd-trade program for power plant CO2 emissions.
The District of Columbia, Pennsylvania and some
Canadian Provinces are participating as observers in the
RGGI process.
The Western Regional Climate Action Initiative was
developed from a Feb. 26, 2007, agreement between
Washington, Oregon, California, New Mexico Arizona
and British Columbia to reduce GHG emissions through
regional reduction goals and the establishment of a
market-based trading system. There are a number of
regional municipalities participating in the u.S. Mayors
Climate Protection Agreement to reduce GHG emissions
to 93 percent of 1990 levels by 2012.
Nationally the Clean Air Mercury Rule (CAMR)
established permanent caps to reduce mercury reduction
from coal-fired power plant emissions. CAMR allows
states to participate in a nation-wide mercury trading
allowance program. States are allowed to determine
their national allocations are distributed among
existing emitters, auctioned or some combination of the
two methods.
IDAHO EMISSIONS LEGISLATION
Idaho does not actively regulate greenhouse gases or
set renewable portfolio standards for its electric utilities.
Idaho governor Butch Otter issued an executive
order in May 2007 directing the Idaho Department
of Environmental Quality (IDEQ) to work on "
policy on the role of state government in reducing
greenhouse gases.The IDEQ is to develop a GHG
emissions inventory and reduction strategy. Idaho has
demonstrated concerns with coal-fired power plants;
most notably, HB 791 (2006) established a moratorium
on new merchant coal-fired power plants for a two-year
period. The state has decided to opt out of CAMR
meaning that a plant located in Idaho could not purchase
mercury credits to offset its emissions. By opting out
of CAMR, the state has effectively stopped coal plant
development.
MONTANA EMISSIONS LEGISLATION
The Montana Global Warming Solutions Act (HB753)
was submitted in late 2006 to establish greenhouse gas
reductions goals through 2020. The legislation did not
make it out of committee. Montana limits mercury
emissions to 0.9 pounds per decatherm for plants using
sub-bituminous coal, and 1.5 pounds for lignite-fired
plants. Montana requires 15 percent of all electricity to
come from new renewables by 2015.
OREGON EMISSIONS LEGISLATION
Oregon has been actively developing greenhouse gas
renewable portfolio standards and mercury emission
legislation. Oregon s climate change legislation goes back
to its December 2004 Oregon Strategy for Greenhouse
Gas Reduction. It called for development of a detailed
GHG report by the end of 2007 and for stabilization of
all six GHGs by 2010, a 10 percent reduction from 1990
levels by 2020 and a 75 percent reduction from 1990
levels by 2050. The goals are in addition to the 1997
regulation requiring utilities to offset CO2 emissions
exceeding 83 percent of the emission level of a state-of-
the-art gas-fired CCCT. State Senate Bill 838 requires
large electric utilities to generate 25 percent of annual
electricity sales with new renewable resources by 2025.
Avista Corp 2007 Electric IRP
Chapter 4- Environmental Issues
Shorter term renewable goals include 5 percent by 2011
15 percent by 2015, and 20 percent by 2020. Oregon
has set mercury emissions levels equaling 90 percent
reduction or 0.60 pounds per Dth by July 1 2012, with
some allowances for compliance alternatives if the targets
cannot be met using best available emissions controls.
WASHINGTON EMISSIONS LEGISLATION
Washington State is quite active on global warming
and renewable energy issues, recently passing an RPS
initiative and GHG legislation. This is in addition to a
2004 law requiring new fossil-fueled thermal electric
generating facilities of more that 25 MW to have a CO
mitigation plan of third-party offsets, purchased carbon
credits or cogeneration.
The Washington Clean Energy Initiative (1-937)
passed in the November 2006 election. This initiative
established an RPS for Washington equal to 3 percent
retail load by 2012, 9 percent by 2016, and 15 percent
by 2020. The 2007 IRP has been developed so that the
937 RPS goals will be achieved by the company for its
Washington retail load.
Governor Christine Gregoire signed Executive Order
07-02 in February 2007 , establishing the following GHG
emissions goals:
. return to 1990 levels by 2020;
. 25 percent below 1990 levels by 2035;
. 50 percent below 1990 levels by 2050, or 75
percent below expected emissions in 2050;
. increase clean energy jobs to 25 000 by 2020; and
. reduce statewide fuel imports by 20 percent.
The goals of this Executive Order became law when
SB 6001 was signed on May 3 2007. The law reduces
the GHG emissions of electric utilities by establishing
an emissions performance standard ofl 100 pounds of
GHG per MWh of new base load generation.
Washington state has proposed mercury legislation
levels of8.7Ib/MWh from all sources by 2013 , with
mandatory plant compliance of utilities by 2017. Trading
is allowed for the first three years. The allocation base
is tentatively set at 70 percent to existing sources
percent to new sources, and the balance held for possible
future distribution. Final mercury rules are expected by
September 2007.
EMISSIONS MEASUREMENT AND MODELING
To evaluate the impact of emissions regulation on market
prices and resource dispatch, estimates of the amounts
of dollars to "tax" certain emissions were made. This
tax is used as an economic indicator oflower emissions.
000
Figure 4.1: Base Case 502 Costs ($/Ton)
000
- Max & Min
- - - ---
A~ffige
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
. 80% Confidence Interval
000
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
000
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
000
000
000
......
C')'OJ'It)r--
......
C')It)r--
............
2007 Electric IRP Avista Corp4 - 6
Chapter 4- Environmental Issues
Valuing emissions is an important part of the IRP
modeling process. Mercury, SO2' and NOx are modeled
using a lognormal distribution, whereas CO2 is modeled
based on a sampling distribution of 300 Monte Carlo
iterations. Each of the four modeled emissions types is
discussed below.
2 emissions average $808 per ton in 2008 and escalate
to $2 571 per ton in 2027 in nominal dollars. SO2 has an
actively traded market so emissions costs and projections
are readily obtained. Figure 4.1 shows the minimum
maximum and average levels ofSO2 emissions costs.
NO x emission costs are $2 248 per ton beginning in
2010 when regulations begin and escalate to $3 875 per
ton in 2027. The NOx market will operate in a manner
that is very similar to the SO2 market. Figure 4.2 shows
the data for NO x cost projections.
Mercury is somewhat problematic to model because
trading does not begin until 2010 and many states have
decided to opt out of the national trading market under
CAMR. Projections of mercury costs are not readily
available. The IRP bases its cost estimates on a variety of
governmental and private sources. Mercury costs start
000
Figure 4.2: Base Case NOx Costs ($/Ton)
000
- Max & Min
- - ---
A~rege
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
. 80% Confidence Interval
000
000
000
000
- -
000
CX)
..-
C')-.:t
..-..-..-
r---CX)C')r---
000
Figure 4.3: Base Case Mercury Costs ($/Ounce)
12,000
- Max & Min
---
A~rege
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
. 80% Confidence Interval
10,000
000
000
000
000 u
CX)
..-
C')-.:t
..-
r---
..-
CX)C')r---
Avista Corp 4 - 72007 Electric IRP
Chapter 4- Environmental Issues
in 2010 at $1 739 per ounce and escalate to $4 863 per
ounce in 2027 (nominal dollars). Mercury emission cost
estimates are shown in Figure 4.
2 emissions are modeled based on a probability
distribution of the 300 Monte Carlo iterations of
AURORAxmp run for the Base Case. The mean value
of the probability distribution equals the projected
cost of the National Commission on Energy Policy
recommendations in their 2004 study. The projected
costs from that study have been escalated to account for
inflation. Figure 4.4 shows the projected CO2 values by
year. Costs average $8.94 per ton in 2015 and increase to
$14.34 per ton in 2027.
Figure 4.4: Base Case CO2 Costs ($/Ton)
- Max & Min
- - - ---
A~mge
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
. 80% Confidence Interval
------------------- --- ------ --- ----- -- -- -- -- -- -- -- -- -- -- -- - - - --- - - - ---- -- --- ----
C\I C\I C\I
C\I
C\I
......
C\I
"'"......
C\I C\I
......
C\I
to-
C\I C\I C\I C\I C\I C\I C\I C\I
to-
C\IC\I
4 - 8 2007 Electric iRP Avista Corp
Chapter 5- Transmission Planning
TRANSMISSION PLANNING
Transmission Construction in the Palouse Region
Southeastern Washington
INTRODUCTION
Comprehensive coordination of transmission system
operations and planning activities with regional
transmission providers is necessary to maintain reliable
and economic transmission service for the region s end-
use customers. Transmission providers and interested
stakeholders are implementing changes in the region
approach to planning, constructing and operating the
system under new rules promulgated by the Federal
Energy Regulatory Commission (FERC) and under state
and local siting agencies. This section was developed
in full compliance with Avista s FERC Standards of
Conduct, governing communications between Avista
merchant and transmission functions.
AVISTA'S TRANSMISSION SYSTEM
Avista owns and operates an electric transmission system
comprised of approximately 623 miles of 230 kilovolt
(kV) line and 1 537 miles of 115 kV line. The company
also owns an 11 percent interest in 495 miles of a 500
kV line between Colstrip and Townsend, Montana. The
transmission system includes switching stations and high-
voltage substations with transformers, monitoring and
metering devices, and other system operation-related
equipment. The system is used to transfer power from
the company s generation resources to its retail load
centers. Avista also has network interconnections with
the following utilities:
. Bonneville Power Administration (BPA)
. Chelan County PUD
. Grant County PUD
. Idaho Power Company
. NorthWestern Energy
. PacifiCorp
. Pend Oreille County PUD
. Puget Sound Energy
In addition to providing enhanced transmission system
reliability, these network interconnections serve as
points of receipt for power from generating facilities
outside the company s service area, including the
Colstrip generating station, Coyote Springs 2 and
the Mid-Columbia hydroelectric facilities. These
interconnections provide for the interchange of power
with entities within and outside of the Pacific Northwest
including the integration oflong-term and short-term
contract resources. Additionally, the company has
interconnections with several government-owned and
cooperative utilities at transmission and distribution
CHAPTER HIGHLIGHTS
. Avista is in the fifth year of a $130 million transmission improvement project.
. Avista has over 2 100 miles of high voltage transmission.
. The company is actively involved in the regional transmission planning efforts of ColumbiaGrid.
. The cost of new transmission lines and upgrades are included in the 2007 Preferred Resource Strategy.
. New construction costs approximately $1.4 million per mile of 500 kV transmission line.
Avista Corp 2007 Eiectric IRP 5 -
Chapter 5- Transmission Planning
voltage levels, representing non-network, radial points of
delivery for service to wholesale loads.
Avista is currently in the fIfth year of a multi-year, $130
million, transmission upgrade project. The planned
upgrades will add over 100 circuit miles of new 230 kV
transmission line to the company s system and increase
the capacity of an additional 50 miles of transmission
line. The transmission upgrade project also includes the
construction of two new 230 kV substations and the
reconstruction of three existing transmission substations.
Upgrades at six 230 kV substations are being undertaken
to meet capacity requirements, to upgrade protective
relaying systems and meet reliability standards. In total
Avista will work on 11 of 13, or 85 percent, of its 230
kV substations. The telecommunication system
also being upgraded with the installation of fiber and
digital microwave systems to improve system control
monitoring and protection. The company s most
significant transmission projects are described below.
BEACON-BELL 230 KV
The company increased the capacity of two parallel path
transmission lines from its Beacon substation to BPA's
Bell substation. The project doubled the line capacity
to 800 MVA and increased equipment ratings from both
substations. The project mitigates overloads between the
largest Avista and BPA substations in Spokane to improve
load service to the Spokane area. The upgrade to
Bell #4 was completed in December 2005 and Bell #5
was energized in April 2007.
BEACON-RATHDRUM 230 KV
Avista recently reconstructed 25 miles of single circuit
230 kV transmission line to a double circuit 230 kV line
between Rathdrum, Idaho, and Spokane, Washington.
DRY CREEK
A second 230/115 kV transformer was added to the Dry
Creek substation to improve load service and system
reliability in the Lewiston-Clarkston area. The new
transformer provides back-up for the North Lewiston
230-115 kV transformer. This project also included
the construction of the 115 kV portion of the Dry
Creek Substation and the loop-in of an area 115 kV
transmission line. This project was completed in the fall
of 2006.
PALOUSE REINFORCEMENT
The company is constructing 60 miles of 230 kV
transmission line between the Benewah and Shawnee
substations to relieve congestion on the existing
Benewah-Moscow 230 kV line. The project provides
a second 230 kV transmission line between the
companys northern and southern load service areas
which significantly improves system reliability. Several
components of the Palouse Project were energized and
placed into service in 2006, including the double circuit
Shawnee-Colfax 230 kV and 115 kV line section and the
Benewah Substation rebuild.
PINE CREEK SUBSTATION
The company reconstructed the Pine Creek 230 kV
Substation in November 2003. This facility is located in
Pinehurst, Idaho.
SPOKANE VALLEY REINFORCEMENT
Avista is adding 500 MVA of230 kV to 115 kV
transformation at the new Boulder Substation.
WEST OF HATWAI TELECOM PROJECT
The ability to communicate, monitor and control
transmission equipment is vital to providing reliable
service. The West of Hatwai (WOH) Telecom Project
is comprised of several sub-projects. The Noxon-Pine
Creek fiber project completes a telecommunication
ring from Spokane to the Noxon Rapids Hydroelectric
Project. The ring provides redundant communication
paths, so the loss of one side of the ring will not
eliminate the ability to control equipment. The ring is
5 -Avista Corp2007 Electric IRP
Chapter 5- Transmission Planning
also required to implement the Clark Fork Remedial
Action System (RAS), which drops generation at the
Clark Fork Projects after critical transmission outages
to ensure system reliability. Another component of the
Clark Fork RAS includes the addition of fiber from the
Cabinet Gorge generation units to the 230 kV Cabinet
Substation. The Hatwai-North Lewiston fiber project
completed a fiber ring around the Lewiston-Clarkston
load service area. This project is also part of a RAS
to improve reliability in the Lewiston area. All three
projects were completed in 2006.
As noted in the August 2002 West of Hatwai letter of
agreement with BPA, these projects are coordinated
to support and enhance BPA transmission projects.
Collaboration has allowed both parties to achieve a
least-cost service plan addressing commercial transactions
load service and regional reliability issues. The Avista
and BPA plan was reviewed by peer utilities, approved
by other Northwest transmission owners and by utility
members of the Western Electricity Coordinating
Council (WECC). The Northwest Power Pool (NWPP)
Transmission Planning Committee agreed that a blended
plan was superior to stand-alone plans separately
executed by the company and BPA.
Avista plans and operates its transmission system pursuant
to applicable criteria established by the North American
Electric Reliability Corporation (NERC),WECC and
the NWPP. Through its involvement in WECC and the
NWPP standing committees and sub-committees, the
company participates in the development of new and
revised criteria, and coordinates planning and operation
of its transmission system with neighboring systems.
The company is subject to periodic performance audits
through participation in these regional organizations.
Portions of the company s transmission system are
fully subscribed for transferring power output
company generation resources to its retail load centers.
Transmission capacity that is not reserved to move power
to satisfy long-term (greater than one year) obligations
is used to facilitate short-term purchases and sales to
optimize the company s resources, as well as to provide
wholesale transmission service to third parties pursuant
to FERC requirements under Orders 888 and 889. It
is important to note that the implementation of FERC
policies and practices under Orders 888 and 889, and
subsequent FERC orders, can occasionally restrict our
ability to optimize transmission system resources in
specific cases. Transmission capacity that might have
been either reserved or recalled to deliver lower-cost
short-term resources for service to native load customers
may not be available because ofFERC policies requiring
transmission capacity to be available for other parties. To
the extent a third party has secured firm capacity rights
on Avista s transmission system, including future rollover
rights, that transmission capacity will not be available for
the company to serve native load.
REGIONAL TRANSMISSION SYSTEM
BPA operates over 15,000 miles of transmission facilities
throughout the Pacific Northwest. BPA's system
represents approximately 75 percent of the region
high voltage (230 kV or higher) transmission grid. The
company uses the BPA transmission system to transfer
output from its remote generation sources to the
company s transmission system, such as Colstrip, Coyote
Springs 2 and the Washington Public Power Supply
System Washington Nuclear Plan No.3 settlement
contract. The company also contracts with BPA to
transfer power from the company s local resources to 10
of its remote retail load areas.
The company participates in a number of regional and
BPA-specific forums to coordinate system reliability
issues and to manage costs associated with the BPA
transmission system. The company participates in BPA
transmission and power rate case processes and in BPA'
Business Practices Technical Forum, to ensure BPA
Avista Corp 2007 Electric IRP 5 - 3
Chapter 5- Transmission Planning
transmission charges remain reasonable and support
system reliability and access. The company also works
with BPA and other regional utilities to coordinate major
transmission facility outages.
Future regional resource development will require new
transmission assets. BPA has indicated that financing
restrictions may hamper its ability to construct
new transmission to support these resources. BPA
transmission customers seeking firm capacity for their
new resources may be required to provide a form
oflong-term financing for BPA to facilitate needed
transmission project construction on its system.
REGIONAL TRANSMISSION ISSUES
Coordinated transmission planning has historically
occurred through various NWPP workgroups.
ColumbiaGrid is a more formalized Northwest
organization that has been created to develop a regional
transmission plan, assess transmission alternatives
(including non-wires alternatives) and provide a
decision-making forum for new projects and cost
allocation methods. ColumbiaGrid was formed
on March 31 2006, as a non-profit, membership,
Washington state corporation. The current members
ColumbiaGrid are Avista, BPA, Chelan County PUD
Grant County PUD, Puget Sound Energy, Seattle City
Light and Tacoma Power.
During the first quarter of 2007, Avista signed a
transmission planning agreement with ColumbiaGrid
to address regional transmission issues. ColumbiaGrid
will perform a number of services under the Planning
Agreement. It will prepare a Biennial Transmission
Plan and, as part of that process, will perform system
assessments of the parties' transmission systems and
identify projected transmission needs. ColumbiaGrid
will also facilitate a coordinated planning process for the
development of multi-transmission system projects.
THE BIENNIAL TRANSMISSION PLAN
Under the planning agreement, ColumbiaGrid will
prepare and adopt a Biennial Transmission Plan during
each two-year planning cycle. The plan will have a 10-
year planning horizon, or longer if required by FERC's
pro forma open access transmission tariff. Throughout
the planning process, drafts of the Biennial Plan will be
posted on the ColumbiaGrid website as they become
available.
As a primary component of the plan, ColumbiaGrid
will perform annual system assessment of the parties
transmission systems. The system assessment will
determine the ability of each planning party to serve
consistent with the planning criteria, its network load
and native load obligations, and other existing long-term
firm transmission obligations anticipated to occur during
the planning horizon. Projected inabilities to meet
such obligations are identified and solutions proposed
outlining those solutions that can be implemented by a
party on a single system basis versus those transmission
solutions that impact the regional transmission grid
multi-system projects ). Those transmission system
modifications that will impact only a single party
transmission system are included in ColumbiaGrid'
biennial plan for informational purposes.
COORDINATED PLANNING OF MULTI-SYSTEM PROJECTS
Columbia Grid will facilitate coordinated planning of
all multi-system transmission projects. If the annual
system assessments identify a need that implicates a
multi-system transmission project, ColumbiaGrid will
develop conceptual transmission solutions through the
creation and use of study teams made up of members
from a number of stakeholder categories. The objective
of a study team will be to develop a transmission plan
that will resolve a reliability need or provide sufficient
capacity for a request for transmission service in a timely
fashion.
5 -Avista Corp2007 Electric IRP
Chapter 5- Transmission Planning
Columbia Grid's unique structure provides a means
for resolving disputes related to multi-system projects.
Transmission system modifications that will impact
more than one transmission system must be approved
by a majority vote of the Columbia Grid board before
they can be incorporated into the final biennial
plan. Projects where all affected parties have reached
agreement will be included in the draft biennial plan
submitted to the board. In the event agreement is not
reached by all affected parties, Columbia Grid staff may
make a recommendation to the Board on whether
to include it in the draft biennial plan and affected
parties may provide comment to the ColumbiaGrid
board. ColumbiaGrid staff's recommendation can
include an equitable allocation of costs to construct
the facilities and an allocation of transmission capacity
increased or maintained. Upon a majority vote by the
Columbia Grid Board, such a project, with its respective
allocations, will be included in the flllal biennial plan
which ColumbiaGrid planning parties are obligated to
uphold. The process provides a means to further address
any such disputes with the Federal Energy Regulatory
Commission.
The ColumbiaGrid coordinated planning process
will be conducted in an open and transparent manner
with Columbia Grid seeking to notifY all affected
and interested parties regarding study team activities.
Additionally, Columbia Grid will also develop a protocol
to foster the collaborative involvement of affected tribes
and states, including agencies responsible for facility
siting, utility regulation and general energy policy.
The Columbia Grid planning process will provide the
necessary coordination and dispute resolution to enable
the construction of necessary transmission facilities to
integrate needed new resources identified in Avista
2007 IRP.
MODELING TRANSMISSION COSTS
Transmission costs to integrate new resources into
the company s system were estimated by Avista
Transmission Department. Estimates were not modeled
in AURORAxmp, but rather in the proprietary PRiSM
model that matches different generating resources with
company-specific resource requirements. Construction
quality estimates have not been completed for any of the
transmission alternatives included in this IRP; estimates
are based on engineering judgment only. There is an
inverse relationship between transmission project size
and the certainty of the estimates. A 50 MW resource
can be integrated in many places on the system. A
400 MW plant can be integrated at some locations
while a 750 MW or 1 000 MW plant has very limited
placement options. A detailed regional process would
probably be undertaken to determine the precise impacts
and integration costs before an actual plant placement
decision would be made.
The Estimated Resource Integration Costs for the 2007
IRP study evaluated 50 MW, 100 MW, 250 MW and
greater than 400 MW generation sizes at 23 different
locations. The study was indifferent to the generation
asset fuel type. Wind projects have a low capacity factor
in the 30-40 percent range, but still require transmission
that corresponds to the nameplate capacity of the project.
This is the same transmission requirement as a natural
gas-fired turbine or any other resource type. The study
was divided into 10 generic project areas located outside
of the company s service territory and nine major areas
within the company s service territory. Areas located
within Avista s service area tend to be higher quality
estimates because of the increased level of system
knowledge.
Avista Corp 5 - 52007 Electric IRP
Chapter 5- Transmission Planning
ESTIMATED RESOURCE INTEGRATION COSTS FOR THE
2007 IRP STUDY
The following sections provide an overview of the Avista
Estimated Resource Integration Costs for the 2007 IRP
Study. A copy of the complete study may be found at
the company s IRPWebsite (www.avistautilities.com).
Several different project sizes were requested for this
work has been done for the alternatives within our
system because detailed machine parameters are only
available when an actual project is specified. In regard to
neighboring system impacts, an approximate worst case
cost estimate has been assigned to these resources based
on engineering judgment. Interconnection costs are
listed for locations within the Avista transmission system.
analysis. Because transmission capability comes in
lumps " and plant sizes may be altered based upon
available transmission capacity at a particular site, the
alternatives were broken into 50 100 400 750 and 1 000
MW sizes.
Integration points were roughly divided into points that
are inside and outside of Avista s transmission system.
There is some overlap for larger amounts of generation
which could have broad impacts to our system as well
as neighboring systems. A rigorous study has not been
completed for any of the foreign system alternatives
because it is impossible to provide meaningful study
results without the knowledge, input and approval
of the owners of those systems. Only limited study
All internal cost estimates are in 2015 dollars and are
based on engineering judgment with a 50 percent error
band. Time to construct is defined from the beginning
of the permitting process to when the line is energized.
An illustration of various northwest transmission upgrade
projects is shown in Figure 5.
External to the Avista System
For areas outside of Avista s transmission system
Avista-LSE would be required to undertake a
transmission request on the BPA or another transmission
system. This work would be required to determine
integration costs and wheeling service to deliver the
energy to the Avista load area. Preliminary construction
estimates are $1.4 million per mile of new 500 kV lines.
2007 Electric IRP Avista Corp
Chapter 5- Transmission Planning
Boardman, Oregon
The present transmission system serving the
Boardman generating complex consists of two 500 kV
circuits which are owned and operated by Portland
General Electric (PGE). The PGE circuits integrate into
several 500 kV circuits owned and operated by the
Bonneville Power Administration (BPA). Boardman
is located to the north and east of several transmission
constraints, which could be an issue with BPA's
transmission pricing and availability policies.
Integrating 400, 750 or 1 000 MW at Boardman would
likely require reinforcement ofPGE's and BPA's
local 500 kV system and might require additional 500 kV
facilities downstream of the plant.
John Day, Washington
The transmission system serving the John Day
generating complex consists of several 500 kV circuits
which are owned and operated by BPA. John Day is
located northeast of several transmission constraints
which could be an issue with respect to BPA's
transmission pricing and availability policies.
The North of John Day Path is constrained
depending upon generation on the upper and mid-
Columbia River. Because of the existing constraints, a
transmission integration study on the BPA system would
be required to determine if 50 to 100 MW could be
integrated at a low cost.
Kalama, Washington
The transmission system serving the Kalama area consists
of two 500 kV and two 230 kV circuits owned and
operated by BPA. This area is located in the center
of several transmission constraints which could be an
issue with BPA's transmission pricing and availability
policies. Integrating 400 MW would most likely require
reinforcement to BPA's local 500 kV system and might
require additional 500 kV facilities "downstream
of the plant. Integrating 750 or 1 000 MW would
require reinforcement to BPA's local 500 kV grid and
additional 500 kV facilities downstream of the plant.
Preliminary construction estimates are $1.4 million for
each mile of new 500 kV line. Because the amount
of new transmission will be unknown until studies are
completed, total integration costs are not known. Costs
for this alternative could easily exceed $1.5 billion.
LaGrande, Oregon
The transmission system serving the LaGrande area
consists of a 230 kV BPA line terminating at McNary
and a 230 kV Idaho Power Company (IPC) line, which
terminates at Brownlee. IPC also owns a 69 kV line
out of LaG ran de which is normally operated in a radial
configuration. LaGrande lies in the center of one of
the four lines which make up the Idaho to Northwest
transmission path (the Brownlee-McNary 230 kV line).
There is presently a WECC rating process that is being
undertaken for the Idaho to Northwest path which
could affect available capacity on these lines. Because of
the rating study, there is no way to perform a reasonable
study for the 50 to 100 MW of additional generation in
this area until that study has been resolved.
Northeast Wyoming
The transmission system serving northeastern Wyoming
consists of several 230 kV circuits, which are owned and
operated by PacifiCorp and Black Hills Power Company.
Additional circuits are owned or planned by Basin
Electric. Northeast Wyoming is presently surrounded by
several transmission constraints.
Moving between 400 and 1 000 MW from this area into
our native system would be difficult, time consuming
and most likely expensive because of all of the constraints
surrounding this area. In the lowest power and lowest
cost case at least one 500 kV line would be required
into the IPC system. In the 1 000 MW case, two
500 kV lines might be required. Depending upon
Avista Corp 5 - 72007 Electric IRP
Chapter 5- Transmission Planning
the arrangements, wheeling expense might also be
incurred. Because the amount of new transmission will
not be known until studies on the area are completed
total integration costs are presently unknown, but are
estimated to be $2.0 to $3.0 billion.
Southeast Idaho
The transmission system serving southeastern Idaho
consists of a 500 kV line, several 345 kV lines, and several
230 kV circuits which are owned and operated by
PacifiCorp and IPc. Southeastern Idaho is east and west
of several transmission constraints. Because Avista owns
no transmission in southeastern Idaho Avista-LSE would
be required to undertake a transmission request on either
the PacifiCorp or IPC systems in the area. This work
would be required to determine integration costs and
wheeling service to deliver energy to the Avista load area.
Because there are constraints from this area to the east
and west, moving 400 to 1 000 MW from this area into
our native system would be difficult, time consuming
and expensive from a construction standpoint. In the
lowest power, lowest cost case at least one additional 345
kV line would be required into the center of the IPC
system. In the 1 000 MW case, two 500 kV lines might
be required to connect the Avista system. Wheeling
expense might also be incurred. Because the amount of
new transmission will not be known until studies on the
area are completed, total integration costs are presently
unknown, but are estimated to be $1.0 to $3.0 billion.
Central Alberta, Canada
There is currently no available transfer capability or
suitable method of inexpensively integrating energy from
central Alberta into the Avista system. Because of the
distances and costs involved, integration into the United
States power grid at capacity levels less than 2 000 to
000 MW is unlikely. Transmission from central Alberta
would probably be a direct current (DC) 500 kV line
because of the capacity required for the economics of
the project. It is assumed that one of the DC terminals
would be either in the Spokane area or at the Mid-
Columbia. Avista could purchase portions of this energy
to be delivered to its system from either location. A
regional scoping effort to estimate costs for this and
similar projects has been completed and may be obtained
from the Northwest Power Pool, assuming that the
Critical Infrastructure Information requirements are met.
Estimates for these projects are $2.0 to $5.0 billion.
A 300 MW transmission interconnection project
between southern Alberta and northern Montana
(MATL) has been proposed. Available capacity on this
project is unknown at this time. However, additional
transmission would be required between central
Alberta and southern Alberta, as well as from northern
Montana to the Spokane area. Until it is known if
the MATL project will be constructed, it is difficult
to provide estimates on whether 50 MW of energy
can be economically integrated into our system from
central Alberta. Avista-LSE would need to undertake a
transmission request on the BPA system to determine
integration costs and wheeling service to deliver the
energy to the Avista load area.
Integrating anything over 300 MW would probably
require a high voltage DC tie directly from the resource
which would most likely be integrated into the Mid-
Columbia area. Integration of more than 400 MW from
the Mid-Columbia could cost $300 to $500 million
exclusive of the 500 kV DC tie project.
Central Washington
The transmission system serving central Washington
consists of multiple 500 kV and 230 kV circuits that are
owned and operated by several entities. One 230 kV
line into the Mid-Columbia area is owned by Avista and
PacifiCorp. Presently there is no long term available
transfer capability from central Washington into the
Avista system via the jointly owned transmission line.
There is a regional study, through the Northwest Power
5 -Avista Corp2007 Electric IRP
Chapter 5- Transmission Planning
Pool in progress, analyzing resource integration in the
Mid-Columbia area (including Avista s system). This
study should be completed in 2007.
The mid-Columbia area is presently in a constrained
state, depending upon generation on the mid-Columbia
River. Because of existing constraints, a transmission
integration study (most likely on the BPA or Avista
system) would be required to determine if 50 to 1 000
MW could be integrated. Integrating more than 400
MW from the Mid-Columbia would be expected to cost
$300 to $500 million.
Eastern Montana
The present transmission system to the west of (and
serving) the present generation in Montana is a double
circuit 500 kV line and two 230 kV lines. In a regional
study, under the auspices of the Northwest Power
Pool (NWPP), NTAC indicated that either additional
transmission or upgrades would be required to integrate
energy from Montana. Eastern Montana also lies east of
several transmission constraints, which could be an issue
with BPA's transmission pricing and availability policies.
A more detailed study effort focusing on constraints
from central and eastern Montana will be released in
2007. This study will identify integration constraints
and costs. Avista-LSE would need to undertake a
transmission request on the NWE system and fund a
study to determine potential impacts on the BPA system.
This work would be required to determine integration
costs and wheeling service to deliver energy to the
Avista load area. Since two transmission systems (BPA
and Northwestern Energy) may be involved in the
integration of this project, the merchant may pay two
wheeling charges for transmission service.
Walla Walla, Washington
The transmission system serving the Walla Walla area is
a single 230 kV line owned by Avista and PacifiCorp.
There is also a 115 kV line owned by BPA and a 69
kV line owned by PacifiCorp. Avista has contractual
transmission rights, but owns no transmission in the Walla
Walla area. Therefore Avista-LSE would be required
to undertake a transmission request on the PacifiCorp
transmission system. This work would be required to
determine integration costs and wheeling service to
deliver the energy to the Avista load area. Due to the
presently constrained paths in the area, such as the Idaho
to Northwest path, a transmission integration study
the PacifiCorp system would be required to determine
integration costs.
INTEGRATION WITHIN THE AVISTA TRANSMISSION
SYSTEM
Table 5.1 provides a summary view of the estimated
integration costs the company would expect for
various resources connected to its transmission system.
Discussions of each interconnection area follow.
N/A N/A $58 $80+
$32 to $500
N/A N/A N/A
N/A N/A N/A
$32 $32 N/A N/A
$13 N/A N/A
$1.N/A N/A
$1.N/A N/A N/A
$1.N/A N/A N/A
N/A N/A $58 $80+
Avista Corp 2007 Eiectric IRP 5 - 9
Chapter 5- Transmission Planning
Sprague, Washington
The transmission system serving the Sprague area is a
low capacity 115 kV line. It is not suited for integrating
250 to 400 MW in its present configuration. Each
connection below (which are the major transmission
interconnection points in the area), would require 230
kV transmission and substation work for the generation
integration. Any added generation greater than 400 MW
will increase costs and have regional impacts.
To integrate 250 MW at Westside, the existing 115 kV
line would have to be rebuilt as 230/115 double circuit
back to the main BPA corridor. An additional 230 kV
line could be constructed utilizing BPA's transmission
corridor or by building a new 230 kV line. This project
would take approximately four years and $58 million to
construct.
To integrate 250 MW at Rosalia on the Benewah-
Shawnee 230 kV line, 30 miles of new 230 kV line
would have to be constructed to Rosalia and a 230 kV
switching station would need to be built. This project
would take about four years and $35 million to complete.
To integrate 400 MW at Westside, the existing 115 kV
would have to be rebuilt as a 230/115 kV double circuit
back to the main BPA corridor. To connect at Westside
an additional 230 kV line would need to be constructed
utilizing BPA's transmission corridor or by building a
new 230 kV line. This project would cost approximately
$80 million and take four years to complete.
In order to integrate 400 MW at Rosalia on the
Benewah-Shawnee 230 kV line, a new 30-mile long 230
kV line would have to be constructed to Rosalia and a
230 kV switching station would also have to be built.
This project would take four years and approximately
$50 million to complete.
Spokane/Coeur d'Alene
There are a number of 230 kV stations and transmission
lines in the Spokane/Coeur d'Alene area that would
make good generation interconnection points. Westside
Beacon, Bell, Boulder and Rathdrum are all large stations
with 230/115 leV transformation in the Spokane/Coeur
d' Alene area. However, integrating large generation
in this area could pose thermal loading problems on
the underlying 115 kV system. Without a specific
interconnect point, all of the needed 115 kV work is an
approximation. The Spokane/Coeur d'Alene area covers
too much land to be more specific on costs. Additional
generation greater than 250 MW will further increase
costs and regional impacts.
Integrating 50 MW of new generation in the Spokane/
Coeur d'Alene area can be done with 10 or less miles of
115 kV reconductor work. This type ofproject would
take approximately one year and $3 million to complete.
100 MW could be integrated into this area with less
than 30 miles of 115 kV line reinforcement. This type of
project would take approximately two years and $7
million to complete.
Integrating more than 250 MW of generation in the
Spokane/Coeur d'Alene area would require 230 kV
work. This would necessitate extensive levels of 115 kV
reconductoring. The radial operation ofAvista s 115
kV lines in Spokane and Coeur d' Alene or generation
dropping for 230 kV outages would probably be needed.
Additional 230 kV work would likely be needed
depending on the interconnection point. This project
could cost $32 to $500 million and take five years to
complete.
Mica Peak
Mica Peak is near existing Avista 115 kV lines with
available capacity. 50 MW could be integrated at the
5 -2007 Electric IRP Avlsta Corp
Chapter 5- Transmission Planning
Post Falls substation with six miles of 115 kV line and a
new breaker position at Post Falls. This project would
cost about $4 million and take one year to complete.
Clark Fork Hydro Upgrades
The present transmission system in the Clark Fork
area consists of both Avista and BPA 230kV lines that
integrate the western Montana hydro (WMH) projects.
The WMH refers to the four major hydroelectric plants
operated in northwestern Montana and on the northern
Montana-Idaho border. These include the federally-
operated Libby and Hungry Horse projects and Avista
Cabinet Gorge and Noxon Rapids (Clark Fork) projects.
After completion of planned upgrades to Cabinet
Gorge and Noxon Rapids, these projects will have
peak generation capacities of268 MW and 558 Mw,
respectively, for a combined capacity of 826 MW
Avista and BPA have a WMH operating agreement that
provides a 50-50 allocation of a 1 700 MW operating
limit between the federal and Avista projects. This
agreement pertains to Avista-LSE's ability to operate its
Clark Fork Projects for service to Avista s bundled retail
native load customers. Mter completion of upgrades
Avista s total Clark Fork hydro generation capacity will
be 24 MW below Avista s WMH operational allocation
of 850 MW Dependent upon continuation of the
operational allocation ofWMH hydro capability between
Avista and BPA, no new transmission upgrades will be
needed for Avista to integrate the planned upgrades of its
Clark Fork hydro projects.
Dayton, Washington
The present transmission system serving the Dayton
Wash., area is a single 230 kV line with dual ownership
by Avista and PacifiCorp. There is also a 115 kV line
in the area owned by BPA and a 69 kV line owned by
PacifiCorp.
Fifty to 100 MW could be integrated on the Dry
Creek-Walla Walla 230 kV line at the ownership change
between Avista and PacifiCorp with a new switching
station and a 15 mile 230 kV line to this location. This
line lacks capacity to support 50 to 100 MW due to
current contractual obligations. Therefore, the Dry
Creek-Walla Walla 230 kV line would need to be re-
conductored to support additional capacity. The project
would take approximately four years and $32 million to
complete. There may be a potential real time solution
using real time thermal monitoring and the Valley
Group s Cat-lor similar technology.
Reardan, Washington
The present transmission system serving the Reardan
Wash. area is a low capacity 115 kV line. Fifty MW
could be integrated at the Reardan substation by re-
conductoring the 115kV line from Garden Springs
to Sunset along with a new air switch at Westside
on the Nine Mile line. This project would require
approximately one year of construction time and
cost about $2 million. One hundred MW could
be integrated by re-conductoring the 115 kV line
from Reardan to Devils Gap along with a new line
out ofReardan. The 100 MW project would cost
approximately $13 million and take two years to
complete.
Lind, Washington
The transmission system serving the Lind area is a low
capacity 115 kV line and two 115 kV lines that are
operated in a radial configuration. Very little new
transmission would be required to integrate 50 MW at
the Lind substation. The project would take about one
year and $1.5 million to complete. Integrating 100 MW
would require re-conductoring the 115kV line from
Lind to Warden. The project would take about one year
and $6 million to complete.
Avista Corp 2007 Electric IRP 5 - 11
Chapter 5- Transmission Planning
Othello, Washington
The transmission system serving the Othello, Wash, area
consists oflow capacity 115 kV lines. Fifty MW could
be integrated at the Othello substation with very little
new transmission. The project would take about one
year to complete at a cost of$1.5 million.
Colfax, Washington
The present transmission system serving the Colfax
Wash., area is a low capacity 115 kV line. Fifty
could be integrated at the East Colfax substation with
very little new transmission being required. The project
would cost about $1.5 million and take approximately
one year to finish.
5 -2007 Electric IRP Avista Corp
Chapter 6- Modeling Approach
MODELING APPROACH
Transformer at Coyote Springs 2
INTRODUCTION
This section discusses market modeling assumptions
used to value each resource option and the combination
of costs and benefits to select the Preferred Resource
Strategy (FRS). The analytical foundation for the 2007
IRP is a fundamentals-based electricity model of the
entire Western Interconnect (WI). Understanding
market conditions in the different geographic areas
of the WI is important because many areas are linked
by transmission facilities and the regional markets are
correlated.
Avista s IRPs prior to 2003 relied on externally generated
market price forecasts that did not consider company
operations. This IRP builds on prior analytical work
by maintaining the link between the WI market and
the changing value of company-owned and contracted
resources. The company s portfolio value is linked to
its loads, resources and contractual arrangements, both
for existing and prospective resource options, and for
meeting future obligations.
The Preferred Resource Strategy is developed using a
multi-step approach. New and existing resources are
combined to simulate hourly operations for the WI to
develop a long-term hourly electricity market price
forecast. This market forecast values each resource
option Avista might select as part of its FRS. Figure 6.
illustrates the company s IRP modeling process.
MARKET MODELING
AURORAxmp is a fundamentals-based electricity
market forecasting tool that tracks the value of the
company s existing resource portfolio as well as potential
new resource portfolios. Additional details about
AURORAxmp can be found in Technical Advisory
Committee presentations at the company s IRP Website.
AURORAxmp is used to simulate the WI for this IRP
The WI includes the states west of the Rocky Mountains
the Canadian provinces of British Columbia and Alberta
and the Baja region of Mexico, as shown in Figure 6.
The WI is separated from the Eastern Interconnect and
ERCOT systems, with the exception of eight inverter
stations. The WI follows operation and reliability
guidelines administered by the Western Electricity
Coordinating Council (WECC).
CHAPTER HIGHLIGHTS
. AURORAxmp is used to model hourly operations for the entire Western Interconnect.
. The company performed 300 iterations of Monte Carlo market analysis with varying wind, hydro, load, natural
gas prices, emissions and thermal outages for each evaluated future.
. The Preferred Resource Strategy was developed using the proprietary Avista Preferred Resource Strategy
Model (PRiSM).
. This IRP considers generation, transmission and emissions costs.
Avista Corp 6 - 12007 Electric IRP
Chapter 6- Modeling Approach
Update Avista s resource
parameters based on proprietary
information; replace any existing
AURORAxmp assumptions
desired.
Scenarior------------------
Examine scenarios in
AURORAxmp if entire west
is affected (re-run
capacity expansion if
necessary) or externally if
scenario is "Avista-only
Calculate
resource valuation
of scenario
-----------
Analyze results from
scenarios and futures.
Figure 6.1: Modeling Process Diagram
Riskr-------- ------- --- ----
Develop stochastics
Base Case Load,fuel price,hydro,wind
Use stochastics to calculate a single generation,emissions thermal
forced outages.average" set of input data
Calculate resource valuation of
each future
Market and cost values of each new and existing
resource for each scenario and future
Base Case
assumptions Avista load requirements
(capacity and energy)
6 - 2 Avista Corp
Capital costs associated with
new resources, including
locational transmission costsII'"Preferred Resource Strategy
2007 Eiectric IRP
Chapter 6- Modeling Approach
Figure 6,2: NERC Interconnection Map
- - - - -, '
FRCC
/ / , ,
Eastern
' ,
InterconnectionERCOT ,
Interconnection
' ,
. e- '
. : .
0'0' s a.o. ne. s
Rocky Desert
Northwest California Mountain Southwest Independent
W. Wash.Northern Wyoming Arizona British Columbia
W. Oregon Central Colorado New Mexico Alberta
E. Wash.South Utah S. Nevada E. Montana
C. OreQon Baja N. Nevada S. Idaho
W. Montana
WesternInterconnection
T bl 6 1 AURORAx
The model separates the WI into 20 zones based on load
concentrations and transmission constraints. Zones are
grouped into pools for regional capacity planning. The
pools do not reflect regional transmission agreements or
reserve sharing but are designed for regional proximity
of resources. Table 6.1 shows the geographic pools and
zones modeled in the IRP. Some zones are modeled
independently due to significant transmission constraints
and/or international boundaries.
Electric models range in their ability to emulate power
systems. Some models account for every bus and
transmission line; others utilize regions or zones. An IRP
requires regional price and plant dispatch information.
Table 6.1 provides a list of zones contained in each pool.
The Northwest is modeled as five separate zones. This
differs from the 2005 IRP where the Northwest was
modeled as a single zone. Montana is split into east and
west load areas to reflect transmission constraints on the
Northwestern system. AURORAxmp has the ability to
model the Northwest as nine separate zones. The nine-
area topology was not selected because oflong solution
times and because the five-area topography was found to
better represents Northwest market operations.
KEY ASSUMPTIONS AND INPUTS
HYDROELECTRIC GENERATION
The Northwest and British Columbia have substantial
hydro generation capacity. A favorable characteristic of
hydro power is the ability to provide short periods of
1 Baja, Mexico, is included in the California pool because of tight interconnection with Southern California. This zone could have been
modeled as an independent zone, but it has no impact on Avistas resource strategy or the Northwest s electricity marketplace.
Avista Corp 2007 Electric IRP 6 - 3
Chapter 6-- Modeling Approach
near-instantaneous generation. This characteristic is
particularly valuable for meeting peak load demands
shaping load and selling surplus energy during peak
hours. A drawback of hydro is the potential lack of
energy, since hydro is constrained by weather patterns
and subsequent stream flows. The amount of energy
available at a particular plant depends on its location and
characteristics of its river system.
This IRP relies on information provided by the
Northwest Power Pool (NWPP) to model regional
hydro resources. The NWPP maintains a hydrological
model providing energy amounts that each hydroelectric
plant could produce from 1928 to 1999. This plan uses
the 2004-05 Headwater Benefits Study. To accurately
model British Columbian hydro projects, historical
generation data from the Canadian Government was
blended with the NWPP data set.
Many of the analyses in this IRP use an average of
the 70-year record; stochastic studies randomly draw
from the 70-year record (see stochastic modeling).
Hydroelectric plants are lumped into geographic regions
and represented as a single plant in each zone. The
company models its Clark Fork, Spokane and Mid-
Columbia projects to extract greater detail for portfolio
modeling.
AURORAxmp represents hydro plants using annual
and montWy information regarding energy generating
capabilities, minimum and maximum generation levels
and abilities to sustain peak generating levels. The
model's objective, subject to the constraints, is to move
hydro generation into peak hours to follow daily load
increases. This maximizes the value of the hydro system
in a manner that approximates actual operations.
FUEL PRICES
The IRP uses fuel price assumptions in the most
up-to-date EPIS database, with the exception of natural
gas and coal prices. The price of fuel is the single most
important modeling assumption in AURORAxmp.
Natural gas sets the market price of power in the
Northwest about three-quarters of the year and in
more hours in other areas of the WI. Coal generally
sets market prices during the spring when significant
hydroelectric generation pushes natural gas-fired plants
off of the margin.
NATURAL GAS PRICES
Avista retains several consultants who specialize in
developing long- and short-term, fundamentals-based
natural gas price forecasts. The company also reviews
the Energy Information Association s Annual Energy
Outlook (AEO) and monitors and participates in the
New York Mercantile Exchange (NYMEX) forward
natural gas price market. Each of these price curves uses
different assumptions and provides the company with
additional data about natural gas pricing.
A multitude of factors were considered before choosing
a price forecast. These factors included assumptions
for economic growth, natural gas production levels
new infrastructure (i.e. Mackenzie Delta and Alaskan
Pipelines), Canadian imports and demand (i.e. residential
commercial, industrial and electric generation). In
particular, the selected consultant's forecast included
more reasonable electric generation demand, liquid
natural gas (LNG) imports, and overall natural gas supply
and demand balance assumptions than the other price
forecasts.
The natural gas price forecast provides annual average
prices per decatherm at the Henry Hub basin in
Louisiana. Annual average prices are converted into
a series of montWy values before being entered into
AURORAxmp. The montWy shape is based on
NYMEX forward prices, which is consistent with Avista
2006 Natural Gas IRP. Table 6.2 presents seasonal natural
gas price factors. MontWy price shapes are derived by
Avista Corp2007 Electric IRP
Chapter 6- Modeling Approach
Table 6.2: Seasonal Natural Gas Price Factors
Percent of Percent ofMonth Annual Month Annual
Janua
Februa
March
A ril
June
113
113
110
Jul
Au ust
Se tember
October
November
December
101
106
Figure 6.3: Henry Hub Natural Gas Forecast ($/Dth)
- -
. Nominal Dollars -
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
. 2007 Dollars
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
C')
"'"
II)
......
applying these percentages to annual average prices. This
approach reasonably reflects the actual seasonal weighting
in the natural gas market.
The natural gas price forecast blended the January 3
2007, NYMEX forward price with the consultant's price
forecast. Blending the two prices acknowledges that
the forward market is the price which can be currently
purchased and that forward and fundamental prices
should converge in the long-run. The weighting of the
NYMEX forward price begins at 50 percent in 2008 and
is decreased by 10 percent annually through 2012. The
Henry Hub price forecast is shown in Figure 6.
Avista has historically used monthly natural gas prices
in its IRP forecasts, but natural gas prices vary daily.
This IRP is our first to include a daily adjustment from
the monthly price forecast. Daily prices are calculated
I'-C')II)
......
using 2003 to 2006 historical prices to determine a daily
percent change from the monthly average price. This
percentage is applied to the monthly price. Figure 6.4
illustrates the variability of daily natural gas prices around
the monthly averages.
The final component of a natural gas price forecast is
development of basis differentials from Henry Hub.
Henry Hub is a trading point in Louisiana on the Gulf of
Mexico, widely recognized as the most important natural
gas pricing point in the United States. Henry Hub holds
this distinction because of its spot and forward market
trading volumes and its proximity to a large portion
u.s. natural gas production. NYMEX uses Henry Hub
as a trading hub for futures contracts. All other
production and market pricing points can be traded with
a "basis differential" on the Henry Hub. The Western
u.S. does not rely on Henry Hub for its physical gas
Avista Corp 2007 Electric IRP
Chapter 6- Modeling Approach
Figure 6.4: Daily Natural Gas Prices Shape ($/Dth)
...,- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -..........
0::(
..c
...,...,
C/)0::(
83.1 86.1 85.1 88.3 86.9 89.
deliveries. Instead it relies on physical supply points
including AECO in Alberta, Canada, and the u.S.
Rockies. Market trading hubs include Sumas Wash.
Malin, are.; Stanfield, are.; and Topock, Calif. Natural
gas at these supply points typically trade at a significant
discount to Henry Hub. This discount is commonly
referred to as the basis differential. Basis differentials
exist because of a more favorable supply/demand balance
in the West, closer physical proximity to these supplies
and longer distances from the large natural gas demand
centers of the Eastern u.S.
Most natural gas price forecasts do not include
Northwest or Western u.s. pricing, so Avista estimates
the basis differential between Henry Hub and the pricing
points the company uses to fuel both its power plants
and other plants across the Western Interconnect. The
company uses an average of recent basis differentials
to estimate price differences between the Henry Hub
forecast and these markets. The company has adopted
the percentages shown in Table 6., consistent with its
2006 Natural Gas IRP.
2 http://www.eia.doe.gov/cneaf/coal/ctrdb/tabSS.htrnl
COAL PRICES
Coal prices and coal transportation costs in this IRP
rely on data provided by the Energy Information
Administration (EIA) in its February 2006 fuels
forecast and its 2002 transportation cost study.2 The
IRP coal price for new coal-fired generation is based
on the forecast ofWestern mine mouth coal prices.
Transportation costs are added based on an assumed plant
distance from its source of coal supply. This plan assumes
three representative coal plant delivery distances for all
plants: mine mouth, short haul (500 miles) and long haul
200 miles). Figure 6.5 shows the coal price forecast
for new coal-fired resources options in the 2007 IRP.
AURORAxmp contains coal price assumptions for
existing coal-fired plants based on existing contracts.
However, some plants also rely on market-based coal.
These contracts are tied to the 2007 IRP coal price
forecast.
EMISSIONS
Environmental factors are an increasingly important
part of resource planning. Emission charges are used
2007 Electric iRP Avista Corp
Chapter 6- Modeling Approach
Figure 5: Coal Prices for New Coal Resources ($/Ton)
~Long Haul (1000+ Miles)
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
........ Short Haul (250 Miles)
Mine Mouth (PRB)
- - - -- -- - - - - - - - - - - - - - - -- - - - - - - - -- - - - -- - - -- -- -- - -- - - -- -'. .. . -. - - - - - - - .. .. . - -- - -;.- -. . ~ .
111 . 8 1111
C")
...
It)
.........
to encourage more environmentally-friendly resource
options. The charge is calculated by estimating the
financial penalty needed on certain types of emissions
to accomplish a stated goal, such as reducing carbon
emissions to 1990 levels. In the 2007 IRp, emissions
charges are assigned to all resources to model the
opportunity cost of generating and producing emissions
or choosing not to generate and selling the right to
produce emissions. This methodology implies that a cap-
and-trade system is in place to trade emissions credits.
Additional emissions discussions are located in Chapter 4.
r--It)r--C")
using the National Commission on Energy Policy
(NCEP) carbon regulation proposal. There is currently
a great deal of state and federal level legislation
regarding carbon emissions which could significantly
impact power prices. The uncertain state of carbon
emissions legislation requires additional analysis to
better understand the issues. This analysis is described in
Chapter 7.
The remaining three emissions charges are estimated
by a third-party consultant. Figure 6.6 shows the Base
Case emission price forecasts. Emissions charges are
set to a level necessary to cause existing plants to install
mitigation equipment to reduce their average emissions
The IRP tracks four emission types: carbon dioxide
(CO ), sulfur dioxide (SO), nitrous oxygen compounds
(NO )' and mercury (Hg). CO2 charges are estimated
Figure 6: Emission Charges Summary
000
000 ........ S02 $ per Ton
~NOx $ per Ton
Hg $ per Ounce
-+-C02 $ per 100 Tons
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
r--
...
C")It)r--C")It)r--
............
2007 Electric IRP
000
000
000
000
000
000
-------------- -- -- :--
Avista Corp
- - - - - - - - - - ----- - - -- -- - -- - - - - - - - - - - - - - - - - - - -
6 - 7
Chapter 6- Modeling Approach
below certain thresholds. Emissions generally do not
have a significant impact on electric market prices in
Western U.S. markets because gas-fired plants usually
set the marginal price of power. These plants have low
overall emission profiles, with the exception of CO
RESOURCES
The AURORAxmp model is populated with all
current power generation resources and the operating
characteristics important for modeling electricity
markets (e., plant capacity, heat rate, and start-up costs).
Resources under construction or otherwise expected
to generate power in the future are also modeled.
The AURORAxmp vendor has a rigorous plant data
collection methodology that makes certain assumptions
for each plant. The company has maintained many of
these assumptions for the IRP model database but has
made various changes where the company has access
to better information. Resources not currently under
construction, or a part of other companies' IRPs or
plans, are modeled indirectly by two methods. The first
method adds resources to meet future load growth for
the West by using expansion logic in AURORAxmp; the
second method adds generation needed to meet active
or impending renewable portfolio standards (RPS). For
example, Washington Initiative 937 requires all utilities
with more than 25 000 customers to serve 15 percent of
their 2020 load with new renewable resources.
The AURORAxmp expansion logic used for this plan
differs from the 2005 IRP. The 2005 plan built a level
of generation across the West to meet the energy needs
of the gross system. The 2007 IRP relies on a capacity
planning target. In general, utilities build resources to
cover adverse load conditions, meaning that resources
are constructed to exceed average needs. This ensures
that adequate resources are available to meet system
requirements in all but the most extreme conditions
driving electric market prices and volatility down. The
937 has earlier targets of3 percent in 2012 and 9 percent in 2016.
availability of firm resources to meet retail loads under
a broad range of operating conditions reduces exposure
to significant purchases of energy from the financially
volatile short-term wholesale energy market.
The resources available to meet regional load growth
are: combined-cycle combustion turbines (CCCTs),
single-cycle combustion turbines (SCCTs), pulverized
coal, integrated gasification combine-cycle (IGCC) coal
IGCC coal with sequestration (certain scenarios) and
wind turbines. Other small renewable resource options
are added using the RPS method discussed in the next
paragraph. New resource options are limited depending
on regional location and the presence of an active
RPS in the region. For example, renewable resource
construction in states with RPS requirements is limited
by their RPS; no additional renewables are constructed.
West coast states cannot rely on coal-fired plants due
to legislative mandates preventing their construction.
Detailed assumptions about these resources are discussed
later in this section. Specific details on which resources
were selected for each study are presented in Chapter 7.
New resource options affect market prices which in turn
affect the resource mix Avista will consider as it makes
investment decisions over its planning horizon.
Renewable portfolio standards change the mix of
resources utilities choose to build. Historically utilities
built resources with the lowest expected future cost and
rate volatility. RPS requirements and other legislative
mandates have changed this approach. Utilities must
build a specified amount of renewable resources or are
limited in their ability to construct certain resource
types. Resources procured under these circumstances
may not be the lowest cost in a traditional sense, but they
will meet a legislative mandate in one or more states
and might reduce rate volatility where free or fIXed fuel
prices and fuel supply are available. Table 6.4 shows the
incremental energy needed to meet existing renewable
2007 Electric IRP Avista Corp
Chapter 6- Modeling Approach
Table 4: New RPS Resources Added to Existin S stem (aMW)
California 187 656 106 991
Oregon 519 914 867
Washington 328 988 260
Nevada 400 684 764 900
Montana 239 271 324
Arizona 187 556 113 964
Colorado 100 606 663 757
New Mexico 177 289 326 389
requirements in the Western Interconnect. Actual
resources in each state will vary depending on how
utilities choose to meet their requirements.
These additions represent company assumptions for the
amount of renewable resources necessary to meet various
state laws. In states where RPS laws were still pending
at the time of the IRP modeling, we made our best
estimate based on draft legislation.
A difficult part of forecasting renewable resources is
determining where they will be located. Some states
require utilities to acquire resources within certain
geographic areas, which can greatly increase the price of
those projects. New regional transmission may also be
required. While recognizing that some regions will meet
their RPS requirements by importing renewable power
from other regions, the 2007 IRP assumes that all RPS
resources are added in the geographic region where they
are required. This simplifying assumption was based on
the lack of a comprehensive study of regional renewable
resource availability. The company does not believe that
this simplifying assumption has any significant impact
on the wholesale marketplace or the value of resource
options available to it.
LOADS
A load forecast is developed for the entire region to
forecast western electric prices. This IRP relies on
several external sources to quantify load growth across
the Western Interconnect. These sources include
integrated resource plans, the Western Electricity
Coordinating Council (WECC) and the Alberta Electric
System Operator (AESO). Peak regional load growth is
shown by area in Table 6.5. New resources are added to
each area to meet capacity planning margins. The 2007
Ta. b. Ie- 6.S:A'nnua,IA've-ra..e- P"kLo.a.d. Gro.wth Yo,
Area Load Area Load
Growth Growth
W. Wash.1.40 California
W. Oregon 1.40 Baja, Mexico
E. Wash.Wyoming
C. Oreqon Colorado
Montana Utah
S. Idaho Arizona
British Columbia New Mexico
Alberta Nevada
4 Southern Oregon is estimated to grow at 1.2 percent and Portland Metro Area is 2.6 percent.
5 Spokane is estimated to grow at 2 percent, other eastern Washington areas 1 percent.6 Southern Nevada peak is expected to grow at 3.2 percent, while northern Nevada is at 2.6 percent.
Avista Corp 2007 Electric IRP
Chapter 6- Modeling Approach
Ta. bile' 6. .6. : A' nnua.1 A' ve.ra. I e- Ene- r I Lo. a. d, Gro.wth o'Yo.
Area Load Area Load
Growth Growth
W. Wash.California
W. Oregon Baja, Mexico
E. Wash.Wyoming
C. OreQon Colorado
Montana Utah
S. Idaho Arizona
British Columbia 1.40 New Mexico
Alberta Nevada
IRP planning margins are assumed to be 25 percent for
the Northwest and Idaho, 17 percent for California and
10 percent for all other zones.
Peak load growth estimates are important for estimating
new capacity; however, market prices are more highly
correlated to actual energy load growth. Energy growth
estimates are shown in Table 6.
RISK MODELING
The power industry has fundamentally changed since
the 2001 energy crisis. Historically, northwest utilities
planned for variability inherent in their hydroelectric
plants and load forecast. Now northwest utilities must
consider natural gas price volatility, thermal plant forced
outages, wind speed, extra-regional load and resource
balances, and the ever changing face of emissions
legislation. This IRP utilizes a Base Case with an
underlying set of assumptions to anchor the modeling
effort. Several alternative scenarios and futures are
modeled to provide information about what could
happen in the electric market under different sets of
assumptions. All of the modeling efforts are combined
with the judgment of planners, senior management
and members of the Technical Advisory Committee to
develop a Preferred Resource Strategy used to guide
company resource acquisitions.
The Base Case for this study uses average values for most
estimates, such as hydro conditions, peak and energy
loads growth, and gas prices. These key market drivers
will probably not be average in every year, but instead
will regress to average levels over the 20-year planning
horizon. Scenarios and stochastic studies help the
company understand how the market might look and
behave if the long-term averages in the Base Case did
not materialize. This section focuses on the stochastic
assumptions for these studies. The IRP models include
several key assumptions that are modeled stochastically,
including natural gas, hydro, load, wind, forced outages
and emissions charges (SO2' NOx' Hg and CO
The 2007 IRP simulates 300 hourly iterations or
games " using the AURORAxmp for the years 2008-
2027. This level of analysis required the use of25
computers writing their results to a SQL database. Each
set of stochastic analysis took the equivalent of four days
or 2 160 computer hours, to complete. The company
prepared four stochastic futures for the IRP, consuming
500 hours of central processing unit time and creating
a 450 gigabyte SQL database.
Running the electricity model stochastically provides a
measure of volatility for forecasted electricity prices and
resource values. This measure is essential to our selection
of new resources, because the company s long-term
objective is to manage rate variability, as well as limit
customer costs.
7 Southern Oregon is estimated to grow at 1.2 percent and Portland Metro Area is 2.6 percent.8 Spokane is estimated to grow at 2 percent, other eastern Washington areas 1 percent.
2007 Electric IRP Avista Corp
Chapter 6- Modeling Approach
Table 6.7: Coefficient of Variation of Forward Sumas Natural Gas Prices
(%)
Month 2005 2006 2007 2008
January
February
March
April
Mav
June
Julv
AuQust
September
October
November
December
Figure 6.7: March 2006 Sumas Natural Gas Contact Price Distribution
C\i
0 0LO 0
o.ri to
0 0LO 0to r-:
0 0
C"S C"S
0 0 0LO 0..,f ..,f o.ri
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
15 8 r-: cO
0 0
0 0
...... ......
0 0 0LO 0
....: ....: C\i
...... ...... ......
C"S
......
0 0LO 0C\i C"S
...... ......
0 0ai
Sumas Natural Gas Price ($ per Dth)
NATURAL GAS PRICES
There are several approaches for stochastically modeling
natural gas prices, as well as a number of assumptions
that need to be made. The 2007 IRP begins with the
deterministic natural gas price forecast discussed earlier
in this chapter. The forecast represents mean prices in
each forecast period. Table 6.7 shows the coefficient of
variation (the standard deviation divided by the mean
value) of historically traded forward natural gas contracts
for the months of2005 through 2008. We believe that
forward market price volatility is a reasonable indicator
of future natural gas price volatility. The Base Case
assumes 30 percent volatility to capture projected market
risk. This assumption differs from the 2005 IRP, which
instead represented natural gas volatility with a 50
percent coefficient of variation.
The Base Case distribution is assumed to be lognormal
based on a statistical review of the forward price datasets.
A review of historical data shows that a majority of
the contracts have lognormal characteristics; Figure
7 presents the distribution of the March 2006 Sumas
forward contract. The Monte Carlo model draws a gas
price curve using the lognormal distribution, but each
Avista Corp 2007 Electric IRP 6 -
Chapter 6- Modeling Approach
~ 8
1ii
.! 6
\Ij
:- 4
20 -
Figure 6.8: 2008 Sumas Natural Gas Price (Deterministic & First 30 Draws)
~--; ~...--:--,,::-- - -- - -- ~ ------ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
(tJ
"")....
(tJ
:::2:
(tJ
:::2:u..::J
"");:."")(,)
Figure 6.9: Annual Average of 300 Iterations of Sumas Natural Gas Prices ($/Dth)
Mean
- Max & Min
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
. 80% Confidence Interval
- -- - - -- ---------------------------------- --- ---- --- ------ -- -- -
C')'O;f'
......
r--
......
C')
............
r--
Figure 6.10: Hydro Capacity Factor and Statistics for Selected Areas (%)
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
. 50% Probability
~25% Probability
- 10% Probability
E Wash WOre MidC SID E Ore W Wash NW Wash SOre
6 -Avista Corp2007 Electric IRP
Chapter 6- Modeling Approach
B 80
I!! 60::I
Figure 6.11: Water Year Distribution
120
100
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
m N('I)m m
...... ......
It)('I)
......~ ~...... ......
It)
......
...... vt;) t;)
...... ......
f0o-
t;)
......
('I)ffi ffi
...... ......
(0 mffi ffi
......
It) com mm m
...... ......
co
......
(f,
...... ......
16 :8m m
...... ............
It)
~ ~...... ......
Water Year
Figure 6.12: Distribution of Stochastic Hydro as a Percent ofthe Mean
e... 14
:e 12
8 10
'0 6
:: 4..QIII..Q~ 2
D..
- - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - -- - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - -
~ ~ ~ 2 ~ ~ ~ ~ ~ ~ ffi 8 8 ~ ~ Q ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ......
Hydro Condition (Percent of Average)
draw has the same shape as the Base Case; the draw is
either above or below the Base Case forecast. See Figure
8 for a graphical illustration.
impacted by hydro generation. Figure 6.10 shows the
hydro capacity factors assumed for zones and sub zones
(areas) that have substantial hydro capacity.
Annual average results of this methodology are
displayed in Figure 6.9. The chart shows the expected
(deterministic) price, the mean of the 300 Monte Carlo
iterations, the 80 percent confidence interval, and
maximum and minimum prices.
To account for hydro variability, a random generator
was used to select different hydro generation amounts
for each year and for each of the 300 iterations. Hydro
available in each draw was selected from 70 historical
water years from 1928/29 to 1998/99. Figure 6.
presents a distribution of the Base Case draws. The draws
HYDROELECTRIC GENERATION show a uniform distribution, or no bias, between water
year selections.The Northwest's electricity market, as well as the
company s own resource portfolio, is significantly
Avista Corp 2007 Electric IRP
Chapter 6- Modeling Approach
Table 8: Selected Zone s Load Correlations to Eastern Wash in
Alberta Not Sig Not Sig Mix Mix Mix 3270
Arizona 3504 3505 Mix Mix 2027 0.4499
Baja Not Sig Not Sig 2109 Not Sig Mix 2171
British Columbia 7856 6762 8047 0997 1058 1089
Colorado 7852 0.4468 Mix Mix Not Sig Mix
E. Oregon 9099 8822 8893 7400 0.4262 8613
Montana 8440 5508 8588 Not Sig Not Sig 3487
N. California Not Sig Not Sig Not Sig Mix Mix Mix
N. Nevada 2456 3232 0.4272 Not Sig 1026 7609
New Mexico Not Sig Mix Mix Mix Not Sig Mix
S. California 1991 Not Sig Not Sig Mix Mix Mix
S. Idaho 6807 7163 6042 3317 2114 7373
S. Nevada 8003 3343 Not Sig Mix Mix 0968
Utah 8988 8770 8435 7345 0.4246 8451
W. Oregon 8177 5723 8781 1043 Mix 3152
W. Washington 8284 0.4689 9031 1043 Mix Mix
Wyoming 9089 9004 9300 6906 0.4186 5850
Alberta 7575 1003 Not Sig Not Sig Not Sig 0.4306
Arizona 2134 Mix Not Sig Not Sig Not Sig 0.4233
Baja 1999 3011 Mix Not Sig Mix 1100
British Columbia 6397 3084 Mix 6985 5887 8158
Colorado Not Sig Not Sig Not Sig Not Sig Not Sig 3321
E. Oregon 7343 7871 5924 8831 8324 0.4573
Montana 8310 2095 2979 8342 8199 8107
N. California 0.4874 Mix Mix Not Sig 2096 2104
N. Nevada 6583 2339 5424 Not Sig 1029 7235
New Mexico Not Sig Mix Not Sig Not Sig 1036 Not Sig
S. California 5017 1044 Mix Not Sig 2025 2284
S. Idaho 2093 6807 7406 2317 8991 0.4475
S. Nevada Not Sig Mix 3208 Not Sig 1020 6617
Utah 6201 7815 8238 8590 8515 5825
W. Oregon 8337 0.4289 0.4410 8547 5755 3413
W. Washington 8645 3171 Mix 8724 8854 0.4803
Wyoming 5902 3100 6721 8919 8685 3487
The historical water record's distribution is shown in methodology developed for the 2003 IRP. The earlier
Figure 6.12. Generation is shown as a percent of the work developed montWy and weekly distributions of
mean for the entire Northwest, encompassing British hourly load data for each Western Interconnect utility
Columbia, Washington, Oregon, Idaho and Montana.using FERC Form 714 data. The 2007 IRP updates
the 2003 data, using FERC Form 714 data for the years
LOAD VARIABILITY 2002-2005. Correlations between the Northwest and
The 2007 IRP relies on Western Interconnect-wide other Western Interconnect load areas were calculated
2007 Electric IRP Avista Corp
Chapter 6- Modeling Approach
Alberta
Arizona 11.
Baja 10.9.4 10.
British Columbia 5.4
Colorado 5.4
S. Idaho
LADWP 10.
Montana
W. Montana
New Mexico
N. Nevada
S. Nevada 6.4 13.8.4
E. Washington 5.4
W. Washington
E. Oregon 5.4 6.4
W. Oregon 7.4
N. California 5.4
S. California 10.
Utah 5.4
Wyoming 6.4 6.4
C. California 5.4
Alberta 2.4
Arizona 10.
Baja 6.4 10.
British Columbia 4.4
Colorado
S. Idaho 5.4
LADWP
Montana 6.4
W. Montana 6.4
New Mexico 6.4
N. Nevada
S. Nevada 10.
E. Washington
W. Washington
E. Oregon
W. Oregon 8.4
N. California
S. California
Utah
Wyoming
C. California
Avista Corp 2007 Eiectric iRP 6 -
Chapter 6- Modeling Approach
and represented in the stochastic load model.
Correlating zone loads avoids oversimplifying the
Western Interconnect load picture. Absent correlation
data, stochastic models would offset load changes in one
zone with load changes in another, thereby virtually
eliminating the possibility of modeling West-wide load
excursIOns. Given the high degree of interdependency
model is necessary to correctly determine its impacts on
the overall market as well as the value of any acquisition.
Accurately modeling a wind resource requires hourly
generation shapes. For regional analyses, wind variability
is modeled in a manner similar to how AURORAxmp
models hydroelectric resources. A single wind plant
and generation shape is developed for each area. This
- -
Figure 6,13: August Hourly Wind Generation Distribution
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
~ 12t...
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
~ 10
D.. 6
;3 4
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
43 48 53 58
Capacity Factor (%)
across the Western Interconnect (e., the Northwest
and California), this additional accuracy is crucial for
understanding variation in wholesale electricity market
pnces.
Tables 6.8 and 6.9 illustrate the correlations used for the
2007 IRP. Tables 6.10 and 6.11 provide the coefficient
of variation (standard deviation devided by the mean)
for each zone in 2007. "NotSig" indicates that no
statistically valid correlation was found in the evaluated
data. "Mix" indicates that the relationship was not
consistent across time and was not used in the 2007 IRP
analysis.
WIND GENERATION
Wind is one of the most volatile energy resources
available to utilities. Since storage, apart from some
integration with hydro, is not a financially viable option
capturing the resource s volatility in the power supply
generation shape is smoother than individual plant
characteristics, but closely represents how a large number
of wind farms across a geographical area would operate
together.
This simplified wind methodology works well for
forecasting electricity prices across a large market, but
it does not represent the volatility of specific wind
resources that the company might select. A different
wind shape was used for each company resource option
in each of the 300 Monte Carlo iterations. This analysis
uses historical wind data for potential wind sites in the
Columbia Basin and eastern Montana. A statistical
analysis of the wind data showed that a wind plant
would either generally be at or near full output or at
no generation most of the time. This U-shaped or beta
general distribution is shown in Figure 6.13. This shape
demonstrates that a wind plant with an annual average
33 percent capacity factor rarely produces energy at this
2007 Electric IRP Avista Corp
Chapter 6- Modeling Approach
Figure 6.14: Actual Stateline Generation August 9th Through 15 , 2006
100
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ------ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
.. 60
.f 50
u 40
Ii' 30
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- -
60 72 84 96 108
Hour of Week in August
120 132 144 156 168
Figure 6.15: Simulated Hourly Columbia Basin Wind Generation for August
100
II..
- ---------- -------------------- ---------- ----------------- - - - - - - - - - - - - - - - - - - - - - - - - - - -
60 72 84 96 108
Hour of Week in August
120 132 144 156 168
Mean
80% Confidence Interval Hi h
80% Confidence Interval Low
level for a specific period of time but does so over an
extended period.
The Monte Carlo model randomly draws a capacity
factor from the distribution for each hour of each
month. This method creates probabilities for good
average and poor wind years. Serial correlation between
hours ensures that the hour-to-hour wind generation
relationship is retained, preventing an entirely random
wind generation prof11e. Figure 6.14 presents actual
Stateline generation from August 2006. The forecast does
not try to replicate historical wind data; instead it tries
to maintain the underlying statistics of the wind patterns.
The Stateline data never reaches 100 percent capacity
9 Includes losses and the mean of stochastic studies does not guarantee the expected value.
Avista Corp 2007 Electric IRP 6 -
Chapter 6- Modeling Approach
factor due to maintenance and forced outages. The
simulated data in Figure 6.15 includes maintenance and
forced outage normalized as part of the average capacity
factor. Table 6.12 presents the average capacity factors
for Columbia Basin and Montana wind sites, along with
their modeled confidence interval.
FORCED OUTAGES
In the 2005 IRP, forced outages were modeled as
de-rates to plant capacity because AURORAxmp was
unable to integrate random forced outages with other
stochastic inputs. The modeling software now has this
capability. Forced outages are based on a rate and a
mean time to repair. Over the 300 iterations forced
outages average mean outage rate levels. The 2007 IRP
models forced outages stochastically for all CCCT, coal
and nuclear plants. These plants represent the marginal
resources running during the majority of the modeled
hours; they are of the most interest. Hydro, wind, SCCT
and other renewables were not modeled stochastically.
EMISSIONS CHARGES
This IRP uses consultant forecasts for SO2' NOx and Hg
emission costs based on current and projected national
emissions policies. Certain state limits, particularly for
Hg, make emissions modeling problematic at best. The
Base Case emission prices described earlier represent
the mean values for each emission. History shows that
emission costs vary depending on market conditions.
For stochastic analysis, each emission price was assumed
to have a 20 percent standard deviation.
10.
15.
50.
15.
Greenhouse gases, or CO2, emission prices were selected
for each iteration by using a probability of different price
levels because of the greater uncertainty of pending state
and federal regulation. Each iteration uses a different
carbon emission charge. Table 6.13 shows the probability
distribution of CO 2 emissions.
NEW RESOURCE ALTERNATIVES
This section describes each of the resource alternatives
considered in the model to meet Avista s future resource
deficits. These resources reflect generic options that
might differ from actual projects for a variety of siting
or engineering reasons. Actual characteristics and
assumptions will likely be developed through a Request
for Proposal (RFP) process.
COMBINED-CYCLE COMBUSTION TURBINES (CCCT)
Combined-cycle combustion turbines are modeled using
a two-on-one configuration. This configuration consists
of two gas turbines using a single heat recovery steam
generator (HRSG), rather than one gas turbine matched
with a HRSG. These plants generally range between 200
and 600 MW Capital cost estimates are based on a 280
MW 7FA General Electric (GE) machine. Operation
and maintenance costs are based on estimates from the
Northwest Power and Conservation Council (NPCC),
adjusted for inflation.
The heat rate modeled for this resource begins at
722 Btu/kWh in 2008 and decreases by 0.5 percent
each year to account for technological improvements.
11.
15.
16.
33.
13.
16.
23.
30.
60.
2007 Electric IRP Avista Corp
Chapter 6- Modeling Approach
Figure 6,16: Capacity Levels for Northwest Gas-Fired Plants (%)
106
104
102
100
- -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
t1I
""')....
0::(
ffi
u..
::J
""')""')
::J
0::(
Table 6,14: Real 2007 Levelized Costs for 2013 CCCT
The plants are modeled so that 7.7 percent of the
capability is for duct firing at a higher heat rate of
300 Btu/kWh. Forced outage rates are estimated as
5 percent per year; 14 days of maintenance will occur
biennially. Cold startup costs are assumed to be $35 and
3 decatherms per megawatt per start. to CCCT plants
are modeled to back down as far as 50 percent of their
nameplate capacity and ramp from zero to full load in
three hours. The maximum capability of each plant is
highly dependent on temperature. Figure 6.16 illustrates
the average capacity by month for a Northwest CCCT
relative to its nameplate rating.
No limitations were placed on the number ofCCCTs
that could be selected for any area.
CCCT Resource Capital and Operating Costs (2007$):
47.
65.
. Capital Cost: $786 per kW
. Fixed O&M: $9.40 per kW-i. .
SIMPLE-CYCLE COMBUSTION TURBINES (SCCT)
The 2005 IRP includes two simple-cycle combustion
turbine options: Frame (GE 7EA) and aero-derivative
(GE LMS 100) machines. Aero-derivative plants can
ramp up quickly and have low heat rates and start-up
costs, but their upfiont costs are significantly higher than
frame units. Operations and maintenance costs are based
on inflation-adjusted NPCC estimates.
The heat rates for SCCT plants are 8 910 Btu/kWh
(Aero) and 10,139 Btu/kWh (Frame) in 2008, decreasing
by 0.5 percent each year to account for technological
improvements. Forced outage rates are estimated at 3.
10 For example, a 250MW plant would cost $18,987.50 to start up: $8 750 ($35 * 250 MW) for O&M and $10 237.
(6.3 Dth * 250 MW * $6.50/Dth) for fuel.
Avista Corp 2007 Electric IRP
Chapter 6- Modeling Approach
a. . e-
: "
a- , I e-ve- Ize-o.sts o.r I U . va. I a, '11Aero FrameItem ($/MWh) ($/MWh)
T bl 615 R I 2007 L r d 2013 SCCT F II A I bTt
Fuel Cost 62.48 72.
VOM 9.40
Fixed O&M
Non-Capital Transmission
Emissions
Generation Capital Recovery and Overheads 7.48
Transmission Capital Recovery and Overheads
Value of Losses
Total 85.89.
. e- '
. ':
o.a.e- c no. o. .a. ra. c e- rlS ICS a. n. . ssume-o.s s
Plant Capital Fixed Variable Forced
Sizes Heat Rate Cost O&M O&M Outage
Technology (MW)(Btu/kWh)(2007$)($/kW/yr)($/MWh)
(%)
Sub-critical 175-1000 371 905 44.
Super-critical 375-1000 955 004 45.
Ultra-critical 600-1000 825 010 46.
CFB 50-425 289 155 48.43
IGCC 250-650 131 378 54.7 or 1011
IGCC wI seQ.250-650 595 045 64.3.45 7 or 10
percent per year, with no modeled maintenance outages
(maintenance will occur in shoulder months where
these plants do not operate). Cold startup costs were not
modeled. The maximum capabilities of these plants are
higWy dependent on temperature conditions and are
assumed to have the shape as CCCT plants, see Table
15. No limits were placed on SCCT construction.
SCCT Resource Capital and Operating Costs (2007$):
. Capital Cost: $628 per kW for Aero, $419 per kW
for Frame
. Fixed O&M: $9.16 per kW /yr for Aero, $7.05 per
kW-yr for Frame
COAL PLANTS
As identified in the 2005 IRP as an action item, in
2005 and 2006 Avista partnered with Idaho Power to
analyze coal plant costs. After the consultant study was
complete, a Request for Qualifications (RFQ) was issued
to learn about coal projects currently in the development
T bl 6 16 C IPI
11 Forced outage rate is lower if a spare gasifier is available.
6 -
pipeline. The RFQ identified projects in Washington
Idaho, Montana, Utah, Wyoming and Nevada. Each
project's cost and non-cost factors were studied. As a
result of this effort, combined with recent legislative
mandates, Avista has decided that it will no longer pursue
a new coal-fired plant. The resource however, does
warrant review in the 2007 IRP.
Two main types of coal plants were studied:
pulverized and IGCc. Pulverized options are sub-
critical, super-critical, ultra-critical and circulating
fluidized bed (CFB). These different technologies have
different boiler temperatures and pressures, resulting in
different capital cost and operating efficiencies. IGCC
plants may include a back-up coal gasifier and/or a
carbon sequestration option.
The market studies limited coal plant construction to the
Rocky Mountains, Canada and the Desert Southwest.
Plants built in these areas were not allowed to serve loads
t '
2007 Electric IRP Avista Corp
Chapter 6- Modeling Approach
o e- 'e- 010. na.o.a.ra. nsmlsslo. n a. 01 a.o.s s
Capital Cost Size Cost
Location ($Millions)(MW)($/kW)
Northwest 500 000 500
Eastern Montana 000 000 000
WyominQ 000 000 500
T bl 617 R tiC
Table 6,18: Real 2007 Levelized Costs for 2013 NW Coal Plants Full Availabilit $/MWh)
Fuel Cost 26.25.24.25.22.27.
VOM 3.40
Fixed O&M
Non-Ca ital Transmission
Emissions 10.10.10.11.
Generation Capital
Recove and Overheads 24.25.25.27.31.42.
Transmission Capital
Recove and Overheads 5.46
Value of Losses
Total 78.77.77.84,80.91.
in other Western Interconnect areas. This plan assumes
that a new coal plant could not be constructed until
2013 at the earliest.
The various coal plant technologies each have unique
characteristics. Table 6.16 illustrates some of these key
operational and cost differences between them.
TRANSMISSION ESTIMATES:
Coal plant costs are highly dependent on the amount
of transmission necessary to bring their power to load
centers. Estimating transmission costs in regions outside
of the Northwest is difficult, as we are not as familiar
with the unique challenges faced by transmission
planners in those regions. Even with good transmission
cost estimates, the method for cost allocation is
unknown. The 2007 IRP relies heavily on other studies
for estimating transmission costs. Table 6.17 illustrates
the transmission costs assumed for the 2007 IRP. Table
18 presents the 2007 reallevelized costs of the various
coal plant technologies.
12 A spare gasifier is not included.
13 This assumes that a plant is built without a spare gasifier in 2018 or later.
Avista Corp
WIND
Concerns over carbon-based generation technologies
impacts on the environment have greatly increased the
demand for wind generation. Governments, through tax
credits, renewable portfolio standards and eminent carbon
caps are also promoting development. Wind is currently
the major renewable resource with commercial-scale
development potential. Strong demand has increased the
price of acquiring these assets by about 70 percent since
the 2005 IRP.
Three wind resource locations were studied: Columbia
Basin, Montana and plants within Avista s service
territory. Each location has a capacity factor and
transmission cost. All locations were assumed to have the
same capital cost.
TRANSMISSION ESTIMATES:
. Columbia Basin: BPA wheel and $50 per kW for
local interconnection
. Montana: Northwestern wheel and $50 per
2007 Electric IRP 6 - 21
Chapter 6- Modeling Approach
Table 6.19: Wind Location Capacity Factors (Excludes Losses)
Capacity
Location Factor
Columbia Basin Tier 1 33.
Columbia Basin Tier 2 27.
Montana Tier 1 40.
Montana Tier 2 32.
Avista Service Territory Tier 30.
Avista Service Territory Tier 2 21.
. e- 'In' n e- . ra- 10' n o.s s
Wind Capacity System
Wind Location (MW)Penetration $/MWh
Columbia Basin 100
50/50 Mix CB & MT 200 10%
Diversified Mix 400 20%
Diversified Mix 600 30%
T bl 6 20 W' d I t t 14
for local interconnection . BPA losses are 1.9 percent
. Northwestern wheel: $40.80 per kW-
. Northwestern losses are 4.0 percent
. No losses or wheel on Avista system
. Avista Service Territory: No wheel and $30-130
per kW for interconnection; it is likely to be
cheaper to integrate a tier 2 wind site than a tier
1 site to Avista due to the distance of existing
transmission
BPA wheel: $16.90 per kW-
Each regional wind area is modeled with two capacity
factor levels: tier 1 and tier 2. Tier 2 wind has a 20
Table 6.21: Real 2007 Levelized Costs for 2013 Wind Plants
. .
Avista Avista
Columbia Columbia Service Service
Basin Basin Montana Montana Territory Territory
Tier 1 Tier 2 Tier 1 Tier 2 Tier 1 Tier 2
Item ($/MWh)($/MWh)($/MWh)($/MWh)($/MWh)($/MWh)
Fuel Cost
VOM and
Inte ration
Fixed O&M 7.49 11.
Non-Capital
Transmission 14.18.
Emissions Taxes
Generation
Capital Recovery
and Overheads 55.68.43 46.62.63.87.
Transmission
Capital Recovery
and Overheads 1.45 3115
Value of Losses
Total 76.77.93.97,80.104.
14 See http:! /www.avistautilities.com/resources/plans/documents/ Avista WindIntegrationStudy.pdf
15 Transmission estimates near Tier 2 wind sites in Avista s service territory tend to be lower than higher capacity factor wind sites due to
the proximity of transmission lines.
6 -2007 Electric IRP Avista Corp
Chapter 6- Modeling Approach
percent lower capacity factor than tier 1 wind. The
capacity factors in Table 6.19 are mean values for each
region; a statistical method based on regional wind
studies was used to arrive at a range of capacity factors
depending on the wind regime in each year. Table 6.
presents the 2007 reallevelized costs of the various wind
plant locations.
Capital and Operating Costs (2007$):
. Capital Cost: $1 884 per kw,
. Fixed O&M: $17.50 per kW-
. Variable O&M: $1.00 per MWh and
. Wind Integration Costs: see Table 6.20.
ALBERTA OIL SANDS
Alberta Oil Sands are potentially an attractive co-
generation resource option for the United States and
Canada. It must overcome the significant transmission
investment required to transport generated power to the
Northwest. It also requires a partnership between oil and
utility firms to make the project viable. For all of the
discussion around this resource, cost and operating data is
hard to come by.
Transmission for this project has been extensively studied
by the Northwest Transmission Assessment Committee
(NTAC) Discussed below are the assumptions used for
modeling the Oil Sands as a resource option for the 2007
IRP.
OIL SANDS TRANSMISSION ESTIMATES (PRIMARILY
FROM NTAC):
. DC Line: $1 365 433 000
Terminals: $500 000 000
. Communications: $30 000 000
. Total Transmission Capital Cost: $1 895,433,000
Capital Cost: $3 963 $/kW (2007$)
. Transmission O&M: $8.90 per kW-
BPA wheel: $16.90 per kW-
. Losses are expected to be 7.7 percent to Celilo
and 1.9 percent back to Spokane
OIL SANDS RESOURCE
The heat rate of this resource is modeled at 5 000
Btu/kWh. This rate allocates potential emission and
fuel costs to the utility.16 The resource would probably
have a gasifier to transform the residual oil to synthetic
gas and a combustion turbine to generate steam for the
oil recovery process. The fuel price equals the fIxed and
operating costs of the gasifier.
An IGCC plant designed for coal gasification is a similar
resource to Alberta Oil Sands because both require
gasification and the use of a combustion turbine unit.
Given a lack of good price information on this resource
we base our estimate on an IGCC plant capital cost of
378 per kw. As one-third of the plant's heat value is
for electric generation, only that portion is applied to
Table 6.22: Real 2007 Levelized Costs for 2013 Alberta Oil Sands Pro ect Full Availability)
7.45
3.48
53.
14.
91,
16 The IRP assumes no fuel costs, but arrangements could have a fuel charge.
Avista Corp 2007 Electric IRP 6 -
Chapter 6- Modeling Approach
the electricity side of the operation. To this cost a heat-
recovery steam generator is added, bringing the total
plant cost to $3,963 per kw.
OTHER MODELED RESOURCES
A number of other resource options are modeled in
this IRP. These include biomass, geothermal, small
cogeneration and nuclear. Nuclear plants are not
Ta. b.le- 6.23: R- e-1201 017 Le-ve-lize-d. Co.sts foe r a-the- r R- e-so. urce-Full A' va. ila, b. iIi
Geo-Small
Biomass thermal Co-Gen Nuclear
Item ($/MWh)($/MWh)($/MWh)($/MWh)
Fuel Cost 33.48
VOM
Fixed O&M 11.
Non-CaDital Transmission
Emissions Taxes
Generation Capital Recovery and Overheads 51.43.24.42.
Transmission Capital Recovery and Overheads
Value of Losses
Total 67.65.62.72.
Operation and maintenance costs are assumed to be
similar to that of an IGCC plant. Fixed O&M is
modeled at $55 per kW-yr and $3.00 per MWh. The
forced outage rate is assumed to be 5 percent, and
planned maintenance occurs biennially for 21 days. Table
22 presents the 2007 reallevelized costs of the Alberta
Oil Sands resource.
currently considered as a resource option to Avista
but, like coal plants, need to be studied for each plan
because they are an option to other areas of the Western
Interconnect. Over time, this could change as national
policy priorities focus attention on de-carbonizing
energy supply. Nuclear capital costs are difficult to
determine, as a new nuclear project has not been built
in the u.S. in more than 25 years. Better nuclear cost
Figure 6.17: Real Levelized Costs for Selected Resources at Full Availability ($/MWh)
Sm Co-Gen (E. Wash.
CCCT (Northwest)
. Ge~eration IGeothermal (Northwest)
. TransmissionBiomass (Northwest)
Subcritial Coal (Montana)
Nuclear (Northwest)
Wind (Columbia Basin)
Wind (Montana)
IGCC (Montana)
Subcritial Coal (Northwest)
Wind (E. Wash.
IGCC (Northwest)
Aero Peaker
IGCC Seq (Montana)
Frame Peaker
Oil Sands (Alberta)
IGCC Seq (Northwest)
2007 Electric IRP Avista Corp6 - 24
Chapter 6- Modeling Approach
Figure 6.18: Real Levelized Costs for Selected Resources with Market Operations ($/MWh)
CCCT (Northwest)
Frame Peaker (Northwest)
Sm Co-Gen (E. Wash.
Aero Peaker (Northwest)
Geothermal (Northwest)
Biomass (Northwest)
Subcritial Coal (Montana)
Wind (Montana)
Wind (Columbia Basin)
Nuclear (Northwest)
Subcritial Coal (Northwest)
IGCC (Montana)
IGCC (Northwest)
Wind (E. Wash.
IGCC Seq (Montana)
Oil Sands (Alberta)
IGCC Seq (Northwest)
. Generation
. Transmission
1:1 Net Operations Cost
estimates should be available for the next IRP because
several plants are being planned to start construction after
2010. Table 6.23 illustrates the levelized cost assumptions
for each of the remaining plant alternatives.
SUMMARY OF RESOURCE OPTIONS
Figure 6.17 provides a comparison of the reallevelized
costs for each modeled resource option. Costs range
from a low of$65 per MWh for a Northwest CCCT
plant to more than $90 per MWh for a Northwest
IGCC plant. Costs are divided between busbar
generation and the transmission necessary to transport or
integrate the new resource into the company s portfolio.
These costs are based on the resource dispatching at full
availability and at expected costs. This chart does not
consider operational dispatch and other risk factors.
All-in levelized costs based on the full availability of a
generating unit can be misleading. Another way to look
at generation cost is to consider what the plant would
cost when operated in a marketplace. In hours where
the plant is uneconomic, it is not operated and market
purchases replace plant output. Total fIXed and variable
costs, including fuel, are then combined with market
displacement purchases to develop an all-in levelized
cost. Figure 6.18 attempts to address these costs; it shows
Generation and Transmission fIXed costs per dispatch
capability. The Net Operations Cost takes into account
operations cost and market value. For example the cost
of a CCCT in Figure 6.17 is $65 per MWh, taking into
account the market value its net cost is $58 per MWh.
Resources that are not commercially viable or are
prohibitively expensive over the IRP planning horizon
are not modeled in this plan. Examples include: pulping
chemical recovery, new hydroelectric facilities, diesel
ocean current, ocean thermal gradients, petroleum
salinity gradients, tidal energy, wave energy and
distributed generation, including small scale solar and
micro-turbines.
THE PRiSM MODEL
The company developed the PRiSM model to help
select its Preferred Resource Strategy. The model
quantifies the cost and risk of Avista s current resource
portfolio and potential new resources. Each existing and
Avista Corp 2007 Electric iRP 6 - 25
Chapter 6- Modeling Approach
future resource option has an expected operating value.
Some resources provide protection against market price
volatility while others do not. Combining the company
current resource portfolio with an optimal mix of new
resources creates the company s Preferred Resource
Strategy. Additional information is needed, including
solves for the optimal mix by year to meet capacity and
energy needs given a specified level of cost and risk
tolerance. The model gives a larger weighting to the first
10 years of the 20-year study. A simplified view of the
linear programming objective function formula is shown
in Equation 6.
Equation 6.1: PRiSM Objective Function
Minimize:
NPV2OO8-2017 DEV2Ol7 F)+ (X * (10% * NPV2Ol8-2027 )+ X * (10% * DEV2O27 )* F)
Where:
X1 = Weight of cost reduction (between 0 and 1)
X2 = Weight of risk reduction (1 - X1)
F = Factor to adjust risk to equal cost in 50/50 case
DEV is the absolute deviation of power supply costs
NPV is the net present value of total cost
Subject to:
Capacity Need +/- deviation
Energy Need +/- deviation
Wash St. Renewable Portfolio Standard
Resource Limitations and Timing
Capital Spending
capital and fIXed operating costs, to determine an optimal
mix. Resource acquisition target amounts must also be
considered along with the net value of the resource
option.
The PRiSM model uses a linear programming routine.
Linear programs help support complex decision making
that have single or multiple objectives. Developing these
tools requires advanced portfolio and market analysis and
can be expensive and complicated. Linear programming
has been used by many industries for decades, although
the utility industry has been slow to adopt it for resource
planning.
OVERVIEW OF THE PRiSM MODEL
PRiSM has four basic inputs: resource shortages for
peak load and energy, existing resource portfolio costs
and volatility, new resource options over the 300 Monte
Carlo iterations market values and capital costs for
potential new resources. With these inputs, the model
The PRiSM model creates a hypothetical resource
selection given that a utility could add resources in exact
increments as needs specify. It relies on a preferred cost
and risk level for the company. The decision on what
level of cost and risk reduction (Xl and X2) can be
studied further using the efficient frontier approach. An
efficient frontier captures the optimal amount of cost
and risk reduction given the constraints of each level
of weighting for cost and risk Figure 6.19 provides an
example of the efficient frontier. The best point to be on
the efficient frontier curve depends on the level of risk
the company and its customers are willing to accept.
Figure 6.19: Efficient Frontier Line
Risk
Cost
Avista Corp2007 Electric IRP
Chapter 6- Modeling Approach
CONSTRAINTS
As discussed above, various model constraints are
necessary to solve for the optimal resource strategy.
Some of the constraints are physical while others are
societal. The major constraints modeled are capacity
needs, energy needs, the Washington state renewable
portfolio standard and resource limitations and timing.
Approximately 65 percent of the company s retail
electricity load is in Washington. New state law requires
that utilities with more than 25 000 customers meet 3
percent of their load by 2012 9 percent by 2016 and
15 percent by 2020 with new renewable resources.
The model selects qualified resources even if they are
more expensive than other alternatives, provided that
the additional cost does not exceed 4 percent of overall
utility revenue requirement. Where costs are more
expensive, the model can instead purchase qualified
green tags; however, in the absence of a liquid forward
market in green tags, their value is assumed to equal the
4 percent cap.
The model has the ability to limit annual capital
expenditures for power plant and associated transmission
construction. Given the resources selected in this study,
we implemented a capital spending constraint. A number
of resource constraints were necessary to ensure the
PRiSM model selected a reasonable portfolio. The
following list of resource constraints were placed on
PRiSM:
. Wind acquisition is limited to 100 MW
nameplate capacity each year.
. Only carbon-sequestered coal plants are allowed.
. Acquisition of other renewables is limited to 35
MW over the first 10 years and 45 MW over the
last 10 years.
. The model can sell in the short-term electricity
marketplace up to 25 MW in all years except 2017
and 2018 , where expiration of the PGE Capacity
Sale creates alSO MW capacity surplus that must
be managed through a larger sale in that year.
The PRiSM model helps make portfolio decisions by
quantifying the costs and risks associated with each
resource option. It does not replace the judgment of
management. Instead, this method more accurately
quantifies the impact of various resource decisions and
once developed, can evaluate alternatives more efficiently
than simplified portfolio analysis.
CHAPTER SUMMARY
The 2007 Integrated Resource Plan is a comprehensive
modeling effort that studies the companys generation
needs and needs of the entire Western Interconnect. This
modeling approach allows us to identify the impacts
of major fundamental changes to the electric industry,
such as fuel price volatility and carbon regulations. The
IRP has three main components: electric market price
forecasting, risk valuation, and a combination of these
two components into the PRiSM model to select the
Preferred Resource Strategy.
Avista Corp 2007 Electric iRP 6 - 27
Chapter 7- Market Modeling Results
MARKET MODELING RESULTS
OVERVIEW
An optimal resource portfolio must account for
optionality inherent in the resource choices. For the
2007 IRp, a simulation was conducted comparing each
resource s expected hourly output at a forecasted Mid-
Columbia hourly price. This exercise was repeated for
300 iterations of Monte Carlo analysis. Resources that
generate during on-peak hours generally contribute a
higher margin to a portfolio than resources that do not.
This enables certain higher average cost resources to be
more cost effective than other options which generate
electricity during off-peak hours.
Mid-Columbia prices are forecasted using
AURORAxmp, an electric market fundamentals model
developed by EPIS, Incorporated. Chapter 6 discusses
the modeling assumptions used to develop the electric
price forecast. In general, the hourly electricity price is
set by either the operating cost of the marginal unit in
the Northwest or the economic cost to move power into
or out of the Northwest.
To create an electricity market price projection, a
forecast of available future resources must be determined.
This study uses regional (instead of the summation
ofindividual utility needs) planning margins to set
minimum capacity requirements. Western regions
can be long on resources, while individual utilities
may need additional resources. This imbalance can be
due to ownership of certain generating resources by
independent power producers and possible differences in
planning methodologies for those utilities.
AURORAxmp does not select Avista s Preferred
Resource Strategy (PRS); rather, it assigns values to
resource alternatives used in the PRS exercise. Using
several market price forecasts can determine the value
and volatility of a resource portfolio. Since we do not
know what will happen in the future with a significant
degree of certainty, it relies on scenario planning to help
determine the best resource strategy. Scenario planning
is done by developing many different market price
forecasts using different assumptions than the Base Case
or by changing the underlying statistics of a study. These
alternate cases are split into two different categories:
futures and scenarios.
A future is a stochastic study using Monte Carlo analysis
to quantify risks. These studies include 300 iterations of
varying gas prices, loads, hydro, thermal outages, wind
shapes and emissions prices. A scenario is a deterministic
study made by changing one or more specific underlying
model assumptions. These cases are generally used
to understand specific changes, but they do not
quantitatively assess all risks facing the company.
STUDIED FUTURES
The company studies four primary futures for the 2007
IRp, including: Base Case Volatile Gas, Unconstrained
Carbon and the Climate Stewardship Act of 2005 (High
Carbon Charges). Each future provides information
to help the company identify its Preferred Resource
Strategy and to help explain the impact of changing
conditions on its Preferred Resource Strategy.
CHAPTER HIGHLIGHTS
. Gas-fired resources continue to serve the majority of new loads in the West through the IRP timeframe.
. Market prices are forecast to fall from today s level through 2011 , and then steadily rise after 2015;
2008-2027; levelized Mid-Columbia prices are forecasted to be $51.25 (real 2007 dollars).
. Electricity and natural gas prices are expected to remain tightly correlated.
. National Commission on Energy Policy s carbon reduction strategy is included in the Base Case.
. This IRP models four stochastic futures.
. Avoided costs consider capacity and risk reduction when the company is resource deficit.
Avista Corp 2007 Electric IRP 7 - 1
Chapter 7- Market Modeling Results
Natural Gas Price Hen
Northwest Load (aMW),
, OR, N.ldaho
BASE CASE FUTURE
The Base Case future study represents Avista s best
estimate of future costs and prices. It uses average
conditions and expected values for its assumptions.
Many of the key assumptions for this case are described
in Chapter 6; a summary of them is shown in Table
1. Future load growth is served primarily by natural
gas-fired, combined-cycle plants, although many simple-
cycle plants are built to meet planning margin targets.
Renewable resources are included to meet various
states' renewable portfolio standards (RPS), as well as to
provide resource diversification. The Base Case assumes
that states with RPS requirements will not construct
renewable resources in exceedance of such requirements
because of the relative scarcity of these resources. The
federal production tax credit, a large subsidy that offsets
a significant portion of the higher development and
584 20,708 715
100,056 120,056 147 348
25,749 29,311 33,863
162 672
067
operation costs of renewable resource, is assumed to be
extended unti12014.
The Base Case assumes that coal resources can be built
only in Rocky Mountain states to serve local electrical
loads; the energy cannot be exported due to various
state import laws preventing it. Constraining coal plant
construction leaves natural gas-fired resources to meet
most of the future load growth in the West. Table 7.
provides cumulative new generation resources assumed in
the Base Case.
As a region, the Northwest is forecast to be in a surplus
position through 2020. New resource construction
before 2020 occurs to meet RPS and sub-regional
requirements. Figure 7.1 illustrates the Northwest
resource position during the system s one-hour peak
Table 2: Cumulative Western Interconnect Resource Additions Name
280 15,360 040 46,080
002 793 661 761
800 600 200
550 900
016 9,499 20,046 29,086
638 177 331 6,457
936 629 100,228 151 484
7 - 2 2007 Electric IRP Avista Corp
Chapter 7- Market Modeling Results
Figure 1: Oregon, Washington and Northern Idaho Resource Positions (GW)
..................
C')I"-
......
C')l"-
......
ton and Northern Idaho Cumulative Resource Selection MW)
920
540
832 835
150 261 017 871
150 305 849 10,166
100
Figure 2: Mid-Columbia Electric Price Forecast ($/MWh)
~Nominal Dollars
_2007 Dollars
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - ---- -- - - - - - -- - - - - ------------ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -......
C')
......
load condition. Regional resource deficiencies begin
in 2021 , and the model begins non-RPS driven
resource construction at this time. Table 7.3 shows new
Northwest resources included in the Base Case.
I"-
......
C')
............
C\i
Individual utilities with short positions are building
additional resources even though the Northwest is in
surplus. Some level of new resource construction is
likely; however, utilities will probably cover at least a
Avista Corp 2007 Electric IRP 7 - 3
Chapter 7- Market Modeling Results
Figure 7.3: Western Interconnect Resource Dispatch Contribution
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
IX)
C\I C\I C\I
C\I
C\I
C')
C\I
......
C\I
It)
......
C\I
............
C\I
portion of their needs by purchasing existing resources
that presently are surplus to the region s needs. Regional
resources not currently owned by local utilities will
probably be less expensive and entail less acquisition risk
than green field options.
Between 2008 and 2027, projected annual average power
prices for the Mid-Columbia market are $51.25 in 2007
real dollars. Taking inflation into account, the cost of
power is forecast at $60.26 in 2007 nominal dollars.
Figure 7.2 illustrates the nominal and real price of Mid-
Columbia power on an annual average basis. Prices are
forecast to decline in real terms until 2015, and then rise
200
180
160
140
120
100
. Other
. Renewables
. Hydro0 Gas
. Coal
C\I
r--
C\I
IX)
......
C\I C\I C\I C\I
C')
C\I C\I
It)
C\I C\I
r--
C\IC\I
with the imposition of carbon taxes and higher natural
gas prIces.
Natural gas plants are the primary source of new
generation in the Western Interconnect forecast. Coal
serves a large portion ofload, though few new plants are
built. Figure 7.3 illustrates how each resource category
contributes to serving loads over the IRP timeframe.
Figure 7.2 shows expected annual prices, but each year
likely will not experience average conditions or witness
each of our modeling assumptions. The company
conducts a stochastic study to quantify the risk of varying
Figure 7.4: Base Case Stochastic Mid-Columbia Prices ($/MWh)
Mean
- Max & Min
. 80% Confidence Interval
------------------------- ----- -- - - - - - --- - - - - -- - --------- - - - -- - - - --- -- - - - -- -- -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
C\I C\I
C\I
......
C\I
C')
C\I
......
C\I
It)
......
C\I C\I
r--IX)
......
C\I
......
C\I C\I
C')C\I
C\I C\I
It)C\I
C\I
C\I
C\I
r--C\I
C\I
............
C\I C\I
7 -Avista Corp2007 Electric IRP
Chapter 7- Market Modeling Results
400
Figure 5: Volatile Gas Future Stochastic Mid-Columbia Electric Forecast ($/MWh)
350
Mean
- Max & Min
. 80% Confidence Interval
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
300
250
- - - - - - - - - - - - - - - - - - - - - - -- -
200
- -- -
150
------- -- -- ----
100
('I')
"'"
It)
......
future prices. Figure 7.4 shows average annual prices for
the deterministic and stochastic studies. In past studies
including the 2005 IRP, stochastic results were slightly
higher than deterministic results. In the current study,
higher planning margins keep the stochastic mean at the
same level as the deterministic values. There is an 80
percent probability that the 2008 annual average price at
Mid-Columbia will be between $35 and $75. The figure
also shows minimum and maximum annual average
prices recorded across the stochastic Base Case study.
VOLATILE GAS FUTURE
To illustrate the potential for greater price volatility in
the natural gas marketplace, a stochastic study assuming
a more volatile gas distribution was developed. The
standard deviation of expected natural gas prices was
doubled to create more volatility. Figure 7.5 shows the
It)r--r--('I')
results of the study. The 80 percent confidence level of
2008 prices increased by slightly more than 50 percent
to between $21 and $82 per MWh.
UNCONSTRAINED CARBON FUTURE
The Unconstrained Carbon future is identical to the
Base Case, except that no carbon emission costs are
included in the market forecast. Table 7.4 presents
Western Interconnect resource selections under this
future. Compared to the Base Case, the Unconstrained
Carbon future builds the same quantity of resources, but
the mix differs. This case selects fewer SCCTs and more
coal-fired power plants.
This future shows that the National Commission on
Energy Policys proposed carbon mitigation strategy,
included in the company s Base Case future, will not
Table 4: Unconstrained Carbon Future Cumulative Resource Selection MW
2,400 15,360 23,040 48,000
19,860 693 45,299 49,031
600 4,400 800
425 375 900
uestration
016 9,499 20,046 29,086
638 177 331 6,457
24,914 62,754 103,491 151,274
Avista Corp 2007 Electric IRP 7 - 5
Chapter 7- Market Modeling Results
Figure 7.6: Unconstrained Carbon Future Mid-Columbia Electric Price Forecast ($/MWh)
200
180
160
Mean
- Max & Min
. 80% Confidence Interval
---------------------------------- ------ - - - - - - - - - --- - - - - - - - - - - - - ------ - - -- -- - -
140
120 u u
100
- -------------- ------------ ------ --- ----- -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
(J)
......
C")-.t
............
I'-
............
(J)C")I'-C\i
240 000 23,520 560
176 33,206 010 50,573
200 200 600
975
203 213
016 499 20,046 086
638 177 331 6,457
24,070 58,082 94,310 142 464
significantly affect the future resource mix, but it will
increase electricity prices by approximately 7 percent, or
$3.69 per MWh levelized real 2007 dollars, as shown in
Figure 7.
THE CLIMATE STEWARDSHIP ACT OF 2005 (HIGH
CARBON CHARGES) FUTURE
The Climate Stewardship Act of2005 (CSA), otherwise
known as the McCain-Lieberman Bill, was first
introduced in the US. Senate in October 2003. This
comprehensive plan was designed to reduce greenhouse
gas emissions to year 2000 levels by 2010. The bill
would reduce emissions through a market-based tradable
allowance system patterned after the sulfur dioxide
emissions permit market established by the Clean Air Act
of 1990.
The company used the results of an EIA study of this
bill for its High Carbon Charges future, as it is the most
comprehensive analysis available. The CSA was used
in this study as a proxy for all of the pending federal
legislation. More up-to-date studies, or possibly federal
laws and subsequent economic analyses, will be available
and used in the Base Case for the 2009 IRP. Large
carbon charges on electricity generating facilities will
likely stop or severely restrict construction of new
non-sequestered coal plants. In this future, utilities will
probably rely most heavily on gas-fired resources, as
shown in Table 7.5.
In this future, existing coal plants dispatch many fewer
hours than in the Base Case, because carbon credits are
more valuable than electricity generated by these plants.
7 -Avista Corp2007 Electric IRP
Chapter 7- Market Modeling Results
Figure 7.7: CSA Carbon Charge Future: WI Resource Dispatch Contribution
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
. .""",:.
. Other
. Renewables
. Hydro0 Gas
. Coal
C\I
..-
C\I
C\I
..-
C\I
C")
..-
C\I
"'"
C\I C\I
r--
C\I
..-
C\I
..-
C\I C\I C\I
C")
C\I
C\I C\I C\I C\I
r--
C\I
Figure 7,8: CSA Carbon Future, Mid-Columbia Electric Price Forecast ($/MWh)
250
Stochastic
Detenninistic
Base Case
- Max & Min
. 80% Confidence Interval
-------------------------- --------- -
200
100
150
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
C\I C\I
..-
C\I
..-
C\I
C\I
C\I
C")
C\I C\I
..-
C\I
r--
..-
C\I C\I C\I C\I C\I C\I C\I C\I
r--C\I
C\I
Figure 7,9: Western Interconnect Total Carbon with Different Futures (Million Tons of CO2)
600
......... No CO2 Taxes_Base Case
-+-
Volatile Gas
........Oimate Stewardship Act
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
500
550
450
400
350
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
300
..-
C\I C")
C\I C\I C\I
..-
C\I
r--
..-
C\I C\I C\I
C")C\I
C\I C\I
r--
C\I
Avista Corp 2007 Electric IRP 7 - 7
Chapter 7- Market Modeling Results
a. b. le- o. m . a. ra. tlve-e- ve- lize- I' - o.lum. la. . rlce- s a, n. - 10. a.
Standard Coeffi :nt of RangeFutureMeanDeviationVanatlon Low High
Base Case $51,$12.24%$35.$66.
Volatile Gas $51.$23.43 46%$20.$81,
Unconstrained Carbon $47.$11.25%$32.$62.42
Climate Stewardship Act $58.$12.22%$42.$75.
. e-o.m .a. ra. Ive-ve- Ize- .I . - o. um. la. . rlce- s a. n. . IS o. mlna.10. a.
Standard Coeffi :nt of Ran e
Future Mean Deviation Vanatlon Low High
Base Case $60.$14.42 24%$41.$78.
Volatile Gas $60.$27.46%$24.$95.
Unconstrained Carbon $55.$13.25%$38.$73.
Climate Stewardship Act $69.$15.22%$49.$88.
76 C d M"d C
T bl 7 7 C dM'd C I
Figure 7.7 highlights a significant reduction in coal
dispatch beginning in 2015 when carbon charges start.
Figure 7.8 illustrates the impact higher carbon charges
would have on the Mid-Columbia price forecast. The
chart shows that prices increase significantly in 2015
when the carbon charges begin.
Higher carbon emission prices significantly decrease
carbon emissions in the Western Interconnect when
compared to the other futures. This reduction is
illustrated in Figure 7.
b" P'd R" k R I 2007 D II
b" P"d R" k N I 2007 D II
FUTURES SUMMARY AND COMPARISON
The results of the futures analyses show that average
electricity prices vary from the Base Case by as much as
15 percent. Tables 7.6 and 7.7 show levelized prices for
each future in real and nominal 2007 dollars. Natural gas
prices are a key volatility driver; though carbon charges
push prices up, they do not significantly affect price
volatility.
The company conducted a regression and correlation
analysis to study natural gas price impacts on the
electricity marketplace. The study was conducted for
140
Figure 7.10: Sumas Gas Price Versus Mid-Columbia Electric Prices
$; 120
:i! 100
------- -
1- - -
CL 40
:5! 20
:::E
Sumas $ per Decatherm (2008 Monthly)
7 -Avista Corp2007 Electric iRP
Chapter 7- Market Modeling Results
Eauation 7.1: 2008 Natural Gas Price to Electric Price Rearession Eauation
PRlCE2OO8 = 6.8436 * G + 7.2168
Where:
G is the estimated annual average 2008 Sumas natural gas price
Eauation 7.2: 2016 Electric Price Rearession Eauation
PRlCE2O16 31.22+6,86*G+O,56*C-25,74* H +361.84*
Where:
G is the nominal Sumas natural gas price in 2016
C is the nominal carbon tax amount in 2016
H is an index of hydro conditions compared to average conditions
D is the annual average demand (load growth) for energy in the Northwest
calendar year 2008 and uses monthly Mid-Columbia
electric and monthly Sumas natural gas prices for all 300
iterations of the Base Case. Figure 7.10 shows the high
level of correlation, 86 percent, with 75 percent of the
variation in electricity prices explained by variation in
natural gas prices. See Equation 7.1 for the regression
equation.
The regression equation shows that electricity prices will
rise by $6.85 for each dollar change in natural gas prices.
By including other independent variables, the regression
equation is able to predict 99 percent of overall price
volatility. Equation 7.2 identifies each additional
variable s coefficient used to forecast the average annual
electricity prices in 2016.
Table 7.8 provides annual average electric price estimates
using the Base Case regression equation for each of the
studied futures. The equation performs well at predicting
electricity prices across the cases, even though the CSA
future uses a different stochastic methodology to model
carbon charges. Further work in this area could simplify
future IRP analyses by limiting the number of stochastic
futures run through AURORAxmp.
SCENARIOS
The 2007 IRP evaluates fewer scenarios than the
2005 IRP. Many of the market structure impacts from
assumption changes were discovered by analysis of those
cases and in the draft 2007 IRP. The following scenarios
were studied for this plan:
. Constant natural gas prices
. 20 percent decrease in gas price escalation
. 20 percent increase in gas price escalation
. Western Interconnect loads increasing 50 percent
faster
. Western Interconnect loads decreasing 50 percent
slower
. Nuclear plant availability beginning in 2015 and
. Electric car.
$6.
$8.
100%
70%
$59,
$59.
$6.
$0.
100%
70%
$54,
$52.
$6.
$34.
100%
70%
$73,
$75.
Avista Corp 2007 Electric IRP 7 - 9
Chapter 7- Market Modeling Results
Figure 7.11: Natural Gas Forecasts, Constant Gas Growth Versus the Base Case ($/Dth)
Base Case
-- -+-Constant Gas Growth --
- -- - -- -- - - -- -- - - -- - -- - - - - - - - - ..-..-
C')
"'"
It)
..-
r--
..-
C')It)r--
Table 9: Constant Gas Growth Scenario, Cumulative Resource Selection MW
2,400 320 17,760 46,080
18,339 645 44,680 556
000 000 400
375 925 750
uestration
016 9,499 20,046 086
638 177 331 6,457
23,393 61,016 99,742 151 329
Figure 7.12: Natural Gas Price Forecast Scenarios Versus the Base Case ($/Dth)
- - -+-
High Gas
- - -- - - - - - - -- -- - -- - -- - - - - -- -- -- - - -- - - - - - -- - - - - - - - - - - - -
- _Base Case
........ Low Gas
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -..-..-
C')
..-"'"
It)
..-..-
r--
..-..-
C')It)r--
7 -2007 Electric IRP Avista Corp
Chapter 7- Market Modeling Results
h Natural Gas Price Scenario: Cumulative Resource Selection MW
280 14,400 20,640 39,840
15,924 33,083 788 096
800 200 800
550 225 16,575
uestration
016 9,499 20,046 29,086
638 177 331 6,457
23,858 64,509 100,230 152 854
For comparative purposes, all market scenario Mid-
Columbia prices are shown in the summary on in Table
19 later in this chapter. A detailed price forecast for
each scenario, including scenarios studied for the draft
IRP, can be found at the company s IRP website.
CONSTANT NATURAL GAS PRICES SCENARIO
This scenario illustrates the effect on electric prices
and the Preferred Resource Strategy if gas prices do
not fall for several years but continue to increase from
the current price level. As discussed in Chapter 5, gas
prices are forecast to fall from 2008 to 2012. Since the
gas forecast relies on many assumptions, this alternative
was studied to quantify the risk of gas prices continuing
to rise throughout the forecast horizon. Figure 7.
illustrates the scenario s gas price assumption and
compares it to the Base Case forecast. Levelized gas
prices rise from $6.85 in the Base Case to $8.19 in this
scenario (nominal 2007 dollars).
Table 7.9 presents incremental resources selected to meet
future loads in this scenario. Fewer combined-cycle
plants are built early in the study compared to the Base
Case. Gas-fired resources are replaced by coal-fired
generation. The Mid-Columbia electricity price forecast
from this scenario can be found in Table 7.17.
INCREASING AND DECREASING NATURAL GAS PRICE
FORECAST SCENARIOS
High and low natural gas price forecasts would
significantly affect resource planning. Figure 7.
illustrates the natural gas prices used in these scenarios;
prices are assumed to be 20 percent higher or lower than
the Base Case forecast.
Table 7.11: Low Natural Gas Price Scenario: Cumulative Resource Selection MW
360 880 000 53,280
19,087 162 307 564
400 200 000
250
016 9,499 20,046 29,086
638 177 331 6,457
25,101 118 98,884 151 637
Base Case
Hi h Load
Low Load
102
103
101
116
126
108
143
172
119
129
147
113
Avista Corp 2007 Electric IRP 7 -
Chapter 7- Market Modeling Results
h Load Escalation Scenario: Cumulative Resource Selection MW
120
080
016
638
854
080
670
000
9,499
177
84,426
600
507
600
650
046
331
142 734
112 320
320
800
150
29,086
6,457
239,133
h Load Escalation Scenario: Chan e Cumulative Resources %
200
144
Table 7.15: Low Load Escalation Scenario: Cumulative Resource Selection MW
2,400 2,400 2,400 160
140 680 28,443 35,052
000 800 600
425 825
uestration
016 9,499 20,046 29,086
638 177 331 6,457
194 756 58,445 86,180
Table 7.16: Low Load Escalation Scenario: Chan e Cumulative Resources %
uestration
2007 Electric IRP Avista Corp7 - 12
Chapter 7- Market Modeling Results
Tables 7.10 and 7.11 present the resources selected for
each of the gas price scenarios. As gas prices increase
new coal generation increases and fewer resources are
built. When gas prices decrease, fewer coal-fired and
more SCCT plants are built relative to the Base Case.
INCREASING AND DECREASING REGIONAL LOAD
SCENARIOS
Increases and decreases to Western Interconnect
load growth will affect future market conditions.
These scenarios were developed to provide a better
understanding of how the market and resource mixes
would change if higher or lower overall load growth
patterns developed across the Western Interconnect.
Table 7.12 compares these scenarios to the Base Case.
Resources selected are similar to the Base Case, but more
or fewer resources are added in the high and low cases
respectively.
Tables 7.13 through 7.16 show the absolute and
percentage changes in the asset mix from the Base
Case. Market prices are also similar to the Base Case, as
seen in Table 7.19. These scenarios did not assume any
adjustments to the RPS levels because the company does
not believe this will significantly impact market prices or
the value of resource options available.
NUCLEAR PLANTS SCENARIO
The Northwest has not considered nuclear plants as
a viable new resource option for over 20 years. This
scenario illustrates the market impact if new nuclear
resources were available. Nuclear plants would not
materially impact Mid-Columbia prices, assuming
nuclear plant capital costs of$3 100 per kW1 Few new
nuclear plants would be constructed at this high capital
cost. The NPCC's Fifth Power Plan estimated nuclear
capital cost to be $1 735 per kW2 Nuclear plants could
significantly impact Mid-Columbia markets at this lower
level. When one or more of the plants proposed in the
Eastern U.S. are constructed, we should have access to
better cost information. Table 7.17 presents the resources
selected for the Nuclear Plant scenario. A single 1 100
MW nuclear plant was selected between 2015 and 2020;
13 nuclear plants were selected between 2020 and 2027
in this scenario.
Nuclear plants would provide substantial fuel savings
relative to the Base Case. Even though few nuclear
plants are constructed because of high capital costs, fuel
savings equal $10 billion net present value over 20 years.
If more nuclear plants were constructed, the fuel savings
would increase linearly. Figure 7.13 shows the fuel
saving from the Base Case between 2015 and 2027.
Table 7.17: Nuclear Plants Scenario: Cumulative Resource Selection M
280 14,400 19,680 640
16,438 27,832 395 885
2,400 800 000
675 625
uestration
100 15,400
016 9,499 20,046 29,086
638 177 331 6,457
372 56,308 96,027 150,093
1 This represents overnight costs.
The NPCC 5th Power Plan estimates a nuclear plant to cost $1 450 per kW in 2000 Dollars.
Avista Corp 2007 Electric iRP 7 -
Chapter 7- Market Modeling Results
Figure 7.13: Western Interconnect Fuel Costs , Nuclear Beginning in 2015 (Nominal $Billions)
_Base Case
-+-Nuclear Available
-- - --------- - -- - - - - - - - - -- - - ---- - ---- ----- - - - - -- - - - - -- - - - -- - - ------- -- - -
C\I C\I
I'-
C\I C\I
(1)
C\I
C\I
C\I
Lower fuel costs are not the only societal benefit of
nuclear power; a commensurate reduction in greenhouse
gases and other emissions would occur if nuclear power
were added to the preferred resource mix. Figure 7.
demonstrates that carbon emissions stabilize across the
Western Interconnect as more nuclear plants come
on-line in the nuclear scenario. While there are clear
financial and societal benefits from nuclear power
- - -
C\I
C\IC\I
C\I
I'-C\I
C\I
C')C\I
C\I
"'"
C\I
C\I
C\I
C\I C\I
the benefits are currently outweighed by capital cost
uncertainties, waste management issues and other public
policy considerations.
ELECTRIC CAR SCENARIO
Rising energy costs combined with concerns over the
energy security of the United States have stimulated
efforts to find alternatives to fueling transportation
The Tesla All-Electric Roadster Photo Credit: Tesla Motors
7 -Avista Corp2007 Eiectric IRP
Chapter 7- Market Modeling Results
600
Figure 7,14: Western Interconnect Carbon Emissions (Million Tons of CO2)
550
-.- No CO2 Taxes-+-Nuclear Available in 2015_Base Case
.........Climate Stewardship Act
500
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
450 --
- -- -
400
~--. ----.--.----
350
- - - -
300
It)r--
......
C\I C\I C\I
......
C\I C\I
- - - - - - - - - - - - - - - - - - - - -
C\I C\I
r--
C\I
(")
C\I C\I
It)C\I
C\I C\I
225
Figure 7.15: Impact of Electric Cars on the Western Interconnect (aGW)
200
Base Case
......... Total With Electric Cars
------- - - - --- -- - - - - - - - - - - - ---- - ----- -
175
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
150
125
100
C\I C\I
......
C\I
C\I
C\I
(")
C\I
"'"
C\I
It)
......
C\I
......
C\I
vehicles with petroleum. There are many significant
subsidies provided for hybrid cars, ethanol and bio-diesel
production, and hydrogen fuel cells. Though significant
subsidies for hybrid cars arguably do not make them
financially attractive to most buyers.
Properly designed, electric cars have the potential to
help optimize electric system infrastructure. Some initial
analyses have been completed, but to-date no study has
attempted to holistically quantify the costs and benefits
......
C\I
r--
C\I C\I C\I C\I C\I
(")
C\I C\I
It)
C\I C\I
r--
C\IC\I
of converting the US. car and light truck fleet to all- or
mostly electric fueP
Avista developed an Electric Car scenario to consider
the potential benefits an electric car fleet might have on
the US. power industry and how some or all of these
benefits might be used to more rapidly transition the
automobile industry toward electric-only or electric-
hybrid technologies.
3 Most other studies on electric vehicles are conducted in foreign countries and focus on social costs and benefits http://www.kfb.se/
pdfer /R -00-46.pdf and http:/ / www.cenerg.ensmp.fr/francais/themes/impact/ pdf/Elec Vehicle(Funk&RabI1999). pdf. Estimates of the
number of vehicles are assumed to be at the 1999-2003 annual rate of vehicle change taken from a recent Polk Company study.
Avista Corp 2007 Electric IRP 7 -
Chapter 7- Market Modeling Results
Scenario Description
The Electric Cars scenario assumes that all passenger
cars and light trucks across the Western Interconnect are
fueled primarily with electricity by 2020.4 The existing
fleet is replaced or retrofitted entirely over this timeframe
at a rate of 10 percent per year, a rate modestly lower
than the natural replacement of vehicles in the United
States.s An estimated 31.8 million electric passenger cars
and 34.8 million electric passenger trucks and SUVs will
be found in the Western Interconnect fleets by 2020.
Each vehicle will travel an average of 12 500 miles per
year and will consume a net (including charging losses)
22 kWh per mile, while heavier trucks and SUVs will
consume 0.39 kWh per mile.6 Figure 7.15 illustrates the
incremental electric-car load.
Total estimated incremental electrical load in 2020 will
equal 85.8 billion kWh (9.8 aGW) and 169.3 billion
kWh (19.3 aGW) for cars and light trucks, respectively.
This creates an increase in total Western Interconnect
load of approximately 25 perce~t in 2020. Because the
projected growth rate of electric vehicle purchases
higher compared to traditional electricity load growth, by
the end of the study electric vehicles will consume one-
third of all electricity. However, as future electric cars
become more efficient, the growth trajectory of the new
demand could become more gradual.
In addition to the benefits electric cars provide to
non-utility interests, electric cars also provide a number
of benefits ITom a utility perspective. The most obvious
of these benefits is the ability to increase load factor
thereby raising the utilization of infrastructure and
lowering per-unit delivered energy costs. Other utility
benefits might be even more significant. The Western
Interconnect electricity grid is currently comprised
of approximately 200 000 MW of generating capacity.
This study estimates that approximately 15 percent, or
000 Mw, of this capacity stands ready to meet load
requirements during extreme weather events or for
back-up when larger plants experience forced outages.
Except during these short intervals, this capacity sits idle.
By 2027 , capacity in the Western Interconnect will grow
to 300 000 MW in the Base Case, with 45 000 MW held
in reserve. Utilities also reserve generation capacity to
follow intra-hour load and resource fluctuations. This
study estimates that the Western Interconnect reserves 6
percent (12 000 MW today, 18 000 MW in 2027) of its
capacity for reserve services.
Raw" capacity-in other words, the portion of a
peaking plant that cannot be recovered through energy
sales over its lifetime-is assumed in this scenario to be
worth $300/kw, or $45/kW-year in 2007 dollars. At
this price, back-up capacity today costs the Western
Interconnect approximately $1.3 billion annually.
Regulation reserves at this price equal an additional $0.
billion annually. Between 2010 and the end of the IRP
study timeframe in 2027 , total savings from reduced
back-up and reserve capacity equals $25 billion on a
present value basis.
An electric automobile fleet also would have the
potential to assist the grid in managing wind integration.
Recent studies confirm that wind generation consumes
increasing amounts of generation flexibility. They show
that wind integration costs range from $2 to $10 per
MWh. This Base Case IRP future estimates that 35,000
MW of wind generation will be installed in the Western
Interconnect by 2027, generating approximately 99.
Though this scenario focuses on the Western Interconnect due to modeling limitations, its results likely could be extrapolated across the u.s.
5 37BetterMotors states the average length of vehicle ownership in the u.S. is between 5 and 10 years. http://37signals.com/better
motors.php. Full Scrappage rate of passenger vehicles in the u.s. was 4.5 percent in 2005 according to Green Car Congress.
http://www.greencarcongress.com/2006/02/us vehicle flee.html
6 This baseline assumption of .22 kWh per mile comes from data released on the Tesla Roadster. A pro-rata increase based on vehicle
weights was applied to SUVs and light trucks.
7 -2007 Electric IRP Avista Corp
Chapter 7- Market Modeling Results
million MWh annually. The wind integration costs
could vary between $0.2 and $1 billion. Between 2010
and 2027, the total value ranges from $1 to $5 billion.
Electric vehicles could eliminate the need for a majority
of transportation-related gasoline and diesel fuel. This
study assumed that gasoline and diesel prices average
$3 per gallon, escalating at 3 percent annually through
the forecast. Total fuel savings from the projected use
electric cars equal 3.6 billion gallons in 2010, rising to
48.0 billion gallons per year by 2020. Over the 2010 to
2027 period, total fuel savings equal approximately $986
billion dollars, net present value.
Transportation in the United States is responsible for
roughly one-third of US. carbon dioxide emissions.
Converting transportation vehicles to electricity should
drastically reduce overall pollutant levels. Assuming a
50 percent reduction in carbon emissions, each electric
vehicle would reduce carbon emissions by approximately
5 tons annually.7 Valuing this savings at $10 per ton
would provide a $25 benefit per year per vehicle. Over
the IRP timeframe, using the Base Case CO2 emission
price would equal a CO2 emission savings of$11.
billion present value for the Western Interconnect.
Converting the Western Interconnect fleet of cars and
light trucks to electricity would require significant new
capital investments. This being said, the study s assumed
the replacement rate falls below the natural rate of
vehicle replacement in the United States; therefore, the
only significant costs resulting from the conversion are
the increased costs of electric vehicles versus traditional
vehicles and the infrastructure necessary to provide for
charging vehicles both at home and away.s Table 7.
details the costs and benefits of the electric car scenario.
Electric vehicles have the potential to provide back-
up capacity, reserves and wind integration services.
Theoretically, each vehicle would be capable of providing
more than 200 kW of instantaneous power to the
electrical grid when connected. However, at this rate
a vehicle would drain its batteries in approximately
15 minutes. A more conservative estimate for vehicle
capacity is 10 kW for cars and 20 kW for light trucks
and SUV s, the approximate charging rate of to day
technology. At this rate of discharge, each vehicle could
provide up to five hours of continuous grid support
though it is unlikely that the electricity industry would
need even a fraction of this capability to support the grid.
In total, electric vehicles could be capable of providing 1
Back-U Ca
Reserves
Emissions
Wind Inte ration
Reduced Petroleum Consum lion
Incremental CarlTruck Cost
New Electrici S stem Infrastructure
Electrici Fuel and O&M
Net Value
Electrici Benefit
986
221
699
7 Emissions based on 2005 EIA study. http://www.eia.doe.gov/oiaf/1605/ggrpt/carbon.html. 50 percent reduction in emissions
assumption based on 2006 study by Sherry Boschert featured in Plug-in Hybrids: The Cars That Will Recharge America.
8 This study assumes that the cost of infrastructure for changing the automobile industry over to electric-fueled vehicles only is covered in
the cost of those vehicles.
Avista Corp 2007 Electric IRP 7 - 17
Chapter 7- Market Modeling Results
Ta. b. le- 7.19': Future- a.nd. Sce-na.rio. Ma.rke-t P"rice- Co.m.a.riso.ns '/MWh
Real Nominal
Scenario 2007 2007 2010 2015 2020 2027
Base Case 51.60.50.55.70.94.
Constant Gas Growth 58.46 68.59.69.78.45 105.
Hiah Gas Price 58.68.58,61.80.82.43
Low Gas Price 43.43 51.41.47.61.44 92.
HiQh Load Growth 51.60.50.57.71.94.
Low Load Growth 50.59,49.45 54.47 69.92.
Nuclear Available 50.43 59.49.54.69,93.
Electric Car 56.66.52.65.81.99.
59.69.46 49.42 68.92.119.
Unconstrained Carbon 47.55.50.49.62.85.
Electric Car
Figure 7,16: Comparison of Total Fuel Costs for the WI in 2017 and 2027 ($Billions)
High Load Growth
High Gas Prices
Constant Gas Growth
Base Case
No CO2 Taxes
Nuclear Available in 2015
Low Gas Prices
Low Load Growth
100
million MW of grid capacity, approximately three times
the total installed capacity of the Western Interconnect in
2020.
Each automobile could be fitted with a device that could
respond to system frequency or other signals to allow
charging to occur with the following order of preference:
(1) meet customer need to maintain a "full tank" of fuel
when needed and (2) provide a storage system to meet
fluctuating changes on the electricity grid.
Charging is expected to occur mainly during lower-cost
off-peak hours of the day, though customers would have
the option of charging their vehicles at other times when
necessary.
.2008-2017 NPV
.2018-2027 NPV
150 200 250 300 350
Impacts on the Larger Economy
The Electric Car scenario would have significant impacts
on the utility, automobile manufacturing and automotive
fueling industries. It would also impact infrastructure
at consumers' homes and where they work and play.
A number of assumptions are necessary to envision
the impacts of the Electric Car scenario. This study is
utility-centric and does not attempt to quantifY all of
the wealth transfers that might occur under the scenario.
However, a return of more than one trillion dollars on an
investment of $350 billion over 20 years is impressive.
FUTURES AND SCENARIOS SUMMARY TABLES AND
CHARTS
A comparison of all of the futures and scenarios run for
7 -Avista Corp2007 Electric IRP
Chapter 7- Market Modeling Results
the 2007 IRP are contained in Table 7.19 below. Total
fuel consumption is included Figure 7.16. The large
increase necessary to support the Electric Car scenario is
offset by even larger reductions in automotive fuel.
AVOIDED COSTS
Avista is obligated to purchase certain third-party
generation under the Public Utility Regulatory Policies
Act of1978 (PURPA). Federal law states that such
purchases will be at prices equal to avoided cost. State
regulatory commissions implement PURPA provisions in
their states.
PURPA developers whose projects exceed certain levels
are eligible for a negotiated rate based on utility avoided
cost, and published rates are provided for smaller PURPA
facilities. In Washington, PURPA resources below one
MW are eligible for published fIXed-rate schedules up
to a five-year term. The five-year schedules are tied to
forward market prices. In Idaho, facilities up to 10 aMW
may obtain published avoided cost rate for up to 20 years.
AVOIDED COSTS VERSUS THE WHOLESALE
MARKETPLACE
There is some disagreement within the industry about
what specifically constitutes avoided cost. In Idaho
administratively determined avoided cost rates use Avista
next lowest cost investment to set rates. The published
figure explicitly includes the cost of installing capacity.
In Washington, published rates are based entirely on the
forward wholesale market price.
AVOIDED COSTS APPROACH
Avoided costs are a function of energy and capacity
cost. Some resources, such as wind, provide little or no
capacity. Most coal- and gas-fired plants provide both
energy and capacity. Other resources, including hydro
and peaking plants, provide a lot of capacity relative to
their expected energy generation profile. Both capacity
and energy have value. Energy is easily valued by electric
market pricing such as the Mid-Columbia index, while
capacity valuation is more difficult because there is not
an active Northwestern capacity market.
Capacity traditionally has been valued at the cost to
build a SCCT plant, even though this plant would
provide some energy value over time. The IRP provides
a better means of extracting capacity value using the
PRSiM Model. As described in Chapter 6, the PRiSM
model helps the company select new resources to meet
future needs. All of the selected resource options are
expected to cost more than the electric market price.
The difference in cost between the Preferred Resource
Strategy and the energy market price represents an
avoided cost for capacity, and the subsequent lowering
of future portfolio risk. Capacity value alone can
be separated from risk by comparing the cost of the
Preferred Resource Strategy to a mix of new resources
that ignore portfolio risk.
The lowest-cost portfolio is made up of simple
cycle turbines and purchasing green tags to meet the
Washington State Renewable Standard. This portfolio
is expected to cost $9.32 per MWh over the market
price, which represents the capacity value of new
generation. The difference between the lowest-cost
portfolio and the FRS indicates the value the company
and its customers are placing on risk reduction. The risk
reduction premium equals $9.39 per MWh. Where a
PURPA resource provides both risk and capacity benefits
on-par with the FRS mix, the avoided cost payment
made under PURPA should equal the cost of the FRS.
If a PURPA resource provides more or less value, the
payment should be adjusted accordingly.
Avista Corp 2007 Electric IRP 7 -
Chapter 8- Preferred Resource Strategy
PREFERRED RESOURCE STRATEGY
INTRODUCTION
The 2007 Preferred Resource Strategy (FRS) differs
substantially from the company s 2005 plan in three
main areas: coal, renewables and gas-fired plants. Avista
is no longer willing to rely on traditional coal-fired
technologies to meet future customer needs. This reflects
recent emissions standards legislation in Washington
imminent federal carbon limiting legislation and higher
coal-fired generation costs. There is a lower contribution
from wind and other renewables due to: (1) recent
legislation promoting renewables in Washington and
Oregon that has reduced the amount of cost-effective
renewables available by increasing demand for such
resources, and (2) wind generation costs have more than
doubled over the past six years and increased more than
50 percent since the 2005 IRP. The final change is that
natural gas-fired plants have returned to the PRS. Gas
resources have not increased as significantly as the other
resource options.
The charts and tables presented in this chapter focus
on the first 10 years of the plan, as these years are the
most relevant for developing our near-term acquisition
strategy. All IRP studies were based on 20-year analyses.
Lancaster Generation Facility
CHAPTER HIGHLIGHTS
. Capital costs for coal and wind generation have increased drastically over the past two years; this greatly
affects our future plans.
. Coal-fired generation in previous plans is replaced entirely with gas plants.
. Preliminary analyses show that fixed-price gas contracts can reduce year-to-year rate volatility substantially;
the PRS "hedges" the portfolio with fixed-price gas even though costs are higher.
. Fewer renewables meet our future loads due to tightening market conditions.
. Conservation acquisition is 25 percent higher than in the 2005 plan and 85 percent higher than in the 2003 IRP.
. The PRS includes 350 MW of gas, 300 MW of wind, 87 MW of conservation, 38 MW of hydro plant
upgrades, and 34 MW of other renewables by 2017.
. Lancaster, a currently running CCCT plant, will be available to the utility in 2010.
Avista Corp 2007 Electric IRP 8 - 1
Chapter 8- Preferred Resource Strategy
The result is a FRS that relies primarily on natural gas
generation, wind and other renewables. The elimination
of coal from our future, combined with reduced
contributions from renewable resources opens the
possibility of more power supply cost volatility relative to
the 2003 and 2005 plans. The costs of these more price-
stable resources simply were too high relative to other
options. In the absence of a new strategy our customers
will be forced to bear this rising volatility. Fortunately,
there appears to be an affordable option to reduce the
volatility of gas-fired generation resources. We are
hopeful that long-term fIXed gas contracts will reduce
overall volatility. Make special note of Figure 8.13 later
in this chapter and consider the superior risk profile of
the FRS relative to the "PRS-No Fixed Gas" portfolio.
Power supply expenses are reduced significantly for a
modest increase in average power supply expense by
locking in" a significant portion of our natural gas
supply under long-term contracts. There is a more in-
depth discussion of how the company might fIX its gas
prices for the long term later in this chapter.
The 2007 IRP finds that recent legislation promoting
renewables and reducing greenhouse gases and other
emissions has driven power supply expenses and
customer rates higher than they would be absent these
mandates and will continue to do so. While sensitive
to and concerned about higher costs that translate into
higher rates, we do not oppose society s desire to reduce
its impact on global warming and diversifY power
production away from carbon-emitting sources. This
plan simply is intended to inform our management
investors, regulators and customers of the costs of
complying with new environmental mandates.
PRiSM DECISION SUPPORT SYSTEM MODEL
As with the 2003 and 2005 IRPs, we continue to use
our decision support system software (PRiSM) to help
guide resource planning decisions. This differs from the
traditional approach many utilities undertake in which
a simplified set of resource portfolios is developed to
illustrate the impacts of one resource decision over
another. 1
The PRiSM model brings together the value of Avista
existing portfolio of resources, its load obligations and
resource opportunities available to meet future load
requirements. To capture the optionality inherent in each
Table 1: Resource 0 tions Available to Avista for the 2005 and 20071RP, First 10 Years
Simple-Cycle Gas
Combined-Cycle Gas
Sub-Critical Pulverized Coal
Critical Pulverized Coal
Super-Critical Pulverized Coal
IGCC Coal , Not Sequestered
IGCC Coal , Sequestered
Alberta Oil Sands
Nuclear
Wind
Biomass
Geothermal
Cogeneration
Simple-Cycle Gas
Combined-Cycle Gas
Wind
Biomass
Geothermal
Cogeneration
1 The company still develops portfolios, both to illustrate the benefits and costs of certain resource decisions and for comparison to the
Preferred Resource Strategy portfolio selected by PRiSM.
8 - 2 2007 Electric IRP Avlsta Corp
Chapter 8- Preferred Resource Strategy
of these categories, the results from of the 300 Monte
Carlo AURORAxmp runs are considered. Capital
transmission and fIXed operations and maintenance costs
attributable to each new resource option are evaluated.
PRiSM reviews our existing portfolio and selects an
optimal mix of new resources from the available options.
A more in-depth discussion of the PRiSM model, and its
inputs and outputs, may be found in Chapter 6.
CHANGING POLITICAL ENVIRONMENT
The 2007 IRP responds to major state and federal
policy changes to reduce greenhouse gas emissions and
encourage development of renewable energy sources.
Avista moved away from natural gas-fired resources in its
2005 IRP because of the fuel's inherent price volatility.
Recent trends and legislation, such as Washington
Senate Bill 6001 (SB 6001), prevent the company from
entering into any long-term financial commitment
for resources that exceed a greenhouse gas emissions
performance standard ofl 100 Ibs/MWh. The bill
provides for the standard to be lowered even further
after 2012, making compliance even more costly. The
emission performance standard effectively precludes
the company from acquiring any new pulverized coal
plant or a long-term contract with an exiting one, and
therefore compels us to rely on natural gas resources.
Table 8.1 illustrates the increasingly limited resource
options available to Avista in this plan.
These limitations stem primarily from new and expected
mandates at the state and federal levels. In the State
ofWashington, limitations have come from Citizen
Initiative 937 (Energy Independence Act, or 1-937), SB
6001, Executive Order No. 07-02 (Washington Climate
Change Challenge) and the Western Regional Climate
Action Initiative signed by the governors of five Western
states. Collectively, the legislation and order seek to
decrease greenhouse gas (GHG) emissions, increase
employment levels in green energy resources, reduce
fuel imports and increase overall renewable generation
levels. Oregon has similar renewable and emissions
goals and laws in place or in development. Other states
throughout the Western Interconnect are also developing
or have already enacted GHG reductions and renewable
portfolio standards. No RPS or carbon emission standard
presently exists in Idaho.
There is a strong regional and national push toward
developing a market-based GHG reduction program.
It involves several competing cap-and-trade legislative
proposals in Congress, as well as an effort to design
and implement a regional mechanism to achieve GHG
reduction goals. It is also apparent that Congress may
enact renewable portfolio standards in the near future.
This IRP assumes that there will be GHG constraints
and models its Base Case on policy recommendations
contained in the National Commission on Energy Policy
December 2004 report.
The combination of actual and pending state and
national legislation creates considerable uncertainty and
novel resource conditions and challenges. First, while
the company anticipates that federal GHG and RPS
legislation will eventually become law, we can neither
accurately predict the final form of these measures
nor can we determine if problems may arise from
complying with state and federal mandates governing
the same subject matter. At this time, the company can
only make general assumptions about future regulatory
requirements, with two exceptions: Washington state
937 and SB 6001. Second, competition and demand
for renewable generating assets has increased substantially
since the 2005 IRP, as will be discussed later. That
competition is principally a factor of five circumstances:
. RPS requirements, including the accelerated
compliance schedule for California s RPS law
. political considerations associated with pending
climate change policies, which, for example, impel
RPS-exempt municipal utilities in California to
Avista Corp 2007 Electric IRP
Chapter 8- Preferred Resource Strategy
acquire renewable generating assets even in the
absence of applicable mandates
. the need for resource diversity to mitigate utility
exposure to volatile natural gas
. the ambition of electric utilities to acquire the
most economical wind generation sites before they
are purchased by competitors, and
. uncertainty about the renewal and duration of
federal tax incentives.
Heightened competition for renewable resources has
caused a dramatic increase in their cost. Short-term
renewals of the federal production tax credit (PTC) also
exacerbate the supply and demand balance for wind
power as developers try to finish projects before the
PTC expires. Lastly, legislation impacts the availability
resources available to serve utilities' retail loads.
Traditional coal-fired generation provides stable
cost-effective energy that meets more than half of
current U.S. power needs. It also emits a tremendous
amount of carbon dioxide (CO ) relative to other
generation options. For every MWh a coal-fired plant
generates, it emits approximately one ton of CO , This
is a level three times higher than from gas-fired CCCT
plants. In a carbon-constrained economy, traditional
coal-fired generation will become expensive as these
generators scramble to acquire carbon offSet credits
weigh the reduced value of generation against the value
of selling carbon offsets into a tight marketplace, or install
carbon mitigation technology. Coal-fired technology is
also significantly more expensive than forecasted in the
2005 IRP.
WASHINGTON STATE RPS
The passage ofI-937 requires all Washington state
electric utilities with more than 25 000 customers to
acquire new "eligible renewable resources" to meet 3
percent of their energy needs by 2012, 9 percent by
2016, and 15 percent by 2020. Figure 8.1 demonstrates
Avista s incremental renewable resource needs. In 2016
more than 80 aMW ofI-937 qualifying renewable
resources are needed; if met by wind resources alone, it
would require Avista to build approximately 240 MW
nameplate capacity. If non-wind renewables options such
as biomass or geothermal can be acquired at an attractive
price, the required renewable resource capacity will be
approximately 90 MW
Wind generation has thus far proven to be the most
commercially viable technology for meeting RPS
requirements. It is necessary to acknowledge the
Figure 8.1: Amount of Renewable Energy Forecasted to Meet Wash. State RPS (aMW)
160
140 . Renewable Need
- - - - - - - - - ----
. Projected Qualified Resources
120
100
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
C\I
CJ)
C\I C\I
C\I
C\I
C')
C\I
"""
C\I
It)
......
C\I
............
C\I
......
C\I
t--
C\I C\I
CJ)
C\I
t:!
C\I C\I
C')
t:!
C\I C\I
It)
t:!
C\I
t:!
C\I
t--
t:!
C\IC\I
Avista Corp2007 Electric IRP
Chapter 8- Preferred Resource Strategy
limitations of relying on wind for these purposes. The
American Wind Energy Association (AWEA) ranks
Washington state 24th in the nation for wind energy
potential. Specifically,AWEA estimates the state s annual
wind energy potential to be 3 740 MW By comparison
Montana is ranked fifth with 116 000 MW of annual
potential. Montana has approximately 10 times the
combined wind potential of the states ofWashington
Idaho and Oregon combined. Unfortunately Montana
wind power potential exists east of the Rocky Mountains
and therefore is not an "eligible renewable resource
under 1-937. This limitation makes compliance more
difficult than it otherwise might be. Transferring
wind energy generated in eastern Montana westward
is also hindered by a present lack of transmission and
integration capacity.
The Fifth Power Plan, published by the Northwest
Power and Conservation Council (NPCC), estimates
the potential wind power capacity of the Pacific
Northwest to be approximately 6 000 MW The
NPCC acknowledges that this potential will have a
capacity factor between 28 and 30 percent. Most
the economically viable and readily developable wind
power sites in the region have already been or are in
the process of being acquired. As Pacific Northwest
electric utilities proceed to comply with RPS mandates
they will be forced to compete for a diminishing pool
of cost-effective wind power sites and to do so within
governmentally-mandated periods of time. This is a
recipe for even higher renewable resource costs and retail
prices in the future.
The limited economic availability of renewable resources
poses planning and regulatory challenges for Avista.
While we are committed to meeting the requirements
ofI-937, we are cognizant of the near-term cost impacts
of those requirements. The company is also concerned
about the potential financial ramifications of failing to
proceed expeditiously to acquire renewable resources
lest their cost continue to rise compared to alternative
resources. This planning uncertainty is compounded
by 1-937, which challenges the conventional regulatory
paradigm. This law dictates the companys "need" to
acquire renewable energy or renewable energy credits.
Though the purchase of renewable energy credits would
enable the company to comply with 1-937, it does
not afford us any certainty about meeting renewable
energy standards in perpetuity. Renewable energy credit
purchases might delay the acquisition of renewable
resources to a point in time when those resources are
more expensive still.
DECREASED RELIANCE ON RENEWABLE
RESOURCES
The 2005 IRP recommended the acquisition of nearly
500 MW of renewable resources between now and 2016
and 750 MW by 2026. Wind resources at that time
though not expected to be inexpensive, were competitive
with other options. Other renewable technologies
including geothermal and biomass, were slated to
make up nearly 20 percent of the renewable resources
contribution in the 2005 plan. The company identified
its overall renewables acquisition strategy as a stretch goal.
Wind plant costs have increased by approximately 50
percent since the 2005 plan, a trend that the 2005 IRP
identified as then beginning to occur. As described
earlier, several factors including RPS requirements have
dramatically increased demand for renewable resources.
Both higher costs and lower availability have reduced
the expected contribution of renewable resources over
the first 10 years of the plan from 500 MW in the 2005
plan to below 350 MW (300 MW wind) in this plan; no
additional wind is selected, where the 2005 IRP included
an additional 350 MW of renewable resources.
To ensure the company has a RPS-compliant portfolio
it is likely that resources will need to be acquired prior
to the traditional load and resource balance metric.
Avista Corp 2007 Electric IRP
Chapter 8- Preferred Resource Strategy
Obtaining resources in an environment with significant
competition has already resulted in a scramble to obtain
the best resources. The company will consider turnkey
or power purchase agreements, as well as investing in
potential renewable energy sites for future development.
We will also consider purchasing qualifying renewable
energy credits to meet our statutory obligations.
NATURAL GAS PLANTS RETURN TO THE
RESOURCE MIX
Natural gas prices rose drastically between the 2003 and
2005 plans. Compared to other resource options, namely
traditional coal-fired resources, natural gas became both
costly and volatile. With a high contribution by wind
and other renewables, natural gas was not selected in the
2005 plan. Conditions are different today. Natural gas-
fired plant costs have not risen as significantly as other
options. In addition, traditional coal,..fired technologies
are not available to the company in this planning exercise
due to recent legislative changes in Washington state.
Figure 8.2 compares capital cost assumptions of various
resource options in the 2005 and 2007 IRPs.
Rising capital costs make gas-fired generation more
attractive because it is a less capital-intensive resource
than coal, wind or other renewable options. CCCT
generation was forecast in the 2005 IRP to cost
approximately $59 per MWh (reallevelized 2007
dollars), while the lowest-cost coal-fired option was
approximately $42.2 The 2007 IRP forecasts equivalent
costs to be $62 and $61 per MWh for CCCT and
Montana-based coal plants, respectively. The gas-fired
CCCT cost rose a modest 5 percent overall, even though
its capital costs are 15 percent higher than in the 2005
plan; the overall cost increase was lower than the capital
cost increase. Coal-fired generation moved in the
opposite direction, rising almost 50 percent compared
with a 35 percent capital cost increase. Gas represents a
comparatively more attractive resource today than it was
in 2005, even absent changing social policies.
Though potentially representing a more volatile
future when compared to the 2005 FRS, the absence
of traditional coal-fired technologies and fewer cost-
effective renewables in the 2007 IRP leave natural gas
the major new resource. The 2007 Preferred Resource
Strategy includes nearly 350 MW of natural gas-fired
CCCT plants in the fIrst 10 years.
DEMAND-SIDE CONSERVATION
PROGRAMS UP 25 PERCENT
The 2005 IRP increased DSM by 50 percent over the
2003 IRp, primarily in response to rising market and
supply-side resource costs. Studies developed by our
conservation groups find approximately 25 percent
500
Figure 8,2: Generation Capital Cost Trends (2007 $/kW)
000
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
500
-- -- --
000
- - - - - - - - - - - - - - - - - -
+1~k
- - - - - - - - - - - - - - - - -
2% ~
500
SCCT CCCT
2 Excluding emission costs.
Coal Wind
2007 Electric IRP Avista Corp
Chapter 8- Preferred Resource Strategy
more conservation potential in 2007 than in 2005. The
avoided costs against which conservation options are
compared continue to rise. As explained above, resource
alternative costs are higher in the 2007 IRP. This raises
the value of energy saved by conservation measures.
Additionally, the 2007 IRP recognizes other factors for
the first time that increase the value of this resource;
namely capacity value, risk reduction, transmission and
distribution savings. These additional factors are inherent
in the selection of supply-side resources. The application
of new analytical techniques enables the company to
assign values for these benefits. Refer back to Chapter
3 for a detailed discussion of the methods we employed
and the values assigned to these new benefit categories.
The company forecasts it will acquire 87 aMW of
conservation over the next decade, thereby reducing the
need for new supply-side resources.
SUPPLY-SIDE CONSERVATION EFFORTS
CONTINUE
The company continues to explore ways to increase
the generation it receives from existing resources and
the efficiency with which it is delivered. Upgrades at
our Cabinet Gorge and Colstrip plants have increased
generation by approximately 20 MW since the 2005 IRP.
The company has evaluated numerous upgrade options
at its hydroelectric projects over the past two years.
This plan incorporates upgrades to the Noxon Rapids
hydroelectric project, increasing generation capacity
by 38 MW Future upgrade evaluations will be made
considering the same new factors being applied to the
conservation resource options.
PREFERRED RESOURCE STRATEGY
SUMMARY AND COMPARISON TO 2005 IRP
The FRS includes wind, other renewable resources
combined-cycle combustion turbines, and supply- and
demand-side efficiency improvements. Table 8.2 provides
the quantity and timing of proposed resources for the
first 10 years of the plan. Comparing this strategy to the
2005 IRP, shown in Table 8., this plan moves away from
coal toward gas-fired resources, scales down wind due to
rising capital costs and lowers the amount of expected
capacity from other renewables. More conservation is
acquired.
Another key difference between this plan and the 2005
plan is that the first new base load resource enters service
CCCT 280 280 280 350 350 350 350
Coal
Wind 100 100 200 300
Other Renewables
Conservation
Total 327 346 356 541 551 661 772
CCCT
Coal 250 250 250 250 250 250
Wind 150 200 250 325 400 400 400
Other Renewables
Conservation
Total 106 198 515 582 674 766 783 800
Avista Corp 2007 Electric IRP 8 - 7
Chapter 8- Preferred Resource Strategy
Figure 8.3: Historical and Future Nameplate Capacity Acquisition (MW)
900
700
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
New Resource Additions
500
300
Historical Resource Additions
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
100
900
700
500
300
100
..-~ ~
m m
..- ..-
1"'-cx)m 0m 0
..- ..-
..- N
(')
0 00 0N N
in 2011 rather than 2012. The 2005 IRP assumed that
a coal resource would not be available until 2012, so the
2011 deficit was filled with short-term contracts until
that resource was available. This IRP selects a natural gas
plant to meet the 2011 shortfall.
RESOURCE ACQUISITION IS LUMPY
PRiSM does not select the Preferred Resource
Strategy; rather it informs the utility on the resources
that should be selected. The exact PRiSM strategy
cannot be used because the model selects resources in
perfect quantities to meet resource deficits. It also lacks
the ability to quantify all of the experience of Avista
management team. Actual resource acquisition will
~ 80 0N N
co I"'-0 00 0N N
CX)
..-(') "'"..- ..-
0 0N N
It) co
..- ..-
0 0N N
I"'-
..-
likely not be so perfect and will be acquired in a lumpy,
or stepwise, pattern. Figure 8.3 shows historical and
future resource acquision. This chart shows that the
company traditionally adds resources in blocks; at times
the company has been able to acquire shares of a plant to
reduce the dependence on large plant acquision. Figure
8.4 shows the total amount of resources selected by
PRiSM's 25/75 risk/cost strategy compared to the PRS.
The key difference is that resources added between 2011
and 2013 by PRiSM are added in 2011 as a single block.
Resource selections in the second 10 years of the plan
are not changed from the PRiSM model selection.
Acquisitions in this timeframe will be quantified in
future plans. Later in this chapter the PRS will be
200
Figure 8.4: Lumpy Resource Acquisition (MW)
000 -+- PRS
25% Risk Reduction
800
- -
600
400
200
- -- -..-(')"'"
It)
..-
2007 Electric IRP8 - 8
---- - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - -
I"'-
Avista Corp
Chapter 8- Preferred Resource Strategy
Table 8.4: Loads & Resources Ener Forecast with PRS aMW
, ,:
II'
Obligations
Retail Load 125 163 196 230 256 326 379 1 ,450 627
90% Confidence Interval 200 199 196 196 192 192 192 156 156
Total Obligations 324 362 392 425 448 518 571 606 783
Existing Resources
Hydro 540 538 531 528 512 510 509 491 491
Net Contracts 234 234 234 129 107 105 105 106 106
Coal 199 183 188 198 187 187 198 199 186
Biomass
Gas Dispatch 280 295 285 295 280 295 295 280 295
Gas PeakinQ Units 145 145 141 146 145 146 145 141 145
Total Existing Resources 446 442 426 342 278 290 299 265 270
PRS Resources
CCCT 253 253 316 316 389 612
Coal
Wind 103 103 103
Other Renewables
Conservation 103
Total PRS Resources 279 291 406 487 587 871
Net Positions 122 196 121 179 215 246 359
Table 8.5: Loads & Resource Ca aci Forecast with PRS MW
,,:
II'
Obligations
Retail Load 703 763 815 868 909 019 103 214 2,492
Planning Margin 260 266 272 277 281 292 300 311 339
Total Obligations 964 029 087 145 190 311 404 525 831
Existing Resources
Hydro 142 154 121 128 084 098 098 070 070
Net Contracts 172 172 173 208 128 128
Coal 230 230 230 230 230 230 230 230 230
Biomass
Gas Dispatch 308 308 308 308 308 308 308 308 308
Gas Peaking Units 211 211 211 211 211 211 211 211 211
Total Existing Resources 111 123 092 999 939 954 104 996 996
PRS Resources
CCCT 280 280 350 350 431 677
Coal
Wind
Other Renewables
Conservation 103
Hydro UpQrades
Total PRS Resources 307 321 410 421 530 839
Net Positions 149 161 122
Planning Margins (%)24.20,15,23.18.17.20.14.13,
Avista Corp 2007 Electric IRP
Chapter 8- Preferred Resource Strategy
Figure 8.5: Loads & Resources Energy Forecast with PRS (aMW)
900
800
-- -- - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- -
700
600
500
1,400
300
200
000
C\I
r::::::J CCCT_Wind
Other Renewables
r:::::::J ConservationTotal Existing Resources
Total Obligations
C\I
..-
C\I
compared to other resource portfolios created by PRiSM.
In these comparisons the PRS will be represented by the
25/75 risk/cost portfolio to ensure an apples-to-apples
comparison (i., not biased by lumpiness).
LOAD & RESOURCE TABULATIONS
Preferred Resource Strategy resources balance the
company position over time, retaining the lowest possible
cost and risk mix of assets to meet customer needs. Table
8.4 and Figure 8.5 illustrate how our present energy
positions will be supplemented with PRS resources
to meet future load growth. Table 8.5 and Figure 8.
illustrate the same information for our capacity positions.
700
500
300
100
900
700
500
C\I
C\I
C\I
C")
C\I C\I
..-
C\I
..-
C\I
I'--
..-
C\I
The PRS affects the company s mix of resources
over time. Today energy needs are met with a mix
resources that is approximately two-thirds fueled by
hydro and natural gas. These resources will contribute
approximately the same level of energy in 2017; however
hydroelectric generation will fall from 35 percent in
2008 to 29 percent in 2017. Remaining needs in both
periods are met by coal, contracts, conservation and
renewable energy sources.
Hydro in 2008 represents approximately 50 percent
of the company s generating capacity. Gas- and coal-
fired plants account for approximately 25 percent and
Figure 8.6: Loads & Resource Capacity Forecast with PRS (MW)
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -..-
C\I C")
..-
C\I C\I C\I
2007 Electric IRP
I:::::J CCCT
Other Renewables
r::::::::J Conservation
Total Existing Resources
Total Obligations
8 -
----------..-
C\I
..-
C\I
..-
C\I
I'--
..-
C\I
Avista Corp
Chapter 8- Preferred Resource Strategy
Figure 7: Company Resource Mix (% of Energy)
. Hydro
0 Conservation
. Contracts
III Coal
. Other Renewables
. Biomass
0 Gas
. Wind
Figure 8: Company Resource Mix (% of Capacity)
. Hydro
0 Conservation
. Contracts
iii Coal
. Other Renewables
. Biomass
0 Gas
Avista Corp 2007 Eiectric IRP 8 -
Chapter 8- Preferred Resource Strategy
10 percent, respectively. Contracts and non-hydro
renewables complete the capacity mix. The 2017
resource mix is more heavily weighted toward gas-fired
generation, as our hydro base does not grow and wind
generation is not included in our capacity tabulation.
See Figures 8.7 and 8.8 for charts of energy and capacity
mixes in 2008 and 2017.
CAPITAL REQUIREMENTS OF THE PREFERRED
RESOURCE STRATEGY
FRS capital requirements equal approximately $782
million between 2008 and 2018. This amount could
increase by as much as 50 percent when the company
finds that the best method for acquiring fIXed-price
gas involves investments in gas fields, a coal gasification
facility and/or other capital-intensive strategies. Table 8.
illustrates the annual capital investments necessary to
support the FRS absent investments in fIXed-price gas.
ANNUAL POWER SUPPLY EXPENSES AND VOLATILITY
Power supply expenses including fuel, variable O&M
and carbon compliance will grow over time at a
compounded annual rate of9 percent between 2008
and 2017; however, market conditions will likely affect
this rate of growth, making some years higher and some
lower. This level might appear high to the casual reader
but this figure does not equate to changes in retail rates.
Retail rate effects will be mitigated by higher retail sales
and lower escalation in non-power supply portions of
our business. The IRP forecasts that the average FRS
change on per-MWh power supply costs will equal 6.
percent per year. This increase should translate into
even lower retail rate impacts, as non-production costs
are expected to increase at a slower rate. Figure 8.
illustrates forecasted annual power supply expenses from
2008 through 2017.
The trade-off for rising power supply expenses is lower
year-on-year volatility. Power supply expense risk
decreases as new resources are brought on-line. Figure
10 illustrates the falling trend in risk measured by the
coefficient of variation of power supply expenses.
CARBON FOOTPRINT
The company has one of the smallest carbon footprints
in the United States because of its renewable energy
resources. Of the top 100 producers of electric power
in the 2006 Benchmarking Air Emissions study by the
Natural Resources Defense Council, only seven other
utilities have a smaller carbon footprint. The company
carbon footprint is forecast to increase over the IRP
timeframe, as it would be nearly impossible to acquire
all future resource requirements from non carbon-
emitting resources. Our per-MWh emissions will remain
essentially flat, and the carbon intensity of our thermal
fleet will fall as natural gas plants are added. Figure
11 forecasts our carbon footprint explaining that our
resources will emit approximately 2.5 million tons of
carbon dioxide in 2008, rising to 3.75 million tons by
2017. Figure 8.12 illustrates our emissions on the basis of
total sales, total generation, and thermal plant generation.
The 2007 FRS emits approximately 6 million fewer tons
2008
2009
2010
2011
2012
2013
27.201498.4 2015
247.2016
36.2017
Net Present Value
60.
270.
37.
249.
218.
781.
3 Coefficient of variation is calculated as the standard deviation of power supply expense divided by the expected (mean or average) power
supply expense in each study year.
8 -2007 Electric IRP Avista Corp
Chapter 8- Preferred Resource Strategy
Figure 8,9: Annual Power Supply Expense ($Millions)
600
- Max & Min
A\erage
. 80% Interval
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
500
400 ~
- -
200
300
100
CJ)
......
C')
......"'".....
Figure 8,10: Annual Portfolio Volatility (%)
Power Supply Expense Change
Coefficient of Variation
- - - - - - - ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ - - - - - - - - - - - - - - - - - - - - - - - - - - --------------------~--
CJ)
............
C')
"'".................
Figure 8.11: Forecasted CO2 Tons of Emissions (Thousands)
000
500
000
500
000
500
000
500
..................
C')
"'".......................
Avista Corp 2007 Electric IRP
Chapter 8- Preferred Resource Strategy
Figure 8,12: Forecasted CO2 (Tons/MWh)
------------
0.40
-+-CO2 Tons per MWh (Thennal Generation)
........CO2 Tons per MWh (All Plants)
_CO2 Tons per MWh Load
- -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
C\I C\I
......
C\I
......
C\I
C\I Ct')
C\I
"'"......
C\I C\I
Figure 8.13: Efficient Frontier and Traditional Resource Portfolios
2008 to 2017 Total Cost Net Present Value ($Millions)550 1 650 1 750 1,850 1 950 2 050 2 1501,450
220
...
0 b
& :: 180
I: 'Iiica -J: 0
(,) :::-::-.
i Q. 140
.. ::::IIII In
II. ..
E ~ 100N ~
seer & Green Tags CCCT & . G
'j; -- -- ~-
- r~en ags
, -- -- :
O%Risk
- - - - - - - - - - - - - - - - -
.- PRS f\b FIXed Gas Wind 20%
- - - - - - _:- - - - - - - ~ -- - - - - -
Cont. to PM Renew abies
Coal Allowed . &CT
, ". '
2005 PRS
- - - - - - -- - - - - - - -; - - - - - - - ~ - -
25O/~kc.- ---- PRS
- - .- - - -:- - ~- - - - ~ - - - - - - -
, 100% Risk
f\b Additions
+--
85 90 95 100 105 110 115
2008 to 2017 Total Cost Percent Change from 75% Cost/25% Risk
of CO2 from 2008 to 2017 than the 2005 PRS.
EFFICIENT FRONTIER ANALYSES
When developing a resource portfolio, two key
challenges must be addressed-how the portfolio
mitigates future costs and how it mitigates year-to-year
volatility. An efficient frontier identifies the optimal
level of risk given a desired level of costs and vice versa.
This approach is similar to finding the best mix of risk
and return when developing a personal investment
portfolio. As the expected average return increases, so
do risks; reducing risk reduces overall returns. Finding
the PRS is very similar to this investors dilemma, but the
......
C\I
.....
250
'iii'.... I:..... 00 =
75~ i
411-1:-0 -
.-
1/1- 0
60 .
!!! (,)
~ b
"C Q... ::::Ica In
45 -g ~
J!! ~In 0II.
120
trade-off is expected average future power supply costs
against future power supply cost variation. Figure 8.
presents the change in cost and risk from the Preferred
Portfolio Strategy on the Efficient Frontier. It also
shows alternative resource portfolios to illustrate various
generic resource strategies. The lower horizontal axis
displays the 2008-2017 percent change in the present
value of existing and future costs from where the PRiSM
model weights its optimization goals 75 percent to cost
reduction and 25 percent to risk reduction (75/25 cost/
risk). The upper horizontal axis presents actual present
value dollars. The right-hand vertical axis shows power
supply volatility as a single standard deviation of the
2007 Electric IRP Avista Corp
Chapter 8- Preferred Resource Strategy
average power supply expense. The left-hand vertical axis
shows the percent change in 2017 power supply volatility
from the 75/25 cost/risk point.
The blue dots represent the efficient frontier of various
resource portfolios developed by PRiSM to meet future
company requirements. Recall that the PRS is not
on the efficient frontier because resource lumpiness is
assumed in the first 10 years of the study. It is based on
the 75/25 portfolio weighting.
ALTERNATIVE FUTURES
The 2007 IRP studied alternative stochastic futures to
measure how the PRS would perform under different
assumptions. Figure 8.14 illustrates these differences.
This chart is similar to Figure 8.13, but it shows how the
efficient frontier would change from the Base Case given
the following three futures:
. unconstrained carbon emissions;
. more volatile natural gas prices; and
. high future carbon constraints.
Figures 8.15 through 8.17 provide a more detailed
comparison of each future, and display the performance
of the various portfolios chosen by the company.
'0 b 260
CD =
0;;
ca ~
c3 ~ 210:0-s: -
B 8:
:& ~
160
D.. ...
~ ;
C) 0('II D.. 110
ALTERNATIVE PORTFOLIO STRATEGIES
This chapter details how the company could serve
future needs using alternative resource portfolios. It
helps benchmark the efficient frontier and the Preferred
Resource Strategy. These portfolios, like the efficient
frontier, assume the company could acquire resources in
perfect increments (i., no lumpiness) and that green tags
are available to meet the Washington State Renewable
Portfolio Requirement. Each portfolio s costs and
benefits are compared to the Preferred Resource Strategy.
The specific resource contributions for each portfolio are
detailed in Table 8.
NO ADDITIONS
This portfolio theoretically assumes that the company
would not acquire any additional resources and instead
would rely on the market for all future capacity and
energy needs. Figure 8.18 shows that this is the lowest
absolute cost portfolio, however, it has the highest level
of risk. Graphically this strategy looks attractive because
it sits to the left of the efficient frontier, but it ignores the
company s responsibility to adequately meet its customer
requirements.
310
Figure 8.14: Efficient Frontier for All Futures
- - - - - - - - J- - - - - - - -
~ - - -
- - - - - - ~ - - - - - - - - J- - - - - - - -
: :-- -- - -- - --- -- - - -
- - ~ - - - - -tit - - - ~ - -
-- - - -- -: - -- - - - -- - +.' : -- - - - - - - - -,- - -.- - - - - "
- - - - - - - r - - - - - - - - -
. - - - - - - -,-
- - - - - - - - - r - - - - - - - - -
: , ~, ., ':--: --.--:,... ~ ..:---. , ...'
90 95 100 105 110
2008 to 2017 Total Cost Percent Change from 75% Cost/25% Risk
115
. Base Case
. Vol Gas
CSA
. No CO2 Taxes
A 75% Cost 25% Risk.PRS
Avista Corp 2007 Eiectric iRP
Chapter 8- Preferred Resource Strategy
...
0 ~ell = 180
CI ,-
'IiiI'a
c3 ~:0-
i 140'"' ::Iell !/)D.. ...... II)
~ ~
100
D..
...
~ ~ 180CI =
! ~ 160
(..) ~
~ 1140
.. ::I
:. !/) 120
.... m
~ ~
100
D..
...
~ ~ 180
CI =C -
.! -5 160
(..) ~
~ 1140
.. ::I
:. !/) 120
.... ~~ ~
100
D..
Figure 8.15: Unconstrained Carbon Future s Efficient Frontier Portfolios
220
N:J Addnions scx:r & Green Tags ax:T & :
.:
Green Tags
- - - - - - - ~ - - - - - ~ - ~ - - - - - - - ~ - - - - - - - - ~ - - - - - - - ~ - - - - - - - - ~ - - - - - - - ~ - - - - - - - -, '
0% Risk
FRS- J\b FIXed Gas
- - - - - - - -- - --- ---:--:--- --
~- ~~~t
~ -
:- R;n~~~bl~~cr - ---
- -:---- -
. I CoalAliowedI 'FRS
- - - ~ - - - - - - - ~ - - - ~ - ~.- -.-. : - - . - - - - - ~ - -. - - - - -: - - - - - - - -
I 25Vo Risk I 2oo5FRS I
: 50% Risk 100% Risk
90 95 100 105 110
2008 to 2017 Total Cost Percent Change from 75% Cost/25% Risk
115 120
Figure 8.16: Climate Stewardship Future Efficient Frontier Portfolios
220
200
- - - -- - - ~- - - - - - - - ~ - - - - - - - ~- - - - - - - - ~ - - - - - - - ~- - - - - -- ~ - - - - - - - ~- - - - - - --, -
N:J Addnions scx:r & Green Tags
e - --
~ - -- -
e -- ax:T&
0% Risk. Green TagsI .
- - - - - - - ~ - - - - - - - - ~ - - ,- -
- FRS- N:J -iMnd 20% - - - - -
; - - - - - - - - ~ - - - - - - - - - - - - - - --- - - - - - - _: - - - - - - - - ~ - - - - ~ .;~~ -~~ -
Cont. to RIII
- - - - -
Renewables & cr
- - - - - _: - - - - - - -
Coal Allowed. .I '
.. . - - -: - - - - - - - - ~ - - - - - - - -: - - - - - - -~ -
- FRS:. - - - - - - - ~ - - 2005 FRS
- - - - - - -
25% Risk 50% Risk ' 100% Risk
- - - - - - - ~ - - - - - - - - ~ - - - - - - - ~ - - - - - - - - r
- - - - - - - - - - - - - - - - - - - - - - - - - - -- -
90 95 100 105 110
2008 to 2017 Total Cost Percent Change from 75% Cost/25% Risk
115 120
Figure 8.17: Volatile Gas Future Efficient Frontier Portfolios
220
200 N:J Addnions scx:r & Green Tags
- - - - - - - - -
1- - - - -
- - - - - - - - - - - - - - ~ - - - - - - - - -
1- - - -
- - - - - - - - - - - - - - - ~ - - - - - - - - -. '. \ ,
ax:T&
- - - - - - - - -
0% Risk -
. - -- - - - - - - - - - -- - - - - - - - - -, - - - - - - - - - ~ - - - - - - -
Green Tags
- - - - - - - - -: - - - - - - -
. - - - - - FRS- N:J ~lXed Gas - - - -
:- - - - - - - - - - - - - - - - - - - ~ - - - - - - - - -
, tit
- - - - - - - - -:- - - - - - - - - ~ - - - -. - - - -
~~ - - - Wnd 20%
- - - - - - - - -
Renewables - - - - -
- - - - - - -
Cont. to RIll & cr
- - - - - - - - _- - - - - - - - - - - - - - - - - - ~- - - - -. - - - _- - - - - - - - - -' - - - - - - - - - ~ - - - - - - - - -
FRS . ' 1
:....- - - - - - - - - -:- - - - - - - - - ~ - - - - - - - - - ~ - - - -
. - - - -6 - - - - - - - -
. - - - - - - - - - ~ - - - - - - - - -
Coal Allowed 25% Risk 100% Risk
- - - - - - - - -:- - - - - - - - - - - - - - - - - - - ~ - - - - - - - -
59% Risk - - -
- - - - - - - - - - -
. ~ - 2005 FRS -90 95 100 105
2008 to 2017 Total Cost Percent Change from 75% Cost/25% Risk
110 115
8 -2007 Electric IRP Avista Corp
Chapter 8- Preferred Resource Strategy
Figure 18: Net Present Value of New Resource and Power Supply Costs by Portfolio
(2007 $Millions)
No Additions
100/0 CosURisk
CCCT
PRS- No Fixed Gas
Wind with 20% CC
Coal Included (75/25)
75/25 CosURisk
75/25 CosURisk RPS
PRS
50/50 CosURisk
25/75 CosURisk
Wind & CT
2005 IRP
0/100 CosURisk
000 000
SIMPLE CYCLE CTS AND GREEN TAGS
This portfolio assumes that the company would acquire
only simple-cycle gas turbines to meet future capacity
needs. Given the high operating costs of these plants
this scenario is actually one where future energy needs
are met through purchases from the volatile wholesale
electricity marketplace. The turbines sit idle a vast
majority of the time. The portfolio meets our capacity
needs unlike the No Additions Portfolio, but it still
contains a high level of volatility due to its heavy reliance
on the marketplace and natural gas. The PRiSM model
identified the timing of SCCT construction to meet
the objectives of this portfolio. Renewable energy
requirements are met by acquiring green tags.
COMBINED CYCLE CTS AND GREEN TAGS
This portfolio assumes that the company only acquires
combined-cycle gas turbines to meet its capacity and
energy needs. The PRiSM model identified the optimal
amount and timing of resource additions to meet
this portfolio objective. Capacity targets are met and
market risk is reduced compared to relying on less-
efficient simple-cycle CTs. Green tags meet our RPS
requirements.
000 000 000 000000
RENEWABLES AND SIMPLE-CYCLE CTS
Future requirements are met only with renewable
resources and simple-cycle CTs in this strategy. The
PRiSM model identifies the optimal amount and timing
of resources to meet this portfolio objective. SCCTs
are included to meet capacity needs, and renewables are
added to serve energy needs and reduce risk. This green
portfolio requires a 1 200 MW wind penetration level
over the next 20 years. Power supply cost variability is
reduced in exchange for higher power supply expenses.
COAL ALLOWED
This portfolio allows coal to be selected by the PRiSM
model rather than fIXed price natural gas plants. The
portfolio is based on the same risk level as the PRS.
The portfolio is made up of a combination of wind
combined cycle CT, other renewables and coal. Coal
is selected after 2013, but not before the 2011 resource
need that is met by a combined cycle CT. Because non-
sequestered coal is not allowed in our analyses except in
this one-off for comparative purposes, this portfolio has a
superior performance to the efficient frontier.
Avista Corp 2007 Electric IRP 8 -
Chapter 8- Preferred Resource Strategy
Base Case: PRS
CSA
Unconstrained CO2
Volatile Gas
Base Case: PRS
CSA
Unconstrained CO2
Volatile Gas
WIND CONTRIBUTES 20 PERCENT TO CAPACITY
PLANNING MARGIN
The IRP assumes that wind generation will provide
no capacity to the portfolio in the near- to medium-
term. This assumption is based on a wind integration
study completed by the company in March 2007.
Ignoring this result and assuming a 20 percent capacity
contribution for wind makes it much more attractive
though it still sits above the points of the efficient
frontier. This portfolio quantifies the impact of the Base
Case wind capacity assumptions.
IMPACT OF RPS REQUIREMENTS ON THE PRS
RPS sensitivity portfolios were developed to illustrate
the impact of renewable resource cost increases on the
level of renewable resources ultimately included in
the FRS. The portfolio analysis is based on the 75/25
cost/risk weighting mix, the same as assumed in the
FRS. The analysis found that in the Base Case, without
a Washington state RPS, the resource strategy would not
change under any of the market futures. This indicates
that renewables were selected primarily to reduce risk
and not to meet the RPS targets. In the unconstrained
2 future, fewer renewable resources are built. The
model purchases green tags because absent the RPS
fewer renewables would be selected. See Table 8.
4 All cases limit wind to 400 MW of capability between 2008 and 2017.
8 -
If the company had an RPS requirement in Idaho that
mirrored the Washington state requirement, the amount
of renewables in our portfolio would not increase
significantly. Instead, we likely would purchase green
tags, as illustrated byTable 8.8. The RPS would cause the
company to build renewable resources that it otherwise
might prefer not to.
RISK-ADJUSTED PORTFOLIO STRATEGIES
Portfolios were selected from the Efficient Frontier
to illustrate various resource combinations and their
performance under alternative market scenarios and
futures. Utility-specified portfolios were created to help
describe the benefits and risk of certain resource mixes.
The portfolios' performances are shown in the figures
below.
The charts quantifY each portfolio s cost, risk and other
factors on a comparative basis. The focus of these charts
is on the 2008-2017 time period, but some information
is provided for the entire 20-year study. These charts are
for the Base Case only. The same information for each
market future is provided in the IRP Appendices.
Table 8.9 first provides an overview of the resources
included in each alternative portfolio. Figure 8.18 shows
the present value of each portfolio s incremental costs
2007 Electric IRP Avista Corp
Chapter 8- Preferred Resource Strategy
Ta. b.le- 8: ,9' : 201018: -17 R- e- so. urce- 5 fo. rEa. ch P- o. rtfo.lio. Ca.. a. b. ilit
Other
Renew-Pulverized Hydro
Portfolio SCCT Wind abies Coal CCCT DSM Upgrades Total
0/100 Cost/Risk 400 350 910
25/75 Cost/Risk 400 350 910
50/50 Cost/Risk 400 350 910
75/25 Cost/Risk 300 350 810
100/0 Cost/Risk 363 507
2005 IRP 650 140 350 265
CCCT 384 509
Coal Included 365 127 228 880
382 507
No Additions 125
PRS 300 460
PRS w/o fixed
!:Ias 300 350 810
RPS 307 467
Wind & CT 350 675 185
Wind & 20% CC 273 433
Figure 8,19: Volatility (Coefficient of Variation) of 2017 Power Supply Expenses (%)
0/100 Cost/Risk
25/75 Cost/Risk
50/50 Cost/Risk
75/25 Cost/Risk RPS
75/25 Cost/Risk
PRS
2005 IRP
Coal Included \15/25)
Wind with 20% CC
Wind & CT
PRS- No Fixed Gas
CCCT
100/0 Cost/Risk
No Additions
including new capital and O&M. The costs represented
by the blue area of the chart bars are the same as those
used on the x-axis of the efficient frontiers.
Risk in the 2007 IRP is measured by the volatility
of annual power supply expenses, driven by modeled
variations in natural gas costs, loads, emission uncertainty,
hydro conditions and forced outages. Figure 8.
illustrates volatility by displaying the coefficient
variation for each portfolio.
The FRS has lower risk because of the investment into
capital intensive and fIXed priced assets. The expected
power supply costs for 2017 are shown in Figure 8.20.
Customer rates will be impacted by new resource
investments. Actual rate increases are likely to be lower
because power supply expense is only one contributor to
rate base. Average power supply cost increases by
scenario are shown in Figure 8., and the highest
single-year increases are shown in Figure 8.22.
5 The coefficient of variation is calculated by dividing the standard deviation of the total annual cost by the expected power supply cost.
Avista Corp 2007 Eiectric IRP
Chapter 8- Preferred Resource Strategy
Figure 8.20: 2017 Total Power Supply Expenses ($Millions)
No Additions
100/0 CosURisk
CCCT
Wind with 20% CC
PRS- No Fixed Gas
2005 IRP
PRS
75/25 CosURisk
75/25 CosURisk RPS
25/75 CosURisk
50/50 CosURisk
Coal Included (75/25)
0/100 CosURisk
Wind & CT
200 250 300 350 400 450 500 550
Figure 8.21: Average Annual Power Cost Component Change 2008-2017 (%)No Additions
100/0 CosURisk
CCCT
Wind with 20% CC
PRS- No Fixed Gas
2005 IRP
PRS
75/25 CosURisk
75/25 CosURisk RPS
25/75 CosURisk
50/50 CosURisk
Coal Included (75/25)
0/100 CosURisk
Wind & CT
Figure 8.22: Maximum Annual Cost Change for Power Supply (%)
100/0 CosURisk
Coal Included (75/25)
CCCT
No Additions
75/25 CosURisk RPS
25/75 CosURisk
50/50 CosURisk
75/25 CosURisk
Wind with 20% CC
Wind & CT
PRS- No Fixed Gas
0/100 Cost/Risk
PRS
2005 IRP
Avista Corp2007 Electric IRP
Chapter 8- Preferred Resource Strategy
Figure 8.23: 2008-2017 NPV of Capital Investment (2007 $Millions)
No Additions
100/0 Cost/Risk
CCCT
Wind with 20% CC
75/25 Cost/Risk
75/25 Cost/Risk RPS
PRS- No Fixed Gas
PRS
50/50 Cost/Risk
25/75 Cost/Risk
0/100 Cost/Risk
Coal Included (75/25)
Wind & CT
2005 IRP
200 400 600 800 000 200 1,400 600
Figure 8.24: Renewable Resources Included in Each Portfolio (Nameplate MW)
No Additions
CCCT
100/0 Cost/Risk
Wind with 20% CC
PRS- No Fixed Gas
75/25 Cost/Risk
PRS
75/25 Cost/Risk RPS
Coal Included (75/25)
0/100 Cost/Risk
50/50 Cost/Risk
25/75 Cost/Risk
2005 IRP
Wind & CT
200 400
Additional capital will be required to meet future
load growth. Each portfolio has a unique capital
requirement. Figure 8.23 shows the present value of
capital requirements for each portfolio option. Capital
requirements shown on this chart are for resource
capital only and do not include associated capital or debt
equivalents needed to firm the price of natural gas
recommended in the FRS.
Figure 8.24 presents new renewable resources included in
each portfolio between 2008 and 2027. These values are
shown in nameplate capacity, not energy or contribution
to system planning margins.
. 2008-2017
.2018-2027
600 800 000 200 1,400
PLANNING CRITERIA
The Northwest continues to debate the proper level
of planning reserves utilities should carry above their
expected peak demand. We also have evaluated
eliminating second quarter resource surpluses to ensure
that resource deficiencies in the remaining three quarters
of the year are not masked by an annual average position
covered with excess second quarter hydro energy. This
planning level would be similar to moving from an 80
percent to a 95 percent confidence interval planning
level.
The FRS currently meets a planning margin equal to 10
percent above expected peak load, plus 90 MW Energy
Avista Corp 8 - 212007 Eiectric iRP
Chapter 8- Preferred Resource Strategy
Figure 8.25: Alternative Resource Planning Criteria (Efficient Frontier Results)
200
....
~ ~160
CI ==
:a ~
6 ~ 140
-;:~
~ 8: 120
... ::s
:. I/)
t:: ; 100
Q 0N a..
- - - - - - - - T - - - - - - - -
, - - - - - - - - - -- - - - - - - - -- - - - - - - - - : - - - - - - - - : - - - - - - - - -. . . . . . . .
I... ..
..
.1\ J .... e f .... r.. .
..... ........ . .:.........
I .. . Base Case
- - - - ~ -- -- '- - ~ - -- -- - - - -:-- - - - -- - - ~ - - - - - - - --
. No 2 Ener Contribution I . 'I :"
("---. '., . ~;~ :::~~:~~ ~:~:~ - - - - - : - - - - - - -~ ~- - j ~~ - - - . - - -.- : - - - - - - - - -
& 75% Cost 25% Risk
-- - -- ~ -- -- -- -- - ~ -- -- -- -- _:--~ _
4!~
-~ _-- - --
ePRS
180
85 90 95 100 105
2008 to 2017 Total Cost Percent Change from 75% Cost! 25% Risk
115
planning margin is currently based on an 80 percent
confidence level of historical hydro and load variance on
an annual basis. An analysis was performed to quantify
the cost and risk of moving to alternative planning
methodologies. Three planning criteria alternatives were
modeled:
. 15 percent planning margin;
. 25 percent planning margin; and
. exclude second quarter energy from the annual
forecast need.
Each of these alternatives has a different impact on
resource acquisition, costs and risks. Figure 8.25 shows
the impacts using efficient frontiers. If the company
moved to a 15 percent planning margin, there would
be little impact on future risks or costs compared to our
current methodology. If the company built additional
capacity to meet a 25 percent planning margin, as the
NPCC recommends in its draft resource adequacy
target, costs would probably increase and risk might
decrease if the selected incremental resources were
one of the lower-risk options. Alternatively, where the
110
company simply met a higher planning margin with
market purchases or spot gas-fueled plants, no additional
benefit would be seen by moving from a 15 percent
to a 25 percent planning margin. Removing second
quarter energy surpluses from the company s load and
resource position would simply increase costs without a
commensurate risk reduction benefit.
CAPITAL COST SENSITIVITIES
Resource capital costs have increased substantially since
the 2005 IRP. The largest impact in this plan is a 50
percent reduction in the amount of wind generation
stemming from an approximate 50 percent increase in
capital costs for wind resources. The Efficient Frontier
can illustrate the impact of varying levels of capital cost.
Table 8.10 identifies the capital cost sensitivities studied
for this IRP. These sensitivities determine how changes
would impact not only the cost of the efficient frontier
but how our resource selections might change.
The sensitivity results are informative and explain
that overall power supply costs change in response to
Wind
Combined C cle
IGCC Coal wISe uestration
Alberta Oil Sands
300
600
500
000
884
786
232
963
500
000
N/A
N/A
8 - 22 Avista Corp2007 Electric IRP
Chapter 8- Preferred Resource Strategy
Base Case
Other
Wind
CCCT
(GCC wISe uestration
Alberta Oil Sands
IGCC $2,500/kW
Other
Wind
CCCT
IGCC wISe uestration
Alberta Oil Sands
Oil Sands $2 OOO/kW
Other
Wind
CCCT
(GCC wISe uestration
Alberta Oil Sands
varying capital cost levels; however, the variations did not
significantly change the overall strategy during the first
10 years of the plan. The one exception is where wind
costs vary significantly. See Table 8.11. Lower wind
acquisition is offset by more green tag purchases.
Sequestered IGCC coal and Alberta Oil Sands would be
selected at the expense of gas resources if their capital
costs were to fall significantly from what is assumed in
the Base Case. See Table 8.12.
FIXED GAS PRICE
Coal-fired generation accounted for a significant
portion of the Avista s FRS mix in both the 2003 and
2005 IRPs. Coal-fired plants provide a hedge against
volatile electricity and natural gas prices because 60
percent or more of their costs are fIXed through large
capital investments. Variable operating and fuel costs at
600 600 600 600
677 657 527 350
130 101
226
600 600 600 600
280
299 101
226
600
467
210
600
350
226
600
350
101
226
600
451
226
a coal plant are modest compared to gas-fired resources.
A resource profile containing coal contributes to stable
power supply expenses.
The cost of operating gas-fired resources, on the
other hand, is highly correlated with the electricity
marketplace. Natural gas prices are very volatile. The
fIXed costs of natural gas plants are low relative to their
all-in cost of generation, approximately 20 percent
reflecting a low capital investment. Utility portfolios
with large concentrations of gas-fired generation suffer
from rates that are less stable than utilities that rely on
other sources of generation.
I .
Gas-fired plants have not experienced the same
capital cost increases seen in new coal-fired plants.
In fact, recent experience by Avista (Coyote Springs
2) and Puget Sound Energy (Goldendale) indicate
that independent power producers in the Northwest
Avista Corp 2007 Electric IRP 8 - 23
Chapter 8- Preferred Resource Strategy
marketplace are willing to sell their gas-fired plants
at prices below the green field costs assumed in this
plan. The enactment of new laws imposing emission
performance standards on fossil-fueled generation
resources acquired by electric utilities in Washington and
California will narrow base load technology options
at least in the short-term, to gas-fired generation. This
restriction, coupled with regional load growth and the
prospect of additional greenhouse gas regulations on
fossil-fueled generation resources, particularly coal-fired
generation, may ultimately increase demand for and the
cost of gas-fired plants.
Locking in natural gas costs through a long-term
fIXed-price contract, an investment in a pipeline-quality
coal gasification plant, an investment in gas fields or
through other means makes a gas-fired combined cycle
combustion turbine (CCCT) behave financially like a
coal-fired resource. Variable costs are greatly reduced and
are much less volatile because a significant portion
its largest variable component-gas fuel-is not tied to
the natural gas market. In both high and low gas market
conditions the price paid by customers is the same. In
years where natural gas prices are high, the fIXed-cost
contract looks very attractive financially and customers
pay less than if the company relied on shorter-term
purchases. On the other hand, years with low natural
gas prices make the fIXed-cost contract look financially
unattractive compared to a short-term purchase. Over
time, the long-run cost of operations with fIXed-price gas
should parallel the cost of operations where a gas plant is
fueled with short-term gas.
Fixing gas prices does not lower absolute cost, but it does
limit price volatility. As with any long-term fIXed price
option, prices over time likely will be higher than if the
company relied exclusively on spot market gas purchases.
Asking a third party to absorb price risk always entails
a premium in exchange for accepting that risk. This is
similar to purchasing an automobile insurance policy.
A policy is not purchased to lower driving costs but to
decrease the amount of financial risk to the driver if
an accident were to occur. A financially-fIXed natural
gas price would be higher than average spot market gas
purchases, but that premium would limit the upside
exposure of the company and its customer to gas price
spikes.
The company has identified three potential avenues to
lower natural gas price risk. There might be more. The
first, and most probable option, would involve purchasing
a long-term fIXed price gas contract. Until recently, the
market did not offer these types of contracts because
of experiences in the 2000/01 energy crisis. Recent
Figure 8,26: Efficient Frontier With and Without Fixed Price Gas Contract Option
200
~T-- --
~~\- --~~--
r--
, ...!--!--.~.! .- -:--: --~ ----,--- .:-----: ::::
~:: No Fixed Priced Gas
- ~ - - - - ~~~
- i
.+- - -! - - - -. - - - - ~ - - - - - - - --
.. 75% Cost 25% Risk - L -- - - -
- - - - - - - - - -- ---- - - - -- -- -- -- -- -- --
.PRS ,
180
....
0 ~
& ii 160I'GI'G .c 0(.) ::- 140
::-.
c -
~ ~ 120
:. In
...~ ~
100
('oj ~
90 95 100 105
2008 to 2017 Total Cost Percent Change from PRS
115
- - -
110
8 - 24 Avista Corp2007 Electric IRP
Chapter 8- Preferred Resource Strategy
informal market surveys have found sellers offering terms
up to 20 years. A second option would involve investing
in a gasification plant to convert coal to pipeline-quality
gas. A third option would be investment in a gas field.
The company tested the benefits of fIXed price contracts
with PRiSM and found a general preference for fIXed
price gas because of its ability to reduce risk. Even with
premiums as high as 75 percent above the short-term
gas prices, the PRiSM model selects fIXed-price gas for
a portion of the preferred portfolio. In the Base Case
where a 30 percent fIXed gas price premium is modeled
risk is reduced by approximately 20 percent, as shown in
Figure 8.26.
AN EMPIRICAL EXAMPLE
Avista has historically purchased fuel for our gas-fired
plants in the short- to medium-term markets, making
purchases from time periods as short as one day up
to 18 months into the future. Generation costs have
varied greatly over this time with the price of natural
gas. Figure 8.27 illustrates historical montWy natural
gas prices at the Stanfield hub, where Coyote Springs 2
procures its natural gas. Prices are shown from January
2002 through March 2008.
As shown, gas prices have been quite volatile. Gas prices
ranged from a low of$1.52 per Dth to a high of$11.29
per Dth. Translated to montWy gas expense, a company
model shows the cost ranges from zero in four months
where market conditions did not support operating the
plant, to as high as $14.4 million in December 2005.
The standard deviation of this hypothetical cost stream
large, at $2.9 million, or 62 percent of the average.
Greater reliance on gas-fired generation has the potential
to introduce significantly more volatility in company
power supply costs than has been witnessed in the past.
The first ten years of the PRS acquires 350 MW of
CCCT capacity, more than doubling the size both of our
CCCT fleet and gas purchasing budget. To illustrate, a
$1.72 per Dth annual increase in natural gas prices would
drive up fuel expenses by approximately $21 million at
Coyote Springs 2; with an additional 350 MW of gas-
fired CCCTs, the exposure would be $48 million.7 The
largest annual swing in gas prices over this period was
Figure 8,27: Historical Monthly Gas Prices at Stanfield ($/Dth)
- - -
C')C')C')C')L()L()L()L()I'-
!!:!!:!!:!!:!!:!!:....,
0::(
....,....,
0::(
....,....,
0::(
....,....,
0::(
....,....,
0::(
....,....,
6 Assuming theoretical operation absent both maintenance and forced outage costs.
7 $1.72 per Dth equals one standard deviation of annual Stanfield natural gas prices between 2002 and 2006. Price swings would be
expected to exceed this amount in one in three calendar years. 160 dth/MW * 280 MW * 365 days * 75 percent capacity factor *
$1.72/Dth = $21.2 million; 160 dth/MW * 630 MW * 365 days * 75 percent capacity factor * $1.72/Dth = $47.8 million.
Avista Corp 2007 Eiectric IRP 8 - 25
Chapter 8- Preferred Resource Strategy
$2.22 per Dth between 2002 and 2003. Reviewing the
2002 through 2006 period, history shows a $48.4 million
range in annual gas procurement costs, and a maximum
year-on-year change of as much as 50 percent. Hedging
a portion or all of our natural gas purchases might reduce
fuel expense volatility by 50 percent where the 2002
through 2006 years provide guidance.
DECIDING THE QUANTITY OF NATURAL GAS TO HEDGE
One challenge of fIXing natural gas prices is deciding how
much of a plant's portfolio should be hedged. Should all
expected generation be hedged? Should the hedge be
placed equally across all months of the year, or differently
in each month to reflect expected generation levels? As
discussed earlier, fIXing gas prices likely will incur higher
average cost. This is illustrated by Figure 8.28. The
lowest average cost is where the plant does not hedge
any of its gas costs with fIXed prices. The mean variable
fuel cost of the plant is approximately $40 per MWh
with a range of $10 to $85 in any given year of the study.
Hedging 25 percent of natural gas consumption reduces
the expected range of operating costs by about a third
but raises the average variable fuel cost of the plant to
about $45 per MWh. Hedging 75 percent of natural
gas consumption tightens the distribution of costs by 75
percent, but it also increases expected variable fuel costs
to $54 per MWh.
The answer to this question is too broad for resolution
in an IRP, and the company will further analyze the
question as part of its action plan. The IRP took a
simpler approach and assumed that the natural gas price
was fIXed for 75 percent of annual average expected
generation.
More analysis of fIXed price options is necessary to
confirm that a fIXed price gas strategy is in the best
interest of our customers. This work is included as an
action item for the 2009 IRP.
PORTFOLIO PERFORMANCE ACROSS
MODELED SCENARIOS
Resource portfolios perform differently in the different
market scenarios detailed in Chapter 7. For example
Figure 8.28: Variable Fuel Costs of CCCT Plant at Various Gas Hedging levels ($/MWh)
S 2.
81.
~1.
C!I
.CI
D.. 0.
Variable Gas
75% Fixed Gas
25% Fixed Gas
------------------ - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- -
8 This analysis is based on dispatching a CCCT plant during the years 2002-06 using daily average Mid-C and Stanfield natural gas prices.
In the case of fixed price gas, fixed price gas was assumed to be purchased in an amount equal to 75 percent of the annual operating
capability of the unit, approximately the level of operation the company would expect out of a CCCT plant. Purchasing between 60 and
75 percent of annual capability provides a similar result. The fixed price was set equal to the average price over the 5-year period. On days
in which the plant operated, the remaining 25 percent of needs not covered by the fixed purchase was purchased at the daily index price.
On days in which the plant was not economical to run, gas was sold into the spot market. Change in volatility is defined as the change in
the standard deviation of fuel expense.
8 - 26 2007 Electric IRP Avista Corp
Chapter 8- Preferred Resource Strategy
Figure 8.29: Portfolio Cost Comparison Versus PRS for Each Market Scenario (%)
1- -1-
: :. .: *
; d
- - - - + - - - - -i - - - - -1- - - - - f- - -t- - -i - - - - -i - - - - -1- - - -
- ~ - - - - ~ - - - - - - - - -
f - ~
- -
I - -:- -1- - ~ - - - - ~ - - - - J
: - - - - _- - - - - : - - - - ~ - - - -
! JIl 1 i 1 ,
t ~: ~ - - ~ J ~ ~ ~ ~ I ~ ~ ~ ~ ~ ~ ~ ~ ~ J ~ ~ ~ ~ J ~ ~ ~ ~ T ~
~ ~ ~: ~ ~ ~ ~ ~ ~ ~ ~ ~
1 I
fib 100/0
Additions Cost/Risk
a:::cr FRS- fib Wind with Coal 75/25 75/25 25(75 50/50 0/100 Wind & 2005 IRPFIXed 20% CC Cost/Risk Cost/Risk Cost/Risk Cost/Risk Cost/Risk Gas RPS
portfolios including higher concentrations of carbon-
emitting resources will perform poorly in a high-cost
carbon environment when compared to portfolios not
relying as heavily on them. The expected costs of gas-
reliant portfolios will vary more under low and high
gas scenarios than portfolios not relying on gas. The
performance of various portfolios studied in the plan is
displayed in Figure 8.29. The figure explains how the
different portfolios compare relative to the Preferred
Resource Strategy, when measured by the 2008-17 NPV
of total power supply expenses. For example, the "
Additions" portfolio is expected to cost as much as 20
percent less than the FRS (shown in this chart as the
25/75 Cost/Risk" portfolio) portfolio under the Low
Gas market scenario. The alternative s savings from the
FRS fall to 15 percent in the Constant Gas Growth
scenarIo.
Figure 8.29 identifies which portfolios are on average
lower and/or more costly than the FRS, and show which
portfolios' expected average costs are more volatile
compared across the market scenarios. Riskier portfolios
have a larger cost range while the performance ofless
risky portfolios does not vary much.
Risk across scenarios is not the same risk being measured
in the efficient frontiers displayed in this section.
- - -~- -"- -: -
& High Load
(:, Low Load
. Constant Gas Growth
. Low Gas
. High Gas
. BaseCase
X Unconstrained Carbon Future -
+CSA
Scenario and paradigm risks help explain how robust
portfolios are where significant changes from the Base
Case occur. Risk measured by the efficient frontier is
how well the portfolio behaves under varying stochastic
parameters (i., natural gas, forced outage, carbon price
and wind and hydro variations). The FRS-No Fixed
Gas portfolio best illustrates this difference. When shown
in Figure 8.29 it appears that the FRS with no fIXed gas
performs exceptionally well across the scenarios while
providing five-percent lower average costs than the FRS.
But in looking back at the efficient frontier of Figure
, not fIXing gas prices actually creates a higher risk
proftle than the FRS (by approximately 35 percent) in
the expected Base Case due to the portfolio s greater
exposure to shorter-term variations in natural gas prices.
THE LANCASTER GENERATION FACILITY
The company announced the sale of its energy
marketing company, Avista Energy, in April 2007. As part
of this transaction Avista Energy s tolling contract for the
Lancaster Generating Plant output will become available
to the utility beginning in 2010. The announcement
came after we had substantially completed our IRP
analysis and FRS. Given that Lancaster is the same
technology as the 280 MW gas-fired combined cycle
resource identified in the FRS at roughly the same
timeframe and is available to the utility, the resource
Avista Corp 8 - 272007 Electric iRP
Chapter 8- Preferred Resource Strategy
Table 8.13: Loads & Resources Ener Forecast with PRS aM
, ,:
Obligations
Retail Load 125 163 196 230 256 326 379 1,450 627
90% Confidence Interval 200 199 196 196 192 192 192 156 156
Total Obligations 324 362 392 425 448 518 571 606 783
Existing Resources
Hydro 540 538 531 528 512 510 509 491 491
Net Contracts 234 234 234 129 107 105 105 106 106
Coal 199 183 188 198 187 187 198 199 186
Biomass
Gas Dispatch 280 295 285 295 280 295 295 280 295
Gas Peaking Units 145 145 141 146 145 146 145 141 145
Total Existing Resources 446 442 426 342 278 290 299 265 270
Net Positions 121 170 228 272 341 513
PRS Resources
Lancaster 254 264 249 264 264 228
CCCT 162 612
Coal
Wind 103 103 103
Other Renewables
Conservation 103
Total PRS Resources 259 290 288 406 487 587 871
Net Positions 122 292 207 117 179 215 246 359
Figure 8.30: Loads & Resources Energy Forecast with Lancaster in PRS (aMW)
900
800
700
600
500
1,400
300
200
100
000
900
ex)
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
I!:::iiiiI Total Existing Resources
Other Renewables
" r:::::J Lancaster
Total Obligations
..--
strategy was not updated. Instead an alternative portfolio
with Lancaster is compared to the FRS to illustrate its
impacts. The Lancaster Generation Facility is a 245
MW gas-fired combined-cycle combustion turbine with
an additional 30 MW of duct firing capability. It is a
General Electric Frame 7FA plant that began commercial
service in 2001. Lancaster is located in Rathdrum
Idaho, in the center of Avista s service territory. It is
..--
C')-.:t 1.0 to-
..--
significantly lower in cost than a green field plant and
would not expose the company to construction risk.
LANCASTER IMPACT ON L&R BALANCES
Lancaster substantially replaces the identified gas-fired
CCCT included in the preferred resource strategy. Tables
13 and 8., and figures 8.30 and 8., present the
FRS with Lancaster replacing a significant portion of
8 - 28 2007 Electric IRP Avista Corp
Chapter 8- Preferred Resource Strategy
Table 8.14: Loads & Resource Ca acit Forecast with PRS MW
II:II'
Obliaations
Retail Load 703 763 815 868 909 019 103 214 2,492
Planning Margin 260 266 272 277 281 292 300 311 339
Total Obligations 964 029 087 145 190 311 404 525 831
Existing Resources
Hydro 142 154 121 128 084 098 098 070 070
Net Contracts 172 172 173 208 128 128
Coal 230 230 230 230 230 230 230 230 230
Biomass
Gas Dispatch 308 308 308 308 308 308 308 308 308
Gas Peaking Units 211 211 211 211 211 211 211 211 211
Total Existing Resources 111 123 092 999 939 954 104 996 996
Net Positions 148 146 251 357 300 530 835
PRS Resources
Lancaster 275 275 275 275 275 275
CCCT 156 677
Coal
Wind
Other Renewables
Conservation 103
Total PRS Resources 280 302 316 410 421 530 839
Net Positions 149 285 156 122
Plannina Marains (%)24.20.30.23.18.17.20.14.13.
700
Figure 8.31: Loads & Resources Capacity Forecast with Lancaster in PRS (MW)
500
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
300
100
900 r:::::=;J CCCT
c:::::::J Lancaster
Other Renewables
r:::::=;J Conservation
Total Existing Resources
Total Obligations
700
500
C\I C\I C\I
......
C\I
the CCCT needs identified for the FRS. The addition
of Lancaster pushes the company s resource need out to
2014.
LANCASTER IMPACT ON PORTFOLIO COSTS AND RISK
The Lancaster plant costs less than an equivalent new
gas-fired CCCT while providing the same benefits.
C\I
C\I
It)
......
C\I C\I
l"-
C\I
C')
......
C\I
"'"......
C\I
Another way to compare the addition of Lancaster to the
Preferred Resource Strategy is to plot a new FRS with
Lancaster s costs on the Efficient Frontier. Figure 8.
provides an updated efficient frontier where Lancaster
replaces a majority of the FRS gas-fired acquisition
during the first decade of the plan. Including Lancaster
reduces costs approximately 6 percent under the original
Avista Corp 2007 Electric IRP 8 - 29
Chapter 8- Preferred Resource Strategy
1,450
220
Figure 8.32: Efficient Frontier with Lancaster Plant
2008 to 2017 Total Cost Net Present Value ($Millions)550 1,650 1 750 1,850 1 950 2 050 2 150 250
& ~ 180
I: III111-.J::. u ;:.
::0.
~ Q. 140
u Co.. :::IIII U)
11. ..
~ ;
100N ~
No Additions seer & Green Tags CCCT &
+---:---- -
GreenTagS :
: 0% RiSk
:- - - - - - - - - -
. PRS No Fixed Gas Wind 20%
, .
ConI. to PM
Coal Allowed
i -- 25%
::-
'-~ ~s
, PR~h 50% Risk
Lancaster
- -
2005 PRS
100% Risk
..... I:... 0
75N ::
::E0 ~I: --0 -.- rn- 060 ,!!! u
::0.Q Q.
.. :::I45.g~
I: IIIJ!! ~U) 0
11.
- - - - - - - - - - - ~- - - - - -
Renew abies
. &CT
85 90 95 100 105 110 115
2008 to 2017 Total Cost Percent Change from 75% Cost/25% Risk
120
FRS for the same amount of risk. Savings are created by
acquiring a more cost-effective plant and an adjustment
to new resource additions.
8 - 30 2007 Electric IRP Avista Corp
Chapter 9- Action Items
ACTION ITEMS
The Integrated Resource Plan (IRP) is an ongoing
and iterative process attempting to balance the need
for regular publications with pursuing the best 20-year
forecast possible. The set biennial publication date
means that there is always room for improvements or
additional research. This section provides an overview of
the progress that has been made regarding the 2005 IRP
Action Plan. The 2007 IRP Action Plan provides details
about the issues and improvements that were developed
or raised during this planning cycle and those that need
to be deferred to the 2009 IRP.
SUMMARY OF THE 2005 ACTION PLAN
The 2005 IRP includes Action Items in four separate
areas: renewable energy and emissions, modeling
enhancements, transmission modeling and research, and
conservation.
RENEWABLE ENERGY AND EMISSIONS
. Commission a study to assess wind potential
within Avista s service territory.
. Continue to monitor emissions legislation and its
potential effects on markets and the company.
. Research clean coal technology and carbon
sequestration.
. Assess biomass potential within and outside of
Avista s service territory.
Avista hired a meteorological consultant who completed
map and aerial studies of wind potential within the
company s service territory. Several promising sites were
located that warrant further consideration and assessment.
The next steps involve contacting landowners to assess
their interest in allowing the installation of anemometers
to test wind speeds and shapes for at least a one-year
period. This research will be ongoing and will be
reported in the 2009 IRP.
Avista has actively monitored state and federal emissions
legislation which has resulted in the company taking
several steps forward in this area. Most notably, an entire
section of this IRP has been dedicated to emissions
issues, greenhouse gas emissions cost estimates have been
included in the Base Case, and an Avista Climate Change
Council has been convened to bring all of the functional
areas of the company together address climate change
Issues.
Wind Turbines Generating Electricity
Avista Corp 2007 Electric IRP 9 -
Chapter 9- Action Items
A variety of different coal technologies have been
researched for this IRP through the joint request
for information with Idaho Power. The research for
this process has resulted in more up-to-date capital
costs for sub-critical, supercritical and ultra-critical
pulverized coal, circulating fluidized bed and integrated
gas combined cycle technologies. These have been
included in the Technical Advisory Committee (TAC)
presentations available at the company s IRP Website.
Presentations on clean coal technologies and carbon
capture and sequestration are also included in the TAC
presentation. The steep increases in capital costs, recent
Washington state legislation and changes in Avista
management directives have moved non-sequestered coal
completely out of the plan. However, we will continue
to research coal technologies to help us better understand
resources throughout the Western Interconnect and in
case new, clean coal technologies become cost effective
in the future.
Some initial assessments of biomass potential within and
outside of Avista s service territory have been researched.
Recent studies have indicated total amounts of biomass
availability by county in Washington, but further work
needs to be done to determine the amount of biomass
that is economically recoverable and feasible to obtain.
One benefit of the recent RPS legislation should be
more research into renewable technologies, including
biomass. This action item will need to be carried
forward to the 2009 IRE
MODELING ENHANCEMENTS
. Evaluate the 70-year water record for inclusion in
2007 IRP studies.
. Add more functionality to the Avista Linear
Programming Model (e., direct consideration
of cash flow and rate impacts versus after-the-fact
reviews).
The 70-year water record has been reviewed and
implemented in the modeling for this IRE The
Avista Linear Programming Model or PRiSM has
been enhanced to handle 300 iterations, cash flow
power supply rate impacts, and improved the overall
functionality and reporting abilities.
TRANSMISSION MODELING AND RESEARCH
. Work to maintain/retain existing transmission
rights on the company s transmission system, under
applicable FERC policies, for transmission service
to bundled retail native load.
. Continue involvement in BPA transmission
practice processes and rate proceedings to
minimize costs of integrating existing resources
outside of the Company s service area.
. Continue participation in regional and sub-
regional efforts to establish new regional
transmission structures (Grid West and TIG) to
facilitate long-term expansion of the regional
transmission system.
. Evaluate costs to integrate new resources across
Avista s service territory and from regions outside
of the Northwest.
Chapter 4 contains details about Avista transmission
modeling and research. These Action Items will continue
to be important in the 2009 IRE
CONSERVATION
. Review the potential for cost-effective load
shifting programs using hourly market prices.
. Complete the conservation control project
currently underway as part of the Northwest
Energy Efficiency Initiative for future evaluation as
a potential conservation resource.
9 - 2 2007 Electric IRP Avlsta Corp
Chapter 9- Action Items
Several new programs and measures are being developed
in addition to enhancements to the company s existing
programs. Load management pilot programs are being
developed for implementation beginning in 2007
in Moscow and Sandpoint, Idaho. Large customer
interruption and distributed generation projects are
also being researched. Nine potential transmission and
distribution efficiency measures were identified and
studied. Three of these projects are currently at the
work-in-progress phase of development.
2007 IRP ACTION PLAN
The companys 2007 Preferred Resource Strategy
provides direction and guidance for resource acquisitions.
The 2007 IRP action plan lists the activities that will be
carried out for inclusion in the 2009 IRP. Progress will
be monitored and reported in Avista s 2009 Integrated
Resource Plan. Each item in the action plan was
developed using input from Commission Staff, the
company s management team and the Technical Advisory
Committee.
RENEWABLE ENERGY
. Continue studying wind potential in the
company s service territory, possibly including the
placement of anemometers at the most promising
wind sites.
. Commission a study of Montana wind resources
that are strategically located near existing company
transmission assets.
. Learn more about non-wind renewable
resources to satisfy renewable portfolio standard
requirements and decrease the company s carbon
footprint.
DEMAND SIDE MANAGEMENT
. Update processes and protocols for integrating
energy efficiency programs into the IRP to
improve and streamline the process.
. Study and quantify transmission and distribution
system efficiency concepts.
Determine the potential impacts and costs ofload
management options currently being reviewed as
part of the Heritage Project.
. Develop and quantify the long-term impacts of
the newly signed contractual relationship with
the Northwest Sustainable Energy for Economic
Development organization.
EMISSIONS
. Continue to evaluate the implications of new rules
and regulations affecting power plant operations
most notably greenhouse gases.
. Continue to evaluate the merits of various carbon
quantification methods and emissions markets.
MODELING AND FORECASTING ENHANCEMENTS
. Study the potential for fIXing natural gas prices
through financial instruments, coal gasification
investments in gas fields or other means.
. Continue studying the efficient frontier modeling
approach to identify more and better uses for its
information.
. Further enhance and refine the PRiSM LP model.
. Continue to study the impact of climate on the
load forecast.
. Monitor the following conditions relevant to the
load forecast: large commercial load additions
Shoshone county mining developments and the
market penetration of electric cars.
TRANSMISSION PLANNING
. Work to maintain/retain existing transmission
rights on the company s transmission system, under
applicable FERC policies, for transmission service
to bundled retail native load.
. Continue involvement in BPA transmission
practice processes and rate proceedings to
Avista Corp 2007 Eiectric IRP 9 - 3
Chapter 9- Action Items
minimize costs of integrating existing resources
outside of the Company s service area.
. Continue participation in regional and
sub-regional efforts to establish new regional
transmission structures (ColumbiaGrid and other
PRODUCTION CREDITS
forums) to facilitate long-term expansion of the
regional transmission system.
. Evaluate costs to integrate new resources across
Avista s service territory and from regions outside
of the Northwest.
Clint Kalich, Manager of
Resource Plannin & Anal sis
John Lyons, Power Supply
Anal st
James Gall, Power Supply
Anal st
Heidi Heath, Power Supply
Anal st
Randy Barcus, Chief Corporate
Economist
Jon Powell , Partnership
Solutions Mana er
Project Manager/Author cl int. kalich~avistacorp. com
Research/Author/Ed itor john .Iyons~avistacorp. com
Modeling and Analysis
/Author
Author/Editor
Load Forecast
Conservation
james.gall~avistacorp.com
heid i. heath~avistacorp. com
randy. barcus~avistacorp .com
jon.powell~avistacorp.com
Other Contributors
Bruce Folsom , Manager of Thomas Dempsey, Manager of
Demand Side Mana ement Thermal En ineerin
Kevin Christie, Director of Gas Scott Waples , Chief SystemSu Planner
Kelly Irvine, Natural Gas Analyst Randy Gnaedinger, Transmission
Plannin En ineer
Sara Koeff, Transmission
Plannin En ineer
Jeff Schlect, Manager
Transmission Services
James McDougall , Regulatory
Anal st
Steve Silkworth, Manager of
Wholesale Power
Jessie Wuerst, Communications
Mana er
Bob Lafferty, Manager of
Wholesale Marketin & Contracts
Todd Bryan, Power Supply
Anal st
Doug Pottratz, Manager
Cor orate Environmental Affairs
Linda Gervais, Regulatory Analyst
Dave Moeller, Market Service
En ineer
Avista Corp2007 Electric IRP