200509012005 IRP.pdf

i
Table of Contents
Executive Summary ...................................................................................................................................................I
Resource Needs ....................................................................................................................................................II
Modeling and Results ............................................................................................................................................II
Electricity and Natural Gas Market Forecasts..................................................................................................... V
Conservation Acquisition ................................................................................................................................... VII
Preferred Resource Strategy .............................................................................................................................. VII
Carbon Emissions .............................................................................................................................................. VII
PRS Acquisition ................................................................................................................................................... IX
Action Items ......................................................................................................................................................... IX
Introduction & Stakeholder Involvement ..............................................................................................................X
IRP Process .......................................................................................................................................................... X
Stakeholder Involvement ...................................................................................................................................... X
Outline of 2005 IRP Report ...............................................................................................................................XIV
1. ..........Electricity Sales Forecast ......................................................................................................................1-1
1.1 ...Economic Conditions in the Service Area ...............................................................................................1-1
1.2 ...Electric Operating Division Economy ......................................................................................................1-1
1.3 ...The Economic Forecasts ..........................................................................................................................1-2
1.4 ...Electricity Customer Forecasts ................................................................................................................1-4
1.5 ...Retail Electricity Sales Forecast ...............................................................................................................1-5
1.6 ...Price Elasticity ...........................................................................................................................................1-6
1.7 ...Alternative Scenarios ................................................................................................................................1-8
1.8 ...Enhancements to the Forecasting Models and Process ........................................................................1-9
2. ..........Resource Requirements ........................................................................................................................2-1
2.1 ...Utility–Owned Resources .........................................................................................................................2-1
2.2 ...Capacity Tabulation ..................................................................................................................................2-8
2.3 ...Energy Tabulation ......................................................................................................................................2-9
2.4 ...Reserve Margins .....................................................................................................................................2-11
3. ..........Conservation Initiatives .........................................................................................................................3-1
3.1 ...IRP Objective.............................................................................................................................................3-4
3.2 ...IRP Methodology and Analysis ................................................................................................................3-4
3.3 ...Conservation Measure Defi nitions ...........................................................................................................3-6
3.4 ...Evaluation of Measures ..........................................................................................................................3-11
3.5 ...Results of the Analysis ............................................................................................................................3-13
3.6 ...Review of the Results .............................................................................................................................3-14
3.7 ...Conservation Business Planning ............................................................................................................3-24
4. ..........Transmission Planning ...........................................................................................................................4-1
4.1 ...Avista Transmission System .....................................................................................................................4-1
4.2 ...Regional Transmission System ................................................................................................................4-4
4.3 ...Regional Transmission Issues ..................................................................................................................4-4
4.4 ...Modeling Transmission Costs in the Integrated Resource Plan .............................................................4-7
ii
5. ..........Modeling Approach ................................................................................................................................5-1
5.1 ...Western Interconnect Simulation: AURORAXMP.......................................................................................5-3
5.2 ...Key Assumptions and Inputs ...................................................................................................................5-5
5.3 ...Risk Modeling .........................................................................................................................................5-11
5.4 ...The Avista LP Model ...............................................................................................................................5-21
5.5 ...New Resource Alternatives ....................................................................................................................5-23
5.6 ...Wind Modeling ........................................................................................................................................5-30
5.7 ...Summary .................................................................................................................................................5-33
6. ..........Modeling Results ....................................................................................................................................6-1
6.1 ...Base Case .................................................................................................................................................6-2
6.2 ...Scenarios .................................................................................................................................................6-11
6.3 ...Carbon Emission Scenarios ...................................................................................................................6-22
6.4 ...Avista-Centric Scenarios ........................................................................................................................6-29
6.5 ...Avoided Costs .........................................................................................................................................6-30
6.6 ...Summary .................................................................................................................................................6-32
7. ..........Preferred Resource Strategy ................................................................................................................7-1
7.1 ...The Preferred Resource Strategy–An Introduction .................................................................................7-1
7.2 ...Preferred Resource Strategy Details ........................................................................................................7-5
7.3 ...Effi cient Frontier .........................................................................................................................................7-7
7.4 ...Twelve Alternative Portfolio Strategies .....................................................................................................7-9
7.5 ...Performance of PRS Compared to 12 Resource Strategies ................................................................7-17
7.6 ...Performance of PRS and 12 Resource Strategies In Market Structure Scenarios .............................7-24
7.7 ...Acquisition of PRS Resources ...............................................................................................................7-29
7.8 ...Adjusted Energy and Capacity Positions ..............................................................................................7-31
8. ..........Action Items .............................................................................................................................................8-1
8.1 ...Summary Report for 2003 Action Plan ....................................................................................................8-1
8.2 ...Action Plan for 2005 .................................................................................................................................8-4
iii
Table of Figures
Figure 1: Load Resource Balance–Energy ..........................................................................................................III
Figure 2: Resource Cost Versus Resource Risk ................................................................................................ IV
Figure 3: Avista Effi cient Frontier ......................................................................................................................... V
Figure 4: Nominal Electricity and Gas Prices ..................................................................................................... VI
Figuer 5: Cumulative Conservation Acquisitions ............................................................................................... VI
Figure 6: Preferred Resource Strategy—Capacity ...........................................................................................VIII
Figure 1.1: Service Territory Map ......................................................................................................................1-2
Figure 1.2: Idaho and Washington Job Change by County ............................................................................1-3
Figure 1.3: Idaho and Washington Population Change by County .................................................................1-3
Figure 1.4: Total Service Territory Population ...................................................................................................1-4
Figure 1.5: Electric Utility Customer Forecast ..................................................................................................1-5
Figure 1.6: 2005 Electric Utility Retail Sales Forecast .....................................................................................1-6
Figure 1.7: Residential Retail Rate Projection for Retail Load Forecast..........................................................1-7
Figure 1.8: Wholesale Natural Gas Price Forecast for Retail Load Forecast .................................................1-8
Figure 1.9: Retail Sales Forecast Scenarios .....................................................................................................1-9
Figure 2.1: Avista’s Hydroelectric Projects .......................................................................................................2-2
Figure 2.2: Energy Load and Resource Tabulation ........................................................................................2-10
Figure 3.1: Historical Electric Conservation Acquisition ..................................................................................3-2
Figure 3.2: Electric Conservation Acquisition Versus Goals ...........................................................................3-3
Figure 3.3: Conservation Supply Curve Stacked by Levelized TRC Cost ...................................................3-19
Figure 3.4: Conservation Supply Curve Stacked by Levelized TRC Cost <0.10 .........................................3-19
Figure 3.5: Conservation Supply Curve .........................................................................................................3-20
Figure 3.6: Conservation Supply Curve .........................................................................................................3-20
Figure 3.7: Aggregate Conservation Goal Comparison ................................................................................3-21
Figure 3.8: Customer Segment Savings Distribution ....................................................................................3-21
Figure 3.9: Industrial Segment Savings Distribution ......................................................................................3-22
Figure 3.10: Commercial Segment Savings Distribution ...............................................................................3-22
Figure 3.11: Residential Segment Savings Distribution.................................................................................3-23
Figure 5.1: Modeling Process Diagram ............................................................................................................5-2
Figure 5.2: NERC Interconnections Map..........................................................................................................5-3
Figure 5.3: NW & Avista Monthly Hydro Capacity Factors Modeled in AURORAXMP .....................................5-6
Figure 5.4: One Week Hydro Dispatch Example..............................................................................................5-6
Figure 5.5: Henry Hub Natural Gas Price Forecast .........................................................................................5-8
Figure 5.6: New Resources Under Construction .............................................................................................5-9
Figure 5.7: Western Interconnect Hydroelectric Generation Distribution .....................................................5-13
Figure 5.8: Mid-Columbia Electricity and Malin Natural Gas Price Correlations .........................................5-13
Figure 5.9: Natural Gas Price Statistics–2007 ................................................................................................5-15
Figure 5.10: Natural Gas Price Statistics–2016 .............................................................................................5-15
Figure 5.11: Actual Wind Data–1000 Continuous Hours ...............................................................................5-19
Figure 5.12: Stochastic Wind Model Absent Serial Correlation ....................................................................5-20
Figure 5.13: Stochastic Wind Model With Serial Correlation ........................................................................5-20
iv
Figure 5.14: Effi cient Frontier Versus Capital Expenditure ............................................................................5-23
Figure 5.15: 2016 Resource Option Costs .....................................................................................................5-29
Figure 6.1: Cumulative Western Interconnect Resource Additions ................................................................6-3
Figure 6.2: Cumulative Western Interconnect Resource Additions Absent California ..................................6-4
Figure 6.3: Mid-Columbia Electric Price Forecast ...........................................................................................6-4
Figure 6.4: Malin Natural Gas and Mid–Columbia Electricity Correlation Plot ...............................................6-5
Figure 6.5: Western Interconnect Resource Contribution ...............................................................................6-5
Figure 6.6: Mid-Columbia Electric Price Forecast Comparison ....................................................................6-7
Figure 6.7: Stochastic Base Case Mid-Columbia Electric Price Forecast .....................................................6-8
Figure 6.8: Base Case and Volatile Gas Mid-Columbia Price Comparison ..................................................6-8
Figure 6.9: Resource Cost and Resource Risk Comparison ........................................................................6-11
Figure 6.10: Cumulative Western Interconnect Resource Additions–High Gas ..........................................6-12
Figure 6.11: Base Case and High Gas Mid-Columbia Electric Price Forecasts .........................................6-13
Figure 6.12: Cumulative Resource Selection for the Western Interconnect—Low Gas ..............................6-14
Figure 6.13: Base Case and Low Gas Mid-Columbia Electric Price Forecasts ..........................................6-14
Figure 6.14: Base Case and High Coal Price Escalation Coal Price Forecasts—Montana Mine Mouth ...6-15
Figure 6.15: Cumulative Resource Selection for the Western Interconnect–High Coal Price Escalation ..6-15
Figure 6.16: Base Case and High Coal Price Escalation Mid-Columbia Electric Price Forecasts .................. 6-16
Figure 6.17: Cumulative Resource Selection for the Western Interconnect–No Capacity Credit ...............6-16
Figure 6.18: Base Case and No Capacity Credit Mid-Columbia Electric Price Forecasts .........................6-17
Figure 6.19: Cumulative Resource Selection for the Western Interconnect—
30% Lower Transmission Capital Cost .....................................................................................6-18
Figure 6.20: Base Case and 30% Lower Transmission Capital Cost Mid-Columbia Electric
Price Forecasts ..........................................................................................................................6-18
Figure 6.21: Cumulative Resource Selection for the Western Interconnect–Hydro Shift ............................6-20
Figure 6.22: Mid-Columbia On & Off Peak Price Comparison For The Hydro Shift ...................................6-20
Figure 6.23: Monthly Market Price Volatility From Increased Wind Penetration ..........................................6-21
Figure 6.24: Base Case and Boom and Bust Mid-Columbia Electric Price Forecasts ..............................6-22
Figure 6.25: Cumulative Resource Selection for the Western Interconnect–SB 342 Carbon Tax ..............6-24
Figure 6.26: Coal Dispatch Between Base Case and SB 342 Scenario ......................................................6-25
Figure 6.27: CO2 Emissions and Cost Forecast for the Base Case and SB 342 .........................................6-25
Figure 6.28: Base Case and SB 342 Mid-Columbia Electric Price Forecasts ............................................6-26
Figure 6.29: Cumulative Resource Selection for the Western Interconnect–NCEP Carbon Tax ................6-27
Figure 6.30: Western Interconnect Generator CO2 Emissions Forecast ......................................................6-28
Figure 6.31: Base Case and NCEP Carbon Tax Mid-Columbia Electric Price Forecasts ..........................6-28
Figure 6.32: Base Case Mid-Columbia Price Forecast and Avoided Costs Comparison ..........................6-31
Figure 7.1: 2005 Preferred Resource Strategy Build ......................................................................................7-3
Figure 7.2: 2003 IRP Preferred Resource Strategy Build ...............................................................................7-3
Figure 7.3: Preferred Resource Strategy Coal Build vs. LP Module Build .....................................................7-4
Figure 7.4: Effi cient Frontier ..............................................................................................................................7-8
Figure 7.5: 50/50 Gas and Coal Build ...........................................................................................................7-10
Figure 7.6: Wind and Gas Build .....................................................................................................................7-11
Figure 7.7: No CO2 Emissions Build ..............................................................................................................7-12
Figure 7.8: All Renewables Build ...................................................................................................................7-12
v
Figure 7.9: Least Cost Build ...........................................................................................................................7-14
Figure 7.10: 25% Risk Build ...........................................................................................................................7-14
Figure 7.11: 50% Risk Build ...........................................................................................................................7-15
Figure 7.12: 75% Risk Build ...........................................................................................................................7-15
Figure 7.13: Least Risk Build .........................................................................................................................7-16
Figure 7.14: Preferred Resource Strategy Build ............................................................................................7-16
Figure 7.15: Renewable Resource Contribution in 2016 ..............................................................................7-17
Figure 7.16: Conservation Acquisition ...........................................................................................................7-18
Figure 7.17: 2007-16 Portfolio Capital Cost ..................................................................................................7-18
Figure 7.18: 2007-26 Portfolio Risk Comparison–Average Covariance .......................................................7-20
Figure 7.19: 2016 Portfolio Risk Comparison-Standard Deviation of Incremental
Power Supply Expense ..............................................................................................................7-20
Figure 7.20: 2007-16 Portfolio Risk Comparision-95th Percentile Difference From Mean Value ...............7-21
Figure 7.21: 2007-16 Average Incremental Power Supply Expense-Induced Rate Increases ...................7-22
Figure 7.22: 2016 Incremental Power Supply Expense ...............................................................................7-22
Figure 7.23: 2007-16 Incremental Power Supply Expense NPV .................................................................7-23
Figure 7.24: 2007-16 Maximum Single-Year Rate Increase ..........................................................................7-23
Figure 7.25: 2007-16 Power Supply Expense Average Covariance—Volatile Gas .....................................7-24
Figure 7.26: 2016 Incremental Power Supply Expense—Low Gas .............................................................7-25
Figure 7.27: 2016 Incremental Power Supply Expense ...............................................................................7-26
Figure 7.28: Maximum Annual Rate Change from Base Case .....................................................................7-26
Figure 7.29: 2007-16 Incremental Power Supply Expense NPV ..................................................................7-28
Figure 7.30: 2007-16 Incremental Power Supply Expense NPV ..................................................................7-28
Figure 7.31: 2007-16 Power Supply Expense NPV Change ........................................................................7-29
Figure 7.32: 2007-16 Incremental Power Supply Expense NPV ..................................................................7-30
Figure 7.33: Preferred Resource Strategy–Energy .......................................................................................7-33
Figure 7.34: Company Resource Mixes (% of Energy) 2007, 2016, and 2026 ............................................7-33
Figure 7.35: Preferred Resource Strategy–Capacity .....................................................................................7-35
Figure 7.36: Company Resource Mixes (% of Capacity) 2007, 2016, and 2026 .........................................7-35
vi
Table of Tables
Table 1: Net Position Forecast ..............................................................................................................................II
Table 2: 2005 to 2003 IRP Comparison ............................................................................................................VIII
Table 3: TAC Participants .................................................................................................................................... XI
Table 4: TAC Meeting Dates and Agendas ....................................................................................................... XII
Table 2.1: Company-Owned Hydro Resources ...............................................................................................2-4
Table 2.2: Company-Owned Thermal Resources............................................................................................2-5
Table 2.3: Signifi cant Contractual Rights & Obligations ..................................................................................2-6
Table 2.4: Mid-Columbia Contract Quantities Summary ................................................................................2-7
Table 2.5: Loads & Resources Capacity Forecast ...........................................................................................2-8
Table 2.6: Loads & Resources Energy Forecast ..............................................................................................2-9
Table 2.7: Reserve Sharing Example ..............................................................................................................2-11
Table 2.8: Capacity L&R Versus Sustained Capacity ....................................................................................2-13
Table 3.1: Summary of Individual Industrial Measures ..................................................................................3-14
Table 3.2: Summary of Individual Commercial Measures .............................................................................3-15
Table 3.3: Summary of Individual Residential Measures ...............................................................................3-16
Table 3.4: TRC Costs and Benefi ts for Industrial Measures .........................................................................3-17
Table 3.5: TRC Costs and Benefi ts for Commercial Measures ....................................................................3-17
Table 3.6: TRC Costs and Benefi ts for Residential Measures ......................................................................3-18
Table 4.1: Avista Generation Integration Cost Estimates ................................................................................4-8
Table 5.1: AURORAXMP Zones ...........................................................................................................................5-4
Table 5.2: Trading Hub and Zone Natural Gas Price Forecast ........................................................................5-7
Table 5.3: Renewable Portfolio Standards by State ........................................................................................5-9
Table 5.4: IRP Differences from Fifth Power Plan ..........................................................................................5-10
Table 5.5: Hydroelectric Generation Statistics by Zone ................................................................................5-12
Table 5.6a: Western Interconnect Load Correlations–Jan through Jun........................................................5-16
Table 5.6b: Western Interconnect Load Correlations–Jul through Dec ........................................................5-16
Table 5.7a: Western Interconnect Load Statistics–Jan through Jun ............................................................5-17
Table 5.7b: Western Interconnect Load Statistics–Jul through Dec ............................................................5-18
Table 5.8: New Resource Alternatives ............................................................................................................5-26
Table 5.9: Forecast Capital Cost Reductions .................................................................................................5-27
Table 5.10: Regional Transmission Cost Estimates .......................................................................................5-28
Table 5.11: Wind Forecasting Accuracy .........................................................................................................5-31
Table 6.1: Electric Market Prices By Western Interconnect Zone ...................................................................6-6
Table 6.2: 2007 Resource Option Costs ........................................................................................................6-10
Table 6.3: Market Value of Noxon Rapids Project .........................................................................................6-29
Table 6.4: Avista Avoided Costs Compared to Mid-Columbia Price Forecast ............................................6-32
Table 7.1: Northwest IOU Loads and Estimated Wind Acquisition Plans through 2016 ...............................7-6
Table 7.2: 2016 Resource Strategies ..............................................................................................................7-13
Table 7.3: 2026 Resource Strategies ..............................................................................................................7-13
Table 7.4: Loads & Resources Energy Forecast with PRS ............................................................................7-32
Table 7.5: Loads & Resources Capacity Forecast with PRS ........................................................................7-34
vii
Appendix Table of Contents
Volume 1
A. Technical Advisory Committee Meeting Agendas
B. Technical Advisory Committee Members
C. Technical Advisory Committee Meeting Presentation Slides
Volume 2
D. Preferred Resource Strategy Detail
E. Scenario and Futures Mid-Columbia Electric Price Forecasts
F. Scenario and Futures Capacity Expansion Results
G. Scenario and Futures Portfolio Results Comparisons
H. Additional Levelized Resource Cost Detail
I. Conservation Details
J. Loads and Resources Detail (Includes High & Low Load)
K. Summary of Draft IRP Comments
viii
List of Acronyms
aMW Average Megawatts
BPA Bonneville Power Administration
CCCT Combined-Cycle Combustion Turbine
CFL Compact Fluorescent Lamp
CO2 Carbon Dioxide
CSA Climate Stewardship Act (also known as
the McCain-Lieberman Bill)
CVR Controlled Voltage Reduction
EF Effi ciency
EIA Energy Information Administration
FERC Federal Energy Regulatory Commission
GHG Greenhouse Gas
GWh Gigawatt-hour
HRSG Heat Recovery Steam Generator
HVAC Heating, Ventilation and Air Conditioning
IGCC Integrated Gasifi cation Combined Cycle
IRP Integrated Resource Plan
IS Information Systems
kV Kilovolt
kW Kilowatt
kWh Kilowatt-hour
LIRAP Low Income Rate Assistance Program
LP Linear Programming
MW Megawatt
MWh Megawatt-hour
NCEP National Commission for Energy Policy
NEB Non-Energy Benefi ts
NPCC Northwest Power and Conservation
Council (formerly Northwest Power
Planning Commission)
NPV Net Present Value
NWPP Northwest Power Pool
O&M Operations and Maintenance
OASIS Open Access Same-Time
Information System
OSU Oregon State University
PC Personal Computer
PGE Portland General Electric
PRS Preferred Resource Strategy
psig Pounds Per Square Inch Gauge
PTC Production Tax Credit
PUD Public Utility District
PURPA Public Utility Regulatory Policies Act
of 1978
RPS Renewable Portfolio Standards
RTO Regional Transmission Organization
SCCT Simple-Cycle Combustion Turbine
TAC Technical Advisory Committee
TIG Transmission Improvements Group
TRC Total Resource Cost
Triple E External Energy Effi ciency Board
VFD Variable Frequency Drive
WECC Western Electricity Coordinating Council
WNP-3 Washington Public Power Supply System
(WPPSS, now Energy Northwest)—
Washington Nuclear Plant No. 3
I
Section Highlights
 Avista has added 35 MW of wind generation, 140 MW of gas-fi red generation and 8 MW of
conservation to its portfolio since the 2003 IRP.
 Energy and capacity defi cits begin in 2010 and 2009, respectively, growing to 640 aMW and
901 MW by the end of the study in 2026.
 Electricity sales are forecast to grow 2.1 percent annually through 2026.
 Avista uses AURORAXMP to model the entire Western Interconnect; market conditions outside
the Northwest affect Mid-Columbia market prices.
 Conservation acquisition is 50 percent higher than in the 2003 IRP.
 Acquiring additional transmission is critical to Company plans.
 The PRS strikes a reasonable balance between keeping average costs and variation in
year-to-year costs low.
 The 2016 PRS includes 400 MW of wind, 250 MW of coal, 80 MW of biomass, 52 MW of plant
upgrades and 69 MW of conservation.
 Over half of future energy needs are met with renewables, plant upgrades and conservation.
The Company’s 2005 Integrated Resource Plan (IRP)
identifi es a strategic resource portfolio that meets
future load requirements, promotes environmental
stewardship and satisfi es regulatory obligations.
A series of robust analyses are used to evaluate
resource options based on expected value and
levels of market volatility over the next 20 years.
These analyses assist in comparing resource
portfolio options, guiding the Company in the
selection of a Preferred Resource Strategy (PRS).
The PRS provides a balance between the objectives
of low cost, reliable service and reasonable future
rate volatility.
Avista’s management and stakeholders in the
Technical Advisory Committee (TAC) play a key
role and have a signifi cant impact in guiding the
plan to its fi nal conclusions. TAC members include
customers, commission staff, consumer advocates,
academics, utility peers, government agencies and
other interested parties. The TAC provides important
input on modeling, planning assumptions and the
general direction of the planning process.
The Company has made signifi cant progress
in resource acquisitions since the last IRP. The
Company demonstrated the need to acquire 75
megawatts (MW) of wind and 140 MW of combined-
cycle combustion turbine generation in the 2003
II
Year
Energy
Position
(aMW)
Capacity
Position
(MW) Year
Energy
Position
(aMW)
Capacity
Position
(MW)
2007 82 118 2011 -157 -256
2008 50 71 2016 -360 -508
2009 12 -5 2021 -491 -673
2010 -40 -75 2026 -640 -901
Table 1: Net Position Forecast
IRP. Avista contracted with PPM Energy for 35 MW
of wind capacity from the Stateline project in 2004.
Upgrades were completed at Cabinet Gorge
Unit 2 in 2004, bringing seven MW of new capacity
and three average megawatts (aMW) of energy.
The Company also reacquired the second half of the
natural gas-fi red Coyote Springs 2 plant from Mirant
Corporation in January 2005.
Incremental upgrades to existing resources are
forecast in this plan to provide additional energy
and capacity at costs lower than acquiring new
generation assets. The Company’s upgrade plans
for the Clark Fork River project forecasts 45 MW
of capacity gains by 2012. Planned upgrades to
Colstrip Units 3 and 4 in 2006 and 2007 will boost
Avista’s output share by 8 MW.
Resource Needs
Recent resource purchases, plant upgrades and
conservation acquisition are inadequate to meet all
future load growth. Annual energy defi cits begin in
2010, with loads exceeding resource capability by
40 aMW. Energy defi cits rise to 360 aMW in 2016
and 640 aMW in 2026. The Company will be short 5
MW of capacity in 2009. In 2016 and 2026 capacity
defi cits rise to 508 MW and 901 MW, respectively.
Table 1 presents Company positions between 2007
and 2026.
Increasing defi cits are a result of forecasted 2.1
percent annual average load growth and expirations
of some long-term contracts. Figure 1 provides
a graphical synopsis of the Company’s load and
resource balances over the next 20 years.
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
West. A Western Interconnect-wide study was
performed to understand the impact of regional
markets on Avista. We believe that the additional
efforts to develop this study were necessary given
the signifi cant impact other western regions can
have on the Northwest electricity marketplace.
Existing resources were combined with the present
transmission grid to simulate hourly operations
for the Western Interconnect from 2007 to 2026.
III
Figure 1: Load Resource Balance–Energy (aMW)
Cost-effective new resources and transmission
were added as necessary to meet growing loads.
Monte Carlo-style analysis varied hydro, wind, load
and gas price data over 200 iterations of potential
future conditions. The simulation results were used
to estimate the Mid-Columbia electric market. The
iterations collectively formed the Base Case for this IRP.
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-Columbia market to identify
the future value of each asset to the Company,
as well as its inherent risk (e.g., year-to-year
volatility). Future market values and risk were
compared with the capital and fi xed operation and
maintenance (O&M) costs that would be incurred.
The Company’s Linear Programming model then
assisted in selecting the PRS for serving future load.
The selection of the PRS was based on forecasted
energy and capacity needs, resource values and
limiting power supply expense variability.
Futures and scenarios were used to identify
performance of the PRS under conditions beyond
the Base Case. Futures are stochastic studies using
a Monte Carlo approach to quantitatively assess
risk around an expected mean outcome.1 This
time-intensive and multi-variable approach is the
most robust method used for risk assessment. Two
futures were modeled for the 2005 IRP: the Base
Case, and a High Gas Volatility case with increased
natural gas price variability.
1 Stochastic studies use a statistical approach using probability
distributions (i.e., means and standard deviations) to forecast variables into
the future.
IV
A scenario is a deterministic study that changes
one signifi cant 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.
Eighteen scenarios were modeled for the 2005 IRP,
including high and low natural gas prices, carbon
emission taxes and the loss of major hydroelectric
generation projects.
This IRP values potential resource options by
considering their costs, defi ned as expected
incremental power supply expenses.2 Financial
risk—variability measured as the standard deviation
2 Incremental power supply expense is defi ned as variable O&M expenses
and fuel for existing Company resources and fi xed and variable O&M and
capital recovery costs for new resources.
of the incremental power supply expense—is also
considered. Figure 2 plots the costs of various
resource options against their inherent risks.
Resources using natural gas and wind are riskier
than those using fuels with more stable prices
and availability, such as coal, nuclear, biomass
and geothermal. The information in Figure 2 does
not attempt to quantify potential risks beyond
operational risk. For example, the potential for
construction cost overruns and nuclear waste
disposal risks are not considered. A geographically
diversifi ed wind portfolio, with ownership across
the Northwest and into eastern Montana, appears
to reduce some of the fi nancial risk created by
intermittent wind availability.
Figure 2: Resource Cost Versus Resource Risk
RISK
TO
T
A
L

C
O
S
T
Pulverized
Coal
Nuclear
IGCC CoalLandfill
Gas
Geo-
thermal
Mt. IGCC
Coal
w/ seq.
Manure
Wood
CCCT
SCCT- Aero
SCCT-
Frame
Co-Gen
Alberta's
Oil Sands
Kennewick
Wind Tier 2
Browning
Depot Wind
Kennewick
Wind Tier 1Diversified
Wind Tier 1
Local Wind
Low Risk - Low Cost
High Risk - High CostLow Risk - High Cost
High Risk - Low Cost
Local
Co-Gen
Diversified
Wind Tier 2
V
The IRP further enhances portfolio analysis by
identifying an “Effi cient Frontier.” The Effi cient
Frontier is a fi nancial theory that develops a curve of
optimal portfolio returns based on the level of risk an
investor is willing to accept. Figure 3 illustrates the
Effi cient Frontier developed for the 2005 IRP. This
fi gure shows the PRS, along with other portfolios
formed for the 2005 IRP, and its position relative to
the Effi cient Frontier.
Resource portfolios in the Effi cient Frontier are
subject to coal and wind limitations; hence some
unrestricted portfolios, like All-Coal, theoretically
can outperform the Effi cient Frontier. The exercise
was limited to 400 MW of wind and 250 MW
of coal in 2016, and 650 MW of wind and 550
MW of coal in 2026. The wind limitation refl ects
Company agreement with the Northwest Power
and Conservation Council (NPCC) that a limited
amount of economically viable wind potential exists
in the Northwest. The NPCC estimates Northwest
wind potential to be 5,000 MW. Avista serves
approximately fi ve percent of Northwest loads; the
prorated Company share is 250 MW. Therefore, the
650 MW target by 2026 is substantially higher than
the Company’s share of Northwest wind potential.
The coal limitation is based on the Company’s
desire to acquire a cost effective and diverse fuel
mix, and the risks of future carbon tax legislation.
Electricity and Natural Gas
Market Forecasts
Our analyses explain that natural gas and Mid-
Columbia electricity market prices are becoming
increasingly correlated because of the increase in
gas-fi red plant construction across the Western
Interconnect. Figure 4 represents the Company’s
electric and natural gas price forecasts. 2003 IRP
forecasts are provided for reference.
Figure 3: Avista Effi cient Frontier ($millions)
VI
Figure 4: Nominal Electricity and Gas Prices
Figure 5: Cumulative Conservation Acquisitions
0
1
2
3
4
5
6
7
8
9
10
20
0
7
20
0
9
20
1
1
20
1
3
20
1
5
20
1
7
20
1
9
20
2
1
20
2
3
20
2
5
$/
d
t
h
30
40
50
60
70
80
90
100
110
120
130
$/
M
W
h
2005 Gas Price Forecast 2003 Gas Price Forecast
2005 Electricity Price Forecast 2003 Electricity Price Forecast
VII
Conservation Acquisition
Figure 5 shows how conservation has lowered
Company requirements by approximately 83 aMW
since programs began in the 1970’s.3 With
additional funding recommended by the IRP, the
Company expects conservation to lower load
growth in its service territory by 6.9 MW per year,
totaling 138 MW over 20 years. The 2005 IRP
conservation acquisition schedule is approximately
50 percent higher than what was included in
the 2003 IRP.
Preferred Resource Strategy
The Company’s Preferred Resource Strategy is
defi ned by fi ve resource categories: conservation,
upgrades to existing generation facilities, wind, other
small renewables and coal. In total, conservation,
plant upgrades and renewables provide more than
half of new load requirements over the IRP time
frame. The 2003 IRP included more coal-fi red
generation to meet requirements. Both the 2005
and 2003 IRPs provide similar insulation from
price volatility. In 2016 newly installed capacity
includes 400 MW of wind, 250 MW of coal and 80
MW of other small renewable projects. Resource
requirements are 69 MW lower because of
conservation measures, and plant upgrades reduce
requirements by an additional 52 MW.
By 2026 new capacity installations equal 1,332
MW: 650 MW of wind generation, 450 MW of
coal-fi red generation, 180 MW of other renewable
generation and 52 MW of plant effi ciency
upgrades. Resource needs are 138 MW lower
because of conservation. Figure 6 illustrates
the Company’s PRS.
A portion of the PRS requires construction of new
transmission capacity. The Company will continue
to work with regional entities and other utilities to
identify low cost solutions to move power across
the Northwest. Without new transmission, the
Company’s future resource portfolio likely will be
different than presented herein.
Carbon Emissions
Two carbon emission scenarios were developed for
the 2005 IRP. The National Commission on Energy
Policy study, completed in late 2004, provided
the basis for the fi rst carbon emission scenario.4
The second looked to an Energy Information
Administration study of the McCain-Lieberman
Climate Stewardship Act.5 These scenarios illustrate
the potential risk inherent in relying too heavily on
traditional coal-fi red technologies.
Table 2 explains how the 2005 plan includes more
non-carbon emitting resources relative to the 2003
IRP. The 2005 plan endeavors to acknowledge
and reduce greenhouse gas emissions by building
signifi cantly more renewable resources than
recommended in the 2003 IRP. Acquisition of the
second half of the Coyote Springs 2 gas plant
fulfi lled much of the 2003 IRP gas goal displayed
in the table.
3 Actual energy savings total nearly 111 aMW; however, due to expected
degradation of historical measures (16-year average measure life),
cumulative savings are estimated at 83 aMW.
4 See www.energycommission.org
5 See www.eia.doe.gov
VIII
Time Period Resource Type 2005 IRP 2003 IRP
2007-2016
Coal 215 350
Wind 122 25
Gas 121 178
Other Renewables 65 0
Conservation and Plant Upgrades 105 46
2007-2026
Coal 388 770
Wind 188 25
Gas 121 178
Other Renewables 145 0
Conservation and Plant Upgrades 174 92
Table 2: 2005 to 2003 IRP Comparison
Figure 6: Preferred Resource Strategy–Capacity (MW) 6
1,500
1,700
1,900
2,100
2,300
2,500
2,700
2,900
3,100
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Upgrades
Conservation
Renewables
Wind
Coal
Market
Existing Resources
Loads
6 Wind capacity is shown at its contribution to meeting system peak
demand. Wind is assumed to contribute 25 percent of nameplate capacity
to peak loads. See “Wind Contribution to Meeting System Peaks” in
Section 5 for further discussion.
IX
This acquisition is shown in the 2005 IRP column for
comparative purposes.
PRS Acquisition
The PRS is very capital intensive. It will require
outlays of approximately $1.5 billion by 2016. This
level equals more than 80 percent of the utility’s
present depreciated book value. The Company
might explore power purchase agreements with
third parties that include options to acquire the
underlying asset as a way to manage the fi nancial
impacts. Medium and short-term market purchases
also are expected to fi ll in modest gaps between
resource acquisitions and load requirements.
The Company believes that acquiring the amount
of wind and biomass included in the PRS will be
challenging, especially in light of our preference to
acquire smaller portions of geographically diverse
projects. Wind and biomass acquisitions therefore
might begin as early as 2007. In the 2005 IRP
Action Plan, the Company commits to continuing
its research into wind and biomass potential,
clean coal technologies, transmission solutions
and conservation. Each of these aspects will
be critical to successful implementation of the
Preferred Resource Strategy.
Action Items
The Company’s 2005 Action Plan outlines the
activities developed by the Company’s staff with
advice from its management and the Technical
Advisory Committee that will be undertaken to
support the PRS and improve the planning process
over the next two years. The Action Plan is found
in Section 8, Action Items. Action Item categories
include renewable energy and emissions,
modeling enhancements, transmission modeling
and research, and conservation. Progress on 2005
action items will be monitored, and the results will
be reported in Avista’s 2007 IRP.
X
The Company submits an Integrated Resource
Plan (IRP) to public utility commissions in Idaho and
Washington every two years as required by state
regulation1. Starting with the 1989 Least Cost Plan,
the 2005 electric IRP represents our ninth plan. It
describes the Preferred Resource Strategy meeting
future customer requirements.
The Company has a statutory obligation to provide
reliable electricity service to customers at rates,
terms and conditions that are fair, just, reasonable
and suffi cient. We assess resource acquisition
strategies and business plans to acquire resources
when our supplies are insuffi cient and to optimize
the value of our current resources. Avista regards
the IRP as a tool for resource evaluation, rather than
an acquisition plan for a particular project. The 2005
IRP therefore focuses on developing a methodology
for evaluating various resource decisions and bids
received in response to requests for proposals and
other resource acquisition efforts.
IRP Process
The Company actively seeks input from customers,
Commission Staff and other stakeholders in the
IRP process and other resource planning activities.
To facilitate stakeholder involvement in the 2005
IRP, the Company sponsored seven Technical
Advisory Committee (TAC) meetings. The fi rst
meeting convened on October 23, 2003, and the
last meeting was held on June 23, 2005. Over 70
people were invited to the meetings. Each meeting
focused on specifi c planning topics, reviewed the
status and progress of planning activities, and
solicited ongoing input as the IRP was developed.
The agendas and presentations for all of the TAC
meetings are located electronically in Appendices
A-C.
Stakeholder Involvement
Opportunity for stakeholder input into Avista’s
planning activities is substantial. The Company’s
public involvement efforts take three forms. First,
the Integrated Resource Planning process
provides several meetings for interested parties,
INTRODUCTION
AND STAKEHOLDER INVOLVEMENT
1 In Washington, IRP requirements are outlined in WAC 480-100-251
entitled “Least Cost Planning.” In Idaho, the IRP requirements are outlined
in Case No. U-1500-165 Order No. 22299, Case No. GNR-E-93-1, Order
No. 24729, and Case No. GNR-E-93-3, Order No. 25260.
XI
generally, the “expert public.” This group reviews
key assumptions, assists in framing IRP studies
and analyses, and reviews the results of the work
performed by the Company. Second, Avista takes
the approach of “niche” public involvement. This
recognizes that some customers and interested
parties are focused on issues important to them.
Examples include transmission corridor planning
and wildlife enhancement efforts. Lastly, Company
representatives participate in regional planning
efforts to obtain critical insights for incorporation into
Avista’s planning efforts. Examples of these forums
include Western Electricity Coordinating Council and
Northwest Power Pool committee involvement.
Public Process
The 2005 IRP was developed very much as a public
document. Each presentation given at the TAC
meetings was made available to the general public
on Avista’s Internet web site shortly after
the meeting. The presentations, along with a
list of active TAC members, may be found at
www.avistautilities.com. The 2005 IRP, including
its technical appendices, can be downloaded at
this location. A copy of our 2003 IRP is also
archived at the site.
IRP Technical Advisory Committee
Avista’s Integrated Resource Plan is informed by
signifi cant public input. The Company scheduled
seven meetings with its TAC during the preparation
of this plan. Topics included conservation, market
drivers, available resource options and technical
modeling issues. The 2001 IRP cycle included
three TAC meetings. The 2003 IRP benefi ted from
four meetings. The larger number of meetings for
Participant Organization Participant Organization
Aliza Seelig PSE John Seymour FPL Energy
Andy Ford WSU Ken Canon ICNU
Charlie Grist NPCC Leonard Coldiron Potlatch
Chris Bevil PSE Liz Klumpp WCTED
Chris Turner Pacifi Corp Lynn Anderson Idaho PUC
Danielle Dixon NW Energy Coalition Mallur Nandagopal City of Spokane
Dave Van Hersett NW Energy Services Patrick Saad Dana-Saad Company
Doug Loreen PSE Richard Nagy U. of Idaho
Hank McIntosh WUTC Rick Sterling Idaho PUC
Harry McLean City of Spokane Terry Morlan NPCC
Howard Ray Potlatch Tom Eckman NPCC
Jamie Stark Idaho Power Tom McLaughlin Potlatch
Joelle Steward WUTC Yohannes Mariam WUTC
Table 3: TAC Participants
XII
the 2005 IRP refl ects the Company’s interest in
obtaining more insight and review from third party
stakeholders, and the number and complexity of
topics and analyses included in the plan.
The TAC mailing list includes more than 70
individuals representing 47 organizations. The
Company recognizes the signifi cant efforts
necessary to participate in its TAC process. Table
1 recognizes individuals who actively participated
in the IRP planning process by attending one or
more of our TAC meetings and their respective
organizations. Table 2 details meeting dates and
Meeting Date Agenda Items
October 23, 2003
• Review of 2003 IRP DSM Approach
• Conservation Integration Methodologies
• Issues to Consider
August 4, 2004
• Review of Process and IRP Schedule
• IRP Topics Brainstorm
• Load Forecast
• 20-Year Loads and Resources Tabulation
January 25, 2005
• Overview of Natural Gas Forecast
• Capacity Planning Overview
• Load Forecast Update
• Loads and Resources Update
• Imputed Debt
February 17, 2005
• IRP Modeling Overview
• Modeling Futures and Scenarios
• More on Modeling Assumptions
• Modeling Emissions in IRP
• Supply-Side Resource Alternatives
• Selection of Future TAC Dates
March 23, 2005
• Conservation Integration in 2005 IRP
• Stochastic Risk Modeling
• Preliminary Capacity Expansion
• Update on Scenarios and Futures
• 2005 IRP Draft Outline
May 18, 2005
• Natural Gas Price Forecast Update
• Base Case Results
• LP Module/Selection Criteria
• Transmission Planning
• Scenario Results
• Avoided Costs
• Action Items for 2005 IRP
June 23, 2005
• Hydro Upgrades
• Emissions
• Conservation Results
• Draft Preferred Resource Strategy
Table 4: TAC Meeting Dates and Agendas
XIII
agenda topics presented by the Company. The
Company has worked diligently to obtain input
from the “general public.” We actively sought
participation through advertisements including
display ads in major circulation newspapers.
Unfortunately, in the past, very few customers
attended our scheduled meetings or otherwise
showed interest in Avista’s long-term planning efforts.
General public customers can be very interested
in collaboration, focusing on issues dear to them.
Some are motivated by a specifi c interest such
as an upgrade to, or construction of, transmission
corridor close to their property. The Company has
provided several opportunities for input on specifi c
issues, as described below.
Issue-Specifi c
Public Involvement Activities
Avista convenes collaborative processes to address
issues that have signifi cant public interest.
External Energy Effi ciency (“Triple E”) Board
The Triple E has met at least twice a year to
guide conservation efforts since 1995. The
Triple E predecessor, the DSM Issues Group, was
instrumental in shaping Avista’s DSM tariff rider,
the country’s fi rst 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 in 1998 to the fi rst all-party settlement
fi led with a FERC relicensing application. The
nationally recognized “Living License” concept
was an outgrowth of this stakeholder process.
The relicensing collaborative formed as part of the
Living License is now in its implementation phase,
with subsets of stakeholders participating in project
mitigation activities including the establishment of
conservation areas for wildlife preservation.
FERC Hydro Relicensing—
Spokane River Projects
The Company has convened a process similar to
the Clark Fork River Projects effort in relicensing
its Spokane River projects. Approximately 100-
stakeholder groups participated in this collaborative
effort. Draft license applications were fi led with
FERC on July 28, 2005.
Transmission Upgrade—Spokane Valley
Avista is constructing two new transmission
substations—Boulder in the Spokane Valley and
Dry Creek in southeast Clarkston, Washington—to
meet growing electricity demand in these areas.
Avista also is reconstructing the 230 kilovolt
(kV) transmission line linking Coeur d’Alene and
Spokane. Construction on each of these projects
began after numerous public meetings. Customer
input led the Company to choose alternative
locations for the Boulder substation and corridor
expansion.
XIV
Transmission Upgrade—Palouse
Avista is working on a new transmission line in the
Palouse region. This project also benefi ts from
public involvement.
Low Income Rate Assistance Program (LIRAP)
LIRAP progress is shared with the four community
action agencies in the Company’s Washington
service territory through regular meetings. At the
inception of the program in 2001, meetings were held
monthly to review administrative issues and needs.
Meetings are now convened on a quarterly basis.
Participation in Regional Planning
The Pacifi c Northwest generation and
transmission system operates in a coordinated
fashion. Avista is an active participant in several
regional organizations with planning efforts
that inform the Company’s integrated resource
planning process. Among the organizations Avista
participates in are:
ƒWestern Electricity Coordinating Council
ƒNorthwest Power and Conservation Council
ƒNorthwest Power Pool
ƒPacifi c Northwest Utilities Conference
Committee
ƒGrid West
ƒTransmission Improvements Group
ƒNorthwest Transmission Assessment Committee
ƒSeams Steering Group-Western Interconnection
ƒNorth American Electric Reliability Council
Future Public Involvement
The Company will continue to actively seek input
from its customers and other interested parties.
Advice will be requested where major impacts are
expected. For the IRP process specifi cally, TAC
meetings will remain open to the general public.
Outline of 2005 IRP Report
The 2005 IRP report contains eight sections,
an executive summary and this introduction.
Technical appendices are included as a supplement
to this report.
Executive Summary
This section summarizes the results and highlights of
the 2005 IRP.
Introduction & Stakeholder Involvement
This section introduces the IRP and explains the
involvement of all interested parties.
Section 1: Electricity Sales Forecast
This section covers the relevant local economic and
Company load forecasts.
Section 2: Resource Requirements
This section provides descriptions of Company-
owned generating resources, major contractual
obligations and rights, capacity and energy
tabulations, and a discussion about reserve margins.
XV
Section 3: Conservation Initiatives
This section covers Avista’s conservation programs,
the methodology and analysis of conservation
measures, descriptions of the conservation measures,
and a discussion of the results.
Section 4: Transmission Planning
This section discusses the Company’s transmission
system, and summarizes the Company’s and
regional transmission issues.
Section 5: Modeling Approach
This section covers the market simulation modeling
assumptions and inputs, risk modeling, the Avista
Linear Program Model, new resource alternatives
available to the Company and wind modeling.
Section 6: Modeling Results
This section covers the results of the Base Case and
scenario analyses for the Western Interconnect and
Mid-Columbia electricity market.
Section 7: Preferred Resource Strategy
This section provides details about the Company’s
Preferred Resource Strategy and how the PRS
compares to theoretical portfolios under stochastic
and scenario analyses.
Section 8: Action Items
This section recaps progress made on 2003 IRP action
items, and details action items for the 2005 IRP.
1-1
This section summarizes a variety of Company,
customer and load forecasts for our service territory.
The section concludes with discussions of both the
high and low load forecasts developed for the 2005
Integrated Resource Plan (IRP) and an overview
of recent enhancements made to the forecasting
models and processes.
1.1 Economic Conditions
in the Service Area
The Avista Utilities electric service territory covers a
wide swathe of Eastern Washington and Northern
Idaho. The geography is as diverse as the economy.
Rugged mountains, fertile river valleys and glacially
created plains provide natural resources, farmlands
and cityscapes for over 800,000 residents of the
Inland Northwest. Avista Utilities serves most of the
urbanized and suburban areas in 24 counties. See
Figure 1.1 for a map of the Company’s service territory.
1.2 Electric Operating
Division Economy
Over the last 20 years, the economy of the
Inland Northwest has transformed from a natural
resource-based manufacturing economy to
diversifi ed light manufacturing and services.
Manufacturing employment has declined along
with mining reserves in Shoshone County, Idaho,
and Stevens County, Washington. Much of the
mountainous area of the region is owned by the
Federal government and managed by the United
States Forest Service. Severe curtailments of timber
harvest on public lands have led to the closure of
many sawmills throughout the region. Two pulp and
paper plants served by Avista Utilities have large
private holdings of forested lands; they continue to
face stiff domestic and international competition for
their products.
1. ELECTRICITY SALES FORECAST
Section Highlights
 Avista will serve 350,000 electric customers in 2007, and nearly 485,000 in 2026.
 135,716 new jobs are forecast for Bonner, Kootenai and Spokane Counties by 2026,
a 61 percent increase from 2004 levels.
 Electric sales are forecast to grow 2.1 percent annually.
 2007 retail load (absent conservation) is forecast at 9,142 gigawatt-hours; 2026 is forecast
at 13,542 gigawatt-hours.
 Several large industrial facilities permanently closed in Washington and Idaho because
of the 2001-02 economic recession; the electric retail sales forecast assumes these closures
are permanent.
1-2
Two national recessions strongly impacted the Inland
Northwest during the 1980s. Economic slowdowns
typically are refl ected in employment data, with
employment expanding during expansionary times
and contracting during recessions. The 1980s
exemplifi ed that pattern with high levels of regional
unemployment. The U.S. recession in the early
1990s bypassed much of the area’s economy. The
most recent recession, beginning in 2001, provided
a harsh reminder of the diffi culty in insulating a
regional economy from national events. Historical
patterns of employment for the three principal
counties in the Company’s electric service area
are shown in Figure 1.2. Population levels often
are more stable than employment levels during
times of economic prosperity and decline; however,
during severe economic downturns, total population
often contracts as people leave in search of job
opportunities. The Company last experienced
population loss during the early 1980s. Figure 1.3
details population changes in Spokane, Kootenai,
and Bonner counties. Figure 1.4 shows total
population in the three counties.
1.3 The Economic Forecasts
Avista Utilities purchases national and county-level
employment and population forecasts from Global
Insight, Inc. (formerly Data Resources, Inc.), an
internationally recognized economic forecasting
consulting fi rm.
The Company purchases data for the three
principal counties comprising over 80 percent of the
service area economy, namely, Spokane County in
Washington; and Kootenai and Bonner Counties
Figure 1.1: Service Territory Map
1-3
Figure 1.3: Idaho and Washington Population Change by County (thousands)
Figure 1.2: Idaho and Washington Job Change by County (thousands)
1-4
in Idaho. The national forecast, on which these
regional forecasts are based, was prepared in March
2004; the county-level estimates were completed in
June 2004.
Employment and population forecasts provide the
basis for electric customer projections. Spokane
County, dominated by the economy of the City
of Spokane, is expected to exhibit moderate and
steady growth for the next 20 years. Kootenai
County, including the City of Coeur d’Alene, was
one of the fastest growing areas in the U.S. during
the 1990s. Our forecast anticipates continued and
signifi cant growth in this area. Bonner County,
located north of Kootenai County, is forecast to
experience steady but more modest growth over the
IRP timeframe.
1.4 Electricity Customer
Forecasts
The key driver of the electricity customer market is
population growth. Population drives the housing
market, a fundamental driver of commercial
customer expansion. Commercial markets expand
as more retail stores, schools, and other businesses
are attracted to an area to serve markets created by
the increased population. Other factors infl uencing
housing demand include interest rates, apartment
vacancy rates and student housing construction on
college campuses. The region’s housing market has
tightened substantially in recently years, absorbing
the surplus generated after the early 1990s
population boom. Low interest rates in 2004
nearly doubled residential building permits in
Spokane and Kootenai County when compared
to 2001 levels, increasing the numer of retail
customers. The unsold housing inventory also is at
Figure 1.4: Total Service Territory Population (thousands)
1-5
a cyclical low. The region’s strong housing market is
expected to continue, at a more modest rate, over
the next decade.
Over the 20-year horizon, overall customer growth is
estimated to average 1.8 percent per year in for the
period 2005-2025. This level of growth is somewhat
faster than the 1.3 percent experienced during the
past fi ve years. Figure 1.5 provides detail about the
forecasted growth in lighting, industrial, commercial
and residential accounts. Relative to the other
customer classes, street lighting loads are very
small and are not included in the fi gure.
1.5 Retail Electricity Sales
Forecast
Between 1997 and 2004, the region was affected
by major economic changes, not the least of
which was a marked increase in retail electricity
prices. The energy crisis of 2000-01 included
the implementation of widespread, permanent
conservation efforts by our customers. In 2004,
rising retail electricity rates reinforced conservation
efforts. Several large industrial facilities served by the
Company permanently closed during the 2001-02
economic recession. The electric retail sales forecast
takes a conservative approach, assuming these
closures are permanent. However, if any of these
major industrial facilities reopen, the annual electric
retail sales forecast will be adjusted accordingly.
Retail electricity consumption rose 1.2 percent
annually from 1998 through 2004. This increase
was in spite of the combined impacts of higher
prices and decreased electricity demand during the
energy crisis. The forecasted annual increase in fi rm
sales over the 2005 to 2025 period is 2.1 percent.
Figure 1.5: Electric Utility Customer Forecast (thousands)
1-6
The forecast is broken into several customer classes
in Figure 1.6.
1.6 Price Elasticity1
Elasticity is one of the central concepts of
economics and must be considered when
forecasting electricity demand. Price elasticity of
demand, or “own price” elasticity, is the ratio of
the percentage change in the quantity demanded
of a good or service, in this forecast electricity, to
a one-percent change in its price. In other words,
elasticity measures the responsiveness of buyers to
a price change. A consumer who is very responsive
to a price change has a relatively elastic demand,
whereas a customer who is unresponsive to price
changes has a relatively inelastic demand.
Consumers illustrated elastic electricity demand
during the 2000-01 energy crisis, reducing 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
coeffi cient indicates that the two products are
substitutes; a negative coeffi cient indicates they
are complementary goods. Substitute goods are
replacements for one another. As the price of the
fi rst good increases relative to the price of the other
good, consumers shift their consumption to the
second good. Complementary goods are used
together, so increases in the price of one good
will result in a decrease in demand for the second
good as consumers reduce consumption of the fi rst
good. For Avista, the dominant impact on electricity 1 The elasticity defi nitions used in this section were paraphrased from
Economics: Principles, Problems, and Policies by Campbell R. McConnell
and Stanley L. Brue, 14th edition.
Figure 1.6: 2005 Electric Utility Retail Sales Forecast (GWh)
1-7
demand is the substitutability of natural gas in some
applications, such as water and space heating.
Income elasticity of demand is the ratio of the
percentage change in the quantity demanded of a
good to a one-percent change in consumer income.
Income elasticity measures the responsiveness
of consumer purchases to income changes. For
electricity demand, there are two impacts on
consumption. The fi rst impact is the affordability
impact. As income increases, a consumer’s ability
to pay for products and services increases. The
second income-related impact is the amount and
number of customers using equipment within
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. Avista estimates price elasticity
by customer class in its computation of electricity
and natural gas usage. Residential customer price
elasticity is estimated at negative 0.15; for each one
percent increase in the price of electricity, usage
falls by 0.15 percent. Commercial customer price
elasticity is negative 0.10. The cross-price elasticity
of natural gas with electricity is estimated to be
positive 0.10. The income elasticity is estimated at
positive 0.75. Figures 1.7 and 1.8 illustrate how the
price projections are used to determine elasticity
impacts. As rates increase or decrease, consumers
will adjust electricity usage according to their elasticity.
Price elasticity at these levels will not greatly
affect the demand forecast. Real income per
household is forecast to increase at an average of
Figure 1.7: Residential Retail Rate Projection for Retail Load Forecast (cents/kWh)
1-8
Figure 1.8: Wholesale Natural Gas Price Forecast for Retail Load Forecast ($/dth)
1.3 percent annually between 2005 and 2025. This
increase results in fl at residential usage and a small
upward drift in commercial usage per customer.
Commercial growth is attributed mostly to a higher
concentration of big box retailers, offi ce buildings
and future school construction.
1.7 Alternative Scenarios
The discussion so far has concentrated on the
“Base Case,” medium or “most-likely” forecast
for electricity consumption by our customers.
Forecasting is necessarily uncertain, so alternative
electricity growth scenarios are used to provide
insight and guidance for resource acquisition plans.
With the advice and consultation of the Technical
Advisory Committee, “high” and “low” economic
forecasts were prepared. The principal determinant
of these alternatives was population change within
the Company’s existing service area. As such,
no assumptions for service area expansion or
integration of existing electricity customers located
within the service area, but served by other utilities,
is expressed or implied by these alternatives. For
example, the Kaiser Aluminum Rolling Mill in the
Spokane Valley is assumed to continue to be served
by the Bonneville Power Administration even though
it is located within our service territory.
The alternative forecasts are presented in Figure 1.9.
The scenarios are specifi c to this IRP; they should
not be confused with other Company or agency
forecasts. The scenarios also are not boundary
forecasts, in that the high forecast should
1-9
not be considered the highest possible load trajectory,
and the low forecast does not represent the lowest
possible forecast.
1.8 Enhancements to
the Forecasting Models
and Process
The forecasting models were updated with the
latest energy consumption profi les for the 2005 IRP.
The model’s coeffi cients were checked for price
elasticity impacts, and the updated values were
incorporated into the forecast. Recent electricity
consumption levels, driven largely by the recent
increase in electricity and natural gas prices, showed
a reduction in price elasticity for our residential
Figure 1.9: Retail Sales Forecast Scenarios (GWh)
and commercial customers. We use conservative
elasticity estimates for industrial customers, because
rising commodity prices can curtail their international
competitiveness.
2-1
A critical aspect of integrated resource planning is
the long-term tabulation of loads and resources.
Loads refer to projections of how much capacity
and energy customers are expected to consume
over the length of the planning period. Resources
refer to the generating assets owned, or controlled
through contracts, by the Company. The differences
between loads and resources illustrate potential
2. RESOURCE REQUIREMENTS
Section Highlights
 The Company requires new generation resources as early as 2009.
 The IRP includes a planning margin of approximately 15 percent.
 Although in balance on an annual basis, every year of the IRP horizon contains monthly defi cits.
 Approximately half of customer requirements in 2007 will be met with renewable resources,
including various hydro plants, our biomass facility at Kettle Falls and a wind contract from the
Stateline Wind Farm.
 Our largest hydroelectric facilities, on the Clark Fork River, operate under a federal license through
2046; the Spokane River project license expires in 2007 and presently is in the renewal process.
 Approximately 25 percent of our portfolio is natural gas-fi red; medium-term market contracts will
serve nine percent of customer requirements in 2007.
needs the Company must address through its
future actions. This section details Company-
projected resources and loads for the next 20
years. A summary of the Company’s conservation
initiatives—they also affect requirements—is
contained in Section 3- Conservation Initiatives.
2.1 Utility-Owned Resources
The Company uses a diversifi ed portfolio of
generating assets to provide electricity to its
customers. Avista owns and operates eight
hydroelectric projects on the Spokane and Clark
Fork Rivers. The Company thermal assets include
partial ownership of two coal-fi red units in Montana,
three natural gas-fi red projects within its service
territory, another natural gas-fi red project in Oregon,
and a wood waste generating plant near Kettle Falls,
Washington. Each resource is described herein.
2-2
Hydroelectric Projects on the
Spokane River
The Company 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.
Following is a short description of the Spokane River
projects with the maximum capacity and nameplate
ratings listed for each. 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 stated by the
manufacturer. Figure 2.1 is a map of all Company
owned hydroelectric projects.
Post Falls
The Post Falls project was completed in 1906
at Post Falls, Idaho. The plant was updated in
1980 with an additional unit. Its fi ve units have a
maximum capacity of 18.0 MW and a nameplate
rating of 14.8 MW.
Upper Falls
The Upper Falls project was completed in 1922 in
downtown Spokane. The single unit project has a
maximum capacity of 10.2 MW and a nameplate
rating of 10.0 MW.
Monroe Street
The Company’s fi rst generating plant, Monroe
Street, was built on the Spokane River in 1890. It
is located in downtown Spokane at Riverfront Park.
Figure 2.1: Avista’s Hydroelectric Projects
2-3
The plant was rebuilt in 1992. Its single unit has a
maximum capacity of 15.0 MW and a nameplate
rating of 14.8 MW.
Nine Mile
The Nine Mile project was constructed in 1908 by a
private developer near Nine Mile Falls, Washington.
The Company acquired Nine Mile in 1925 from the
Spokane & Eastern Railway. The four units at the
facility have a combined maximum capacity of 24.5
MW and nameplate rating of 26.4 MW.
Long Lake
The Long Lake project was built in 1915 above Little
Falls. It was “the world’s highest spillway dam” with
the largest turbines in existence at that time. The
plant was upgraded in 1999 with the installation of
new runners. The total maximum capacity of its
four units is 88.0 MW; it has a nameplate rating of
70.0 MW.
Little Falls
The Little Falls project is located on the Spokane
River near Ford, Washington. Completed in 1910, it
has four units with a combined maximum capacity
of 36.0 MW and a nameplate rating of 32.0 MW.
Clark Fork River Projects
The Clark Fork River Project consists of two large
hydroelectric plants in Clark Fork, Idaho, and Noxon,
Montana. The two plants operate under a FERC
license that was extended in 1999 to 2046.
Cabinet Gorge
Cabinet Gorge began generating power in 1952.
Two additional units, bringing the total to four, were
added in 1953. The current maximum capacity
of the plant is 261.0 MW; its nameplate rating is
265.2 MW.
Noxon Rapids
Noxon Rapids consists of four units installed
between 1959 and 1960. A fi fth unit was installed in
1977. The plant currently has a maximum capacity
of 527.0 MW and a nameplate rating of 466.2 MW.
Total Hydroelectric Generation
In total, our hydroelectric plants are capable of
generating as much as 979.7 MW. Table 2.1
summarizes the Company’s hydro projects.
2-4
Thermal Resources
The Company owns or leases and maintains several
thermal resources across the Northwest. Each
plant is expected to remain available through the
duration of the Integrated Resource Plan (IRP) study
period. The Company’s thermal resources provide
dependable low cost resources that serve base
load needs as well as provide peak load serving
capabilities. Table 2.2 provides a summary of the
Company’s thermal projects.
Colstrip
The Colstrip plant,
located in eastern
Montana, consists of
four coal-fi red steam
plants owned by
a group of utilities.
PPL Global operates the facility. The Company
owns 15 percent of Units 3 and 4. Unit 3 was
completed in 1984 and Unit 4 was fi nished in 1986.
The Company’s share of each Colstrip unit has a
maximum capacity of 111.0 MW and a nameplate
rating of 116.7 MW.
Rathdrum
Rathdrum is a two-unit, simple-cycle, gas-fi red plant
near Rathdrum, Idaho, that entered service in 1995.
The plant has a maximum capacity of 176.0 MW
and a nameplate rating of 167.2 MW.
Northeast
The Northeast plant, located in northeast Spokane,
is a two-unit aero-derivative simple-cycle plant
completed in 1978. The plant can burn either
natural gas or fuel oil, although current permits
prevent the use of fuel oil. The combined maximum
capacity of the units is 66.8 MW with a nameplate
rating of 61.8 MW.
Table 2.1: Company-Owned Hydro Resources
Project Name
River
System Location
Project
Start
Date
Nameplate
Capacity
(MW)
Maximum
Capability
(MW)
60-Year
Energy
(aMW)
License
End
Date
Monroe Street Spokane Spokane, WA 1890 14.8 15.0 13.2 7/2007
Post Falls Spokane Post Falls, ID 1906 14.8 18.0 9.9 7/2007
Nine Mile Spokane Nine Mile Falls, WA 1925 26.4 24.4 16.4 7/2007
Little Falls Spokane Ford, WA 1910 32.0 36.0 22.8 N/A
Long Lake Spokane Ford, WA 1915 70.0 90.4 52.4 7/2007
Upper Falls Spokane Spokane, WA 1922 10.0 10.2 8.8 7/2007
Cabinet Gorge Clark Fork Clark Fork, ID 1952 265.2 261.0 122.2 3/2046
Noxon Rapids Clark Fork Noxon, MT 1959 466.2 527.0 202.9 3/2046
Total All Hydro 879.3 979.7 442.9
2-5
1 The Rathdrum generating plant is operated under a third party lease.
Table 2.2: Company-Owned Thermal Resources
Project Name Location Fuel
Start
Date
Nameplate
Capacity
(MW)
Maximum
Capability
(MW)
Energy
Capability
(aMW)
Colstrip 3 (15%) Colstrip, MT Coal 1984 116.7 111.0 93.3
Colstrip 4 (15%) Colstrip, MT Coal 1986 116.7 111.0 93.3
Rathdrum1 Rathdrum, ID Gas 1995 166.5 176.0 135.6
Northeast Spokane, WA Gas/Oil 1978 61.8 66.8 9.8
Boulder Park Spokane Valley, WA Gas 2002 24.6 24.6 23.2
Coyote Springs 2 Boardman, OR Gas 2003 287.0 274.0 233.8
Kettle Falls Kettle Falls, WA Wood 1983 46.0 50.7 42.2
Kettle Falls CT Kettle Falls, WA Gas 2002 6.9 6.9 6.1
Total All Thermal 886.2 821.0 651.4
Boulder Park
The Boulder Park project was completed in Spokane
Valley in 2002. The site has six natural gas-fi red
internal combustion engines. The combined
maximum capacity and the nameplate rating of all of
the units is 24.6 MW.
Coyote Springs 2
Coyote Springs 2 is a natural gas-fi red combined-
cycle combustion turbine located near Boardman,
Oregon. The plant entered service in 2003. The
maximum capacity is 269.0 MW. Its nameplate
rating is 287.0 MW. A duct burner provides the unit
with an additional capability of up to 25.0 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 hog fuel (wood).
It has a maximum capacity of 50.0 MW and a
nameplate rating of 50.7 MW.
Kettle Falls CT
The Kettle Falls CT is a natural gas-fi red
combustion turbine that began service in 2002. It
has a maximum capacity rating of 6.9 MW. Exhaust
heat from the plant is routed into the Kettle Falls
biomass plant boiler to increase its effi ciency. The
plant is capable of running independent of the
biomass steam plant.
Power Purchase and Sale Contracts
The Company utilizes several power supply
purchase and sale arrangements of various lengths
to meet a portion of its load requirements. This
section describes various contracts in effect during
the IRP timeframe. The contracts provide a number
of benefi ts to the Company, including low-cost
hydro and wind power. An annual summary of our
contracts is contained in Table 2.3.
2-6
Bonneville Power Administration (BPA) –
Residential Exchange
The Company entered into a settlement agreement
of the BPA Residential Exchange Program effective
on October 1, 2001. Over the fi rst fi ve years of the
ten-year settlement, the Company receives fi nancial
benefi ts equivalent to purchasing 90 aMW at BPA’s
lowest cost-based rate. Beginning October 1, 2006,
the Company’s benefi t level increases to 149 aMW.
At BPA’s option, the 149 aMW may be provided in
whole or in part as fi nancial benefi ts or as a physical
power sale; the IRP assumes the former based on
regional discussions.
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 for an energy exchange through June 30,
2019, with an agreement to reimburse the Company
Contract Name Start Date Capacity (MW) Energy (aMW) End Date
Grant County Purchase 2005 129.3 71.0 TBD
Rocky Reach Purchase 1961 37.7 19.3 Oct-2001
Wells Purchase 1967 28.6 9.9 Aug-2018
PGE Capacity Sale 1992 150.0 0.0 Dec-2016
Upriver Dam Purchase 1966 14.4 10.0 Dec-2011
WNP-3 Purchase & Sale 1987 82.0 48.0 Jun-2019
Medium-Term Purchase 2004 100.0 100.0 Dec-2010
PPM Wind Purchase 2004 35.0 9.8 Mar-2013
Total Contract 577.0 268.0
Table 2.3: Signifi cant Contractual Rights & Obligations
for certain WPPSS – Washington Nuclear Plant No.
3 (WNP-3) preservation costs and an irrevocable
offer of WNP-3 capability for acquisition under the
Regional Power Act.
The energy exchange portion of the settlement
contains two basic provisions. The fi rst provides
the Company with approximately 42 aMW of
energy from BPA through 2019, subject to a
contract minimum of 5.8 million mega-watt hours
(MWh). The Company is obligated to pay BPA
operating and maintenance costs associated
with the energy exchange as determined by a
formula that has a range of $16 to $29 per MWh,
expressed 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
2-7
under the Confi dence Interval Planning heading of
this section 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
of these plants was very large compared to the
loads then served by the PUDs. To assist with
fi nancing, and to ensure a market for the surplus
power, long-term contracts were signed with other
public, municipal, and investor-owned utilities
throughout the Northwest.
The Company entered into long-term contracts for
the output of four of these projects “at cost.” In
2007, the contracts provide energy, capacity, and
reserve capabilities; they provide approximately 138
MW of capacity and 70 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 2005 IRP does
not include energy or capacity for these contracts
beyond their expiration dates.
The Company recently renewed its contract with
Grant PUD 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,
is 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
The Company purchased 100 MW of “fl at” power
from 2004 through 2010 from several suppliers in
late 2001 and early 2002.2
Nichols Pumping Station
The Company provides energy to operate its share
of the Nichols Pumping Station, the supplier of water
for the Colstrip plant. The Company’s share of the
Nichols Pumping Station load is approximately
one aMW.
Table 2.4: Mid-Columbia Contract Quantities Summary
Project Name
2007 2012 2017 2022 2026
MW aMW MW aMW MW aMW MW aMW MW AMW
Rocky Reach 37.7 19.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Wells 28.5 15.4 28.5 15.4 28.5 15.4 0.0 0.0 0.0 0.0
Grant County 72.2 35.1 21.8 10.8 17.1 8.3 11.9 5.8 7.8 3.8
Totals 138.4 70.1 50.3 26.2 45.6 23.7 11.9 5.8 7.8 3.8
2 Delivery will occur in every hour of the contract term.
2-8
Portland General Electric
The Company provides Portland General Electric
(PGE) with 150 MW of fi rm capacity under contract
through December 31, 2016. PGE may schedule
deliveries up to its capacity limit during any ten
hours of each
weekday. Within
168 hours PGE
returns energy
delivered under the
contract.
Stateline Wind
Energy Center
The Company
entered into a contract with PPM Energy in 2004 for
35 MW of wind capacity out of the Stateline Wind
Energy Center located on the border of Oregon and
Washington. This 35 MW contract does not include
fi rming services. It was entered into in part to meet
a 2003 IRP Action Item.
2.2 Capacity Tabulation
The Company develops a twenty-year service
territory forecast of peak capacity loads and
resources for the IRP. Peak load is the maximum
one-hour obligation on the expected average
coldest day in January, including operating reserves.
Peak resource capability is the maximum one-
hour generation capability of Company resources,
Table 2.5: Loads & Resources Capacity Forecast (MW)
2007 2008 2009 2010 2011 2016 2021 2026
Obligations
Retail Load3 1,704 1,754 1,799 1,860 1,898 2,137 2,343 2,573
Operating Reserves 260 265 269 274 278 299 317 338
Total Obligations 1,964 2,019 2,068 2,134 2,176 2,436 2,660 2,911
Existing Resources
Hydro 1,100 1,100 1,066 1,059 1,028 1,016 983 978
Conservation 5 9 14 18 23 46 69 92
Net Contracts 159 159 165 164 48 49 118 118
Coal 222 222 222 222 222 222 222 222
Biomass 50 50 50 50 50 50 50 50
Gas Dispatch 303 308 303 303 307 303 303 308
Gas Peaking Units 243 243 243 243 243 243 243 243
Total Existing Resources 2,082 2,090 2,062 2,059 1,920 1,928 1,988 2,010
Net Position 118 71 -5 -75 -256 -508 -673 -901
Planning Margin 21.8% 18.5% 13.7% 9.6% -0.1% -11.7% -17.6 -24.6%
3 Retail load is absent historical conservation acquisitions levels. Historical conservation levels are counted as a resource, thereby increasing retail load for
purposes of the load and resource charts presented in this plan. This treatment has no impact on power generation acquisitions going forward.
2-9
including net contract contribution at the time of
the one-hour system peak. This calculation is
performed to insure that the Company has suffi cient
resources to meet its load obligations.
The Company has surplus capacity through 2008.
Annual capacity defi cits begin in 2009, with loads
exceeding resource capabilities by fi ve MW. The
defi cits continue to grow as peaking requirements
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. Table 2.5
summarizes the forecast.
The Company currently has suffi cient 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. In general, future capacity
requirements will be addressed by acquiring new
resources that provide both energy and capacity.
2.3 Energy Tabulation
Table 2.6 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 confi dence level. Confi dence
interval planning is discussed later in this section.
Table 2.6: Loads & Resources Energy Forecast (aMW)
2007 2008 2009 2010 2011 2016 2021 2026
Obligations
Retail Load 1,125 1,160 1,197 1,232 1,268 1,424 1,566 1,725
90% Conf. Interval 193 193 193 189 188 184 148 148
Total Obligations 1,318 1,353 1,390 1,420 1,456 1,608 1,715 1,873
Existing Resources
Hydro 510 510 506 487 483 464 447 444
Conservation 5 9 14 18 23 46 69 92
Net Contracts 234 234 234 235 131 104 57 57
Coal 182 193 181 181 193 181 181 193
Biomass 42 44 40 44 42 43 42 44
Gas Dispatch 282 268 282 272 282 268 282 272
Gas Peaking Units 145 145 145 141 145 142 146 132
Total Existing Resources 1,400 1,403 1,402 1,380 1,299 1,248 1,224 1,233
Net Position 82 50 12 -40 -157 -360 -491 -640
2-10
Figure 2.2: Energy Load and Resource Tabulation (aMW)
Similar to Table 2.5, maintenance schedules affect
the output of various plants over the IRP timeframe.
Specifi cally, coal, biomass, gas dispatch and gas
peaking units are affected.
As shown, only after 2009 are new resources
necessary to continue meeting the 90 percent
confi dence interval criterion. The table shows that
the Company is in a surplus position through 2009
on an annual basis. Figure 2.2 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
included as reductions to retail load. The 2005
IRP instead includes 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
in Table 2.6 are cumulative beginning in 2007 and
illustrate the Company’s commitment to continued
acquisition of cost-effective conservation. Activities
beyond current levels are discussed in Section 3-
Conservation Initiatives and are shown as new
resources in later tabulations.
The Company expects to encounter energy defi cits
during some months in all forecast years. As an
example, the Company anticipates defi cits in
January, March, August, September, October and
December of 2007 even though the annual position
is surplus by 82 aMW. Surplus positions occur in
the remaining months, particularly during spring
runoff. The Company balances its monthly positions
2-11
through short-term market purchases or sales,
exchanges or other resource arrangements.
The annual energy load and resource projections
are used to determine when the Company needs to
acquire additional resources to meet fi rm loads. The
fi rst annual energy defi cit of 40 aMW is expected in
2010. This defi cit is forecasted to grow to 360 aMW
by 2016 and 640 aMW by 2026. A signifi cant portion
of the projected defi cits results from the loss of Mid-
Columbia contracts as well as retail load increases.
2.4 Reserve Margins
Planning reserves accommodate situations at
times when loads exceed 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 of the disagreements
stem from differences between systems, such
as resource mix, system size and transmission
interconnections. For example, a hydro-based utility
generally has a higher ratio of capacity to energy
than a thermal-based company. Some advocate
carrying reserve levels equal to the largest resource
on a specifi ed system. Others, including the authors
of FERC’s recent Standard Market Design, believe
that margins should be set between 12 and 18
percent of forecast peak load. California requires
that all load serving entities under its jurisdiction
carry a 15 percent planning margin calculated as a
percentage of peak load.
Reserve margins, on average, increase customer
rates when compared to resource portfolios without
reserves. A 100 MW block of reserve resources
currently costs between $35 and $50 million in
capital expenditure, or $5 to $7 million per year.
Reserve resources have the physical capability
to generate electricity, but their high operating
costs limit economic dispatch and the potential
to create revenues to offset capital costs. Some
argue that regions with deregulation, or “customer
choice,” provide strong incentives for industry
participants to underestimate their reserve
obligations and lower costs.
Reserve margin obligations can be reduced
in a larger system comprised of many market
participants. Table 2.7 uses an operating reserve
example to explain how margins can be reduced
for all participants when entities commit to sharing
reserve obligations. The example is based on one
matrix of operating reserve margin—reserves should
be carried in an amount equal to a company’s
single largest resource. Total resource obligations
are reduced by one-third to 9.1 percent from 11.4
percent in the example.
When one load serving entity violates its reserve
Table 2.7: Reserve Sharing Example
Total
Resources
(MW)
Largest
Resource
(MW)
Margin
(%)
Utility A 10,000 1,000 10.0
Utility B 1,000 250 25.0
Total 11,000 1,250 11.4
Utilities
A&B 11,000 1,000 9.1
2-12
margin obligation, especially under a larger multi-
entity reserve sharing agreement, there likely will
not be a system-wide emergency during tight
market conditions. The violating company, as well
as its customers, will benefi t as free riders from
lower system costs at the expense of other market
participants. If several entities simultaneously
violate their planning margin obligations, high
wholesale prices and/or load curtailment might
occur. Therefore, it is important for utilities to be
diligent in carrying adequate reserve levels to insure
system reliability. To this end, many in the industry
advocate for the defi nition and enforcement of
reserve levels.
Avista 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 ten percent of our one-hour
system peak load plus 90 MW. The 90 MW fi gure
accounts for approximately 60 MW of hydro and
30 MW of Colstrip reserves. During extremely cold
conditions, fl ows into our hydroelectric plants taper
off as ice forms along the river banks. Experience
shows that fuel-handling problems can limit Colstrip
production during cold snaps. This amounts to
roughly a 15 percent planning reserve margin during
the Company’s peak load hour.
Confi dence Interval Planning
The Company uses confi dence interval planning
to insure it has resources adequate to meet its
customers’ energy requirements. Extreme weather
conditions can affect monthly energy obligations by
up to 30 percent. If the Company lacks generation
capability to meet high load variations, it exposes
the Company to increased volatility in the short-
term marketplace.
Evaluation of historical data indicates that an optimal
criterion is the use of a 90 percent confi dence
interval based on the monthly variability of load
and hydroelectric generation. This results in a ten
percent chance of the combined load and hydro
variability exceeding the planning criteria for each
month. In other words, there is a ten percent chance
the Company would need to purchase energy from
the market in any given month. The criterian is
identical to the 2003 IRP level of 80 percent. Based
on 2003 IRP feedback, the Company learned that
using a two-tail statistical measurement was confusing
to readers. Shifting to a single-tail test better
illustrates the concept of a one-in-ten probability.
The Company has considered using larger
confi dence intervals, but analysis suggests that
the cost of adding additional resources to cover
higher levels of variability would exceed the potential
benefi ts. Building to the 99 percent confi dence
interval could signifi cantly decrease the frequency of
market purchases, but such a criterion 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.
2-13
The 90 percent confi dence level varies between 94
aMW and 258 aMW on a monthly basis in 2007,
or 160 aMW across the twelve-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 chance of exceedance rather than an
annual fi gure. Additional variability is inherent in the
WNP-3 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 confi dence interval are increased by 33
aMW to account for the WNP-3 obligation through
its expiration in 2019.
Sustained Peaking Capacity
Parallel to planning margins lies the “gray area”
between energy and capacity planning termed
sustained peaking capacity. Sustained peaking
capacity is a tabulation of loads and resources
over a period exceeding the traditional one-hour
defi nition. It is also a measure of reliability and
recognizes that peak loads do not stress the system
for just one hour. Table 2.8 details the assumption
differences between the Company’s planning
approach and the sustained capacity approach.
The preliminary results gathered from work on the
2005 IRP suggest the Company should study this
topic further. It is included as an action item in
Section 8. Where the additional study supports
changing the planning criteria, we will review such a
move with our Technical Advisory Committee.
Table 2.8: Capacity L&R Versus Sustained Capacity
Item Capacity L&R Sustained Capacity
Period One Hour One Hour to Three Days or More
Peak Load Average Coldest Day Temperature Highest Load on Record
Thermals Average Temperature & Colstrip
Reduced for Freeze (~30 MW)
Lowest Temperature & Colstrip
Reduced for Freeze (~30 MW)
Hydro Maximum Capability
Reduced for Freeze (~60 MW)
Maximum Capability
Reduced for Freeze (~60 MW)
Contracts Actual Forecast Actual Forecast
2-14
3-1
Avista Utilities began offering conservation programs
in 1978 to encourage effi cient energy use. Since
1978, 111 aMW of energy has been acquired
through Company
programs.1 In 1995, the
Company initiated the
nation’s fi rst non-by-
passable distribution
charge, otherwise
known as the DSM tariff
rider, to ensure long-
term stable conservation
funding. Avista’s current
conservation programs are operationally divided
into commercial/industrial, residential and limited
income portfolios. Figure 3.1 details the Company’s
acquisition successes over time.
The fl exible nature of Avista’s programs allows it to
offer customized conservation services and technical
assistance for any cost-effective commercial or
industrial electric effi ciency measure. The Company
also provides prescriptive conservation programs for
specifi c common measures.
The comprehensive nature of Avista’s commercial
and industrial programs impacts the methodology
used to evaluate conservation options in this IRP
and the evaluation of future business planning.
The limited income program is offered through
several community agencies with broad discretion
to pursue energy-effi ciency measures among limited
income and vulnerable customer groups. There
also is limited funding for health and human safety
measures designed to enhance the life of effi ciency
measures, the habitability of the residence, and
3. CONSERVATION INITIATIVES
Section Highlights
 In 1978 Avista began acquiring conservation, focusing on residential audits, and providing
incentives for shell and water heater insulation.
 Residential programs were ramped up in 1980 to focus on weatherizing, infi ltration reduction,
windows and water heater insulation measures.
 Avista regulators approved the nation’s fi rst non-by-passable distribution charge in 1995.
 Responding to the 2000-01 Western Energy Crisis, the Company acquired over 20 aMW of
conservation in 2001 alone.
 Avista reached a milestone in 2002—100 MW of conservation.
 The 2005 IRP increases our conservation acquisition goal by 50 percent.
1 Due to expected degradation of historical measures (16-year average
measure life), cumulative savings in effect today are estimated at 83 aMW.
3-2
energy safety. Avista augmented agency funding
with Conservation and Renewable Discount dollars
received from the Bonneville Power Administration
beginning in 2003.
Residential programs are exclusively prescriptive
in nature because of the relatively small nature of
residential electric usage. The Company offers
a number of programs in this class, including
improved space and water heating effi ciencies,
improved shell effi ciency and more effi cient
residential lighting. The space and water heating
components of these programs include the
conversion of space and water heating appliances
from electricity to natural gas. All existing and
several new and promising residential measures are
incorporated in the 2005 IRP evaluations.
The Company launched a major conservation
response to the 2001 western states energy crisis.
The acceleration was a cost-effective strategy that
helped mitigate the impacts of abnormally high
wholesale energy prices. Program funding was
derived from the DSM tariff rider. As a result of
this extraordinary utility effort, the Company spent
$12.4 million more on conservation measures
than was collected from the tariff rider in 2001. To
address the resultant tariff rider defi cit, the Company
established a 2002-2005 business plan designed to
meet regulatory obligations, to fi eld a cost-effective
conservation portfolio and to expeditiously return the
tariff rider balance to zero.
The return to a zero balance was and continues to
be achieved through a series of sustainable and
non-sustainable cost containment measures and
through the targeting of low- or no-cost measures
and lost opportunities. As individual tariff rider
balances approach zero in each state, the target
markets of each component are redefi ned to include
Figure 3.1: Historical Electric Conservation Acquisition
3-3
all cost-effective measures, and program support is
increased to meet available opportunities.
Even with the Company’s recent cost-containment
measures, it has continued to materially achieve
the conservation goal specifi ed in the electric tariff
rider, as illustrated in Figure 3.2.2 Avista’s prorata
share of the Northwest Power and Conservation
Council’s Fifth Power Plan conservation goal is shown
for reference.
The Company reviewed its 2002-05 business plan
in early 2004, concluding that a 10 to 25 percent
conservation funding increase was needed to
support the 2005 electric IRP. The anticipated
increase led to program revisions and to the
acceleration of selected program components
in anticipation of additional cost-effective
opportunities. The ramp-up included the launch of
several projects piloting alternative implementation
strategies for prescriptive air conditioning and
lighting measures, as well as larger commercial and
industrial site-specifi c projects.
Analyses of these pilots, along with an assessment
of contracts acquired under Avista’s 2000 all-
resource request for proposal process, indicates that
direct customer incentives are insuffi cient to support
the programs necessary to achieve future goals.
Revisions to the Company’s electric conservation
tariff that would roughly double customer direct
incentives was approved in Idaho (effective March
2005) and Washington (effective July 2005).
The aggregate tariff rider defi cit approached zero in
August 2005. The Company is in the fi nal stages of
transitioning to the 2006 conservation business plan. 2 Figure 3.2 includes resources acquired through a cooperative program
with the Northwest Energy Effi ciency Alliance.
Figure 3.2: Electric Conservation Acquisition Versus Goals (aMW)
0
5
10
15
20
25
Actual aMW
aMW goal
With Regional
NPCC
20041999200020012002 20052003
3-4
It will focus on acquiring all cost-effective
conservation opportunities given the results of the
2005 IRP. The IRP process enables the Company
to determine the level of conservation acquisition,
the target markets and the measures that will be
incorporated into the future business plan.
3.1 IRP Objective
The primary purpose of the IRP evaluation for
conservation is to:
• Establish an aggregate level of cost-effective
projects for acquisition through local utility
programs. This becomes the future
conservation goal.
• Assess individual markets and measures on which
to focus future acquisition efforts. This is
applied to future business planning efforts,
including marketing and staffi ng decisions.
• Identify specifi c prescriptive conservation
programs for the residential sector. All
measures will be thoroughly defi ned as part of
the 2006 conservation business plan.
Results of the IRP do not displace tariff rider
obligations. There is signifi cant variation within the
measure categories evaluated in the IRP process.
It is not uncommon for specifi c applications of
generally cost-ineffective measures to be individually
cost-effective. Similarly, not all applications of
generally cost-effective measures will always be
cost-effective for individual projects. The Company
has incorporated in our incentive calculation model
an assessment of a “sub TRC” calculation to
provide cost-effectiveness feedback on an individual
project basis. The “total resource cost” (TRC) test
is designed to ascertain whether an investment
is economically justifi ed when all of its costs and
benefi ts are included. The “sub TRC” calculation
excludes relatively fi xed non-incentive utility costs
that are diffi cult to ascribe to individual projects.
The sub TRC represents each project’s individual
contribution to portfolio cost-effectiveness. This
level of detail augments general fi ndings of the IRP
process with individual customer data for continuous
program refi nement and target marketing.
3.2 IRP Methodology
and Analysis
The resources acquired in our current conservation
portfolio generally are not dispatchable and are
acquired in small quantities on a continuous basis.
Consequently, the aggregate level and type of
acquired conservation resources do not affect the
generating resources used to establish market
prices. Under these circumstances conservation
is a price-taker. In other words, lower or higher
acquisition levels are not expected to change overall
prices in the wholesale electricity marketplace.
Conservation resources were modeled
independently of supply-side resources in the IRP
due to the complexity and the relatively small size of
the conservation resources, because it is suffi cient
to acquire all cost-effective resources relative to the
IRP market price signal.
IRP market prices were used at a fi ner level of detail
for conservation planning than in the past.
3-5
A 20-year hourly avoided cost price signal was used
to determine the cost-effectiveness of individual
conservation measures and the aggregate level
of cost-effective conservation available within the
service territory. A ten percent adder was tiered on
all hours of the avoided cost to refl ect transmission
and distribution savings, and the risk reduction
values inherent in conservation resources.
Using a more detailed avoided cost required the
development of unique hourly load shapes for
each conservation measure. Load shapes were
developed through comparable engineering
simulations of base case and high-effi ciency
scenarios. Hourly load shapes allowed for an
evaluation of load-shifting opportunities. This was
not possible in past IRPs, since, for the most part,
load–shifting measures can increase overall kWh
usage as loads are shifted to off-peak periods.
Without hourly prices to value the shift, higher usage
did not appear cost-effective.
The initial survey of conservation inventory was
subdivided into an assortment of independent
measures. Potentially feasible measures were then
added to the list. Particular attention was paid to
residential measures, as they are an exception to
the all-inclusive conservation portfolio approach
and are not evaluated on a customer-by-customer
basis. Engineers and program planners involved
in this process were encouraged to err on being
overly inclusive in their evaluations of different
conservation measures.
The 2005 IRP exercise resulted in an initial defi nition
of 52, and the subsequent evaluation of 51,
conservation measures. The controlled voltage
reduction and rooftop air conditioning measures
were excluded from further consideration because
both measures are currently being piloted.3 Each
will be evaluated further when the pilots are
complete; results will be included in the 2007 IRP.
TRC inputs were collected for the remaining 51
measures, including customer cost, non-incentive
utility cost, non-energy benefi ts, natural gas impact,
electric energy savings and avoided cost. During
the initial iteration of the 51-measure package,
inputs for cost and benefi t characteristics were
reasonably close to those observed in the 2003
conservation program portfolio. Acquirable resource
potential therefore was indexed to 2003 levels. This
initial iteration provided a realistic baseline assessment
to compare against actual historical operations.
Subsequent iterations involved reassessment
of each measure and modifi cations to all inputs,
including acquirable potential, with the intent of
maximizing net TRC benefi ts. The measures were
defi ned assuming that each was independent;
however, it was necessary to perform a collective
assessment of non-incentive utility costs to ensure
that they were reasonably allocated across measures.
3 Insuffi cient results were available for evaluation of these two measures
because of delays in the completion of pilot studies for each respective
measure.
3-6
A “stacking” of the measures was completed for
each iteration of the 51-measure portfolio.
This stacking ensured reasonableness and consistency
in the overall analysis. Measures were stacked in
order of total resource benefi t-to-cost ratio. This
helped defi ne acceptable measures and determined
the shape of the IRP conservation supply curve.
3.3 Conservation
Measure Definitions
A brief description of each measure considered for
the IRP is presented below. The measures are
divided into three main categories: industrial measures,
commercial measures and residential measures.
Industrial Measures—
24,523 MWh Annual Potential
Industrial Refrigeration –
6,062 MWh Annual Potential
Cooling systems are used in a variety of processes
including food storage and preparation, ice making
and other large scale cooling requirements.
Savings potential includes tighter control of coolant
pressures and temperatures, the use of variable
frequency drives (VFD), operation of ancillary fans
and new control options.
Industrial Hydraulics – 667 MWh Annual Potential
Industrial hydraulics systems utilize high-pressure
fl uids for power transmission in a variety of
industries, including wood products, plastics and
mining. Hydraulic systems are used for precise
control and applications requiring high power
density, such as extruding, lifting or pressing.
Potential savings exist in a number of ways,
including better-part or no-load controls.
Industrial Pumps – 4,775 MWh Annual Potential
Industrial pumps refer to all processes designed
to move fl uids. This includes, but is not limited
to, process, irrigation, and heating, ventilation and
air conditioning (HVAC) applications. Savings
potential exists in tighter control of pressures
and fl ows, the use of VFDs for fl ow control and
optimized pump selection.
Industrial Fans and Blowers –
2,808 MWh Annual Potential
Industrial fan and blower applications denote all
processes that include the movement of a gas up to
about 30 pounds per square inch gauge (psig). This
includes, but is not limited to, a variety of industrial
processes, HVAC and conveying applications.
Potential savings exists in tighter pressure control
and fl ows, the use of VFDs for fl ow control and
system designs using high effi ciency fans and blowers.
Industrial Compressed Air –
8,711 MWh Annual Potential
Industrial compressed air refers to all processes
that include the movement of a gas above 30
psig. Savings potential exists in better-part or
no-load controls, the use of VFDs and high-
effi ciency compressors. Demand-side application
optimizations reduce actual consumption without
affecting system production.
Industrial Lighting – 1,500 MWh Annual Potential
Three industrial lighting measures were evaluated:
3-7
• Metal halide to T-5 fl uorescent lighting in
manufacturing facilities
• T-12 to T-8 fl uorescent lighting retrofi ts in
industrial facilities
• Metal halide to pulse start lighting in
manufacturing facilities
T-5 fl uorescent lamps are the basis for a new
generation of fl uorescent lighting products. The
smaller lamp diameter provides good optical control
and may be used in applications traditionally lit by
alternate systems, such as metal halide. The most
signifi cant barrier for T-5 systems is the initial cost
associated with replacing existing fi xtures. Utility
rebates help overcome the T-5 conversion cost barrier.
T-12 fl uorescent lighting is far less effi cient than
T-8 technology. Pulse start technology provides
improved light output from metal halide fi xtures
and longer lamp life. The measure is most cost
effective when existing metal halide lamps need to
be replaced for reasons other than energy effi ciency.
Commercial Measures –
15,641 MWh Annual Potential
Commercial conservation measures are performed
in or on commercial properties, including schools.
This group comprises the bulk of conservation
project potential. Commercial measures generally
require and utilize engineering resources because of
the sheer size and magnitude of this segment.
Commercial Lighting –
7,641 MWh Annual Potential
The incandescent light bulb is the least effi cient
form of electric lighting. It wastes most of the
energy it uses in the form of heat, increasing air
conditioning loads. Furthermore, the life of an
incandescent bulb is very short when compared to
a compact fl uorescent lamp (CFL). An equivalent
CFL can last an average of 10 times longer than its
incandescent counterpart. CFL measures generally
are implemented through prescriptive incentives.
There are many existing commercial buildings not
yet retrofi tted to T-8 technology. Incentives for
retrofi tting T-12 to T-8 lighting are offered primarily
through a prescriptive program. T-12 fl uorescent
lighting often is used in schools and there are many
opportunities to retrofi t T-8 fi xtures.
The different categories of commercial lighting
retrofi ts are identifi ed as follows:
Incandescent to Compact Fluorescent Lighting –
1,200 MWh Annual Potential
• CFLs in commercial buildings
• CFLs in schools
Metal Halide to Pulse Start Lighting –
1,100 MWh Annual Potential
• Metal halide to pulse start lighting in
commercial buildings
• Metal halide to pulse start lighting in gymnasiums
• Metal halide to pulse start lighting in parking lots
Metal Halide to Fluorescent Lighting Conversions –
800 MWh Annual Potential
• Metal halide to T-5 in commercial buildings
• Metal halide to T-5 in gymnasiums
3-8
Incremental Fluorescent Lighting Retrofi ts –
4,541 MWh Annual Potential
• T-12 to T-8 retrofi ts in convenience stores
• T-12 to T-8 retrofi ts in commercial buildings
• T-12 to T-8 retrofi ts in schools
Commercial Air Conditioning Measures –
2,500 MWh Annual Potential
Buildings that require mechanical cooling are
identifi ed in two different ways, skin load or internal
load facilities. High-effi ciency air conditioning
measures for both building types were evaluated
for the IRP. A skin load building is one that is highly
sensitive to environmental or weather conditions.
Internal processes operating within a structure
impact internal facility load. An internal load building
can require mechanical cooling year round if its
internal processes create waste heat.
One facility can have characteristics of both internal
and skin load structures, but when defi ning the
system being changed, one type generally is
predominant. A skin load building requires less air
conditioning when compared to an internal load
building, because it requires mechanical cooling only
when the outside environment is near to or hotter
than the building’s temperature set point.
Corporate Network Personal Computer Controls –
800 MWh Annual Potential
Present Information Systems (IS) require processing
actions to take place many times during the day
in present network systems. Employees are often
asked to leave their computer running after hours so
that software and security systems may be updated.
A personal computer (PC) consumes between
60 and 120 watts in standby mode, even when
the monitor is shut off. New network software-
hardware combinations allow IS to turn on and shut
off PCs during maintenance cycles, saving up to
12 hours of run time per night per PC. Individual
personal computer control options were combined
with corporate personal computer control
conservation options.
Building Exit Signs – 1,000 MWh Annual Potential
Exit signs are excellent targets for energy savings, as
they are illuminated 24 hours a day, 365 days a year.
Replacing existing exit signs with more effi cient
models generally is cost effective.
Variable Frequency Drives –
2,550 MWh Annual Potential
VFDs are used to control motors on fans and pumps
to optimize the fl ow of fl uid. Two fl uid types (liquid
and vapor) are used in these applications. Liquid
VFDs operate continually and generally have higher
savings than vapor VFDs.
Commercial High-Effi ciency Heat Pumps –
150 MWh Annual Potential
High-effi ciency air source heat pumps are cost-
effective only in areas without natural gas service.
Natural gas furnaces and heat pumps have similar
operating costs. As heat pumps have higher upfront
costs than gas systems, heat pumps are not cost-
effective where natural gas is available.
3-9
Non-Residential Appliance Effi ciency Measures –
200 MWh Annual Potential
Non-residential appliance effi ciency measures
include water heating, cooking, and refrigeration
end-uses. Restaurant and hospitality segments are
primary targets for these measures.
Non-Residential Shell Effi ciency Measures –
800 MWh Annual Potential
Shell measures increase building envelope
effi ciencies. Measures include insulation upgrades
and window replacements.
Rooftop HVAC Measures – Annual Potential
Currently Being Studied
The Company is piloting a rooftop maintenance
program in our Idaho service territory. Certifi ed
contractors are using the latest tools and technology
to diagnose and service problems in rooftop units.
Program cost effectiveness will be determined after
the pilot ends in December 2005.
Residential Measures –
10,632 MWh Annual Potential
Residential customers make up the largest group
in our system, but savings opportunities on a per-
customer basis are small. Therefore, it is necessary
to offer residential measure through prescriptive
programs. Prescriptive programs are calculated
using historical average unit savings and costs.
Incentives are provided based on the device
being replaced or retrofi tted. Customers send in
documentation to verify that they have installed the
measure prior to receiving an incentive.
Residential Compact Fluorescent Lamps –
3,600 MWh Annual Potential
Residential CFLs generally are offered through
point-of-purchase coupons, bulb giveaways and
manufacturing buy downs. In any case, replacing
incandescent bulbs with CFLs appears cost-
effective in a residential conservation portfolio.
Residential Shell Measures –
703 MWh Annual Potential
Residential shell measures include changes to the
building shell, HVAC systems or envelope, which
reduce energy use without affecting customer
comfort. Residential window measures were
evaluated on both a new and retrofi t basis. Many
of the measures in this segment use the R-Value
as a measurement. The thermal resistance
normally indicated in insulation as the R-Value
gives a higher value for more thermal resistance.
Residential shell measures include duct, wall, roof
and fl oor insulation.
A rebate of 75 cents per linear foot of R-10 insulation
presently is available for installing insulation on
heating ducts in unconditioned areas, such as attics
and crawlspaces. A 12 cents-per-square-foot
rebate is available for the addition of new insulation
that increases R-Value by R-10 or greater. Rebates
are available if existing insulation is less than R-22
in attics, R-11 in walls and R-11 in fl oors. Attic,
fl oor, and wall insulation must be installed only
where cavities separate areas that either have or do
not have air conditioning. Any insulation installed
outside the cavity, such as siding, does not meet
rebate requirements.
3-10
Residential Programmable Thermostat Programs –
659 MWh Annual Potential
Residential programmable thermostat programs
offer incentives to residents who control heating
with a set-back thermostat. Three residential
programmable thermostat measures were evaluated
for the IRP: electric resistance heating, heat pumps
and air conditioning. The Company used to offer
a rebate of up to $40 to homeowners replacing
their manual thermostats with an approved
programmable thermostat. The program has been
reevaluated for the IRP.
Residential HVAC Effi ciency Measures –
3,889 MWh Annual Potential
This group of residential effi ciency measures
includes high-effi ciency air conditioning, electric-to-
natural gas space heat conversion in ducted homes,
electric-to-natural gas space heat conversion in
non-ducted homes and heat pumps.
A rebate offering could be developed for
homeowners who install an air conditioner with 12.0
SEER (cooling effi ciency) or greater. We will evaluate
whether to offer an incentive to new construction
customers, retrofi t customers or both. A $200
rebate is currently available to homeowners who
replace primary electric heat (forced air furnace or
baseboard heat) with a central natural gas heating
system. A $100 rebate is available to replace
electric heat with a natural gas wall heater. This
rebate can be claimed in addition to the $150 high-
effi cient natural gas furnace rebate. A $300 rebate
is available to homeowners whose primary heating
source is electric heat and who install an air-source
heat pump of 8.0 HSPF (heating effi ciency) with 13.0
SEER or greater. Homeowners are eligible at the
7.5 HSPF and 12.0 SEER levels for manufactured
homes. Replacement of an existing heat pump
qualifi es for a $50 rebate.
Residential Water Heating Measures –
1,475 MWh Annual Potential
Three residential water-heating measures were
evaluated for this study. The measures included
water heating appliance effi ciency, electric-to-natural
gas water heating conversion, heat pump water
heaters and water heating blankets.
These measures are designed to upgrade existing
water heaters to more effi cient units or to improve
the effi ciency of an existing water heater by adding
additional insulation. A $50 rebate is currently
available to install tank-type electric water heaters
that are at least 0.91 effi ciency (EF) or to tank-type
natural gas water heaters that are at least 0.62 EF for
40-gallon and at least 0.60 EF for 50-gallon units. A
rebate of $60 is available to electric customers who
replace an electric water heater with a new tank-
type natural gas water heater. The $60 rebate can
be claimed in addition to the $50 high-effi cient water
heater rebate. A rebate offering could be developed
to provide an incentive for increasing exterior
insulation of water heater tanks.
Residential Windows – 305 MWh Annual Potential
Residential windows initially were evaluated based
on the direction they were installed: north, south,
3-11
east or west. The categories ultimately were
combined because of an inability to adequately
distinguish the difference between them. A rebate
could be developed for the addition of energy
effi cient windows installation with increased U-Value.
The U-value is the measure of thermal conductivity.
A higher value means a material is more thermally
conductive. For example, a lost opportunity
is targeting new construction with incentives
encouraging installation of windows with U-values
above current building code. Bringing older
windows up to current standards would also provide
energy savings and signifi cant non-energy benefi ts.
Distribution Measures Impacting
Customer End-Use Effi ciency
Controlled Voltage Regulation (CVR) –
Annual Potential Currently Being Studied
CVR incorporates a variety of measures that may be
physically located on the customer or utility side of
the meter to control end-use voltage.
Maintaining voltage levels closer to the appropriate
levels for end-use equipment generally improves
effi ciency and increases equipment life. Avista is
participating in a regional market transformation
venture, incorporating 17 pilot sites and several
alternative technologies, to determine the cost-
effectiveness, non-energy impact, total energy
savings and the load shape of savings under
various circumstances. All of this information is
highly dependent on the end-use mix and utility
distribution characteristics. At this time there is
insuffi cient data to characterize CVR for evaluation
in the IRP process.
3.4 Evaluation of Measures
Each measure was evaluated based on
characteristics relevant to total resource cost
analysis. A description of these characteristics, and
the approach used to quantify the inputs, is briefl y
described below.
Measure Load Shape
Measure load shapes are engineering calculations of
the shape of effi ciency measure savings. Generally,
savings shapes mimic the end-use load shape.
Exceptions, such as heat pumps and programmable
thermostats, were modeled to only include energy
savings. Industrial measure load shapes benefi ted
from actual metering data acquired from various
industrial end-use projects. The load shapes are
characterized as 8,760-hour but are often of a
repetitive nature (e.g., similar weekday or weekend
shapes repeated throughout the year).
Non-Energy Benefi ts
The fi rst iteration of non-energy benefi ts (NEB) for
each measure was based on the 2003 historical
non-energy benefi ts per kilowatt-hour (kWh),
disaggregated by customer segment and measure
type based on the External Energy Effi ciency (Triple-
E) board-reporting format. The measures defi ned
for the IRP analyses were not necessarily consistent
with those used in past Triple-E board reports, so
it was necessary to modify these in later iterations.
Avista traditionally reports only quantifi able NEB for
purposes of providing external cost-effectiveness
3-12
analysis of past program activity. This primarily
consists of maintenance savings, reduction in usage
of other inputs to the production process, and other
quantifi able benefi ts. Other NEB that may not have
been observed in the past or suitable for inclusion in
Triple-E board analysis were included to the extent
that they were appropriate for individual measures.
The technology for several measures has been
changing so rapidly that it is necessary to modify
even recent calculations to refl ect the nature of the
current and near-future market.
Natural Gas Impact
Several of the evaluated electric effi ciency measures
impact natural gas usage. This could result in
increased or decreased natural gas usage. Natural
gas impacts were quantifi ed and incorporated into
the analysis of applicable measures. The seasonal
nature of the natural gas impact, either “annual”
or “winter,” was characterized by measure, and a
natural gas avoided cost forecast was applied over
the estimated life of the measure.
Customer Cost
Customer cost has been at least 75 percent of the
total resource cost of Avista’s historical conservation
portfolio. The incremental cost over the appropriate
baseline scenario was quantifi ed for each measure.
The assumption of base case and high-effi ciency
scenarios was consistent for the calculation of
customer costs and energy savings.
Non-Incentive Utility Cost
Non-incentive utility costs incorporate labor and
non-incentive expenses associated with utility
acquisition programs. Direct customer incentives
are not incorporated in this calculation. Initial
iterations applied historic average non-incentive
utility costs to each measure. As programs were
optimized over subsequent iterations, costs were
changed to recognize program design revisions.
Measure Life
Measure life represents the life of the energy savings
inherent in the defi ned measure. For the most
part, the measure life is equal to the shorter of the
physical or economic life of the end-use equipment.
Utility Incentive Cost
Utility incentive cost is not part of the total resource
test, but incentive level and structure assumptions
were incorporated into alternative program designs
to create a complete program. This was necessary
to provide a basis for an informed estimate of
energy savings. Incentive assumptions were not
necessarily limited to a particular tariff structure,
but Avista’s current Idaho Schedule 90 and fi led
Washington Schedule 90 incentive structures were
used as a guide. Incentives were not permitted to
exceed 100 percent of measure cost, and in most
cases customer direct incentives of 40 to 50 percent
were deemed to be adequate.
Energy Savings
Based on inherent measure characteristics and
program design developed per iteration, an estimate
of annual energy acquisition for each measure was
developed. Generally speaking, annual acquisition
levels were considered to be a reasonable estimate
for a fi ve-year period.
3-13
Based on these measure categories and
characterizations, a TRC analysis was performed on
each measure. In addition to traditional cost-benefi t
analysis two different calculations of TRC levelized
cost were performed. The fi rst calculation applied
the customer and non-incentive utility cost, measure
life, discount rate and annual energy savings.
This calculation excludes the benefi t (or cost) of
non-energy benefi ts and the impact on natural
gas usage from the calculation, because these are
not considered costs for purposes of the cost-
benefi t analysis. An alternative calculation of the
TRC levelized cost treats non-energy benefi ts and
natural gas impact as offsets (or additions to) the
TRC cost of the measure. The latter, more inclusive
TRC levelized cost calculation, is more suitable for
evaluating the total resource value of the measure in
almost any circumstance.
3.5 Results of the Analysis
The fi nal evaluation accepted 36 measures as cost-
effective, which resulted in 5.5 aMW of aggregate local
conservation acquisition. This excludes acquisition
attributed to Avista through participation the Northwest
Energy Effi ciency Alliance effi ciency programs. The
total energy acquisition evaluated for all programs
(including non-cost effective programs) ranged from
4.1 to 7.0 aMW. Tables 3.1 through 3.6 summarize
the results of the analysis of individual measures.
Ranking measures by cost-benefi t ratio is related,
but not identical, to ranking the same measures
by TRC levelized cost. This is due to the inclusion
of the value of alternative load shapes in the
cost-benefi t analysis; it is not considered in the
calculation of the TRC levelized cost. For example,
a measure with a TRC levelized cost of $37 per
MWh may actually have a more favorable cost-
benefi t ratio than another measure costing $35 per
MWh. This would happen if the energy savings of
the higher-cost measure occurred during relatively
higher-value periods of the year. For these reasons,
the cost-benefi t ratio is a superior means of ranking
measures, but it is also true that load shapes are
generally not different, nor the hourly avoided cost
differentials so extreme, to result in a signifi cant
difference in the ranking of the measures.
Seven measures have a negative total resource
cost, as non-energy benefi ts fully offset customer
and utility costs. These measures include all three
compact fl uorescent lighting measures and four
industrial measures.
Figure 3.3 is a graphical representation of the supply
curve “stacked” in descending order of cost-benefi t
ratio. The descending order of this ratio, with the
most cost-effective measure to the left, results in
an untraditional downward sloping supply curve.
Measures where the total resource costs were
less than zero are not represented as points on
this curve, but the savings are incorporated into
the acquisition potential. A negative incremental
replacement cost will create values less than zero.
Figure 3.4 is a graphical representation of measures
with cost-benefi t ratios below 10. This view provides
more detail on the majority of evaluated measures.
3-14
The TRC levelized cost of these measures, sorted in
descending order, is represented in Figure 3.5. The
aberrations in this TRC levelized cost supply curve
are the result of the distinction between the rankings
of measures by cost-benefi t ratio vs. ranking by TRC
levelized cost previously mentioned.
Figure 3.6 represents TRC levelized cost, excluding
residential window and non-residential shell
measures. The fi gure allows for a more detailed
scale of the majority of the measures.
3.6 Review of the Results
The 5.5 aMW (47,500,000 fi rst year kWh), identifi ed
as cost-effective and appropriate for local
acquisition, represents a 19 percent increase above
Avista’s current Schedule 90 tariff goal. Additionally,
Avista has 1.4 aMW of attributed resource
acquisition based on participation in regional energy-
effi ciency ventures through the Northwest Energy
Effi ciency Alliance. This avoids double counting by
attributing all effi ciency measures, participated in
by local utility programs, entirely to the local utility.
A residential compact fl uorescent program is not
currently offered by Avista to any signifi cant extent
but is currently offered as a regional program.
Figure 3.7 describes three goals: Avista’s 2003
current tariff goal labeled “Current Tariff,” an
extrapolation of Avista’s share of the NPCC goal
labeled “NPCC,” and the aggregation of the cost-
effective potential for our local acquisition program,
the overlapping adoption of previously regional
programs into a local utility program (residential
CFLs) and additional regionally-acquired energy
Measure
Savings
(MWh)
Measure
Life
(Years)
Electric
Avoided
Cost
($000s)
Non-
Energy
Benefi ts
($000s)
Gas
Avoided
Cost
($000s)
Non-
Incentive
Utility Cost
(000s)
Customer
Cost
($000s)
Hydraulics 667 15 261 64 0 33 20
Fans Blowers 2,808 15 1,101 270 0 140 86
Pumps 4,775 15 1,867 459 0 239 146
Refrigeration 6,062 15 2,364 583 0 303 185
Compressed Air 8,711 15 3,411 285 0 436 500
T12-T8 Fluor. 500 12 182 75 -7 10 160
MH to T5 Fluor. 500 15 207 75 -8 10 185
MH to PS Fluor. 500 15 207 75 -8 10 200
Total 24,523 9,601 1,887 -24 1,181 1,483
Table 3.1: Summary of Individual Industrial Measures
4 This fi gure is based on Avista being 4.0 percent of the regional
end-use load.
3-15
Table 3.2: Summary of Individual Commercial Measures
Measure
Savings
(MWh)
Measure
Life
(Years)
Electric
Avoided
Cost
($000s)
Non-
Energy
Benefi ts
($000s)
Gas
Avoided
Cost
($000s)
Non-
Incentive
Utility Cost
(000s)
Customer
Cost
($000s)
School CFL 200 10 66 30 -3 4 0
Commercial CFL 1,000 7 253 150 -10 20 30
A/C, Internal Load 1,455 15 597 154 0 29 131
Avista Network Comp 800 20 339 0 0 16 8
Exit Signs 1,000 12 339 150 -15 20 170
T12-T8 Conv. Retail 2,000 12 679 300 -29 40 400
VF Drives, Liquid 1,050 20 478 0 0 21 168
MH to PS Fluor. 500 15 208 75 -8 10 145
MH to T5 Fluor. 500 15 208 75 -8 10 145
Heat Pumps 150 15 63 16 -3 3 44
VF Drives, Vapor 1,500 20 683 0 0 30 360
T12-T8 Fluorescents 2,041 12 756 306 -30 41 714
A/C, Skin Load 1,045 15 424 111 -18 21 355
MH to PS Park Lots 300 15 106 45 0 6 144
MH to T5 Gyms 300 15 127 45 -5 6 165
Appliances 200 20 94 40 -11 4 128
MH to PS Gyms 300 15 127 0 -5 6 165
T12-T8 Schools 500 12 188 75 -7 10 350
Shell 800 25 403 64 0 16 7,856
Total 15,641 6,137 1,637 -152 313 11,478
extrapolated from 2004 activity4 labeled “IRP.”
The distribution of the 39 cost-effective measures
is approximately 50 percent industrial, 30 percent
commercial and 20 percent residential. The
plan is signifi cantly more reliant on industrial
acquisition than in the past. Commercial acquisition
has decreased as a share of the total but is
approximately equal to recent acquisition levels
on an energy basis. Residential acquisition is not
signifi cantly revised, except by the addition of the
residential CFL program. Figure 3.8 represents the
distribution of energy saving by customer segment.
This distribution of energy savings into more detailed
categorizations by segment is represented in Figures
3.9 through 3.11.
3-16
Measure
Savings
(MWh)
Measure
Life
(Years)
Electric
Avoided
Cost
($000s)
Non-
Energy
Benefi ts
($000s)
Gas
Avoided
Cost
($000s)
Non-
Incentive
Utility Cost
(000s)
Customer
Cost
($000s)
CF Lighting 3,600 10 1,215 549 -46 62 288
Duct Insulation 285 25 144 0 0 6 31
Roof Insulation 108 25 55 0 0 2 18
Water Htr Blanket 121 12 41 0 0 2 17
Wall Insulation 158 25 79 0 0 3 46
W/H Elec-Gas Conv. 606 12 212 0 -84 12 73
Prog Ts, Elec Resist. 295 20 109 0 0 6 89
Air Conditioning 353 0 147 0 0 7 120
FAE-G Conv. Ducted 2,606 0 1,264 0 -567 52 521
Prog Ts, Heat Pump 198 20 74 0 0 4 69
Res Heat Pump 470 15 196 0 0 5 207
Floor Insulation 128 25 64 0 0 3 68
FAE-G Conv. No Duct 460 0 223 0 -100 9 170
W/H Appliance Eff 485 12 170 0 0 10 310
Prog Ts, Air Cond 167 20 66 0 0 3 135
East Windows, retro 89 12 30 0 0 2 311
West Windows, retro 98 12 33 0 0 2 346
South Windows, retro 49 12 17 0 0 1 212
North Windows, retro 69 12 23 0 0 1 677
East Windows, new 8 12 3 0 0 0 1
West Windows, new 8 12 3 0 0 0 1
South Windows, new 6 12 2 0 0 0 1
North Windows, new 3 12 1 0 0 0 1
Heat Pump Water
Heaters 263 12 89 0 0 5 121
Total 10,633 4,260 549 -749 197 3,832
Table 3.3: Summary of Individual Residential Measures
3-17
Table 3.4: TRC Costs and Benefi ts for Industrial Measures
Measure
TRC AC
Benefi ts
($000s)
TRC Net of Gas AC
and NEB Benefi ts
($000s)
Net TRC
Benefi ts
($000s)
TRC Benefi t
to Cost Ratio
TRC Levelized
Cost ($/MWh)
Hydraulics 287 -10 298 Infi nite -2.0
Fans Blowers 1,211 -44 1,255 Infi nite -2.0
Pumps 2,053 -75 2,128 Infi nite -2.0
Refrigeration 2,601 -95 2,695 Infi nite -2.0
Compressed Air 3,752 651 3,101 5.76 9.0
T12-T8 Fluor. 200 102 98 1.96 28.0
MH to T5 Fluor. 228 128 100 1.78 31.0
MH to PS Fluor. 228 143 85 1.59 35.0
Total 10,561 801 9,760
Table 3.5: TRC Costs and Benefi ts for Commercial Measures
Measure
TRC AC
Benefi ts
($000s)
TRC Net of Gas
AC and NEB
Benefi ts ($000s)
Net TRC
Benefi ts
($000s)
TRC Benefi t
to Cost Ratio
TRC Levelized
Cost ($/MWh)
School CFL 73 -15 89 Infi nite -12.0
Commercial CFL 279 -90 369 Infi nite -18.0
HE A/C, internal load buildings 657 6 651 110.72 0.0
Network computer 373 24 349 15.54 3.0
Exit signs 373 54 319 6.85 7.0
T12-T8 convenience retail 746 169 578 4.42 12.0
VFD, liquid 526 189 337 2.78 19.0
MH to PS, commercial 228 88 140 2.58 21.0
MH to T5, commercial 228 88 140 2.58 21.0
HE heat pumps 69 34 36 2.07 27.0
VFD, vapor 751 390 361 1.93 28.0
T12-T8 commercial 831 479 353 1.74 32.0
HE A/C, skin load buildings 466 284 182 1.64 33.0
MH to PS, parking lots 117 105 12 1.11 42.0
MH to T5, gyms 139 131 8 1.06 53.0
Non residential appliances 103 103 1 1.01 54.0
MH to PS, gyms 139 176 -37 0.79 71.0
T12-T8 schools 207 292 -85 0.71 80.0
Non residential shell 443 7,808 -7,365 0.06 956.0
Total 6,750 10,314 -3,564
3-18
Table 3.6: TRC Costs and Benefi ts for Residential Measures
Measure
TRC AC
Benefi ts
($000s)
TRC Net of Gas AC
and NEB Benefi ts
($000s)
Net TRC
Benefi ts
($000s)
TRC Benefi t
to Cost Ratio
TRC
Levelized
Cost ($/MWh)
CF Lighting 1,336 -153 1,489 Infi nite -6.0
Duct Insulation 158 37 121 4.26 13.0
Roof Insulation 60 21 39 2.91 19.0
Water Htr Blanket 45 19 26 2.33 22.0
Wall Insulation 87 49 38 1.79 30.0
W/H Elec-Gas Conv. 234 169 65 1.39 38.0
Prog Ts, Elec Resist. 120 94 26 1.27 34.0
Air Conditioning 162 127 34 1.27 43.0
FAE-G Conv. Ducted 1,391 1,140 251 1.22 46.0
Prog Ts, Heat Pump 81 73 8 1.11 39.0
Res Heat Pump 215 212 4 1.02 54.0
Floor Insulation 71 70 0 1.01 54.0
FAE-G Conv. No Duct 245 279 -34 0.88 64.0
W/H Appliance Eff 187 320 -133 0.58 90.0
Prog Ts, Air Cond 72 138 -66 0.52 88.0
East Windows 33 313 -280 0.11 481.0
West Windows 37 347 -311 0.11 481.0
South Windows 18 213 -195 0.09 590.0
North Windows 26 678 -652 0.04 1,342.0
East Windows, new 3 2 1 1.91 27.0
West Windows, new 3 2 1 1.91 27.0
South Windows, new 2 1 1 1.58 32.0
North Windows, new 98 126 -28 0.78 65.0
Heat Pump Water Heaters 1 1 0 0.73 70.0
Total 4,579 4,147 431
3-19
Figure 3.3: Conservation Supply Curve Stacked by Levelized TRC Cost ($)
-0.20
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
1.60
- 1 2 3 4 5 6
aMW
Figure 3.4: Conservation Supply Curve Stacked by Levelized TRC Cost <0.10 ($)
(0.04)
(0.02)
-
0.02
0.04
0.06
0.08
0.10
- 1 2 3 4 5 6
aMW
3-20
Figure 3.5: Conservation Supply Curve (TRC B/C Ratios)
(10)
10
30
50
70
90
110
130
150
2 3 4 5 6
aMW
Figure 3.6 Conservation Supply Curve (TRC B/C Ratios < 10.0)
-
1
2
3
4
5
6
7
8
2 3 4 5 6
aMW
3-21
Figure 3.7: Aggregate Conservation Goal Comparison (aMW)
Residential
Commercial
Industrial
Figure 3.8: Customer Segment Savings Distribution
3-22
Figure 3.9: Industrial Segment Savings Distribution
Residential
Commercial
Pumps
Fans/blowers
Non-process
Refrigeration
Hydraulics
Compressed air
Industrial
Figure 3.10: Commercial Segment Savings Distribution
Residential
Lighting
HVAC
Motors
Controls
Appliances
Industrial
Commercial
3-23
Figure 3.11: Residential Segment Savings Distribution
Industrial
Commercial
Lighting
HVAC
Water Heater
Shell
Prog Thermostat
Residential
3-24
3.7 Conservation
Business Planning
Avista has recently assumed that the 2005 IRP
would identify a 10 to 25 percent increase in cost-
effective conservation potential. In late 2003, the
Company began ramping up conservation programs
to coincide with the Idaho electric tariff riders
reaching a zero balance. It is anticipated that the
aggregate tariff rider balance will reach zero in 2005.
Once the balance reaches zero, the Company will
transition to a long-term business plan structured
toward acquiring all cost-effective conservation
potential available through local programs.
As part of the 2004 ramp-up process the Company
piloted several alternative implementation
approaches intended to enhance cost-effective
acquisition. Based on an analysis of the
conservation pilot projects, a review of existing
Avista implementation efforts and conservation
contracts acquired under the 2000 All-Resource
Request For Proposals, it was determined current
incentive levels are insuffi cient to meet future
conservation acquisition goals.
The Company requested revisions to Idaho
Schedule 90 in early 2005 to approximately double
the incentive levels offered in Idaho. The revised
schedule became effective in March 2005. A similar
fi ling has been made in Washington to become
effective July 2005. The Company anticipates
annual revisions of the tariff rider funding mechanism
to provide adequate funding for future programs
and to recover any individual tariff rider balances,
positive or negative, carried into a calendar year.
The Company has increased staffi ng in 2004 and
will continue to evaluate additional staff in 2005 and
beyond. The results of the IRP, and in particular the
identifi cation of signifi cant increases in cost-effective
industrial conservation potential, will play a key role
in the development of infrastructure that is capable
of delivering our new conservation goals.
Avista will continue to work with regional entities,
and in particular the Northwest Energy Effi ciency
Alliance, to acquire cost-effective conservation
resources. This is likely to play its greatest role in
the acquisition of residential resources. Based on
a review of historical Northwest Energy Effi ciency
Alliance venture success there is a strong indication
that residential programs are typically more cost-
effectively acquired through a combined local utility
and regional market transformation approach.
4-1
Comprehensive coordination of transmission
system operations and planning activities among
the region’s transmission providers is necessary
to maintain reliable and economical transmission
service and to integrate the output of generation
resources to serve the region’s end-use customers.
Regional transmission providers and interested
stakeholders are working toward implementing
changes in the region’s approach to planning,
constructing and operating the regional transmission
system under new rules promulgated by the Federal
Energy Regulatory Commission (FERC), and under
state and local siting.
This section was developed in full compliance with
Avista’s FERC Standards of Conduct governing
communications between Avista Utilities Merchant
and Transmission functions.
4.1 Avista 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, Montana,
and Townsend, Montana. The transmission system
includes switching stations and high-voltage
substations with transformers, monitoring and
metering devices, and other equipment related to
the operation of the system. It is used to transfer
power from the Company’s generation resources
to its retail load centers. The Company also has
network interconnections:
• Bonneville Power Administration (BPA)
• Idaho Power Company
Section Highlights
 Avista has over 2,200 miles of high voltage transmission.
 The Company is involved in many regional transmission organizations and studies.
 Regional transmission groups, Grid West and the Transmission Improvement Group (TIG) are
continuing development.
 New transmission construction costs associated with the integration of new generation projects
can vary greatly, ranging from $10 million to $1.5 billion depending on location and project size.
 New transmission upgrade costs are included in the Preferred Resource Strategy.
4. TRANSMISSION PLANNING
4-2
• Northwestern Energy
• Pacifi Corp
• Puget Sound Energy
• Chelan County PUD
• Grant County PUD
• Pend Oreille County PUD
In addition to providing enhanced reliability in
the operation of the transmission system, these
network interconnections serve as points of receipt
of power from generating facilities outside the
Company’s service area, including the Colstrip
generating station, Coyote Springs 2 and Mid-
Columbia hydroelectric generating facilities. These
interconnections provide for the interchange of
power with entities within and outside the Pacifi c
Northwest, including the integration of long-term
and short-term contract resources. Additionally,
the Company has a number of interconnections
with government-owned or cooperative utilities
at transmission and distribution voltage levels,
representing non-network, radial points of delivery
for service to wholesale loads.
Avista is in the process of implementing a
transmission upgrade plan to add over 100 circuit
miles of new 230 kV transmission line to its system
and will later increase the capacity of another 50
miles. Avista is also constructing two new 230 kV
substations and is reconstructing three existing
transmission substations. Related projects at six
230 kV substations are necessary to meet capacity
requirements, upgrade protective relaying systems,
and to meet regional and national reliability standards.
In total, Avista will perform work in 11 of its 230 kV
substations or 85 percent of its system. The most
signifi cant projects are described below.
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
Avista constructed a new 230 kV substation near
Clarkston, Washington, that enables existing
transmission lines to form a 35-mile transmission
“ring” around the Lewiston, Idaho, and Clarkston,
Washington, areas. The project serves load and
improves reliability by reducing congestion during
peak energy fl ows.
Spokane Valley Reinforcement
Avista is adding 500 million voltamps (MVA) of
230 kV to 115 kV transformation at the new Boulder
Substation.
Pinecreek Substation
The Company recently completed the reconstruction
of this 230 kV facility located in Pinehurst, Idaho.
Palouse Reinforcement
The Company plans to construct 60 miles of 230
kV transmission line between the Benewah and
Shawnee substations to relieve congestion on
the existing Benewah-Moscow 230 kV line and
to provide an alternative source of power to the
Shawnee Substation.
4-3
Beacon-Bell 230 kV
The Company is increasing the capacity of two
parallel path transmission lines from its Beacon
substation to BPA’s Bell substation.
The overall cost of the above-mentioned transmission
projects is estimated at over $100 million.
As set forth in an August 2002 agreement with BPA
known as the West of Hatwai letter agreement, these
projects are coordinated with the federal entity.
Company upgrades support and enhance BPA
transmission projects. By working together, both
parties have achieved a least-cost service plan that
addresses commercial transactions, load service
and regional reliability issues.
This Avista and BPA plan was reviewed by
peer utilities and 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 Company and BPA stand-alone plans
separately executed.
The Company plans and operates its transmission
system pursuant to applicable criteria established
by the North American Electric Reliability Council,
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 or revised criteria and
coordinates the planning and operation of its
transmission system with neighboring transmission
systems. The Company is subject to periodic
performance audits through participation in these
regional organizations.
Portions of the Company transmission system are
fully subscribed for the purpose of transferring the
power output of 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 by the Company necessary to optimize its
resource portfolio, 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 in specifi c cases
can occasionally restrict our ability to optimize our
system resources. 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
of FERC policies making transmission capacity
available to other parties. Furthermore, to the extent
a third party has secured fi rm capacity rights on
Avista’s transmission system, including future roll-
over rights, that transmission capacity will not be
available for Company use to serve native load.
4-4
4.2 Regional
Transmission System
BPA operates more than 15,000 miles of
transmission facilities throughout the Pacifi c
Northwest. BPA’s system represents approximately
75 percent of the region’s 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 and its 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 nine of its remote retail load areas.
The Company participates in a number of regional
and BPA-specifi c forums to coordinate system
reliability issues and planning issues, and to manage
costs associated with the BPA transmission system.
NWPP forums include the following work groups:
the Transmission Planning Committee provides
coordinated analysis of proposed transmission
projects in the Northwest sub-region and resolves
technical transmission planning issues; the
Northwest Transmission Assessment Committee
reviews transmission needs in a broad sense,
performing studies and developing cost estimates
for future resource development alternatives; and the
Northwest Operations and Planning Study Group
reviews near-term seasonal operating capacity on
constrained portions of the Northwest grid.
The Company also participates in BPA transmission
and power rate case processes, and in BPA’s
Business Practices Technical Forum, to ensure BPA
transmission charges remain reasonable and that
they support system reliability and access.
The Company also works with BPA and other
regional utilities to coordinate major transmission
facility outages.
4.3 Regional
Transmission Issues
While coordinated transmission planning takes place
through various NWPP workgroups, process
improvements can further increase responsiveness
and timeliness of major regional transmission project
decisions. A more formalized organization is under
consideration in the Northwest to develop a regional
transmission plan, assess transmission alternatives
(including non-wires alternatives) and provide a
forum for decision-making for new projects and cost
allocation methods.
Future regional resource development will require
new transmission assets. BPA has indicated
that fi nancing restrictions may hamper its ability
to construct new transmission to support these
resources. BPA transmission customers seeking
fi rm capacity for their new resources may be
required to provide what is essentially long-term
fi nancing for BPA in order to facilitate needed
transmission project construction on its system.
The formation of a regional transmission
organization (RTO) to address the transmission
4-5
issues discussed above has been studied for some
time. State and/or federal jurisdiction over such a
regional transmission organization has also been the
subject of much debate. Accordingly, at the end of
September 2005, regional parties are slated to make
a determination as to whether to move forward
with either of two alternatives to address a number
of regional transmission issues: Grid West or the
Transmission Improvements Group (TIG) proposal.
Grid West
FERC Order 2000 requires all jurisdictional utilities
either to fi le a proposal to form an RTO, or a
description of efforts to participate in an RTO, or a
list of any existing obstacles to RTO participation.
FERC Order 2000 is a follow-up to FERC Orders
888 and 889 issued in 1996. It requires transmission
owners to provide non-discriminatory transmission
service to third parties.
The Company participated in a negotiation process
with nine Western state utilities, incorporating the
involvement of a broad spectrum of additional
regional stakeholders, on the possible formation of
“RTO West,” a non-profi t organization. The utilities
and regional stakeholders have since shifted to an
approach intended to respond to identifi ed problems
and ineffi ciencies in how the region’s integrated
transmission grid is managed, as opposed to
attempting to develop an RTO that is fully compliant
with specifi ed functions and characteristics outlined
by FERC. This revised process has resulted in
the adoption, on December 9, 2004, of interim
bylaws governing continuing developmental
activities for this non-profi t corporation under the
new name Grid West.
Building on earlier RTO development work,
regional stakeholders participating in the Grid
West process identifi ed a number of transmission-
related “problems and opportunities” that need
to be addressed. Among these are current rules
and practices that prevent full utilization of the
transmission infrastructure and impede the ability
to facilitate more effi cient, region-wide transactions.
Congestion management by curtailment was viewed
as problematic. Additionally, diffi culties in effi ciently
and effectively planning and constructing needed
transmission infrastructure in the region were
identifi ed, and the lack of an independent market
monitor was raised as an issue.
The Grid West proposal seeks to improve
transmission services and infrastructure development
through the establishment of a new, non-profi t
corporation with board membership independent
of any specifi c electric wholesale or retail market
interest. The Grid West proposal intends to
1. Implement a system to manage and offer
transmission rights to attain greater utilization of
the transmission grid while preserving existing
transmission rights;
2. Provide voluntary consolidation of control area
operations to create organized market
structures for the provision of ancillary services;
3. Implement a regional transmission system
planning process and provide for backstop
authority to resolve issues regarding;
4-6
fi nancing, cost allocation and construction of
new transmission facilities;
4. Provide a market monitoring function.
By the end of 2005, participants in the development
of the Grid West proposal are expected to determine
if Grid West will hold elections to seat the independent
board and move forward with further developmental
activities in preparation for reaching operational status.
Transmission Improvements Group
In its review of whether or not to move forward with
Grid West, the Company recognizes the prudence
in assessing other alternatives to address regional
transmission issues. Other regions of the U.S. that
have implemented RTO structures have experienced
signifi cant costs associated with such organizations.
Many regional stakeholders are skeptical as to
whether implementation of the Grid West proposal
will ultimately provide meaningful and sustainable
net benefi ts to customers. Several regional parties
explored how regional transmission issues might
be addressed using a coordination contract model
and relying upon the enhancement of existing
organizational structures to mitigate some of the
jurisdictional and cost control concerns associated
with broader RTO structures, specifi cally Grid West.
In March 2005, a group of regional stakeholders,
TIG, agreed to fund the development of proposals
for improving the planning, operation and oversight
of the Northwest transmission system.1
TIG intends to identify effective, low-cost solutions
to known transmission issues within the general
geographic area covered by the NWPP. TIG
participants plan to make immediate, substantive,
incremental steps to improve access to, and the
effi ciency of, the region’s transmission system.
The TIG approach intends to address the same
transmission-related “problems and opportunities”
outlined in the GridWest process. TIG is focusing on
fi ve areas of development:
1. A common region-wide Open Access
Same-time Information System (OASIS) to
manage access to the systems of all regional
transmission providers;
2. A regional transmission planning and expansion
model for coordinated planning and the authority
to resolve decisions regarding what new
transmission facilities are be constructed,
who should fi nance and construct these facilities,
and to whom such costs should be allocated;
3. Enhanced reliability and security functions,
including broader functionality of the Pacifi c
Northwest Security Coordinator and providing
for the voluntary consolidation of certain control
area operations functions;
4. Region-wide implementation of a fl ow-based
determination of available transmission
capacity;
5. The implementation of a market monitoring
function.
Parties developing the TIG proposal have
established work groups to address these fi ve areas.
The work groups hope to develop their proposals in
1 TIG participants include Avista, BPA, Chelan County PUD, Clark County
PUD, Cowlitz County PUD, Douglas County PUD, Grant County PUD,
Portland General Electric Company, Power Resource Managers, Public
Power Council, Puget Sound Energy, City of Seattle, Tacoma Power and
the Washington PUD Association.
4-7
suffi cient detail to allow for a reasonable comparison
between the TIG and Grid West proposals by the
end of 2005. To the extent possible, approaches
developed by the TIG work groups will utilize
existing organizations and contracts and avoid
creating new institutions.
4.4 Modeling Transmission
Costs in the Integrated
Resource Plan
Transmission costs to integrate new resources into
the Company’s system were estimated by Avista’s
Transmission Department. Estimates were not
modeled in AURORAXMP, but rather in the proprietary
LP model that matches resources with Avista’s
resource requirements. A rigorous study has not
been completed for any of these transmission
alternatives; estimates are engineering judgment
only and are not “construction estimate” quality. As
the size of the resource increases, the certainty of
the estimates diminishes. A 50 MW resource can
be integrated in many places on Avista’s (or another)
system. A 350 MW plant can be integrated at some
locations, while a 750 MW plant has very limited
placement options. At the 1,000 MW plant level,
a generic integration cost of $1.5 billion has been
assigned because of the uncertainty of impacts
to the Company’s system and/or the neighboring
systems. A detailed regional process likely would
be undertaken to determine the precise impacts and
integration costs before an actual plant placement
decision would be made.
Table 4.1 describes the location for potential
resources, capacity, required upgrades, and the
cost of the upgrade for the requested locations.
Transmission costs are allocated on a per-kilowatt
basis. For example, if Avista purchased half of a 750
MW plant with an estimated transmission expense
of $400 million, the portion allocated to Avista would
be $200 million.
In summary, there are a number of issues and
uncertainties regarding future expansion of the
Northwest transmission system to accommodate
the integration of future resources needed to serve
the region’s load growth. Among these are the
following:
1) The Northwest transmission system is fully
subscribed in many areas with scarce fi rm
transmission capacity to accommodate the
integration of new large-scale resources;
2) Current FERC policies and practices restrict
the fl exible use of transmission assets to
facilitate resource portfolio optimization in
hydro-based systems;
3) There is no comprehensive and authoritative
regional planning process for transmission
expansion issues, including transmission
siting, fi nancing, construction, ownership and
cost recovery;
4) Restrictions on federal borrowing authority hinder
BPA’s fi nancing of new transmission construction;
4-8
5) There are multi-jurisdictional siting and
permitting issues for new large-scale
transmission expansion;
6) The regional transmission organization forum
is still being resolved, as is the subsequent
jurisdiction over the organization.
Table 4.1: Avista Generation Integration Cost Estimates (2005$)
From
Capacity
(MW) Potential Upgrade
Approximate Capital
Cost ($millions)
Eastern MT
350 Install 500 kV series capacitors on
existing lines 100-150
750
Install 500 kV series capacitors
& reinforcements such as 230 kV
reinforcements in Eastern WA
400-450
1,000 New 500kV line 1,500
Eastern WA to Mid-
Columbia
350 N/A 100
750 N/A 150
1,000 N/A 600-800
Eastern WA – Adjacent
to Existing 230kV
System
350 Additional substation 10
750 Additional 230 kV reinforcement 80
Northern ID – Adjacent
to Existing 230kV
System
350 New substation 10
750 230 kV reinforcement 70
Eastern WA – Remote
From Existing 230kV
System
350 New double circuit 230 kV line east
of Spokane 50
750 New double circuit 230 kV line -
Spokane to Mid-Columbia 100
Eastern WA (Wind) 80-150 Depending on the size 10-70
5-1
The analytical foundation for this IRP was to model
the Western states’ electric system and markets
to quantify impacts on Avista. The Company used
this approach to derive electric prices for the Mid-
Columbia market, taking into account physical
systems outside the Northwest. Understanding
all the geographic areas within the Western
Interconnect is important because the area functions
as one larger market with various sub-markets.
Prior to 2003, Company IRPs relied on market price
forecasts modeled exogenously, breaking the link
between the market price forecast and modeling
of Company operations. This IRP combines these
efforts by tracking Company-owned and contracted
resources as they dispatch into the modeled
marketplace. The resource portfolio then is linked to
its loads, resources and contractual arrangements to
calculate expected power supply costs.
The Company used a multi-step approach to
develop the Preferred Resource Strategy (PRS).
5. MODELING APPROACH
Section Highlights
 Avista uses AURORAXMP to model hourly operations of the entire Western Interconnect;
market conditions outside the Northwest affect local market prices.
 The Company performed Monte Carlo market analyses, varying load, hydro, wind and natural
gas price data over 200 iterations.
 The 2005 IRP benefi ts from signifi cant wind modeling enhancements.
 The proprietary Avista Linear Programming Model helped direct the Preferred Resource Strategy.
 The IRP adopts many assumptions from the Northwest Power and Conservation Council’s Fifth
Power Plan.
 The federal production tax credit for renewables is assumed throughout the IRP timeframe,
except in carbon tax scenarios where the credit terminates.
 The IRP accounts for transmission costs necessary to bring distant generation sources into the
Northwest (e.g., Montana coal and wind).
5-2
Figure 5.1: Modeling Process Diagram
5-3
Potential new resources were identifi ed to serve
future demand in the Western Interconnect. New
resources were combined with existing resources
and then used to simulate hourly operations from
2007 to 2026, using a Monte Carlo analysis varying
hydro, wind, load, and gas prices. The simulation
results were used to estimate Mid-Columbia
electric market prices. These prices were used
to analyze potential new conservation initiatives
and supply side resources. This step values plant
operations and weighs those values against capital
requirements using Avista’s Linear Programming
(LP) model; the LP model selects optimal resources
to serve load based on energy and capacity needs,
cost, value and risk. Figure 5.1 presents a visual
interpretation of the modeling process.
5.1 Western Interconnect
Simulation: AURORAXMP
The AURORAXMP model was used to simulate
the Western Interconnect market for the 2005
IRP. The Western Interconnect includes the states
west of the Rocky Mountains, as well as British
Columbia, Alberta, and Baja, Mexico. This area
is highlighted on the map in Figure 5.2. The
Western Interconnect is separated from the Eastern
Interconnect and ERCOT systems except for eight
inverter stations between the three systems. The
Western Interconnect follows operation and reliability
guidelines administered by the Western Electric
Coordinating Council (WECC).
1 Graphic courtesy of NERC and can be found at http://www.nerc.com
Figure 5.2: NERC Interconnections Map 1
5-4
AURORAXMP separates the Western Interconnect
into sixteen “zones” based on load concentration
and transmission constraints. Table 5.1 lists the
Western Interconnect zones included in AURORAXMP.
This table also provides a reference to the zone
acronyms used later in this document.
The AURORAXMP database contains hourly loads
and resources for each zone in Table 5.1. These
components along with fuel prices, transmission
constraints, hydro conditions and wind conditions
allow the model to simulate the Western Interconnect
system on an hourly basis. This simulation is used
to derive market-clearing prices for each zone.
Market-clearing prices are derived from the marginal
cost to supply the next megawatt of energy plus any
applicable wheeling charges for each unit.
The model meets future loads by choosing new
generating assets from a pool of hypothetical user-
defi ned resources. Hypothetical construction of new
resources is referred to as “capacity expansion.”
In capacity expansion, the model calculates a net
present value for each new resource by subtracting
fuel costs, variable operations and maintenance
(O&M), fi xed O&M, emissions costs and capital
investment from its expected market value. The
model uses an iterative process that places plants
into the system and selects those with positive net
present values. After the expansion studies are
completed, the model simulates the system using
the optimal set of new resources for all 175,320
hours of the 20-year study.
After capacity expansion, a stochastic analysis is
performed in AURORAXMP to incorporate market
uncertainty. Stochastic analysis is performed
using probability distributions for load, fuel price,
hydroelectric and wind generation data, rather
than by simply using single point estimates. The
Company generated 200 sets of unique inputs for
200 distinct 20-year iterations of AURORAXMP. In
Zone Area(s) Included Zone Area(s) Included
AB Alberta IDS Southern Idaho
AZ Arizona MT Montana
BAJA Baja Mexico NM New Mexico
BC British Columbia NNV Northern Nevada
NCAL Northern California SNV Southern Nevada
CCAL Central California OWI OR, WA, & Northern Idaho
SCAL Southern California UT Utah
CO Colorado WY Wyoming
Table 5.1: AURORAXMP Zones
5-5
total, the Company simulated more than 70 million
market hours for the 2005 IRP, requiring nearly 5,000
hours of computer processing and 300 gigabytes of
data storage for each stochastic study. In addition
to stochastic studies, Avista looks at individual
deterministic scenarios to understand how certain
variables drive results.
5.2 Key Assumptions
and Inputs
AURORAXMP contains a database with generic data
developed by EPIS, Inc. The database provides
a reasonable approximation of future market
conditions. The Company modifi ed many of the
base data sets to obtain more robust results. The
following section describes the changes made by
the Company for the 2005 IRP.
Hydroelectric Generation
The AURORAXMP model is shipped with hydrological
data sets for the entire Western Interconnect. For
the Northwest, data includes average monthly
generation levels taken from Bonneville Power
Administration (BPA) 50-year hydrologic studies.
The Company uses hydrologic data from the
Northwest Power Pool (NWPP) rather than BPA data
for planning and ratemaking. Presently, the NWPP
performs 60-year headwater benefi t studies annually
for the Northwest hydroelectric system.
Data from the 60-year NWPP Headwater Benefi ts
Study was converted into an AURORAXMP format
and Northwest data sets for IRP modeling.
AURORAXMP data for zones outside the Northwest
(e.g., California) were not modifi ed.
AURORAXMP models hydroelectric generation
by load area or zone. This means that every
hydroelectric facility located within a zone utilizes
the same shaping factors.2 The results for the entire
hydroelectric system are accurate, but individual
projects may not be correctly represented. To
track Company-owned hydroelectric resources
more accurately, each Company river system was
separated from the base hydroelectric data set. A
unique set of shaping factors, based on historic
generation, was assigned to each project. Figure
5.3 demonstrates monthly capacity factors for the
OWI zone, and the Company’s hydroelectric projects
in an average water year.
The model dispatches hydro resources based on
demand changes. Hydro units are dispatched
before thermal, wind or other resources. To
dispatch hydro, the model takes several factors
into consideration including available annual and
monthly energy, minimum and maximum capacity,
and load following ability. Figure 5.4 demonstrates
hydro load following in one hypothetical week.
Natural Gas Prices
The price of natural gas is a key model assumption
because gas-fi red resources presently set the
marginal electricity price for the majority of hours
at trading hubs across the Western Interconnect.
2 Shaping factors determine how much each hydroelectric facility can vary
its operations to serve peak loads.
5-6
400
500
600
700
800
900
1,000
1,100
1,200
Time
Load served by
other resources
Load served by
hydro dispatch
Total
Load
Figure 5.4: One Week Hydro Dispatch Example (MW)
Figure 5.3: NW & Avista Monthly Hydro Capacity Factors Modeled in AURORAXMP (%)
0
20
40
60
80
100
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
OWI
Mid C
Spokane
Clark Fork
5-7
The gas price forecast was developed in April
2005. It uses a blend of NYMEX forward prices
and Global Insight Inc.’s Gas Escalation Forecast.
NYMEX monthly forward prices for Henry Hub were
obtained on April 6, 2005, for 2007 through 2010.
Global Insight’s escalation rates are used from 2011
through the duration of the forecast period.
To accurately model the Western Interconnect,
additional natural gas basin forecasts are required.
Northern basins at AECO, Malin, and Sumas,
and southern basins at Opal, Topock, and San
Juan were added to the model. Northern basins
use forward market differentials and southern
basins use generic differentials included with the
AURORAXMP database. The difference in handling
northern and southern basins is due to the minimal
relative impact of southern gas on Company
costs and southern basin differentials not being
readily available to the Company. The Company’s
natural gas procurement group reviewed the
differentials provided by the AURORAXMP database
and determined that they were reasonable for
IRP modeling purposes. Table 5.2 contains the
natural gas price forecasts used for the 2005 IRP.
An additional transportation charge was added to
move gas between basins and plant locations.
Hub/Zone 2007 2008 2009 2010 2011 2012 2016 2020 2024 2026
AECO Hub 6.68 6.25 5.75 5.36 5.39 5.41 5.87 6.46 6.70 7.05
Henry Hub 7.37 6.92 6.41 6.01 6.07 6.15 6.81 7.57 8.07 8.60
Malin Hub 7.01 6.55 6.03 5.62 5.66 5.72 6.33 7.04 7.49 7.97
Sumas Hub 6.86 6.40 5.88 5.46 5.49 5.54 6.13 6.83 7.25 7.72
AB 6.80 6.37 5.87 5.48 5.52 5.54 6.01 6.62 6.88 7.23
AZ 6.58 6.16 5.65 5.26 5.28 5.29 5.74 6.32 6.55 6.89
BAJA 7.27 6.81 6.30 5.89 5.95 6.03 6.68 7.43 7.94 8.46
BC 6.80 6.37 5.87 5.48 5.52 5.54 6.01 6.62 6.88 7.23
CCAL 7.44 6.98 6.48 6.08 6.14 6.22 6.89 7.67 8.19 8.73
CO 6.55 6.13 5.62 5.22 5.25 5.25 5.70 6.28 6.51 6.84
IDS 7.09 6.64 6.12 5.71 5.74 5.80 6.42 7.14 7.60 8.09
MT 6.92 6.49 5.99 5.61 5.65 5.67 6.16 6.78 7.05 7.42
NCAL 7.12 6.67 6.16 5.75 5.78 5.85 6.47 7.20 7.66 8.16
NM 6.55 6.13 5.62 5.22 5.25 5.25 5.70 6.28 6.51 6.84
NNV 6.52 6.04 5.51 5.08 5.13 5.18 5.82 6.58 7.06 7.58
OWI 6.97 6.52 6.00 5.59 5.61 5.67 6.27 6.98 7.42 7.90
SCAL 7.44 6.98 6.48 6.08 6.14 6.22 6.89 7.67 8.19 8.73
SNV 6.63 6.15 5.63 5.20 5.24 5.30 5.96 6.73 7.22 7.74
UT 6.46 5.98 5.45 5.01 5.06 5.11 5.75 6.50 6.96 7.48
WY 6.40 5.92 5.39 4.95 5.00 5.05 5.68 6.42 6.88 7.39
Table 5.2: Trading Hub and Zone Natural Gas Price Forecast ($/dth)

5-8
Figure 5.5 shows annual average natural gas prices
at Henry Hub used in the Base Case analysis. The chart
shows prices in both 2005 and nominal year dollars.
Resources
A Company review of existing Western Interconnect
resources included in the AURORAXMP database
found it to be comprehensive and accurate for IRP
purposes after some modifi cation. Two substantial
changes were made to the AURORAXMP database
for new construction and Renewable Portfolio
Standard (RPS) resources. New generating
resources currently under construction and likely to
be constructed as defi ned by the California Energy
Commission were included in the resource base.
RPS resources were included based on data from
the Northwest Power and Conservation Council
(NPCC) Fifth Power Plan.
Plants under Construction
Figure 5.6 describes approximately 9,900 aMW
of resources presently under construction and
expected to be online during the study’s time frame.
These resources were included in all studies and
scenarios. New gas represents 88 percent of the
new energy, while wind accounts for three percent
and coal seven percent.
Renewable Portfolio Standards
States with RPS legislation were explicitly modeled
in AURORAXMP The methodology to select
renewable resource types is either consistent
with the NPCC’s Fifth Power Plan or follows state
statute. Plants identifi ed as RPS resources are
fi xed within the model and are consistent across all
studies and scenarios.
Figure 5.5: Henry Hub Natural Gas Price Forecast ($/dth)
4.00
4.50
5.00
5.50
6.00
6.50
7.00
7.50
8.00
8.50
9.00
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Nominal
2005$
NYMEX Forwards Global Insights Escalations
5-9
Figure 5.6: New Resources Under Construction (MW)
Gas
87.9%
Hydro
1.3%
Wind & Solar
3.3%
Geothermal
0.2%
Coal
7.3%
Table 5.3 shows states that have renewable portfolio
standards and the RPS requirement that was modeled.
Future Resource Alternatives
As part of the AURORAXMP simulation, new
resources are identifi ed to meet future load growth.
This IRP considers generic resource alternatives
identifi ed in the NPCC Fifth Power Plan that could
be built across all AURORAXMP zones. The Company
believes that NPCC resource assumptions provide
greater transparency in the IRP process. The NPCC
resources were formulated through a committee of
regional experts drawn from utilities, developers,
regulators and other interested parties.
The Company does not have a resource defi ciency
until 2009; therefore the Company has not recently
studied site-specifi c projects. This IRP provides a
framework of analysis that the Company expects
to utilize for future resource procurements.
Assumptions will be updated at that time to
include site-specifi c resource alternatives. Specifi c
resource alternatives drawn from a Request for
Proposals, or other acquisition process, would
be evaluated in the same manner as the NPCC
resources used in this study.
AURORAXMP Modeling Divergences
from the NPCC
The Company diverged modestly from NPCC
resource assumptions in three areas: the federal
production tax credit (PTC) for renewables;
transmission costs for new coal, wind and oil
sand plants; and the use of capacity credits.
The Company also has updated certain datasets
with more recent information than was available to
the NPCC.
State RPS Date Level (%)
Arizona 2007 1.10
California 2017 20.00
Colorado 2015 10.00
Nevada 2013 15.00
New Mexico 2011 10.00
Table 5.3: Renewable Portfolio Standards by State
5-10
Data Source Used
Infl ation Company Forecast is Based on Global Insight, Inc.
Load Escalation WECC 2004 Load & Resource Report and the 2004 Pacifi Corp IRP for Utah
Coal Escalation EIA’s Annual Energy Outlook 2005
Wind Monthly Generation Replaced by Hourly Shapes
Start Up Costs Fuel Price Adders Replaced With Start Fuel and O&M Start-Up Costs
Table 5.4: IRP Differences from Fifth Power Plan
Production Tax Credit
The NPCC models the wind PTC as an offset to
variable O&M costs directly within AURORAXMP.
The Company chose to reduce fi xed costs in each
year by an amount equal to the tax credit in its
revenue requirements model. The ultimate impact
of this change was negligible, but it more accurately
accounted for the credit value, including the impact
on the Company’s federal income tax obligations.
In addition, the modeling accounts for the PTC
as extended to other renewables (geothermal,
biomass, solar) by the Federal 2004 HB 4520 Jobs
Act. The PTC is assumed to be available throughout
the timeframe of the study, except where carbon
legislation is enacted. Where carbon legislation is
enacted, the PTC is assumed to expire.
Incremental Transmission
The Company sought to improve the NPCC’s
incremental transmission cost estimates for
integrating plants into the Northwest and the
Western Interconnect. Existing transmission lines
out of eastern regions in the Western Interconnect
to the Northwest do not have adequate capacity to
integrate large coal or wind plant developments. A
combination of new and upgraded transmission
facilities likely will be required to integrate
such plants. To account for new transmission
construction, the capital and operating costs of the
new transmission are added to the costs of new
generation resources.
Capacity Credits
Capacity credits provide a fi nancial incentive for
the model to build more generation than is needed
under average conditions. The AURORAXMP model
has perfect foresight and builds just enough
resources to meet future load growth assumptions.
It does not build additional resources for planning
margin. Providing credits is similar to the regulated
environment where planning margins are retained to
meet load under adverse conditions. The capacity
credit is applied by reducing the capital cost of new
generating resources. A fi nal credit was developed
by testing various values until the wholesale
marketplace reached a balance.
The credit amount is equal to $31.22 per kilowatt-
year for a plant with 90-percent availability. The credit
is smaller for plants with lower availability such as wind
or solar plants. For example, a 25-percent availability
wind plant is credited $8.22 per kilowatt-year.
5-11
Other Changes
The Company chose to incorporate other data that
became available after the NPCC Fifth Power Plan
was drafted. Table 5.4 lists the remaining major
differences between this IRP and the NPCC Fifth
Power Plan.
5.3 Risk Modeling
The 2005 IRP relies on work initially developed
for the 2003 IRP. It continues to enhance the risk
evaluation capabilities of Company models.
In addition to stochastically modeling hydroelectric
output, natural gas prices and load variability,
the 2005 IRP models wind plant generation
stochastically. Natural gas prices also were
reevaluated, and a new approach to obtaining
stochastic variables was pursued.
Background
Stochastic risk analysis offers a powerful means to
understand the potential impact of portfolio options
under various “draws” of future conditions. The
life-cycle costs of long-lived resources are critical
to the Company and its customers. For example,
the Company’s oldest active generation facility the
Monroe Street hydroelectric project was built in
1890. Company investments in Colstrip Units 3 &
4, made in the mid-1980s, generate cost-effective
electricity for our customers today.
Resource decisions therefore must provide cost-
effective power for years to come. Reducing cost
volatility for customers and shareholders is also
important when considering long-term investments.
The energy crisis in 2000-01 changed utility
planning views of electric market price volatility risk.
Stochastic analysis helps us understand possible
variations inherent in future resource options
and how to diversify resource types to arrive at a
portfolio that reduces cost and minimizes variation.
Implementation
Preparing a stochastic analysis requires a large
number of unique datasets. To understand the
impact of varying customer load conditions on the
resource decisions made for this IRP, 200 unique
20-year datasets for each zone in the Western
Interconnect were created. More than 46 million
daily loads were ultimately evaluated through the
stochastic process. Similar work was performed
for natural gas prices, hydroelectric generation and
wind. A separate model was developed to evaluate
historical relationships and project possible futures
for each stochastic variable. Each stochastic
variable is further described below.
Hydroelectric Generation
The Company portfolio is dominated by
hydroelectric generation. Over 40 percent of
customers’ electricity is generated by hydroelectric
projects today. NWPP estimates of hydroelectric
generation over the 1929–1988 period were used
to develop the stochastic variables for Northwest
hydroelectric generation. As the Company learned
in the 2003 IRP process, streamfl ows are normally
distributed but hydroelectric generation is not.
Therefore, using the simplifi ed mean/standard
deviation approach to create hydroelectric datasets
5-12
was not possible. Generation levels were estimated
by taking random draws from the 60-year NWPP
dataset, with each draw containing a full year of the
hydroelectric record.
Hydroelectric generation levels outside of the
Northwest were held constant throughout the
stochastic process due to a lack of available data.
The Company believes this decision still provides
a robust analysis of hydroelectric generation since
Northwest hydroelectric plants account for 85
percent of all hydroelectric generation in the Western
Interconnect. Table 5.5 illustrates that the OWI
zone by itself accounts for more than half of all
hydroelectric generation. Figure 5.7 presents the
distribution of hydroelectric generation modeled for
the Western Interconnect.
Natural Gas Prices
Natural gas and electricity prices are highly
correlated across the Western Interconnect. The
correlation refl ects the region’s increased reliance on
natural gas-fi red generation, a relationship expected
to continue, because natural gas-fi red plants set
marginal electricity prices in most hours.
Figure 5.8 shows the relationship of prices in
the Northwest as the correlation between Mid-
Columbia electricity prices and the Malin hub gas
prices in January, June and August over the IRP
time horizon. Correlations rise modestly over time,
especially in the month of June. The change in
June refl ects forecasted additions of gas-fi red
generation in the Southwest as the Western
Interconnect continues to outgrow its hydroelectric
generation base.
Changes Since The 2003 IRP
Two natural gas assumptions were changed for
this IRP: 1) Hydroelectric conditions are no longer
modeled to affect natural gas prices directly; and
2) the distribution is log-normally distributed
rather than normally distributed. Evaluations of
the wholesale marketplace since the 2003 IRP
indicate that hydroelectric generation levels do not
signifi cantly impact natural gas prices.
Zone %WI Avg Min Max
OWI 54 14,091 10,604 17,672
BC 23 6,048 5,588 6,558
NCAL 7 1,850 1,850 1,850
IDs 5 1,331 885 1,850
AZ 3 829 829 829
MT 3 709 526 866
SCAL 2 583 583 583
SNV 2 429 429 429
AB 0 126 126 126
CO 0 81 81 82
UT 0 54 54 54
NM 0 17 17 17
WY 0 15 15 15
NNV 0 6 6 6
CCAL 0 1 1 1
BAJA 0 0 0 0
Total3 100 26,171 21,801 30,515
Table 5.5: Hydroelectric Generation Statistics
by Zone (aMW)
3 Minimums and maximums are Western Interconnect-wide
coincident totals
5-13
Figure 5.7: Western Interconnect Hydroelectric Generation Distribution
0
100
200
300
400
500
600
21
.
5
22
.
0
22
.
5
23
.
0
23
.
5
24
.
0
24
.
5
25
.
0
25
.
5
26
.
0
26
.
5
27
.
0
27
.
5
28
.
0
28
.
5
29
.
0
29
.
5
30
.
0
30
.
5
aMW (thousands)
Fr
e
q
u
e
n
c
y
Figure 5.8: Mid-Columbia Electricity and Malin Natural Gas Price Correlations (%)
40
50
60
70
80
90
100
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Aug
Jan
Jun
5-14
A “critical” water year in the Northwest reduces
hydroelectric generation by approximately four
thousand average megawatts. Replacing this
hydroelectric generation with gas-fi red generation
would increase total U.S. natural gas consumption
by less than one percent. Smaller reductions
seen in other hydroelectric generation years would
impact the U.S. marketplace even less. Given
the modest impact of hydroelectric conditions on
natural gas consumption, natural gas prices and
hydroelectric generation levels are not correlated in
2005 IRP analyses.
The decision to adopt a lognormal distribution for
natural gas prices refl ects input received by the
Company since the 2003 IRP was published.
Many peer utilities, and other groups evaluating
wholesale natural gas markets, assume a lognormal
price distribution.
As with any stochastic forecast, the 2005 IRP
necessarily must assume a mean (average) price
and sigma (standard deviation). The 2005 IRP
continues with the 2003 IRP natural gas sigma
assumption of 50 percent. This means that two-
thirds of all gas prices in the study fall within 50
percent of the mean. Because the 2005 mean
forecast for natural gas prices has increased by
approximately one third from the 2003 IRP,
nominal sigma values are also increased.
Figure 5.9 and Figure 5.10 illustrate statistics
for 2007 and 2016.
Load Variability
Loads across the Western Interconnect are not
independent. In other words, often times heat
waves and cold snaps occur at the same time
in the Northwest and Southwest. Representing
this relationship is important when developing a
representation of the future wholesale marketplace.
The 2005 IRP relies on Western Interconnect-wide
statistical relationships developed for the 2003 IRP.
The earlier work developed monthly and weekly
distributions based on hourly data from all utilities
obtained from FERC Form 714. Correlations
between the Northwest and other Western
Interconnect load areas were found and represented
in the stochastic load model. Correlating zone loads
avoids oversimplifi cation. Absent correlation data,
the stochastic models would offset load changes
in one zone with load changes in another zone.
Given the high degree of interdependency across
the Western Interconnect (e.g., the Northwest and
California), this additional accuracy is considered
crucial for understanding wholesale electricity
market price variation.
Tables 5.6a and 5.6b illustrate the correlations used
for the 2005 IRP. Tables 5.7a and 5.7b provide
mean and sigma values for each zone in 2007. The
BAJA area has no load and was not included in this
study. “NotSig” indicates that no statistically valid
correlation was found in the evaluated data. “Mix”
represents that the relationship was not consistent
across time, and that it was not used.
5-15
Figure 5.9: Natural Gas Price Statistics-2007 ($/dth)
4
5
6
7
8
9
10
11
12
13
14
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Mean 80% CI High
80% CI Low MAX
MIN
Figure 5.10: Natural Gas Price Statistics-2016 ($/dth)
3
4
5
6
7
8
9
10
11
12
13
14
15
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Mean 80% CI High
80% CI Low MAX
MIN
5-16
Area Jan Feb Mar Apr May Jun
AB 0.659 NotSig 0.481 NotSig Mix 0.635
AZ 0.44 0.664 NotSig Mix -0.29 0.666
BC 0.918 0.838 0.825 0.733 0.617 NotSig
CCAL NotSig 0.734 NotSig NotSig NotSig 0.771
CO 0.623 NotSig 0.567 Mix Mix NotSig
IDs 0.673 0.747 0.882 NotSig NotSig 0.758
MT 0.894 0.773 0.755 0.651 0.405 0.599
NCAL NotSig 0.734 NotSig NotSig NotSig 0.771
NM 0.384 Mix Mix NotSig NotSig Mix
NNV Mix NotSig NotSig NotSig NotSig NotSig
SCAL NotSig Mix NotSig NotSig Mix 0.68
SNV NotSig 0.641 0.513 Mix NotSig 0.729
UT 0.816 NotSig 0.669 0.697 0.61 0.698
WY 0.765 Mix 0.641 NotSig Mix Mix
Table 5.6a: Western Interconnect Load Correlations—Jan through Jun
Area Jul Aug Sep Oct Nov Dec
AB 0.668 Mix Mix 0.479 NotSig NotSig
AZ NotSig NotSig NotSig NotSig Mix NotSig
BC 0.56 NotSig 0.638 0.809 0.525 0.89
CCAL Mix 0.757 0.789 NotSig Mix NotSig
CO NotSig NotSig NotSig 0.655 0.629 0.571
IDs Mix 0.789 0.733 0.561 0.587 0.813
MT 0.786 0.648 0.752 NotSig 0.856 0.898
NCAL Mix 0.757 0.789 NotSig Mix NotSig
NM NotSig Mix NotSig NotSig Mix Mix
NNV NotSig NotSig NotSig Mix 0.476 NotSig
SCAL Mix 0.5 0.778 NotSig NotSig NotSig
SNV Mix NotSig Mix NotSig 0.461 Mix
UT 0.703 0.604 0.611 NotSig 0.561 0.837
WY NotSig NotSig 0.483 NotSig 0.522 0.633
Table 5.6b: Western Interconnect Load Correlations—Jul through Dec
5-17
Area Value Jan Feb Mar Apr May Jun
AB Mean 7.98 7.19 7.61 6.98 7.12 7.00
StDev 0.29 0.22 0.25 0.22 0.22 0.25
AZ Mean 8.26 7.53 7.48 7.28 8.22 9.39
StDev 0.56 0.50 0.32 0.47 0.69 1.15
BC Mean 8.76 7.74 8.06 7.27 7.20 6.82
StDev 0.43 0.31 0.40 0.35 0.41 0.29
CCAL Mean 1.52 1.42 1.44 1.40 1.43 1.55
StDev 0.10 0.08 0.09 0.10 0.10 0.16
CO Mean 6.37 5.96 5.87 5.57 5.49 5.81
StDev 0.28 0.25 0.26 0.24 0.27 0.37
IDS Mean 2.50 2.17 2.20 2.10 2.29 2.52
StDev 0.11 0.07 0.11 0.09 0.16 0.23
MT Mean 1.48 1.31 1.33 1.22 1.19 1.17
StDev 0.05 0.03 0.04 0.03 0.04 0.04
NM Mean 2.74 2.49 2.56 2.47 2.62 2.83
StDev 0.09 0.08 0.07 0.08 0.11 0.15
NVN Mean 1.12 1.00 1.04 1.03 1.18 1.34
StDev 0.02 0.02 0.02 0.03 0.03 0.06
NVS Mean 2.66 2.38 2.46 2.45 2.80 3.19
StDev 0.10 0.09 0.06 0.14 0.24 0.47
NCAL Mean 14.82 13.87 14.04 13.65 13.95 15.09
StDev 1.00 0.74 0.85 0.95 1.01 1.58
OWI Mean 20.43 18.99 18.47 16.96 16.55 16.13
StDev 1.38 1.01 1.12 1.18 1.20 1.69
SCAL Mean 22.43 20.99 21.24 20.66 21.10 22.83
StDev 1.72 1.45 1.47 1.57 1.75 1.92
UT Mean 3.32 3.01 3.05 2.91 3.04 3.18
StDev 0.27 0.23 0.31 0.28 0.30 0.37
WY Mean 2.13 2.06 2.09 1.96 1.94 1.93
StDev 0.02 0.02 0.02 0.01 0.02 0.01
Table 5.7a: Western Interconnect Load Statistics–Jan through Jun (2007 aGW)
5-18
Area Value Jul Aug Sep Oct Nov Dec
AB Mean 7.38 7.37 7.11 7.54 7.74 8.23
StDev 0.29 0.29 0.23 0.23 0.22 0.29
AZ Mean 11.00 11.36 10.66 8.87 7.50 8.10
StDev 0.70 0.55 0.80 0.68 0.35 0.54
BC Mean 6.93 7.03 6.88 7.65 8.14 8.77
StDev 0.32 0.35 0.32 0.34 0.34 0.43
CCAL Mean 1.65 1.75 1.71 1.61 1.46 1.50
StDev 0.18 0.19 0.16 0.11 0.09 0.10
CO Mean 6.34 6.47 6.23 5.76 5.74 6.35
StDev 0.36 0.33 0.35 0.25 0.29 0.32
IDS Mean 2.83 2.62 2.24 2.16 2.19 2.46
StDev 0.13 0.13 0.17 0.08 0.09 0.13
MT Mean 1.23 1.27 1.19 1.19 1.30 1.40
StDev 0.06 0.04 0.04 0.02 0.04 0.05
NM Mean 3.05 3.04 2.86 2.64 2.53 2.74
StDev 0.12 0.11 0.11 0.09 0.09 0.10
NVN Mean 1.55 1.55 1.31 1.13 1.07 1.18
StDev 0.06 0.06 0.06 0.03 0.04 0.04
NVS Mean 3.70 3.70 3.13 2.69 2.55 2.82
StDev 0.25 0.21 0.27 0.18 0.07 0.12
NCAL Mean 16.10 17.06 16.64 5.68 14.20 14.59
StDev 1.78 1.87 1.56 1.05 0.92 0.93
OWI Mean 16.41 16.32 15.63 15.90 16.93 19.43
StDev 1.82 1.79 1.46 1.07 1.09 1.24
SCAL Mean 24.35 25.81 25.18 23.72 21.48 22.07
StDev 2.50 2.19 2.81 1.79 1.61 1.61
UT Mean 3.55 3.55 3.18 3.12 3.16 3.50
StDev 0.38 0.35 0.36 0.24 0.29 0.38
WY Mean 1.88 2.01 1.88 2.00 2.06 2.12
StDev 0.02 0.01 0.02 0.01 0.02 0.02
Table 5.7b: Western Interconnect Load Statistics—Jul through Dec (2007 aGW)
5-19
Figure 5.11: Actual Wind Data - 1000 Continuous Hours (aMW)
-
20
40
60
80
100
120
140
160
180
200
Time
Wind Generation
The 2005 IRP benefi ts from the addition of
stochastic wind analysis. Evaluations of wind
traditionally have oversimplifi ed assumptions due
to a lack of data. Some analyses have simplifi ed
to the point of assuming that wind generation is fl at
over all hours of each month. Other analyses have
developed stochastic relationships that ignore serial
correlation. Ignoring serial correlation disregards the
fact that current-hour generation tends to be highly
correlated with what happened in the previous hour.
The importance of this is illustrated in the next three
tables. Each table includes 1,000 hours of wind
generation values.
Figure 5.11 provides actual generation from a
Northwest wind facility. The scale has been
adjusted to a maximum capability of 200 MW to
protect the identity of the site. Figure 5.12 provides
a simple stochastic representation of the data with
no serial correlation assumed.
Figure 5.13 shows the results of the model
developed by the Company for the 2005 IRP. The
time period of the Company-generated data is
different in Figure 5.13 than in Figure 5.11; even
though total generation is higher, the general pattern
is similar. In summary, the three charts explain
that a simplifi ed wind model signifi cantly misstates
the variability when compared to a simulation of
actual operations. The Company’s modeling more
accurately refl ects wind output.
In today’s marketplace oversimplifi cation might
not greatly affect wind resource decisions, as it is
not a signifi cant enough power source to affect
5-20
Figure 5.12: Stochastic Wind Model Absent Serial Correlation (aMW)
-
20
40
60
80
100
120
140
160
180
200
Time
Figure 5.13: Stochastic Wind Model With Serial Correlation (aMW)
-
20
40
60
80
100
120
140
160
180
200
Time
5-21
market prices; however, wind energy is expected to
represent a growing part of the Western Interconnect
resource mix throughout the planning horizon.
Continuing to simplify the characteristics of this
resource is no longer appropriate.
The 2003 IRP action plan stated that Avista would
study wind further. Based on analyses completed
since the fi ling of our 2003 IRP, more robust wind
data underlie assumptions used for the 2005 IRP.
Northwest wind speed data from Oregon State
University (OSU) was evaluated hourly from 1985
through 2000 to develop statistical distributions
for sample wind sites. Five separate wind sites
were assessed to develop a combined distribution
for Northwest wind generation. This approach is
similar to how AURORAXMP dispatches hydroelectric
generation. Market prices are affected by the total
dispatch of hydro resources in any given hour,
irrespective of how one hydroelectric plant operates.
The same logic holds true for wind power.
The stochastic wind model developed by Avista for
the 2005 IRP produces daily generation levels and
shapes them based on the monthly average hourly
wind shape over the 1985-2000 period. With daily
generation levels changing, the model more accurately
represents the variability inherent in the resource.
Absent equivalent data for areas outside the
Northwest, Avista looked to the Seams Steering
Group-Western Interconnect to obtain average
monthly generation and variance levels.4
Due to the lack of multi-year datasets, each load
area outside of the Northwest was assigned a
sigma value equivalent on a percentage basis to
the information obtained from the OSU database.
Independent models were run to generate
synthesized stochastic wind data for the entire
Western Interconnect.
5.4 The Avista LP Model
The Company uses a proprietary linear programming
(LP) model to assist in developing its Preferred
Resource Strategy, rather than relying on a set of
predetermined resource portfolios. Avista believes
that using this approach is superior to simply using
portfolios. Predetermined portfolios are simpler
to understand, but they ignore the thousands of
potential resource mixes available to the Company
to serve future loads. For example, a wind portfolio
can be comprised of many different wind projects,
each with varying characteristics (e.g., location). The
Avista Linear Programming model approach looks
at 180 different wind options nine different wind
basins, each available for selection over the 20-year
forecast horizon. The LP model does not preclude
the Company from using portfolios to help readers
understand the effect of different market scenarios
on several generic resource types. Instead, the
LP model helps develop portfolios used later in
the results section to help illustrate the relative
performance of specifi c resource strategies.
The LP model relies on three primary datasets:
1) Avista load requirements (capacity and energy)
over time; 2) capital recovery costs associated with 4 http://www.ssg-wi.com
5-22
new resource alternatives, inclusive of locational
transmission pricing; and 3) the value of each new
resource alternative over 200 iterations of 20-
year stochastic analysis performed in the market
forecasting model AURORAXMP.5 The LP model
is guided by various constraints to arrive at a
least-cost solution defi ned in terms of the present
value of expected power supply expenses and
risk, measured as the standard deviation of
the same expenses.
Constraints
Various constraints were placed on the LP model.
The model ensures that suffi cient capacity and
energy are constructed in every year to serve
annual customer demand. Energy quantities
were defi ned as minimum levels, allowing more
energy than necessary to be constructed. This
assumption refl ected actual utility planning
requirements. Capacity was a capped constraint.
The model matched forecasted capacity
requirements in every year.
An optimization algorithm gave the model a strong
bias to limit market purchases and sales. Legislation
and regulation in the Northwest is not favoring
further rulemakings that would limit or discourage
utility resource acquisition or construction to serve
load growth; therefore, it is unlikely in the current
environment that independent generators will
develop adequate resources on a speculative
basis to serve utility requirements. For this and for
other fi nancial and credit reasons, the Company
believes it would be inappropriate to rely on large
purchases from the market in the long term to
serve fi rm load obligations.
Wind generation has become more attractive
relative to other resource options because of rising
natural gas costs and the related rise in wholesale
electricity prices. Wind power economics were
questioned in the past because of their similar
busbar cost to natural gas-fi red generation and
the uncertainty surrounding additional integration
costs necessary to “fi rm” the resource. Now that
natural gas-fi red projects have become costlier,
the difference between the busbar cost of gas
and wind has grown to a point that likely exceeds
wind integration costs. The model recognizes this
condition and builds signifi cant amounts of wind
resources absent constraints.
Though preferred by the Avista LP model, it is
unlikely that the high level of wind resources
identifi ed in early runs would be obtainable. The
model was constrained to select no more than 650
MW of wind in any given resource mix. The limit is
discussed later in the section. The LP model also
was constrained to allow no more coal reliance than
350 MW in 2016, 450 MW in 2021, and 550 MW in
2026. Ultimately, the 550 MW coal constraint was
not necessary, as the Preferred Resource Strategy
identifi ed a need for 450 MW.
5 AURORAXMPaccounts for variable O&M and fuel costs for each resource
valuation
5-23
Effi cient Frontier
The Avista LP model was used to defi ne a 1,000-
point “effi cient frontier” of resource options over the
full range of risk and cost. This method provides an
optimal resource build for each level of willingness
to accept more volatility. Figure 5.14 provides the
effi cient frontier. To create an effi cient frontier, the
LP model is directed to fi nd the lowest cost resource
mix for each level of risk or variation in power
costs. Capital costs generally tend to be inversely
correlated with risk.
5.5 New Resource Alternatives
Each zone modeled in AURORAXMP has the potential
to build a wide variety of resource types. Resource
availability varies between geographic areas
because of the potential for renewables, the cost of
new transmission and the difference in each region’s
attitude toward certain fuel types. For example,
Wyoming and Montana are open to new coal plants.
California is not assumed to be building any new
coal plants in the 2005 IRP. This assumption is
based on decisions made by the NPCC in its Fifth
Power Plan.
Underlying assumptions for each new resource
are based on recent work by the NPCC. For
further detail on generator assumptions see the
NPCC website.6 In addition, transmission cost
estimates used to set the market price forecast
are based on research from regional transmission
studies such as those prepared by the Rocky
Mountain Area Transmission Study7 and the
Northwest Transmission Assessment Committee.8
Regional transmission assumptions were derived
from a working forum of utility experts, merchant
6 http://www.nwcouncil.org
7 http://psc.state.wy.us/htdocs/subregional/home.htm
8 http://www.nwpp.org/ntac/
Figure 5.14: Effi cient Frontier Versus Capital Expenditure ($millions)
1,450
1,500
1,550
1,600
1,650
1,700
1,750
1,800
1,850
6.
5
%
7.
0
%
7.
5
%
8.
0
%
8.
5
%
9.
0
%
9.
5
%
10
.
0
%
10
.
5
%
11
.
0
%
11
.
5
%
12
.
0
%
Covariance (stdev/mean)
Po
w
e
r

S
u
p
p
l
y

E
x
p
e
n
s
(2
0
-
y
e
a
r

N
P
V
)
0
150
300
450
600
750
900
1,050
1,200
Ca
p
i
t
a
l

C
o
s
t
Power Supply Expense
Capital Cost
5-24
plant developers, BPA, Western Area Power
Adminstration and other interested parties.
The Avista Transmission Department developed
estimates to provide a better understanding of
the transmission costs associated with building
generators to serve Avista’s native load. These cost
estimates may be found in Table 4.1 in Section
4-Transmission Planning. Resource options
available to serve future demand in the West, as well
as those available to meet resource defi cits that face
Avista in the future, are discussed below.
Combined-Cycle Combustion Turbines (CCCT)
Combined-cycle combustion turbines were
modeled using a two-on-one (2x1) confi guration.
This confi guration consists of two gas turbines
exhausting waste heat into a single heat recovery
steam generator (HRSG), rather than one gas
turbine matched to the HRSG as in the traditional
one-on-one confi guration. The NPCC assumes
that modest cost effi ciencies are gained through the
2x1 confi guration. All CCCT plants that could be
selected by the model have 610 MW of capacity;
540 MW is assumed to be base load and 70 MW
is duct fi re. The actual monthly capability of these
plants varies across regions based on NPCC
assumptions. The Company did not restrict the
quantity of CCCT plants built by AURORAXMP for
market forecasting.
Simple-Cycle Combustion Turbines (SCCT)
Two simple-cycle technologies were modeled,
aero-derivative and frame machines. These two
resources have a trade off between capital and
effi ciency. Aero-derivative machines are more
capital intensive and have higher operating costs
than frame machines, but their heat rate is lower and
the plants have more fl exible start times of 10 to 15
minutes. The Company did not restrict the quantity
of SCCT plants that could be built by AURORAXMP
for market forecasting purposes.
Coal Plants
Three types of coal technologies were modeled
for the 2005 IRP, pulverized, integrated gasifi cation
combined cycle (IGCC), and IGCC with carbon
sequestration.
Pulverized: Sub-critical pulverized plants were
modeled with low-NOX burners, nitrogen oxide and
mercury controls. Capital cost estimates assume
that more than one unit will be built on a site and
that the plant is wet cooled. Air-cooled plants
reduce thermal effi ciency by about 10 percent.
IGCC: Integrated Gasifi cation Combined Cycle
plants convert coal into a gas, and then burn it using
technology similar to CCCT plants. IGCC plants
signifi cantly reduce emission levels through the
gasifi cation process. These plants are expected
to cost 15 to 20 percent more than their pulverized
counterparts, but they offer greater effi ciency and
the opportunity to sequester carbon emissions.
IGCC with Carbon Sequestration:These plants use
the same technology as the IGCC plant, except
that carbon is captured and sequestered into deep
geological pockets in the earth or in the ocean.
ICGG with sequestration has the potential to capture
approximately 90 percent of plant carbon emissions.
5-25
As constraints prevented new coal plants from
being built in California, coal plants built to serve
the Golden state are constructed in Wyoming or
Utah, and power is transmitted on new or upgraded
transmission lines. Other regions, including Idaho,
the Northwest and Utah, have the option to build
coal plants locally or construct plants in other
states (e.g., Wyoming or Montana) and transmit the
power over transmission lines. Colorado, Northern
Nevada, Arizona and New Mexico built coal plants
locally. The Northwest was limited to two pulverized
and fi ve IGCC coal units. The Company did not
include any clean coal tax benefi ts that may become
available by pending federal legislation or a carbon
tax adder for the Base Case. Since no federal law
limits carbon emissions, the Company chose not
to include any carbon tax in the Base Case, though
the Company will continue to study potential carbon
taxes for future resource acquisitions. The 2005
IRP includes two carbon-limited scenarios to help
understand the potential impacts of such a tax.
Wind
Improving wind modeling has been a focus for this
resource plan. The major improvement is the use of
hourly generation shapes rather than a fl at monthly
capacity factor. Consistent with the NPCC, wind
plants are assumed to require new transmission
with the exception of the fi rst 1,000 MW of wind
generation added to the Northwest Region. It is
also assumed that the capacity factor for wind
will fall after the fi rst 1,000 MW are installed. For
example, the fi rst sites (Tier 1) have 33 to 37 percent
capacity factors. Tier 2 are assumed to be 80
percent of Tier 1 potential. Furthermore, Tier 1 sites
can be integrated without signifi cant transmission
construction, whereas Tier 2 sites require new
transmission construction.
Alberta Oil Sands
The oil sands of Northern Alberta are often called
“tar sands.” According to the NPCC Fifth Power
Plan, the oil sands have an estimated 1.6 trillion
barrels of petroleum deposits. The petroleum is
in the form of bitumen and methane contained
in the sands. The bitumen can be extracted
and processed to create synthetic crude oil.
The process used to extract the bitumen uses
steam produced from natural gas or coke plants.
Developers of the oil sands would like to build
additional co-generating natural gas plants near
Fort McMurray, Alberta, to process the bitumen,
and then sell the electric byproduct to markets in
the Northwest and/or California. This plan requires
the costly construction of new high-voltage DC
transmission lines to reach U.S. markets.
Nuclear
Nuclear-powered generation is a prominent energy
source throughout the world and the United States
(approximately 20 percent of generation in the U.S.).
The U.S. has not licensed a new plant since 1978.
Based on new cost data provided by the NPCC,
nuclear generation appears competitive compared
to coal. Great uncertainty remains with new nuclear
plants because of concerns with plant siting,
historical issues with cost overruns, and long-term
waste storage. Future carbon emission standards,
5-26
potential tax benefi ts and federal loan guarantees
could make nuclear power more economically
attractive in the future. Given uncertainties
surrounding nuclear plants, the model was allowed
to construct only one plant in Arizona after 2020.
The option was provided only to test the relative
economics of the nuclear option.
Other Resources
Several other resources also were modeled in this
study. The resources are relatively small compared
to the other options described by the NPCC, and
were limited based on location. Solar projects were
limited only to the southern states. Wood, landfi ll
gas and manure biomass were only allowed in the
Northwest to simplify the modeling exercise. Table
5.8 provides a brief description of each technology
type and key underlying assumptions. The resource
assumptions were taken from the NPCC except
where noted. Capital is shown as overnight cost,
meaning that allowances for funds used during
construction are not included.
Unit availability accounts for both maintenance
and forced outage and is based on NPPC
assumptions. Wind plant availability varies by
region and season; on average, wind plants are
modeled with a 34 percent capacity factor. Solar
is shaped by hour over the year with an average
availability of 22 percent.
Resource
Gen. Cost
($/kW)
Unit
Capacity
(MW)
Heat Rate
(Btu/kWh)
Unit
Availability
(percent)
Fixed
O&M
($/kW/yr)
Variable
O&M
($/MWh)
CCCT/Duct (2x1) 567 540/70 7,030 90 8.76 3.02
SCCT- Aero 648 94 9,900 90 8.64 8.64
SCCT- Frame 405 94 10,500 90 6.48 4.32
Coal- Standard 1,343 400 9,550 84 43.19 1.89
Coal- IGCC 1,512 425 7,915 83 48.59 1.62
Coal- IGCC w/seq 1,949 401 9,290 83 57.23 1.73
Wind 1,191 100 N/A Avg. 34 18.90 1.08
Geothermal 1,976 50 9,300 92 103.66 0
Adv. Nuclear 1,566 1,100 9,600 88 43.19 1.08
Solar 7,558 2 N/A Avg. 22 34.55 4.32
Oil Sands 611 180 5,800 85 0 3.00
Landfi ll Gas 1,468 1 11,100 80 134.97 1.08
Manure 3,347 1 11,100 90 72.34 0
Wood 2,159 25 14,500 90 86.38 9.72
Co-Gen 1,080 25 5,500 85 31.31 2.16
Table 5.8: New Resource Alternatives (2005$)
5-27
Heat rates for CCCT, SCCT, and coal plants are
expected to improve over time. For example, the
NPCC assumes that CCCT heat rates will improve
by 10 percent from an average of 7,030 Btu per
kWh today to 6,359 Btu per kWh in 2026. Coal
plant heat rates are expected to improve by 6
percent over the same period.
Fixed O&M fi gures include maintenance and
transmission costs of $15 per kW-year, except for
SCCT plants where non-fi rm transmission service
is assumed. These assumptions are based on
NPCC datasets.
Certain resources benefi t from capital cost
reductions over time due to anticipated technology
improvements. These reductions are shown in
Table 5.9
Resources Not Evaluated for the
Western Interconnect
There are many resources that could be a vital part
of an energy future but were not modeled in this
analysis because of problems with commercial
viability at this point in time. The resource types
include nuclear, 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
model was allowed to build a single nuclear plant in
Arizona, with the assumption that a new unit could
be added to the Palo Verde generation station after
2020. The lone facility was included to show the
potential for cost effective nuclear power provided
that safety, security, waste storage and political
issues can be rectifi ed in the future.
Resources Not Evaluated for the Northwest
Table 5.8 includes many resources that were not
included in the Western Interconnect. Large-scale
solar, nuclear and coal with carbon sequestration
most likely will not be constructed in the Northwest
because of cost, siting or other concerns.
Western Interconnect Generic Transmission
Cost Estimates
New resources built far from load centers, such
as coal and oil sands, will require transmission
investments. Cost adders to account for
transmission were included in the IRP analysis. In
the Northwest, several resources are available that
would require new transmission:
• New coal plants located in the
Oregon/Washington region
• New wind farms located in Oregon/Washington
that are in excess of the 1,000 MW of Tier 1 wind
• New coal plants located outside the Northwest
• New wind farms located outside the Northwest
• Oil sands located in Northern Alberta
Resource
Type
2007-
2009
2010-
2014
2015-
2026
Wind 3.1 2.3 1.9
Solar 8.0 8.0 8.0
CCCT - Gas 0.5 0.5 0.5
CT - Gas 0.5 0.5 0.5
IGCC 1.5 1.5 1.5
Table 5.9: Forecast Capital Cost Reductions (%)
5-28
Other regions, including California and Southern
Nevada, also may import coal generation from
other regions or oil sands generation from Alberta.
In these cases, new transmission costs are
included in the delivered price of energy.
Transmission estimates generally are based on
fi gures from regional transmission studies, such
as the Rocky Mountain Area Transmission Study
and the Northwest Transmission Assessment
Committee. The values used for the 2005 IRP are
shown in Table 5.10. These studies provide rough
estimates for specifi c lines of various sizes and
locations. Transmission estimates use approximate
mileage of the new transmission lines, the size
required for the connecting plant and the locations
of the new line. Specifi c transmission costs for
Avista resource options are provided in Section 4-
Transmission Planning.
Levelized Costs
Figure 5.15 is a graphic showing levelized costs
of each resource alternative assuming full plant
capability. At full capability, certain Tier 1 wind and
Montana coal are still the lowest cost resources,
excluding any carbon mitigation costs. Nuclear is
also a low cost resource based on cost estimates
provided by the NPCC. Renewable resources are
competitive when the federal the production tax
credit is applied. Similar values in real levelized
dollars are presented in Apendix H.
Resource
Type To From
Line
Size
(KV)
Capacity
(MW) Miles
Cost
Per Mile
($Mil)
Substation
Costs
($Mil)
Total
Cost
($Mil) ($/kW)
Fixed
O&M
(kW/yr)
Coal Inter-regional 500 1,200 300 1.20 40 400 333 8.9
Wind Inter-regional 500 1,200 100 0.90 40 130 108 8.9
Coal OWI MT 500 1,200 672 0.85 50 621 518 8.9
Wind OWI MT 500 1,200 600 0.85 50 560 467 8.9
Coal IDs WY 500 1,200 450 1.20 10 550 458 8.9
Coal UT WY 500 1,200 200 1.50 20 320 267 8.9
Coal UT UT 500 1,200 100 1.20 15 135 113 8.9
Coal SCAL WY 500 1,200 1,500 1.80 100 2,800 2,333 8.9
Coal NCAL WY 500 1,200 1,600 1.80 100 2,980 2,483 8.9
Coal SNV WY 500 1,200 1,100 1.70 100 1,970 1,642 8.9
Oil Sands OWI AB 500 DC 1,500 1,200 N/A N/A 1,400 933 8.9
Oil Sands SCAL AB 500 DC 2,000 1,730 N/A N/A 2,000 1,000 8.9
Oil Sands AB AB 500 DC 500 475 N/A N/A 500 1,000 8.9
Gas/Other Inter-regional N/A N/A N/A N/A N/A N/A N/A 16.8
Table 5.10: Regional Transmission Cost Estimates (2005$)
5-29
Figure 5.15: 2016 Resource Option Costs (2005$/MWh, Levelized)
5-30
5.6 Wind Modeling
Avista has a distinguished history of using renewable
energy to serve its customers. Hydroelectric and
wood-fi red generation currently accounts for more
than half of all electricity consumed in our service
territory. Wind power presents an avenue for the
Company to generate more renewable energy.
Wind resources benefi t from having no fuel costs
and low operations and maintenance costs
when compared to other renewable generation
technologies. Wind power has similar fi nancial
benefi ts to traditional generation facilities like coal
and nuclear plants, as well as other renewable
facilities, because its costs are not highly correlated
to the wholesale electricity marketplace.
The 2003 IRP identifi ed 75 MW of wind as a part of
the Preferred Resource Strategy. The Action Plan
for the 2003 IRP also committed the Company to
study wind generation further. Over the past two
years the Company has researched wind generation
and the potential fi nancial and operational impacts
of wind integration. In early 2004, the Company
signed a ten-year wind power contract for 35
MW of installed capacity from the Stateline Wind
Energy Center. The contract is for busbar (i.e.,
delivered when the wind blows) power, this allows
the Company to experience and evaluate the
actual impacts of a wind resource on its system.
It also provided actual data to assist in evaluating
wind, including access to a state-of-the-art wind
generation forecasting package.
Wind Integration
Wind integration entails costs associated with
fi rming and shaping the resource to meet customer
needs. Wind energy is “controlled” by weather
patterns rather than utility operators, thereby
creating a generation resource that is signifi cantly
different than traditional types. To integrate wind,
other resources must be dispatched in a different
and often costlier manner. Wind behaves more
like a load, because it requires other resources
to follow its intermittent output. This impacts the
opportunity costs of operating non-wind generating
assets differently than they would be absent wind
generation.9 This higher incurred cost is attributed
to the wind resource and charged against its
generation value.
Various studies have been performed to address
wind integration costs. Actual integration costs
have been estimated from less than one dollar
per MWh to more than $20 per MWh. Company
studies have shown that integration costs can
range upward of $20 per MWh where penetrations
exceed 20 percent of total system installed
capability. A wind integration model developed by
the Company showed that modest levels of wind
installation, around 50 MW, were expected to incur
integration costs below $3 per MWh. Levels near
100 MW incurred costs closer to $5 per MWh. The
model showed that both system capabilities and
9 Opportunity cost is the cost of an item in terms of the next best-
forgone alternative. In the case of wind power, an alternative asset, such
as a hydroelectric project must be taken off of optimal economic dispatch
and be used to shape the non-fi rm power coming from the wind project.
5-31
transmission location and costs affect the actual
level of wind integration cost. For the 2005 IRP,
the Company used wind integration assumptions
from the NPPC. The NPPC assumes integration
costs of $4.50 (2005$) per MWh for Tier 1 wind
resources and $9.00 (2005$) per MWh for Tier 2
wind resources.
Wind Forecasting
Many in the wind industry tout the ability of wind
forecasting to reduce wind integration costs. If
wind generation could be forecast on a day-ahead
basis with higher levels of accuracy, then the
resource would become more reliable and valuable.
Preliminary Company analysis found that wind
forecasting does not enhance the ability to forecast
wind generation; therefore wind integration costs
cannot presently be reduced substantially through
forecasting. Ideally, a forecast would provide
accurate information about wind generation for the
period four hours or more into the future so that
utility operations can be modifi ed to accommodate
the wind energy. In its study, the Company
found that the results of a third-party weather
forecast were not superior to simple persistence
forecasting.10 Table 5.11 compares the accuracy of
wind forecasting methods to persistence over the
July 2004 through March 2005 time period for the
Stateline Wind Energy Center. Avista believes that
there is room for improvement in wind forecasting
and is analyzing additional data. It is hoped that
wind-forecasting methods can be improved to bring
down wind integration costs.
Wind Contribution to Meeting System Peaks
The Company must own or control adequate
resources to serve customer loads during adverse
weather or other events. The Company’s last
IRP gave wind resources a zero value for meeting
system peaks because of the erratic nature of
wind and our lack of experience with the resource.
Using data from the Oregon State University Wind
Research Project, Avista was able to estimate the
contribution of wind resources to meeting system
peaks. The evaluation of individual sites, such
as Stateline, supports our 2003 IRP assumption
that wind does not possess signifi cant capacity
value. An analysis of wind sites located across
the Northwest showed that a portion of installed
capability could be relied on to meet system peak.
Using a method called “Energy Load Carrying
Capability,” the Company found that a mix of fi ve
Northwest sites scattered from the Oregon Coast
to the eastern side of the Rocky Mountains could
support a capacity level of approximately 25
percent. 11 The level appears low in relation to a
Hours
Ahead
Weather
Forecast
Simple
Persistence Difference
1 94.9 95.6 -0.7
2 95.0 88.9 6.1
4 90.0 76.2 13.8
8 42.0 57.4 -15.4
12 32.4 44.3 -11.9
24 19.6 32.3 -12.7
48 11.8 11.4 0.4
Table 5.11: Wind Forecasting Accuracy (%)
10 Persistence forecasting assumes that the last hour of generation will
represent future hours of generation.
11 Energy load carrying capability represents the expected portion of
nameplate capacity available during a system’s peak demand.
5-32
plant’s nameplate rating, but not when compared
to the expected capacity factor. The 25 percent
peak capacity value is used when evaluating wind
contribution to meeting system coincident peaks for
the 2005 IRP.
Wind In The Preferred Resource
Strategy
Cost
Wind energy costs are driven by several factors: the
capacity factor of the wind site, availability of the
federal PTC, integration costs, and transmission
costs to deliver wind energy to customers. Capacity
factor accounts for the largest difference between
wind site values. Changing from a 33 percent
capacity factor to 25 percent equates to an energy
loss of around 25 percent. The 2005 IRP assumes
that two tiers of wind energy exist in the Northwest
and in eastern Montana: Tier 1 equals 33 percent in
the Northwest and 35 percent in eastern Montana;
Tier 2 equals 26.4 percent in the Northwest and 28
percent in eastern Montana. Montana wind sites
have a higher wind capacity factor, which explains
their modestly higher generation levels. Tier 2 wind
levels are estimated at 80 percent of Tier 1 levels.
The lower generation level, while based on limited
information, refl ects data obtained from various
sources over the past few years. The reference
to eastern Montana in the IRP is for illustrative
purposes only. There are various regions remote
to Avista (including eastern Montana) that have
better wind patterns than found in the Northwest.
This plan does not preclude the Company from
purchasing wind energy from these sites.
The federal PTC plays a signifi cant role in wind
project economics. To determine the importance
of the PTC, a Base Case scenario using a 50/50
weighting of cost, measured as the net present
value of expected power supply expenses, and
risk, measured as the standard deviation of the
expected power supply expense, was run absent
the PTC. The result showed that no Tier 2 wind
resources were selected, which reduced overall
wind penetration by one-third. The Base Case runs,
and most scenarios, assume the PTC remains at
its 2005 level through the 20-year study. The PTC
might be eliminated or modifi ed, but Avista believes
that the PTC is a good alternative to a carbon-based
fee, and it likely will remain absent carbon legislation.
However, the PTC is phased out in scenarios where
carbon emissions are regulated.
Availability of Wind Generation
The 2003 IRP limited total installed wind generation
to 75 MW of installed capability. The PRS would
have selected more wind without the constraint.
The reasoning for the limitation was based on many
perspectives at that time, including:
1. The Company had limited experience with wind;
2. The two-year planning cycle allowed for later
revisions to the estimate without compromising
the Company’s future resource mix because no
new resources were required before 2008;
3. Avista’s analysis of integration costs found that
higher wind penetration resulted in higher costs
than assumed by the resource selection model;
4. Signifi cant modeling changes for the 2003 IRP
precluded the Company from fully addressing
5-33
capacity planning and therefore was cautious
about selecting resources with low abilities to
contribute to system peaks;
5. Based on preliminary work, the capacity
contribution to system peak was assumed
to be zero, which compromised the value of
wind generation to the Company.
The 2005 IRP analyses benefi t from substantially
more information than was available for the 2003
effort. Studies have shown that wind integration
costs are more manageable than forecast at that
time. The NPCC has evaluated wind integration and
the costs have been included in the present analysis.
Results of the 2005 IRP indicate that substantial
amounts of wind would be cost-effective within
certain limits. The NPCC Fifth Power Plan discusses
a potential of 5,000 MW of installed wind capacity
in the Northwest. Avista’s pro-rata share of wind
generation would be approximately 250 MW.
The Company has determined that it makes sense
to limit the overall level of wind energy within Avista’s
resource portfolio due to our concern over adequate
levels being available to serve our requirements. To
enforce the limit, the Avista Linear Programming
model allows 250 MW of wind generation from
the Northwest, plus 150 MW of wind capability
in Avista’s own service territory over the 20-year
study. Two hundred fi fty additional megawatts are
assumed to be available to Avista from outside
the Northwest (e.g., eastern Montana). The total
potential wind resource available to Avista is 650
MW over the 20-year IRP timeframe.
5.7 Summary
The 2005 Integrated Resource Plan is a
comprehensive modeling effort that not only studies
Avista’s generation needs but also those of the entire
Western Interconnect. The modeling approach
allows the Company to identify costs and benefi ts
of large changes to the electric industry, such as fuel
price volatility, carbon emission standards and lower
future hydro energy.
The modeling approach relies heavily on estimates
provided by the NPCC Fifth Power Plan and the
resource database provided by EPIS, Inc. The
Company’s approach differs from other integrated
resource plans by deriving price forecasts from the
same model that evaluates resource option values
instead of simply inputting electricity prices into
the study as exogenous variables. This approach
more fully accounts for changes made to the input
assumptions and eliminates the need to make
assumptions about the correlations and statistics
of and between natural gas and electricity
prices. This creates a fully integrated generation
evaluation of how Company resources would act
in the marketplace.
5-34
6-1
Avista’s Preferred Resource Strategy (PRS) is tied
to the Mid-Columbia electric market forecast more
than to any other variable. The Modeling Approach
section describes four major drivers of market
prices: natural gas prices, electricity demand, hydro
generation levels and wind generation levels. There
are several ways to evaluate future market prices
with AURORAXMP. The most common approach is
to forecast the market using averages.
In this case, average hydro conditions, base line
fuel prices, average wind conditions, average load
projections and other variables discussed in the
key assumptions portion of Section 5- Modeling
Approach are input into the model. The Company
used this approach to develop the Base Case
electricity forecast.
After the Base Case forecast was completed, two
methodologies for risk assessment were utilized:
Futures: stochastic studies that use a Monte Carlo
approach to quantitatively assess the risk around an
expected mean outcome.1 This time-intensive and
multi-variable approach is the most robust method
used for risk assessment.
Section Highlights
 Gas-fi red resources continue to serve the majority of new loads in the West through the IRP
timeframe; however, load growth in Washington, Oregon and Northern Idaho is primarily served
by new wind and coal-fi red resources.
 Market prices are forecast to fall from today’s level through 2010, rising approximately with
infl ation thereafter.
 Electricity and natural gas prices are expected to remain highly correlated in the future.
 The IRP analyses are based on more than 300 gigabytes of data generated by 24 computers
running continuously for nine days.
 The 2005 IRP modeled 18 unique market scenarios.
 Utility avoided costs are modestly higher than the electricity market price forecast because of
resources built to support planning margins.
6. MODELING RESULTS
1 A stochastic study is a statistical approach that uses probability
distributions to forecast the future.
6-2
Scenarios: deterministic studies that change one
signifi cant underlying assumption to assess the
impact of the change.2 This approach is easier to
understand and takes less time to prepare than a
future, but does not quantitatively assess risk.
This section is split into three parts: Base Case
results, futures results, and scenario results. It
discusses resources AURORAXMP built to serve
load growth in the Western Interconnect over the
next 20 years, the Northwest electric market price
forecast, and variables driving the results. All fi gures
representing prices are in nominal (i.e. not infl ation
adjusted) dollars unless otherwise stated.
6.1 Base Case
The Base Case is a deterministic study with a
baseline set of assumptions for each variable
entered into the AURORAXMP model. They are
described in the Key Assumption section of Section
5-Modeling Approach. This case also assumes
continued availability of the federal production
tax credit (PTC) for renewable resources and that
no carbon or greenhouse gas (GHG), emissions
legislation will be enacted. The PTC and GHG
legislation are interconnected because the PTC
provides a fi nancial incentive to build plants with
low or no GHG emissions. If carbon legislation
were enacted, the PTC for renewable resources
would most likely be terminated because the new
legislation would provide an incentive similar to
the present PTC. Two different GHG emissions
scenarios were developed for this IRP to provide
a better understanding of the fi nancial impacts of
potential emissions legislation. They are discussed
later in this section.
AURORAXMP builds future resources to serve regional
load growth based on construction costs, return
on capital, availability, and operation costs, before
it can create a price forecast. Understanding the
new resources built by AURORAXMP is the key to
understanding what drives future prices. The Base
Case price forecast includes two 400 MW coal units
for the Northwest in 2012; 500 MW of wind capacity
will be constructed in 2016 and 2017.
New resources shown in Figure 6.1 are primarily
natural gas-fi red. In addition to the gas plants,
the model built some coal and wind. Fixed RPS
resources account for 10 percent of new generation
capacity and 9 percent of total energy.
The model chose two coal plants outside of
their native load area, connected by new or
upgraded transmission facilities. The fi rst coal
plant, constructed in Utah, is wheeled to Southern
California via an upgrade to the IPP DC line that runs
from Intermountain, Utah to Adelanto, California.
The second is a new coal plant in Wyoming that
serves load in southern Idaho.
The large penetration of gas-fi red generation is
driven by the IRP assumption that no new coal-fi red
plants will be constructed within the state. Gas-fi red
generation, even with its higher fuel costs, is less
expensive than transporting coal-fi red generation
from other states such as Montana.2 A deterministic study assumes there is only one future and uses single
point estimates to determine it.
6-3
Figure 6.1: Cumulative Western Interconnect Resource Additions (GW)
-
20
40
60
80
100
120
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Pulverized
Nuclear
SCCT- Frame
CCCT
Wind
RPS-Other
RPS-Wind
Figure 6.2 details resource additions absent California.
This fi gure is presented to explain that while gas-fi red
generation is signifi cant across the West, absent
California its contribution is more modest.
Figure 6.3 shows that Base Case electric prices are
expected to fall between 2007 and 2010 in line with
falling natural gas prices. The large coal plant added
in 2012 will help keep prices relatively fl at in real-
dollar terms.
Northwest electricity prices in the future will be
highly correlated with the natural gas market.
Since market prices are set by the operating cost
of the resource that is on the margin, recently built
gas turbines will continue to set market prices.
Company analysis found that 79 percent of the time
electric market prices are correlated to natural gas
markets in the future. Figure 6.4 is a scatter plot
that shows the correlation between Malin natural
gas prices and Mid-Columbia electricity prices over
the IRP timeframe. Excluding the second quarter of
the year, when hydro contributes large amounts of
energy to the system, the correlation is 89 percent.
These correlations indicate that natural gas plants
will continue to set the marginal price of electricity,
and that the dependence on natural gas likely will
cause future market prices to be substantially higher
than historic levels.
Another useful price forecast statistic is what
resources are being used to serve loads. The Base
Case forecast uses existing resources, along with
new natural gas, coal, and renewables, to serve
electricity demand. Figure 6.5 illustrates resources
used to meet requirements over the forecast period.
6-4
Figure 6.3: Mid-Columbia Electric Price Forecast ($/MWh)
30
35
40
45
50
55
60
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Nominal Dollars
2005$
Figure 6.2: Cumulative Western Interconnect Resource Additions Absent California (GW)
0
10
20
30
40
50
60
70
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Pulverized
Nuclear
SCCT- Frame
CCCT
Wind
6-5
Figure 6.4: Malin Natural Gas and Mid-Columbia Electricity Correlation Plot
Figure 6.5: Western Interconnect Resource Contribution (%)
6-6
Table 6.1 presents annual average electricity
market price forecasts for each zone modeled
in AURORAXMP. Zone prices differ because of
transmission and congestion costs to move power
from one zone to another. Market prices also are
affected by natural gas transportation and delivery
costs across the Western Interconnect. Congestion
costs are economic costs derived when a transfer
line between two areas is fully utilized, effectively
closing the path between two areas. When the
path is closed, the higher-cost zone is islanded from
the rest of the system, and it must rely on higher
cost internal resources to meet load requirements.
Transportation costs are the physical cost or rent to
move power and natural gas from the supplier to the
end-use consumer.
Stochastic Results
Stochastic studies are necessary to understand
the probability, or risk, of an outcome. A
stochastic Base Case electric price forecast is
created with AURORAXMP by simulating the study
period multiple times and varying key input values
in each simulation.
Natural Gas prices, hydro conditions, load and
wind conditions were allowed to vary in each
study. Using different probability distributions,
200 random draws were transferred to a database
linked to the AURORAXMP model. Using 24
processors over four and a half days, the model
was run 200 times to create 200 unique price
forecasts in each stochastic study. The results
Table 6.1: Electric Market Prices By Western Interconnect Zone ($/MWh)
Area 2007 2008 2009 2010 2011 2012 2016 2020 2024 2026
AB 52.57 45.23 43.28 41.74 42.94 39.35 43.36 49.28 52.51 56.40
AZ 50.23 45.18 42.30 40.25 40.33 39.93 44.97 48.65 51.39 53.50
BAHA 53.51 49.99 46.84 44.66 45.37 45.75 50.26 54.64 59.24 61.47
BC 53.09 46.50 44.30 42.67 43.61 40.73 42.94 49.50 51.95 55.58
CCAL 53.23 48.73 45.94 43.99 44.31 43.92 48.80 53.61 57.07 60.32
CO 49.51 45.21 42.56 40.59 40.61 38.70 41.61 47.56 47.84 49.44
MT 49.01 44.62 42.05 40.09 40.53 38.21 29.86 38.40 41.11 39.37
NCAL 53.52 48.88 46.15 44.23 44.45 43.93 48.39 53.09 56.70 59.39
NM 48.90 44.34 41.73 39.71 39.86 39.24 44.34 48.21 50.27 52.94
NNV 50.90 45.96 43.29 41.24 41.68 40.19 43.58 49.43 53.18 55.69
OWI 50.35 46.12 43.60 41.69 42.16 40.47 44.05 50.20 53.83 57.22
SCAL 54.34 49.64 46.88 45.01 45.45 45.23 50.80 55.58 59.00 62.35
IDs 49.90 45.49 43.01 41.10 41.53 39.62 42.87 48.85 52.52 55.65
SNV 52.46 47.25 44.37 42.29 42.44 42.07 47.13 51.46 54.51 57.68
UT 49.48 45.00 42.39 40.38 40.73 38.91 41.69 47.19 50.37 53.55
WY 48.96 44.55 41.96 39.96 40.25 38.07 39.24 42.23 47.82 49.10
6-7
were queried from a single SQL Sever database
containing all iterations. The Avista LP model
quantifi ed the return and risk of each new resource
option available to Avista.
Two stochastic studies were completed for the
2005 IRP. The fi rst was the Base Case. It used
assumptions described in the Modeling Approach
section. The second used the same underlying
assumptions as the Base Case, but gas prices were
assumed to be twice as volatile. Both studies were
variances on the deterministic Base Case, and each
of their mean price forecasts is within 2.7 percent of
the deterministic Base Case study. See Figure 6.6.
Stochastic studies are necessary to quantify the
standard deviation, or risk, around the expected
outcome, or mean value. To quantify the standard
deviation at Mid-Columbia, the AURORAXMP model
was run 200 times. The risks surrounding average
expected market prices for the Base Case run are
shown in Figure 6.7. The solid line represents the
average market price, while the inner tick marks are
the 80 percent confi dence interval. This interval
describes the range within where 80 percent of
all observations lie. The outer tick marks are the
maximum and minimum average annual prices
observed in the study.
To approximate recent natural gas price volatility, the
natural gas price standard deviation was increased
from 50 to 100 percent of the mean in the Volatile
Gas Case. Mean market prices in this case are
similar when compared to the Base Case, but the
80 percent confi dence interval has a larger range, as
shown in Figure 6.8.
Figure 6.6: Mid-Columbia Electric Price Forecast Comparison ($/MWh)
6-8
Figure 6.7: Stochastic Base Case Mid-Columbia Electric Price Forecast ($/MWh)
25
35
45
55
65
75
85
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Max 80% CI High Mean 80% CI Low Min
Figure 6.8: Base Case and Volatile Gas Mid-Columbia Price Comparison ($/MWh)
6-9
Table 6.2 displays levelized cost data for each
resource option modeled in AURORAXMP at full
capability and at modeled operation levels, both
with and without the renewable production tax credit
levels. AURORAXMP calculated levelized costs using
the model’s expected dispatch levels to value each
resource; natural gas plants such as SCCT and
CCCT do not operate at full capacity. This skews
their levelized costs, since their fi xed costs are
levelized over the small number of operating hours.
Base load plants, including coal and nuclear, are
running at full capacity, and their levelized costs at
expected dispatch levels compare almost equally
to levelized costs at full output. Wind plants are not
dispatched by AURORAXMP, due to their very low
operating cost; their values under each method are
equal.
Table 6.2 illustrates plant costs, but it does not
detail risks inherent to them. Figure 6.9 allows
both cost and risk to be evaluated in one view. The
fi gure compares the average cost of each resource
necessary to acquire one average megawatt of
electricity over a year. For example, a wind plant
produces one-third of a megawatt of energy for each
megawatt of installed capacity. Three megawatts
of wind capacity are necessary to average one
megawatt of energy. A coal plant with an 85 percent
capacity factor is assumed to require approximately
1.2 megawatts to generate one average megawatt
of energy. Only natural gas-fi red resources, due to
their high capacity factors, are not scaled up.
Costs in Figure 6.9 are defi ned as all fi xed and
variable operation and maintenance costs plus
fuel and capital recovery. Risk is measured as the
variation around the expected average value of
these costs over the 200 Monte Carlo iterations.
The fi gure accounts only for operational risks from
changing fuel and market prices. Other risks,
such as nuclear waste disposal or construction
cost overruns for new coal or nuclear plants, are
not accounted for in this view. These risks are
quantitatively addressed in the selection of the PRS.
Higher-cost resources are shown in the upper
regions of Figure 6.9. The plant costs assume
“economic” dispatch of each resource type, with
market purchases replacing operating costs during
times where the resources are not running. The
horizontal axis represents risk, with higher risk
resources landing to the right side of the fi gure. Risk
is derived from fuel cost variation, such as gas price
volatility and wind speed variations.
In the Base Case, coal and renewable resources
such as manure and geothermal provide the most
risk protection, though their costs are somewhat
higher than other alternatives. Wind is one of the
lowest cost resources when the federal production
tax credit is accounted for. Wind power has fuel risk
because of the variation of wind or weather. Wind
variation can be avoided to some extent if a utility
purchases portions of several wind sites across
the Northwest to create a diversifi ed portfolio. Like
equity portfolios, wind diversifi cation adds cost,
as seen in Figure 6.9. Gas-dependent resources
such as CCCT, SCCT, and co-generation can have
low cost; however, correlation to electric and gas
markets results in riskier returns.
6-10
Table 6.2: 2007 Resource Option Costs (2005$/MWh)
AURORAXMP Full Output
w/ PTC W/O PTC w/ PTC W/O PTC
Coal Pulv MT 50.13 50.13 50.07 50.07
Nuclear 52.77 52.77 52.80 52.80
Coal Pulv OWI 53.81 53.81 53.44 53.44
Wind- Kennewick Tier 1 55.88 68.63 55.88 68.63
Coal IGCC MT 57.56 57.56 57.54 57.54
Coal IGCC OWI 59.85 59.85 59.61 59.61
Local Manure 61.40 65.22 61.44 65.27
Wind- OWI Tier 1 62.06 76.40 62.06 76.40
Local Landfi ll Gas 63.14 66.94 63.17 66.98
Wind- MT Tier 1 63.23 75.46 63.23 75.46
Local Co-Gen 63.48 63.48 60.34 60.34
Wind- Browning Depot 1 64.85 78.36 64.85 78.36
Geothermal 65.55 73.55 65.58 73.58
Coal IGCC SQ MT 66.88 66.88 66.83 66.83
Manure 67.60 71.42 67.64 71.47
Landfi ll Gas 70.12 73.92 70.15 73.96
Co-Gen 70.63 68.15 66.54 64.40
Wind- OWI Tier 2 70.78 82.15 70.78 82.15
Wind- MT Tier 2 71.32 80.99 71.32 80.99
Local Wind 71.90 84.46 71.90 84.46
Alberta’s Oil Sands 75.35 75.35 75.37 75.37
Wind- Kennewick Tier 2 76.95 89.52 76.95 89.52
Wind- Browning Depot 2 80.94 93.18 80.94 93.18
CCCT (2x1) 100.09 100.09 72.45 72.45
Local Wood 152.09 160.97 89.27 93.10
Wood 166.48 175.36 95.47 99.30
SCCT- Frame 3,534.29 3,534.29 79.55 79.55
SCCT- Aero 6, 337.13 6,337.13 80.62 80.62
6-11
6.2 Scenarios
Scenarios are non-stochastically modeled futures
that rely on average hydro generation, wind
generation, natural gas prices and load conditions
with a single signifi cant change to the future. This
type of analysis is performed to better understand
the impact of a fundamental change to one of the
Base Case assumptions. Scenario analysis allows
for quicker solutions, and the results are easier to
understand. The major disadvantage with scenarios
is their inability to quantitatively assess market
volatility risks.
Some scenarios are calculated using AURORAXMP
because the entire Western Interconnect
marketplace is affected. Other scenarios are more
easily and quickly solved outside of the AURORAXMP
model because the change only impacts the
Company’s resource portfolio.
Fuel Risk Scenarios
One of the biggest unknown variables in the future
is the price of fuel. Whether the fuel is coal, natural
gas, uranium, or manure, the price paid will depend
on supply and demand for the fuel. In the 2005 IRP
the Company chose to test natural gas prices under
high and low price scenarios to understand how
the Preferred Resource Strategy stands up against
natural gas variation.
Coal may be an option for the Northwest in the
future, and it is part of many resource plans across
Figure 6.9: Resource Cost and Resource Risk Comparison
RISK
TO
T
A
L

C
O
S
T
Pulverized
Coal
Nuclear
IGCC CoalLandfill
Gas
Geo-
thermal
Mt. IGCC
Coal
w/ seq.
Manure
Wood
CCCT
SCCT- Aero
SCCT-
Frame
Co-Gen
Alberta's
Oil Sands
Kennewick
Wind Tier 2
Browning
Depot Wind
Kennewick
Wind Tier 1Diversified
Wind Tier 1
Local Wind
Low Risk - Low Cost
High Risk - High CostLow Risk - High Cost
High Risk - Low Cost
Local
Co-Gen
Diversified
Wind Tier 2
6-12
the west. The cost of the abundant resource
has been stable or declining for several years.
A scenario was tested to see how the electric
market and Preferred Resource Strategy would
be affected if coal prices started to rise rather than
follow historical patterns. Other fuels such as
wood, manure and refuse have price risk, but the
small overall contribution of these resources when
compared to the entire market limits their impact on
market prices.
High Gas
The High Gas scenario was designed to understand
the market impacts of a permanent increase in
natural gas prices. This scenario started with
Base Case gas price assumptions and increased
price by 50 percent. The Company expected this
scenario to push some natural gas projects further
from economic viability and increase the viability of
alternative resources such as wind and coal.
Figure 6.10 shows that if high natural gas prices
were expected to persist, fewer natural gas
resources would be built when compared to the
Base Case. Natural gas would be replaced with
additional wind, coal, and geothermal resources. In
the High Gas case, AURORAXMP built a substantial
amount of coal across the West, as well as a
signifi cant amount of new transmission to wheel
power into southern California and Nevada.
In 2012, large quantities of Rocky Mountain coal
come online to serve West Coast load centers.
Electric prices are driven down to within 20 percent
of Base Case levels. This scenario provides a
glimpse of how the region might respond to a
permanent increase in overall natural gas prices.
Figure 6.11 shows the electric market price forecast
resulting from a 50 percent increase in natural
gas prices.
Figure 6.10: Cumulative Western Interconnect Resource Additions–High Gas (GW)
-
20
40
60
80
100
120
140
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Manure
Solar
Geothermal
IGCC
Pulverized
Nuclear
SCCT- Frame
CCCT
Wind
RPS-Other
RPS-Wind
6-13
Low Gas
The Low Gas scenario was designed to refl ect changes
in the market resulting from a permanent decrease
in natural gas prices. This scenario started with
Base Case natrual gas price assumptions and then
decreased the cost of natural gas by 50 percent. The
Company expected this scenario to maintain current
practice and build only new CCCT projects. Figure
6.12 shows the annual resources that were built
across the west in the Low Gas scenario. Natural
gas resources were built exclusively except for RPS
resources required by some states in the Western
Interconnect. Low natural gas prices contribute to
low market prices when compared to the Base Case,
as shown in Figure 6.13.
High (Doubled) Coal Price Escalation
The High Coal Price Escalation scenario is designed
to show the possible impact of higher coal prices
over time. This particular scenario doubles coal
price escalation as a response to increased demand
for coal in the West. Figure 6.14 shows the price of
coal in the Base Case, and in the High Coal Price
Escalation scenario.
Where coal prices dramatically increase in the future,
the likely result will be fewer new coal plants. The
scenario results show a 35 percent reduction in new
coal plant construction in the West when compared
with the Base Case under the High Coal Price
Escalation scenario. Figure 6.15 shows the resource
mix with higher coal price escalation.
Higher coal prices result in slightly more expensive
market prices when compared to the Base Case.
Figure 6.16 provides an electric price forecast
comparison to the Base Case.
Figure 6.11: Base Case and High Gas Mid-Columbia Electric Price Forecasts ($/MWh)
40
45
50
55
60
65
70
75
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
High Gas
Base Case
6-14
Figure 6.12: Cumulative Resource Selection for the Western Interconnect–Low Gas (GW)
-
20
40
60
80
100
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
SCCT- Frame
CCCT
RPS-Other
RPS-Wind
Figure 6.13: Base Case and Low Gas Mid-Columbia Electric Price Forecasts ($/MWh)
20
25
30
35
40
45
50
55
60
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Base Case
Low Gas
6-15
Figure 6.14: Base Case and High Coal Price Escalation Coal Price Forecasts-
Montana Mine Mouth ($/dth)
0.60
0.70
0.80
0.90
1.00
1.10
1.20
1.30
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
2x Coal Price Escalation
Base Case
Figure 6.15: Cumulative Resource Selection for the Western Interconnect–
High Coal Price Escalation (GW)
-
20
40
60
80
100
120
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Pulverized
Nuclear
SCCT- Frame
CCCT
Wind
RPS-Other
RPS-Wind
6-16
Figure 6.16: Base Case and High Coal Price Escalation Mid-Columbia Electric Price Forecasts ($/MWh)
40
45
50
55
60
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Base Case
High Coal Esc
Figure 6.17: Cumulative Resource Selection for the Western Interconnect–No Capacity Credit (GW)
-
10
20
30
40
50
60
70
80
90
100
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
IGCC
Pulverized
Nuclear
CCCT
Wind
RPS-Other
RPS-Wind
6-17
Market Structure Scenarios
Though the AURORAXMP model makes sound
economic decisions for the marketplace under
assumptions derived by the Company and its
Technical Advisory Committee, some market drivers
and assumptions are only estimates of possible
futures. Market structure scenarios target macro
changes to the electric market, including low capacity
planning margins, federal or state legislation capping
carbon emissions, more effi cient transmission
construction, climate change forcing long-term hydro
conditions down, Northwest wind-heavy construction,
and companies following a boom-bust build cycle
similar to the 1998-2001 time period.
No Capacity Credit
Capacity credits are a fi nancial incentive for the
model to build more generation than is needed to
serve forecasted load under average conditions.
This is similar to building a planning margin into
utility resource portfolios. Excess capacity stabilizes
prices even in cases of extended outage or spiked
demand. Extra capacity results in slightly higher
average costs, but it spares customers from large
price swings. Removing the capacity credit also
provides an estimation of avoided costs for a utility.
Figure 6.17 shows the results of this scenario.
The model builds approximately 23 GW less
capacity than in the Base Case. This scenario
illustrates what could happen in a marketplace if
utilities had no incentive to build planning margins
into their forecasts. Analysis shows that fewer
resources result in greater market volatility,
higher prices and greater price risk as a result of
extended shortages. Figure 6.18 shows that Mid-
Columbia market prices are higher than in the Base
Case in most years, and that prices are more volatile
from year to year.
Figure 6.18: Base Case and No Capacity Credit Mid-Columbia Electric Price Forecasts ($/MWh)
35
40
45
50
55
60
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Base Case
No Capacity Credit
6-18
Figure 6.19: Cumulative Resource Selection for the Western Interconnect –
30% Lower Transmission Capital Cost (GW)
-
20
40
60
80
100
120
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
IGCC
Pulverized
Nuclear
SCCT- Frame
CCCT
Wind
RPS-Other
RPS-Wind
Figure 6.20: Base Case and 30% Lower Transmission Capital Cost Mid-Columbia
Electric Price Forecasts ($/MWh)
35
40
45
50
55
60
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Base Case
Low Tx Capital
6-19
30 Percent Lower Transmission Capital Costs
The 30 Percent Lower Transmission Capital Costs
scenario assumes a 30 percent reduction in
transmission construction costs due to possible
effi ciencies gained from a regional approach to
transmission siting. This scenario benefi ts capital-
intensive resources like wind and coal. Providing
a lower cost transmission scenario enables the
Company to see how resource selections will
change under a range of transmission costs.
With lower transmission capital costs, the
expectation was that AURORAXMP would build
additional wind and coal units. The model built 21
percent more wind capacity, 42 percent more coal
capacity and decreased gas construction by nine
percent. Figure 6.19 shows the annual resource
builds for this scenario.
The model built more coal outside native load regions,
including a plant in Montana to serve the Northwest,
and several plants in Wyoming to serve Utah and
southern Idaho. The scenario even allowed construction
of a new IGCC coal plant in the Northwest.
The Mid-Columbia price results of this scenario are
shown in Figure 6.20. Though market prices did
not drop substantially, total fuel costs across the
Western Interconnect dropped by an average of 2.8
percent, or $563 million annually, because of the
switch from natural gas to coal and wind.
The results of this scenario explain that transmission
costs are a barrier to bringing power over mid-range
distances of 400 to 600 miles. It also shows that
where regional estimates for transmission costs are
too high, models will underestimate the level of new
wind and coal project construction.
Hydro Shift
The Hydro Shift scenario was developed to help
the Company understand the ramifi cations of a
long-term shift to lower hydroelectric generation
levels witnessed over the past half decade across
the Western Interconnect. This scenario was
accomplished by reducing average hydro generation
by 10 percent during the IRP study horizon.
Moving average hydro energy down by 10 percent
did not have a large effect on the resource selection,
as shown in Figure 6.21. Lower hydro levels allowed
IGCC coal plants into the regional resource mix.
Reducing hydroelectric energy by 10 percent
lowers hydro output in the Western Interconnect
by 1,400 aMW. It also moves a higher percentage
of available hydro energy from low load hours to
high load hours as shown by the change in market
prices in Figure 6.22. As in the Base Case, market
prices did not change substantially under this
scenario because gas resources continue to set
market prices like in the Base Case. To understand
the monetary effect of lower hydro generation in
the West, the total incremental fuel expense to
replace lost hydroelectric generation was calculated.
Fuel expenses increased by $642 million annually
(2005$), a 3.5 percent increase.
6-20
Figure 6.21: Cumulative Resource Selection for the Western Interconnect–Hydro Shift (GW)
-
20
40
60
80
100
120
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
IGCC
Pulverized
Nuclear
SCCT- Frame
CCCT
Wind
RPS-Other
RPS-Wind
Figure 6.22: Mid-Columbia On & Off Peak Price Comparison For The Hydro Shift ($/MWh)
6-21
High Wind Penetration
The High Wind Penetration scenario was developed
to understand potential costs and market effects
of integrating a large amount of wind generation
into the Northwest grid. In this case, the resource
build was modifi ed by the addition of 5,000 MW of
wind placed in service in 2007. Intra-month market
volatility rose by an average of 15 percent. The
higher variation is highly dependent on hydro levels.
See Figure 6.23.
Boom and Bust
The Boom and Bust scenario models a potential
future where the electricity industry behaves more
like the real estate market—speculation and under-
investment initially drive prices to spectacular highs.
These highs are followed by equally spectacular
lows as over-investment pushes speculators
out of the marketplace. The Technical Advisory
Committee requested this scenario to help
understand the ramifi cations of market cycles like
those experienced between 1998 and 2001 across
the Western Interconnect.
The resource build for this scenario is the same
as the No Capacity Credit scenario, except
that resources are only allowed to come online
every fi ve years. The movement of resource
development schedules strains the marketplace.
Figure 6.24 shows market prices for the Northwest
in the Boom and Bust scenario. When resources
come online in 2010, 2015, 2020, and 2025 market
prices fall, while the resource-constrained years
attain higher prices.
Figure 6.23: Monthly Market Price Volatility From Increased Wind Penetration (%)
6-22
6.3 Carbon Emission
Scenarios
The Company developed two carbon scenarios to
address increasing concern over the environmental
effects of greenhouse gases, including carbon
dioxide, methane, nitrous oxide, hydrofl uorocarbons,
perfl uorocarbons, and sulfur heaxafl uoride. The
GHG issue is often referred to as a carbon, carbon
dioxide (CO2), or carbon dioxide equivalents
problem. Internationally, GHG emissions are
regulated by the Kyoto Treaty. The treaty was
developed in 1997 and implemented in February
2005. The Kyoto Treaty established a carbon
emissions trading market in Europe, which began
trading this year. The U.S. did not ratify the Kyoto
Treaty and current laws in the United States do
not regulate GHG emissions. The main legislative
proposal for limiting GHG emissions is the McCain-
Lieberman bill in the U.S. Senate. This bill is
described later in this section.
Several states in the West are starting to regulate
GHG emissions through law or policy. California
laws limit carbon and noxious oxide emissions in
vehicles. The California Public Utility Commission
also requires the inclusion of a carbon adder in any
thermal based generation proposals to account for
the potential future costs of GHG emissions.
Oregon established the fi rst CO2 standards in
the U.S. in 1997, requiring new carbon emitting
generation projects to offset a portion of their
CO2 emissions through effi ciency improvements,
cogeneration projects, other offset projects like
tree planting, or payments into the Climate Trust of
Oregon. Washington State requires CO2 mitigation
for new fossil-fueled thermal electric generation
plants exceeding 25 MW of nameplate capacity.
Figure 6.24: Base Case and Boom and Bust Mid-Columbia Electric Price Forecasts ($/MWh)
40
45
50
55
60
65
70
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Base Case
Boom-Bust
6-23
Though there is no national GHG law or policy
today, the Company believes that some form of
GHG emissions regulation will occur at some point
in the future. The challenge arises in assessing
when the new requirements might begin and how
expensive future emissions of GHG will be. Large
costs enacted early in the IRP timeframe would
push the Company away from high carbon emitting
resources. A carbon tax implemented late in the
forecast horizon would not signifi cantly impact the
economics of carbon emitting resources.
It is diffi cult to analyze carbon emissions, absent
a specifi c federal law or mandate. However, the
Company believes that it is prudent to study the
potential impact of carbon regulation on its Preferred
Resource Strategy. If there is a clear mandate at
the federal or state level to reduce carbon emissions
so that the higher costs associated with greener
generation can be calculated in the future, the
Company will be able to forecast its impact on future
generating capacity choices.
SB 342 Carbon Tax
SB 342, otherwise known as the McCain-Lieberman
Bill or Climate Stewardship Act (CSA) of 2005,
initially was introduced to the Senate in October
2003. It was intended as a comprehensive plan
for the U.S. to reduce heat-trapping gas emissions
to year 2000 levels by 2010. The bill would have
reduced emissions through a market-based tradable
allowance system patterned after the sulfur dioxide
emission permit market established by the Clean
Air Act of 1990. It was expected to make carbon
emissions costly enough to shift our economy away
from carbon producing technologies.
There are several different opinions on the necessity
of the CSA, ranging from it refl ecting a crisis
that requires immediate action, to it needlessly
destroying the national economy. Several groups
and governmental agencies have studied the
CSA and have attained different results. The
Massachusetts Institute of Technology performed
an economic study of the CSA and found the overall
cost would be $20 per household per year. Charles
River Associates determined that the cost would be
$350 per household in 2010 and would increase to
$530 per household by 2020, with the potential for
costs to increase to $1,300 per household per year.
The Energy Information Administration (EIA) also
performed an analysis of the CSA. It found that the
discounted per capita cost would be $56 annually
per person (2005$).
The Company chose to use results of the EIA
study for this carbon tax scenario. It appears
to be the most comprehensive analysis and has
more information on the effects to the electricity
marketplace.
A large carbon tax on electric generating facilities
implemented in 2010 would likely stop or severely
restrict construction of new carbon-emitting coal
plants. The new resource mix would still rely on
natural gas, as shown in Figure 6.25; however wind,
solar, geothermal and carbon sequestration coal
plants would also enter the mix. If the CSA passes,
6-24
many existing coal plants may shut down because
carbon credits likely would be more valuable than
the electricity they produce. This is described in
Figure 6.26. When the carbon tax peaks in 2023, 60
percent of remaining coal output is from plants with
carbon sequestration technology.
An additional assumption of note is that renewables
in this scenario do not receive the PTC. The 2005
IRP follows the NPCC Fifth Power Plan assumption
that the PTC would not be renewed once a carbon
tax is enacted. The incentive to generate power
through renewable resources would be replaced
by the fi nancial disincentive of a carbon tax on
fossil fueled assets. If the PTC for wind continued
after a carbon tax was added, it would effectively
double the net incentive to construct renewable
resources. The Company does not believe this is
likely over the long run.
Carbon dioxide emissions might fall 20 percent in
2014 from the Base Case and 50 percent by 2022
for the Western Interconnect generating fl eet under
the CSA. See Figure 6.27.
A carbon tax likely will not end carbon production
by the U.S. electricity industry. New wind and other
renewable resources are not capable of serving the
entire need of the Western Interconnect. Without a
fundamental change in the industry, such as a shift
to nuclear power, market prices still will be set by
carbon emitting combined-cycle gas plants.
Our modeling shows that lowering emission levels
across the Western Interconnect will come at a
high cost to customers. Figure 6.28 illustrates
that Mid-Columbia electric prices could increase
by 47 percent from the Base Case in 2014 and
66 percent in 2020. Increased market prices
are driven by higher taxes and higher fuel costs.
Figure 6.25: Cumulative Resource Selection for the Western Interconnect – SB 342 Carbon Tax (GW)
-
20
40
60
80
100
120
140
160
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
IGCC SQ
Nuclear
SCCT- Frame
CCCT
Wind
RPS-Other
RPS-Wind
6-25
Figure 6.26: Coal Dispatch Between Base Case and SB 342 Scenario (millions of tons)
0
5
10
15
20
25
30
35
40
45
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Base Case
SB 342
Figure 6.27: CO2 Emissions and Cost Forecast for the Base Case and SB 342
6-26
Nearly 14 billion dollars of CO2 allowances would
be exchanged annually between 2010 and 2026
to keep the western United States within carbon
limits. The Western Interconnect would see $2.5
billion in increased fuel costs every year as a result
of switching from coal to gas-fi red plants. Higher
electricity prices, driven by a carbon tax, will
decrease future loads as customers respond to
higher prices. Based on work from the EIA study,
Western Interconnect loads are forecast to fall by
0.33 percent annually after 2010 to refl ect reduced
demand caused by higher electricity prices.
National Commission for Energy Policy
Carbon Tax
The National Commission for Energy Policy (NCEP)
is a non-governmental group of 18 energy experts
funded by several private foundations and trusts to
develop a national energy strategy for the United
States. In December 2004, NCEP published
“Ending the Energy Stalemate: A Bipartisan Strategy
to Meet America’s Energy Challenges.” A section
of the report is devoted to the risks of climate
change and calls for the establishment of a national
tradable–permits program for GHG. The Company
considered an alternative because a carbon tax has
not been established in the U.S. at this time and
because of the signifi cant impacts of the SB 342
Carbon Tax Scenario described above. The NCEP
study calls for an initial cost of around $7 per metric
ton of CO2 equivalent beginning in 2010. The price is
forecast to rise to approximately $15 per ton in 2026.
The Company assumed that legislation based on
the NCEP analysis would eliminate the federal
Figure 6.28: Base Case and SB 342 Mid-Columbia Electric Price Forecasts ($/MWh)
30
40
50
60
70
80
90
100
110
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Base Case
SB 342
6-27
production tax credit for renewables. The results
of the study found that the tax would essentially
eliminate new pulverized coal plants. The study
also found that the loss of the federal PTC under
the NCEP Carbon Tax scenario disadvantaged
wind relative to the Base Case. Figure 6.29 shows
the resource build for this scenario. It maintains
the status quo with continued construction of
natural gas resources and modest investments in
other resources.
If Congress passes a carbon allowance program
that result in a CO2 tax similar to that of the NCEP
forecast, carbon emissions would continue to rise
because new natural gas resources would be built
and existing coal resources would remain online.
NCEP carbon tax levels likely would succeed in
prohibiting new coal-fi red resources that did not
sequester their carbon emissions. Figure 6.30
shows that carbon emissions are expected to
increase from Base Case levels, but at a slower rate
of growth. NCEP carbon tax levels will still affect
marginal electric prices signifi cantly. See Figure 6.31.
Emission Scenarios Summary
Where federal legislation limits carbon emissions,
electricity prices are likely to increase sharply. The
carbon tax likely will eliminate proposals to build
new coal plants unless future technologies reduce
carbon emission levels from these plants. In today’s
tight natural gas market, it is plausible that the
necessary large shift to natural gas-fi red resources
would drive natural gas prices to new highs not seen
before. In this case, electricity prices might rise even
more substantially than presented in this study.
Figure 6.29: Cumulative Resource Selection for the Western Interconnect – NCEP Carbon Tax (GW)
-
20
40
60
80
100
120
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
IGCC SQ
Pulverized
Nuclear
SCCT- Frame
CCCT
Wind
RPS-Other
RPS-Wind
6-28
Figure 6.30: Western Interconnect Generator CO2 Emissions Forecast
Figure 6.31: Base Case and NCEP Carbon Tax Mid-Columbia Electric Price Forecasts ($/MWh)
35
40
45
50
55
60
65
70
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Base Case
NCEP
6-29
6.4 Avista-Centric Scenarios
Avista-centric scenarios are scenarios that do not
affect the marketplace but do affect the Company.
Because the marketplace isn’t affected, the
scenarios were not modeled with AURORAXMP. The
following is a list of the Avista-centric scenarios:
• Green Growth Initiative
• Loss of Spokane River Projects
• Intermediate-Term Loss of Noxon Rapids
Powerhouse
• High Load Growth Trajectory
• Low Load Growth Trajectory
• Long Haul Coal Option
• Double Conservation Acquisition
Green Growth Initiative
The Green Growth Initiative became the “All
Renewables” resource portfolio discussed in
Section 7- Preferred Resource Strategy.
Loss of Spokane River Projects
The Spokane River projects are licensed through
June 2007. The Company expects to renew
its federal license to operate these facilities.
The Technical Advisory Committee asked the
Company to assess the fi nancial impact of losing
these assets. The Company found that a loss of
the projects would increase power supply costs
by $458 million net present value over the 20-year
IRP timeframe.3
Intermediate-Term Loss of
Noxon Rapids Powerhouse
Noxon Rapids is the Company’s largest
hydroelectric resource and its most fl exible
asset. A short-term loss likely would be offset
with intermediate-term market purchases.
The Technical Advisory Committee asked the
Company to produce scenarios detailing potential
causes for such an outage and the fi nancial impact
of the outages. Avista’s engineering department
identifi ed three possible outage scenarios:
earthquakes causing the wash out of earthen
embankments (two to three years), powerhouse
fl ooding (nine months), and a major transformer
or switchyard failure (nine months).4 Table 6.3
illustrates the value of Noxon Rapids in each year
of the IRP study.
4 The capital replacement costs for these outages depends on the level
of damage to existing assets. These costs are not included in the cost
estimates of Table 6.3
Year Cost Year Cost
2007 89.5 2017 79.6
2008 82.6 2018 81.8
2009 77.3 2019 86.4
2010 73.6 2020 89.1
2011 74.7 2021 90.0
2012 71.4 2022 92.6
2013 72.2 2023 95.1
2014 73.2 2024 95.9
2015 74.9 2025 99.4
2016 77.3 2026 101.8
Table 6.3: Market Value
of Noxon Rapids Project ($millions)
3 This estimate does not consider signifi cant cost reductions stemming
from ceasing operations at the projects.
6-30
High and Low Load Trajectory
The Electricity Sales Forecast section discussed high
and low Company load scenarios. Avista currently
has adequate resources until 2009. As the IRP is
updated every two years, the Company will have
the opportunity to adjust its load forecast based on
changes in expected load levels. We believe that a
shift in load growth will not substantially change the
mix of resource types, but potentially could change
the quantity.
Long Haul Coal Option
The Company studied the potential for locating a
new coal plant in or near its service territory. The
Company believes that plant capital costs will not
be substantially different whether located outside of
the Northwest or closer to our load. A plant located
in Montana, for example, will require substantially
higher transmission investment than a plant located
closer to Avista. A plant located in our service
territory will have a higher fuel expense driven
primarily by rail transportation costs necessary to
bring in coal from distant mining regions. Overall,
the long-haul coal option appears cost competitive
when compared to a mine-mouth coal plant
located outside of the Northwest. The Company
will continue to study various coal plant locations,
including local sites, as part of its action plan.
Double Conservation Acquisition
Section 3 of this IRP explained that the Company
would work to acquire 6.9 aMW of conservation in
each year of the IRP study period. The Company
was asked during its Technical Advisory Committee
meeting to quantify the cost were the Company
to double its conservation acquisition levels. The
Company found that if it acquired 13.8 aMW
annually, then program costs would rise to 2.5
times the Preferred Resource Strategy level. The
increase in conservation also would reduce the
need for new resources.
6.5 Avoided Costs
Avista is obligated to purchase from certain third-
party generation projects under the Public Utility
Regulatory Policies Act of 1978 (PURPA). The
federal law states that such purchases will be at
prices equal to avoided cost. State regulatory
commissions implement PURPA provisions in
their states.
The Washington and Idaho Commissions interpret
which resources are eligible for PURPA avoided cost
rates. PURPA developers with projects that exceed
certain levels are eligible for a negotiated rate based
on utility avoided cost. Published rates are provided
for smaller PURPA facilities. In Washington PURPA
resources below 1 MW are eligible for published
fi xed rate schedules with a term of up to fi ve years.
The fi ve-year schedules are tied to forward market
prices. In Idaho, facilities up to 10 aMW may obtain
published avoided cost rates for up to 20 years.
Avoided Costs Versus the
Wholesale Marketplace
There is some disagreement in the industry
over what constitutes avoided cost. In Idaho,
administratively-determined avoided cost rates
6-31
presently are based on a gas-fi red CCCT as a
surrogate to represent the Company’s next
lowest cost investment. The published fi gure
explicitly includes the cost of installing capacity.
In Washington, published rates are based entirely
on the forward wholesale market price.
Avoided Costs Approach
The 2003 IRP ignored planning margins and only
built resources that could recover all costs, including
capacity payments, in the marketplace. The 2003
IRP market prices included all costs associated with
constructing new resources; the market equaled
avoided cost.
The 2005 IRP uses capacity credits to insure
planning margins adequate to prevent large price
spikes during various adverse market conditions.
With capacity credits lowering the installed cost
of new resources, the wholesale marketplace
modeled for the IRP more accurately represents
the wholesale electricity marketplace we witness
today. The drawback is that the modeled wholesale
marketplace does not represent full utility avoided
costs. A secondary step, essentially reverting to
the 2003 IRP methodology, is necessary to extract
avoided costs from the IRP modeling.
Once all 2005 IRP modeling assumptions were
fi nalized, an additional run was launched without
capacity credits reducing resource construction
costs. The Base Case run, the basis for our 20-year
market price forecast, and the new avoided cost run
are displayed below in Figure 6.32.
The same data may be found in tabular format
in Table 6.4, along with both ten- and 20-year
levelized costs.
Figure 6.32: Base Case Mid-Columbia Price Forecast and Avoided Costs Comparison ($/MWh)
35
40
45
50
55
60
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Base Case
No Capacity Credit
6-32
6.6 Summary
Using a regional approach to calculate market prices,
and to calculate the value of resource options,
provides for more robust results when compared to
an analysis that separates resource modeling from
price forecasting. The Company also believes that
using a stochastic approach to evaluate risk is more
valuable than simply creating scenarios.
Year
BC Market
Forecast Avoided Cost Year
BC Market
Forecast Avoided Cost
2007 51.25 49.99 2017 46.56 49.12
2008 46.91 47.04 2018 47.49 50.59
2009 44.57 44.42 2019 50.17 51.71
2010 42.82 42.80 2020 51.71 52.37
2011 43.11 44.21 2021 52.63 54.67
2012 41.22 44.11 2022 53.75 54.62
2013 42.04 44.83 2023 54.88 56.11
2014 42.71 45.73 2024 55.35 57.32
2015 44.08 46.63 2025 57.57 57.90
2016 45.09 47.84 2026 59.07 59.42
10-Yr.
Lev. Cost 44.78 45.84 20-Yr.
Lev. Cost 47.05 48.28
Table 6.4: Avista Avoided Costs Compared to Mid-Columbia Price Forecast ($/MWh)
This section focused on market prices used
to select the Preferred Resource Strategy, and
discussed many regional costs and benefi ts of
certain market actions. The next section will
focus on how the Company used this information
in creating the PRS, and the effect of the various
scenarios and futures on the PRS.
7-1
The Preferred Resource Strategy (PRS) contains
the Company’s forecasted preferred mix of new
resources over the IRP time horizon. The PRS must
strike a balance between the many (and oftentimes
confl icting) criteria of resource planning. One potential
future mix of resources might result in the lowest
absolute cost over time but does so at the expense
of volatile costs from one year to the next. Another
future might keep rates reasonably stable over time
but suffer from an unacceptably higher average rate
level. The PRS generally is not capable of providing
an optimal outcome when measured against each
resource-planning criterion and/or market condition
individually. Instead, a PRS should perform strongly
across the various criteria and the range of possible
future market conditions, when compared to other
resource strategies. Herein lies the largest challenge
facing electric utility resource planners today.
This section will introduce and then later detail the
Company’s 2005 IRP PRS. It will introduce 12
alternative resource strategies developed to illustrate
the relative strengths and weaknesses of resource
options under varying models of future market
conditions. Next, the Company’s work to develop
an Effi cient Frontier is detailed. The last few pages
tabulate the Company’s load and resource balance
with the inclusion of PRS resources.
7.1 The Preferred Resource
Strategy—An Introduction
The wholesale marketplace is comprised of
thousands of generating assets located across the
western United States. This market is available
to the Company to help manage its assets to the
benefi t of retail customers. At certain times it is less
costly to shut down owned generation plants and
purchase power from other market participants.
At other times Company-owned assets provide
electricity at the least cost.
Section Highlights
 The Preferred Resource Strategy meets more than 50 percent of load growth with conservation,
plant effi ciency upgrades, and renewables.
 Our annual conservation target is 50 percent higher than in 2003.
 The 2005 IRP Effi cient Frontier, a tool for comparing the tradeoff between price volatility and
expected cost, is the product of 1,000 Avista Linear Programming model simulations.
 The PRS reduces portfolio price volatility by 55 percent in 2016 when compared to relying
exclusively on market purchases.
7. PREFERRED RESOURCE STRATEGY
7-2
Prior to the energy crisis of 2000-01, many within
our industry, including policy makers, utilities,
and customers believed that the wholesale
marketplace could serve customers at costs below
traditional regulation. To varying degrees, utilities
relied more heavily on the energy market than they
had in the past.
Avista believes that a prudent strategy for serving
its customers in the future contains a mix of
resources and/or contracts backed by generation
assets. A portfolio comprised substantially of
actual generation property, owned or held under
contract, is necessary to ensure reliable service at
risk-adjusted least-cost. The Preferred Resource
Strategy was developed in part by using results from
Avista’s Linear Programming model discussed in
Section 5- Modeling Approach.
The Company’s Preferred Resource Strategy was
developed after careful review of the Effi cient
Frontier, the relative performance of 12 alternative
resource strategies (described later in this section),
and results of 18 alternative marketplace scenarios
and Avista-centric possibilities. The PRS is defi ned
by three generation categories: wind generation,
coal-fi red generation and other small renewables. It
contains upgrades to existing Avista resources and
a signifi cant increase in conservation acquisition
from today’s level.
The PRS does not recommend additional natural
gas-fi red generation due to the high level of gas-
fi red generation already in the Company’s portfolio,
the high price of natural gas, and the resource’s
tendency to introduce additional volatility into
Avista’s portfolio. In 2016 total installed capacity
is 400 MW of wind, 250 MW of coal, and 80 MW
of other small renewable projects. Resource
requirements are 69 MW and 52 MW lower
because of conservation and effi ciency upgrades
to existing resources, respectively. By 2026, the
end of the IRP study timeframe, total installed
capacity equals 1,332 MW and is comprised of
650 MW of wind generation, 450 MW of coal-
fi red generation, 180 MW of other renewable
generation, and 52 MW of plant effi ciency
upgrades. Needs are 138 MW lower because of
conservation. Figure 7.1 illustrates the Preferred
Resource Strategy developed by the Company.
This PRS mix differs from the 2003 IRP primarily
by the replacement of a signifi cant portion of the
coal-fi red resource with wind and other renewable
generation projects. The 2003 IRP Preferred
Resource Strategy is shown below in Figure 7.2.
Three factors explain the differences between the
2003 IRP and this plan. First, the acquisition of the
second half of Coyote Springs 2 in January 2005
brought 140 MW of natural gas-fi red combined-
cycle combustion generation into the Company’s
portfolio. That purchase met the natural gas-fi red
component of the 2003 IRP.
Second, higher natural gas and electricity market
prices have allowed resources that previously were
uncompetitive, namely wind and other renewable
7-3
0
200
400
600
800
1,000
1,200
2008 2010 2012 2014 2016 2018 2020 2022
Coal
Wind
Peakers
CCCT
Figure 7.2: 2003 IRP Preferred Resource Strategy Build (MW)
Figure 7.1: 2005 Preferred Resource Strategy Build (MW)
0
200
400
600
800
1,000
1,200
1,400
1,600
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Market Sales
Market Purchases
Renewables
Wind
Coal
Upgrades
Conservation
7-4
resources, to now become competitive. Finally,
wind integration studies and actual experience
with integrating wind into the Company’s system
lead us to believe that we can rely more heavily on
this resource.
The PRS adds “lumpiness” to the acquisition
schedule when compared to Effi cient Frontier and
alternative scenario resource mixes. The lumpy
nature more closely refl ects how the Company
might expect to add resources. This contrasts with
the portfolios selected by the Avista LP model for
the Effi cient Frontier. The Effi cient Frontier resource
selections are not constrained by lumpiness. For
example, the 50 Percent Risk mix allows annual
acquisition levels of 49.3 MW, 2.6 MW, and 36.6
MW of new coal in various years of the study.
It would be nearly impossible either to construct
plants, or obtain contract provisions, allowing for
these capacity levels. Instead, resource acquisitions
likely will occur as shown in the Preferred Resource
Strategy, with blocks of no less than 100 MW in any
given year for coal plants, blocks of no fewer than 25
MW for wind, and no fewer than fi ve MW for biomass
plants. Medium term market purchases of up to fi ve
years can also be made to allow added fl exibility on
the timing of new power plant acquisitions. Modest
market purchases included in the PRS have blocks
no smaller than 25 MW. Bringing new resources into
the portfolio on a less granular schedule increases
costs slightly when compared with resource mixes
developed for the Effi cient Frontier and other
resource mix alternatives that are not constrained by
lumpiness. Figure 7.3 compares the 50 Percent Risk
and PRS acquisition patterns.
Figure 7.3: Preferred Resource Strategy Coal Build vs. LP Module Build (MW)
0
50
100
150
200
250
300
350
400
450
500
2009 2011 2013 2015 2017 2019 2021 2023 2025
50% Risk
Preferred Resource Strategy
7-5
7.2 Preferred Resource
Strategy Details
Wind Resources in the PRS
Wind comprises the largest nameplate capacity
component of the Preferred Resource Strategy,
contributing 400 MW by 2015 and 650 MW by
2024. The Company’s reliance on wind technology
in the 2005 plan also represents a large strategic
shift from the 2003 IRP and is the result of two
major changes since the last plan further research
on wind resources and wind integration and rising
wholesale market prices.
The Company committed to “continu[ing] to
evaluate the effects and costs of integrating wind
generation...” in its 2003 IRP Action Plan. Various
data and internal evaluations, combined with actual
experience gained from integration of 35 MW of
wind generation from the Stateline Wind Energy
Center, were completed since the release of the
2003 plan. This work indicated that the Company
might be able to include signifi cant, but not
unlimited, additional wind resources into its future
plans. It also was learned that the Company might
need to purchase wind integration services from
third parties for some or all of its wind resource due
to rising integration costs incurred as installed wind
capacity levels grow. The 2005 IRP adopts NPCC
wind integration cost estimates.
How much wind can Avista reasonably expect
to include in its future? Exhausting wind’s cost-
effective regional potential becomes a concern as
utilities in the Northwest, including Avista, begin
to include wind plants in their future plans. Idaho
Power’s 2004 Integrated Resource Plan identifi es
350 MW of wind over their ten-year planning
horizon. Pacifi Corp plans to include 600 MW for its
west-side service territories. Puget Sound Energy
has committed to 845 MW of wind. Portland
General Electric includes 200 MW in its latest IRP.
Add Avista’s 400 MW by 2016 and the region’s
investor-owned utilities are looking to add 2,395 MW
of nameplate wind capacity. Table 7.1 details the
fi ve utilities, with a comparison of their loads and
wind plans.
The NPCC estimates that total Northwest wind
generation potential is 5,000 MW. The fi ve
Northwest investor-owned utilities are planning to
develop nearly 50 percent of regional potential over
the next ten years.
Though aggressive, Avista believes it is possible to
acquire 400 MW by 2016 and 650 MW by the end
of 2026 by pursuing three different wind resource
strategies. First, the 2005 IRP assumes that
Avista will acquire 250 MW of Northwest regional
wind generation outside of its service territory.
This amount approximately equates to its pro-
rata share based on Northwest loads. Second,
the PRS selects 150 MW of wind within Avista’s
service territory. While in-territory wind resources
are estimated on average to generate 21 percent
less energy than sites presently being developed
across the Northwest, transmission savings are
signifi cant and make the sites potentially attractive.
7-6
Finally, the plan assumes another 250 MW of
wind generation will be available from outside the
Northwest (e.g., eastern Montana or Wyoming).
These sites have relatively higher capacity factors
when compared to Northwest wind sites. Higher
transmission costs therefore can be offset
somewhat by higher generation levels. Acquisitions
from outside the Northwest will be dependent on the
availability of transmission at costs allowing them to
be acquired economically.
Other Renewables in the PRS
The LP model selected a mix of renewables
besides wind power, namely landfi ll and manure
biomass, in many of the Effi cient Frontier
portfolios. This result indicates the possibility of
further cost-effective investments in renewable
energy technologies above 650 MW of wind. The
PRS includes 80 MW of biomass resources, both
from landfi ll gas and manure methane, by 2016.
Other renewables are forecasted to provide 180
MW by the end of the IRP timeframe.
As with wind, Avista’s ability to include signifi cant
renewable resources in its future resource portfolio
ultimately will depend on how close NPCC cost
estimates for these resources come to actual offers
received by the Company. Integration also will
depend on commercial availability. The NPCC Fifth
Power Plan expresses concern over the viability and
potential of biomass renewable resource options.
The Company will explore this issue as an action
item in the 2005 IRP and provide further information
in its 2007 plan.
Conservation in the PRS
The 2005 IRP supports increasing annual
conservation acquisitions from approximately 4.6
aMW today, to 6.9 aMW. This equals a nearly 50
percent increase, due primarily to higher avoided
Utility
IRP Wind
Capacity (MW)
2016 Load
(aMW)
IRP Wind
Energy1 (aMW)
Wind Contribution
to Load (percent)
Avista 400 1,424 132 9.3
Idaho Power2 350 2,187 116 5.3
Pacifi Corp West3 600 2,678 198 7.4
Portland General Electric4 200 3,075 66 2.1
Puget Sound Energy5 845 2,790 279 10.0
Total 2,395 12,154 790 6.5
Table 7.1: Northwest IOU Loads and Estimated Wind Acquisition Plans through 2016
1 Assumes all wind resources have a 33 percent capacity factor for comparative purposes.
2 2013 levels from 2004 Integrated Resource Plan: pages 2 and 30.
3 See pages 30 and 38 of 2004 Integrated Resource Plan Appendix.
4 Load is found on page 100 of 2002 Integrated Resource Plan.
5 2013 statistics. Includes existing wind at Hopkins Ridge and Wild Horse. See PSE 2005 IRP, pages 1X-8 and X-22
7-7
cost estimates that include a ten percent adder
over generation-based acquisition, and a
movement toward 8,760-hour evaluation of the
various available measures. Refer to Section
3-Conservation Initiatives for further detail on the
signifi cant enhancements made in conservation
program analyses for the 2005 IRP.
On a cumulative basis, the acquisition of
conservation will offset 69 aMW of new generation
by 2016; in 2026 customer loads are estimated to be
138 aMW lower than absent conservation efforts.
Project Upgrades
The Preferred Resource Strategy includes
upgrades at both its Cabinet Gorge and Noxon
Rapids hydroelectric facilities, as well as at
Colstrip. These modifi cations will bring additional
energy and capacity with no incremental fuel
costs. The various improvements will be
completed between 2005 and 2011.
Coal in the PRS
In reviewing forecasted future customer
requirements it becomes clear that conservation
and renewable resources, while having the
potential to contribute signifi cantly to our future
mix, cannot fi ll the gap entirely. We believe that
conservation and renewables have the potential to
meet approximately two-thirds of our capacity and
one-half of our energy requirements in 2016. After
selecting cost-effective conservation and renewable
resources, the Company looks to more traditional
base load supply-side resources. As discussed in
Section 5, these options include natural gas, nuclear,
Alberta oil sands, and coal located in and outside
of the Northwest. The best option among these
resources for Avista’s resource mix is coal-fi red
generation. Coal benefi ts from low variable costs,
helping to keep power supply expense volatility low.
Coal was selected in the 2003 IRP, though customer
costs were expected to be modestly higher than an
all-gas plan. With higher natural gas prices, coal-
fi red generation also brings lower customer costs.
The PRS contains 250 MW of coal-fi red generation
entering service during the middle of next decade.
In 2026, coal–fi red generation equates to 450 MW,
or 30 percent of our new requirements.
7.3 Efficient Frontier
The Effi cient Frontier is a key component of an
academic body of work in “portfolio theory.” First
applied to fi nance, the Effi cient Frontier measures
tradeoffs between expected return and risk inherent
in securities portfolios. With IRP planning, a similar
exercise in portfolio management, the concept is
applied when selecting future mixes of supply-
and demand-side resources. Figure 7.4 illustrates
the concept by showing risk on the vertical axis,
measured as the 2016 standard deviation of
incremental power supply expenses, and cost on
the horizontal axis, measured as the 2007-16 net
present value of the incremental power supply
expenses.6 Risk and cost are both at their lowest
point in the bottom-left corner of the chart.
6 Incremental power supply expense includes fuel and variable O&M for
existing resources, as well as fuel, variable O&M, fi xed O&M, and capital
recovery for new resources.
7-8
The blue curve explains simultaneously the optimal
measure of risk at any cost point, and conversely
the optimal cost point given a level of acceptable
risk. Notice how it is impossible to attain a lowest
risk and lowest cost position concurrently. This
curve represents the quandary facing resource
planners—selecting a position on the Effi cient
Frontier. The 12 alternative scenarios selected
by the Company to compare to the Preferred
Resource Strategy are displayed as orange
diamonds. The Preferred Strategy itself is shown
as a large green circle.
Effi cient Frontier Concerns
The Effi cient Frontier may work well where risk and
cost matrixes are reasonably well known. For
example, natural gas spot markets have a many-
year historical data series. Natural gas price and
volatility therefore can be reasonably estimated.
The Effi cient Frontier concept does have
limitations. Future carbon emission regulations are
not easy to defi ne. There presently is signifi cant
disagreement about the magnitude and timing of
future carbon regulations.
Should the planner then assign carbon tax levels
and associated probabilities to add this variable into
an Effi cient Frontier? The Effi cient Frontier also has
limitations when considering risk such as nuclear
plant siting and waste disposal, carbon emissions
and project cost overruns. The Effi cient Frontier
approach appears unable to address future cost
and risk challenges like these in a meaningful way.
The Company will continue to evaluate the Effi cient
Frontier as a means to measure the tradeoffs
inherent in resource decisions.
Figure 7.4: Effi cient Frontier ($millions)
7-9
7.4 Twelve Alternative
Portfolio Strategies
This section details 12 portfolios developed while
defi ning the PRS. Each provides a different mix
intended to illustrate the strengths and weaknesses
of resource strategies.
1) No Additions (Market Purchases
Backed by Peaking Plants)
In the No Additions portfolio, the Company would
plan to rely on the wholesale marketplace to meet
all of its future load requirements. No additional
investments in generation plants or transmission
are envisioned. This strategy, by defi nition, would
limit Company ownership of generation assets to
its existing mix of resources. Customer rates would
vary depending on price levels in the wholesale
marketplace. In higher priced years customers
would see their power bills rise, potentially
substantially. For example, prices in calendar
year 2002 averaged $22 per MWh, below the
Company’s present production cost. However, in
both 2003 and 2004 average prices were nearly
double 2002 levels. With direct exposure to the
wholesale marketplace, customer rates would
have the potential to rise or fall substantially in any
given year. Customers would also be exposed
to extreme market conditions where prices could
range well above $100 per MWh. With half of the
Company’s power supply expenses tied directly to
the wholesale marketplace, such a condition could
increase power supply expenses in a given year by
250 percent.
By its nature of having no new assets, the No
Additions alternative is not a portfolio the Company
will pursue. Instead, the strategy is included as
a benchmark for comparison against the other
portfolios evaluated for the 2005 IRP.
2) All Coal
Coal-fi red generation serves more than 50 percent
of the nation’s electricity needs today Avista relies
on coal-fi red generation to meet approximately
18 percent of its needs. Coal reserves in the
United States are so vast that some industry
experts believe they will extend to the middle of
this millennium, an attractive feature given recent
run-ups in the prices of commodities tied to crude
oil and natural gas. Coal generation benefi ts
from its historical independence of the oil and
natural gas markets, and its relatively low fuel
cost. There is some risk that this independence
might be compromised over time as existing
and new technologies for converting coal into
various synthetic petroleum products are driven to
commercialization by rising crude oil and natural
gas prices.
The All Coal portfolio meets all new load
requirements with coal-fi red generation. Coal-fi red
generation presently provides 222 MW of generating
capacity in the Company’s portfolio of resources
and approximately 185 aMW of energy. Under the
All Coal portfolio, coal’s contribution would rise to
714 MW in 2016 and 1,078 MW in 2026. At the end
of the study, coal would meet 43 percent of all utility
capacity requirements.
7-10
3) All Gas
Natural gas-fi red generation has been the
predominant resource constructed across the
United States in the past decade. Its benefi ts
include low capital costs, simpler permitting and
engineering, and moderate emission levels. Recent
rises in the price of natural gas are forecast to
continue well into the future. This option does
not provide much customer protection against
market volatility because natural gas-fi red
generators are the marginal resource of today’s
wholesale marketplace.
In the All Gas portfolio, the Company would add
492 MW and 856 MW of natural gas-fi red combined
cycle natural gas-fi red generation by 2016 and
2026, respectively. Natural gas would become the
dominant generating resource used by the Company
to meet customer requirements. In 2016, fully 46
percent of the Company’s generation capacity
Figure 7.5: 50/50 Gas and Coal Build (MW)
0
200
400
600
800
1,000
1,200
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Market Sales
CCCT
Coal
Upgrades
Conservation
would be from gas-fi red generation; 52 percent
would come from natural gas in 2026.
4) 50/50 Gas and Coal
The 50/50 Gas and Coal portfolio would split
capacity additions equally between these resources,
providing some balance between the lower capital
costs of natural gas-fi red generation and the
lower fuel cost of coal-fi red generation. Figure 7.5
illustrates the 50/50 Gas and Coal portfolio.
5) Wind and Gas
The Wind and Gas portfolio benefi ts from a greater
reliance on wind, a resource absent fuel costs and
air emissions. It is comprised of approximately 400
MW each of wind and combined-cycle combustion
turbines (CCCT) in 2016. In 2026, the end of the
IRP study timeframe, wind provides 650 MW of
nameplate capacity with CCCTs contributing nearly
700 MW. Wind generation is limited to 650 MW
7-11
Figure 7.6: Wind and Gas Build (MW)
0
200
400
600
800
1,000
1,200
1,400
1,600
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Market Sales
CCCT
Wind
Upgrades
Conservation
of capacity over 20 years, a level the Company
believes is aggressive for this resource. The
Company’s decision to constrain wind’s contribution
to our future mix is based on the limited availability
of this resource. More detailed discussions of the
limit are contained in Sections 5 and 6. The Wind
and Gas portfolio may be found in Figure 7.6.
6) No CO2
The No CO2 scenario was developed to illustrate
a mix of net-zero CO2-emitting resources that may
consist of wind, nuclear, other renewables, and
cogeneration additions. This strategy brings wind
generation into the Company’s portfolio sooner
and more aggressively, reaching the full 650 MW
potential by 2016. The other major contributor to
the portfolio is nuclear energy, at 176 MW in 2016.
Renewables besides wind contribute 70 MW by
2016, while cogeneration adds another 25 MW. By
2026, the contributions of nuclear and renewables
rise to 494 and 170 MW, respectively. Cogeneration
grows slightly to 30 MW. The No CO2 portfolio is
shown below in Figure 7.7.
7) All Renewables
The All Renewables case ignores potential wind
generation limitations, and constructs a portfolio
mix comprised of 1,406 MW wind and 140 MW
of other renewable resources by 2016. The totals
rise to 2,225 MW of wind and 300 MW of other
renewables in 2026. The large wind capacity
requirements of this scenario were necessary given
the limited on-peak capacity contribution of wind
(25 percent of nameplate capacity). More than
2,500 MW of nameplate generation was constructed
to meet load growth of just less than 900 MW in
the All Renewables portfolio. Substantial surplus
generation therefore must be sold into the volatile
wholesale marketplace. All Renewables is shown
7-12
Figure 7.8: All Renewables Build (MW)
0
500
1,000
1,500
2,000
2,500
3,000
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Market Sales
Renewables
Wind
Upgrades
Conservation
Figure 7.7: No CO2 Emissions Build (MW)
0
200
400
600
800
1,000
1,200
1,400
1,600
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Market Sales
Market Purchases
Renewables
Nuclear
Wind
Upgrades
Conservation
7-13
Resource Least Cost 25% Risk 50% Risk 75% Risk Least Risk
Coal 0 205 243 243 133
Gas CT 444 105 31 0 0
Gas CCCT 0 0 0 21 21
Wind 0 275 400 400 400
Other Renew/Cogen 0 75 80 90 200
Market 47 38 38 38 38
Total 492 698 792 792 792
Table 7.2: 2016 Resource Strategies (MW)
Resource Least Cost 25% Risk 50% Risk 75% Risk Least Risk
Coal 0 550 550 550 323
Gas CT 856 105 31 0 0
Gas CCCT 0 0 0 21 21
Wind 0 400 650 650 650
Other Renew/Cogen 0 75 105 165 400
Nuclear 0 77 58 8 0
Market 0 0 0 0 0
Total 856 1,206 1,394 1,394 1,394
Table 7.3: 2026 Resource Strategies (MW)
in Figure 7.8. Like the No Additions case, the
Company does not believe that the All Renewables
portfolio is realistic due to a lack of adequate wind
sites and the intermittent nature of the resource;
however the scenario is included in this IRP at the
request of the Technical Advisory Committee.
8-12) Risk-Adjusted Portfolio Strategies
Five portfolios were selected from the Effi cient
Frontier exercise to illustrate various resource
combinations and their performance under the
alternative market scenarios and futures. The
points on the Effi cient Frontier represent varying
combinations of risk, defi ned as the standard
deviation of expected incremental power supply
expenses, and cost, defi ned as the expected
net present value of incremental power supply
expenses, between 2007 and 2016. The 2003
IRP Preferred Resource Strategy was based
predominantly on a mix of resources defi ned by
weighting cost and risk at 50 percent each. Each
risk-adjusted portfolio resource mix is shown in
Tables 7.2 and 7.3 for calendar years 2016 and
2026, respectively.
7-14
Figure 7.9: Least Cost Build (MW)
0
200
400
600
800
1,000
1,200
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Market Sales
Market Purchases
CT
Upgrades
Conservation
Figure 7.10: 25% Risk Build (MW)
0
200
400
600
800
1,000
1,200
1,400
1,600
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Market Sales
Market Purchases
Cogen
Renewables
CT
Nuclear
Geothermal
Wind
Coal
Upgrades
Conservation
7-15
Figure 7.11: 50% Risk Build (MW)
0
200
400
600
800
1,000
1,200
1,400
1,600
1,800
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Market Sales
Market Purchases
Cogen
Renewables
CT
Nuclear
Geothermal
Wind
Coal
Upgrades
Conservation
Figure 7.12: 75% Risk Build (MW)
0
200
400
600
800
1,000
1,200
1,400
1,600
1,800
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Market Sales
Market Purchases
Cogen
Renewables
CCCT
Nuclear
Geothermal
Wind
Coal
Upgrades
Conservation
7-16
Figure 7.13: Least Risk Build (MW)
0
200
400
600
800
1,000
1,200
1,400
1,600
1,800
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Market Sales
Market Purchases
Renewables
CCCT
Geothermal
Wind
Coal
Upgrades
Conservation
0
200
400
600
800
1,000
1,200
1,400
1,600
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Market Sales
Market Purchases
Renewables
Wind
Coal
Upgrades
Conservation
Figure 7.14: Preferred Resource Strategy Build (MW)
7-17
Figure 7.9 through Figure 7.13 illustrate the
various Effi cient Frontier resource strategies
described above.
7.5 Performance of PRS
Compared to 12 Resource
Strategies
The Preferred Resource Strategy developed for
the 2005 IRP provides the following benefi ts to
customers when compared across the alternative
resource strategies:
• Large contribution from renewable resources
• 50 percent higher level of conservation
acquisition
• Signifi cant reduction in year-on-year rate volatility
• Reasonable rate impacts when compared to
other alternatives
The PRS is shown graphically in Figure 7.14.
Renewable Resource Contributions
The Preferred Resource Strategy contains among
the highest contribution of renewable resources
in the 13 resource strategies. The three portfolios
with higher levels of renewables were allowed
to violate the wind limitation of 400 MW by
2016 and were developed to illustrate certain
characteristics of wind resources. The 100 percent
Risk strategy does contain 100 MW more of
non-wind renewables in 2016. Figure 7.15 shows
the renewables contribution of the 13 alternative
resource portfolios.
Conservation
Conservation plays an increased role in the 2005
Integrated Resource Plan compared to the 2003
IRP. Acquisition levels are increased 50 percent,
from approximately 4.6 aMW per year to 6.9
aMW per year. Figure 7.16 details the impact of
- 100 200 300 400 500 600
PRS
No Additions
No CO2
All Renew
Least Risk
75/25 Risk/Cost
50/50 Risk/Cost
25/75 Risk/Cost
Least Cost
All Coal
All Gas
Wind/Gas
Coal/Gas
Figure 7.15: Renewable Resource Contribution in 2016 (MW)
7-18
- 500 1,000 1,500 2,000 2,500 3,000
PRS
No Additions
No CO2
All Renew
Least Risk
75/25 Risk/Cost
50/50 Risk/Cost
25/75 Risk/Cost
Least Cost
All Coal
All Gas
Wind/Gas
Coal/Gas
Figure 7.17: 2007-16 Portfolio Capital Cost ($millions)
0
20
40
60
80
100
120
140
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Existing
2005 IRP
2026 Difference = 46 aMW, or 50%
Figure 7.16: Conservation Acquisition (aMW)
7-19
higher conservation acquisition targets. More
detailed information may be found in Section 3-
Conservation Initiatives.
Capital Intensity
Resource strategies require differing levels of capital
investment over the IRP timeframe. Lower-risk
portfolios tend to be more capital intensive than
higher-risk ones. The portfolios illustrate this in
Figure 7.17. The capital requirement for the 0%
Risk strategy, composed exclusively of simple-cycle
combustion turbines, requires a comparatively
modest $246 million investment over the fi rst ten
years of the IRP. The All Coal portfolio requires
$1.1 billion. Capital requirements of the PRS—$1.5
billion in nominal dollars by 2016—will be signifi cant
for Avista. The Company might explore power
purchase agreements with third parties that include
purchase options as a way to manage the fi nancial
impacts of the overall acquisition strategy. Medium-
and short-term market purchases are also expected
to fi ll in small gaps between resource acquisition and
load requirement timelines.
Rate Volatility
The Preferred Resource Strategy contains a mix of
resources with low and stable fuel prices. The mix
helps the Company’s resource portfolio reduce year-
on-year power supply expense rate volatility. Figure
7.18 compares the risk inherent in the various portfolio
strategies developed for the 2005 IRP. The statistics
presented are the 20-year average covariance of
each portfolio strategy. Covariance is the quotient
of the standard deviation divided by the mean.
Higher covariance indicates a higher risk profi le. For
example, a 10 percent covariance means that two-
thirds of all expected outcomes will fall between plus
and minus 10 percent of the expected value.
Covariance is a somewhat abstract concept, but
useful for comparing portfolio strategy risk in a
consistent manner over time. Figure 7.19 illustrates
risk in calendar year 2016 by displaying the actual
standard deviation of the incremental power supply
expense. 2007 risk levels, adjusted to 2016 dollars,
are provided to illustrate the risk-reduction benefi ts
of the various resource acquisition strategies.
2007 risk under this measurement is constant, as
the Company has not added any new resources.
Notice that the Least Cost and No Additions
strategies provide modest reductions from the risk
of today’s portfolio mix. This result is due to heavy
reliance on the wholesale marketplace or SSCTs.
The Preferred Resource Strategy reduces risk from
more than $42 million (2016$) to $28 million.
The risk picture is consistent when looking at
“tail distribution.” Figure 7.20 illustrates the risks
inherent in each portfolio at the extreme end of
the distribution curve. The 95th percentile cost
statistic explains that costs in any given portfolio
are not expected to exceed the presented value
except once in 20 years. The fi gure presents the net
present value (NPV) of the difference between the
average and the 95th percentile revenue requirement
for each portfolio.
7-20
Figure 7.18: 2007-26 Portfolio Risk Comparison–Average Covariance (%)
9 10 11 12 13 14 15 16 17
PRS
No Additions
No CO2
All Renew
Least Risk
75/25 Risk/Cost
50/50 Risk/Cost
25/75 Risk/Cost
Least Cost
All Coal
All Gas
Wind/Gas
Coal/Gas
Figure 7.19: 2016 Portfolio Risk Comparison–Standard Deviation of
Incremental Power Supply Expense ($millions)
7-21
Figure 7.20: 2007-16 Portfolio Risk Comparision-95th Percentile Difference From Mean Value (%)
20.0 22.5 25.0 27.5 30.0 32.5 35.0
PRS
No Additions
No CO2
All Renew
Least Risk
75/25 Risk/Cost
50/50 Risk/Cost
25/75 Risk/Cost
Least Cost
All Coal
All Gas
Wind/Gas
Coal/Gas
Rate Impacts
The Preferred Resource Strategy forecasts rate
impacts from incremental power supply expenses
only (i.e., incremental variable O&M and fuel for
new and existing resources, as well as capital and
fi xed O&M for new resources). Other cost increases
are not included in these rate impact estimates.
Figure 7.21 shows an average rate increase of
approximately 4.4 percent between 2007 and
2016 due to new resource construction and
increases in variable costs associated with existing
generation assets. This relative level of increase
is consistent across all resource portfolios that do
not rely on gas-fi red resources, on coal resources
exclusively, or the marketplace entirely. Annual rate
increases could be modestly less than 4.4 percent
were the Company to choose one of these plans;
however, the Company believes that the overall
cost increases associated with the PRS are
reasonable given its ability to greatly reduce risk
and its renewables resource levels. While the No
Additions case appears attractive in this view,
and in Figures 7.22 through 7.24, its underlying
assumptions are unrealistic. No new resources are
constructed thereby leaving the portfolio exposed
to wholesale marketplace volatility. Additionally,
no new transmission costs are included to allow
increasing market purchases. Please refer back
to the No Additions scenario discussion presented
earlier in this discussion.
Another way to look at the cost of the PRS is to
consider annual power supply expense levels.
Figure 7.22 contains a summary of “Incremental
Power Supply Expense,” defi ned for the 2005 IRP
to be the summation of the variable O&M and fuel
costs of existing portfolio resources and the total of
capital, variable O&M, fuel, and fi xed O&M costs of
new resources.
7-22
200 250 300 350 400 450
PRS
No Additions
No CO2
All Renew
Least Risk
75/25 Risk/Cost
50/50 Risk/Cost
25/75 Risk/Cost
Least Cost
All Coal
All Gas
Wind/Gas
Coal/Gas
Figure 7.22: 2016 Incremental Power Supply Expense ($millions)
Figure 7.21: 2007-16 Average Incremental Power Supply Expense–Induced Rate Increases (%)
2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
PRS
No Additions
No CO2
All Renew
Least Risk
75/25 Risk/Cost
50/50 Risk/Cost
25/75 Risk/Cost
Least Cost
All Coal
All Gas
Wind/Gas
Coal/Gas
7-23
1,200 1,300 1,400 1,500 1,600 1,700 1,800
PRS
No Additions
No CO2
All Renew
Least Risk
75/25 Risk/Cost
50/50 Risk/Cost
25/75 Risk/Cost
Least Cost
All Coal
All Gas
Wind/Gas
Coal/Gas
Figure 7.23: 2007-16 Incremental Power Supply Expense NPV ($millions)
Figure 7.24: 2007-16 Maximum Single-Year Rate Increase (%)
5 10 15 20 25
PRS
No Additions
No CO2
All Renew
Least Risk
75/25 Risk/Cost
50/50 Risk/Cost
25/75 Risk/Cost
Least Cost
All Coal
All Gas
Wind/Gas
Coal/Gas
7-24
The picture is similar for the NPV of incremental
power supply expenses. Figure 7.23 details the
statistics over the fi rst ten years of the IRP.
Rate increases generally are not level over time;
instead they refl ect the inherent lumpiness of
resource additions. The maximum rate impact in
the PRS is 9.4 percent in 2012. See Figure 7.24.
As with power supply expenses, those portfolios
relying exclusively on the wholesale marketplace
and natural gas- or coal-fi red resources perform
modestly better than the PRS under this measure.
All other strategies that include renewables and
modest levels of coal-fi red generation have rate
impacts that exceed the PRS.
7.6 Performance of PRS and
12 Resource Strategies In
Market Structure Scenarios
The Preferred Resource Strategy was compared to
18 market structure scenarios detailed in Section
6-Modeling Results. Similar scenarios are grouped
into categories.
Fuel Risk Scenarios
Volatile Natural Gas
The most interesting result of the Volatile Natural
Gas market scenario was its lack of impact on the
resource portfolios. Although the risk measures rose
across the board, the relative position of the various
portfolios did not change. See Figure 7.25.
Low Natural Gas Prices
When natural gas prices fall by 50 percent over the
IRP timeframe, many portfolios relying on the natural
Figure 7.25: 2007-16 Power Supply Expense Average Covariance–Volatile Gas (%)
13 14 15 16 17 18 19 20 21 22 23 24 25
PRS
No Additions
No CO2
All Renew
Least Risk
75/25 Risk/Cost
50/50 Risk/Cost
25/75 Risk/Cost
Least Cost
All Coal
All Gas
Wind/Gas
Coal/Gas
7-25
gas marketplace fare well when compared to the
Preferred Resource Strategy and other portfolios
without new gas generation. Incremental power
supply expenses in the All Gas portfolio fall from
$378 million under the Base Case in 2016 to $275
million, a change of just over $100 million. The PRS
experiences a correlated reduction due to the gas
resources already in the Company’s portfolio, but
the fi gure is much smaller: $38 million. Figure 7.26
compares the Base and Low Gas Cases.
High Natural Gas Prices
The High Natural Gas Prices market scenario
increases natural gas prices from the Base Case
by 50 percent. Instead of portfolios relying heavily
on gas-fi red generation performing well as in the
Low Natural Gas Prices case, the opposite occurs.
Incremental power supply expenses rise in the All
Gas portfolio mix from $378 million in the Base
Case in 2016 to $431 million, or by $53 million.
The PRS rises by a more modest $30 million.
Figure 7.27 provides a comparison across all of the
illustrative portfolios.
Many of the portfolios saw higher maximum single
year rate impacts under the High Gas scenario.
For example, the No Additions case saw its maximum
single-year increase rise by 5.3 percent, from 7.5
percent in the Base Case to just under 12.8 percent
under the High Gas scenario. Many of the portfolios
saw similar increases, including the PRS, which saw
its maximum one-year rate increase rise by around
four percent, from nine percent to 13 percent. Figure
7.28 displays the differences between the Base Case
and the High Gas market scenarios.
Figure 7.26: 2016 Incremental Power Supply Expense–Low Gas ($millions)
7-26
200 250 300 350 400 450 500
PRS
No Additions
No CO2
All Renew
Least Risk
75/25 Risk/Cost
50/50 Risk/Cost
25/75 Risk/Cost
Least Cost
All Coal
All Gas
Wind/Gas
Coal/Gas
Figure 7.27: 2016 Incremental Power Supply Expense ($millions)
Figure 7.28: Maximum Annual Rate Change from Base Case (%)
-3 -2 -1 0 1 2 3 4 5 6
PRS
No Additions
No CO2
All Renew
Least Risk
75/25 Risk/Cost
50/50 Risk/Cost
25/75 Risk/Cost
Least Cost
All Coal
All Gas
Wind/Gas
Coal/Gas
7-27
Carbon Emission Scenarios
The carbon emission scenarios illustrate how the
various resource mixes might perform were a
carbon-limited future imposed. Carbon emission
scenarios drive Company costs higher under any
future resource strategy that is pursued. Avista’s
existing portfolio of resources contains both coal-
and gas-fi red resources that emit carbon into
the atmosphere.
National Commission on Energy Policy
Carbon Emissions
The National Commission on Energy Policy (NCEP)
Carbon Emissions market future drives portfolio
power supply expenses up under all portfolio
options. Carbon emissions are forecast in this
future to begin at $7 per ton in 2010, rising in a linear
fashion to equal $15 per ton in 2026. Even the All
Renewables and No CO2 portfolios see increases
under this case due to the Company’s present
ownership of carbon-emitting resources.
The Preferred Resource Strategy remains
competitive under the NCEP Carbon Emissions
case; however, the No CO2 and All Renewables
cases become more competitive with the PRS.
Power supply expenses under the PRS are $41
million and $143 million higher under the NCEP
Carbon Emissions case than under the Base Case.
This equates to cost increases of 12 and 24 percent,
respectively. Figure 7.29 provides a comparison
of 2007-16 incremental power supply expenses
under both the Base Case and the NCEP Carbon
Emissions case.
SB 342 Carbon Emissions
The SB 342 Carbon Emissions market scenario
assumes carbon emission rates that on a present
value basis are three times the level of those
assumed in the NCEP Carbon Emissions scenario.
Prices start at $22 per ton in 2010, rising to $60 per
ton by the end of the IRP timeframe. Incremental
power supply expenses are signifi cantly higher
across the board in this market scenario,
as shown in Figure 7.30. The All Renewables
and No CO2 portfolios fared better under this
market scenario. Heavily coal-dependent resource
strategies did less well.
Avista-Centric Scenarios
Avista-centric scenarios are scenarios that do not
affect the marketplace but do affect the Company.
A number of the Avista-centric scenarios did not
affect resource acquisition. They were provided to
illustrate certain future conditions (for example, the
loss of the Noxon Rapids powerhouse for some
period of time). The impacts of these scenarios are
covered in Section 6– Modeling Results. The Avista-
centric scenarios were used to help develop the
Preferred Resource Strategy are discussed below.
Base Case Monte Carlo–No Production Tax Credits
Removing production tax credits drives the cost of
most example resource portfolios higher because
each of them contains a signifi cant amount of wind
and other renewables. Only the All Gas, All Coal,
Coal/Gas, and No Additions cases are insulated.
The PRS incremental power supply expenses,
measured as NPV between 2007 and 2016, rises by
1.6 percent or $24 million, from $1.47 billion to $1.49
7-28
Figure 7.29: 2007-16 Incremental Power Supply Expense NPV ($millions)
Figure 7.30: 2007-16 Incremental Power Supply Expense NPV ($millions)
7-29
-60 -40 -20 0 20 40 60 80 100 120
PRS
No Additions
No CO2
All Renew
Least Risk
75/25 Risk/Cost
50/50 Risk/Cost
25/75 Risk/Cost
Least Cost
All Coal
All Gas
Wind/Gas
Coal/Gas
Figure 7.31: 2007-16 Power Supply Expense NPV Change ($millions)
billion without production tax credits. The magnitude
over 20 years is about the same at $58 million, or 2
percent. See Figure 7.31.
Hydro Shift (90% Base Case Hydro)
Scenario
The Hydro Shift scenario reduces hydro capability
across the Western Interconnect by 10 percent. The
relative results of the various resource portfolios
were consistent with the Base Case results. Overall
average rate increases were lower on a percentage
basis due to average power supply expenses being
higher initially. The higher initial cost is due to the
loss of approximately 50 MW of Avista hydroelectric
generation in this case. 2016 power supply
expenses are shown to be approximately $15-$20
million higher in Figure 7.32.
7.7 Acquisition of
PRS Resources
The 2005 IRP envisions a diversifi ed mix of new
resources acquired beginning as early as 2007. Each
of the fi ve major categories of resource acquisition—
conservation, plant upgrades, wind, biomass, and
coal—is described below.
Conservation
The 2005 PRS relies on conservation to meet
69 aMW of future load growth by 2016 through
reductions in existing customer usage. Analyses
developed for the 2005 IRP found potential savings
in all customer sectors. The Company expects to
acquire this resource through both utility-sponsored
programs and programs acquired on its behalf by
third parties through an RFP process. An initial RFP
for conservation resources is expected to follow
Commission acknowledgement of the plan.
7-30
Coal Acquisition
New coal is forecast to enter the Company’s
portfolio in the 2012-15 timeframe at an initial level
of 250 MW. Similar to our assumptions around
wind and renewable resources, we believe that
bringing new coal-fi red resources into our portfolio
by 2012 will be a challenge. Lead times for green
fi eld coal development range between seven and
ten years. Some time might be shaved off of this
estimate were the Company to join with partners in
a project already under way. The Company will have
to remain fl exible when acquiring this resource given
the need to work with partners to gain necessary
economies of scale.
The amount of coal ultimately might be modestly
lower or higher than included in the PRS. In any
event, the Company will continue to evaluate
timelines for coal development and update its plans
as necessary. The Company does not expect
that its acquisition of coal-fi red generation will be
completed through a “traditional” RFP process
that includes “turn-key” bid prices. Instead, the
Company envisions a screening process that will
include a “due diligence” process comparing key
cost and feasibility factors between projects.
Wind Acquisition
Through study, the benefi t of geographical diversity
becomes evident. Having a single erected wind
turbine brings greater variability day-to-day, hour-to-
hour, and minute-to-minute, than erecting 100 wind
turbines at a given site. Similarly, spreading wind
turbine sites apart from one another geographically
appears to lessen variability. The Company believes
that taking modest ownership shares in multiple
wind projects will benefi t its customers by reducing
the variability within its wind generation portfolio.
This reduction in variability will lower integration
Figure 7.32: 2007-16 Incremental Power Supply Expense NPV ($millions)
250 270 290 310 330 350 370 390 410 430 450
Base Case
Hydro Shift
7-31
Northwest likely can be leveraged to provide
low incremental cost transmission capacity for
wind projects. Absent construction for coal-fi red
generation, the Company believes that transmission
costs would be too high to provide cost-effective
transfer of the extra-regional 250 MW of wind
generation planned in the PRS.
Biomass Acquisition
The performance of landfi ll gas and manure
biomass projects indicates that renewables besides
wind generation have the potential to meet future
customer requirements. The Company is hopeful
that it can acquire as much as 70 MW between the
years 2010 and 2016; however, the potential for
landfi ll gas and manure biomass could be limited.
Manure biomass, while having a signifi cant potential,
has not been proven on a large commercial scale.
Landfi ll gas also has limited potential given its fuel
source. The Company will continue to monitor the
potential for biomass resources.
7.8 Adjusted Energy and
Capacity Positions
With the addition of new PRS resources, the
Company ensures adequate resources for serving
customers through the IRP timeframe. Table 7.4
details the energy forecast, and resources planned
to meet it. The PRS envisions modest market
purchases during the IRP timeframe; they are
necessary to balance the level of annual load
growth with the lumpiness of resource acquisition.
costs, provide a higher level of dependable capacity,
and help lower power supply expense volatility.
To acquire a 400 MW portfolio of diversifi ed wind
generation assets by 2016 the Company might
begin acquiring this resource as early as 2007. The
early start date refl ects the Company’s belief that
acquiring 400 MW of wind from multiple projects
over a fi ve-year period beginning in 2010 may not
be possible. While this acquisition schedule might
bring new generation into the portfolio slightly
ahead of new load requirements, the level should be
modest and within an historical range of reasonable
utility surplus. An early start to wind resource
acquisition should assist the Company in the event
where coal-fi red generation acquisition slips beyond
2012. Acquisition of wind generation likely will occur
both within and without an RFP process, based on
Company experience with this resource.
Wind and Coal Link
Initial wind acquisition in the PRS is expected to
occur in the Northwest and within Avista’s service
territory. Tier 1 Northwest wind and local wind
projects are expected to require modest levels of
new transmission. This assumption is in-line with the
NPCC Fifth Power Plan. Later acquisitions of Tier 2
wind, and wind located outside the Northwest, likely
will require signifi cant new transmission investment.
Construction of new coal facilities outside of the
Northwest will provide an excellent opportunity for
wind resource development. The large transmission
investments necessary to import coal into the
7-32
2007 2008 2009 2010 2011 2016 2021 2026
Obligations
System Retail Load7 1,125 1,160 1,197 1,232 1,268 1,424 1,566 1,725
90% Conf. Interval 193 193 193 189 188 184 148 148
Total Obligations 1,318 1,353 1,390 1,420 1,456 1,608 1,715 1,873
Existing Resources
Hydro 510 510 506 487 483 464 447 444
Conservation 5 9 14 18 23 46 69 92
Net Contracts 234 234 234 235 131 104 57 57
Coal 182 193 181 181 193 181 181 193
Biomass 42 44 40 44 42 43 42 44
Gas Dispatch 282 268 282 272 282 268 282 272
Gas Peaking Units 145 145 145 141 145 142 146 132
Total Existing Resources 1,400 1,403 1,402 1,380 1,299 1,248 1,224 1,233
PRS Resources
New Conservation 2 5 7 9 12 23 35 46
Plant Upgrades 7 11 23 36 36 36 36 36
Wind 0 0 0 23 63 122 162 188
Other Renewables 0 0 0 0 16 65 97 145
Coal 0 0 0 0 0 215 302 388
Market 0 0 0 0 125 25 0 25
Total PRS Resources 10 16 30 68 251 486 630 828
Net Position 92 66 42 28 94 126 139 188
Table 7.4: Loads & Resources Energy Forecast with PRS (aMW)
7 Retail load is absent historical conservation acquisitions levels. Historical conservation levels are
counted as a resource. This treatment has no impact on power generation acquisitions going forward
• Figure 7.33 details the Company’s resource mix
graphically over time.
• Figure 7.34 details Company resource mixes of
energy in 2007, 2016 and 2026 graphically.
• Table 7.5 illustrates the Company’s capacity
forecast and resources forecast to meet it.
• Figure 7.35 provides capacity forecast and
resources graphically.
• Figure 7.36 explains the Company’s mix of
capacity resources in 2007, 2016, and 2026.
7-33
Figure 7.33: Preferred Resource Strategy—Energy (aMW)
1,000
1,100
1,200
1,300
1,400
1,500
1,600
1,700
1,800
1,900
2,000
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Existing Resources Coal
Wind Other Renewables
Conservation Upgrades
Market Load & Conf. Interval
Hydro
Coal
Gas
Contract
Wind
Other Renews
DSM
Upgrades
2007
2016
2026
Figure 7.34: Company Resource Mixes (% of Energy) 2007, 2016, and 2026
7-34
2007 2008 2009 2010 2011 2016 2021 2026
Obligations
Retail Load 1,704 1,754 1,799 1,860 1,898 2,137 2,343 2,573
Operating Reserves 260 265 269 274 278 299 317 338
Total Obligations 1,964 2,019 2,068 2,134 2,176 2,436 2,660 2,911
Existing Resources
Hydro 1,100 1,100 1,066 1,059 1,028 1,016 983 978
Conservation 5 9 14 18 23 46 69 92
Net Contracts 159 159 165 164 48 49 118 118
Coal 222 222 222 222 222 222 222 222
Biomass 50 50 50 50 50 50 50 50
Gas Dispatch 303 308 303 303 307 303 303 308
Gas Peaking Units 243 243 243 243 243 243 243 243
Total Existing Resources 2,082 2,090 2,062 2,059 1,920 1,928 1,988 2,010
PRS Resources
New Conservation 2 5 7 9 12 23 35 46
Upgrades 20 34 41 52 52 52 52 52
Wind8 0 0 0 19 50 100 138 163
Other Renewables 0 0 0 0 20 80 120 180
Coal 0 0 0 0 0 250 350 450
Market 0 0 0 0 125 25 0 25
Total PRS Resources 22 39 48 80 259 530 694 916
Net Position 140 110 42 5 3 22 21 14
Planning Margin 23.1% 20.7% 16.4% 13.9% 13.4% 15.7% 14.7% 14.0%
Table 7.5: Loads & Resources Capacity Forecast with PRS (MW)
8 Wind is presented as its contribution to meeting system peak. The IRP assumes a peak contribution for
wind of 25 percent. For example, the 100 MW value shown in 2016 equals 400 MW (400 x 25% = 100 MW)
7-35
Hydro
Coal
Gas
Contract
Wind
Other Renews
Conservation
Upgrades
2007
2016
2026
Figure 7.36: Company Resource Mixes (% of Capacity) 2007, 2016, and 202610
1,500
1,700
1,900
2,100
2,300
2,500
2,700
2,900
3,100
2007 2009 2011 2013 2015 2017 2019 2021 2023 2025
Market
Upgrades
DSM
Renewables
Wind
Coal
Existing Resources
Loads
Figure 7.35: Preferred Resource Strategy—Capacity (MW)9
9 Ibid
10 Ibid
7-36
8-1
This section reviews the
2003 IRP Action Plan
and provides an update
concerning how the
Company addressed
each item in the 2003
plan. The Action Plan
for 2005 provides details
concerning the research and actions the Company
will take as it prepares the 2007 IRP.
8.1 Summary Report for
2003 Action Plan
In the 2003 IRP, the Company listed several
activities to be accomplished during the two-
year planning cycle. The items in the Action
Plan included activities to further develop the
Company’s planning and resource acquisition
processes. The Action Items for 2003 are
listed below, followed by an explanation of the
Company’s progression for each item.
Public Process Action Items
Two action items were identifi ed to support and
develop the public process of integrated resource
planning. The action items follow:
1. Propose changes to the WUTC on the IRP/RFP
process that will provide improvements.
2. Continue to manage the free fl ow of information
with TAC participants.
Avista is working with state regulators to improve
the IRP/RFP process. In May 2005, the Company
participated in a hearing to make editorial revisions
to the Washington IRP and bidding rules and to
request additional language to address long lead-
time assets.
Avista continued to expand the Technical Advisory
Committee process by increasing the number
of meetings from three for the 2001 IRP to four
for the 2003 IRP to seven for the 2005 IRP. The
number of invitees has also increased from 53 in
the 2003 IRP process to 73 invitees on the current
list. The TAC meetings enhanced the Company’s
relationship with the academic community,
resulting in several additional meetings regarding
future collaboration with Washington State
University’s Program in Environmental Science and
Regional Planning. Avista is extremely grateful
for the core group of members who made a
sincere effort to attend most of the TAC meetings
and provided thoughtful and meaningful input.
However, we would like to increase the overall
number of stakeholders who actively attend TAC
meetings in the future. The current TAC members
will be queried about other people they would like
to have invited to the process in the future and
what changes we can make to the process to
improve attendance.
8. ACTION ITEMS
8-2
Conservation Action Items
The 2003 IRP identifi ed six areas in the DSM arena:
1. Evaluate the cost-effectiveness and resource
potential of conservation voltage reduction
(CVR) on the Company’s system;
2. Acquire electric resources that are at least
proportionate to the percentage of DSM
revenues being expended;
3. Field a DSM portfolio that continues to be cost-
effective on a societal and utility basis;
4. Prepare contingency plans for future emergency
responses to unexpected fl uctuations in
wholesale electric markets;
5. Prepare for a reevaluation of continued
participation in the Northwest Energy Effi ciency
Alliance upon expiration of the current contract
period (expiring at the end of 2004);
6. Convene a TAC meeting in the fall of 2003 to
discuss the various alternatives for integrating
DSM into the 2005 IRP process.
Avista has instituted a CVR pilot project at the
Francis & Cedar substation as part of a 17-site
regional evaluation of several different approaches
to voltage control. The project is funded and
sponsored by the Northwest Energy Effi ciency
Alliance. Avista’s project, along with other regional
pilots, has been delayed due to unexpected system
communication infrastructure issues. The NPCC
was also unable to include CVR in its Fifth Power
Plan for similar reasons.
The Company calculates cost-effectiveness as part
of our ongoing Triple-E Reporting process. The
portfolio has remained cost-effective since the last
IRP and is projected to continue to be cost-effective
into the future.
During 2001, the Company initiated an emergency
business plan for conservation operations that
resulted in the acquisition of over three times our
goal. The contingency plan for this response has
been re-initiated, on a much smaller scale, in our
2005 Drought Contingency Plan. This response is
an example of our continuing ability to respond on a
real-time basis to market conditions.
The Company has reevaluated its participation in the
Northwest Energy Effi ciency Alliance. Based on that
evaluation, Avista signed funding contracts with the
Alliance for an additional fi ve-year period.
In October 2003 Avista convened a joint meeting
of the IRP TAC and the Triple-E Board to discuss
issues relating to the future integration of
conservation into the IRP. Consensus achieved in
that meeting led to the integration methodology
used in the 2005 IRP.
Action Items for Supply Side
Resource Options
There were seven action items for the Supply Side
Resource Option area:
1. Pursue a new license for the Spokane River
projects by fi ling a new license application by
July 31, 2005;
2. Continue to evaluate the effects and costs of
integrating wind generation into the Company’s
electrical system;
8-3
3. Consider and evaluate the potential to add coal
facilities to the Company’s mix of existing
generating resources;
4. Determine the feasibility of entering into a
medium-term fi rm power sale during the
Company’s surplus years;
5. Initiate a study to determine the optimal reserve
margin for the Company, including the benefi ts
of additional peaking capacity;
6. Continue to assess the cost-effectiveness of
new resource additions;
7. Continue to work with Commission Staff on
methods whereby the Company can acquire
resources with development timelines beyond
one or two years and increase the probability for
full rate recovery.
The Company continues its pursuit of a new license
for the Spokane River projects. In July 2005 Avista
fi led a draft license and is hopeful that the process
will be completed before the existing license expires
in July 2007.
Wind integration and cost studies performed
since the 2003 IRP support the inclusion of 650
MW of wind generation into the Company’s 2005
IRP Preferred Resource Strategy (PRS). This
compares to 75 MW included in the 2003 IRP.
The studies found that integration costs can be
signifi cant at high penetration levels; however, at
lower levels costs can be more manageable. The
Company also learned that geographical wind
diversifi cation can help reduce wind risk, both
fi nancially and operationally.
Coal-fi red generation still makes a signifi cant
contribution to the PRS. The Company continues its
work with partners to solve the locational challenges
associated with this resource. During the past two
years the Company has reviewed proposals for six
coal sites.
The Company considered and ultimately rejected
signing a medium-term fi rm power sale based on
its resource position. Recent poor hydro conditions
have limited our surplus generation potential.
The Company has performed various analyses in
its effort to defi ne an optimal reserve margin. The
2005 IRP looked at sustained peaking capability and
concluded that our existing method of determining
planning margins will continue for at least the
next two years. Results of the sustained capacity
exercise did lead to questions that could not be
answered promptly. The 2005 Action Items include
further study on this signifi cant issue. Additionally,
the Company continues to work in the various
regional forums in this area.
Evaluating cost-effective new resource options is a
continual process. The 2005 IRP includes signifi cant
additional work beyond the 2003 plan to assess the
potential for various new resource alternatives. The
Company will continue this exercise going forward.
The Company believes that risks associated with
long lead-time assets may not be adequately
addressed in present regulations. We are actively
participating in regulatory proceedings that strive to
clarify this issue further.
8-4
Action Items for Resource
Management Issues
1. Analyze the uncertainty of decisions as the
Company confronts risks and opportunities.
2. Continue to assess the electric marketplace and
its effect on the Company.
The Action Items concerning resource management
issues are intertwined with the IRP process. The 2005
IRP built on work prepared for the 2003 IRP, further
enhancing the evaluation of market interactions across
the Western Interconnect. The Effi cient Frontier
provides another method to evaluate the PRS and
compare it to other resource portfolios.
8.2 Action Plan For 2005
The Company’s Preferred Resource Strategy provides
direction for long-term activities. The Company’s
2005 Action Plan outlines activities that will be
undertaken to support this strategy and improve
the planning process over the next two years.
Progress will be monitored and reported in Avista’s
2007 Integrated Resource Plan. Each item was
developed with the advice of the Technical Advisory
Committee or by Company staff during the IRP process.
Renewable Energy and Emissions
1. Commission a study to assess wind potential in
Avista’s service territory;
2. Continue to monitor emissions legislation and its
potential effects on markets and the Company;
3. Research clean coal technology and carbon
sequestration;
4. Asses biomass potential within and outside
Avista’s service territory;
5. Continue to study various, including local sites.
Modeling Enhancements
1. Evaluate 70-year water record for inclusion in
2007 IRP studies.
2. Add more functionality to the Avista Linear
Programming model (e.g., direct consideration
of cash fl ow and rate impacts versus after-the-
fact reviews).
Transmission Modeling and Research
1. 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;
2. Continue involvement in BPA transmission
business practice processes and rate
proceedings to minimize costs of integrating
existing resources outside of the Company’s
service area;
3. 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;
4. Evaluate costs to integrate new resources
across Avista’s service territory and from regions
outside of the Northwest.
Conservation
1. Review the potential for cost-effective load
shifting programs using hourly market prices.
2. Complete the conservation control project
currently underway as part of the Northwest
Energy Effi ciency Initiative for future evaluation
as a potential conservation resource.
Individual Contribution Contact
Clint Kalich,
Manager of Resource Planning & Analysis
Project Manager/
Modeler/Author [e-mail]
John Lyons,
Power Supply Analyst
Author/Editor/
Researcher [e-mail]
James Gall,
Power Supply Analyst Modeler/Author [e-mail]
Randy Barcus,
Chief Corporate Economist Load Forecast/Author [e-mail]
Jon Powell,
Partnership Solutions Manager
Conservation/
Author [e-mail]
Primary 2005 IRP Team
Other Contributors
PRODUCTION CREDITS
Bruce Folsom,
Manager of Regulatory Compliance
Bob Anderson,
Director of Environmental Affairs
Scott Waples,
Chief System Planner
Todd Bryan,
Power Supply Analyst
Ed Groce,
Manager of Transmission Planning
Bob Lafferty,
Manager of Wholesale Marketing & Contracts
Jeff Schlect,
Manager Transmission Services
Scott Steele,
Marketing Communication Manager
Steve Lester,
Database Administrator
Heidi Heath,
Power Supply Analyst
Dick Winters,
Gas Acquisition & Planning Administrator
Diane Thoren,
Assistant Treasurer & Director of Corporate Finance
George Perks,
Manager of Generation Joint Projects
Jessie Wuerst,
Communications Manager