20100310AVU to Staff 1-15.pdf

Avista Corp.
1411 East Mission P.O. Box 3727
Spokane. Washington 99220-0500
Telephone [phone]
Toll Free 800-727-9170
Rt.~r'"
J.~'V'STJI.
Corp.
iD\Û r\~.R \ 0 ~,,\O: 33
March 9, 2010
Idaho Public Utilities Commission
472 W. Washington
Boise, ID 83702-5918
Attn: Donald L. Howell
Deputy Attorneys General
Re: Production Request of the Commission Staff in Case Nos. A VU-G-09-06
Dear Mr. Howell,
Enclosed are an original and two copies of Avista's response to IPUC Staffs production request
in the above referenced docket. Included in this mailng is Avista's response to production
request 001 through 015. The electronic version of the response was emailed on 03/09110 and is
also being provided in electronic format on the CDs included in this mailing.
Also included is Avista's CONFIDENTIA response to PR 03C and 013C. These responses
contain TRAE SECRET, PROPRIETARY or CONFIDENTIAL information and are
separately filed under IDAPA 31.01.01, Rule 067, and Section 9-340D, Idaho Code. It is being
provided under a sealed separate envelope, marked CONFIDENTIA.
If there are any questions regarding the enclosed information, please contact me at (509) 495-
4584 or via e-mail atpaui.kimball~avistacorp.com
Sincerely,#7-#
Paul Kimball
Regulatory Analyst
Enclosures
AVISTA CORPORATION
RESPONSE TO REQUEST FOR INFORMATION
JUSDICTION:
CASE NO:
REQUESTER:
TYE:
REQUEST NO.:
IDAHO
A VU-G-09-06
IPUC
Production Request
Staff-001
DATE PREPARD: 03/08/2010WITSS: nla
RESPONDER: Greg Rah
DEPARTMENT: State & Federal Regulation
TELEPHONE: [phone]
REQUEST:
As par of demand forecasting, please explain the "demand influencing factors" that were
reviewed in an attempt to quantify customer growth and usage per customer. Where applicable,
please provide the executable electronic analysis of your review.
RESPONSE:
At the Company's first TAC meeting, we defined demand influencing factors as conditions that
directly affected core customer (natural gas) volume consumed and identified conceptually four
broad areas we considered for analysis which were 1) customer growth 2) customer mix shifts 3)
weather and 4) technology. With input from the TAC, we analyzed several factors as sensitivities
relative to our. reference case including high and low customer growth, alternate weather
standards, and significant adoption of compressed natural gas vehicles. We also discussed with the
TAC global warming, northern migration, stagnant (zero) growth, as possible sensitivities to our
reference case but did not specifically model these factors. Appendix 3.6 summarzes the factors
reviewed and the input assumptions used while Appendix 3.7 Sensitivities Section includes detail
discussion on the above demand influencing sensitivities we considered which are as follows:
Low & High Customer Growth - In our low customer growth Sensitivity, anual customer
growth rates under perform the reference rate of growth by 50% over our 20 year planing horizon
while anual customer growt rates exceed the reference rate by 50% in our high growth
Sensitivity.
1HDD Lower Weather Standard - Peak Day weather temperature is reduced by 1 heating
degree (Fahenheit) in each service region. This sensitivity, although instrctive on understanding
underlying incremental change in demand, is not used in any Scenaro.
Coldest Day 20yrs Weather Standard - Peak Day weather temperature reduced to coldest
average daily temperature (HDDs) experienced in the most recent 20 years in each region. Note
this sensitivity only affects our W A1ID, Medford and Roseburg service regions as Klamath Falls
and La Grande have experienced a coldest day on record within the last 20 years.
Compressed Natural Gas (CNG) Vehicles - CNG vehicles assumed to produce a 15%
cumulative incremental demand over our 20 year planng horizon. Our assumption utilzed
market consumption estimates from an independent analysis on CNG vehicle viability. The
analysis indicates significant challenges exist to widespread adoption but did provide a scenaro
for significant market penetration (10% in 10 years). Although we concur significant system
demand from CNG vehicle purchases in our service terrtories is unlikely at this time, we were
Response to Staff Request No.001
Page 2
motivated to run this sensitivity to lear how our system would respond to an emerging application
that would grow significant new natural gas demand. This sensitivity, although instructive on
understanding underlying incremental change in demand, is not curently used in any Scenaro.
Global Warming - Adjust the regional peak day weather temperatures lower to account for global
waring. Although we have developed analysis supporting adjustment to historical average daily
temperatures for our forecasted average daily temperatures, we searched unsuccessfully for
information that would provide a basis for adjusting peak day temperatures. Our data does suggest
more volatile temperatures recently but is inconclusive on a trend of lower (or higher) peak
temperatues. One TAC member provided information from a study that could not conclude global
waring influenced peak day temperatures. Another TAC member offered reliable assessments
of global warming applied to specific service regions would be challenging given local weather
dynamics and conjectured overall global waring weather dynamics might produce possible peak
day cooling trends for regions situated in transition areas. After discussion and feedback, we
determined that a reliable basis for global warming temperatue adjustment is too uncertain. We
also believe the Alternate Weather Standard sensitivity may encompass many possible demand
impacts for this sensitivity therefore we did not pursue further analysis.
Northern Migration - Economic and water issues in south western states spur increased
migration to Pacific Northwest states. After discussion, it was determined that the High
Customer Growth sensitivity would likely encompass this sensitivity's demand impacts
therefore we did not pursue further analysis.
Stagnant Growth - Current economic conditions spur much slower and possibly negative
customer growth rates for an extended perod with a retu to trend rates at some point. It was
noted that we have not experienced widespread negative growth in our actual recent data. Our
significant residential customer base has historically been very stable and not prone to extreme
boom or bust cycles in four of our five service regions. MedfordIoseburg would appear most
vulnerable to a severe impact though a sustained negative growth trend appears remote. Also
noted were the ver low long term growth rates in our Low Customer Growth sensitivity. After
discussion, it was determined that the Low Customer Growth sensitivity would likely encompass
this sensitivity's demand impacts therefore we did not pursue fuer analysis.
AVISTA CORPORATION
RESPONSE TO REQUEST FOR INFORMTION
JUSDICTION:
CASE NO:
REQUESTER:
TYE:
REQUEST NO.:
IDAHO
AVU-G-09-06
IPUC
Production Request
Staff-002
DATE PREPARD:
WITSS:
RESPONDER:
DEPARTMENT:
TELEPHONE:
03/08/2010
nla
GregRah
State & Federal Regulation
[phone]
REQUEST:
Please explain why peak day demand (Dthday) is expected to increase at a faster rate than anual
average daily demand (Dthlday) from 2009-2029.
RESPONSE:
Peak day demand contains a higher weather sensitive demand component versus base (non
weather) demand relative to the ratio for average daily demand. The historically derived demand
coefficients indicate weather sensitive demand has grown at a faster rate than base demand.
A VISTA CORPORATION
RESPONSE TO REQUEST FOR INFORMATION
JUSDICTION:
CASE NO:
REQUESTER:
TYE:
REQUEST NO.:
IDAHO
A VU-G-09-06
IPUC
Production Request
Staff-003
DATE PREPARD:
WITESS:
RESPONDER:
DEPARTMENT:
TELEPHONE:
03/0812010
nla
GregRah
State & Federal Regulation
[phone]
REQUEST:
On page 3.10, when specifyng the price elasticity factor used in the IR, it's mentioned that the
AGA price elasticity study compared favorably to your past estimates. Please provide an
explanation of how the AGA price elasticity study compared to your past estimates. In your
response, please provide executable electronic copies of both studies.
RESPONSE:
Please see Avista's response 003C, which contains TRAE SECRET, PROPRIETARY or
CONFIDENTIAL information and exempt from public view and is separately filed under
IDAPA 31.01.0l, Rule 067, and Section 9-340D, Idaho Code.
AVISTA CORPORATION
RESPONSE TO REQUEST FOR INFORMATION
JUSDICTION:
CASE NO:
REQUESTER:
TYE:
REQUEST NO.:
IDAHO
A VU-G-09-06
IPUC
Production Request
Staff-004
DATE PREPARD:
WITSS:
RESPONDER:
DEPARTMNT:
TELEPHONE:
03/08/2010
nla
GregRahn
State & Federal Regulation
[phone]
REQUEST:
On page 1.5, when modeling the "Expected Case" price elasticity, the IR states: "We have
assumed a low response to prices in our Expected Case, parly based on a conservative assumption
that tight economic conditions and declining real estate values may deter many customers from
investing in long-term capital intensive conservation measures in the near term." When
determining customer wilingness to invest in long-term capital intensive conservation measures,
do you quantify the impact of Federal Stimulus measures associated with energy efficiency? If
you do, explain how? If not, explain why?
RESPONSE:
Although our Expected Case assumed a low price elastic response, we did model additional
scenarios that assumed a higher price elastic response (see Updated Price Forecast section in
Chapter 7 pg 7.4). We also specifically analyzed two DSM sensitivities relative to our Expected
Case to better understand how DSM might be affected by uncertain economic conditions (Chapter
4 pg 4.10). The DSM Accelerated sensitivity assumed stimulus incented DSM measures were
acquired beyond what was assumed in our Expected Case and was quantified in Table 4.4 on page
4.11.
AVISTA CORPORATION
RESPONSE TO REQUEST FOR INFORMATION
JUSDICTION:
CASE NO:
REQUESTER:
TYE:
REQUEST NO.:
IDAHO
A VU-G-09-06
IPUC
Production Request
Staff-005
DATE PREPARD:
WITSS:
RESPONDER:
DEPARTMENT:
TELEPHONE:
03/08/2010
nla
GregRah
State & Federal Regulation
[phone]
REQUEST:
On page 1.12, one of the 2010-2011 Action Plan items is to "continue to monitor the discussion
around diminishing Canadian natural gas imports and look for signals that indicate increased risk
of disrupted supply." Since a substantial portion of the Company's supply comes from Canadian
sources, please explain the "signals" A vista wil monitor indicating increased risk of disrupted
supply.
RESPONSE:
The Company regularly monitors curent data, forecasts and commentary regarding Canadian
imports from various sources including the EIA, NEB, our consultant services, publications, and
seminars for indications of constrained supply from Canada. Since our 2007 Natural Gas IR, the
general consensus outlook for imports has improved primarly as a result of continuing positive
trends in British Columbia shale gas prospects and lowered expectations on oil sands demand for
natural gas.
AVISTA CORPORATION
RESPONSE TO REQUEST FOR INFORMATION
JUSDICTION:
CASE NO:
REQUESTER:
TYE:
REQUEST NO.:
IDAHO
A VU-G-09-06
IPUC
Production Request
Staff-006
DATE PREPARD:
WITSS:
RESPONDER:
DEPARTMNT:
TELEPHONE:
03/0812010
nla
GregRah
State & Federal Regulation
[phone]
REQUEST:
On page 2.2, the IR states: "Because our transportation only customers purchase their own
natural gas and delivery on our distrbution system is non-firm, we exclude these customers from
our long-term resource planing." Do you forecast the number of non-firm transportation
customers and plan for distribution to sere them? If you do, explain how? If not, explain why?
RESPONSE:
Avista does forecast the number of non-firm transportation customers for corporate budgeting and
revenue forecasting, however we do not include non-firm transportation customers for IR
planing purposes and do not plan incremental distrbution enhancements to sere their
incremental demand. Non-firm transportation customers pay a rate which is less than firm core
customers given this demand can be curailed when distrbution resources are needed to sere core
customers.
JUSDICTION:
CASE NO:
REQUESTER:
TYE:
REQUEST NO.:
REQUEST:
AVISTA CORPORATION
RESPONSE TO REQUEST FOR INFORMATION
IDAHO
A VU-G-09-06
IPUC
Production Request
Staff-007
DATE PREPARD:
WITSS:
RESPONDER:
DEPARTMENT:
TELEPHONE:
03/0812010
nla
GregRah
State & Federal Regulation
[phone]
Please provide a 2007 IR comparson for Idaho and Washington in a format similar to Figues 2.5
and 2.6.
RESPONSE:
Please see the residential and commercial graph tabs in the accompanying fie
STAFF PR 007-Attachment.
AVISTA CORPORATION
RESPONSE TO REQUEST FOR INFORMATION
JUSDICTION:
CASE NO:
REQUESTER:
TYE:
REQUEST NO.:
IDAHO
A VU-G-09-06
IPUC
Production Request
Staff-008
DATE PREPARD:
WITSS:
RESPONDER:
DEPARTMNT:
TELEPHONE:
03/0812010
nla
Randy Barcus
State & Federal Regulation
[phone]
REQUEST:
Please provide an explanation of how long-term Global Insight data was combined with local
knowledge about sub-regional construction activity, age, demographic trends, and historical data
to develop a 20-year customer forecast. Please provide the Global Insight data, sub-regional data,
and forecast analysis in executable electronic format.
RESPONSE:
A vista's 2009 Natural Gas IR Appendix 3 .1-Customer Forecasts provides the detailed forecasts
for the service area. The files underlying the charts and tables are as follows:
STAFF _PR_008-Attachment A;
STAFF _PR_008-Attachment B; and
STAFF PR 008-Attachment C.
A VISTA CORPORATION
RESPONSE TO REQUEST FOR INFORMATION
JUSDICTION:
CASE NO:
REQUESTER:
TYE:
REQUEST NO.:
IDAHO
A VU-G-09-06
IPUC
Production Request
Staff-009
DATE PREPARD:
WITSS:
RESPONDER:
DEPARTMENT:
TELEPHONE:
REQUEST:
03/08/2010
nla
GregRah
State & Federal Regulation
[phone]
Please provide a scatter plot comparson of daily demand and temperature for each class in a
format similar to Figure 3.2?
RESPONSE:
Figure 3.2 reflects daily historical gas flow at our city gates which we use as the first step in
developing base and weather sensitive demand coeffcients. City gate data is used because its daily
volume data can be directly compared to daily weather data. To derive weather sensitive demand
coeffcients we remove base demand from the total and plot usage by HDD in a scatter char. We then
apply linear regression to the data to captue the linear relationship of total daily usage to HDD. The
slopes of the resulting lines are our weather sensitive demand coeffcients. Because we do not captue
daily demand by customer class, to derive factors by customer class, we use allocations based on
customer biling data demand ratios and biling cycle HDDs.
To check class demand coeffcient reasonableness, we apply the calculated coeffcients to actual
customer count and weather data to back cast demand.
WAllO Residential Demand vs. Temperature
100 o80604020
Average Temperature (Fahrenheit)
160,000
140,000
120,000
100,000
80,000
60,000
40,000
20,000
o
-20
.cÕ
"
Response to Staff Request No. 009
Page 2
WAllO Commercial Demand vs. Temperature
100 20 o806040
Average Temperature (Fahrenheit)
90,000
80,000
70,000
60,000
50,000 .c
40,000 Õ
30,000
20,000
10,000
o
-20
WAllO Industrial Demand vs. Temperature
100 o80604020
Average Temperature (Fahrenheit)
6,000
5,000
4,000
3,000 .cÕ
2,000
1,000
0
-20
AVISTA CORPORATION
RESPONSE TO REQUEST FOR INFORMATION
JUSDICTION:
CASE NO:
REQUESTER:
TYE:
REQUEST NO.:
IDAHO
A VU-G-09-06
IPUC
Production Request
Staff-010
DATE PREPARD:
WITSS:
RESPONDER:
DEPARTMENT:
TELEPHONE:
03/08/2010
nla
GregRah
State & Federal Regulation
[phone]
REQUEST:
When modeling use per customer by class, and then checking the modeling coeffcients for
reasonableness, please explain why use of a ver low adjusted r-square statistic for July and
August is reasonable. In your response, please include an explanation of how this might impact
the modeled results and how it wil be reconciled in the next IR.
RESPONSE:
Base demand is not heat sensitive therefore we would not expect there to be a strong correlation
between heating degree days and demand. To calculate base demand we use historical demand for
July and August, when there is insignificant, if any, heating demand. We then calculate a per
customer base usage coeffcient as detailed in Appendix 3.3.
If you overestimate the amount of demand in base usage you underestimate your heat sensitive usage
and vice versa. In the case of underestimating heat sensitive usage you may need to acquire resources
sooner than anticipated and on the flip side by overestimating heat sensitive usage you may actually
acquire resources you don't need. To check class demand coeffcient reasonableness, we apply the
calculated coeffcients to actual customer count and weather data to back cast demand.
AVISTA CORPORATION
RESPONSE TO REQUEST FOR INFORMATION
JUSDICTION:
CASE NO:
REQUESTER:
TYE:
REQUEST NO.:
IDAHO
A VU-G-09-06
IPUC
Production Request
Staff-011
DATE PREPARD:
WITSS:
RESPONDER:
DEPARTMNT:
TELEPHONE:
03/08/2010
nla
GregRah
State & Federal Regulation
[phone]
REQUEST:
On page 3.5, it's mentioned that in Oregon, multiple weather stations are used for weather
normalization because the weather in locations where natural gas services are provided don't
correlate. However in easter Washington and norther Idaho, the only location used for
normalization is the Spokane airport. What is the correlation between the Spokane airport weather
station and other weather stations within Washington and northern Idaho where natural gas
services are provided? In your response, please include a comparson of how the Washington and
northern Idaho weather stations correlate compared to how the four weather stations in Oregon
correlate.
RESPONSE:
The attached file STAFF _PR_011-Attachment computed monthly correlation among our service
terrtories for weather data January 2004 through March 2009. Spokane and Lewiston had total
average correlation of .93 versus Spokane and the Oregon service areas ranging from .69 (Medford
and Klamath Falls) to .84 (LaGrande). Winter months average correlation for Spokane and
Lewiston was .90 while Oregon areas ranged from .61 (Medford) to .82 (LaGrande). This is
mainly attributable to the Oregon areas being geographically furter from Spokane where several
factors affect correlation including elevation differences and proximity to mountain ranges
(Cascades and Blue).
AVISTA CORPORATION
RESPONSE TO REQUEST FOR INFORMATION
JUSDICTION:
CASE NO:
REQUESTER:
TYE:
REQUEST NO.:
IDAHO
A VU-G-09-06
IPUC
Production Request
Staff-012
DATE PREPARD:
WITSS:
RESPONDER:
DEPARTMENT:
TELEPHONE:
03/08/2010
nla
GregRah
State & Federal Regulation
[phone]
REQUEST:
When preparng the peak day demand forecast, please explain why Februar 15th is the peak date
modeled for the IdaholW ashington service terrtories, but December 20th is the peak date modeled
for the Oregon serice terrtories. When reviewing the summar of record setting HDD for
IdaholWashington service tertories, the dates are more aligned with December.
RESPONSE:
The Company models February 15th as the peak date for our Washington and Idaho serice
terrtories and December 20th for Oregon in order to stress test our system related to storage. By
modeling the peak day late in the winter season we are able to assess how much we need in storage
to meet our peak day demand obligations.
AVISTA CORPORATION
RESPONSE TO REQUEST FOR INFORMTION
JUSDICTION:
CASE NO:
REQUESTER:
TYE:
REQUEST NO.:
IDAHO
A VU-G-09-06
IPUC
Production Request
Staff-013
DATE PREPARD:
WITSS:
RESPONDER:
DEPARTMNT:
TELEPHONE:
03/08/2010
nla
GregRah
State & Federal Regulation
[phone]
REQUEST:
Please explain how the 2005 Oregon conseration analysis completed by RLW Analytics was used
to extrapolate the technical potential in Washington and Idaho. In your response, please provide
the RLW Analytics conservation analysis and the corresponding executable electronic files
associated with extrapolating the Oregon results to Washington and Idaho.
RESPONSE:
Please see Avista's response 013C, which contains TRADE SECRET, PROPRIETARY or
CONFIDENTIA information and exempt from public view and is separately filed under
IDAPA 31.01.01, Rule 067, and Section 9-340D, Idaho Code.
AVISTA CORPORATION
RESPONSE TO REQUEST FOR INFORMATION
JUSDICTION:
CASE NO:
REQUESTER:
TYE:
REQUEST NO.:
IDAHO
A VU-G-09-06
IPUC
Production Request
Staff-014
DATE PREPARD:
WITSS:
RESPONDER:
DEPARTMNT:
TELEPHONE:
03/08/2010
nla
Lori Hermanson
State & Federal Regulation
[phone]
REQUEST:
On page 4.8, when describing the Nort Division's conservation goals, the IR states: "As more
participation occurs in specific applications and technologies, program implementers and
engineers use results to establish more prescriptive approaches in order to increase paricipation
without having to add additional infrastrcture." Please give examples ofthese approaches.
RESPONSE:
Examples of this approach would be programs such as prescriptive food service equipment or
prescriptive steam trap replacement where program managers have worked with engineers to
standardize either an incentive per type of application or a standardized incentive per therm saved
for that project. These types of projects used to be handled individually through the site-specific
program, however, after rebating a large enough sample of these types of projects, engineers are
able to develop a typical scenaro that can be applied to similar measures in the future. This
streamlines the process for customers as well as for the engineering and support staff to process
these projects from initiation to completion.
AVISTA CORPORATION
RESPONSE TO REQUEST FOR INFORMATION
JUSDICTION:
CASE NO:
REQUESTER:
TYE:
REQUEST NO.:
IDAHO
A VU-G-09-06
IPUC
Production Request
Staff-015
DATE PREPARD:
WITSS:
RESPONDER:
DEPARTMENT:
TELEPHONE:
03/0812010
nla
GregRah
State & Federal Regulation
[phone]
REQUEST:
On page 7.2, when modeling "Peak Day Demand by Case," the IR states: "The price increase in
2015 is a result of significant carbon cost adders for climate change policy going into effect,"
please explain how the carbon adder was modeled, how the 2015 estimated cost per ton of carbon
was chosen, and how the contingencies associated with your forecasts were identified.
RESPONSE:
In evaluating various price forecasts, we selected a price forecast from an independent consultant
(Consult! in Figure 6.4) for our medium case forecast. Their analysis included a base case forecast
and separate analysis for effects of carbon mitigation legislation. We also developed a second
carbon case based on data from a northwest peer utilty to capture a range of possible outcomes
with respect to carbon mitigation. Appendix 3.7 Sensitivities Section includes detail discussion on
the carbon case sensitivities we considered which are as follows:
Carbon Mitigation 1 - Utilzes carbon cost adders quantified by independent analysis from an
independent consultant (Consult 1). They identify both an adder reflecting carbon allowances as
well as an adder to captue the effect of increased natural gas demand as more gas turbines come
online to replace coal plants and back up wind generation. The allowance adder escalates from
$5/ton in 2012 to $67/ton by 2030 while the increased demand adder climbs from $.50/mmbtu to
$1.00 over our planing horizon.
Carbon Mitigation 2 - Recognizing significant uncertainty exists regarding the amount, scope,
and timing of carbon regulation, we utilze a second alternate range of cost adders to develop a
high carbon cost case. We escalate an allowance adder from $37/ton in 2012 to $ 140/ton by 2030
as forecasted in a Pacific Nortwest electric utilty's integrated resource plan. The increased
demand adder is consistent with our Carbon Mitigation 1 case.