snohomish county pud no. 1the timing and scale of carbon costs in each scenario affected the...

Snohomish County PUD No. 1

2017 Integrated Resource Plan

2018 through 2037

Adopted May 8, 2018

SNOHOMISH PUD FINAL 2017 INTEGRATED RESOURCE PLAN, ADOPTED MAY 8, 2018

ii


ii

TABLE OF CONTENTS

1 EXECUTIVE SUMMARY SECTION 1-1

GUIDING PRINCIPLES FOR 2017 IRP 1-2 LOAD RESOURCE BALANCE 1-2 HIGHLIGHTS OF THE 2017 IRP 1-4 2017 IRP SCENARIOS 1-6 WINTER ON-PEAK CAPACITY NEED 1-7 LONG TERM RESOURCE STRATEGY 1-9 PROPOSED ACTION PLAN 1-13

2 WHO WE ARE SECTION 2-1

LOAD GROWTH 2-2 CURRENT TRENDS 2-2 HISTORICAL PERSPECTIVE 2-3 OVERVIEW OF THE PUD’S PORTFOLIO 2-5

3 PLANNING ENVIRONMENT SECTION 3-1

PUD’S STRATEGIC PRIORITIES 3-1 THE ECONOMY – PUGET SOUND AND BEYOND 3-2 ELECTRIC INDUSTRY – INITIATIVES AND EFFORTS 3-2 ENERGY POLICY & REGULATORY REQUIREMENTS 3-7 FEDERAL COLUMBIA RIVER POWER SYSTEM 3-15

4 SCENARIOS AND PLANNING ASSUMPTIONS SECTION 4-1

SCENARIOS 4-1 SENSITIVITIES 4-6 LOAD FORECASTS 4-6 PLANNING ASSUMPTIONS 4-10

5 ANALYTICAL FRAMEWORK SECTION 5-1

IDENTIFYING FUTURE RESOURCE NEED 5-1 PROBABILISTIC ANALYSIS OF EXISTING LOAD RESOURCE BALANCE 5-2 PLANNING STANDARDS 5-5 RESOURCE OPTIONS 5-7 RESOURCE MODELING ASSUMPTIONS 5-23 OVERGENERATION EVENTS 5-24 UNBUNDLED RENEWABLE ENERGY CREDITS 5-24 DEVELOPING INTEGRATED PORTFOLIOS 5-27


ii

6 PORTFOLIOS AND PROPOSED LONG TERM RESOURCE STRATEGY SECTION 6-1

PORTFOLIO DEVELOPMENT 6-1 PORTFOLIO FINDINGS 6-3 PORTFOLIO RESULTS 6-4 DETERMINATION OF THE PROPOSED LONG TERM RESOURCE STRATEGY 6-12 PROPOSED LONG-TERM RESOURCE STRATEGY 6-16

7 KEY INSIGHTS AND PROPOSED ACTION PLAN SECTION 7-1

KEY INSIGHTS 7-2 PROGRESS ON 2015 IRP ACTION ITEMS 7-14 PROPOSED 2017 ACTION PLAN 7-15

APPENDICES (AVAILABLE ELECTRONICALLY UPON REQUEST)

Appendix A: Probabilistic Load Resource Balance Model & Determining Need

Appendix B: Portfolio Optimizer Model

Appendix C: Portfolio Results for Scenarios & Sensitivities

Appendix D: Climate Change Analysis

Appendix E: Snohomish PUD’s 2017 Conservation Potential Assessment (CADMUS Report)

Appendix F: Avoided Costs for Conservation Methodology

Appendix G: Glossary

Section 1: Executive Summary

Snohomish PUD - 2017 Integrated Resource Plan 1-1

1 EXECUTIVE SUMMARY

Integrated resource planning is a comprehensive process that considers how a utility will meet

its objective to provide reliable electric service to its customers at the lowest, reasonable cost

under a variety of futures. This process must also consider the risks and uncertainties inherent

in a rapidly changing and complex industry.

To achieve this objective, a range of

alternatives are considered. Accordingly, an

integrated resource plan (IRP) must be flexible

and allow the utility to adapt to changing

circumstances, without adverse financial

impacts.1

Key elements in an IRP include:

1. A variety of futures or scenarios the utility could face;

2. An analysis of the utility’s existing and committed resources under each scenario to

determine the potential range of future energy and capacity needs;

3. The types of demand-side and supply-side resources considered to be reliable and

commercially available over the study period to meet future need;

4. Candidate portfolios that identify the mix of resources to meet future energy and capacity

needs by scenario, based on lowest reasonable cost criterion;

5. A selected portfolio or resource strategy that best positions the utility to meet future needs;

and

6. A near term action plan with steps the utility can take to implement the plan over the next

two to four years.

The PUD’s 2017 IRP covers the 20-year planning horizon of 2018 through 2037.

1 Revised Code of Washington, Chapters 19.280 and 19.285 prescribe the statutory requirements of an integrated

resource plan



Guiding Principles for 2017 IRP

The guiding principles for the PUD’s 2017 IRP

effort were to:

1. Meet load growth first by pursuing all cost-

effective conservation;

2. Understand the probabilistic range of available

energy and capability from the PUD’s existing

and committed resources and the overall impact

on the load resource balance across the 20-year

study period;

3. For future load growth not met by the PUD’s

existing/committed resources and new conservation acquisitions, pursue clean, renewable

resource technologies. However, planning must take into consideration resource options

“that provide the optimum balance of environmental and economic elements.” 2

4. Comply with all applicable Board policies, regulations, state laws and established IRP

planning standards; and

5. Preserve the PUD’s flexibility to adapt to changing conditions..

Load Resource Balance

Under the 2017 IRP planning assumptions for the PUD’s existing and committed resources,3 the

PUD expects it will remain in a surplus average annual energy position across the 2018 to 2037

study period, after the acquisition of new cost-effective cumulative conservation. This remains

true for all 2017 IRP scenarios, except for the High Growth Load scenario. Figure 1-1 shows

the PUD’s 2016 actual monthly load and generation from its existing and committed resources.

2 The Board of Commissioners adopted its Climate Change Policy and Strategies in March 2007. Full text

available at http://www.snopud.com/AboutUs/environment/climate.ashx?p=1233 3 The 2017 IRP has assumed a subsequent BPA long-term power supply contract for the post-2028 period. See

Section 5 for planning assumptions for the BPA power supply contract.

http://www.snopud.com/AboutUs/environment/climate.ashx?p=1233



Figure 1-1

2016 Actual Snohomish PUD Load and Existing and Committed Resources

(by Month, in aMW)

The PUD has been a winter peaking utility that experiences its highest peak customer demand

during the on-peak periods during the winter months of November through February. The

highest customer demand typically occurs in the month of December. The PUD’s highest

recent peak winter demand occurred in December 2008 at 1560 MW, which is more than

double the 2016 actual average annual system demand of 748 aMW. The PUD’s all-time

summer peak occurred in July 2009 with a system demand of 946 MW.

0

200

400

600

800

1,000

1,200

Jan Feb Mar Apr May June July Aug Sept Oct Nov Dec

aMW

BPA Slice BPA Block Market Purchases

Wind Other Renewables PUD-Owned Hydro

PUD System Load



Highlights of the 2017 IRP

Two significant elements in the 2017 IRP analysis were the probabilistic analysis of the PUD’s

load and its existing and committed resources, and revisions to the planning standards and the

associated metrics used to measure resource need. Two new planning standards based on the

probabilistic analysis were used to measure the PUD’s future capacity need during the on-peak

hours of its historically most deficit month and most deficit week, across the 20-year study

period. In this way, candidate portfolios were developed to satisfy the PUD’s two most deficit

periods, 19 out of 20 times.4

Other highlights include:

New energy efficiency and conservation is the single largest resource addition for every

portfolio for each scenario. Conservation is estimated to serve 74% of the PUD’s future

load growth,5 resulting in the PUD having no annual energy need over the 2018 through

2037 study period under average hydrological conditions, except under the High Growth

scenario.6

The Monthly On-Peak and Peak Week planning standards that defined the capacity need

over the study period drives the need for capacity resource additions in all scenarios.

Short-term and long-term capacity resources7 provide seasonal and peak load matching

capabilities to augment the PUD’s owned and contracted resources and ensure the winter

planning standards are met 19 out of 20 times; these resource additions resulted in the

lowest cost portfolios.

4 Section 5 - Analytical Framework, details the Planning Standards used in the 2017 IRP analysis. 5 The amount of achievable economic potential identified over the 20 year study period as a percent of total

resource need under the Long Term Resource Strategy. 6 After new conservation additions, the PUD forecasts no average annual energy need through 2037 unless 1) poor

hydrological conditions occur/persist; 2) there is a fundamental change in Federal hydro operations that affect the

PUD’s long term BPA power contract; 3) the post 2028 BPA products differ from existing product offerings. 7 A capacity resource refers to a generator or source of electricity that can be turned on or off, or otherwise

adjusted up or down as needed (“dispatched”) at the request of power grid operator or plant owner. The fastest

plants to dispatch are hydroelectric projects with storage and natural gas.

https://en.wikipedia.org/wiki/Natural_gas



The PUD will continue to meet its Washington state annual renewables requirement through

a combination of renewable energy credits (RECs) from contracted for renewable resources,

incremental hydro, and RECs allocated through the BPA long-term contract. Given the

PUD’s forecast surplus annual energy position under average water conditions, procuring

some portion of compliance RECs from third parties in the post-2020 period was identified

to be the most cost effective way to meet renewables requirements at this time.8

The climate change analysis showed impacts to both the PUD’s existing resources and its

future resource needs. With changes expected to regional precipitation and temperature

patterns over time, winter needs are expected to gradually decline as a result of increased

hydro production during the November through February period. Summer needs will

increase over time with warmer temperatures and increased air conditioning load, while

spring and summer hydro production levels decline due to reduced snowpack.9

The timing and scale of carbon costs in each scenario affected the forecast market price and

the fuel costs for certain supply side resource options available to the candidate portfolios.

The higher the level of carbon pricing, the further up the conservation supply curve (cost)

the model acquired. In general, the larger the amount of total new cumulative conservation

found to be cost effective for the portfolio, the later in the study period new supply-side

resource additions occurred, and the addition was typically smaller in size.

In the event of an unanticipated need for generating resources, the 2017 IRP identified

numerous commercially available and reliable resources. If and when a new resource

addition is determined to be needed, the PUD would perform a thorough due diligence

review, and conduct a comprehensive economic evaluation for a site specific resource.

8 The 2017 IRP Action Plan contains an action item to develop a least cost approach to meet its annual state

renewables requirement, including monitoring applicability of the no load growth and financial cost cap methods,

as well as procuring RECs from third parties for eligible renewable resources situated in Washington, Oregon and

Idaho. 9 The Climate Change analysis is based on regionally downscaled data consistent with Representative

Concentration Pathway 4.5 of the UN Intergovernmental Panel on Climate Change’s Fifth Assessment Report

(2014).



2017 IRP Scenarios

The 2017 IRP utilized five scenarios that considered the range of possible futures the PUD

could face for the 2018 through 2037 study period. Figure 1-2 below summarizes the primary

socio-economic drivers considered by the five scenarios evaluated in the 2017 IRP analysis:10

Figure 1-2

Snohomish PUD’s 2017 IRP Scenarios

Primary

Socio-Economic

Drivers

Low Load

Growth

Business as

Usual with

No Carbon

(BAU)

Business a-Usual

with California

Carbon in 2022

High Load

Growth

Climate Change

Avg Load Growth 0.5% 1.1% 1.1% 2.3% 1.0%

Natural Gas Price

($/MMBtu) $2.89-$5.80 $3.25-$6.64 $3.54 - $8.60 $3.83 - $10.56 $2.89-$5.80

Electricity Prices11

($/MWh) $37.43-$81.71 $27.59-59.24 $27.17-$115.47 $68.03-$163.86 $37.43-$81.71

Carbon Price

($/ton) $13.19 -$29.95 $0.32 $20.68-$59.27 $43.96-$89.84 $13.19-$29.95

Electric Vehicle

Adoption12 1-9 aMW 1-25 aMW 1-25 aMW 3-60 aMW 1-25 aMW

Indoor Ag,

Cannabis Load13 3-4 aMW 4-22 aMW 4-22 aMW 5-41 aMW 4-22 aMW

Weather14 Normal Normal Normal Normal Climate Change

3 degrees15

10 The 2017 IRP scenarios are described in Section 4. 11 Section 4 describes the planning assumptions associated with natural gas, carbon and market price forecasts. 12 Electric vehicle adoptions rates were derived from a 2017 joint study performed by Energy and Environmental

Economics (E3), “Economic & Grid Impacts of Plug-In Electric Vehicle Adoption in Washington & Oregon,”

March 2017, sponsored by Snohomish PUD, Chelan County PUD, Puget Sound Energy, Tacoma Power, Avista

Utilities and Seattle City Light. 13 Informed by data gathered from Washington State Liquor & Cannabis Board found at https://lcb.wa.gov/. 14 Normal weather is based on the PUD’s actual historical weather for the period 1991-2015. 15 The Climate Change weather forecast assumed 1 degree warming has already occurred and 2 degrees are

expected by 2037; estimate informed by regional downscaling of Representative Concentration Pathway 4.5 from

the United Nation’s Intergovernmental Panel on Climate Change’s 5th Assessment Report. See Appendix D for

additional Climate Change information.

https://lcb.wa.gov/



Winter On-Peak Capacity Need

Similar to the adopted 2013 IRP and the adopted 2015 Update to the 2013 IRP, the 2017 IRP

analysis shows that after all cost-effective conservation is acquired, a long term capacity need

exists in all of the scenarios except for the Low Growth scenario. In the Low Growth scenario,

a short term capacity product meets the near term 2018 through 2022 need, with new

cumulative conservation meeting the long term need after 2022.

In general, the 2017 IRP analysis found that the acquisition of new cumulative conservation

defers the PUD’s need for any new long-term capacity resources until the late 2020’s. The

addition of a long term capacity resource is expected to meet seasonal and peak loads and also

serves to limit the PUD’s exposure to price volatility and delivery risk associated with the short

term energy market.16 The 2017 IRP analysis does not select or predict what long term capacity

resource may eventually be acquired to meet this forecast seasonal load and peak matching

need. The renewable resources predominantly available in the Northwest today (wind and

solar) do not possess the operating characteristics necessary to meet the PUD’s on-peak capacity

need at this time, in a reliable and cost competitive manner.17 Staff does expect continued

advancement in demand response, energy storage, and pumped hydro storage – technologies

that hold tremendous promise for the future.

The 2017 IRP’s candidate portfolio for each scenario selected varying amounts of new

conservation, and some combination of supply-side resources that provided capacity to meet the

PUD’s future seasonal and peak loads. Figure 1-3 illustrates the PUD’s December On-Peak

load resource balance for each scenario, before the addition of any new conservation. The bars

represent the PUD’s existing and committed resources, and the gap between those bars and load

16 A Winter Capacity Product (WCP) was added to the list of dispatchable resource options for candidate portfolios

to evaluate as part of the adopted 2015 IRP Update. The WCP represented a short-term seasonal contract for firm

energy and capacity, backed by a specific generating resource. The levelized cost of the WCP for IRP modeling

purposes was based on the long-term fixed costs of a simple cycle combustion turbine acquired for the November

through February period, for each year of the study. The 2015 IRP Update’s Preferred Portfolio added a 25 MW

capacity resource beginning in 2021. 17 Carnegie Institution for Science study of meeting U.S. electricity needs with wind and solar, study summary at

https://www.brightsurf.com/news/article/022718450726/wind-and-solar-could-meet-most-but-not-all-us-

electricity-needs.html



forecast for each scenario represents the load resource balance deficit that could be expected

one out of 20 time, expressed as the P5 or 5% probability of occurrence.18

Figure 1-3

Snohomish PUD’s December On-Peak Load Resource Balance with

Existing/Committed Resources before New Conservation by Scenario (in aMW)

The new resource additions by portfolio for the five scenarios is summarized in Figure 1-4:

Figure 1-4

Summary of Resource Additions by Scenario (in aMW)

Scenario

Total

Cumulative

Conservation

(aMW)

Short Term

Capacity Contract

(Dec HLH aMW)

Long Term

Capacity

(Dec HLH

aMW)

Renewables

(aMW)

RECs

(aMW)

Low Growth w/Low Carbon 121 25 n/a n/a 68

BAU w/No Carbon 92 25 232 n/a 78

Climate Change w/Low Carbon 114 50 116 3 68

BAU w/California Carbon 2022 152 n/a 97 1 72

High Growth w/Mid High Carbon 152 n/a 396 68 22

18 The PUD’s load resource balance for the December on-peak period were modeled probabilistically across

multiple scenarios and time periods. The Probabilistic Load Resource Balance Model is detailed in Appendix A.



Long Term Resource Strategy

The two primary sources of uncertainty and risk that most affected the 2017 IRP candidate

portfolios were the timing and level of carbon policy, and the impacts related to climate change.

The candidate portfolio that best addressed carbon uncertainty in the absence of a defined state

or federal policy at the time of this analysis, and the uncertainty associated with regional climate

change, is the Climate Change scenario with the low carbon pricing.19

The Climate Change scenario with the low carbon cost and market environment was determined

to be the PUD’s 2017 IRP Long-Term Resource Strategy.20 This strategy pursues the

acquisition of 114 aMW of cumulative conservation over the 20year study period as the

resource of choice. The acquisition of conservation defers the need for a long term capacity

resource addition until 2028. This delay provides the PUD with the time and flexibility to

further assess the impacts of any future carbon legislation and determine the type of load and

peak matching resource or resources it wishes to consider, including the timing or development

of any future resource addition.

Carbon Policy Uncertainty

Staff evaluated the Climate Change scenario under various carbon policies to gain an

understanding of the range of potential impacts on the candidate portfolios. To mitigate carbon

uncertainty until more is known about a future carbon policy, the Low Societal Cost of Carbon,

effective in 2018, was selected.

Climate Change Uncertainty

The University of Washington’s Climate Impacts Group has performed extensive work on

regional climatology. The Climate Impacts Group suggests that across time, the region will

19 Beginning in 2018, the PUD incorporated elements of climate change into its energy risk portfolio management

processes and financial planning and ratemaking purposes. 20 Section 5 details the IRP’s Analytical Framework, Section 6 details portfolio results and selection of the Long

Term Resource Strategy.



experience milder winters with increasing amounts of precipitation and less snowpack. The

expectation is for increases over time in winter hydroelectric production, and reductions over

time to summer hydro production, due to lack of snowpack build and reduced spring snowmelt.

The forecast impact on the PUD’s load resource balance is a winter capacity need that

gradually decreases in the post-2027 period, while a new summer need begins to emerge in the

late 2030’s, with warmer temperatures and increases in air conditioning load, combined with a

regional decline in summer hydro production.

Figure 1-5

Forecast Impact of Climate Change on PUD’s Summer and Winter

Load Resource Balance (in aMW)

(400.0)

(300.0)

(200.0)

(100.0)

-

100.0

200.0

Net

On

-Pea

k P

osi

tio

n a

t P

5 (

in a

MW

)

2017 HLH @P5 2027 HLH LRB @P5 2037 HLH LRB @P5



Planned Resource Additions

The incremental planned demand-side and supply-side resource additions to serve the December

On Peak need in the Long Term Resource Strategy are shown in Figure 1-6, with load growth

reflecting the effects of climate change, over time.

Figure 1-6

2017 IRP Long Term Resource Strategy

December On-Peak (in aMW)

The addition of 114 aMW of new cost-effective conservation over the 20-year planning

horizon, with a 10-year conservation potential of 92.7 aMW by 2027. The expected 20-

year cumulative contribution of new conservation to the winter on-peak period is 152 aMW.

A 50 MW short-term capacity contract for the 2018 through 2022 period addresses near

term seasonal capacity needs.

Distributed generation renewable resources like rooftop solar and other technologies,

bundled with the associated environmental attributes help diversify the PUD’s portfolio and

aid the PUD in meeting the annual renewables requirement.

2018 2020 2022 2024 2026 2028 2030 2032 2034 2036

Forecast Market Purchase 96 80 57 97 87 - - 15 23 36

Distributed Generation - 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4

Short-Term Capacity Contract 50 50 50 - - - - - - -

Long-Term Capacity Resource - - - - - 116 116 116 116 116

Conservation 12 36 61 86 112 129 133 141 146 150

Existing Resources 869 895 908 915 918 908 905 898 899 904

Load 1,027 1,061 1,076 1,098 1,118 1,130 1,150 1,170 1,184 1,207

0

200

400

600

800

1,000

1,200

1,400

aMW



The long-term capacity resource addition for seasonal load and peak matching is

deferred until 2028, as a result of new cumulative conservation and the benefits to the

winter on-peak period. The long term capacity addition also addresses the emerging

summer need identified late in the study period. This deferred need allows the PUD time to

investigate and evaluate the various types of long-term capacity and demand response

resources that may be available to meet this need, including future BPA products.

Procure unbundled RECs to meet forecast annual State renewables compliance

requirements. The PUD’s forecast REC need across the study period varies from

approximately 450,000 in 2021, to approximately 960,000 in 2037.

Regulatory Planning Standard

The 2017 IRP analysis and Long Term Resource Strategy comply with the state Energy

Independence Act’s (EIA), consistent with RCW Chapter 19.285, for: 1) cost-effective

conservation identified through a utility specific analysis to determine the 10-year conservation

potential over a range of forecasts; and 2) serving at least 9% and then 15% of the PUD’s total

annual retail load with a combination of eligible renewable resources, renewable energy credits

(RECs) or a combination of both.21 The 2017 IRP portfolio analysis also considered the impacts

of overgeneration, consistent with RCW 19.280.030(1)(e).

21 Based on Revised Code of Washington, Chapter 19.285.050, and per the prescribed methodology in Washington

Administrative Code, Chapter 194.37.070.



Proposed Action Plan

The 2017 Integrated Resource Plan has identified several near term actions to ensure the PUD

can meet the needs of its customers in a rapidly changing environment, well into the future:

1. Pursue all cost effective conservation and further explore capacity contributions, including

the feasibility of demand response as a future utility-scale capacity resource for the PUD.

2. Explore the best mix of resource alternatives in the Northwest for capacity resources to

meet peak needs, including the ongoing evaluation of battery and pumped hydro storage,

and discussions with BPA for seasonal load and peak matching products.

3. Ensure customer owned and distributed renewables programs are complementary to the PUD’s

overall power supply portfolio strategy.

4. Develop a least-cost renewables compliance approach to meeting the state’s renewables

requirements under the Washington Energy Independence Act (EIA).

5. Enhance short and long-term resource portfolio modeling capabilities; expand cost and risk

tradeoff analyses.

6. Conduct an internal survey about the IRP to determine how the reference document is used;

validate key findings and incorporate into PUD’s next IRP process.

7. Re-assess the methodology used to determine the value associated with the deferral of PUD

transmission and distribution investments; monitor the Northwest Power & Conservation

Council’s regional review of same.

8. Continue to participate in regional forums and assess impacts associated with climate

change, reduction in greenhouse gas emissions, renewable portfolio standards, and regional

power and transmission planning efforts.



Organization of the Document

The organization of the 2017 IRP document is as follows:

Section 1 is this Executive Summary.

Section 2 describes the PUD, including current load forecast and trends, existing and

committed power supply resources, and demand side programs.

Section 3 discusses the industry’s changing dynamics and planning environment, including

recently adopted or proposed legislation that may affect utility operations and costs. These

set the stage for the IRP planning process.

Section 4 details the scenarios, range of forecasts and planning assumptions incorporated in

the 2017 IRP analysis.

Section 5 summarizes the analytical framework and planning standards used to examine the

PUD’s load resource balance and identify future resource need.

Section 6 describes the portfolio results by scenario and selection of the Long Term

Resource Strategy.

Section 7 describes the key insights of the 2017 IRP analysis and the near-term Action Plan

to implement the selected Long Term Resource Strategy.

Section 2: Who We Are


2 WHO WE ARE

The Public Utility District No. 1 of Snohomish County (the PUD) began utility operations in

1949 by purchasing the electric distribution facilities for Snohomish County and the Camano

Island portion of Island County from Puget Power & Light. The PUD is the 12th largest

public utility in the U.S. and the second largest in Washington state serving more than

341,000 electric customers and about 20,000 water customers.

The PUD is committed to delivering the best possible service, keeping rates competitive and

maintaining the highest levels of reliability for our customers. As stewards of critical

community resources, the PUD takes its responsibility seriously.

The PUD is governed by a Board of Commissioners, which is composed of three members.

They represent separate commissioner districts, and are elected at-large for staggered six-

year terms. The legal responsibilities and powers of the PUD, including the establishment of

rates and charges for services rendered, reside with the Board of Commissioners. The PUD

is a not-for-profit utility and takes great pride in serving our customers in our community.

Figure 2-1

Snohomish County PUD’s Service Area



Load Growth

The PUD’s load growth from 1970 to 2017 averaged 1.7% annually, with residential,

commercial and industrial loads growing at average annual rates of 1.8%, 3.7% and -0.8%

respectively, as shown in Figure 2-2. From 2010 through 2016, the PUD acquired 73 average

megawatts of new conservation, or roughly 10% of the PUD’s total retail load. For the 2008

through 2017 period, after new conservation, the PUD’s average annual rate of load growth

was -0.5%.

Figure 2-2

Snohomish PUD’s Historical Annual MWh Retail Sales

Current Trends

The current economic environment for Snohomish County and Washington state is

exceedingly strong. In 2016, Washington’s real GDP grew at a rate of 3.7%, the fastest of

any state in the nation. Snohomish County itself ranked second nationally behind Pierce

County in the net number of people moving into the county at 10,500, with historically low

4.0% unemployment rate at the end of the year. In 2016 and 2017, the PUD connected an

average of 4,100 new premises per year, as compared to the 2,200 per year pace seen during

0

1,000,000

2,000,000

3,000,000

4,000,000

5,000,000

6,000,000

7,000,000

8,000,000

MW

h

Residential Load Commercial Load Industrial Load Total Load



the 2011 and 2012 timeframe. This pace is expected to slow to a long-term trend of

approximately 2,800 new premises per year by 2021.

Snohomish County’s main employment base is in aerospace manufacturing, primarily

Boeing’s Everett Plant, and hundreds of small aerospace companies delivering parts for the

747, 767, 777, and 787 programs. Naval Station Everett, Snohomish County and Providence

Hospital are major employers in the region. Growth also continues in the biotech sector in

South Snohomish County, and continued changes to the manufacturing sector in the Everett

area. The Port of Everett’s development of the Waterfront Place Central and Riverfront is

also underway and is expected to provide jobs and easy access to the waterfront. This effort

located east of downtown Everett will transform the waterfront into a sustainable and unique

commercial, recreation, and residential community.22

Historical Perspective

Figure 2-3 shows that historically, the PUD’s total retail sales rebound and resume their

prior, upward slope, following recessionary periods (see first two recession periods circled).

Given the strong economic conditions in Snohomish County and Washington state, the

historical trend would suggest the same would occur after the third recession. Instead,

recovery from the recent “Great” recession has been markedly different for the PUD; its

retail sales have remained flat through year 9 of the recovery. The flattening of retail sales in

recent years is likely due to a number of factors – some new to the post-recession world –

and others, such as culminations of decades of energy efficiency acquisitions, and the

growing impact of building codes and standards improvements.

22 Section 2 – Who We Are, discusses the PUD’s load forecast methodology and current trends. Section 4 –

Scenarios and Planning Assumptions, describes the various future socio-economic factors and elements

considered in the five different scenarios studied in the 2017 IRP analysis.



Figure 2-3

Historical and Base Case Demand Forecast Retail Sales by Sector

Before New Conservation (in aMW)

The PUD expects to experience positive load growth for the foreseeable future. This is a

reflection of population inflows and strong economic conditions in the Puget Sound area.

After acquiring the forecast level of new cost effective conservation, staff anticipates the

recent trend of flat to declining retail sales will persist, as detailed in Figure 2-4.

Figure 2-4

Historic Snohomish PUD Load by Sector in Annual MWh

0

100

200

300

400

500

600

700

800

900

1,000

19

69

19

72

197

5

19

78

19

81

19

84

198

7

19

90

19

93

19

96

19

99

20

02

20

05

20

08

20

11

20

14

20

17

20

20

20

23

20

26

20

29

20

32

20

35

Ave

reag

e M

egaw

atts

Residential Commercial Industrial Total Retail Sales

Recessions with Rising Retail Electricity Costs

0

1,000,000

2,000,000

3,000,000

4,000,000

5,000,000

6,000,000

7,000,000

8,000,000

MW

H

Industrial Commercial Residential Total Forecast (No Conservation)

Total Without New Conservation

Total With New Conservation



Overview of the PUD’s Portfolio

The PUD relies on a diversified power portfolio consisting of a broad range of conservation

and energy-efficiency programs, a long-term power supply contract with the Bonneville

Power Administration (BPA), PUD-owned hydroelectric projects, and several long-term

renewable power supply contracts. The PUD buys and sells power in the short-term energy

market to balance daily and seasonal variations in its customer loads and owned and

contracted resources.

In 2016, the BPA contract provided over 82% of the PUD’s power needs, primarily sourced

from the Federal hydro system;23 nearly 7% from the PUD’s owned hydroelectric

resources;24 approximately 7% from a combination of long-term wind contracts and

customer-owned, renewable distributed energy resources; and approximately 4% came from

short-term market purchases (Figure 2-5). The PUD’s 2016 Fuel Mix is 98% carbon free.25

Figure 2-5

23 BPA markets the output of the Federal Columbia River Power System and delivers firm power to the PUD at

cost, under its long-term contract for the Block and Slice products. 24 PUD-owned hydroelectric resources include: 112 MW Jackson Hydroelectric Project; 7.5 MW Youngs Creek

Hydroelectric Project; .65 MW Woods Creek Hydroelectric Project; and a 20% share of the 27 MW Packwood

Lake Hydroelectric Project, located in Packwood, WA. 25In accordance with RCW 19.29A.060, the PUD reports its fuel mix annually to the Washington State Department

of Commerce. The PUD’s 2016 annual fuel mix report can be found at http://www.commerce.wa.gov/growing-

the-economy/energy/fuel-mix-disclosure/.

2016 PUD Portfolio 2016 BPA Fuel Mix

http://www.commerce.wa.gov/growing-the-economy/energy/fuel-mix-disclosure/

http://www.commerce.wa.gov/growing-the-economy/energy/fuel-mix-disclosure/



The shape of the PUD’s load resource balance is an important consideration in long term

resource planning. The PUD’s loads have been historically highest during the winter, while

existing and committed resources have produced more energy in the spring. The net result is

monthly energy surpluses and deficits the PUD must manage. Figure 2-6 illustrates the shape

of the PUD’s 2016 actual load and existing resources:26

Figure 2-6

2016 Actual Firm Monthly Loads with PUD’s Existing/Committed Resources (in aMW)

The dotted line in Figure 2-6 shows the PUD’s average load by month during calendar year

2016. The PUD’s annual load shape is driven largely by electric heating loads during the

winter months – this has historically made the PUD a “winter peaking” utility. Monthly,

daily and hourly energy imbalances are balanced by selling or purchasing energy from the

short-term wholesale power market. The majority of market purchases in 2016 were made

during the winter period when resource supply timing did not always match the increased

customer need on an hour-to-hour basis, even though it was sufficient on an average monthly

basis.

26 Water Year 2016 as measured at The Dalles was 96% of average for the Jan-July period, based on the 1960-

2018 period.

0

200

400

600

800

1,000

1,200

Jan Feb Mar Apr May June July Aug Sept Oct Nov Dec

aMW

PUD-Owned Hydro Other Renewables Wind

Market Purchases BPA Block BPA Slice



Existing & Committed Resources The PUD relies on a portfolio of resources to meet customer demands. These include:

Supply side resources

o BPA power contract

o PUD-owned generating resources

o Long-term renewable power supply contracts

o Small renewables program and customer-owned generation

o Short-term market purchases

o Regional transmission contracts

Demand side resources

o PUD energy efficiency programs

o Demand response programs

Existing Supply Side Resources

BPA Power Contract

The PUD meets its load obligations by managing the energy available from the BPA power

contract in concert with its owned resources and other long-term power supply contracts.

The BPA is a revenue-financed federal agency under the Department of Energy that markets

wholesale electricity to more than 135 utility, industrial, tribal and governmental customers

in the Pacific Northwest. Its service area covers more than 300,000 square miles with a

population of approximately 12 million in Idaho, Oregon, Washington and parts of Montana,

Nevada, Utah and Wyoming.

The BPA sells, at wholesale rates, electric power generated from 31 federal hydroelectric

projects in the Columbia River basin, including one nonfederal nuclear plant and several

other small nonfederal power plants. The federal hydroelectric projects and the related

electrical system are known collectively as the Federal Columbia River Power System (the

“Federal System”), which has an expected aggregate output of approximately 9,089 annual



average megawatts under average water conditions and approximately 8,135 annual average

megawatts under adverse water conditions. The Federal System produces more than one-

third of the region’s electric energy supply.

Block-Slice Product

The PUD currently purchases the “Block-Slice” product from BPA for the contract term of

October 1, 2011 through September 30, 2028. The PUD purchases more than 80% of its power

supply from the BPA under this long-term power contract. The Block-Slice product is a

combination of two energy products:

Block Product: The Block product provides the PUD with power in flat monthly amounts

that are determined based on the PUD’s average monthly load. In 2017, the Block product

provided the PUD with 495 aMW in January when customer heating demand is seasonally

high, while in June when temperatures are more moderate, the Block amount is 346 aMW.

For all of 2017, the PUD received 3,363,849 MWh from the Block product.

Slice Product: The Slice product provides the PUD with variable amounts of power that

reflect the output of the Federal System. The PUD takes responsibility for managing this

product within the hourly contractual constraints and physical limits of the Federal System.

This product provides the PUD with the ability to follow its customer loads and resources by

storing and dispatching energy. The majority of the PUD’s short-term wholesale market sales

are from surplus Slice energy, which varies with the seasonal and daily variations in the Slice

product’s output. If snowpack and water conditions are above average in the region, the

energy output is also above average. If snowpack and water conditions are low, the PUD’s

energy supply is correspondingly reduced.

Every two years, BPA determines the total of its customers’ loads and the size of the Federal

hydro or “Tier 1 System,” in order to allocate costs for the next two year rate period. This

Rate Period High Water Mark process establishes the maximum amount of energy the PUD

is eligible to purchase from the BPA at cost, or the Tier 1 rate. The size of the Tier 1 System

varies due to changes in BPA’s system obligations, customer load growth, and maintenance

outages and refurbishments to the Federal hydro system. Figure 2-7 shows the actual BPA



Tier 1 System Size and Tier 1 contract allocation amount to the PUD for the 2012 through

2017 period:27

Figure 2-7

BPA Tier 1 System Size and Contract Allocation to Snohomish PUD

Fiscal Year

BPA Tier 1

System Size

(in aMW)

Maximum Tier 1

Available to PUD

Rate Period High Water Mark

(in aMW)

Actual BPA Tier 1

Contract Allocation

to Snohomish PUD

(in aMW)

2012 7181 811 785

2013 7181 811 788

2014 7240 811 753

2015 6992 811 755

2016 6983 791 759

2017 6983 791 778

Figure 2-8 shows the actual annual average megawatt hours (aMW) provided to the PUD by

BPA under the long term Block-Slice contract by fiscal year, the December on-peak aMW,

and the December Peak Week on-peak hours for 2012 through 2017:28

Figure 2-8

Snohomish PUD BPA Contract Actual Annual MWh

(Block and Slice Combined)

Fiscal

Year

Annual aMW

December

On-Peak aMW

December Peak

Week (aMW)

2012 941 1,076 1,141

2013 859 886 963

2014 859 1,016 1,047

2015 824 924 983

2016 865 1,032 1,074

2017 941 1,076 1,141

27 The BPA Slice product is allocated contractually based on the customer’s Slice percentage with monthly

output based on critical water; actual amounts will vary. 28 Peak Week is defined to be the on peak hours represented by hour ending 0700 through hour ending 2200,

Monday through Friday, for a total of 80 on peak hours for the peak week in a month.



PUD-Owned Generating Resources

Jackson Hydroelectric Project

The Jackson Hydroelectric Project (Jackson Project) is located on the Sultan River, north of

the City of Sultan, and is owned and operated by the PUD. The project has two large 47.5

MW nameplate Pelton generating units and two small 8.4 MW Francis generating units for a

total nameplate capacity of 111.8 MW. The firm energy for the project, based on the 1940-41

water year, is ~29.5 aMW. The average annual or expected output is approximately 49

aMW. Project output is delivered directly into the PUD’s electric system.

The Jackson Project is operated to produce the optimum amount of electrical energy, subject

to specified minimum releases of water into the Sultan River for maintenance of fish and the

diversion of water into the City of Everett’s water reservoir system. An agreement from

1961, with subsequent amendments, established the rights and duties of the City of Everett

and the PUD to the uses of water from the project. The City of Everett receives its water

supply from Lake Chaplain Reservoir, which the project feeds through the two 8.4 MW

Francis units. The PUD received a new 45-year project license as the sole licensee in

September 2011. The new license did not alter how the project is operated.

Historical output for the project varies with the amount and timing of rainfall that affects

stream flows that fuel the project. Power production is typically highest in the late fall

through late spring periods due to precipitation and snowmelt. The shape of the project’s

output roughly matches the PUD’s seasonal load shape. The project has some seasonal

ramping capability, depending on time of year, and also has some ability to be dispatched in

conjunction with storage in the Spada Lake Reservoir. That said, license requirements to

maintain stream flows and supply the City of Everett’s potable water supply do limit the

project’s ability to follow the PUD’s load within a day.

For the 2012 through 2016 period, the Jackson Project generated an annual average of

426,208 MWh, with a minimum of 308,865 MWh in 2012 and a maximum of 512,423 MWh



in 2014. During a regional drought in 2015, the project generated 375,100 MWh, or 43

aMW. Figure 2-9 denotes the project’s average generation by month for this period:

Figure 2-9

Jackson Hydroelectric Project Actual Production– in MWh by Month

Woods Creek Hydroelectric Project

The Woods Creek Hydroelectric Project is located in Snohomish County, north of the city of

Monroe, with a nameplate capacity of 0.65 MW. The PUD purchased the powerhouse and

adjoining acreage in February 2008. Prior to its acquisition, the PUD had been purchasing the

output from this plant. This project is adjacent to Woods Creek, a tributary of the Skykomish

River, with the powerhouse located at the base of a natural impassible barrier to anadromous

fish, and typically produces the majority of its generation during the November through April

period.

Since acquiring the project, the PUD has made numerous engineering and efficiency

improvements which has increased annual production from the historical 10 year average

production of 497 MWh to just under 1,800 MWh, depending on hydrological conditions.

These improvements to the project that result in increased hydro production from the existing

project with no additional diversion or impoundment of water are considered to be

“incremental hydro.” Incremental hydro qualifies and can be applied toward the PUD’s

0

10,000

20,000

30,000

40,000

50,000

60,000

70,000

1 2 3 4 5 6 7 8 9 10 11 12

MW

h

Month



annual renewables requirement.29 For the 2012 through 2016 period, Woods Creek has

generated an annual average of 1,774 MWh. Figure 2-10 shows the generating profile for

this resource:

Figure 2-10

Woods Creek Hydroelectric Project Actual Production– in MWh by Month

Youngs Creek Hydroelectric Project

In 2008, the PUD purchased the unconstructed Youngs Creek Hydroelectric Project located

on Youngs Creek, a tributary of Elwell Creek near Sultan in Snohomish County. The project

is situated above a natural impassable barrier to anadromous fish. Commissioning of this

new run of river resource, with single Pelton unit at 7.5 MW nameplate, occurred in

November 2011.

Youngs Creek was the first new hydroelectric resource to be constructed in the region in

more than 17 years. It is licensed through 2042. For the 2012 through 2016 period, the

project generated an annual average of 18,312 MWh, with the majority generated during the

winter and spring months (Figure 2-11).

29 Washington Administrative Code (WAC) Section 194-37-040 (13)(b) provides: “Incremental electricity

produced as a result of efficiency improvements completed after March 31, 1999, to a hydroelectric generation

project owned by one or more qualifying utilities [see definition of qualifying utility in RCW 19.285] and

located in the Pacific Northwest or to hydroelectric generation in irrigation pipes and canals located in the

Pacific Northwest, where the additional electricity generated in either case is not a result of new water

diversions or impoundments.”

0

50

100

150

200

250

300

350

1 2 3 4 5 6 7 8 9 10 11 12

MW

h

Month



Figure 2-11

Youngs Creek Hydroelectric Project Actual Production– in MWh by Month

Calligan Creek Hydroelectric Project

In 2015, the PUD received an original 40-year license for the Calligan Creek Hydroelectric

Project located on Calligan Creek, a tributary to the North Fork Snoqualmie River in King

County. The project is located above Snoqualmie Falls, a natural barrier to anadromous fish.

Construction on this 6.0 MW Pelton unit, run of river facility began in 2015 and began

commercial operation in February 2018. Based on historic hydrology records, the output

from this project is 20,700 megawatt-hours (MWh) on average per year, the majority of

which will be generated during the months of November through April.

Hancock Creek Hydroelectric Project

In 2015, the PUD received an original 40-year license for the Hancock Creek Hydroelectric

Project located on Hancock Creek, a tributary to the North Fork Snoqualmie River in King

County. The project is located above Snoqualmie Falls, a natural barrier to anadromous fish.

Construction on this 6.0 MW, run-of-the-river facility began in 2015 and began commercial

operation in February 2018. The powerhouse has a single 6 MW Pelton unit. Based on

historic hydrology records, the anticipated output from this project is an average of 22,100

megawatt-hours (MWh) annually, the majority of which will be generated during the months

of November through April.

0

500

1,000

1,500

2,000

2,500

3,000

3,500

1 2 3 4 5 6 7 8 9 10 11 12

MW

h

Month



Arlington Microgrid Solar Array

The PUD announced plans in 2017 to build the Arlington Microgrid Solar Array, as part of

its new local office complex in Arlington, Washington, located east of the Arlington

Municipal Airport. This facility will demonstrate multiple new energy technologies,

including energy storage paired with a solar array, interconnected to form a “microgrid’ – or

system that can be “islanded” and run independently from the electrical grid.

The project will be funded in part through a Clean Energy Fund II grant provided by the

Washington State Department of Commerce. The microgrid project will consist of a:

500 kW utility scale solar array;

Modular Energy Storage Architecture (MESA) compliant 500 kW/1000 kWh lithium

ion battery;

Micro-turbine for back-up generation;

Several vehicle-to-grid (V2G) charging systems; and

Clean Energy Technology Center (CETC) to provide the load and demonstration

area.

All of these components will be interconnected and controlled via a central control system.

Figure 2-12 shows an overview of the design and components that will be integrated at the

Arlington Microgrid Solar Array:



Figure 2-12

Diagram of the Arlington Microgrid Components

Key milestones for the Arlington Microgrid at the time of this writing are:

Design & Equipment Procurement - 2018

Site Preparation & Construction - 2019

Commissioning & Reports - 2020

The installed battery system at the Arlington Microgrid would be available to be called on to

provide grid support and provide ancillary services via the PUD’s Distributed Energy

Resource Optimizer (DERO). The DERO is a software system. Other components

envisioned for the Arlington Microgrid include a demonstration of how electric vehicles and

vehicle to grid (V2G) charging systems can potentially provide a benefit to the grid by

providing another source of stored energy in the event of a power outage.



At some future time, the Clean Energy Technology Center (CETC) could also serve as the

new North County local office – in the event of a large scale power outage, such as a

Cascadia Rising major earthquake event. The general concept is that in the event of a power

disruption, the microgrid would disconnect from the utility grid and act as the back up power

supply to the CETC and/or local office. While connected to the grid, the solar array is

expected to provide approximately 544 MWh per year according to NRELs PV Watts

calculator as shown in Figure 2-13.

At the time of this writing, the PUD is in the preliminary stages of planning a community

solar program associated with the 500 kV solar array at this location. This program would

continue to support the PUD’s clean renewable energy development efforts within its service

territory, and provide opportunities for PUD customers who otherwise may not be able to

participate or benefit from solar energy at this time. Typically community solar programs

have offered customers the ability to either lease or purchase “shares” of the solar project,

without having to provide their own rooftop, or fund and install their own solar panels.

Figure 2-13

Preliminary Arlington Microgrid – Estimated Monthly Output

500kW Solar Array (MWh)

0

10

20

30

40

50

60

70

80

90

MW

h



MESA Battery Storage Initiative

The PUD and several project partners have

been engaged since 2011 in the design of a new

model of battery architecture, known as the

Modular Energy Storage Architecture (MESA).

MESA is a set of nonproprietary design and

connectivity standards that provide a scalable

approach for energy storage control system

integration and optimization. The goal of the

PUD's energy storage or MESA project has

been to standardize the electric and

communications interfaces between the battery

components and the utility’s control systems, to

drive down the integration costs for this flexible

technology. Energy storage has the capability to be “dispatched” when needed, and the PUD

this technology a key component in the future integration of renewable resources and

standard resources such as run-of-river hydro, and helping to serve peak loads.

MESA 1 Battery System

The MESA 1 project was installed in 2015 and 2016 in the PUD’s service territory. It has a

nameplate of 2 MW and is comprised of two types of lithium-ion battery systems. The first

battery system is a 1 MW, .5 MWh, utilizing GS Yuasa batteries, and the second is a 1 MW,

.5 MWh system utilizing LG Chem batteries. Both systems use a power conversion system

from Parker-Hannifin. Since completion the project has undergone use case testing with

Pacific Northwest National Laboratories, been enrolled in demand response program with the

BPA, and used in a BPA Technology Innovation Fund project studying the sharing of energy

storage between transmission and distribution use cases. The battery has also been utilized

by the PUD for energy shifting and energy imbalance mitigation. The Washington State

Department of Commerce provided $2.4M to help fund this project.



MESA 2 Battery System

The MESA 2 project has a 2.2 MW nameplate and was installed in July 2017. It utilizes

vanadium flow battery technology, manufactured locally by UniEnergy Technologies, in

Mukilteo, Washington. The vanadium flow battery technology was developed at the Pacific

Northwest National Laboratory and provides twice the energy density of other flow batteries.

It is a promising new entrant in the utility-scale battery storage market. The power

conversion system is provided by the German company, AEG Power Solutions. The MESA

2 system is currently undergoing use case testing with Pacific Northwest National

Laboratories. The Washington State Department of Commerce provided $4.4M to help fund

this project.

Distributed Energy Resource Optimizer (DERO)

The DERO project was installed in 2017 and consists of controls integration to allow the

PUD’s Power Schedulers to remotely manage energy storage. DERO automatically provides

optimized schedules for review and deployment by Power Scheduling and allows for

schedules to be remotely loaded into individual energy storage systems. The software and

integration was provided by Doosan GridTech. The Washington State Department of

Commerce provided $1.8M to help fund this project.



Long-Term Power Supply Contracts

The PUD has several long-term contracts for energy, each associated with a specific

generating resource. The PUD has no ability to shape deliveries under these contracts.

Hampton Lumber Mill – Darrington Cogeneration Contract

In 2006, the PUD executed a 10-year contract with Hampton Lumber Mills-Washington,

Inc., for 100% of the electrical output from the 4.5 MW cogeneration project that utilizes

wood waste. The project is a primary employer for residents in the town of Darrington, WA.

The project began commercial operation in February 2007 and produces approximately 2

aMW. The contract was amended in December 2011 to reflect acquisition by the PUD of

both the energy and RECs from the project for the 2012 through 2016 term; a 2016

amendment extended the contract term through 2021. This project is recognized as an

eligible renewable resource under the EIA, and also qualifies for the two times distributed

generation multiplier for every MWh generated.

Packwood Lake Hydroelectric Project

This small hydroelectric project is located at Packwood Lake, 20 miles south of Mount

Rainier in Packwood, Washington, and began operating in 1964. This project is managed and

operated by Energy Northwest and has a nameplate capacity of 27.5 MW. The PUD is a

participant in this project and contracts for a 20% share, or 1.3 aMW, on a firm energy basis.

Since October 2011, the PUD has been taking delivery of its 20% contractual share, which it

plans to maintain for the foreseeable future. The PUD’s 20% share of the project’s output has

averaged just under 20,000 MWh for 2012 through 2016 period.



Contracted Wind Fleet

The PUD purchases the wind energy and environmental attributes or RECs from three wind

projects under four long-term contracts. The White Creek, Hay Canyon and Wheat Field

wind projects are situated in the Pacific Northwest and have a combined nameplate rating of

217 MW. Historical production for the contracted wind fleet is reflected in Figures 2-14 and

2-15. The aggregate historical annual capacity factor for the PUD’s contracted wind

resources is approximately 25%.

The wind contracts were modelled as a single fleet based on their aggregate historical actual

production by month, in the 2017 IRP analysis. These long term contracts expire during the

2024 through 2029 period.

Figure 2-14

Actual Average Monthly Wind Fleet Production 2010 - 2016 (in aMW)

Figure 2-15

Actual Annual Wind Fleet Production 2010 - 2016 (in aMW)

-

10.0

20.0

30.0

40.0

50.0

60.0

70.0

80.0

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

aMW

0.0

10.0

20.0

30.0

40.0

50.0

60.0

2010 2011 2012 2013 2014 2015 2016

aMW



White Creek Wind Project

In 2007, the PUD executed a 20-year power purchase contract with LL&P Wind, a wholly

owned subsidiary of Lakeview Light and Power, Tacoma, Washington, for approximately

10% of the output and RECs from the White Creek Wind Project. The project is located in

south-central Washington along the Columbia River Gorge. The PUD’s share of the White

Creek output is equivalent to 20 MW of wind capacity, with 6 aMW of wind energy

forecasted each contract year. The project is an eligible renewable resource under I-937 and

began commercial operation in November 2007; the PUD began taking output and RECs

from the project in January 2008. This contract expires in 2027.

Hay Canyon Wind Project

The PUD executed two power purchase agreements in February 2009 for 100% of the wind

energy and RECs from the Hay Canyon Wind Project. This 100.8 MW nameplate project

interconnects to BPA’s transmission system and is located in north central Oregon along the

Columbia River Gorge. It was developed by Hay Canyon Wind, LLC, a subsidiary of

Iberdrola Renewables, Inc.30 The PUD contracts for 50% of the project’s output under a 15-

year power purchase agreement, and 50% under an 18-year power purchase agreement.

These contracts expire in 2024 and 2027, respectively. The Hay Canyon Wind Project is an

eligible renewable resource under the EIA; the PUD began talking delivery of energy and

RECs from the project in March 2009.

Wheat Field Wind Project

The PUD signed a 20-year power purchase agreement for the entire output and RECs

associated with the 97 MW nameplate Wheat Field Wind Project in 2008. The project is

located in north central Oregon and interconnects to the BPA’s transmission system. The

project was developed by Wheat Field Wind Project, LLC, in conjunction with Horizon

Wind Energy, LLC, a subsidiary of Energías de Portugal.31 The Wheat Field Wind Project is

30 In December 2015, Iberdrola USA finalized acquisition of UIL Holdings to create a new company, Avangrid.

The Hay Canyon contracts are now managed by Avangrid out of its Portland, OR offices. 31 In July 2011, Horizon Wind Energy changed its name to EDP Renewables North America LLC.



an eligible renewable resource under the EIA; the PUD began taking delivery of energy and

RECs from the project in April 2009. This contract expires in 2029.

Small Renewables Program

The Small Renewables Program was adopted by the Board of Commissioners in August

2011 to encourage development of customer-owned, distributed generation inside the service

area. The program established a standard methodology for determining the price the utility

may pay for the energy and environmental attributes produced by the customer-owned

resource. The contract term ranges from one to five years. Participation in this program is

limited to renewable resource technologies between 100 kilowatts and 2 megawatts (MW)

nameplate, with a total program limit of 10 MW aggregated nameplate capacity.

Customer-owned Renewables

The PUD introduced its Solar Express program in March 2009 to incent the development of

renewable distributed generation by residential customers. This program sunset at the end of

2017 after having reached a total of 1,167 photovoltaic systems and a total of 11.3 MW of

installed rooftop solar. In aggregate, these PV systems produced 7,768 MWh in 2017.

Despite the sunset of the Solar Express program, the PUD continues to interconnect

customer-owned, generally rooftop, distributed generation systems upon request.

Short Term Wholesale Power Market Purchases and Sales

The PUD is a net seller of energy when annual snowpack and precipitation results in at or

above average water years . For the 2012 through 2016 period, the PUD purchased an

average of 340,000 MWh of energy, and sold an average of 2,190,000 MWh in the short-

term wholesale power markets. PUD staff make short-term energy purchases from the

wholesale power market during the winter months when peak demand is expected to exceed

the capabilities of the PUD’s owned and contracted resources, and as needed to balance

seasonal variations in loads and resources. Sales are made when the PUD’s contracted

resources and surpluses associated with the BPA Slice product exceed the PUD’s need. The



PUD’s short-term market purchases and sales fluctuate each year, reflecting variations in

customer demand, weather, market and hydrological conditions.

Firm Transmission Contracts

The PUD relies on long-term firm transmission capacity across the BPA transmission system

through its long-term firm point-to-point agreement with BPA. This firm transmission is

used to schedule and deliver the PUD’s power supply from the source of the generation it

purchases and contracts for, to the homes and businesses it serves in Snohomish County and

Camano Island. The PUD currently contracts for 1,969 MW of firm point-to-point capacity

with BPA. This contract includes 16 different points of receipt (where the BPA picks up

power for the PUD) and six points of delivery (where the BPA will deliver power for the

PUD). Of the total, 1,365 MW is designated for delivery directly to the PUD’s service

territory. The remaining 601 MW is used to transport power supplies that are surplus to the

PUD’s needs, primarily during the spring and summer periods, to the wholesale power

market. When the PUD needs more than 1,365 MW delivered to its service area, the Power

Scheduling staff formally request the BPA, through its Open Access Same-time Information

System (OASIS), to “redirect” its contract capacity to PUD points of interconnection with

BPA. With limited exception, the BPA has typically granted these requests.

In 2008, to ensure the ability to meet the PUD’s long-term peak demand across time, the

PUD requested, and was granted, an additional 350 MW of firm transmission capacity from

the BPA. In total, the contracts for 2,166 MW of firm transmission capacity across the BPA

transmission network.

The contract term expiries for the PUD’s firm transmission contracts with BPA range from

2026 through 2043; under BPA’s transmission business practices, said contracts are eligible

for the PUD to request renewal (rollover rights) with the first right of refusal.



Existing Demand Side Resources

Conservation

The PUD has actively engaged in conservation and demand-side management for over 35

years. Since 1980, conservation and energy efficiency programs have resulted in the

cumulative acquisition of more than 180 aMW of conservation resources, or enough to power

more than 125,000 homes. Figure 2-16 shows the gross annual and cumulative savings

accomplishments for the PUD through 2016:32

Figure 2-16

PUD Gross Annual and Cumulative Conservation Savings in aMW

32 As illustrated here, the cumulative savings calculation does not include degradation of savings as energy

efficiency measures reach the end of their useful life.



The acquisition of new conservation through energy efficiency programs encourages

customers to use energy more efficiently, which can defer the acquisition of new supply side

resources; defer the need for new transmission and distribution system upgrades; create value

for customers; increase affordability for households; and reduce operating costs for

businesses. Conservation is a low cost resource with minimal environmental impacts.

The PUD offers financial incentives, technical assistance and educational services for all

customer classes. For residential customers, the PUD provides a comprehensive set of

energy efficiency programs targeting single and multi-family residences, new construction

and low-income households. Financial incentives are offered for efficiency products

including new heating systems, window and insulation upgrades, LED lighting, and home

appliances. For commercial and industrial customers, the PUD offers financial incentives

and technical assistance to help reduce energy use and annual operating costs. Efficiency

products include HVAC, high-efficiency lighting, insulation, process load efficiencies,

motors, and equipment controls. Figure 2-17 highlights key programs and the sector served:

Figure 2-17

PUD Energy Efficiency Programs by Target Sector

Program Description Target Sector

Residential Residential Multi-Family Commercial Industrial

Single Family Weatherization

Multi-Family Weatherization

High Efficiency Lighting

New Home Construction

Matchmaker

X

X

X

X

X

X

X

Commercial & Industrial

Custom Projects

Energy Smart Industrial

Commercial Kitchen Equipment

Lighting Rebates

Strategic Energy Management

New Building Construction

Pay-For-Performance

X

X

X

X

X

X

X

X

X

X

X

X



Program Innovation

In addition to the PUD’s traditional conservation programs, the PUD actively seeks out new

approaches to markets and emerging technologies. Examples include:

Pay-for-Performance pilot that allows commercial building owners to acquire energy

efficiency as a service. The program is designed to pay for capital improvements through

energy savings over a period of ten years.

In partnership with Snohomish County, the PUD secured state matching funds to help

improve efficiency in low income housing.

With market transformation in the area of efficient lighting, the PUD was able to revise

its incentives to focus on how best to increase other efficiency opportunities for its

commercial and industrial customers. Savings from these other areas can reduce peak

demand periods and aid in reducing the PUD’s winter capacity needs.

Bundled lighting and weatherization in multi-family dwellings significantly increased the

comfort and efficiency of 29 multifamily complexes, while reducing overall PUD

program implementation costs.

The PUD recently developed its Smart Rewards online platform. Smart Rewards provides

customers a resource to research the purchase of energy efficient home appliances and

products. The site serves as an aggregate location that allows customers to compare the

energy efficiency, price, customer reviews, operating cost and utility incentives for

models. The platform delivers an Amazon-style experience and it has helped many of our

customers improve the efficiency of their homes.

The PUD recently added numerous new technologies to its program offerings. Emerging

products such as direct outside air systems for HVAC, heat pump water heaters, high

efficiency control systems, and advanced lighting controls provide exciting new

opportunities for energy savings and often provide important secondary benefits to

customers.



Community Programs

The PUD places high value on offering programs and measures to serve all customers in our

community. Recently, staff worked with the Northwest Power and Conservation Council

(NWPCC or Council) to study whether the PUD’s programs were reaching all customers and

markets. Specific attention was given to the more hard to reach populations (low income

customers, moderate income customers, multifamily tenants, manufactured home dwellers,

small business owners, commercial tenants, and industrial customers). In general, the study

showed that most of the hard-to-reach markets were well served by the PUD’s energy

efficiency programs. Low and moderate income residential customers participated at rates

roughly equal to their distribution in the customer population. Manufactured home dwellers

and rural residential customers had proportionally high participation rates. As a group, small

business owners, commercial tenants, and industrial customers, participated proportionally

throughout PUD’s service territory.

Regional and National Efforts

The PUD remains actively engaged in regional and national conservation activities to

identify new technologies, develop new delivery strategies and affect policy related to energy

efficiency and conservation.

The PUD actively participates and provides financial support for market transformation

efforts through the Northwest Energy Efficiency Alliance, Consortium for Energy

Efficiency and the Electric Power Research Institute.

The PUD is a member of the Regional Technical Forum and the Snohomish County

Sustainable Development Task Force and supports the Pacific Northwest Integrated

Lighting Design Labs.

The PUD actively participated in the development and review of the conservation supply

curves developed by the Council for its Seventh Power Plan adopted in 2016. The PUD

supports establishing achievable energy efficiency targets and recognizes the need to

conduct research, development and demonstration activities to ensure a sustainable

pipeline of future energy efficiency resources.



Demand Response Program and Strategy

Demand response involves the development of programs, pricing structures and technologies

to influence when and how customers use electricity. By shifting electricity demands from

periods when loads and power prices are high to periods of lower loads and prices, the PUD

can reduce its costs and maintain or increase reliability, all of which can reduce customers’

power bills.

Demand response programs take multiple forms: dispatchable load controls, scheduled load

controls, voluntary calls to action, and price incentives. Dispatchable load control programs

give utilities the ability to call on resources without any action by the customer. Dispatchable

resources are often available within 10 or 15 minutes after being requested or “dispatched”

by a utility. Scheduled load control programs require customers to temporarily change

business processes and typically require advance notice by the utility ahead of a request for

load reduction.

The PUD’s adopted 2013 IRP included an action item for staff to conduct a situational scan

of demand response technologies and applications. Staff completed this work in 2014 and

found that the Northwest’s lack of a well-established capacity market to help determine the

value of demand response and that demand response technologies in general are still

evolving, to be limiting. The majority of demand response efforts in the Northwest were

driven primarily by the need to: 1) demonstrate technology; 2) test customer acceptance;

and/or 3) explore demand response costs and potential. National programs – largely from

summer peaking utilities – were found to be more mature yet still considered ‘developing,’

and not fully mature.

An action plan item in the PUD’s adopted 2015 IRP Update to the 2013 IRP directed that

demand response be pursued and a work plan established. A demand response potential

assessment was conducted by CADMUS in 2014, limited to the products and technologies

identified in the situation scan, and to determine the value demand response provides by



developing a methodology to identify peak hour needs and a valuation of energy savings in

those hours.

In 2016, the PUD supported customer participation in a demand response pilot offered by

BPA and implemented by EnerNOC. The PUD’s objectives in participating in the BPA pilot

included: general assessment of customer interest in demand response; discover and examine

customer opportunities and issues associated with implementing demand response; consider

potential program modifications that could improve program performance and participation;

and assess the performance and reliability of such a program in delivering peak savings. The

BPA pilot allowed all involved parties to test the physical and economic dispatch of demand

response technologies and evaluate the impacts upon facilities and operations, and explore

pathways to promising demand response program design elements for future programs.

At the present time, the PUD is developing its Demand Response Strategy that will serve as a

roadmap for the PUD’s demand response efforts with its customers, going forward. This

initiative is expected to develop a comprehensive approach that will include clear objectives,

reviewing current programs, products and work to date, exploration of program options,

establishing near-term pilots and program offerings to deliver reliable, measureable and cost-

effective capacity savings, and reviewing peer utility experiences with demand response

programs and offerings across the country.

Demand response is viewed as having the potential to serve as a reliable resource alternative

to capacity made available from supply-side resources. Demand response may also impact

and potentially defer transmission and distribution investment needs over time, as well as

serve as a customer engagement offering. A comprehensive strategy will incorporate the

benefits and assess the value that demand response products and programs can bring to the

PUD and power supply portfolio. This effort is expected to develop specific demand

response options - with quantified cost and performance attributes – that can be incorporated

into the list of available demand side resource options for future IRP processes.



Demand Response in the Council’s Seventh Power Plan

In the Council’s Seventh Power Plan (Plan), demand response was identified as the region’s

least-cost solution for providing new peaking capacity, particularly when hydro conditions

are low. The Council’s action plan recommended that “the annual regional resource

adequacy assessment compare the cost and economic risk of increased reliance on external

market purchases to developing demand response resources to meet capacity.” As part of the

Plan’s Mid-Term Assessment in 2018, the Council will determine if the region has made

sufficient progress toward acquiring cost-effective demand response or confirm the ability to

import a minimum of at least 600 megawatts of additional peaking capacity.33

The Council reviewed 11 different demand response program options, largely focusing on

space and water heating technologies. Commercial sector programs included lighting

controls, and industrial programs encompassed agricultural technologies such as irrigation

pumping and refrigerated warehouses.

Demand response was incorporated into the Council’s Regional Portfolio Model and

evaluated for its peak load reduction. The results were that some non-modeled uses of

demand response may prove useful in the future: non-firm demand response (new pricing

structures), dispatchable standby generation, and providing ancillary services to the grid

(including contingency reserves, operating reserves, frequency regulation, and locational

congestion relief). The Plan notes that these programs, while omitted from the Regional

Portfolio Model, may still provide cost-effective services depending on other options in

meeting those needs. This work is expected to be part of the Council’s analysis included in

the 2018 Mid-Term Assessment.

33 Northwest Power and Conservation Council, Seventh Power Plan, Executive Summary, page 1-6, found at

https://www.nwcouncil.org/media/7149937/7thplanfinal_chap01_execsummary.pdf


Section 3: Planning Environment


3 PLANNING ENVIRONMENT

Part of the process for determining the best way to meet future customer needs and demands

involves establishing an environment in which the PUD sees itself operating. This

environment must consider both the current landscape of policy and trends, and how they

evolve over time. To evaluate these trends, the more significant factors have been

categorized by their sphere of influence on the PUD:

The PUD’s Strategic Priorities

The Puget Sound Economy

Electric Industry Initiatives and Efforts

Energy Policy and Regulatory Requirements

Climate Change

PUD’s Strategic Priorities

The Board of Commissioners expects the PUD to deliver power and water to its customers in

a safe, sustainable and reliable manner while successfully navigating complex change in our

industry. The PUD accomplishes this by empowering its teams to provide quality service to

its community and prudently managing costs while investing for the future. The Strategic

Priorities, developed in 2016-17 and updated annually, are designed to support the PUD’s

missions of providing quality water and electric energy products and services and include a

distinct focus on 5 key areas: Team PUD, Customer Experience, Delivering Now & For the

Future, Responsible Cost & Fiscal Management, and Continual Improvement.



The Economy – Puget Sound and Beyond

Snohomish County, one of four central Puget Sound counties – King, Pierce, Snohomish and

Kitsap – gained few new jobs in 2017, with an annual growth rate of 0.2%. This marks the

seventh consecutive year of job growth in the region accompanied by the lowest

unemployment rate in the last decade at 3.9%. Manufacturing saw a loss in jobs with no gain

in 2017 mostly due to the Asian economic slowdown and political uncertainty overall.

Snohomish County’s neighbor to the south, King County, has been experiencing record

growth in the technology sector. Companies such as Amazon, Google, and Expedia are

settling in the South Lake Union neighborhood, with no signs of slowing. As the cost of

living in Seattle continues to increase, and housing inventory cannot meet demand,

employees of these tech firms continue to search for affordable housing, and parent

companies search to reduce overhead costs.

Nationally, the unemployment rate fell 0.5% in 2017, from 4.9% to 4.4%. Employment grew

at 1.5%, which was slightly less than the 2016 growth rate of 1.8%. GDP growth however

grew at 2.2%, a higher rate than the 2016 rate of 1.9%. The electric industry was generally

affected by an abundant supply of natural gas helping to keep power prices low in many parts

of the country.

Electric Industry – Initiatives and Efforts

The electric industry in the Pacific Northwest is facing dynamic changes. As the PUD plans

for the future and assesses the state of the industry, the following regional policies and

guidelines relevant to utility resource planning come to the forefront. These include the

Bonneville Power Administration, the Northwest Power and Conservation Council, and the

potential for newly forming markets.

The Bonneville Power Administration

The Bonneville Power Administration (BPA) is a significant supplier of power to the region;

as such its success and long term viability is of great importance to public utilities like the

PUD and its customers. In 2015, BPA launched Focus 2028, to establish a common

understanding with regional leaders as to the types of industry changes, challenges and



strategic choices BPA may face as it tries to maintain its financial strength and cost

competitiveness in the post 2028 period. September 2028 is when the current long-term

agreements with customers like the PUD are set to expire.

Focus 2028 also looked at how the agency could better manage its costs to maintain its

overall financial health. BPA explained that in order to stay competitive, it must keep costs

low and effectively manage its capital and program expenditures. As a follow on in 2016,

BPA developed a set of key strategic initiatives (KSIs) intended to revamp legacy processes,

systems and capabilities integral to BPA’s corporate and commercial functions. The four

KSIs with funding as part of the 2016 Integrated Program Review included: Asset

Management, Commercial Operations, Long-Term Financial Health and Rates, and Business

Information Systems. The PUD expects BPA will discuss funding levels to support the KSIs

in the 2018 Integrated Program Review as part of its budget development process held in

Summer 2018; this process review will identify the programmatic costs and inform revenue

requirements ahead of the fiscal year 2020-2021 rate case.

In January 2018, BPA released its Strategic Plan for the 2018 through 2023 period. This

Strategic Plan focuses on how to strengthen the agency’s financial health, modernize the

grid, and provide competitive power and transmission products to “deliver on (BPA’s) public

responsibilities through a commercially successful business.”

Federal Interactions with BPA

In the 2018 Presidential budget, President Trump included provisions to sell BPA’s

transmission assets, resulting in a one-time influx of cash to the federal government. This

highlights the current administration’s focus on privatization of government assets. Because

the Transmission system is a highly valuable asset to Northwest customers, the impacts of a

potential sale, and the uncertainty around the scope of these offerings must be considered

while evaluating future resource choices. These budget provisions met fierce resistance from

the Northwest Delegation, arguing that the public value of this infrastructure far outweighed

any potential short-term gain from privatization. The proposal currently is not expected to

move forward.



Northwest Power and Conservation Council

The Northwest Power and Conservation Council (NWPCC or Council) is a public agency

created by the Pacific Northwest Electric Power Planning and Conservation Act of 1980. The

agency’s three primary functions include:

1. Develop 20-year electric power plans for the Northwest that guarantees adequate and

reliable energy at the lowest economic and environmental cost;

2. Develop programming to protect and rebuild fish and wildlife populations affected by

hydropower development in the Columbia River Basin; and

3. Educate and involve the public in the Council’s decision-making processes.

Due to the nature of the Council’s work and its structure within the Northwest Power Act, its

five year power plan serves as a guidebook to resource planning in the region. Many utilities,

as well as BPA, look to the Council’s Power Plans as a key source of information for their

own planning needs.

The Council’s Seventh Power Plan covered 2015 through 2035, and was adopted in February

2016. The plan was developed at a time when the Northwest power system was facing

uncertainties such as how federal carbon dioxide emissions regulations might be

implemented, changes in future fuel costs, baseload resource retirements, salmon recovery

actions, the pace of economic growth, and the cost and capacity required to reliably integrate

increasing amounts of renewable resources.

Key findings for the Seventh Power Plan analysis include:

1. Energy efficiency was the least expensive resource available to the region;

2. Developing demand response capabilities or rely on increased market imports to meet

system capacity needs when adverse or low water and extreme weather conditions occur.

3. New natural gas-fired generation is the most cost-effective resource option for the region

in the near-term.



4. After energy efficiency, the increased use of existing natural gas generation offers the

lowest cost option for reducing regional carbon emissions.34

5. New methods to store electric power, such as pumped storage or advanced battery

technologies may enhance the value of existing variable generation, like wind.

6. Modest development of renewable generation is expected to satisfy the various renewable

portfolio standards;35

The Plan also called for the development of 600 MW of demand response to help meet peak

demand needs; noted that with respect to carbon reduction, increasing regional renewable

portfolio standards was not an effective option for reducing emissions; and encouraged

investment in research, development, and demonstration projects for emerging technologies

to help reach their full potential.

Council’s Regional Resource Adequacy Standard

In 2013, the Council created the Resource Adequacy Advisory Committee (RAAC) to aid in

the assessment of regional power supply adequacy. The Council, with the aid of the RAAC,

annually assesses the adequacy of the regional supply five years out. The adequacy standard

limits the likelihood of a supply shortage or “loss of load probability” (LOLP), to a

maximum of 5 percent. The state of the system is determined using only existing resources,

planned resources that are sited and licensed, and the energy efficiency savings targeted in

the Council’s power plan.

The Council’s annual assessment serves as an early warning to the region, in the event

energy efficiency and resource acquisitions don’t keep pace with load growth. This provides

sufficient time to perform any actions that might be needed. Other adequacy metrics (size of

potential shortages, how often forecast to occur, how long event could last) are also part of

the annual report.

34 Northwest Power & Conservation Council’s Seventh Power Plan, Executive Summary, pages 1-2 through 1-4,

located at https://www.nwcouncil.org/media/7149937/7thplanfinal_chap01_execsummary.pdf 35 Ibid., pages 1-5 through 1-6.




In 2016, the 2021 resource adequacy assessment was determined to have a LOLP of 10%,

exceeding the Council’s 5% LOLP standard, primarily due to the scheduled retirements of

the Centralia 1 and Boardman coal plants (1,330 megawatts combined).36

The Council updated its assessment in July 2017, determining that the 2021 LOLP was just

under 7%, while for 2022, the LOLP was slightly higher at just over 7%. This update re-

evaluated regional load forecasts, hydro operations for the U.S. and Canada, wholesale

energy market supplies available from the Desert Southwest, and an assumption that the

Council’s energy efficiency targets are achieved through 2022. To comply with the

Council’s 5% LOLP adequacy standard, the region will need to add an estimated 400

megawatts of new effective capacity by 2021.37

Energy Markets

The Energy Imbalance Market (EIM), is operated by the California Independent System

Operator (CAISO). Since 2015, several Northwest utilities either joined or signaled their

intention to join. While the Northwest energy market has traditionally traded on an hourly

basis, the EIM is designed to balance energy and capacity needs on a sub-hourly basis. The

region is monitoring the results and cost/risk tradeoffs associated with joining an EIM,

particularly as to how it can help contribute flexibility and value to the region. The region is

also currently discussing expanding the EIM to a day-ahead market.

Cyber Security and the Grid

The safety and economic security of the nation depends on the reliable functioning of critical

infrastructure such as energy delivery systems. Cybersecurity threat actors may exploit the

increased complexity and connectivity of these systems, placing the nation’s security,

economy, public safety and health at risk. Similar to financial and reputational risk,

cybersecurity risk affects a company or government’s financial health. It can drive up costs,

impact revenue, and harm an organization’s ability to innovate and to gain and maintain

customers. Over the past several years, the PUD has evolved into a local, regional and

36The Council’s 2016 Adequacy Assessment Report can be found at the following link:

https://www.nwcouncil.org/media/7150591/2016-10.pdf 37The Council’s Pacific Northwest Power Supply Adequacy Assessment for 2022 can be found at the following

link: https://www.nwcouncil.org/media/7491213/2017-5.pdf

https://www.nwcouncil.org/media/7150591/2016-10.pdf

https://www.nwcouncil.org/media/7491213/2017-5.pdf



national leader planning against cyber threats in partnership with both the public and private

sectors; including the military. Adhering to the National Institute of Standards and

Technology (NIST) framework, the PUD and the State of Washington are well positioned to

plan and manage cyber activities and the risk environment.

In May 2017, President Trump signed an executive order on cybersecurity designed to

protect federal government networks and critical infrastructure, including the nation’s power

grid. The executive order, similar to one signed by President Obama in 2013, directs the

Secretary of Energy and Secretary of Homeland Security, in consultation with national

intelligence and other governments, to jointly assess the potential scope and duration of

prolonged power outage associated with a significant cyber incident. The executive order

largely supports NIST standards of which the PUD has been following for several years.

Energy Policy & Regulatory Requirements

Future legislative policy and regulatory requirements can have a profound effect on the

PUD’s existing power supply and any future resources it may consider, acquire or operate.

While no formal carbon policy has been enacted within the State of Washington at the time

of this analysis, the 2017 IRP considers that some form of emissions reduction or regulatory

policy likely will be implemented in the future. In addition, there are several ongoing

regulatory processes that may have a significant impact on the PUD’s existing resources,

such as the litigation surrounding the Federal Columbia River Power System, and the

discussions over modernizing or terminating the Columbia River Treaty.

Below are some of the regulatory initiatives in the planning environment that were

considered to have potential impacts on the PUD’s existing and future resource portfolio.

Greenhouse Gas Emissions and Climate Change

One of the largest uncertainties utilities face today is the impact of and response related to

climate change. Whether examining climate change at either a national, state or local level, it

is an important component to consider when analyzing future resource performance and/or

new resource acquisitions.



Local Responses to Climate Change

In the Northwest, our local community is very environmentally minded and conscious of the

ongoing effects of climate change, and includes robust academic community. The University

of Washington Climate Impacts group is an internationally recognized organization

researching the causes, effects, and mitigation tools for climate change. The PUD recognizes

our local community’s commitment to environmental stewardship, which the PUD has

adopted as one of its strategic priorities.

In response to growing public concerns over climate change, a few Puget Sound area

municipalities, along with many corporate entities, are exploring options to reduce their

carbon footprints through programs and partnerships with their local utilities. The PUD

anticipates this trend will continue as residents, businesses, and local governments have

interest in moving toward green energy options.

Executive Orders and Emissions Performance Standards

In 2007, Governor Christine Gregoire issued an executive order challenging the state to

reduce Greenhouse Gas emissions (GHG) on a timetable leading through 2050. The

milestones laid out were:

By 2020, reduce overall emissions of GHG in the state to 1990 levels

By 2035, reduce to 25% below 1990 levels.

By 2050, reduce to 50% below 1990 levels.

The passage of ESSB 6001 through the legislature in 2007 codified these standards into law,

along with performance standards for existing and new natural gas generating plants. ESSB

6001 established an emission performance standard of 1,100 lbs of GHG per MWh that all

electric baseload generators must meet in order to be permissible for long term financial

commitments. This emission level was, at the time, thought to be an approximate average

natural gas emission rate. In 2013, the Department of Commerce reduced this standard to 970

lbs per MWh. Exemptions were also built in for unexpected reliability needs on a case-by-

case basis.

Governor Jay Inslee issued another Executive Order in 2014, outlining a series of steps to

reduce carbon pollution in Washington, including:



Establishing a Carbon Emissions Reduction Taskforce;

Committing to reduce imports of coal-fired electricity from other states; and

Working to reduce emissions in the transportation sector, which is recognized as the

greatest percentage of carbon emissions in the state.

While this executive order was not incorporated into statute, it provides clear direction that

the state takes carbon reduction seriously and has long term plans to help curb emissions.

Governor Inslee was one of three flagship members of the U.S. Climate Alliance – a group of

states committed to upholding the Paris Climate Agreement emission standards. Since its

creation on June 1, 2017, thirteen other states have joined the Alliance. The states maintain

that they “…are committed to achieving the U.S. goal of reducing emissions 26–28 percent

from 2005 levels and meeting or exceeding the targets of the federal Clean Power Plan."

Clean Air Rule

In 2016, the Washington State Department of Ecology (DOE) enacted the Clean Air Rule

(CAR), effective beginning in 2017. This rule established a cap and a reduction schedule on

GHG emissions from in-state sources (natural gas plants), petroleum product producers and

importers, and natural gas distributors. However, new emission sources, like new baseload

gas plants, were permitted.

The CAR was challenged by a number of groups, and on December 15, 2017, a Thurston

County Superior Court Judge invalidated large portions of the CAR, finding that the

Department of Ecology lacked authority to impose the CAR without legislative approval.

Pending an appeal, the Department of Ecology has suspended the rule's compliance

requirements. Facilities covered by the rule are still required to report their emissions for the

Greenhouse Gas Reporting program.

https://ecology.wa.gov/Air-Climate/Climate-change/Reducing-carbon-pollution/Facility-greenhouse-gas-reports



E3 Carbon Study

In 2017, a group of utilities, including the PUD, commissioned a study through Energy +

Environmental Economics Inc. (E3), to analyze the most cost effective method to achieve

meaningful carbon reduction in the electric power sector.38 The study itself did not begin

with any presuppositions about what that method would be – the goal was to have an

unbiased analysis to determine the least cost way to reduce emissions for the electric sector,

given a set of policy goals, from present through the year 2050.

Using E3’s regional model to simulate energy markets, generation dispatch, and electric

loads, and layering policy decisions such as renewable portfolio standards, carbon taxes, or

resource restrictions, two outputs were determined for each set of policies: the amount of

GHG reduced, and the cost associated with that reduction.

After evaluating numerous policy scenarios, the E3 study concluded that the most cost

effective method of reducing emissions is to implement a price on carbon. A carbon price

creates the most reduction of GHG emissions, while maintaining a low societal cost.

Conversely, policies that call for a higher renewable portfolio standard (RPS) or that restrict

the construction of new natural gas plants, result in a significantly smaller reduction in GHG

emissions, and proves to be incrementally more expensive for electric consumers. Figure 3-1

shows the scenarios analyzed in the study and the relationship between cost and emissions

reductions.

38 The full report, “Pacific Northwest Low Carbon Scenario Analysis” published by E3, can be found at

http://www.publicgeneratingpool.com/wp-content/uploads/2017/12/E3_PGP_GHGReductionStudy_2017-12-

15_FINAL.pdf

http://www.publicgeneratingpool.com/wp-content/uploads/2017/12/E3_PGP_GHGReductionStudy_2017-12-15_FINAL.pdf




Figure 3-1

E3 Study - Annual Cost vs Reduction in GHG Emissions

Comparison of Regional RPS or Carbon Policy

Washington State’s Energy Independence Act (EIA) - RCW Chapter 19.285

In 2006, the voters of Washington State approved the Energy Independence Act (the EIA)

through the state’s initiative process (Initiative 937). This Act requires electric utilities with

25,000 or more customers to pursue all cost-effective energy conservation measures, and to

acquire and include in their portfolios a mandated amount of eligible renewable resources,

renewable energy credits, or combination of the two. The amount of eligible renewable

resources required scales to the utility’s retail load.

Utilities have three methods for complying with the renewables portion of the EIA: meeting

the load-based goals with resources or RECs, demonstrating investment of 1% of its retail

revenue requirement in eligible renewable resources or RECs without load growth, or

demonstrating investment in excess of 4% of the utility’s annual retail revenue requirement

(commonly referred to as the “cost cap” method) in eligible renewable resources or RECs.



The post-2020 landscape for the EIA has been discussed by the Washington State Legislature

for several sessions now, with a number of bills aiming to redefine and expand the scope of

the EIA. Two main approaches have been discussed: update or change the EIA to account for

the post-2020 timeframe, or sunset the existing statute and pursue alternative efforts related

to carbon reduction.

During the 2018 legislative regular session, the Governor advance a carbon tax proposal.

This proposal was estimated to generate up to $2 billion in revenue for the state. The initial

proposal advocated for a $20 per metric ton of carbon dioxide tax, with revenues directed

toward carbon reduction projects, water quality projects, and transition aid for those hit

hardest by the tax. Lawmakers are aware of potential initiative efforts in the state as well. In

the fall of 2016, a proposed carbon tax initiative (Initiative 732) did not prove successful.

Future legislation to address carbon is expected.

The Trump Administration

President Trump’s budget proposal for 2018 envisions deep across-the-board agency cuts;

including many of the Department of Energy’s current programs (research labs, renewables,

energy efficiency, electrification, carbon capture) along with proposals to sell off energy

resources and infrastructure such as BPA transmission. The budget proposal also includes

provisions to restart the investigation of Yucca Mountain as a repository for nuclear waste

created across the country. This development could have an impact on the Columbia

Generating Station and the rates and fees associated with the generation of nuclear energy.

Energy Secretary Perry suggests the budget reflects a focus on nuclear capabilities and early

stage energy research and development. The Environmental Protection Agency (EPA) has

the largest percentage funding decline of any agency with a focus on infrastructure and

air/water quality.

President Donald Trump has made it clear in his administration that climate change is not a

high priority. Since his election, President Trump has withdrawn from the Paris Agreement, a

global accord addressing climate change due to concern for the levels of United States

financial contribution. In March 2017, President Trump released an Executive Order



signaling a pullback from the Clean Power Plan (CPP) put into place by the Obama

administration. The CPP would have mandated that the States achieve specified reductions in

greenhouse gas emissions through a variety of mechanisms, including heat rate

improvements at coal and gas plants, replacement of coal with gas, energy efficiency

measures, and greater reliance on renewable resources. The EPA has announced its intend to

pull back the Clean Power Plan, and to replace it with a much narrower regulation that is

aimed at GHG emissions from power plants. Court challenges are likely to continue, though

the Administration’s prioritization of the financial strength of the coal industry above other

drivers will likely result in continued Federal support for undoing recent power industry

developments.

The Trump Administration also recently announced a 30% tariff on foreign solar cell

imports. Solar installers have stated that they believe that this will greatly harm their

business. By making the least expensive option for installing solar more expensive, demand

for solar cells could decrease and harm installers. The PUD monitors shifts in renewable

energy markets and this policy could have an effect on future resource decisions as its

impacts on the solar market become better known.

FERC’s 2016 Notice of Proposed Rulemaking on Electric Storage

In a November 2016, the Federal Energy Regulatory Commission (FERC) issued a Notice of

Proposed Rulemaking (NOPR), stating that market rules can create barriers to entry for

emerging technologies like energy storage resources. The proposed rule would require

regional transmission operators and independent system operators to propose tariff revisions

to recognize the physical and operational characteristics of electric storage resources, and to

allow those resources to participate in organized markets. The proposed tariff revisions are

based on a “participation model,” to ensure that a resource using the model: (1) is eligible to

provide all capacity, energy, and ancillary services that it is technically capable of providing;

(2) can be dispatched and is a price maker in the wholesale market as both a seller and buyer,

consistent with existing market rules; (3) accounts for the physical and operational

characteristics of electric storage resources through bidding parameters or other means; and,

(4) establishes a minimum size requirement. In addition, the sale of electric energy from the



wholesale electricity market to an electric storage resource, that the resource then resells back

to those markets, must be at the wholesale locational marginal price.

During drafting of the PUD’s 2017 IRP document, FERC adopted Final Rule Order No. 841

in February 2018. The Order does not force grid operators to change technical requirements

or compensation mechanisms for existing products, introduce new products, or exempt

energy storage resources from performance requirements. According to a Brattle Group

report, “…Order 841 does not address state or retail level challenges or reduce barriers

that would allow for energy storage to capture distribution level or customer benefits.

That underscores the important role state participation will play in the development of

energy storage. Order 841 will have a significant effect, but the greater and wider effect

will come from state policies.”39

Six states in the U.S. have taken steps to incorporate energy storage. A California mandate

calls for utilities to install 1,325 MW of storage by 2020; Massachusetts has a 200 MWh by

2020 goal; Oregon has a 5 MW by 2020 goal per utility; New York is working on an energy

storage target with a governor proposed 1,500 MW target by 2030; and legislation has passed

in Nevada and Arizona that asks regulators to investigate energy storage targets. The states

of Colorado, Illinois, Indiana, Minnesota, Missouri, New Mexico, Ohio and Vermont also

have active proceedings involving energy storage policies.

39 “The flip side of FERC's landmark storage order: A call for states to take action,” Utility Dive, March 6,

2018, URL at:

https://www.utilitydive.com/news/the-flip-side-of-fercs-landmark-storage-order-a-call-for-states-to-take-

a/518497/

https://www.ferc.gov/market-oversight/guide/glossary.asp#L

https://www.utilitydive.com/news/the-flip-side-of-fercs-landmark-storage-order-a-call-for-states-to-take-a/518497/

https://www.utilitydive.com/news/the-flip-side-of-fercs-landmark-storage-order-a-call-for-states-to-take-a/518497/



Federal Columbia River Power System

Endangered Species Act and NEPA Litigation

Litigation over the operation of the Federal Columbia River Power System and associated

Biological Opinions (BiOp) has been ongoing for the past 40 years. In 2014 parties

challenged the sufficiency of the 2014 Supplemental BiOp, alleging that the BiOp violated

the Endangered Species Act (ESA) and that adoption of the BiOp by the action agencies (the

Bonneville Power Administration (BPA), Army Corps of Engineers (COR), and Bureau of

Reclamation (BOR) violated the National Environmental Policy Act (NEPA).

Judge Simon in the District Court of Oregon concluded that the National Oceanic and

Atmospheric Administration (NOAA) Fisheries violated the ESA by adopting the 2014

Supplemental BiOp. The Court determined that the mitigation in the BiOp was insufficient

to avoid jeopardy of the listed species, particularly for salmon and steelhead in the Columbia

and Snake Rivers. The Court left the 2014 Supplemental BiOp in place while NOAA

Fisheries prepares a new BiOp to be released in late 2018. The Court also ordered

compliance with NEPA, launching a public process by the action agencies (BPA, COR,

BOR) with a final Environmental Impact Statement due by 2021.

Some parties, led by the State of Oregon and the National Wildlife Federation, subsequently

sought an injunction requiring maximum spill at eight Federal hydroelectric projects during

spring months. The injunction was granted, but has been appealed by the action agencies. If

the appeal is unsuccessful and the increased spill goes into effect during the spring of 2018, it

could require notable hydro operational changes that would affect the amount and timing of

electric power generated by the Federal System.

The PUD has no capability to predict what the outcome of this litigation will be, nor how it

will affect development of the future BiOp and Environmental Impact Statement. The 2017

IRP analysis used existing and known Federal hydro system operating assumptions, based on



the 2014 Supplemental BiOp, to model the PUD’s offtake under the Slice product portion of

its long term BPA power supply contract.40

Columbia River Treaty

The Columbia River Treaty is a 1964 treaty agreement between Canada and the United States

addressing the flood control and power benefits derived from the development and operation

of dams in the upper Columbia River basin. Either nation can terminate certain provisions of

the Treaty at any time, with a ten year notice, either on or after September 16, 2024.

In 2013, the U.S. Entity developed a Regional Recommendation in collaboration and

consultation with states, tribes, and stakeholders within the Northwest Region. The Regional

Recommendation concluded that a modernized Treaty framework - that includes ecosystem

considerations – is necessary to reflect the actual value of coordinated power and flood

control operations with Canada.

The Canadian and United States governments will soon begin a formal review of the Treaty

and began negotiating changes for the joint operation of the system. These negotiations

could result in modifications to the flood control and power obligations for each nation,

resulting in potential impacts to the hydroelectric power produced by the Federal System,

that BPA markets. At this time, PUD staff cannot predict with any certainty the outcome of

these negotiations, or the impact or impacts to the Federal hydro system or the PUD’s long

term BPA power supply contract and power costs.

40 The 2017 IRP analysis performed a sensitivity analysis to the BAU scenario labelled “No Snake River Dams”

to better understand the potential costs and ramifications to the PUD’s existing portfolio if the four Lower

Snake River dams were removed. The results of this analysis can be found in Appendix C.

Section 4: Scenario & Planning Assumptions


BAU w/No Carbon

BAU w/Calif Carbon 2022

Climate Change

Low Growth High Growth

4 SCENARIOS AND PLANNING ASSUMPTIONS

Scenarios

Scenarios help explain how changes in economic, social, technical and environmental trends

could affect the PUD’s future load growth, and the cost and risk of various resource plans

developed in response. Scenarios also provide useful insights into potential uncertainties and

broad sets of risks the PUD could face under each of these futures. The 2017 IRP evaluated

five scenarios that considered the range of futures the PUD could face for the 2018 through

2037 study period. Staff also conducted two sensitivities to the Business as Usual scenario.41

Figure 4-1

2017 IRP Scenarios & Sensitivities

41 Results of the two sensitivities can be found in Appendix C.

Sensitivities - No Snake River Dams - Renewables Only



Low Growth Scenario

The Low Growth scenario reflects a future where Executive Branch decisions affect the

country’s position in the global marketplace.

This is marked by the eventual withdrawal of

the U.S. from Trans-Pacific Trade pact,

including repeals of other national

environmental policies and regulations. The

effect on Washington state is, according to a

statement by the Washington Council on

International Trade, that 2 in 5 jobs is linked

to global trade. The result is the loss of

living wage jobs, and slower than historic growth rates

for Washington state and the Snohomish County

economy, population and employment. With other

changes to federal environmental policies, horizontal

shale extraction expands and the natural gas price

forecast is the lowest of all the scenarios.

Despite these national policy changes, Washington state retains its leadership role in state

environmental policy, and a carbon tax is instituted beginning in 2018.42 The state’s annual

renewables requirement remains at 9% of total retail load for utilities with more than 25,000

customers through 2019, and increases to 15% in 2020, remaining at this level through 2037.

The elements in this scenario inform the PUD’s future resource need in light of increasing

California and Oregon renewable portfolio requirements and the PUD remaining in a net no

load growth status, after new conservation has been acquired. Other factors include reduced

levels of electric vehicle adoption due to increased consumer price sensitivity; a decline in

industrial load growth and in the production and processing of cannabis.

42 The EPA’s Low Societal Cost of Carbon at $13.19 per metric ton begins in 2018. Source: “The Social Cost of

Carbon,” EPA. Found at https://19january2017snapshot.epa.gov/climatechange/social-cost-carbon.html

Courtesy of Deutsche Welle, www.DW.com

https://19january2017snapshot.epa.gov/climatechange/social-cost-carbon.html

http://www.dw.com/en/the-pros-and-cons-of-the-trans-pacific-partnership-pact/a-18597149

http://www.dw.com/



Business as Usual with No Carbon Scenario

The future under the Business as Usual with No Carbon scenario (“BAU w/No Carbon”)

reflects average annual load growth of 1.1% under normal historic weather conditions, before

new conservation. This scenario reflect moderate socio-economic conditions. Increased

production at the Boeing Everett plant associated with the 777X project is expected, and both

Naval Station Everett and Boeing maintain steady employment levels through the late

2020’s. The Port of Everett is actively

developing its Waterfront Place – a

mixed commercial, residential and

marina complex, located west of

historic downtown Everett,

Washington. Small businesses and

services continue to grow across time,

while growth in the industrial and

large commercial sectors remains low across the study period.

The basis for the load forecast in this scenario assumes actual historic weather and

temperatures based on data for the 1991 through 2015 period. No changes to future weather

patterns or temperatures were incorporated in the BAU scenario. Other growth factors such

as population, employment, new customer connections and electric vehicle adoption were at

the midpoint of the range (see Figures 4-6 and 4-7).

The current Washington State law that established a de minimus carbon price was assumed

across the 20-year study period, resulting in the lowest forecast fuel costs and wholesale

energy market prices of all of the scenarios.43 The scenario also assumed a future where

regional coal plants are retired44 and that the state’s Clean Air Rule will help achieve the

desired level of greenhouse gas emissions reductions. Therefore, no new state or federal

carbon policy was considered in this scenario.

43 Revised Code of Washington Chapter 80.80, Greenhouse Gas Emissions, Baseload Electric Generation

Performance Standards, located at http://app.leg.wa.gov/RCW/default.aspx?cite=80.80&full=true#80.80.040 44 The BAU scenario assumed retirement of Colstrip 1 and 2 was accelerated from 2021-2022 to 2018, see p.

10.

http://app.leg.wa.gov/RCW/default.aspx?cite=80.80&full=true#80.80.040



Business as Usual with California Carbon Scenario

The Business as Usual with California Carbon

scenario (“BAU w/CA Carbon”) assumes the

same load growth rate and socio-economic

conditions described above in the BAU w/ No

Carbon scenario. But this scenario evaluates

carbon costs at the California Cap and Trade

level, effective 2022.

This scenario considered the tradeoffs

associated with a carbon price consistent with California’s existing policy to better

understand the policy’s impacts on forecast fuel costs and wholesale energy prices. This

scenario also helped identify the most suitable mix of future demand- and supply-side

resource additions to the PUD’s existing portfolio and other considerations if a carbon policy

at this level began in 2022.

High Growth Scenario

The High Growth scenario is marked by 2.2%

average annual load growth for Snohomish

County. The socio-economic factors of

population, employment and income growth

for the Puget Sound exceed the national

average across the 20 year planning horizon.

The County’s leadership in technology and

innovation enhances its position in the global economy and the Boeing Everett Plant and

Naval Station Everett complete facility expansions. The increased cost of housing in the

greater Seattle area spurs residential development to more affordable Snohomish County.

The advancement and application of innovative new technologies makes Puget Sound a

hotbed for high tech industry, and South Snohomish County booms with new businesses and

residents. Washington State University’s newly sited campus in the county expands its

footprint to offer new STEM education and training programs to support the needs of the

region’s top employers.

https://www.bing.com/images/search?view=detailV2&ccid=LFO%2bMNSw&id=C1AFEC781FB3B8FE999B4A565B5BC2574696868C&thid=OIP.LFO-MNSwdZUMO51yeljPpAHaEc&mediaurl=http://www.abam.com/sites/default/files/project-images/portfolio-pmw-142-02.jpg&exph=600&expw=1000&q=naval+station+everett&simid=608007074519517445&selectedIndex=5



The natural gas price forecast in the High Growth scenario is the highest of the scenarios, and

includes increased environmental regulations through implementation a mid-high level of

carbon policy that averages $65 per ton. As a result, the High Growth scenario has the

highest forecast fuel costs and market prices of all the scenarios.

The volume of the demand- and supply-side resources to meet the load growth and capacity

needs after new conservation in this scenario, given the higher price market environment and

higher level of annual renewables requirements, proves challenging.

Climate Change Scenario

The Climate Change scenario assumes an average annual load growth rate of 1.0% and

applies the same socio-economic factors of population, employment and income growth

identical to those that underpin the Business As Usual scenario. The scenario incorporated

low natural gas prices and a Low Societal Cost of Carbon beginning in 2018; both affect

forecast regional fuel costs and market prices.

What’s different in the Climate Change from the BAU scenario is the use of current climate

change science to forecast future customer demand and load patterns based on expected

changes in weather patterns, and to quantify impacts to the PUD’s existing and committed

resource portfolio.45 Based on extensive work performed by the University of Washington’s

Climate Impacts Group on regional climatology, the region is expected to experience milder

winters with increasing amounts of precipitation and less snowpack. The expectation is for

increases over time in winter hydroelectric production, and reductions over time in summer

hydroelectric production due to lack of snowpack build and reduced spring snowmelt.

The Climate Change scenario highlights a different future resource need for the PUD as a

result of changes in load patterns, and later, due to changes in hydroelectric production

patterns.

45 The PUD’s load use patterns assumed an increase in baseline temperature of one degree effective 2017 and

two additional degrees by 2037. This approach best adapted the internationally and regionally-recognized best

science to the PUD’s own weather normalization model. Appendix A describes the climate change analysis in

greater detail.



Sensitivities

The 2017 IRP evaluated two sensitivities in order to query the range of impacts that could

occur as compared to the Business-as-Usual w/No Carbon scenario (BAU w/No Carbon). A

sensitivity analysis varies an element in the analysis for the purpose of asking “what if”

questions. Discussion on the two sensitivities can be found in Appendix C.

Load Forecasts

The range of load forecasts developed for the 2017 IRP rely on a mix of econometric and

deterministic approaches. An econometric approach was used for modeling historical

weather, consumption, and customer information to build a baseline from which future years

can be predicted. In building this baseline, the PUD relies on actual consumption data from

the past several years by sector, and then holding other things constant, forecasts what

consumption would have been under normal or expected historical weather.

With the baseline established, PUD staff then adjusted for expected future conditions

including changes in: population, housing type and efficiency, electric vehicle adoption,46

assumptions based on permitting by the Washington State Liquor and Cannabis Board on

grow or processing locations, county employment and projections in the goods-producing,

service-producing and military sectors, known industrial developments, and other factors.

These changes are summed and net effects are applied over the forecast period.

Figure 4-2 shows the average annual load forecast by scenario for the 2018 through 2037

study period, before new conservation.

46 Estimates for electric vehicle adoption (plug-in electric and battery electric technologies) in the PUD’s

service territory were derived from a 2017 joint study performed Energy and Environmental Economics (E3),

“Economic & Grid Impacts of Plug-In Electric Vehicle Adoption in Washington & Oregon,” March 2017. This

study was sponsored by Snohomish PUD, Chelan County PUD, Puget Sound Energy, Tacoma Power, Avista

Utilities and Seattle City Light.



Figure 4-2

2017 IRP Average Annual Load Forecast by Scenario

before New Conservation (in aMW)

The near term load growth rate between Climate Change and the Low Growth cases are

similar for different reasons. The Climate Change load growth assumed that near-term

weather will be closer to recently experienced milder winter weather, which would reduce

average annual loads. The Low Growth forecast includes lower population growth,

disposable income, and economic activity, which has a similar scaled effect in the 2018

through 2025 period. The assumption for Low Growth, BAU and High Growth load

forecasts reflect weather normalized for the 1950-1999 period, resulting in colder “normal”

winters than under the Climate Change forecast.

Figure 4-3 shows the winter load forecast by scenario for the December On-Peak hours,

before new conservation. The BAU and Climate Change forecasts share many of the same

socio-economic assumptions, but use different expectations for weather. The weather

assumption for the Climate Change scenario gradually reduces expected winter demand over

the 2018 through the 2028 period, since winter heating demand is highly dependent on

temperature.

700

800

900

1,000

1,100

1,200

1,300

aMW

Low Climate Change Business As Usual High



Figure 4-3

2017 IRP December On-Peak Load Forecast by Scenario

before New Conservation (in aMW)

Electric Vehicle Adoption

Electric vehicle (EV) adoption assumptions were built into each of the scenario load

forecasts and reflect the PUD’s expectation that EV’s may become a significant component

of future load growth. Figure 4-4 illustrates the adoption rates used in the Low, BAU and

High scenarios: the EV share of total load growth by 2037 is 12%, 14% and 14%

respectively, for these scenarios.47

Figure 4-4

Electric Vehicle Adoption Rate Assumptions by Scenario (in aMW)

47 The estimates used for electric vehicle adoption (plug-in electric and battery electric technologies) in the

PUD’s service territory were derived from a 2017 joint study performed Energy and Environmental Economics

(E3), “Economic & Grid Impacts of Plug-In Electric Vehicle Adoption in Washington & Oregon,” March 2017,

sponsored by Snohomish PUD, Chelan County PUD, Puget Sound Energy, Tacoma Power, Avista Utilities and

Seattle City Light.

1,000

1,100

1,200

1,300

1,400

1,500

1,600

1,700

1,800

aMW

Climate Change Low Business As Usual High

0

10

20

30

40

50

60

aMW

Low Business As Usual High



The inflection point in the year 2020 reflects near-term EV market development through

2020, with divergent assumptions of light duty vehicle market share and vehicle sales

thereafter. This assumption is intended to reflect anecdotal market development

observations, such as Volvo’s transition to offering only hybrid and electric light-duty

vehicles by 2019.

Cannabis Production and Growth

The PUD has observed changes in this emerging industry since Washington voters approved

Initiative 502 in November 2012. Staff monitor the Washington Liquor and Cannabis Board

website for source data of approved and pending state permits of cannabis production

facilities, and contemplated square footage for these facilities. The rates of growth included

in the Low Growth, BAU and High Growth load forecasts are shown in Figure 4-5.

Figure 4-5

Indoor Agriculture (Cannabis) Load Growth by Scenario

-

10

20

30

40

50

60

Load

in a

MW

Low Business As Usual High



Planning Assumptions

BPA Long Term Contract

As described in Section 2 – BPA Contract, the PUD contracts with BPA for the Block and

Slice products under a long term power supply agreement. The Block product supplies the

PUD with firm energy in flat monthly amounts based on the PUD’s average monthly load

shape. The Slice product provides the PUD with variable amounts of energy that depend

upon the output of the Federal System. Under the Slice product, the PUD takes responsibility

for managing its share of the output from the Federal Base System by month, day and hour,

also assuming the inherent risks. If snowpack and water conditions are above average in the

region, the energy output typically is also above average. If regional snowpack and water

conditions are low, the amount of energy the PUD derives from the Slice product would also

be reduced.

For ratemaking purposes, BPA determines the total of its customers’ loads and the Federal

System size in order to allocate costs over the two year rate period. This Rate Period High

Water Mark process establishes the maximum amount of energy the PUD is eligible to

purchase from the BPA at cost, or the Tier 1 rate. Since the new contract term began in

October 2011, the size of the Tier 1 System has varied. Tier 1 System size variations occur

due to changes in BPA’s system obligations and hydro operations, and maintenance outages

and refurbishments to the federal hydro system. Figure 4-6 shows the actual BPA Tier 1

System Size and Tier 1 contract allocation to the PUD for the 2012 through 2017 period:48

Figure 4-6

BPA Tier 1 System Size and Tier 1 Allocation to Snohomish PUD

Fiscal Year

BPA Tier 1

System Size

(in aMW)

Maximum Tier 1

Available to PUD

Rate Period High Water Mark

(in aMW)

Actual BPA Tier 1

Contract Allocation

to Snohomish PUD

(in aMW)

2012 7181 811 785

2013 7181 811 788

2014 7240 811 753

2015 6992 811 755

2016 6983 791 759

2017 6983 791 778

48 The BPA Slice product is allocated contractually based on the customer’s Slice percentage with monthly

output based on critical water; actual amounts will vary.



The 2017 IRP assumed that the BPA Tier 1 System would decline from 6,944 in 2018 to

6,850 in 2028. This assumption is based on the BPA’s most recent BP-18 rate case

documentation and studies.

Figure 4-7 shows the actual generation provided by the BPA Block/Slice contract by fiscal

year, by monthly average for December on-peak, and for the average on-peak, 80 Peak Week

Hours (Monday through Friday, on-peak hours of hour ending 0700 through hour ending

2200) for 2012 through 2017:

Figure 4-7

Snohomish PUD BPA Contract - Total Actual Generation

(Block and Slice Combined)

Fiscal

Year

Annual aMW

December

On-Peak aMW

December Peak

Week (aMW)

2012 941 1,076 1,141

2013 859 886 963

2014 859 1,016 1,047

2015 824 924 983

2016 865 1,032 1,074

2017 941 1,076 1,141

With no other information available at this time as to the types of products or contract term

length the BPA may offer in the post-2028 period, the 2017 IRP assumed that the Tier 1

System size would be allocated to customers in a similar fashion as today under a 10 year

contract term. These assumptions result in the BPA Tier 1 System size held constant at ~

6,850 aMW for the 2028 through 2037 period, resulting in the maximum amount of energy

the PUD could purchase from BPA at the Tier 1 rate of 775 aMW for the 2028 through 2037

period.

The distribution of the PUD’s contract allocation between the Slice and Block products was

assumed to remain similar to today’s at ~48% Block and ~ 52% Slice, where BPA allocates a

Slice contract amount based on the firm energy content it would yield under adverse water

conditions. The result is Block amounts ranging from 353 aMW in 2018 to 403 aMW in

2037. Slice product deliveries were simulated thousands of times using a 66 year regulated



hydro study for the water years 1950 through 2015.49 The combination of the Block and

Slice products results in a total BPA contract allocation of 733 aMW in 2018, growing to 775

aMW in 2037.50

Natural Gas

Natural gas price assumptions varied by scenario and served as both an input to the forecast

of wholesale electricity prices modeled at the Mid-Columbia trading hub, and to the

underlying fuel costs associated with certain supply side resource options. The 2017 IRP

analysis used natural gas price forecasts derived from AURORAXMP, the Council’s Seventh

Power Plan, and the Energy Information Administration’s Annual Energy Outlook for

September 2016 and January 2017 as follows:

Low Growth: Based on Council’s Seventh Power Plan’s Low Case natural gas price

forecast, ranging from $2.89 in 2018 to $5.80/MMBtu in 2037.

Business as Usual (BAU): Based on the average of the AURORAXMP 2016 base

assumption and the January 2017 Annual Energy Outlook, ranging from $3.25 in 2018 to

$6.64/MMBtu in 2037.

High Growth: Based on the September 2016 Annual Energy Outlook ranging from

$3.83 in 2018 to $10.56/MMBtu in 2037.

49 Hydro regulation data reflects operating constraints for the 2016 Water Year, informed by the 2014

Biological Opinion for fish and wildlife. See Appendix A for more on Probabilistic Load Resource Balance. 50 The PUD’s existing BPA power contract does not preclude other bilateral commercial arrangements the PUD

could make with BPA for different or future products.



Figure 4-8

2017 IRP Range of Natural Gas Prices by Scenario

(without carbon policy effects)

Carbon Costs

Carbon cost assumptions were applied to the fuel costs for supply-side resources that may

use natural gas, and to the forecast of wholesale electricity prices. This created increased

costs for the dispatch of some supply side resources within the AURORAXMP model for the

Western Electricity Coordinating Council (WECC). Increased dispatch costs within the

simulated WECC impacted the cost of wholesale electricity relative to the natural gas

forecast and carbon policy forecast, depending on the scenario. Figure 4-9 represents the

range of carbon costs considered in the 2017 IRP analysis:

$0

$2

$4

$6

$8

$10

$12

$/M

MB

Tu

Low Case Business As Usual High Case



Figure 4-9

2017 IRP Range of Carbon Prices by Scenario

Forecast Market Electricity Prices

Staff modified inputs in the AURORAXMP model to reflect natural gas and carbon cost

assumptions for each scenario with carbon costs applied to all generators in the WECC

region. To guide resource build decisions in the future, staff incorporated California,

Washington and Oregon renewable portfolio standards (RPS) and applied them to new

generating resources so sufficient energy would be available from the resources being built to

meet the states’ increasing RPS targets. At the time of modeling in early 2017, future

retirements were assumed to occur in 2018 for coal plants Colstrip 1 and 2. Load growth was

also modified to reflect scenario assumptions. These inputs were used to run a long-term

capacity study WECC-wide. The diagram of AURORA modeling and forecast electricity

market prices for the Mid-Columbia trading hub are shown in Figures 4-10 and 11:

$0

$10

$20

$30

$40

$50

$60

$70

$80

$90

$100

$/T

on

of

CO

2

Low Load Growth High Load Growth

Business as Usual w / No Carbon Business As Usual w/ CA Carbon



Figure 4-10

Diagram of AURORAXMP Modeling

Figure 4-11

2017 IRP Wholesale Market Price Forecast by Scenario

$0

$20

$40

$60

$80

$100

$120

$140

$160

$180

20

18

20

19

20

20

20

21

20

22

20

23

20

24

20

25

20

26

20

27

20

28

20

29

20

30

20

31

20

32

20

33

20

34

20

35

20

36

20

37

$/M

Wh

Business As Usual Low Case Business As Usual w/ CA carbon High Case

•Natural gas prices

•Emissions prices

•RPS constraints

•Coal retirements

•Regional load

INPUTS

•Long-Term Capacity Expansion study

AuroraXMP

• Mid-C Power Prices

Forecast Market Prices


5 ANALYTICAL FRAMEWORK

The development of an Integrated Resource Plan (IRP) involves six broad steps:

1. Creating a range of scenarios, a range of load forecasts and other planning assumptions

2. Identifying resource needs and establishing planning standards

3. Evaluating resources to meet those needs

4. Developing integrated portfolios that combine demand and supply side resources

5. Evaluating the costs and benefits of the candidate portfolios

6. Recommending and documenting a long term resource strategy

Scenarios, load forecasts and other key planning assumptions are described in Section 4. This

section addresses the analytical framework used to identify the PUD’s forecast resource need

for each scenario and established planning standards; how new conservation and energy

efficiency measures, demand response and supply side resource options were evaluated; and

how candidate resource portfolios were modeled. Section 6 discusses the resulting candidate

resource portfolios and selection of the preferred or long term resource strategy.

Identifying Future Resource Need

A significant effort in the long term resource planning process is for the utility to assess how

long it can meet its customers’ future needs with its existing energy and capacity resources,

and when it will need to plan for new resource additions. The timing of when this future need

will occur depends on the future the PUD may face – slow, moderate or more robust load

growth than the past – and the characteristics of the utility’s existing resource portfolio,

including any existing or new regulatory requirements. A utility also considers and

incorporates other criteria to mitigate portfolio risk, such as low generation years (typically

due to poor hydro conditions) and exposure to short-term market and price volatility and

other uncertainties.

The PUD’s existing power supply portfolio is predominantly comprised of hydroelectric

generation, with over 80% of its energy provided via a long term power supply contract with

the Bonneville Power Administration (BPA). BPA is the balancing authority area that the

PUD resides in, and as such, the PUD contracts with BPA for the needed reliability and

Section 5: Analytical Framework


ancillary services so it can provide reliable electric service to its customers.51 The most

significant portfolio risk the PUD faces in meeting its customer needs is the impact of low

hydro conditions.

Probabilistic Analysis of Existing Load Resource Balance

In order to understand the risks associated with low hydro conditions, the 2017 IRP analysis

incorporated a probabilistic approach to assessing the variability of its existing and

committed resources against the load forecast for each scenario. The probabilistic approach

considered the range of possible combinations generation from the PUD’s existing and

committed resources and the scenario’s load forecast, and simulated them together in an in-

house model52 that identified the timing, scale, and likelihood of the sufficiency of the PUD’s

existing portfolio to meet customer need across the planning horizon.

Probabilistic View of Annual Average Customer Demand (Load)

Figure 5-1 illustrates the probabilistic range of the PUD’s average annual customer demand

or load for the BAU scenario, where the yellow line represents the mean of the distribution,

and the lower green band denotes that the average customer load is expected to exceed this

level 95 percent of the time, before new conservation.

Figure 5-1

Probabilistic View of Snohomish PUD’s Average Annual Load Forecast

51 Since the PUD contracts with BPA for the required balancing, reliability and ancillary services associated

with grid reliability, the PUD includes no reserve margins as part of its overall resource planning process. 52 The Probabilistic Load Resource Balance Model is more fully described in Appendix A.

700

750

800

850

900

950

1000

1050

aMW

5% - 95%

+/- 1 Std. Dev.

Mean



Probabilistic View of Annual Average Existing/Committed Resources

Figure 5-2 illustrates the probabilistic range of average annual generation produced by the

PUD’s existing and committed resources, used to meet average annual customer load. The

yellow line shows the average annual portfolio production, for the 2018 through 2037 period.

The lower green band on the chart indicates that 95 percent of the time the PUD’s existing

resources will exceed this average annual production level, while the level denoted by the

upper green band will exceed this upper level of average annual production only 5 percent of

the time. The illustration depicts the decline in annual resource production over time as a

result of wind and other renewable contracts expiring, denoted by the black vertical lines

shown for years 2024, 2027, 2028 and 2029.53 The BPA Block/Slice allocations are

associated with the load growth in the BAU scenario.

Figure 5-2

Probabilistic View of Snohomish PUD’s Existing/Committed Resources

53 The 2017 IRP analysis assumed that the PUD’s long-term power supply contract with BPA continues post

2028.



Probabilistic Load Resource Balance

Combining the PUD’s probabilistic range of annual customer load with its existing and

committed resources results in a net annual load resource balance or “net position.” Figure

5-3 shows the net position under the BAU scenario before the acquisition of any new

conservation over the 20 year study period. The short and long positions shown on the left

side of the figure indicate that the PUD’s available resource position becomes deficit or

“short” when it trends below zero, and hence is not able to meet customer need on a planning

basis. The net position is said to be “long” when it exceeds zero, meaning the PUD can meet

its customers’ annual energy needs and may have some level of surplus to sell after meeting

these needs on a planning basis. The yellow line is the mean of the distribution. The lower

green band represents the P5 value for the annual net position; 95% of the time (19 out of 20

times), the PUD’s load resource balance will exceed this level. The upper green band is the

P95 value, indicating that the PUD’s net position is expected 95% of the time, and will only

exceed this level 5% of the time.

Figure 5-3

Probabilistic Average Annual Load Resource Balance before New Conservation

Business as Usual Scenario



Forecast Renewables Requirement under the EIA

To comply with the state’s EIA annual renewables target, measured as a percentage of

customer load, the PUD’s portfolio must have sufficient eligible renewable resources,

renewable energy credits (RECs), or the combination of both. Figure 5-4 shows the PUD is

forecast to meet its renewables requirement with its existing resources through 2020, when

the requirement increases to 15%. The portfolio is deficit thereafter. Satisfying the annual

renewables requirement for the 2018 through 2037 period has been incorporated into the

portfolio development model through an established planning standard (see below).

Figure 5-4 Forecast Renewables Compliance Requirements

Planning Standards

Prior IRPs established planning standards or guidelines to ensure future customer load

growth would be met on an annual and winter on-peak basis. These previous planning

standards relied on static, arithmetic metrics for determining annual hydro production. These

limited metrics tended to overstate future annual energy need and did not accurately reflect

the potential range of seasonal impacts to the PUD’s existing portfolio under poor hydro

conditions. The probabilistic approach to the PUD’s load resource balance54 provided the

platform upon which to establish new planning standards for the 2017 IRP analysis. In this

54 The PUD’s in-house Probabilistic Load Resource Balance Model is more fully described in Appendix A.

-

200,000

400,000

600,000

800,000

1,000,000

1,200,000

1,400,000

Tota

l REC

s in

MW

h

Existing REC Bank Existing Portfolio RECs Annual EIA Target

9%

2020 Renewables Target

15% of Total Retail Load



way adequate resources would be available on a planning basis to meet annual, monthly and

weekly customer demand across the IRP study period. A planning standard to measure

compliance with the annual renewables requirements prescribed under the EIA was also

included in this framework. The four planning standards established in the 2017 IRP analysis

to provide for an objective comparison of the impacts of various scenario assumptions on

future resource need are:

1. The Annual Energy Planning Standard measures the ability of the PUD to meet

average annual energy demand across the entire year. The PUD is deemed to have an

energy need if average annual load exceeds average annual resource production.

2. The Monthly On-Peak Planning Standard measures the ability of the PUD to meet

monthly on-peak demand, 19 out of 20 times, with its existing and committed resources.

Given the PUD’s existing portfolio is predominantly hydro based, the Monthly On-Peak

standard is reflective of exposure to the combination of high load and poor or adverse

water hydro conditions. This planning standard also limits the quantity of on-peak

energy/capacity purchased from the short-term wholesale energy market to no more than

100 aMW in a given month to satisfy portfolio deficits.

3. The Peak Week Planning Standard measures the ability of the PUD to reliably meet its

highest on-peak demand during the most deficit week of the month, 19 out of 20 times,

with its existing and committed resources. The highest on-peak demand has historically

occurred during December. Given the PUD’s existing portfolio is predominantly hydro

based, the Monthly Peak Week standard for on peak hours is reflective of exposure to the

combination of high load and poor or adverse water hydro conditions. This planning

standard limits the quantity of on-peak energy/capacity purchased from the short-term

wholesale energy market to no more than 200 aMW in a given month to satisfy portfolio

deficits.

4. The Regulatory Compliance Standard measures the portfolio’s compliance with the

provisions for determining cost effective conservation and annual renewables target set

forth under the Washington state Energy Independence Act (EIA). Other regulatory

requirements including consideration of overgeneration and renewable and nonrenewable

resources are also addressed through this planning standard:55

55 RCW Chapter 19.285 details conservation and renewables’ compliance requirements and RCW Section

19.280.030 addresses developing a resource plan and considering overgeneration events.



Resource Options

It is important to understand the differences among resource options available to serve future

energy and capacity needs and meet the objective of providing reliable, lowest reasonable

cost electric service to the PUD’s customers under a variety of futures. In many cases,

renewable resources have a lesser impact on the environment while non-renewable resources

are considered to be more reliable and cost-effective.

The 2017 IRP evaluated the relative costs and benefits of different types, sizes and time

constraints of commercially available resources. Supply side and demand side resources were

evaluated using the same measurements: their potential contributions to capacity, energy,

and satisfying annual renewable compliance requirements. In this way, the PUD was able to

use an integrated portfolio approach for each scenario, creating candidate portfolios that

combined the best mix of demand and supply side resources to meet future need, based on

least cost criterion.

Demand Side Resource Options

Conservation

The PUD contracted for a utility-specific analysis with the CADMUS Group, who conducted

a 2017 Conservation Potential Assessment (CPA) study. The CPA identified all achievable

technical conservation within the PUD’s service territory over the 20 year study period.56

The CPA was informed by: the PUD’s past conservation achievements; the preliminary

regional 2016 Residential Building Stock Assessment (RBSA); a preliminary oversampling

of the PUD’s service territory for certain RBSA conservation measures; and measures

identified in the Northwest Power & Conservation Council’s (Council’s) Seventh Power

Plan. The CPA informs the amount, type, and availability of conservation measures, their

associated savings, and costs.

The CPA assessed each achievable technical conservation measure, and sorted the measures

into eight different bundles by levelized cost. These bundles were then used to determine the

56 A full description of the conservation resources available to the PUD can be found in the PUD’s 2017 CPA

Report prepared by the CADMUS Group, Appendix E.



amount of conservation that is cost-effective, alongside supply side resource options, using

an integrated portfolio approach for each scenario.57

Figure 5-5 illustrates the 2017 CPA’s conservation supply curve, separated by bundle. This

supply curve facilitates comparison of demand-side resources to supply-side resources. Each

bar in the chart below represents the amount of achievable technical conservation potential

(annual and winter as measured during December On-Peak Hours based on end use profiles)

and a demand side resource option available for selection in the IRP analysis. Bundle 1

represents the conservation measures identified at a levelized cost of $45/MWh that have a

total of achievable technical potential of 88 aMW in annual energy savings over 20 years,

and 117 aMW of winter on-peak benefit over 20 years. The stacked bars in Figure 5-5 below

show total cumulative conservation grouped by levelized cost in $/MWh. The last bar to the

right represents a total of 220 aMW of cumulative conservation. This represents the

maximum amount of annual achievable technical conservation savings that could be

achieved over 20 years (through 2037).

Figure 5-5

20 Year Cumulative Conservation Supply Curve – 2018 through 2037

(Achievable Technical Potential)

57 The integrated portfolio analysis was performed in the development of the portfolios via the optimization

process. The Portfolio Optimization Model is more fully described in Appendix B.

88105 110 114

139152

177

220

117135 145 152

210228

259

321

-

50

100

150

200

250

300

350

aMW

Measures grouped or "bundled" by Levelized Cost - $/MWh

Annual aMW December HLH aMW

Bundle 1 Bundle 2 Bundle 3 Bundle 4 Bundle 5 Bundle 6 Bundle 7 Bundle 8



The residential sector accounts for approximately 68% and the commercial and industrial

sectors account for 23% and 6% of achievable technical conservation potential, respectively.

Figure 5-8 shows cumulative achievable technical potential in 2037, distributed by sector:

Figure 5-6

20 Year Achievable Technical Potential by Sector

The primary measures for the various sectors are listed below by sector. The PUD’s 2017

Conservation Potential Assessment Report, conducted by CADMUS, can be referenced in

Appendix E.

In the PUD’s 2017 CPA, incremental achievable potential was determined for each year of

the study’s planning horizon by measure turnover rates and measure-specific ramp rates.

Figure 5-7 shows cumulative 10-year and 20-year achievable technical potential by sector:

Figure 5-7

2017 CPA - Achievable Technical Potential by Sector

Sector 10–Year (2018–2027) Achievable Technical

Potential (aMW)

20-Year (2018–2037) Achievable Technical

Potential (aMW)

Residential 90 149

Commercial 40 50

Industrial 14 14

Agriculture 0 0

DEI58 7 7

Total 151 220

58 Distribution Efficiency investment (DEI) measures improve the efficiency of utility distribution systems by

operating in the lower end of the acceptable voltage range (126 to 144 volts).



Figure 5-8 and 5-9 show cumulative and incremental achievable technical potential by sector,

over the study’s 20-year horizon.

Figure 5-8

Cumulative Achievable Technical Potential by Sector

Figure 5-9

Incremental Achievable Technical Potential by Sector

CADMUS applied an even, 10-year ramp rate for all discretionary measures, resulting in

most savings occurring within the first 10 years. Under this 10 year deployment rate,

approximately 32% of the 20-year achievable potential is forecast to be acquired in the first

five years, while 69% of 20-year achievable potential is forecast to be acquired over the first

10 years. Both measure turnover and ramp rates drive acquisition rates. Incremental



achievable potential from 2028 through 2037 is driven by interactions between ramp rates for

lost opportunity measures and their natural turnover rates (determined by measure lives).

Conservation continues to be a low-cost resource, with roughly 88 aMW of achievable

technical potential available at a levelized cost of less than $45 per MWh for nearly 40% of

total cumulative 20-year achievable technical potential. The 20 year conservation supply

curve below (Figure 5-10) shows the total amount of cumulative achievable potential,

grouped by levelized cost into bundles.59

Figure 5-10

20 Year Cumulative Conservation Supply Curve by Sector

(Achievable Technical Potential)

59 Late in the IRP process a cost adjustment error was identified in the development of the conservation supply

curve associated with the 2017 CPA. Correcting the cost adjustment did not affect the total 20 year cumulative

achievable technical potential of 220 aMW in annual savings, or the 321 aMW of cumulative December On

Peak savings. Rather, it affected how measures were grouped by levelized cost. The correction resulted in

more conservation potential being available at a lower cost, which reduced overall portfolio costs, and increased

the annual to winter savings ratio from 1:1.3 to 1:1.4, on average The Long-Term Resource Strategy and 10

year conservation potential estimate was based on the corrected data shown in Figures 5-5 and 5-10.



Residential Sector

Residential customers in the District’s service territory account for 55% of total baseline

sales. The sector, divided into single-family, multifamily, and manufactured homes, provides

a variety of potential savings sources, including equipment efficiency upgrades (e.g., water

heaters, appliances), improvements to building shells (e.g., windows, insulation, air sealing),

and increases in lighting efficiency.

Figure 5-11

Residential Technical and Achievable Potential by Segment

Segment

2037

Baseline

Sales

Technical Potential—

Cumulative 2037

Achievable Technical

Potential—Cumulative 2037

aMW Percentage of

Baseline Sales aMW

Percentage of

Baseline Sales

Single Family 401 143 36% 105 26%

Multifamily—High Rise 12 4 37% 3 29%

Multifamily—Mid Rise 48 20 42% 16 33%

Multifamily—Low Rise 59 22 36% 17 28%

Manufactured 36 11 30% 8 22%

Total 556 200 36% 149 27%

Within the residential sector, central heating accounts for approximately one-third (50 aMW)

of the total cumulative achievable technical potential by end use, followed by appliances (29

aMW) and water heating (29 aMW).

Figure 5-12 indicates the cumulative and incremental residential achievable technical

potential by segment for the residential sector, across the study period.

Figure 5-12

Residential Incremental Achievable Technical Potential



Figure 5-13 lists the top 10 residential energy-conservation measures. Together, these

measures account for roughly 70% of the total residential conservation potential. Tier 3

clothes dryers—the highest energy-saving residential measure—have a weighted, average

levelized cost of approximately $130/MWh. Efficient windows also produce substantial

savings, with an average cost of $70/MWh.

Figure 5-13

Top 10 Residential Measures

Measure Name

Achievable Technical Potential—aMW

Percent of Total (20-Year)

10-year 20-Year

Clothes Dryers Any Residential BPA Tier 3 1.21 3.61 14%

Windows Multifamily - Existing Window Single Pane Base to 0.22 Window Any Electric Heat

3.3 3.3 13%

WINDOW Low-E Storm Window - Single Pane Base_Any Electric Heat

2.58 2.58 10%

Clothes Washers Any Residential CEE Tier 3 Any Water Heater/Any Dryer

1.3 1.8 7%

Home Energy Reports 1.53 1.53 6%

Windows Multifamily - Existing Window Single Pane Base to 0.30 Window Any Electric Heat

1.11 1.11 4%

Fixtures Linear Fluorescent Fixture Hard-wired 0.37 1.07 4%

Windows Multifamily - Existing Window Double Pane Base to 0.22 Window Any Electric Heat

1.04 1.04 4%

Heat Pump Water Heaters Tier 4 0.31 0.92 4%

Ductless Heat Pump 0.27 0.83 3%

Commercial Sector

The District’s commercial sector accounts for 31% of baseline sales in 2037 and 23% of total

achievable technical potential. Figure 5-14 summarizes the distribution of achievable

technical potential by commercial segment.



Figure 5-14

Commercial Achievable Technical Potential by Segment

Collectively, small and large offices account for 26% of commercial achievable technical

potential. The miscellaneous commercial segment accounts for 12% of commercial

achievable technical potential, and grocery segments account for 18% (combined).

Lighting accounts for the highest portion of total achievable technical potential. Figure 5-15

shows 20-year cumulative commercial potential by end use.

Figure 5-14

Commercial Achievable Technical Potential by End Use

Segment Baseline Sales

Technical Potential—

Cumulative 2037

Achievable Technical

Potential—Cumulative 2037

aMW Percentage of

Baseline Sales aMW

Percentage of

Baseline Sales

Cooking 8 3 41% 3 35%

Cooling 30 8 25% 6 21%

Data Center 17 6 36% 5 31%

Heat Pump 8 0 5% 0 4%

Heating 37 3 8% 2 6%

Lighting 109 31 28% 24 22%

Miscellaneous 39 3 7% 2 6%

Refrigeration 22 6 29% 5 24%

Ventilation 33 1 3% 1 2%

Water Heat 15 1 5% 1 4%

Total 316 61 19% 50 16%



Figure 5-15 shows the cumulative and incremental, achievable technical potential for the

commercial sector by segment.

Figure 5-15

Commercial Cumulative Achievable Technical Potential

Figure 5-16 summarizes the top 10 commercial conservation measures, sorted by 20-year

achievable technical potential.

Figure 5-16

Top Commercial Conservation Measures

Measure Name

10-Year

Achievable

Technical

Potential

(aMW)

20-Year

Achievable

Technical

Potential

(aMW)

Percentage

of Total

(20-Year)

LED—Linear 3.41 7.32 15%

Exterior Lighting: Façade—LED 3.22 3.82 8%

Advanced Rooftop Controller 3.17 3.17 6%

Server Virtualization/Consolidation 2.46 2.46 5%

LED—Recessed Can 1.08 2.11 4%

Electric Commercial Steam Cookers—Weighted

Average of Pan Capacities 1.38 1.65 3%

Grocery Retro-commissioning 1.58 1.58 3%

TLED Over Ballast on SP32WT8 1.29 1.53 3%

VRF 1.05 1.46 3%

Parking Garage Bi-Level LED 1.34 1.34 3%



Distribution Efficiency Investment (DEI)

DEI measures improve the efficiency of utility distribution systems by operating in the lower

end of the acceptable voltage range (126 to 144 volts). Figure 5-17 shows the distribution

system efficiency potential based on the Council’s Seventh Power Plan measures.

Figure 5-17

Cumulative Achievable Technical Potential—DEI

Industrial Sector

CADMUS estimated conservation potential for the industrial sector using the Council’s

Seventh Power Plan measures and selections from BPA’s UES measure list. The study

assessed potential for 15 industrial segments within the PUD’s service territory, based on

allocations developed from the PUD’s nonresidential database. Figure 5-18 shows the

industrial achievable technical potential by end use, and Figure 5-19 details the cumulative

and incremental, achievable technical potential for the sector.



Figure 5-18

Industrial Achievable Technical Potential by End Use

Figure 5-19

Industrial Cumulative Achievable Technical Potential



Demand Response

Demand Response programs entail coordination with customers to alter their energy

consumption patterns to help the PUD defer customer demand in a time period with peak

load pressure to a time period with less peak load pressure. An example of this type of

program is the recent BPA Commercial & Industrial Load Curtailment pilot program, in

which some large scale customers reduced their energy consumption during peak periods in

exchange for monetary compensation. Demand Response is being increasingly viewed as a

significant resource in the region to temporarily assist with meeting peaking and system

flexibility and reliability needs.

2017 Demand Response Potential Study

As part of the 2017 IRP effort, the PUD contracted with The CADMUS Group for a 20 year

demand response potential assessment to identify demand response potential by products and

levelized cost to inform the demand side resource options evaluated in the 2017 IRP analysis.

The study identified five specific product types and potential in the PUD’s service territory:

1. Residential space and water heat direct load control (DLC);

2. Residential water heat DLC;

3. Residential Wi-Fi thermostat DLC;

4. Residential smart water heaters; and

5. Commercial and industrial load curtailment.

Figure 5-20 below identifies the 20 year cumulative demand response potential available by

program, to serve the December Peak Week hours, or the 80 on-peak hours during the Peak

Week, Monday through Friday, from hour endng 0700 through hour ending 2200.



Figure 5-20

2017 Demand Response Assessment – Cumulative DR Potential by Program

Demand Response (DR) Program

Cumulative DR Potential

through Dec 2037

December Peak Week

(in aMW)

20-YearProgram

Net Present Value60

(NPV)

$ per Peak MW

(in $/MW-Year)61

“Smart” Thermostats 12.8 $16,221,880 $148,347

Commercial/Industrial Controls 6.3 $12,025,108 $179,086

Thermostats & Water Heaters 83.3 $153,938,733 $216,233

Water Heaters (<80 gallons) 78.4 $249,071,654 $371,788

“Smart” Water Heaters (>80

gallons) 34.3 $152,309,616 $728,778

Typically, participants in demand response programs are limited on the number of times they

are willing to be called on to provide the DR service. The amount of demand response

potential available to serve the 80 hours during the peak week of December is notable, as

shown in Figure 5-21. The demand response potential and associated capacity costs were

included as individual demand side resource options and evaluated as part of the integrated

portfolio optimization process for each of scenario.

Figure 5-21

2017 Demand Response Potential Assessment

December Peak Week (Mon-Fri, On-Peak Hours)

60 The CADMUS Demand Response Potential study did not include advanced metering infrastructure (AMI)

costs. 61 The $ per Peak MW for DR was computed by assessing the 20 year program administration costs and Peak

Week contributions for each program

0

50

100

150

200

250

aMW

Direct Load Control -Smart Thermostats

Direct Load Control -HVAC & Water Heaters

Direct Load Control -Water Heaters

Smart Water Heaters

Commercial &Industrial LoadCurtailment



Supply Side Resource Options

Commercially Available Resources

The PUD surveyed other Pacific Northwest utilities’ IRPs, the Council’s Seventh Power

Plan, research papers and other sources to gather cost and operations data on renewable and

nonrenewable generating resources. Resources were then screened based on their commercial

availability for the IRP study period. Commercial availability was based in part on whether

there was sufficient information available to estimate the resource’s production costs, and

whether the resource could be reasonably expected to be permitted and constructed within

the next few years, with consideration given to whether the resource technology was

sufficiently mature in the regional marketplace and available to the PUD.

Developing and emerging technologies not yet commercially available for consideration in

this IRP are being monitored and include: offshore wind, wave projects, and small modular

nuclear reactors.62 Costs associated with repowering existing wind projects when they reach

the end of their asset life was not considered at this time due to lack of cost and operational

data.

Costs

Development and production costs used to evaluate renewable and non-renewable resources

included the recovery of the capital investment, operations and maintenance costs, and the

cost to transmit or deliver the energy or capacity to the PUD. Renewable resource costs

represent the total investment for all of the resource characteristics (energy and capacity

bundled with the environmental attributes or RECs) and assumed the Federal production tax

credit or investment tax credit would not be renewed or extended beyond 2020. Figure 5-22

shows the supply side resource options’ Nameplate, 20 year Levelized Cost ($/MWh), and

Levelized Cost per MW of Peak Week Capacity ($/MW).

62 Modular nuclear reactors are currently being tested in the region, however this new technology was not

included due to insufficient pricing information associated permitting, construction and development of this

type of facility.



The Levelized Cost of Capacity reflects both the cost of resource development, and the

capacity available with 95% certainty during the Peak Week period in the month of

December, when the PUD’s peak demand and capacity needs historically have been the

largest. This metric, expressed in Levelized Cost per December Peak Week MW, provides a

useful comparison of the cost and reliability among the various capacity resource options.

End effects were included in resource costs, such that resources of differing life spans were

evaluated on the same basis. [See Appendix B, page 18, for more detail.]

Figure 5-22

Supply Side Resource Options

Resource Nameplate

(MW)

Levelized Cost

($/MWh)

Levelized Cost

Peak Week

Capacity ($/MW)

Simple Cycle Combustion Turbine (SCCT) 239 $ 137 $ 118,163

25 MW Short Term Capacity Contract (5 year) 25 $ 155 $ 129,496

DR – Direct Load Control Smart Stat 20.5 $ 1,854 $ 148,347

Dual Fuel Reciprocating Engine 50 $ 199 $ 170,841

DR - C&I Curtailment 0.1 $ 2,239 $ 179,086

DR – Direct Load Control Air & H20 Heat 0.8 $ 2,703 $ 216,233

Pumped Hydro Storage Low 100 $ 248 $ 286,148

DR – Direct Load Control H20 Heat 0.7 $ 4,647 $ 371,788

Pumped Hydro Storage High 100 $ 330 $ 381,559

Landfill Gas 10 $ 75 $ 545,327

Biomass 15 $ 91 $ 572,799

Geothermal (traditional) 25 $ 87 $ 578,937

DR - Smart H20 & Heat 34.3 N/A $ 728,778

Energy Storage – Battery 25 $ 373 $ 764,313

Long Distance Wind (Montana) 50 $ 69 $ 881,933

Run of River Hydro (small hydro) 30 $ 113 $ 2,451,503

WA/OR New Wind* 50 $ 77 $ 6,685,044

Lower Cost Utility Scale Solar (E Wash) 25 $ 94 $ 8,229,719

Customer Owned DG 15 $ 125 $ 8,304,656

Utility Scale Solar (U/S Solar E Wash) 25 $ 97 $ 13,912,590

Utility Scale Solar (U/S Solar W Wash) 5 $ 182 $ 39,824,048

For consistency with the probabilistic modeling of the PUD’s load resource balance, staff

forecast the probabilistic generation for each supply side resource option, across different

time periods. The information in Figure 5-23 shows the expected or annual production

modeled for each resource type (P50 value). The contribution of the resource to meeting the



Peak Week on-peak hours was assumed to be P5 value for the resource output, or the

minimum amount of generation the PUD could expect from the resource 19 out of 20 times.

This approach was intended to reflect forced outage rates and fuel uncertainty (water, wind,

and solar).63

Figure 5-23

Supply Side Resource Options - Probabilistic Generating Characteristics

Resource Nameplate

(in MW)

Average Annual Energy

(in aMW)

Winter December On-Peak (in aMW)

Winter December Peak Week (in aMW)

Summer August

On-Peak (in aMW)

Summer August

Peak Week (in aMW)

Long

Ter

m C

apac

ity

Res

ourc

es

Simple Cycle Combustion Turbine (SCCT) 239 23.9 231.8 231.8 231.8 231.8

25 MW Short Term Capacity Contract (5 year contract term) 25 2.5 25.0 25.0 25.0 25.0

Dual Fuel Reciprocating Engine 50 5 48.5 48.5 48.5 48.5

Pumped Storage Hydro 100 18 95.0 97.0 97.0 97.0

Energy Storage – Battery 25 5 12.5 12.5 12.5 12.5

Dem

and

Res

pons

e DR – Direct Load Control Smart Stat64 20.5 0.1 2.5 12.8 - -

DR - C&I Curtailment 0.1 0.1 1.2 6.3 - -

DR – Direct Load Control Air & H20 Heat 0.8 0.8 16.0 83.3 - -

DR – Direct Load Control H20 Heat 0.7 0.7 15.1 78.4 - -

DR - Smart H20 & Heat 34.3 34.3 34.3 34.3 34.3 34.3

Bas

eloa

d R

enew

able

s

Landfill Gas 10 8.5 8.5 9.7 8.5 9.7

Biomass 15 12.8 12.8 14.6 12.8 14.6

Geothermal (traditional) 25 22.5 22.5 24.3 22.5 24.3

Var

iabl

e R

enew

able

s

Long Distance New Wind (Montana) 50 21.8 11.6 11.6 11.6 11.6

Run of River Hydro (small hydro) 30 13.6 4.3 4.3 - -

WA/OR New Wind 50 17.5 5.9 1.7 16.1 5.2

Customer Owned Distributed Generation 15 1.7 0.4 0.2 2.4 1.2

Utility Scale Solar (U/S Solar E Wash) 25 6.8 0.8 0.4 2.8 1.4

Utility Scale Solar (U/S Solar W Wash) 5 0.6 0.1 0 0.3 0.2

Emissions

The 2017 IRP analysis made broad and general assumptions about emissions attributable to

certain supply side resource options as described below in the Resource Modeling

Assumptions. Data sources used included the EPA’s carbon emission factor set and the

emissions modeled in the AURORAXMP software that represent the WECC-wide market.

63 These costs are generic by resource type. If the PUD were to acquire an actual resource, it would perform its

due diligence and conduct a comprehensive economic analysis on a site specific resource, which may look

different from the costs and characteristics shown in Figures 5-9 and 5-10. 64 Demand response resources were modeled as available on a call limited basis and used to serve Peak Week in

the month of December; they were not considered to be available in any other time period.



Resource Modeling Assumptions

The following represent the resource modeling assumptions used in the 2017 IRP analysis for

the supply side resource options:

Short Term Capacity Contract (5 year contract term)

• Based on annual levelized capacity costs of a Simple Cycle Combustion Turbine (SCCT)

• Annual dispatch assumed at ~10%

• Emissions based on estimated regional Northwest Power Pool Net System Market Mix

Renewable Resources (Solar, Wind, Small Hydro, Landfill Gas, Biomass, Geothermal)

• Resource costs based on survey of other regional utility IRPs and market research

• Assumed federal tax incentives not continued after 2020 expiry

• Annual dispatch reflects ‘must run’ resource, less forced outage rate

• Emissions based on resource and fuel type

Fossil Fueled Resources (Simple Cycle Combustion Turbine and Reciprocating Engine)

• Resource costs based on survey of other regional utility IRPs and market research

• Annual dispatch assumed at ~10%, fueled by natural gas

• Emissions based on plant efficiency and carbon content of fuel (~117 lbs/MMBtu)

Pumped Hydro Storage (100 MW nameplate)

• Resource costs based on market research and internal studies

• Annual dispatch assumed at ~18%

• Roundtrip efficiency estimated at ~80%

• Fuel costs assumed to be market purchases at forecast market price

• Emissions based on estimated annual WECC Market Mix



Overgeneration Events

The provisions under RCW 19.280.030 – Developing a Resource Plan, were expanded in

2013 and require the utility’s IRP to describe not only the mix of supply and demand side

resources but, where applicable, how its resource plan addresses overgeneration events. The

PUD’s service area resides in and is served by the Bonneville Power Administration’s

(BPA’s) footprint or BPA Balancing Authority Area. As the Balancing Authority Area, BPA

is responsible for moment-to-moment balancing of loads and resources within its footprint,

including for the PUD. BPA mitigates overgeneration conditions or oversupply events on a

regional basis through its Oversupply Management Protocol. An oversupply event is an

event that historically occurs in the late spring, and is marked by moderate temperatures that

reduce demand at the same time regional snow melt and spring rains resulting in high

hydroelectric energy production that combine with high energy production from regional

wind projects. The PUD’s portfolio is subject to BPA’s Oversupply Management Protocol

and pays the oversupply rate assessed by BPA.65

Unbundled Renewable Energy Credits

The cost to purchase the environmental attributes or renewable energy credits (RECs)

associated with a renewable resource were modeled and made available in the 2017 IRP

analysis as an investment option for meeting the PUD’s annual EIA renewables compliance

requirement. The environmental attributes or RECs associated with energy produced by a

Washington state eligible renewable resource can be purchased separately from the energy

itself. The modeling assumption for unbundled RECs was that the seller of the REC owns or

contracts for the renewable resource and may have RECs surplus to their own compliance

need or are trying to maximize revenue from the energy and REC streams for their project

portfolio.

Today, the Northwest has a reasonably liquid bilateral market for unbundled RECs, with

REC prices forecast for the 2018 through 2022 period near $5.00 per REC. Staff anticipate

that as renewables compliance requirements in Washington state increase from 9% to 15% of

65 BPA’s Oversupply Management Protocol and Oversupply Rate can be found at

https://www.bpa.gov/Projects/Initiatives/Oversupply/Pages/default.aspx.

https://www.bpa.gov/Projects/Initiatives/Oversupply/Pages/default.aspx



load in 2020, and both California and Oregon experience increases in their annual renewables

requirements, the availability of surplus RECs in the region will begin to diminish, increasing

the cost for unbundled RECs beyond levels being quoted by brokers in the short-term REC

markets today. Because the long term market for RECs in the 2020’s through 2037 is less

liquid and price discovery is limited, the PUD chose a conservative approach to modeling

REC prices across the 20 year IRP study period.66

REC Availability Assumption

The volume of unbundled RECs assumed to be available from the market were in increments

of 100,000 MWh, available under a 25 year contract term. The modeling limitation was a

modeled maximum of 1,000,000 unbundled RECs during any calendar year.

REC Price Assumption

REC prices were modeled based on the question - if REC prices over the 20 year IRP study

period were modeled at their theoretical maximum, would the PUD have a preference to

purchase unbundled RECs or to invest in a renewable resource (energy, capacity and REC)

as a future generating asset. With support to consider unbundled RECs in the 2017 IRP

analysis, REC prices were subsequently modeled at the cost to develop a new Northwest

wind project, where project development costs were reduced by the value of the project’s

energy sold at forecast market prices over a 25 year asset life. The remaining value was

allocated to the environmental attributes or RECs associated with the project’s annual

production as follows:

REC Price = Levelized cost of wind – Levelized cost of energy

$43/REC in 2018 to $53/REC in 2037 67

Staff recognizes this is a conservative approach to modeling the long term price for RECs,

and yet, each portfolio selected some amount of unbundled RECs for meeting the EIA annual

66 In selecting the Long Term Resource Strategy, staff performed a sensitivity with REC prices at varying levels

to test the impact on the portfolio. Results are detailed in Section 7, Figure 7-10. 67 REC pricing and renewables compliance modeling are more fully described in Appendix B.



renewables target. This is primarily due to the lack of annual energy need by the PUD after

new conservation under average water conditions, across the 20 year study period, for all but

the High Growth scenario.68

Figure 5-24 below shows the forecast incremental REC need by year for the Climate Change

scenario, after new cumulative conservation:

Figure 5-24

Forecast Incremental REC Need by Year after New Conservation

(Climate Change Scenario)

68 The High Growth scenario has an annual energy need after new conservation, beginning in 2034. This

combined with the cap on the number of unbundled RECs that could be acquired in a year drove acquisition of

Washington state eligible renewable resources and RECs to satisfy the forecast annual renewables compliance

obligation.

0

2

4

6

8

10

12

14

16

18

20

REC

s ex

pre

ssed

in 1

00

k M

Wh

s

REC Bank Anticipated Unbundled REC Purchase

Physical REC Adds (Customer Owned DG) Existing Portfolio REC Generation

REC Target



Developing Integrated Portfolios

The goal of the 2017 IRP analysis, consistent with the requirements in RCW 19.280.030 –

Developing a Resource Plan, is to integrate into a long-range assessment the lowest

reasonable cost mix of supply and demand side resources that meets current and future needs

under a range of scenarios or futures. To perform this analysis, the PUD used an integrated

portfolio approach, established parameters based on the planning standards, selected from the

demand and supply side resource options, and developed candidate portfolios for each

scenario.

Integrated Portfolio Modeling Approach

An in-house portfolio optimization model was developed to solve for the lowest reasonable

cost portfolio, that satisfied all planning standards and constraints, for each scenario. This in-

house model69 calculated thousands of possible combinations of supply side resources,

conservation by cost bundle, demand response programs and unbundled RECs, and solved

for the optimal combination of demand- and supply-side resources, resulting in the lowest

reasonable cost, identified as the incremental net portfolio costs net present value (NPV) for

the scenario. Section 6 and Appendix C provide additional detail on the new resource

additions for each portfolios in the 2017 IRP analysis.

69 The in-house Portfolio Optimization Model was used to develop candidate portfolios and is described in

Appendix B.

Establish Planning

Standards

•Annual energy

•Monthly On-Peak (Dec)

•Peak Week (Dec HLH)

•Regulatory (I-937)

Optimize Portfolios by

Scenario

•Select from demand & supply side resource options

•Assumes 10 year conservation ramp

•Derives optimal combination of new resource additions within modeling constraints and satisfies all Planning Standards

Generate Lowest Cost Portfolio by

Scenario

•Net Portfolio Cost (NPV)

•10 and 20 Year Economic Conservation Potential

•Renewables Req’t (I-937)

•Incremental Emissions

Parameters for satisfying I-937 with

RECs


6 PORTFOLIOS AND PROPOSED LONG TERM RESOURCE STRATEGY

The 2017 IRP developed candidate portfolios designed to meet a range of futures that differed in

the combination of load growth, carbon policy and other socio-economic conditions. These

varying futures and conditions are described in Section 4. This section characterizes and contrasts

the following five constructed integrated resource scenarios:

1. Low Growth w/Low Societal Cost of Carbon

2. Business as Usual with No Carbon (BAU w/No Carbon)

3. Business as Usual with California Carbon (BAU w/CA Carbon)

4. Climate Change w/Low Societal Cost of Carbon

5. High Growth w/Mid-High Societal Cost of Carbon

These portfolios were evaluated with a broad range of renewable and nonrenewable resources

required by state rules for IRP planning,70 which informed the selection of a preferred or Long

Term Resource Strategy. The Long Term Resource Strategy represents the most effective mix of

demand and supply side resources that consider supply availability, delivery risk, energy-related

regulatory policies and other uncertainties.

Portfolio Development

The process used to construct the portfolios for each scenario was the same. An in-house model

simultaneously identified the optimal mix of conservation measures, demand response, supply-side

resource and renewable energy or REC options to augment deficits or shortfalls identified by the

established planning standards and market environment. The portfolios were evaluated under

expected conditions and adverse conditions, or conditions that deviated from expected.71 The

scenarios helped identify impacts and test the resilience of each portfolio.

70 Referenced requirements are detailed in the Revised Code of Washington, Section 19.280.030. 71 Section 5 details the Analytical Framework and Planning Standards and probabilistic approach to the PUD’s

existing/committed load resource balance, including the expected (average or P50)and adverse conditions (P5) for

testing the portfolios.

Section 6: Portfolios and Proposed Long Term Resource Strategy


The in-house portfolio model identified the incremental costs and benefits associated with each

candidate portfolio. Portfolio costs were measured as the net present value (NPV) of the

incremental cost of new resource additions to the portfolio over the 2018 through 2037 study

period. Portfolio revenues, if any, were determined by forecasting any portfolio energy surpluses

due to the timeing of additions. These surpluses were modeled as sold into the market at the

forecast market price. Revenues, if any, were then used to offset portfolio costs, resulting in a net

portfolio cost, measured in NPV. Net portfolio costs included fuel costs and the cost of carbon, if

applicable, to supply side resources in the portfolio, based on the carbon policy assumed by the

scenario.

Figure 6-1 illustrates net present value (NPVs) of the total portfolio costs and the net portfolio

costs by scenario. The total portfolio cost NPV for incremental resource and REC additions is the

sum of the stacked bars for each scenario. The dashed line notes the net portfolio cost NPV, which

includes the offset due to modeled NPV portfolio revenues created from surplus portfolio energy

sales over the study period.

Figure 6-1

Comparison of 2017 IRP Portfolios by Scenario

Total vs Net Cost NPV

$0

$500

$1,000

$1,500

$2,000

$2,500

$3,000

$3,500

$4,000

Low Load Growth Case Business As Usualw/No Carbon

Business As Usualw/CA Carbon

Climate Change w/Low Societal Cost of

Carbon

High Load GrowthCase

Tota

l Po

rtfo

lio C

ost

(M

illio

ns)

New Cumulative Conservation New Supply-Side Resource Additions

New DG Additions Forecast RECs

Net Portfolio Costs (NPV)

$317

$702

$481

$422

$1,369



Portfolio Findings

Several general trends and insights emerged in the development of the portfolios:

1. Conservation is the single largest resource addition across all scenarios.

2. The PUD has no annual energy need after conservation, except in the High Growth case

unless:

o Region experiences poor or adverse hydro conditions;

o Operational changes are made to Federal System operations that in turn affect the PUD’s

offtake under its BPA power supply contract;

o BPA changes product offerings post 2028.

3. RECs were the most cost effective way to satisfy the PUD’s annual renewables requirements,

except in the High Growth case.

4. The High Growth portfolio required a mix of renewables and capacity, or seasonal and peak

matching resources, after cumulative conservation to satisfy annual energy and long term

capacity needs; this portfolio had fewest unbundled REC purchases, since annual renewables

requirements were met in part with renewable resource additions.

5. Planned new resource additions in the 2017 IRP analysis were driven by:

Forecast shortfalls after the acquisition of new cumulative conservation and short-term

market allowance, that exceeded the Monthly On Peak and Peak Week planning standards;

Need that was best met by the addition of capacity or seasonal and peak matching resources

to augment the PUD’s existing portfolio; and

The need to satisfy the PUD’s annual renewables requirements (EIA).

These findings and other key insights are more fully explored in Section 7.



Portfolio Results

The final candidate portfolio for each scenario represents the lowest reasonable cost combination

of resources that satisfied all established planning standards. Figure 6-2 denotes the ranking of the

scenario and sensitivity portfolios developed in the 2017 IRP analysis, by net portfolio cost NPV.

The portfolios for the five scenarios are denoted by the blue bars. The green bars show the

sensitivity portfolios.

Figure 6-2

2017 IRP Final Portfolio Ranking by Net Portfolio Cost NPV

Figure 6-3 summarizes the resource and REC additions for the five scenarios:

Figure 6-3

Summary of Portfolio Resource Additions by Scenario (in aMW)

Scenario Total

Conservation

Short Term

Capacity

Contract

(Dec HLH

aMW)

Long Term

Capacity

(Dec HLH

aMW)

Renewable

s (aMW)

RECs

(aMW)

Low Growth w/Low Carbon 121 25 n/a n/a 68

BAU w/No Carbon 92 25 232 n/a 78

Climate Change w/Low Carbon 114 50 116 3 68

BAU w/CA Carbon 152 n/a 97 1 72

High Growth w/Mid High Carbon 152 n/a 396 68 22

$317

$422

$481

$653

$702

$760

$891

$1,369

$- $500 $1,000 $1,500

Low Growth Scenario

Climate Change with Low SCC

BAU w/ California Carbon

BAU w/ California Carbon - Renewables Only

BAU w/ No Carbon

BAU w/ No Carbon and No Snakes

BAU w/ No Carbon - Renewables Only

High Growth Case

Net Portfolio Cost NPV in Millions



The following section details the portfolios by scenario, and illustrates how the planned resource

additions address the PUD’s future December On-Peak need:

Low Growth w/Low Societal Cost of Carbon

Figure 6-4 details portfolio resource additions for the Low Growth scenario to address forecast

December On-Peak deficits across the 2018 through 2037 study period. This scenario incorporates

the EPA’s Low Societal Cost of Carbon, which increases fuel costs and forecast market prices,

making more cumulative conservation cost effective. The acquisition of larger quantities of new

conservation frees up existing/committed PUD resources, which are then modeled as surplus

portfolio energy sales. Revenues derived from such sales are used to directly offset portfolio costs.

The slow rate of annual load growth in this scenario is met by the addition of 121 aMW total

cumulative conservation (160 aMW total cumulative in winter); a 25 MW short term capacity

contract for the 2018-2022 period; and the addition of demand response (C&I Curtailment

program). Cumulative conservation in this scenario helps meet the forecast winter capacity need

post 2022. Renewables requirements are met through the acquisition of ~ 68 aMW of unbundled

RECs over the study period, with initial purchases beginning in 2020.

Figure 6-4

Low Growth Portfolio - December On-Peak (in aMW)

-

200

400

600

800

1,000

1,200

1,400

aMW

Existing Resources New Conservation

Short Term Capacity Contract Demand Response- C& I Curtailment

Forecast Market Purchases Load

New Resource Nameplate MW 1st Yr Added

Short Term Capacity Contract 25 2018

Demand Response 1.2* 2018

New Cum. Conservation 121 2018

Forecast RECs 68 2020

Net Portfolio Cost NPV ($000s) $317,088

*DR contribution measured in terms of December on peak hours



Business as Usual with No Carbon

Figure 6-5 details the BAU w/No Carbon portfolio. This scenario reflects the current planning and

market environments with no carbon policy to affect fuel costs or market prices. The scenario

reflects historic load growth and resource production with static or normal weather, and gives no

consideration to the impact of climate change on weather, load or annual resource production.

Net portfolio costs for this portfolio are high at $702 million NPV, and the amount of cumulative

conservation is lower than in other scenarios. A 25 MW short term capacity contract and 92 aMW

of new cumulative conservation (107 aMW total cumulative in winter), defers the need for long

term capacity resource to 2023 in this scenarios. After conservation, there is no annual energy need

in this scenario; approximately 78 aMW of RECs from Washington-eligible renewables projects is

acquired over the 20 year period beginning in the year 2020 to meet the EIA requirements.

Figure 6-5

BAU w/No Carbon – December On-Peak (in aMW)

-

200

400

600

800

1,000

1,200

1,400

aMW

Existing Resources New Conservation Short-Term Capacity Contract

Long Term Capacity Resource Forecast Market Purchases Load



New Conservation 92 2018

Long Term Capacity Resource 232 2023

RECs 78 2020




Business as Usual with California Carbon in 2022

The Business As Usual with California Carbon in 2022 scenario (BAU w/CA Carbon) reflects the

same socio-economic and load growth assumptions as the BAU w/No Carbon scenario, and

incorporates a California carbon policy cost beginning in 2022. Figure 6-6 shows the portfolio for

the December On-Peak period across the years 2018 through 2037. As modeled, these carbon costs

increased fuels costs and market prices. The higher the fuel costs and market prices, the larger the

amount of cumulative conservation that is deemed to be cost effective. [BAU w/No Carbon

identified 92 aMW of cost effective conservation.]

New resource additions include: 152 aMW of new cumulative annual conservation (228 aMW

cumulative toward winter) that defers the need for a long term capacity resource until 2029 and

2031; 1.2 aMW of demand response is added from commercial and industrial programs; and 5

MW from utility scale solar projects are added in 2029 and in 2032, for a total of 10 MW (Western

WA solar). With no explicit energy need in the portfolio after conservation, the renewables

requirement is met through acquisition of ~72 aMW unbundled RECs and from the two utility

scale solar projects.

Figure 6-6

BAU w/ CA Carbon – December On-Peak (in aMW)

-

200

400

600

800

1,000

1,200

1,400

aMW


Demand Response - C&I Curtail Western WA Solar

Western WA Solar 2 Long Term Capacity Resource 1

Long Term Capacity Resource 2 Forecast Market Purchases

Load


Demand Response 1.2* 2018

New Cum. Conservation 152 2018

Util Scale Solar (W WA) 5 2029, 2032

Long Term Capacity Resource 100 2029, 2031

RECs 72 2019


*DR contributes 1.2 aMW to December On-Peak Hours



Climate Change with Low Societal Cost of Carbon

For the Climate Change portfolio, there are two fundamental differences in the operating and

market environment for this scenario that impact the future resource need and subsequent portfolio

additions. The first difference is that the portfolio reflects the impact of climate change on the

PUD’s existing resources and its forecast average load over the 20 year study period. In broad

terms, the PUD requires a smaller amount of future resources to serve winter needs under the

Climate Change scenario. This is due to the expected increase in winter hydroelectric production –

more precipitation is expected to fall as rain, and regionally snowpack is expected to decline. The

gradual decline in snowpack, which is a natural storage of water between winter and spring

periods, is expected to reduce stream flows, making less hydroelectric production available during

the summer months.72 This, combined with forecast warmer temperatures, creates an emerging

summer need for the PUD near the end of the study period. The net effect on the portfolio is a

subtle decrease in overall winter capacity, but with a new requirement for capacity that can

serve both winter and summer seasonal and peak customer demand across time. Figure 6-7

highlights changes to the PUD’s net monthly net load resource balance for 2027 and 2037,

compared to 2017:

Figure 6-7

Impact of Climate Change on Summer and Winter Load Resource Balance

72 The scale of the climate change impacts modeled in the 2017 IRP were consistent with the regionalized effects of

Representative Concentration Pathway (RCP) 4.5, characterized as “low climate change” in the 5th Assessment

Report, UN Intergovernmental Panel on Climate Change, at https://ipcc.ch/report/ar5/.

(400.0)

(300.0)

(200.0)

(100.0)

-

100.0

200.0

Net

On

-Pea

k Po

siti

on

at

P5

(in

aM

W)

2017 HLH @P5 2027 HLH LRB @P5 2037 HLH LRB @P5

https://ipcc.ch/report/ar5/



The second important difference in this scenario is the introduction of costs reflective of the EPA’s

Low Societal Cost of Carbon effective in 2018.73 This carbon policy increased fuel costs and

wholesale energy prices, which in turn increased the amount of cumulative conservation acquired

compared to the BAU w/No Carbon scenario, changing the timing and amount of future supply-

side resources that would otherwise be required.

This scenario includes the addition of a 50 MW short term capacity contract beginning in 2018,

116 aMW of long term capacity, and annual cumulative conservation savings of 152 aMW over

the 20 year planning horizon. The conservation acquisition defers the PUD’s need for a long term

capacity resource until 2028.

Figure 6-9

Climate Change with Low Societal Cost of Carbon – December On-Peak (aMW)

73 See Section 4 Planning Assumptions for more discussion on the EPA societal cost of carbon.

-

200

400

600

800

1,000

1,200

1,400

aMW


Long Term Capacity Resource Short-Term Capacity Contract

Distributed Generation Forecast Market Purchases

Load



New Cum Conservation 152 2018

Distributed Generation 15 2019


RECs 68 2021

Net Portfolio Cost NPV ($000) $422,159



The Climate Change w/Low Carbon has no annual energy need after new cumulative conservation.

Therefore, no baseload energy or renewable resources were added to the portfolio. Annual

renewables requirements are satisfied through the procurement of ~68 aMW of RECs from

Washington-eligible renewable energy projects beginning in 2021, and the addition of customer-

owned distributed generation resources residing within the PUD service territory (likely rooftop

solar).

High Growth with Mid-High Societal Cost of Carbon

The High Growth scenario is characterized by the highest rate of average annual load growth in the

2017 IRP analysis. Even after the identified cumulative conservation additions, the need for

resources across the study period to meet average annual energy and winter capacity needs is

notable. The High Growth case is the only scenario with a measurable annual energy need after

new cumulative conservation. This need for annual energy changed the type of resources selected

by the portfolio model to satisfy the annual energy planning standard.

This scenario also included the highest natural gas price forecast and carbon costs in all of the

2017 IRP scenarios. The results was higher fuel costs and sharp increases in the cost of supply

side capacity resources, and increases to the forecast wholesale market prices. Higher fuel costs

and market prices incentivized higher levels of conservation acquisition compared to the BAU

w/No Carbon case. A diverse set of resource types were added in order to satisfy both the annual

energy, winter capacity and larger EIA renewables requirements beginning in 2018. Conservation,

Biomass, Geothermal, Demand Response, New Wind, and Long Term Capacity Resources were

all selected for this portfolio.

Figure 6-10 details the new resource additions and year added for the annual average position

versus annual load forecast. Figure 6-11 illustrates the same resource additions and how they serve

the December On-Peak need. Approximately 68 aMW of renewable resources were added to meet

annual energy needs, and aided in meeting the EIA renewables requirement. Only 22 aMW of

RECs from Washington-eligible renewable projects were required to meet the EIA beginning in

the 2021.



Figure 6-10

High Load Growth Portfolio – Annual Energy (aMW)

Figure 6-11

High Load Growth Portfolio – December HLH

-

200

400

600

800

1,000

1,200

1,400

aMW

Existing Resources

New Conservation

Demand Response -Smart StatDemand Response -Air& H20 HeatDemand Response - C&ICurtailBioMass 1

BioMass 2

WA Wind

Long Term CapacityResource 1Long Term CapacityResource 2Long Term CapacityResource 3Geothermal 1

Geothermal 2

Load

-

200

400

600

800

1,000

1,200

1,400

1,600

1,800

aMW

Geothermal 2

Geothermal 1

Long Term CapacityResource 3Long Term CapacityResource 2Long Term CapacityResource 1WA Wind

Biomass 2

Biomass 1

Demand Response -C&I CurtailDemand Response - Air& H20 HeatDemand Response - T-StatsForecast MarketPurchasesNew Conservation

Existing Resources

Load


Demand Response 19* 2018

New Conservation 152 2018


New Wind - NW 50 2023

Geothermal 50 2022 (2)

Biomass 30 2023(2)



RECs 22 2019

Net Portfolio Cost NPV ($000s) $1,369,600

*DR contributions during December On-Peak Hours



Determination of the Proposed Long Term Resource

Strategy

A Long Term Resource Strategy must reflect and balance the potential for risk and uncertainty,

while identifying a lowest reasonable cost portfolio approach. The two driving factors identified

during the course of IRP analysis were the impact of climate change on the PUD’s existing

resource portfolio and forecast future needs, and the impact of a potential carbon policy on the

market environment the PUD operates in.

Climate change effects the PUD’s portfolio over the study period by shifting how and when energy

is delivered by the PUD’s existing owned and contracted hydropower resources, and changing the

timing and scale of customer need and use patterns. Generally, the climate change impact on

hydropower shifts more of the potential energy into the winter period from summer due to warmer

weather that reduces snowpack at increasingly higher elevations and suggests that more

precipitation will fall as rain instead of snow. This helps the PUD better meet winter customer

demand needs, but does so at the expense of reduced summer hydropower production largely due

to less snowpack and runoff. Customer demand is expected to gradually decline in winter due to

milder temperatures, and increase in summer due to warmer temperatures that will increase

cooling load (air conditioning).

The combined effect of these impacts is shown in Figure 6-12, as an impact to the monthly load

resource balance or net position over time.74 The light blue line for 2018’s Monthly Load Resource

Balance shows a winter resource deficit of about -150 aMW as measured in December (P5), which

is significantly larger than any other month. By 2037 (represented by the black line), the winter

deficit is about -300 aMW. However, the August deficit in 2037 exceeds -300 aMW. From a

portfolio planning perspective, whatever long-term capacity resource(s) is eventually identified to

serve the PUD’s forecast capacity need, must also be available to serve both the winter and

summer peak periods.

74 The Load Resource balance shown in Figure 6-12 is based on a P5 metric, or the combination of high load and poor

hydro conditions.



The inference too is that the modeling of historical weather as static moving forward, done in the

past and reflected in many of the 2017 IRP scenarios (Low Growth, BAU, BAU w/CA Carbon and

High Growth), no longer reflects the same future portfolio needs as the one that anticipates

ongoing climatic changes. The Long Term Resource Strategy incorporates Climate Change

effects on the existing portfolio as it most accurately reflects the PUD’s future expectations.

Figure 6-12 Changes in the PUD’s Forecast Peak Need under Climate Change Scenario

(Monthly On Peak Load Resource Balance @P5)

The affect and impact of a carbon policy or price affected the PUD’s net portfolio cost NPV and

level of conservation acquired under the integrated portfolio approach used in the 2017 IRP. This

was most evident in the comparison between the BAU w/No Carbon Scenario and the BAU w/CA

Carbon scenarios. These scenarios are identical except for their differing carbon assumptions. The

introduction of carbon pricing as modeled, affected fuel costs and increased the forecast wholesale

market prices used by the Portfolio Optimization Model.

Figure 6-13 shows a comparison of the two scenarios in terms of their total December On Peak

hour capacity resource additions, average annual resource additions, and net portfolio costs. What

became evident was the trend in higher carbon environments, demonstrated by the BAU w/CA

(400)

(300)

(200)

(100)

0

100

200

Mo

nth

ly H

LH L

oad

Res

ou

rce

Bal

ance

@

P5

in a

MW

Snapshot of Monthly On-Peak Net Position

2018 2027 2037



Carbon scenario, was more conservation became cost effective, deferring and changing the amount

of supply side capacity resources.



Figure 6-13 Comparison of Portfolio Additions – Energy and Capacity

BAU w/No Carbon vs BAU w/CA Carbon

BAU w/No Carbon

BAU w/CA Carbon

20 Yr Conservation December On-Peak Cap (aMW)

107

228 Long-Term Capacity Adds -December On-Peak (aMW)

232

97

Total Capacity Adds- December On-Peak (aMW)

338

325

20 Yr Conservation Annual Energy Adds (aMW)

92

152

20 Yr Supply Side Annual Energy Adds (aMW)

24

11

Total Annual Energy Adds (aMW)

116

164

Total Net Portfolio Costs NPV ($000) $ 702,088 $ 481,360

This result is apparent in Figure 6-13 when comparing the demand-side (107aMW vs. 228aMW)

and supply-side additions (232aMW vs. 97aMW) for December On-Peak. The preferences for

demand and supply side resources to meet capacity needs are nearly inverse, and the only variable

driving these differences between BAU w/No Carbon (current carbon law for Washington state)

and BAU w/CA Carbon is the cost of carbon applied to fuel costs and market prices, or the

“market environment.” Because carbon policy assumptions drive such divergent preferences for

conservation acquisition and resulting portfolio builds, a balanced approach was developed to find

an optimal portfolio strategy across carbon policies.

The Long Term Resource Strategy reflects the Climate Change Scenario with a market

environment reflective of the Low Societal Cost of Carbon and low natural gas prices. This

planning environment was arrived at through an analysis for conservation and net portfolio cost

NPV across plausible carbon policies for the Climate Change scenario.

Portfolios were then developed for the Climate Change scenario at each of the conservation

bundles under a No Carbon (current law), Low Societal Cost of Carbon, and California Carbon

policy environments, and then ranked by cost effectiveness within that policy environment. The

PUD was seeking to find the optimal conservation bundle across the plausible carbon policies



given the absence of a current carbon policy in Washington state, understanding that previous

analysis had shown that in all cases, tested portfolios were balanced after conservation with an

optimal level of long term capacity resource that could meet future seasonal and peak loads.

After all policy environments were populated, the conservation bundles were ranked for their

resulting average portfolio cost across the carbon policies, shown in Figures 6-14. The conclusions

was that conservation measures up to $75/MWh performed the best across the potential carbon

policies, and performed the best in the mid-priced carbon policy, that of the Low Societal Cost of

Carbon. For this reason, the PUD considers an optimal portfolio built around conservation up to

$75/MWh to be both the optimal portfolio composition in a Low Societal Cost of Carbon

environment, providing the PUD with a balanced, lowest reasonable cost approach in the absence

of a known carbon policy

Figure 6-14

Ranked Evaluation of Least Cost Portfolios under Three Different Carbon Policy

Environments

Proposed Long-Term Resource Strategy

Based on the planning assumptions used in the 2017 IRP analysis, the Long Term Resource

Strategy shows the PUD has sufficient resources to meet its annual energy needs over the 20 year

study period, under average water conditions. The underlying carbon assumption is the EPA’s

Low Societal Cost of Carbon that ranges from $13.19 per ton of carbon dioxide emissions in 2018,

to $29.95 per ton in 2037, for an average 20 year carbon cost of ~$20.38 per ton.

New resource additions under the Long Term Resource Strategy includes the addition of 93 aMW

of new cumulative annual conservation by 2027, and 114 aMW by 2037, and a 50 MW short term

Conservation BundleClimate Change

(BAU Carbon)

Climate Change

(Low SCC Carbon)

Climate

Change (CA

Carbon)

Overall Rank

<$65/MWh 3 4 5 4

<$75/MWh 1 1 2 1

<$85/MWh 2 2 3 2

<$100/MWh 4 3 1 3

<$120/MWh 5 5 4 5



capacity contract to meet near term winter capacity needs for the 2018 through 2022 period. The

level of cumulative conservation is expected to provide 126 aMW of winter conservation by 2027,

and 152 aMW of winter conservation by 2037. This defers the need for a long-term capacity

resource until 2028. Annual EIA renewables requirements are satisfied through a combination

unbundled REC purchases and 15MW of customer owned distributed generation. Figure 6-15

illustrates the resource additions that satisfy the December On-Peak planning standard.

Figure 6-15

Long-Term Resource Strategy - December On-Peak (in aMW)

The incremental net portfolio cost for the Long Term Resource Strategy has a net present value of

$422 million as shown below in Figure 6-16, and assumes the PUD will meet its EIA renewables

requirements with its existing resources through 2020, and contract for the incremental acquisition

of approximately ~68 aMW unbundled RECs across the study period.

2018 2020 2022 2024 2026 2028 2030 2032 2034 2036

Forecast Market Purchase 96 80 57 97 87 - - 15 23 36

Distributed Generation - 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4

Short-Term Capacity Contract 50 50 50 - - - - - - -

Long-Term Capacity Resource - - - - - 116 116 116 116 116

Conservation 12 36 61 86 112 129 133 141 146 150

Existing Resources 869 895 908 915 918 908 905 898 899 904

Load 1,027 1,061 1,076 1,098 1,118 1,130 1,150 1,170 1,184 1,207

0

200

400

600

800

1,000

1,200

1,400

aMW



Figure 6-16 Long-Term Resource Strategy Portfolio Costs, Benefits and Net Present Value

Cost for New Conservation $ 147,987,664

Cost of New Capacity Additions $ 259,431,989

Cost of New DG Additions $ 34,781,279

Cost of Procured RECs $ 453,201,140

Total Portfolio Costs $ 895,402,072

Less Portfolio Revenues $ 473,243,503

Net Portfolio Cost NPV $ 422,158,569

The incremental costs attributed to REC acquisitions were modeled for the Long Term Strategy at

the 2017 IRP’s assumed fundamental REC Price, and also tested at $20/REC, $15/REC, and

$10/REC.75 If REC prices had been modeled closer to today’s near term compliance REC prices

across the 20 year study period, the Long Term Resource Strategy would still have selected the

same amount of conservation and supply side resource additions, and had no change in timing of

the acquisitions – but overall, portfolio costs would have been lower.

Figure 6-17 shows the portfolio costs and benefits for a range of RECs prices, including the

fundamental REC price developed as a conservative estimate in the 2017 IRP analysis. Resource

costs and revenues remain the same, with costs changing only for RECs. Net Portfolio Cost NPVs

range from of high of $422 million to a low of $62 million based on a modeled price of $10/REC.

75 See Section 5 – Analytical Framework, for additional discussion on how unbundled RECs were modeled.



Figure 6-17

Portfolio Costs with a Range of Compliance REC Prices

Fundamental REC Price* $15/REC $10/REC

Cost for New Conservation $ 147,987,664 $ 147,987,664 $ 147,987,664

Cost of New Capacity Additions $ 259,431,989 $ 259,431,989 $ 259,431,989

Cost of New DG Additions $ 34,781,279 $ 34,781,279 $ 34,781,279

Est'd Cost of RECs $ 453,201,140 $ 139,842,011 $ 93,228,007

Total Portfolio Costs $ 895,402,072 $ 582,042,943 $ 535,428,939

Less Portfolio Revenues $ 473,243,503 $ 473,243,503 $ 473,243,503

Net Portfolio Cost (NPV) $ 422,158,569 $ 108,799,440 $ 62,185,437

The Long Term Resource Strategy does not commit the PUD to any long term resource decisions

that could prove uneconomic if carbon policy changes significantly; instead, the strategy invests in

a conservation level that works well across carbon policies, invests in a term-limited, short-term

capacity contract to provide for near-term capacity needs, and anticipates investments in

unbundled RECs as part of a strategy to cost-effectively comply with RPS standards. The time

before a long-term resource decision needs to be made also means the Long Term Resource

Strategy can be revaluated in the next IRP, when there is a possibility a defined carbon policy can

help the PUD to focus its Long Term Resource Strategy choice to optimize a specific, structured

market environment.


7 KEY INSIGHTS AND PROPOSED ACTION PLAN

The PUD expects its historical achievements and continued acquisition of conservation, combined

with the adoption of federal energy codes and standards, will temper future load growth. The

result is no significant increase in future annual energy needs across the 20 year planning horizon

for all but the High Load Growth scenario. That said, the PUD’s future could unfold differently.

The 2017 IRP analysis also studied the forecast capacity need beyond its existing resources and

after the acquisition of conservation, to meet periods of peak customer demand. Depending on the

scenario, the identified capacity need changes across time due to several factors:

1. Limitations inherent in the existing portfolio; several owned and contracted resources are

limited in their ability to be dispatched up and down, hour to hour, and within an hour.

2. Established planning standards limit the volume of on-peak market purchases the PUD is

willing to make on a planning basis from the wholesale energy market, particularly during peak

demand periods. These limitations are intended to reduce exposure, on a planning basis, to

market and price volatility and delivery risk; and

3. Modeling assumptions for the post-2028 BPA power contract are limited by the nature of what

future products the agency may offer.

Conducting an integrated resource planning process every two years yields insights for the utility

about its existing portfolio, planning assumptions used, and the alternate portfolios evaluated,

regardless of which future unfolds. This is an added benefit of the IRP process. The following

represents some of the key insights from the 2017 IRP analysis:

Section 7: Key Insights and Proposed Action Plan


Key Insights

1. Conservation is the PUD’s Resource of Choice

New conservation is the single largest portfolio addition in every scenario evaluated in the 2017

IRP. The portfolio analyses again confirmed that new conservation provides significant benefit to

the PUD – it defers the need to acquire new supply-side resources, reduces the overall cost of

power and annual EIA renewables compliance, and defers the need for additional transmission and

distribution capacity.

After the acquisition of new cumulative conservation, the PUD forecasts that it will have no annual

energy need over the 2018 through 2037 study period under all of the scenarios, except in the High

Load Growth case, and long term resource additions are deferred until the mid-2020s.76 The

addition of new conservation eliminate the need for baseload or must-run energy resources in the

Low Growth, BAU, and Climate Change with Low Carbon scenarios shown in Figure 7-1.

Figure 7-1

Annual Load Resource Balance after New Conservation by Scenario – in aMW

76 The PUD has no expected annual energy need after conservation except in the High Growth case unless: the region

experiences poor hydrological conditions, operations changes occur that affect the Federal System and the PUD’s BPA

power contract, or BPA’s post 2028 product offerings are different than today.

-

200

400

600

800

1,000

1,200

Existing Resources @P50 BAU w/ CA Carbon BAU w/ No Carbon

Low Growth High Growth Climate Change



The new cumulative conservation potential identified for each scenario is shown in Figure 7-2.

Figure 7-2

Comparison of New Cumulative Conservation by Scenario

2017 IRP Scenario

10-Year Cumulative

Economic Achievable

Potential (annual)

20-Year Cumulative Economic Achievable

Potential (annual)

BAU w/No Carbon 73 aMW 92 aMW

Climate Change w/Low Societal Cost of Carbon 93 aMW 114 aMW

Low Growth w/Low Societal Cost of Carbon 97 aMW 121 aMW

BAU w/CA Carbon in 2022 114 aMW 152 aMW

High Growth w/Mid-High Societal Cost of Carbon 114 aMW 152 aMW

2. Future Need is for Capacity Resources

The Monthly December On Peak and Peak Week winter planning standards were the drivers for

capacity additions to augment the PUD’s existing portfolio and match seasonal and peak demand,

after conservation. The granularity explored in the 2017 IRP analysis helped to identify the nature

of the future capacity need – whether for on-peak hours during a given month, or for a sustained

period or ‘peak week’ within a month. In nearly every case, the portfolios selected short-term

and/or long-term capacity resources, after conservation, to satisfy peak demand periods.

Figures 7-3 below shows the PUD’s December On-Peak net position after new cumulative

conservation, with the PUD’s existing and committed resources, for the Long Term Resource

Strategy.77 The December On-Peak deficitranges from 146 aMW in 2018 to 153 aMW in 2036.

After applying the Monthly On-Peak planning standard that limits reliance on the short term

energy market on a planning basis to no more than 100 aMW, the near term deficit ranges from 46

to 7 for 2018 to 2022, and in 2030 at 12 to 53 aMW in 2036.

77 The 2017 Conservation Potential Assessment conducted by The CADMUS Group provided conservation savings at

the hourly level across a calendar year. This enabled the analysis to identify the amount of economic achievable

potential available by month. The months of December and August therefore have different cumulative conservation

totals.



Figure 7-3

December Monthly On-Peak Net Position for Long Term Resource Strategy

after New Conservation (in aMW)

DECEMBER 2018 2020 2022 2024 2026 2028 2030 2032 2034 2036

Load Resource

Balance (LRB) with

Existing Resources

-158 -166 -168 -183 -200 -222 -245 -272 -286 -303

New Cumulative

Conservation 12 36 61 86 112 129 133 141 146 150

LRB After New

Cum. Conservation -146 -130 -107 -97 -88 -93 -112 -131 -140 -153

Allowance for

Monthly Planning

Standard

100 100 100 100 100 100 100 100 100 100

Forecast Need

(-) Deficit -46 -30 -7 0 0 0 -12 -31 -40 -53

Figure 7-4 shows that the August On-Peak deficit, after new cumulative conservation, begins at 20

aMW in 2030 and increases to 86 aMW in 2036.

Figure 7-4

August Monthly On-Peak Net Position for Long Term Resource Strategy

after New Conservation (in aMW)

AUGUST 2018 2020 2022 2024 2026 2028 2030 2032 2034 2036

Load Resource

Balance (LRB) with

Existing Resources -26 -34 -58 -88 -116 -167 -224 -252 -283 -305

New Cumulative

Conservation 9 28 47 67 87 100 105 111 114 119

LRB After New

Cum. Conservation -17 -6 -11 -21 -28 -68 -120 -141 -168 -186

Allowance for

Monthly Planning

Standard

100 100 100 100 100 100 100 100 100 100

Forecast Need

(-) Deficit 0 0 0 0 0 0 -20 -41 -68 -86

The forecast Peak Week deficit occurs in the month of December, with an on-peak deficit for the

week ranging from 234 aMW in 2018 to 268 aMW in 2036 (Figure 7-5). After applying the

Monthly Peak Week planning standard that limits reliance on the short term energy market on a

planning basis to no more than 200 aMW, the deficit shows a near term need for the 2018-2020

period of 12 to 34 aMW, and 20 to 68 aMW for the 2030 through 2036 period.



Figure 7-5

Peak Week Net Position for December On-Peak

Long Term Resource Strategy after New Conservation (in aMW)

DECEMBER 2018 2020 2022 2024 2026 2028 2030 2032 2034 2036

Load Resource Balance (LRB) with Existing Resources

-246 -247 -249 -272 -292 -323 -357 -376 -388 -411

Peak Week Cum. Conservation 11 34 58 82 107 123 127 141 139 143

LRB After

Conservation -234 -212 -191 -189 -185 -200 -230 -235 -248 -268

Allowance for Peak Week Planning

Standard 200 200 200 200 200 200 200 200 200 200

Forecast Need

(-) Deficit -34 -12 0 0 0 0 -20 -35 -48 -68

3. Capacity Need Drives Type of Resource Additions

The resources selected in development of the candidate portfolios were those that best matched the

PUD’s forecast need using least cost criterion.78 To address the forecast capacity need, supply

side resources with the highest availability that could be used to match periods of seasonal or peak

demand were selected by the portfolio model over baseload or other must run resources. Capacity

resources or contracts for flexible resources that can provide capacity, offer a distinct set of

operating characteristics to the portfolio – they can be turned on and off, adjusted, or ‘dispatched’

by the grid operator or generator owner, when system needs or load conditions dictate.79

Figure 7-6 shows the new resource additions for the candidate portfolios by scenario. Each

portfolio selected some level of conservation determined to be cost effective in combination with

the best mix of a short term capacity contract and/or long term capacity resources. The exception

is the Low Growth scenario, where new cumulative conservation is forecast to meet the forecast

capacity need after 2022.

78 Section 5 describes the analytical framework and four planning standards used to develop the candidate portfolios. 79 Long term capacity resource additions after conservation are deferred until the mid to late 2020’s, depending on the

scenario. Long term capacity resources were modeled (price, cost and availability) as fueled by fossil fuels due to lack

of a Northwest capacity market. Section 5 details supply side resource options and assumptions used to model long

term capacity resources in the 2017 IRP.



Figure 7-6

Summary of New Resource Additions by Scenario in aMW

Scenario

New Cumulative

Conservation

(Annual aMW)

Short Term

Capacity

Contract

(Dec HLH

aMW)

Long Term

Capacity

Resource

(Dec HLH

aMW)

Renewable

Resources

(Annual

aMW)

BAU w/No Carbon 92 25 232 0

Climate Change w/Low Carbon 114 50 116 3

Low Growth w/Low Carbon 121 25 0 0

BAU w/CA Carbon 152 0 97 1

High Growth w/Mid-High Carbon 152 0 396 68

While the analysis in this IRP demonstrates that the PUD has no need to acquire long term supply-

side resources the next five years, it is important to note that in the unlikely event conditions

change, the 2017 IRP has identified and evaluated numerous commercially available and reliable

resources. Should a resource need become imminent, the PUD would perform a due diligence

review and conduct a comprehensive economic and environmental assessment for a specific

resource option.

Candidate portfolios were analyzed to determine impacts associated with meeting the PUD’s

forecast capacity needs exclusively with renewable resources like wind, solar, landfill gas and

biomass, geothermal, pumped hydro storage and battery storage. While not 100% conclusive, and

because the cost and benefits related to using re-powered wind resources was not evaluated in the

2017 IRP analysis, indications were that net portfolio costs tended to be higher when traditional

capacity resources were excluded from the set of commercially available supply side resource

options. Results of this sensitivity analysis are shown in Figure 7-7.



Figure 7-7

Comparison of Summary of New Resource Additions by Scenario in aMW

Scenario

Net Portfolio Costs

All Available Supply

Side Resource Options

(NPV in $ Millions)

Net Portfolio Costs

Renewable Only

Resources80

(NPV in $ Millions)

%

Difference

in NPV

Low Growth w/Low Carbon $ 317 $ 317 0%

BAU w/No Carbon $ 702 $ 891 27%

Climate Change w/Low Carbon $ 422 $ 489 16%

BAU w/CA Carbon $ 481 $ 653 36%

High Growth w/Mid-High Carbon $1,369 $2,433 78%

Based on the planning and modeling assumptions in the 2017 IRP analysis, the addition of

baseload or must run renewable resources, when the PUD is forecast to have an annual energy

surplus (except in the High Growth case), was not a cost competitive solution for meeting the

PUD’s forecast winter and summer capacity needs. The range of increased portfolio costs varied

from zero to 78 percent, depending on the scenario.81 The Low Growth case notes no difference in

net portfolio costs since under this scenario, since the forecast capacity need after 2022 is assumed

to be met solely by new cumulative conservation, with no other post 2022 resource additions.82

It is first important to understand the differences in the cost and availability for supply-side

resources and cost in evaluating how to best meet the PUD’s forecast future capacity need at the

lowest reasonable cost. Figure 7-8 below lists this comparison for a select group of resources to

meet the PUD’s peak demand, 19 out of 20 times:83

80 Renewable resources for the purpose of this analysis include new wind, utility scale solar, landfill gas and biomass,

traditional geothermal, pumped hydro storage and battery storage. 81 With exception of a 25 MW short term capacity contract 2018 through 2022, new conservation met the forecast

capacity needs in the Low Load Growth portfolio. 82 The Low Growth scenario adds a near term 25 MW capacity product for the 2018 through 2022 period, until

sufficient new conservation has accrued and can satisfy the forecast post 2022 capacity need. 83 Metric definitions used in Figure 7-8: Peak Week Credit expressed in MW is an availability metric to reflect the

capacity a generating resource can contribute to meeting the PUD’s Peak Week need. Reflects an adjustment to the

generator’s nameplate to account for the amount of time resource not available, due to lack of fuel or forced outage, to

generate during the 80 on-peak hours during the peak week, Monday through Friday, in the month of December.

The Levelized Cost per megawatt-year metric represents the levelized or annualized total cost (capital and operating)

of the resource, divided by the capacity provided by the project over its useful life.



Figure 7-8

Select Resource Comparison – Availability and Cost for Peak

Resource Fuel Nameplate Peak Week

Credit (availability)

Levelized Cost per

MW-Year

Nameplate to meet 100 MW

Peak Week Need

Simple Cycle Comb. Turbine Fossil 239 MW 232 MW $ 104,205 103 MW

Dual Fuel Recip Engine Fossil 50 MW 48.5 MW $ 159,205 103 MW

Pumped Hydro Storage Hydro

and Mkt Purchases

100 MW 97 MW $ 299,567 103 MW

Landfill Gas LF Gas 10 MW 9.7 MW $ 545,327 103 MW

New Wind, Northwest Wind 50 MW 1.7 MW $ 6,685,044 2,941 MW

New Wind, Montana Wind 50 MW 11.6 MW $ 881,933 431 MW

West. WA Utility-Scale Solar Solar 5 MW .03 MW $39,824,048 16,667 MW

The conclusion of the 2017 IRP analysis as modeled, is that the addition of new baseload or

renewable resources like landfill gas, solar, biomass or wind to meet forecast peak capacity needs,

is forecast to be more expensive than meeting the same need with a long term capacity resource.

Long term capacity resources were modeled at the price, cost and availability of gas-fired

generating resources like a simple cycle combustion turbine or reciprocating engine. This finding

is consistent with the Council’s Seventh Power Plan84 and the recent “Pacific Northwest Low

Carbon Scenario Analysis,” published by Energy+ Environmental Economics.85

The PUD will continue to investigate the types of resources that would be available and economic

to meet forecast future capacity needs, and will re-assess the cost of available resource options in

future IRPs, as markets, technology and demand response continues to evolve.

84 Seventh Power Plan, “Executive Summary,” Figure 1-1, page 1-2, published February 2016. Conclusion based on

800 futures tested by the Council’s Resource Portfolio Model, found at

https://www.nwcouncil.org/media/7149940/7thplanfinal_allchapters.pdf 85 “Pacific Northwest Low Carbon Scenario Analysis,” published December 2017 by consulting firm E3, found at

http://www.publicgeneratingpool.com/wp-content/uploads/2017/12/E3_PGP_GHGReductionStudy_2017-12-

15_FINAL.pdf

https://www.nwcouncil.org/media/7149940/7thplanfinal_allchapters.pdf





4. RECs to meet Annual Compliance Requirements

The PUD expects to continue meeting its annual renewables compliance requirement through a

combination of the RECs from its contracted renewable resources, its incremental hydro, and with

RECs allocated by BPA through the long-term power supply contract. The PUD’s existing

portfolio meets the 15% requirement in 2020. Because the IRP analysis shows the PUD expects to

be in a surplus annual average energy position through 2037, after new conservation, the lowest

cost approach to satisfying the EIA was the addition of unbundled RECs (except in the High

Growth scenario). Figure 7-9 summarizes the resources and REC additions by scenario.

Figure 7-9

Renewables and REC Additions (after Conservation) by Scenario (in aMW)

Scenario Renewables

Annual aMW

RECs

Annual aMW

Low Growth w/Low Carbon 0 68

BAU w/No Carbon 0 78

Climate Change w/Low Carbon 3 68

BAU w/CA Carbon 1 72

High Growth w/Mid-High Carbon 68 22

The forecast cost to comply with the annual EIA requirement with supplemental REC purchases is

notable for the PUD, and will depend on the price per REC across time. The IRP analysis

evaluated a variety of REC prices for the Long Term Resource Strategy, detailed in Figure 7-10.



Figure 7-10

Comparison of Forecast REC Costs as Percent of Total Portfolio Costs

5. Scale and Timing of Carbon Costs

The scale and the timing of future carbon costs as modeled in the 2017 IRP analysis affected the:

Forecast fuel costs and wholesale market prices developed for each scenario;

Amount of new conservation acquired, directly affecting the timing and quantity of supply side

resource additions; and

Revenues derived from the sale of any portfolio energy surpluses created over the study period.

Overall, the amount of cost effective conservation and net portfolio costs were sensitive to the

level of carbon costs evaluated in the 2017 IRP analysis. As modeled, the higher the carbon cost

in dollars per ton, the higher the level of conservation acquired, and the smaller and later in the

study period were supply side resources added additions. Higher carbon costs that affected fuel

costs and market prices also affected portfolio revenues created from the sale of portfolio energy

surpluses (example, High Growth scenario, Figure 7-11). Satisfying the Monthly On Peak and

Peak Week planning standards and achieve lowest reasonable cost portfolios did not vary the mix

of supply side resource additions; rather the supply side resource additions consistently were those

that provided capacity, not energy, and augmented the PUD’s existing portfolio.

Figure 7-11 details the portfolio revenues created from any energy surpluses as the portfolio was

modeled. Revenues offset portfolio cost NPVs to arrive at the net portfolio cost NPV.

51%

39%

30% 24%17%

0%

10%

20%

30%

40%

50%

60%

$-

$100,000,000

$200,000,000

$300,000,000

$400,000,000

$500,000,000

$600,000,000

$700,000,000

$800,000,000

$900,000,000

$1,000,000,000

FundamentalREC price

$30/REC $20/REC $15/REC $10/REC

Tota

l Po

rtfo

lio C

ost

($

NP

V)

Forecast REC Costs Total Portfolio Cost Percent of Total Cost



Figure 7-11

Comparison of Portfolio Revenues and Net Portfolio Cost by Scenario

6. Climate Change Effects

The potential effects of climate change on the future of the PUD’s power supply portfolio –

comprised of over 85% hydroelectric generation – create uncertainty and risk. The 2017 IRP

initially assessed climate change as a sensitivity to the Business as Usual case, and later as its own

scenario.

PUD staff’s research and analysis focused on the impacts and effects of climate change on the

existing power supply fueled by hydro and on customer load patterns. Staff used current science

from the University of Washington Climatology Impact Group’s studies, scaled down and

consistent with the regionalized effects consistent with the Representative Concentration Pathway

(RCP) 4.5, characterized as “low climate change” in the 5th Assessment Report, produced by the

United Nations Intergovernmental Panel on Climate Change. 86 These sources informed weather

and hydrology adjustments for the 20 year IRP study period.

86 The scale of the climate change impacts modeled in the 2017 IRP were consistent with the regionalized effects of

Representative Concentration Pathway (RCP) 4.5, characterized as “low climate change” in the 5th Assessment

Report, UN Intergovernmental Panel on Climate Change, at https://ipcc.ch/report/ar5/.

$317

$702

$481$422

$1,369

0

200

400

600

800

1,000

1,200

1,400

1,600

0

500

1,000

1,500

2,000

2,500

Low Case Business As Usual Business As Usualw/ CA Carbon

Climate Changew/ Low SCC

High Case Net

Po

rtfo

lio C

ost

s (N

PV

in $

Mill

ion

s)

Po

rtfo

lio R

even

ues

(in

$M

illio

ns)

Portfolio Revenues Net Portfolio Costs (NPV)

https://ipcc.ch/report/ar5/



This analysis found that effects attributable to climate change altered the PUD’s future load

patterns and the timing of the hydroelectric generation produced by both PUD-owned resources

and the Federal System that the PUD contracts for the output through its long term BPA power

contract. In general, regional temperatures are expected to gradually warm and customer demand

is forecast to decline in winter periods due to milder temperatures, while warmer summer

temperatures are expected to increase demand as a result of increasing air conditioning load.

Shifts over time due to warmer temperatures are forecast to increase regional precipitation and

alter annual snowpack levels historically seen build in the Cascade Mountains. Under the climate

change analysis, winter hydro production increase by 12% due to increased stream flows during

the winter periods when more precipitation falls as rain at higher elevations. This increased

hydroelectric production combines with milder temperatures to reduce the PUD’s forecast peak

need during the winter months. Compared to the BAU w/No Carbon scenario, the Climate Change

scenario reduced forecast load growth by 1%, reduced future winter peak load by 8% percent, and

resulted in a $112 million net portfolio cost savings NPV.

The change in and gradual decline of snowpack builds in the Cascade Mountains is expected to

decrease the natural storage of water between the winter and the spring and summer periods. This

results in declining stream flows affecting the month of August with a 17% decline in average

hydroelectric production by 2037. This change in summer hydro production under the Climate

Change scenario is forecast to create an emerging summer capacity need during on-peak hours in

the month of August by the end of the IRP study period, after cumulative conservation.

As a result of climate change effects, the PUD’s future need is different from the other scenarios,

and is one of the reasons the Long Term Resource Strategy was informed by this scenario. In

general, under the Climate Change scenario, the PUD has a:

Reduced annual energy need;

Reduced winter need (less load and more hydro); and an

Emerging summer need (more load and less hydro).



The PUD is not the first utility to consider the implications and effects of climate change on its

load resource balance during the long term planning process. Several other regional utilities,

including BPA, have or are actively assessing the potential range of impacts associated with

climate change.



Progress on 2015 IRP Action Items

The following updates progress on the Action Items from the 2015 Update to the 2013 IRP:

1. Pursue all cost-effective conservation measures. Continue to monitor demand-side technologies and pursue where applicable.

The PUD achieved its two-year target of 14.04 aMW for CY 2016 and CY 2017.

2. Continue the District’s Demand Response initiative in accordance with the work plan.

Three PUD customers participated in BPA’s 2016 DR pilot program; the PUD launched its Pay for Performance pilot in 2016. Efforts are underway to finalize the PUD’s DR strategy.

3. Continue to evaluate the benefits of energy storage and portfolio optimization through the Modular Energy Storage Architecture (MESA) projects.

The PUD has brought online and is continuing to manage two battery storage projects operated using the MESA framework – 2 MW of storage at the Hardeson substation and a 2.2 MW project at our Everett substation. These projects continue to provide valuable information about managing energy storage and its relative portfolio benefit.

4. Participate in regional forums and assess impacts associated with climate, carbon and renewable portfolio standard legislation and any associated rulemaking processes.

PUD staff has remained engaged in regional processes addressing climate issues, and has been actively involved in helping to craft least-cost carbon reduction policy.

5. Conduct a situational scan of best practices to address “portfolio resiliency” that goes beyond typical IRP scenario modeling.

The 2017 IRP includes a series of sensitivities that tested the portfolio’s resiliency to carbon costs and climate change, which were identified as two notable uncertainties.

6. Expand the utility’s resource modeling and assessment capabilities to better assess risk and uncertainty.

The 2017 IRP incorporated probabilistic modeling of the PUD’s existing and committed resources and load forecast by month and by peak week. This granularity helped identify resource need within year and within month.

7. Conduct an analysis to assess the impacts and/or benefits of to the District of switching from the existing BPA Slice/Block product to a BPA Load Following product for the 2020-2028.

After performing this analysis and presenting the findings to the Board of Commissioners in March 2016, the Board elected to remain with the BPA Block/Slice product through 2028.

8. Participate in regional discussions concerning a Northwest Power Pool imbalance market.

The PUD was an active participant in the NWPP MC Initiative process from 2012 to 2016, when the effort concluded.

9. Explore procurement of the Winter Capacity Product to address winter period deficits. Evaluate pumped hydro potential in PUD service area and the Western Interconnect.

The PUD issued an RFP to procure 50 MW of a winter capacity product from NW power markets for the winter 2016-2019 period. Evaluation of potentially viable Pumped Hydro Storage resources is ongoing.



Proposed 2017 Action Plan

The 2017 Integrated Resource Plan has identified several near term actions to ensure the PUD can

meet the needs of its customers in a rapidly changing environment, well into the future:

1. Pursue all cost effective conservation and further explore peak benefits, including the

feasibility of demand response as a future utility-scale capacity resource for the PUD.

Conservation is the single largest portfolio addition for every scenario evaluated in the 2017

IRP. It remains the PUD’s resource of choice for meeting future load growth. The acquisition

of conservation savings reduces the demand for electricity, delaying the need to acquire or

develop new resources, which can reduce the overall cost of energy and capacity, including

deferral for additional transmission and distribution capacity upgrades.

The PUD has been a regional leader in its acquisition of conservation for over 35 years. It has

successfully developed and operated numerous cost-effective programs that help customers of

all types use energy more efficiently. Staff forecasts the PUD will acquire 114 aMW of new

cumulative annual energy savings and 152 aMW of new winter on-peak energy savings over

the 2018 through 2037 period. The 10 year conservation potential for the Long Term Resource

Strategy (Climate Change) was identified at 92.7 aMW.

To attain this level of conservation achievement, PUD staff will continue to develop strategies

and programs that reach all sectors, with special focus on implementation strategies for

conservation that brings capacity contributions. Accelerated customer adoption of new

technologies and other energy conservation practices based on sound business principles will

continue.

2. Explore low cost, low emissions alternatives in the Northwest for capacity resources to

meet peak needs, including the ongoing evaluation of battery and pumped hydro storage,

and discussions with BPA for peaking and capacity products.

Under the Long Term Resource Strategy (Climate Change), the 2017 IRP analysis forecasts a

future capacity need that changes across time. With the addition of 114 aMW of new



cumulative conservation that defers the need for a long term capacity resource solution until

the 2028,the decision on how this capacity need will be met is not imminent. The PUD has

adequate time to explore the cost and availability of low emissions capacity alternatives in the

Northwest that are well suited environmentally and economically to meet the PUD’s forecast

future needs. The identified need for resources to meet periods of forecast peak customer

demand is not new – both the adopted 2013 and 2015 IRP’s identified this after conservation.

In the 2015 Update, this need was satisfied in part by conservation, and through the addition of

a short term capacity product.87 By end of calendar year 2018, staff will have conducted a

market survey including, but not limited to, requests for information or requests for proposals

as to the quantity, cost and availability of capacity alternatives in the region.

3. Ensure customer owned and distributed renewables programs are complementary to the

PUD’s overall power supply strategy.

Serving the PUD’s customers to provide outstanding value relative to their cost, and delivering

now and for the future, are two key PUD strategic priorities. This action item reflects efforts by

staff, in consultation with a broader stakeholder group, to update the PUD’s solar, electric vehicle

and demand response programs. Staff will continue to identify and assess opportunities to engage

in pilot and demonstration programs that test acquisition strategies, incentives, costs and customer

reactions to such programs. This may include standardization and mechanisms associated with

customer owned distributed generating resources, for the purpose of developing a more

coordinated, comprehensive and complementary approach for the customer in conjunction with the

PUD’s overall portfolio strategy.

4. Develop a least-cost compliance strategy for meeting the state’s renewables requirements

under the Washington Energy Independence Act (EIA or Initiative 937).

In light of the PUD’s current forecast of being in a surplus annual average energy position after

conservation through 2037, the lowest cost compliance approach at this time is to satisfy the

87 2015 Update to the 2013 IRP, Section 3.3 Supply-Side Resource Options, page 26 located at

https://www.snopud.com/PowerSupply/irp.ashx?p=1161

https://www.snopud.com/PowerSupply/irp.ashx?p=1161



PUD’s 15% renewable target post 2020 by acquiring unbundled RECs. Staff will develop its

renewables compliance strategy and update it every two years as part of the IRP and budget

planning process. This strategy does not preclude the PUD from pursuing other EIA

compliance methods or procuring additional renewable resources.

5. Enhance short and long-term resource portfolio modeling capabilities, expand cost and

risk tradeoff analyses.

During 2016 and 2017, staff has incorporated probabilistic modeling of its load resource

balance to complement its short term hedging program and inform its long term resource

portfolio modeling by scenario. At this time, managing portfolio risk has occurred through

establishing planning standards. As part of a continual improvement process review, staff will

evaluate how the in-house portfolio model could be enhanced to streamline cost/risk tradeoffs

across the long term planning horizon as part of the next IRP update.

6. Conduct an internal survey about the IRP to determine how the reference document is

used; validate key findings and incorporate into District’s next IRP process.

Long term resource planning is a significant undertaking and investment for a utility. The

analysis informs the utility of any future resource need and how that need could be met with a

mix of demand and supply side resources using least cost criterion. The final IRP document

describes the range of futures considered, identifies the resource need, and selects a portfolio or

long term resource strategy that describes how the PUD anticipates it will meet future needs,

until the next planning cycle. This reference documents facilitates alignment for utility

employees at all levels of the organization and serves as a reference document that takes a long

term view in how the PUD forecasts it will prudently meet future customer needs. This action

item envisions conducting an employee survey to provide feedback for the IRP technical and

steering teams to learn how the IRP document is used and referenced, and what areas are most

helpful or have been overlooked that will grow employees general understanding of the PUD’s

mission and the Delivering Now and for the Future strategic priority. Survey timing is

estimated to be mid-2018.



7. Re-assess the methodology used to determine the value associated with the deferral of

PUD distribution and transmission investments; monitor the Northwest Power &

Conservation Council’s regional review.

The cost of distribution and transmission investments by a utility is a significant investment.

Deferral of such investments by the incremental addition of capacity provided from other

alternatives sources can provide value to the utility. The dollar value used to represent the

benefit to the utility for deferring distribution and transmission investments is used in

developing the avoided cost for conservation. It is also a key element in providing the

appropriate price signal for the value of the capacity an alternate resource brings to bear. The

PUD’s methodology used to develop its deferred T&D costs has not been reviewed in some

time. A methodological review was also identified by the Council as an action item in the

Seventh Power Plan. Staff will conduct a review of the PUD’s existing methodology,

informed by regional discussions, to arrive at a final methodology that can be incorporated into

the 2019 conservation potential assessment and IRP effort.

8. Continue to participate in regional forums and assess impacts associated with climate

change, reduction in greenhouse gas emissions, renewable portfolio standards, and

regional power and transmission planning efforts.

Executive leadership and staff will continue to monitor and actively participate in regional

forums, legislative policy discussions and rulemaking initiatives, and BPA power and

transmission planning initiatives in support of Board policies and the PUD’s Mission and

Strategies Priorities.

snohomish county pud no. 1the timing and scale of carbon costs in each scenario affected the...

Documents