Alan F. Hamlet, Nathalie Voisin,

Dennis P. Lettenmaier, David W. Pierce, Tim Barnett

Center for Science in the Earth System Climate Impacts Group

and Department of Civil and Environmental EngineeringUniversity of Washington

Scripps Institute of OceanographyCalifornia Energy Commission

April, 2005

The Role of Climate Forecasts in Western U.S. Power Planning

Background

150000

200000

250000

300000

350000

400000

450000

190

0

191

0

192

0

193

0

194

0

195

0

196

0

197

0

198

0

199

0

200

0

Ap

r-S

ept F

low

(cfs

)

Effects of the PDO and ENSO on Columbia River Summer Streamflows

Cool CoolWarm Warm

PDO

Red=warm ENSO Green=ENSO neutral Blue=cool ENSO

Value of Long-Range Streamflow Forecasts for PNW Hydro Marketing

~ $150 million/yr

This value is created in part by marketing additional energy in late summer (when energy is more valuable) in expected wet years.

Hamlet, A.F., Huppert, D., Lettenmaier, D.P., 2002, Economic Value of Long-Lead Streamflow Forecasts for Columbia River Hydropower, ASCE J. of Water Res. Planning and Mgmt, 128 (2), pp 91-101

http://www.abb.com/global/abbzh/abbzh251.nsf!OpenDatabase&db=/global/gad/gad02181.nsf&v=17EA&e=us&m=100A&c=C1256D71001E0037C1256B8000371E41

~8000 MW IntertieCapacity

In normal or above normal water years in the PNW, net transfers are typically from PNW to CA

In below normal water years in the PNW, transfers from CA to PNW can occur in winter and transfers from PNW to CA in spring and summer may be

reduced or suspended.

Research Questions:

How do hydropower resources in the PNW and CA vary with climate and how predictable are these variations?

How does electrical demand vary with climate in the PNW and CA and how predictable are these variations? Are extreme loads predictable?

How do supply and demand co-vary in time? How predictable are energy surpluses and deficits in each region?

How can we best use this information to maximize the benefits of the PNW - CA electrical intertie?

Tools and Methods

Snow Model

Schematic of VIC Hydrologic Model and Energy Balance Snow Model

PNW

CACRB

GB

Evaluation of VIC Streamflow Simulations

Schematic Overview of CVMod and ColSim Reservoir Models

CVMod

ColSim

IrrigationDominates

Flood ControlWinter HydroDominate

Evaluation of Daily Time Step Energy Demand Model

Demand vs Tmax

Model Validation

Daily Peak Demand Models

PNW(Nov-March)

CA(May-Sept)

Met Data1915-2003

VIC

Overview of Simulation and Analysis

CVMod

ColSim PNW Hydro

SacramentoSan Joaquin Hydro

StatisticalDemand Models

PNW Demand

CA Demand

Population Weighted

TemperatureTime Series

Urban PopulationData

Observed DemandData

Results

Relationship Between Annual Discharge at The Dalles and Annual Hydropower Production in the PNW

-3.000

-2.000

-1.000

0.000

1.000

2.000

3.000

191719221927193219371942194719521957196219671972197719821987199219972002

PNW*

CA*

Covariation of Normalized PNW and CA Hydropower Production

California hydropower is less than 10% of total CA demand, but may play a significant role in energy trading on short time scales. Transferable energy from the PNW is also about 10%.

R2 =0.541

Predictability of Seasonal Demand in the PNW and CA

•Winter demand in the PNW is predictable with long lead times via ENSO forecasts:

Warm ENSO = lower winter electrical demandCool ENSO = higher winter electrical load

•Load is in phase with water availability.

•Summer demand in CA has been recently demonstrated to be predictable with several months lead time using the NPO (PDO) index in spring (Alfaro et al. 2005). •

B.Normal NPO (MAM) = B. Normal CDD in S. CA. (JJA)A.Normal NPO (MAM) = A. Normal CDD in S. CA. (JJA)

•Load is out of phase with PNW water availability.

30000

30500

31000

31500

32000

32500

33000

-3 -2 -1 0 1 2 3 4

Nino3.4 Anomaly (C)

Ave

rag

e N

DJF

Pea

k E

lect

rica

l Lo

ad (

MW

)Nino3.4 Anomaly vs Avg NDJF Peak Electrical Demand

MAM-PDO IndexExplains about 20% of the Variance in monthly average TMAX in CA.

y = 189.49x + 36876

R2 = 0.0014

20000

25000

30000

35000

40000

45000

50000

55000

60000

-2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5

MAM avg PDO

CA

pk

dem

Relationship Between MAM avg PDO and CA Daily Peak LoadApproximate Threshold of Capacity Problems

May-June-July Hydropower available for Transfer

0

1

2

3

4

5

6

7

19

17

19

22

19

27

19

32

19

37

19

42

19

47

19

52

19

57

19

62

19

67

19

72

19

77

19

82

19

87

19

92

19

97

20

02

Cli

ma

te s

ign

als

0

2,000

4,000

6,000

8,000

10,000

12,000

PDO

ENSO

Available Energy

Surplus Energy in Spring the PNW is Predictable with Long Lead Times

Above normal surplus hydro from the PNW is more likely in cool PDO and cool ENSO years, less likely in warm PDO and warm ENSO years.

Probability of Exceedance for Spring Surplus Energy Resources in the PNW

Simple Monthly Energy Transfer Model

April May June July August

CA Natural Gas electricity $/MWh 59.52 59.52 59.52 59.52 59.52

PNW hydropower sale (HHL + load variance)

$/MWh 19.15 19.08 23.63 30.71 44.94

(PNW transmission) $/MWh 3.39 3.39 3.39 3.39 3.39

(PNW high load hour sale (HHL), PNW benefit)

$/MWh 18.05 17.98 22.53 29.61 43.84

If surplus energy is available (i.e. system-wide supply > PNW load) then energy can be shipped from the PNW to CA subject to the following constraints.

•Transferred energy does not exceed the intertie capacity during daylight hours ( instant intertie cap*10 hr*days per month)

•Transferred energy does not exceed the hydro production at John Day and The Dalles hydro plants

•Transfer is economically feasible (less cost than gas turbine generation in CA)

April Benefits:

CA = $36.98/ MW-hr

PNW = $18.05/ MW-hr

Apr-July Surplus System Wide Energy Available for Transfer(value is constrained by intertie capacity and the capacity of John Day and The Dalles hydro plants)

Extra Benefits and Costs over April-May-June-July in millions $

CA BENEFIT

PNW BENEFIT

WARM

ENSO 132 77

  PDO 135 78

  ENSO/PDO 120 69

COLD ENSO 172 102

  PDO 186 111

  ENSO/PDO 189 112

AVG   159 95

Estimate of Economic Benefits of Current Energy Transfers

Why is transferable energy so much higher in the most recent decades?

150000

200000

250000

300000

350000

400000

450000

190

0

191

0

192

0

193

0

194

0

195

0

196

0

197

0

198

0

199

0

200

0

Ap

r-S

ept F

low

(cfs

)

Effects of the PDO and ENSO on Columbia River Summer Streamflows

Cool CoolWarm Warm

PDO

Red=warm ENSO Green=ENSO neutral Blue=cool ENSO

y = 632.52x + 196974

R2 = 0.0598y = 1257.6x + 379657

R2 = 0.0842y = 2077.8x + 389645

R2 = 0.1032y = 780.89x + 244524

R2 = 0.0478

0

200,000

400,000

600,000

800,000

1,000,000

1,200,000

1917

1920

1923

1926

1929

1932

1935

1938

1941

1944

1947

1950

1953

1956

1959

1962

1965

1968

1971

1974

1977

1980

1983

1986

1989

1992

1995

1998

2001

April

May

June

July

Linear (April)

Linear (May)

Linear (June)

Linear (July)

Reservoir Storage “Resets”

Sys. Storage 80%

Sys. Storage 90%

Reservoir system storage in the most recent decades has been relatively high, which increases efficiency.

A Forecast Timeline

Jun Aug Jun

On ~June 1 we have:

•CurrentReservoirContents (PNW and CA)

•Current SummerStreamflow Forecast (PNW and CA)

•ENSO forecast•PDO forecast

CA Demand

Surplus PNWEnergy forComing Spring

PNW Winter Demand

Jan

Forecasts:

Two Primary Energy Planning Applications:

1) Predicting Natural Gas Consumption in CA

CA hydro and transferable surplus hydro in spring from the PNW to CA represents gas that will not be burned in CA to produce peaking energy in spring and summer. The probability distribution of this number is known with about 12-months lead time. Because CA and PNW hydro are typically in phase the effect is typically amplified.

2) Changes in Water Management in the PNW

Releases of water in late summer are at a premium in the PNW. Using streamflow forecasts, an available energy pool can be estimated that will not jeopardize reservoir refill in spring. Power from this pool could then be sold to spot market customers in CA depending on their needs. This effective transfer of energy from the coming spring to the current summer would benefit the PNW in terms of achieving more reliable fish flows in late summer and higher power revenues on average.

•Such actions would not reduce drought risks to fish (no energy to ship to CA).

•CA would almost always produce a marginal economic benefit by buying power from the PNW, but since the net $ benefit to CA is higher in spring than summer, CA would not necessarily want to trade spring energy for summer energy.