Alan F. HamletPhilip W. Mote

Dennis P. Lettenmaier

JISAO Center for Science in the Earth System Climate Impacts Group

and Department of Civil and Environmental EngineeringUniversity of Washington

September, 2004

Effects of Climate Variability and Change on Pacific Northwest Rivers:

Implications for Water Management and Salmon in the Columbia River Basin

http://www.hydro.washington.edu/Lettenmaier/Presentations/2004/hamlet_salmon_workshop_sept_2004.ppt

Example of a flawed water planning study:The Colorado River Compact of 1922

The Colorado River Compact of 1922 divided the use of waters of the Colorado River System between the Upper and Lower Colorado River Basin. It apportioned **in perpetuity** to the Upper and Lower Basin, respectively, the beneficial consumptive use of 7.5 million acre feet (maf) of water per annum. It also provided that the Upper Basin will not cause the flow of the river at Lee Ferry to be depleted below an aggregate of 7.5 maf for any period of ten consecutive years. The Mexican Treaty of 1944 allotted to Mexico a guaranteed annual quantity of 1.5 maf. **These amounts, when combined, exceed the river's long-term average annual flow**.

      

Hydroclimatology of the Pacific Northwest

Annual PNW Precipitation (mm)

Elevation (m)

The Dalles

Columbia River Basin Useable Storage ~35 MAF~50% of storage is in Canada~Storage is 30% of annual flowSnowpack functions as a natural reservoir

0.0

0.5

1.0

1.5

2.0

2.5

3.0

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

Month

No

rmal

ized

Str

eam

flow

SnowDominated

Transient Snow

Rain Dominated

Hydrologic Characteristics of PNW Rivers

Temperature warms,precipitation unaltered:

•Streamflow timing is altered• Annual volume stays about the same

Precipitation increases,temperature unaltered:

•Streamflow timing stays about the same•Annual volume is altered

Sensitivity of Snowmelt and Transient Riversto Changes in Temperature and Precipitation

0

100000

200000

300000

400000

500000

600000

700000

800000

900000

19

73

19

73

19

73

19

73

19

73

19

73

19

74

19

74

19

74

19

74

19

74

19

74

Water Year

Flo

w (

cfs

)

0

100000

200000

300000

400000

500000

600000

700000

800000

900000

19

73

19

73

19

73

19

73

19

73

19

73

19

74

19

74

19

74

19

74

19

74

19

74

Water Year

Flo

w (

cfs

)

A history of the PDOwarm

coolwarm

A history of ENSO

1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000

Pacific Decadal Oscillation El Niño Southern Oscillation

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

high highlow low

Ocean Productivity

PDO

1750 1775 1800 1825 1850 1875 1900 1925 1950 1975 2000Year

5.0

5.1

5.2

5.3

5.4

5.5

Lo

g1

0 m

ea

n f

low

, Th

e D

alle

s, O

R (

cfs

)

Source: Gedalof, Z., D.L. Peterson and Nathan J. Mantua. (in review). Columbia River Flow and Drought Since 1750. Submitted to Journal of the American

Water Resources Association.

Tree ring reconstructions of Columbia River flows showthe Dust Bowl was probably not the worst drought sequence

in the past 250 years.

red = observed, blue = reconstructed

Global Climate Change Scenarios and Hydrologic Impacts for the PNW

Humans are altering

atmospheric composition

The earth is warming -- abruptly

Precipitation Fraction, 2020s

0.5

0.75

1

1.25

1.5

1.75

J F M A M J J A S O N D

Frac

tion

hadCM2

hadCM3

PCM3

ECHAM4

mean

Delta T, 2020s

-1

0

1

2

3

4

5

J F M A M J J A S O N D

De

gre

es

C

hadCM2

hadCM3

PCM3

ECHAM4

mean

Delta T, 2040s

-1

0

1

2

3

4

5

J F M A M J J A S O N D

De

gre

es

C

hadCM2

hadCM3

PCM3

ECHAM4

mean

Precipitation Fraction, 2040s

0.5

0.75

1

1.25

1.5

1.75

J F M A M J J A S O N D

Fra

ctio

n

hadCM2

hadCM3

PCM3

ECHAM4

mean

Four Delta Method Climate Change Scenarios for the PNW

~ + 1.7 C ~ + 2.5 C

Somewhat wetter winters and perhaps somewhat dryer summers

ColSimReservoir

Model

VICHydrology Model

Changes in Mean Temperature and

Precipitation or Bias Corrected Output

from GCMs

Current Climate 2020s 2040s

Snow Water Equivalent (mm)

VIC Simulations of April 1 Average Snow Water Equivalentfor Composite Scenarios (average of four GCM scenarios)

The main impact: less snow

April 1 SWE (mm)

Current Climate “2020s” (+1.7 C) “2040s” (+ 2.5 C)

-44% -58%

Changes in Simulated April 1 Snowpack for the Cascade Range in Washington and Oregon

Regulated Flow

Historic Naturalized Flow

Estimated Range of Naturalized FlowWith 2040’s Warming

Naturalized Flow for Historic and Global Warming ScenariosCompared to Effects of Regulation at 1990 Level Development

Changes in Natural Streamflow for the “Middle of the Road” Scenarios

Current Climate--Blue2020s--Green2040s--Red

Impacts in the upper basin (Canada) are delayed in comparison with the lower basin (USA).

0

1000

2000

3000

4000

5000

6000

7000

8000

900010

/1

10/2

9

11/2

6

12/2

4

1/21

2/18

3/18

4/15

5/13

6/10 7/8

8/5

9/2

Date

Infl

ow

(ac

re-f

t) Simulated 20thCentury Climate

2020s ClimateChange Scenario

2040s ClimateChange Scenario

Effects to the Cedar River (Seattle Water Supply)for “Middle-of-the-Road” Scenarios

Frequency of Drought in the Columbia River Comparable to Water Year 1992

(data from 1962-1997)

0

2

4

6

8

10

12

14

16

Base Mean2020s

Mean2040s

ECHAM42040s

PCM2040s

Scenario

Nu

mb

er

of

Oc

cu

ren

ce

s

x 2

x 4.7

x 1.3 x 1.3

Effects of Hydrologic Changes

Increased Winter FlowIncreases winter flooding in some basins

Increased scouring events and sediment loads(?)

Potential benefits to winter hydro production

Reduced Snowpack and Earlier Snow MeltReduces spring flooding in some basinsReduces summer water availability (limited storage)Reduces summer hydro production

May change structure of mountain ecosystems Longer dry season may intensify forest disturbance

(e.g. fire, pests)Late summer streamflows systematically lowerIncreased water temperatures (?)

Decadal Climate Variability and Climate Change

Will Global Warming be “Warm and Wet” or “Warm and Dry”?

Answer: Probably BOTH!

150000

200000

250000

300000

350000

400000

45000019

00

1910

1920

1930

1940

1950

1960

1970

1980

1990

2000

Ap

r-S

ept F

low

(cfs

)

0

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

O N D J F M A M J J A S

Ave

rag

e S

trea

mfl

ow

(cf

s)

Base

2040_warm_PDO

2040_cool_PDO

Natural Streamflows at Dworshak

Sustainable management of PNW salmon populations will very likely have to cope with flow variability associated with both “warm and wet” and “warm and dry” scenarios at different times. Such conditions can be incorporated in planning as a test for sustainability though adverse periods, rates of recovery during favorable periods, etc.

Warm PDO2040

Cool PDO2040

Monitoring Observed Climate Change Impacts

Source: Mote et al., Declining Mountain Snowpack in Western North America (BAMS, 2004)

Trends in April 1 SWE 1950-1997

Trends in timing of peak snowpack are

towards earlier calendar dates

Change in Date

As the West warms,winter flows rise and summer flows drop

Figure by Iris Stewart, Scripps Inst. of Oceanog. (UC San Diego)

Cedar River: -30.7%

y = -0.0020x + 4.3416

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

1945 1955 1965 1975 1985 1995 2005

May

-Sep

t fr

acti

on

of

ann

ual

flo

w

May-Sept frac

Linear (May-Septfrac)

SFTolt River: -15.7%

y = -0.0010x + 2.2890

0

0.1

0.2

0.3

0.4

0.5

0.6

1945 1955 1965 1975 1985 1995 2005

May

-Sep

t fr

acti

on

of

ann

ual

flo

w

May-Sept frac

Linear (May-Septfrac)

Summer Water Availability is Declining

55 years

Figures courtesy of Matt Wiley and Richard Palmer at CEE, UW

Dworshak

y = -0.0006x + 0.506

0.3

0.4

0.5

0.6

0.7

0.8

streamflow

Linear (streamflow)

Fra

ctio

n of

Ann

ual S

trea

mflo

w O

ccur

ring

from

Jun

e-S

ept

Trends in Simulated Summer Water Availability for the N.F. Clearwater River at Dworshak Dam

Impact Pathways for Salmon Management and Recovery Efforts in the Columbia Basin

and Opportunities for Adaptation

•The flow needed to provide acceptable flow velocity for juvenile transport is frequently higher than natural flow, particularly in late summer (I.e. use of storage is required). Climate change increases the amount of storage required to meet flow targets.

•Currently very little storage is allocated to fish in comparison with hydropower.

•In a conflict between hydro or irrigation and fish flow, the current reservoir operating policies are designed to protect hydro and irrigation (fish flow storage allocation for main stem and Snake River flow targets is at the top of a shared reservoir storage pool)

•The Columbia River Treaty does not provide explicitly for summer flow in the U.S. (transboundary issues). Compare with guaranteed winter releases associated with flood control.

0

5000

10000

15000

20000

25000

30000

35000

40000

1

Sys

tem

Sto

rag

e (

kAF

) Hydro storage

Fish flow storage

Managed Flow Augmentation

Source: Payne, J.T., A.W. Wood, A.F. Hamlet, R.N. Palmer, and D.P. Lettenmaier, 2004, Mitigating the effects of climate change on the water resources of the Columbia River basin, Climatic Change, Vol. 62, Issue 1-3, 233-256

Adaptation to climate change will require complex tradeoffs between ecosystem protection and hydropower operations

2070-2098

60

80

100

120

140

FirmHydropower

Annual FlowDeficit atMcNary

Perc

en

t o

f C

on

tro

l R

un

Cli

mate

PCM Control Climate andCurrent Operations

PCM Projected Climateand Current Operations

PCM Projected Climatewith AdaptiveManagement

Flood Control vs. Refill

Because so little storage is currently allocated to fish flows, reliability of refill is crucial to achieving acceptable levels of flow augmentation in summer.

As streamflow timing shifts move peak flows earlier in the year, flood evacuation schedules may need to be revised both to protect against early season flooding and to begin refill earlier to capture the (smaller) spring freshet.

Model experiments (see Payne et al. 2004) have shown that moving flood evacuation two weeks to one month earlier in the year helps mitigate reductions in refill reliability associated with streamflow timing shifts.

Payne, J.T., A.W. Wood, A.F. Hamlet, R.N. Palmer, and D.P. Lettenmaier, 2004, Mitigating the effects of climate change on the water resources of the Columbia River basin, Climatic Change, Vol. 62, Issue 1-3, 233-256

Water Temperature

Higher air temperatures and increased residence time in reservoirs due to summer streamflow reductions are likely to systematically increase water temperatures throughout the basin. In unmanaged tributaries these impacts may be difficult or impossible to mitigate. (land use)

In managed basins, stored cold water in reservoirs may be exhausted more rapidly than now, reducing the ability to mitigate high stream temperatures using releases from storage, particularly in late summer. Cold water storage at Dworshak dam is a particular concern since it is one of the few dams available to control stream temperatures in the lower Snake and is sited in a sensitive area.

Broad Strategies for Incorporating Climate Variability and Climate Change in Long-Term Planning

Identify and Assess Climate LinkagesIdentify potential linkages between climate and resource management that could affect outcomes in the long term. What’s being left out? Are there future “deal breakers” in these omissions? (e.g. ocean productivity, glaciers maintaining summer streamflow in the short term)

Design for Robustness and SustainabilityUse modeling studies to test preferred management alternatives for robustness in the face of climate variability represented by paleoclimatic studies, conventional observations, decadal variability, and future climate change projections.

Identify Limits and Increase Response CapabilityUse estimates of uncertainties or “what if” scenarios to find the performance limits inherent in preferred management alternatives. How can response capability be increased?

Expect Surprises and Design for Flexibility to Changing ConditionsDesign contingency planning into management guidelines to allow for ongoing adaptation to unexpected (or uncertain) conditions without recursive policy intervention.

Selected References and URL’s

Climate Impacts Group Website

http://www.cses.washington.edu/cig/

White Papers, Agenda, Presentations for CIG 2001 Climate Change Workshop

http://jisao.washington.edu/PNWimpacts/Workshops/Skamania2001/WP01_agenda.htm

Climate Change Streamflow Scenarios for Water Planning Studies

http://www.ce.washington.edu/~hamleaf/climate_change_streamflows/CR_cc.htm

Refs on Climate Variability and Climate Change

http://www.ce.washington.edu/~hamleaf/hamlet/publications.html