Reducing Water Resources Vulnerability of Climate Change through Adaptive Management

8/9/2019 Reducing Water Resources Vulnerability of Climate Change through Adaptive Management

1/23

Aris GeorgakakosHuaming YaoMartin Kistenmacher

Integrated Decision Support: Reducing Water Resources Vulnerability toClimate Change through Adaptive Management

Integrated Decision Support FrameworkGCM Scenarios, Downscaling, Hydrology, Water Resources

Climate Change Assessments for Northern CaliforniaCurrent vs. Adaptive Policies; Historical vs. Future Scenarios; Vulnerability

Conclusions/Further AssessmentsMitigation potential of adaptive, risk based management

Kosta GeorgakakosNick GrahamFang-Yi Cheng

Cris Spencer


2/23

Generate consistent climate forcing sequences ofRainfall and temperature.

River/ReservoirPlanning & Management

Simulate soil moisture,evapotranspiration, runoff,and streamflow.

GCM ScenariosDownscaling

Simulate current andadaptive mgt. policiesand assess impactson water uses.

Watershed Hydrology

VulnerabilityAssessmentandMitigation Potential

Integrated Modeling Framework

SanJ

oaqu

inRive

r

San Luis

Clair Engle Lake

Trinity Power Plant

Lewiston

Lewiston

JF Carr

Whiskeytown

Shasta

Keswick

ShastaSpringCr

Keswick

Oroville

Thermalito

Folsom

Natoma

New Melones

Tulloch

Goodwin

Oroville

Folsom

Nimbus

Melones

Tracy

Pumping

Banks

Pumping

SanJ

oaqu

inRive

r

AmericanRiv

er

Feath

erRiv

er

Sacram

entoRiver

TrinityRiver

ClearCreek

YubaR

iver

BearRi

ver

Delta-Mendota Canal

California Aqueduct

ONeillForebay

To Dos Amigos PP

To Mendota Pool

Sacramento San JoaquinRiver DeltaReservoir/

Lake

Power Plant

Pumping Plant

River Node

Reservoir/

Lake

Power Plant

Pumping Plant

Reservoir/

Lake

Power Plant

Pumping Plant

River Node

ISV

IFT

IES,IMC,IYB,ITI

DDLT,DBS,DCCWD,DNBA

DDM

DFDM

DDA

DSF

DSB

Black Butte

NewBullards Bar


3/23

CLIMATEMODEL:NCARCCSM3.0 (COUPLEDMODEL)

SCENARIO: A1B MIDDLELEVELSCENARIO

DECLININGEMISSIONSAFTER2050

MAXCO2CONCENTRATONOF~715PPMAT2100

RESOLUTION: ~120KMHORIZONTALRESOLUTION

26VERTICALLAYERS

6HRSTEMPORALRESOLUTION

VARIABLESUSED:

3D

ATMOSPHERIC

VARIABLES

TWOINPUTSETS:1970 2019 AND2050 2099

GoodLargeScalePrecipitationCorrespondence

ofHistorical19501999runwithNCEP

Reanalysis1948

1997

for

West

Coast

Climate Scenarios


4/23

OROGRAPHICPRECIPITATION

GRIDDED

MODEL

WINDFLOWFROMQUASISTEADYSTATEPOTENTIALTHEORYFLOW

WATERSUBSTANCESOURCE/ADVECTIONMODELWITHKESSLERMICROPHYSICS

10X10SQKMSPATIALAND6HOURLYTEMPORALRESOLUTION

SURFACETEMPERATUREGRIDDEDMODEL

INTERPOLATION/ADJUSTMENTOF

CCSM3.0

LOW

LEVEL

TEMPERATURE

OVER

TERRAIN

SURFACEENERGYBALANCEMODEL(OROGRAPHICANDSNOW/SOILMODELCOUPLING)

10X10SQKMSPATIALAND6HOURLYTEMPORALRESOLUTION

GISBASEDSYSTEMFORCATCHMENTDELINEATIONANDPRODUCTIONOFMAPAND MAT

Reference:HRCGWRI:http://www.energy.ca.gov/pier/project_reports/CEC5002006109.html

Dynamic Downscaling: Mean Areal Precipitation and Temperature


5/23

ADAPTATIONOF

NWS

OPERATIONAL

SNOW

ACCUMULATION

AND

ABLATION

MODEL

ADAPTATIONOFNWSOPERATIONALSOILWATERACCOUNTINGMODEL

KINEMATICROUTINGTHROUGHRIVERNETWORKFORALLBASINS

HydrologicModelDomain

OrovilleSubcatchments

Oroville HistoricalESPReliabilityDiagrams

Watershed Hydrology: Snow, Soil, and Channel Modeling System



6/23

PrecipitationDifference(mm/6hrs) FebTemperatureDifference(oC) Feb1800Z

Futurewintersarewarmerandwetter(athigherelevations)thanthehistorical.

SimulatedFlows:

Future

Historical

Futureflowsaresomewhathigher

andoccur

earlier

(Shasta/Oroville)

Selected Results: Temperature, Precipitation, Streamflow


7/23

SanJ

oaqu

inRive

r

San Luis

Clair Engle Lake

Trinity Power Plant

Lewiston

Lewiston

JF Carr

Whiskeytown

Shasta

Keswick

ShastaSpringCr

Keswick

Oroville

Thermalito

Folsom

Natoma

New Melones

Tulloch

Goodwin

Oroville

Folsom

Nimbus

Melones

Tracy

Pumping

Banks

Pumping

SanJ

oaqu

inRive

r

AmericanRiv

er

Feath

erRiv

er

Sacram

entoR

iver

Trinity

River

ClearC

reek

Yuba

River

BearRi

ver

Delta-Mendota Canal

California Aqueduct

ONeill

Forebay

To Dos Amigos PP

To Mendota Pool

Sacramento San Joaquin

River DeltaReservoir/Lake

Power Plant

Pumping Plant

River Node

Reservoir/

Lake

Power Plant

Pumping Plant

Reservoir/

Lake

Power Plant

Pumping Plant

River Node

ISV

IFT

IES,IMC,IYB,ITI

DDLT,DBS,DCCWD,DNBA

DDM

DFDM

DDA

DSF

DSB

Black Butte

New Bullards Bar

Northern California River and ReservoirSystem Schematic

Trinity River System (Clair Engle Lake,Trinity Power Plant, Lewiston Lake, LewistonPlant, JF Carr Plant, Whiskeytown, ClearCreek, and Spring Creek Plant);

Shasta Lake System (Shasta Lake, ShastaPower Plant, Keswick Lake, Keswick Plant, andthe river reach from Keswick to Wilkins);

Feather River System (Oroville Lake, OrovillePower Plants, Thermalito Diversion Pond,Yuba River, and Bear River);

American River System (Folsom Lake,Folsom Plant, Natoma Lake, Nimbus Plant,Natoma Plant, and Natoma Diversions);

San Joaquin River System (New MelonesLake, New Melones Power Plant, Tulloch Lake,Demands from Goodwin, and Inflows fromthe main San Joaquin River); and

Bay Delta (Delta Inflows, Delta Exports,Coordinated Operation Agreement--COA, andDelta Environmental Requirements).

River and Reservoir Modeling System

Objectives:Water SupplyEnergy GenerationEnvironment

EcologyRecreation


8/23

202615.5546300San Luis80

20106757Tulloch70

25712731100800New Melones60

102283470327Folsom50

3537.8852.2900640Oroville40

434711671068900Shasta302840.2312231000Whiskeytown20

261731323802145Clair Engle Lake10

SmaxSm inHmaxHm inReservoir NameReservoir

ID

202615.5546300San Luis80

20106757Tulloch70

25712731100800New Melones60

102283470327Folsom50

3537.8852.2900640Oroville40

434711671068900Shasta302840.2312231000Whiskeytown20

261731323802145Clair Engle Lake10

SmaxSm inHmaxHm inReservoir NameReservoir

ID

1502New Melones13.52Nimbus

2103Folsom

6006Oroville

753Keswick

6595Shasta

1502Spring Creek141.42JF Carr

0.351Lewiston

1402Trinity

Capacity (MW )UnitsPower Plant

1502New Melones13.52Nimbus

2103Folsom

6006Oroville

753Keswick

6595Shasta

1502Spring Creek141.42JF Carr

0.351Lewiston

1402Trinity

Capacity (MW )UnitsPower Plant

River Nodes

Lewiston

JF Carr

Clear Creek

Spring Creek

Keswick

WilkinsFeather

American River

Freeport

Goodwin

SJR above Stanislaus

SJR at Vernalis

Antioch

Delta Exit

Tributary Inflows

Trinity

Whiskeytown

Shasta

Keswick-Wilkins

Oroville

Yuba River

Bear River

Folsom

Sacramento Miscellaneous

Eastside streams

Delta Miscellaneous streams

New Melones

San Joaquin River

Water SupplyThermalito

Folsom Pumping

Folsom South Canal

OID/SSJID

CVP Contractors

CCWD

Barker Slough

Federal Tracy PP

Federal Banks On-Peak

Federal Banks Off-Peak

Federal Banks PP Total

Federal Banks PP CVC

Federal Banks PP - Joint Point

Federal Banks PP Transfers

North Bay Aqueduct

State Banks PP

State Tracy PP

Delta Mendota Canal

Federal Dos Amigos

Federal O'Neil to Dos Amigos

San Felipe

Cross Valley Canal

Federal Exchange O'Neil

Federal Exchange San Luis

South Bay/San Jose

State Dos Amigos

Delta Consumptive Use

Freeport Treatment Plant

Yolo Bypass

Transfer Inflow

AFRP (Anadromous Fish Restoration Plan)

Clear Creek Below Whiskeytown Lake (Trinity)

Below Keswick Dam (Sacramento)

Below Nimbus Dam (American)

River and Reservoir Modeling System (2)


9/23


10/23

Current Policy

Generate inflow forecastsmedian trace (HA).

Determine water year type (DWR: C/D/N/AN/W).

Adjust base demands based on year type. Determine next month reservoir releases to

- meet water delivery targets and minimumrequired flows at various river nodes,

assuming no extra releases are required to meetDelta demands (X2) and pumping to South CA.

If X2 requirements and south CA delivery targetsare not met, increase releases according to COA(roughly 25/75 rule).

If deficits persist, allocate water to meet X2 first,then south CA water deliveries.

Repeat at the next month.

River and Reservoir Modeling System (4)Current and Adaptive Management Policies

Adaptive, Risk-based Policy

Generate inflow forecastsfull ensemble (HA).

Determine reservoir releases for the next 9

months to- meet water delivery targets and minimumrequired flows at various river nodes,

- meet environmental and ecological Deltarequirements associated with the X2,location and Delta outflow,

- generate as much energy as possible, and

- maintain high reservoir levels andsufficient carry-over storage.

(System-wide, stochastic optimization;Not according to the COA. )

Apply first month release and repeat.

Main Policy Differences

Current Policy Adaptive Policy Focuses on current month. Optimizes over the next 9 months. Deterministic. Risk based.

Adjusts demand targets twice a year. Re-optimizes every month. Follows COA in extra water allocation. Finds optimal allocation strategy each time.


11/23

Water DistributionFlow RegulationHydro Plant Operation

Emergency Response

Monthly Decisions Releases/Energy

Target Conditions

State VariablesPlanning Tradeoffs

Water Supply/Allocation Energy Generation Carry-over Storage Env.-Ecosystem Management

Development Tradeoffs

Urban/Industrial Agriculture Power System Socio-economic & Ecological

Sustainability

Operational Tradeoffs

Flood Management Water Distribution Energy Generation Env.-Ecosystem Management

Benefit/Impact Functions Water Supply Energy Flood Damage Env.-Ecosystem

Scenario/Policy Assessment

Monthly / Several Decades

Actual HydrologicConditions

Actual Demands

Climate-HydrologicForecasts

Demand Forecasts Water Food Energy Env.-Ecosystem

Climate-HydrologicForecasts

Demand Forecasts Water Supply Power Load/Tariffs Flood Damage Env.-Ecosystem Targets

Inflow Scenarios

Development/DemandScenarios Water/Energy Water/Benefit Sharing Environmental Sustainability

Daily Decisions Releases/Energy

Target Conditions State Variables

Benefit/Impact Functions Water Supply Energy Flood Damage

Env.-Ecosystem

Near Real Time Decision Support

Hourly / 1 Day

Mid/Short Range Decision Support

Daily, 6-Hourly, or Hourly / 1 Month

Long Range Decision Support

Weekly, 10-Day or Monthly / 1-2 Years

Infrastructure Develpmnt.Water Sharing CompactsSustainability Targets

Management Policy

INFORM DSS: OverviewMultiple Objectives, Time Scales, & Decision Makers

ManagementAgencies

/Decisions

PlanningAgencies/Decisions

OperationalPlanningandManage

ment

Off-lin

e

Assessments

Adaptive Management System (INFORM DSS)


System-wide, stochastic optimization


12/23

Assessment Criteria

Lake Levels, SpillageWater Supply Reliability

Energy Generation

Bay Environment (X2)

others.

Simulation Horizon1970 to 2019 (Historical)

2050 to 2099 (Future)

9-month Forecast-Decision Horizon

(Monthly time steps)

Management Policy

Inflow Forecasting

River/ReservoirSimulation

Reservoir Mgt.Basin wide

One Step System

Simulation

Inflow Scenario

Demand Scenario

Regulation Policy

Assessment Process

SanJ

oaqu

inRive

r

San Luis

Clair Engle Lake

Trinity Power Plant

Lewiston

Lewiston

JF Carr

Whiskeytown

Shasta

Keswick

ShastaSpringCr

Keswick

Oroville

Thermalito

Folsom

Natoma

New Melones

Tulloch

Goodwin

Oroville

Folsom

Nimbus

Melones

Tracy

Pumping

Banks

Pumping

SanJ

oaqu

inRive

r

AmericanRiv

er

Fea

therRiv

er

SacramentoR

iver

TrinityRiver

ClearCreek

Yuba

River

BearRi

ver

Delta-Mendota Canal

California Aqueduct

ONeillForebay

To Dos Amigos PP

To Mendota Pool

Sacramento San Joaquin

River DeltaReservoir/Lake

Power Plant

Pumping Plant

River Node

Reservoir/

Lake

Power Plant

Pumping Plant

Reservoir/

Lake

Power Plant

Pumping Plant

River Node

ISV

IFT

IES,IMC,IYB,ITI

DDLT,DBS,DCCWD,DNBA

DDM

DFDM

DDA

DSF

DSB

Black Butte

NewBullards Bar


13/23

Inflow Comparison (Historical vs. Future Scenario)

Average future inflows are somewhat higher.(Trinity 6.3%; Oroville 10%; Shasta 4.3%; Folsom 5.6%.)

Minimum future inflows are considerably lower indicating more severe droughts(27% reduction).

Future inflows are more variable.

Wet season shifts earlier.

Trinity Monthly Mean

0

500

1000

1500

2000

2500

3000

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

Trinity His

Trinity Fut

Total Reservoir Inflows

0

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

0 10 20 30 40 50 60 70 80 90 100

Exceedance of Probability (%)

TAF/Year

Future

Historical

Trinity Historical vs. Future Inflows

System Historical vs. Future Inflows


14/23

Lake Levels: Current vs. Adaptive Policies forHistorical and Future Scenarios

Lake levels exhibit considerably greater seasonal and annual variability in the futurescenario.

System conservation storage is used up in the future scenario. Drought vulnerabilityincreases.

Adaptive DSS policy exhibits higher lake levels and less spillage than current policy.

System Storage Sequences; Historical Period

0

2000

4000

6000

8000

10000

12000

Jan-7

4

Jan-7

6

Jan-7

8

Jan-8

0

Jan-8

2

Jan-8

4

Jan-8

6

Jan-8

8

Jan-9

0

Jan-9

2

Jan-9

4

Jan-9

6

Jan-9

8

Jan-0

0

Jan-0

2

Jan-0

4

Jan-0

6

Jan-0

8

Jan-1

0

Jan-1

2

Jan-1

4

Jan-1

6

Jan-1

8

TAF

His/DSS

His/CurrentPolicy

System Storage Sequences; Future Period

0

2000

4000

6000

8000

10000

12000

Jan-74

Jan-76

Jan-78

Jan-80

Jan-82

Jan-84

Jan-86

Jan-88

Jan-90

Jan-92

Jan-94

Jan-96

Jan-98

Jan-00

Jan-02

Jan-04

Jan-06

Jan-08

Jan-10

Jan-12

Jan-14

Jan-16

Jan-18

TAF

Future/DSS

Future/CurrentPolicy

Historical Future


15/23

Water Deliveries: Current vs. Adaptive Policies forHistorical and Future Scenarios

Current policy provides higher amounts during wet years and lower during dry years.Adaptive DSS policy is more balanced and reliablereduces vulnerability.

Current policy WS during most severe drought (TAF): 4,798 (Historical); 2,545 (Future)Adaptive DSS WS during most severe drought (TAF): 4,923 (Historical); 4,949 (Future)

System Water Delive ries, Historical Inflows

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

0 10 20 30 40 50 60 70 80 90 100

Exceedance of Probability(%)

TAF/Year

His/DSS

His/CurrentPolicy

System Water Deliveries, Future Inflows

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

0 10 20 30 40 50 60 70 80 90 100

Exceedance of Probability(%)

TAF/Year

Future/DSS


Historical Future


16/23

Energy Generation: Current vs. Adaptive Policies forHistorical and Future Scenarios

Annual System Energy, Historical Period

0

1000

2000

3000

4000

5000

6000

7000

0 10 20 30 40 50 60 70 80 90 100


GWH/Year

His/DSS

His/CurrentPolicy

Annual System Energy, Future Period

0

1000

2000

3000

4000

5000

6000

7000

0 10 20 30 40 50 60 70 80 90 100


GWH/Year

Future/DSS


Historical Future

Average energy generation increases by 5% in the future scenario under both policies.

Firm energy generation decreases by 10% under the Adaptive Policy and 29% underthe Current Policy.


17/23

Delta Outflow and X2 Location: Current vs. Adaptive Policies forHistorical and Future Scenarios

Adaptive DSS policy meets Delta outflow and X2 requirement in both scenarios.

Current Policy violates Delta outflow and X2 requirement (by 28 kilometers) in futuredroughts.

Current Policy Adaptive DSS Policy

Delta Outflow

X2 Location


18/23

Performance Differences (%) ofFuture relative to Historical Scenario

Future vs. Historical Period

27.9

3.0

-47.0

3.5

-21.4

35.0

14.5

7.7

0.5

3.9

-9.2

0.0

-60

-50

-40

-30

-20

-10

0

10

20

30

40

Precent(%)

Current Policy

DSS

Avg. Spillage Avg. WS

Min. WS

Avg. Energy

Firm Energy

X2 Violation

Future vs. Historical Period

27.9

3.0

-47.0

3.5

-21.4

35.0

14.5

7.7

0.5

3.9

-9.2

0.0

-60

-50

-40

-30

-20

-10

0

10

20

30

40

Precent(%)

Current Policy

DSS

Avg. Spillage Avg. WS

Min. WS

Avg. Energy

Firm Energy

X2 Violation

Current policy worsens in the future scenario: More spillage (27.9%), less minimum water deliveries (47%),

less firm energy (21.4%), and significant X2 and delta outflow violations (35%).+ Increased average water deliveries (3%) and energy generation (3.5%).

Adaptive DSS policy more robust between historical and future scenarios:+ Increased average water deliveries (7.7%), increased minimum water deliveries

(0.5%), increased average energy (3.9%), and no X2 and delta outflow violations. Increased spillage (14.5%), less firm energy (9.2%).


19/23

DSS vs. Current Policy

-2.1 -3.5

2.6-0.1

2.80.0

-12.3

0.9

94.4

0.2

18.8

-35.0

-50

-30

-10

10

30

50

70

90

110

Precent(%)

Historical

Future

Avg. Spillage Avg. WS Min. WS Avg. Energy Firm Energy

X2 Violation

DSS vs. Current Policy

-2.1 -3.5

2.6-0.1

2.80.0

-12.3

0.9

94.4

0.2

18.8

-35.0

-50

-30

-10

10

30

50

70

90

110

Precent(%)

Historical

Future

Avg. Spillage Avg. WS Min. WS Avg. Energy Firm Energy

X2 Violation

Adaptive DSS vs. Current Policy differences are minor in the historical scenario.

Adaptive DSS policy is notably more robust in the future scenario with respect toall criteria, especially minimum water supply, firm energy, Delta requirements, andspillage.

There are nonlinear interactions and tradeoffs underlying system response and

performance against the different criteria. Such assessments serve to quantify andcommunicate these interdependencies.

Performance Differences (%) ofDSS relative to Current Policy


20/23

Future A1B scenario portents intensifying water stresses (due to seasonal inflowshifts and higher inflow variability) and higher vulnerability to extreme droughts.

Adaptive, risk based, reservoir regulation strategies are self tuning to the changingclimate, deliver more robust performance than current management practices, and

can considerably mitigate the negative impacts of increased water stresses.

Effective implementation of adaptive, risk based, reservoir regulation strategies require

more flexible laws and policy statutes (COA, heuristic rules, etc.), a new level of institutional cooperation for water resources management, and

capacity building of agency personnel in modern decision support methods.

Conclusions


21/23

Climate

Forecasts

DecisionRules

Impact Forecasts

Flood Damage, Water Supply,Energy Generation, AgriculturePublic Health, etc.

HydrologicForecasts

Impacts

p

Uncertainty Management: Climate Hydrology Management

u: Static/Fixed

u: Dynamic/Adaptive

Adaptive decision rules can manage forecastuncertainty.

Heuristic regulation rules cannot.

H H H (P, P ) ,Q (T,, Q T{ } f [ S , , k ]),

=

{ , (P, P ) (T, T ) }

S S SS(Q,Q ),(

{ }

f [ ...)S , , ,

I, I

u(S ) k, ]

=

=


22/23

Further Work

Bracket cloud influence in climate downscaling component.

Incorporate impacts ofsea-level rise.

Assessments ofother GCM scenarios (A2, B1, etc.) to investigate the sensitivity of thefindings presented herein.

Assessments with daily and sub-daily temporal resolution to quantify climate changeimpacts on other system functions and outputs (flooding, energy generation markets,ecosystem response, etc.).

Multi-stressor assessments including demand and land use change.

Conjunctive, statewide surface water groundwater assessments.


23/23

The INFORM project was sponsored by the NOAA Climate Program Office, the CaliforniaEnergy Commission PIER Program, and the CALFED Program.

Contributors included several scientists and managers from the California Department ofWater Resources, the Bureau of Reclamation, the US Army Corps of Engineers,NOAA/NWS/CNRFC, NOAA CPO, and others.

The Climate Change INFORM application was funded by the Energy Commission PierClimate Change Program. We thank Guido Franco for his support and guidance, and RobHartman of CNRFC for making available operational historical data.

Acknowledgements

Reducing Water Resources Vulnerability of Climate Change through Adaptive Management

Documents

Transcript of Reducing Water Resources Vulnerability of Climate Change through Adaptive Management