Piloting Utility Modeling Applications: The San Francisco Public - PowerPoint PPT Presentation

Piloting Utility Modeling Applications: The San Francisco Public Utilities Commission David Behar Climate Program Director Alexis Dufour, P.E. Hydrology and Water Systems Modeling PUMA Workshop December 1, 2010 San Francisco

The San Francisco Public Utilities Commission 1. Department of City and County of San Francisco 2. 2.5 million customers 3. 28 wholesale customers in four counties, 800,000 retail customers in City of San Francisco 4. 2300 employees 5. Three Enterprises: Water, Wastewater, Power 6. FY 2011-12 budget: $867 million

SFPUC assessment work to date 1. Statistical modeling, simple delta method, temperature only, to 2025 2. Developing hydrological modeling tools (Dufour) 3. Treasure Island development – Assisting in framing adaptation/ adaptive management approach for SLR 4. Co-founder and Chair, Water Utility Climate Alliance

Audiences for this work 1. Utility Staff 2. Senior Managers 3. Commission 4. Board of Supervisors 5. Mayor 6. RATEPAYERS (wholesale and retail)

Our goals 1. Learn 2. Stay abreast of the science 3. Develop ongoing expert relationships 4. Understand and synthesize uncertainty 5. Advance impacts analysis step-by-step 6. Be the most informed constituency in San Francisco 7. Plan adaptation

SFPUC Water System 85% 5% 10%

Water Supply Vulnerability to Changing Climatic Conditions: Modeling Approach Climate Modeling & Downscaling Tools Hydrologic models for Sierra Nevada & Bay Area Watersheds Planning Model (HH/ LSM) Decision Making Policy Infrastructure (Decision Support Planning Planning Planning Methods)

Assess sensitivity of SPUC water entitlement to changing climatic conditions 800 Unimpaired flow at LG Districts Max Rights 700 20,000 City entitlement 600 Unimpaired 15,000 500 Flow Water Available at La Grange 400 to the City TI D/ MI D (1,000 AF) 10,000 Entitlements 300 (cfs) 200 5,000 100 0 0 Dec Jul Oct Nov Jan Feb Mar Apr May Jun Aug Sep Water Year 2010

Snowline in 2000 Hetch Hetchy Reservoir

Monthly Hetch Hetchy Inflow Exceedences WY 1931 to 2005 450 5% 400 25% 350 50% 300 75% 95% Monthly 250 Inflow 200 (TAF) 150 100 50 0 Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep

Water Supply Vulnerability to Changing Climatic Conditions: Modeling Approach Climate Modeling & Downscaling Tools Hydrologic models for Sierra Nevada & Bay Area Watersheds Planning Model (HH/ LSM) Decision Making Policy Infrastructure (Decision Support Planning Planning Planning Methods)

Water supply planning model: the Hetch Hetchy and Local System Model – HH/ LSM • mass balance model • representation of each operational objective and constraint • Excel VBA platform • monthly time step • recurrence of historical hydrology (1920-2002)

Water Supply Vulnerability to Changing Climatic Conditions: Modeling Approach Climate Modeling & Downscaling Tools Hydrologic models for Sierra Nevada & Bay Area Watersheds Planning Model (HH/ LSM) Decision Making Policy Infrastructure (Decision Support Planning Planning Planning Methods)

HFAM Hydrologic Modeling Tool • physically-based conceptual hydrologic model to simulate: snow accumulation and melt, soil moisture and runoff • model time step: hourly • input: temperature, precipitation, solar radiation, wind movement, and potential evapotranspiration • output: snow pack, soil moisture, actual evapotranspiration, streamflow

Land Segments defined by: topography (elevation, aspect, slope), mean precipitation, soils/ rock, vegetation Size: 2 – 6 km 2 (500 – 1,500 acres)

HF AM Natur al F low Re ac he s Legend Rafferty 3005 3000 Tuolumne 3020 reach Dana 3010 3020 3015 3110 reservoir Delaney 3110 Cathedral 3024 3025 reservoir represented as a reach 3040 3045 3030 3185 gaged reach direction of flow 3044 3050 3055 3060 Matterhorn 3054 3059 River reaches Register 3064 3065 Piute Cr. 3070 3075 3080 3085 receive Benson Lake 3079 3084 Rancheria 3090 land segment 3124 3125 3094 3095 runoff 3135 Falls Creek 3100 3105 3112 Hetch Hetchy 3139 3140 3104 3109 3147 Lake Eleanor and Cherry Creek 3155 3165 3152 3117 Cherry Lake 3160 3170 3182 3187 3122 Diversion to Mountain Tunnel transport water 3169 3190 3195 3200 3237 South Fork, Tuolumne 3205 3210 3220 3225 to reservoirs 3215 3230 Clavey River 3239 3240 3250 3255 3262 Pine Mountain Lake 3245 3265 3270 Big Creek North Fork, Tuolumne 3275 3280 3285 3302 Turnback 3290 Moccasin 3295 Sullivan 3305 3317 Don Pedro Woods Creek 3310 LaGrange 3317

Weather Stations currently used in HFAM 11 stations Period of interest 1975-2008 1/ 8 deg( ~ 12x12km)

Water Supply Vulnerability to Changing Climatic Conditions: Modeling Approach Climate Modeling & Downscaling Tools Hydrologic models for Sierra Nevada & Bay Area Watersheds Planning Model (HH/ LSM) Decision Making Policy Infrastructure (Decision Support Planning Planning Planning Methods)

How to use climate data and climate projections? 1) Methodological approach 2) Hydrologic modeling implementation

Methodological approach questions 1. What can we expect from climate models given the uncertainties (forcing, initial conditions, model imperfections)? -> How to use the projections? 2. Should we use ensemble data from the model with the greatest skill or use multi-model ensemble data? 3. Dynamical vs statistical downscaling? 4. Top-down vs Bottom-up

Hydrologic modeling questions 1. Temporal distribution of temperature projections on a seasonal and diurnal time scale 2. Sensitivity of temperature projections to elevation 3. Trends for precipitation projections 4. Projections for other variables: Solar radiation, wind, relative humidity

Trend in Daily Minimum Temperature at Hetch Hetchy (1930-2005)

QUESTIONS

Current approach to climate change projections 10 1A 2A 2B 2C 3A 3B 5 0 Precipitation Change (% of MAP) -5 -10 -15 6 5 4 Temperature Change 3 (deg C) 2 1 0 2040 2070 2100

• NWS Climate Prediction Center, • SFPUC, • USGS, • NWS River Forecast Center, Data ata: • NRCS Snotel, • IRI Seasonal Forecast Other products from academic institutions Up Count untry • WSFM • Base Ops Bay A y Area • MOS • Hydrologic Frequency Analysis • Qualitative Demand Forecasting • Local Reservoir System Model Time scal scale: Monthly to Seasonal Decis isio ions: releases, pipeline rates, treatment plant rates, instream flow, maintenance outages, transfers

Water Supply Forecast Model