FORECAST-BASED OPERATIONS AT FOLSOM AND LAKE

237 217 200 80 252 237 217 200 119 174 237 217 200 27 .59

255 0 163 131 239 110 112 62 102 130 Bridging255 the0 Gap163 Conference132 65 135 92 102 56 120 The255 Dana on0 Mission163 Bay122 53 120 56 130 48 111 San Diego, CA January 28, 2019

Joe Forbis, P.E. Chief, Water Management Section Sacramento District, U.S. Army Corps of Engineers

“The views, opinions and findings contained in this report are those of the authors(s) and should not be construed as an official Department of the Army position, policy or decision, unless so designated by other official documentation.” AGENDA

• Background of watershed and • Need for Joint Federal Project (JFP) • New forecast-based operations at Folsom Dam AMERICAN RIVER HYDROLOGY

• Steep watershed

• Rain-on-snow potential

• Winter snowpack WATER SUPPLY SYSTEM CONTEXT

Central Valley Project State Water Project US Bureau of Reclamation CA Department of Water Resources FOLSOM DAM

Folsom Lake • Gross Pool (100% full): 967,000 ac-ft • Control Space: up to 600,000 ac-ft • Avg. Annual Unregulated Runoff: 2,788,000 ac-ft

Folsom Dam • Main spillway with eight radial gates • Elevation ~420 ft; Max release capacity ~567,000 cfs • River outlets • Max release capacity ~28,000 cfs • Auxiliary spillway (JFP) • Elevation ~370 ft; Max release capacity ~314,000 cfs FOLSOM DAM FLOOD HISTORY

• 1944 – Folsom Dam authorized – Designed to provide 500- year level of flood protection

• 1951 – Record Flood

• 1956 – Record Flood

• 1964 – Record Flood – New level of protection is 120-year

• 1986 – Record Flood – New level of protection is 60-year

• 1997 – Record Flood 350,000NEED FOR JFP 5 300,000 Peak Annual Inflow (Unimpaired) 3 4 250,000 2 1

200,000 6

Flow (cfs) 150,000

100,000

50,000

0

Year Construction of Original Dam Complete NEED FOR JFP

• Problems with the existing dam: • Only 400,000 acre-feet of authorized flood storage (wasn’t sized to include largest storms) • Can’t pass the Probable Maximum Flood without overtopping • 30% of flood storage used when downstream objective release (115,000 cfs) can be achieved

• Proposed solutions: • Additional upstream flood storage • Expansion of existing outlets • Auxiliary spillway, additional 200,000 ac-ft of variable flood storage, and forecast-based operations potential EXISTING AND JFP OUTLETS

Flood space (KAF)

Main Dam JFP Remaining Total Top of flood pool 0 115K cfs – without JFP 283 400 115K cfs – with JFP 569 600

Main Dam stilling basin floor 967 HISTORY OF FOLSOM OPERATIONS FORECAST-BASED OPERATIONS

• Theory: foreknowledge of runoff volume and timing enables optimal use of storage and release decisions

• Concerns: forecast uncertainty generates risk • Insufficient releases (increased flood risk) • Excessive releases (water supply risk)

• Challenges: • Limited forecast data record • Never been done before TWO BASELINES/THREE ALTERNATIVES

Flood Space (TAF) 1,000 0 Existing (USACE) 400 TAF Flood Space Basin Wetness 800 Existing (BOR-SAFCA) 400/670 TAF Flood Space Storage Upstream Storage Credit (TAF) Alternative 1 400/600 TAF Flood Space 600 Upstream Storage Credit 400 Alternative 2 400/600 TAF Flood Space Upstream Storage Credit + Basin Wetness 400 600 Alternative 3 400/600 TAF Flood Space 670 Forecasted Inflow Volume

200 Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct 1986 event pattern scaled to 200-yr

Forecast operation reflects perfect forecast

Existing (BOR/SAFCA)

JFP + US credit

JFP + Forecast-based NEW FOLSOM WCD NEW FOLSOM WCD FORECAST ENSEMBLE – 12/22

59 ensemble traces FORECAST ENSEMBLE – 12/23 FORECAST ENSEMBLE – 12/24 FORECAST ENSEMBLE – 12/25 FORECAST ENSEMBLE – 12/26 FORECAST ENSEMBLE – 12/27 FORECAST ENSEMBLE – 12/28 FORECAST ENSEMBLE – 12/29 FORECAST ENSEMBLE – 12/30 FORECAST ENSEMBLE – 12/31 FORECAST ENSEMBLE – 1/01 FORECAST ENSEMBLE – 1/02 FORECAST ENSEMBLE – 1/03 FORECAST INFLOW VOLUMES 3-day volume updated daily

59 ensemble traces

21 probabilistic traces Each trace reflects one value of Non- Exceedance Probability (NEP) ROBUSTNESS TESTING

• Identify smallest NEP that routes design event at target release.

• Identify greatest NEP that routes design event without drawing down to bottom of variable space.

• Identify smallest NEP that routes design event given 24- hour time shift in forecasted inflow volumes.

• Identify greatest NEP value that does not result in drawdowns that do not refill (false positives). ROBUSTNESS – 1/200 EVENT (1997) ROBUSTNESS – 1/200 EVENT (1997) ROBUSTNESS RESULTS

Minimum Flood Performance Metric Successful Event, Target Release NEP 1/100 1986 pattern, 115 kcfs 0% (ALL PASS) 1/100 1997 pattern, 115 kcfs 0% (ALL PASS) 1/200 1986 pattern, 160 kcfs 0% (ALL PASS) 1/200 1997 pattern, 160 kcfs 0% (ALL PASS)

Maximum Drawdown Metric Successful Minimum storage > 367 TAF NEP 1/100 1986 pattern 60% 1/100 1997 pattern 80% 1/200 1986 pattern 70% 1/200 1997 pattern 85% QUESTIONS?