Modeling the Boise Reservoir System with Climate Change Leslie Stillwater, Pacific NW Region

1. Modeling the Boise Reservoir System with Climate Change Leslie Stillwater, Pacific NW Region

2. Reclamation’s Boise Project • The Boise Project provides water to lands in southwestern Idaho and eastern Oregon • Boise Basin Project Reservoirs, • active capacities: • Anderson Ranch, 413 kAF • Arrowrock, 272 kAF • LuckyPeak (COE facility), 264 kAF • Lake Lowell, 159 kAF • Irrigated Lands: • 224,000 acres primary supply • 173,000 acres supplemental supply

4. Climate Change Is Likely to Bring… • Warmer Temperatures • Less Snowpack • More Precipitation (some say) • Greater variability in flows • Earlier runoff • Higher winter flows; lower summer flows

5. Preliminary Assessment Questions • will Reclamation meet it’s current contractual obligations for water storage and delivery? • will Reclamation meet it’s environmental obligations? • will current flood risk management practices be adequate?

6. The IPCC is a scientific intergovernmental body set up by the World Meteorological Organization (WMO) and by the United Nations Environment Programme (UNEP). The IPCC provides an objective and neutral source of information about climate change. The IPCC does not conduct research nor does it monitor climate related data or parameters.   Intergovernmental Panel on Climate Change (IPCC)

7. Selected IPCC Model Output Three projection scenarios for the year 2040 were selected from among the suite of IPCC models. Each applies an aggressive emissions scenario, assuming increasing emissions based on population growth and current technological and economic trends. • ECHAM • developed by the Max Planck Institute for Meteorology, Germany • Echam simulations produce moderate temperature and precipitation trends when compared to other IPCC models. • IPSL • developed by the IPSL Institut Pierre Simon Laplace, France. • IPSL simulations produce the greatest warming and increased precipitation trends. • GISS • developed by NASA / Goddard Institute for Space Studies, USA • GISS simulations produce the lowest warming trends and a small decrease in precipitation.

8. Selected IPCC Model Output Modeled Predictions for the Year 2040

9. 6 Scenarios were developed from 3 Climate Change Projections for Year 2040 • Echam - Temp Adjustments Only • Echam - Temp and Precip Adjustments • GISS - Temp Adjustments Only • GISS - Temp and Precip Adjustments • IPSL - Temp Adjustments Only • IPSL - Temp and Precip Adjustments

10. Climate Impacts Group(CIG) The projected precipitation and temperature changes were downscaled to the Pacific Northwest by the Joint Institute for the Study of Atmosphere and Oceans Climate Impacts Group (CIG) at the University of Washington.

11. How Climate AffectedNaturalized Streamflows were Developed T and P Adjustments were obtained from the Climate Impacts Group for each Climate Scenario T and P Adjustments were applied to the National Weather Service River Forecast System model (NWSRFS) to produce daily local naturalized streamflows (WY1949-1996)

12. NWSRFS Model • Forecasts streamflow • Collection of hydrologic/hydraulic models calibrated to observed snow, soil moisture, river, and reservoir conditions and historical meteorological data • Model input (for this study): T and P • Model components: • Sacramento Soil Moisture Accounting Model • SNOW-17

13. NWSRFS Concepts http://meteora.ucsd.edu/~knowles/html/land/mod_descr.html

14. NWSRFS Concepts

15. Daily Naturalized Flows at Lucky Peak using Temperature Predictions for 2040(NWSRFS model results)

16. Daily Naturalized Flows at Lucky Peak using Temperature and Precipitation Predictions for 2040 (NWSRFS model results)

17. How Climate Affected Streamflows were applied in thePlanning and Operations Models • Naturalized streamflows from each NWSRFS Climate Scenario were compared to the unadjusted NWSRFS Scenario (calibrated to historic streamflows) to produce daily local naturalized streamflow adjustments • The daily streamflow adjustments were applied to the Monthly Planning Model (Snake River Revised MODSIM Model) • The daily streamflow adjustments were also applied to the Daily Boise Operations Model (MODSIM)

18. Model Purposes and Assumptions • The monthly planning model addresses storage, Reclamation contract obligations, water rights, irrigation deliveries and minimum streamflows • Modeled irrigation deliveries are based on present level delivery requirements • Environmental obligations are based on current practices • The daily operations model addresses flood risk management and refill

19. Results (page1) • The ability to refill Project reservoirs may not be significantly impacted by Climate Change, but refill is dependent on the successful response to changing flood risks • Project deliveries are likely to not be significantly affected by Climate Change (natural flow diversions decline, Project storage deliveries increase) • Environmental obligations for storage and streamflows are likely to not be significantly affected by Climate Change

20. Planning Study: Modeled DiversionsTemperature Adjustments Only

21. Planning Study: Modeled DiversionsTemperature and Precipitation Adjustments

22. Planning Study: Modeled RefillTemperature Adjustments Only

23. Planning Study: Modeled RefillTemperature and Precipitation Adjustments

24. Results (page2) • However, winter and spring flooding and flood control operations are more likely with Climate Change and will be the major issue in the Treasure Valley • Flood risk management and reservoir refill go hand-in-hand

25. Observations on Flood Risk Management and Reservoir Refill (page1) Current COE regulations, guidelines and space requirements are outdated, having been developed using data from 1895 through 1980. Starting about 1980, the guidelines under-predict inflows prior to April 1 and over-predict inflows after April 1 With Climate Change, the under- and over- predictions are even larger

26. Observations on Flood Risk Management and Reservoir Refill (page2) Accurate forecasts will become more difficult develop to due to the influences of precipitation on the basin and increased flow variability

34. Boise Project Flood Operations Starting on January 1, a volume forecast from ‘now’ until the end of July is developed Part of that forecast will arrive April – July producing the April 1 space requirements (rule curves) The remainder of that forecast will arrive ‘now’ – March determining the rate of release to get down to the April 1 space requirements Measure of success is discharge < 7,000 cfs at Glenwood Bridge

35. No Adjustment 1970/1971 Apr 1 Space (AF) required = 950,150 achieved = 885,780

36. Echam T 1970/1971 Apr 1 Space (AF) required = 664,530 achieved = 663,780

37. Giss T 1970/1971 Apr 1 Space (AF) required = 822,790 achieved = 820,390

38. Ipsl T 1970/1971 Apr 1 Space (AF) required = 588,970 achieved = 582,100

39. Echam TP 1970/1971 Apr 1 Space (AF) required = 784,410 achieved = 613,210

40. Giss TP 1970/1971 Apr 1 Space (AF) required = 834,650 achieved = 831,930

41. Ipsl TP 1970/1971 Apr 1 Space (AF) required = 852,310 achieved = 244,280

42. No Adjustment 1970/1971 Apr 1 Space (AF) required = 950,150 achieved = 885,780

43. Impacts to Reservoir Refill Refill capabilities go hand-in-hand with flood control operations When modeled, perfect forecasts (!) and revised operating rules produce excellent refill capability in all Climate Change Scenarios studied

44. So, will Reservoirs fill? Streamflow forecasts will need to be early – that seems unlikely Greater streamflow variability produces greater uncertainties in streamflow forecasts Operators will be hesitant to draw down early and start an early fill

45. So, will there be flooding in Boise? Most likely* *given our current assumptions

46. Comments on the Daily Operations Studies Assumptions drive the study results A2 IPCC scenarios (aggressive emissions) T and P results are scalable to PN Region and to the Boise Basin Starting storage conditions of Nov 2001 (historic median) Perfect forecasts We addressed uncertainties by employing results from a range of IPCC models

47. Daily Operations Study Results Reliable forecasts will be even more critical Early forecasts (prior to Jan 1) will be required Drawdown needs to start before Jan 1 Space requirements need to start earlier than Apr 1 Maintaining 55% space in Lucky Peak and Arrowrock may not be possible Glenwood Bridge discharge Jan – Apr will be higher Glenwood Bridge discharge > 7,000 cfs can be anticipated if the wettest scenario is realized

48. Web sites • Hydromet teacup diagrams: • http://www.usbr.gov/pn/hydromet • Intergovernmental Panel on Climate Change • http://www.ipcc.ch/ • NWS models • http://www.nws.noaa.gov/ • Climate Impacts Group, U of Washington • www.cses.washington.edu/cig/