The Florida Water and Climate Alliance: A Collaborative Working Group for the Development of Climate Predictions for Improved Water Management

Transcription

1 The Florida Water and Climate Alliance: A Collaborative Working Group for the Development of Climate Predictions for Improved Water Management Wendy Graham, Ph. D., Director, UF Water Institute Tirusew Assefa, Ph. D., Tampa Bay Water Alison Adams, Ph. D., Tampa Bay Water

2 The Florida Water and Climate Alliance Initially funded by NOAA Climate Program Office: CSI and SARP Now funded by local partners. Goal: To increase the regional relevance and usability of climate and sea level rise models for the specific needs of water suppliers and resources managers in Florida.

3 Project Activities! Develop a collaborative Working Group comprised of public water suppliers, water resource managers, climate scientists, and hydrologic scientists! Evaluate the practical applicability of current climate data/models predictions at utility relevant space-time scales! Evaluate the usefulness of these data/models for minimizing current and future public water supply risks associated with climate variability/climate change and/or sea level rise Academic Partners: UF Water Institute; Southeast Climate Consortium; UF Center for Public Issues Education; FSU COAPS; U Miami RSMAS Public Utilities: Miami-Dade County; Broward County; Palm Beach County; Peace River Manasota Regional Water Supply Authority; Tampa Bay Water; Orlando Utilities Commission; Gainesville Regional Utilities Water Management Districts: SFWMD, SWFWMD; SJRWMD

4 Develop a collaborative working group Actionable climate science - Data/models/tools relevant to water supply operations and management Domain Community Learning together Practice Public water suppliers, resource managers, climate, social and hydrologic scientists, local planners Workshops, projects, research, website, reports, s, personal communication, outreach Based on: Wenger, Communities of Practice, May 29

5 Evaluate applicability and usefulness of climate models for water supply operations & planning SEASONAL PREDICTIONS Diagnose skill of NMME seasonal precipitation and temperature forecasts and their utility for forecasting seasonal streamflow in Florida LONG TERM CLIMATE PROJECTIONS Evaluate ability of reanalysis data and retrospective GCM output to reproduce current climate and hydrologic patterns, and implications of future GCM projections on climate and hydrologic patterns SEA LEVEL RISE Evaluate salt water intrusion and coastal flooding risks for a suite of sea level rise predictions

6 Tampa Bay Water Project Objectives! Evaluate the ability of GCM retrospective predictions to reproduce observed temperature, precipitation and reference evapotranspiration in the Tampa Bay region! Evaluate the ability of downscaled retrospective GCM predictions to reproduce historic hydrologic behavior when used with Tampa Bay Water s Integrated hydrologic model! Evaluate changes in hydrologic behavior that result from GCM future projections! Evaluate impact of future climate scenarios on future water supply availability in the Tampa Bay region

7 Long-term Climate Projection Analysis Framework General Circula/on Models reanalysis data, retrospec,ve predic,ons, future projec,ons 2~3km resolu,on Dynamic Downscaling RSM 1 km resolu,on BCSD,SDBC, BCCA,BCSA 12 km resolution Sta,s,cal Downscaling Evalua,on of climate informa,on over Florida Integrated Hydrologic Model Applica/ons Impact assessment

8 Phase 1: Hydrologic Implications of Dynamically Downscaled Climate Predictions! What we did Used dynamically-downscaled bias corrected retrospective and future climate projections (CMIP3) to evaluate potential impacts of future climate change on hydrology in the Tampa Bay region! Why we did it Want to understand implications of future changes in temperature and precipitation over Tampa Bay region on long term water supply planning! What we found

9 Validation of the Downscaled Retrospective GCM Output for Streamflow Prediction Key Finding Historic Streamflow seasonal cycles preserved using retrospectiv e GCM predictions Mean streamflow (m 3 /s) Mean streamflow (m 3 /s) Obs (1989-) Cal (1989-) CCSM_retrospec,ve_BC HadCM3_retrospec,ve_BC GFDL_retrospec,ve_BC Alafia River Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Obs (1989-) Cal (1989-) CCSM_retrospec,ve_BC HadCM3_retrospec,ve_BC GFDL_retrospec,ve_BC Hillsborough River Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

10 Temperature Dynamic Downscaling Climate Modeling Results obs_tmax Retro. _Tmax Future_Tmax obs_tmin Retro. _Tmin Future_Tmin mean temperature ('C) 15 Bias-corrected results T max T min CCSM mean temperature ('C) 15 T max T min HadCM3 mean temperature mean temperature ('C) ('C) T max T max T min T min GFDL GFDL 5 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 5 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 5 Jan Feb Mar Apr Ma Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Key Findings: 1. Global Climate Models reproduce seasonal cycles of observed mean monthly temperatures 2. Global Climate Models predict about 1 to 3 degree C increase in average monthly Temperature for the future period

11 Precipitation Dynamic Downscaling Climate Model Results Bias-corrected results 8 mean precipiatation (mm) obs Retro. _BC Future_BC CCSM obs Retro. _BC Future_BC HadCM obs Retro. _BC Future_BC GFDL Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Future projections for rainfall vary considerably in the summer rainy season

12 Evapotranspiration Hydrologic Modeling Results Evapotranspiration (mm/year) Annual average ET (mm/year) CCSM HadCM3 GFDL ET/P (%) ET frac;on (ET/Precip) CCSM HadCM3 GFDL th_BC BC delta 2th_BC BC delta 2th_BC 2th_BC BC delta 2th_BC BC delta 2th_BC BC delta BC delta 21st 21st 21st cal. CCSM GFDL HadCM3 21st 21st 21st cal. CCSM GFDL HadCM3 IHM predicts lower actual ET, but higher ET/P ratios for low rainfall scenario (CCSM) system becomes water rather than energy limited

13 Future Predictions of Streamflow (Alafia River station) CCSM Very low river flow predictions 3 Mean streamflow (m3/s) 2 1 Future simulations Obs (1989-) Alafia River Cal (1989-) CCSM_21st_BC1 CCSM_21st_delta Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Mean streamflow (m3/s) Streamflow Change (Future-retro.) CCSM_21st-2th_BC1 CCSM_21st-2th_delta Alafia River Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec HadCM3 About the same as today Mean streamflow (m3/s) Obs (1989-) Cal (1989-) HadCM3_21st_BC1 HadCM3_21st_delta Alafia River Mean streamflow (m3/s) HadCM3_21st-2th_BC1 Alafia River HadCM3_21st-2th_delta Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec -2 GFDL Higher late spring and summer flows Mean streamflow (m3/s) Obs (1989-) Cal (1989-) GFDL_21st_BC1 GFDL_21st_delta Alafia River Mean streamflow (m3/s) GFDL_21st-2th_BC1 Alafia River GFDL_21st-2th_delta Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec -2

14 Phase 2: Hydrologic Implications of Statistically Downscaled Climate Predictions! Statistically downscale CMIP5 retrospective and future predictions for 1-15 GCMs and 3 RCP trajectories! Quantify the uncertainty in future temperature, precipitation and reference ET projections for the Tampa Bay Region among the GCMs and RCPs! Estimate agricultural and urban irrigation demand projections for retrospective vs future climate in the Tampa Bay Region! Evaluate potential impacts of climate change on future water supply availability in the Tampa Bay region

15 Step 1: Comparison of Statistical Downscaling Methods! What we did Developed a new statistical downscaling method (BCSA) and compared it to existing methods (BCSD, SDBC, BCCA)! Why we did it Existing statistical downscaling methods did not reproduce spatiotemporal rainfall characteristics in Florida very well! What we found Choice of statistical downscaling method matters in Florida.

16 Wet season average daily rainfall Wet season standard deviation of daily rainfall Wet season spatial correlation structure variogram (mm2) Gobs BCSD_daily SDBC BCSA distance (km)

17 Monthly average streamflow Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Streamflow (m3/s) BCSD_GCMs Obs. Cal. Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec SDBC_GCMs Obs. Cal. Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec BCSA_GCMs Obs. Cal Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Streamflow (m3/s) BCSD_GCMs Obs. Cal. Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec SDBC_GCMs Obs. Cal. Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec BCSA_GCMs Obs. Cal. Alafia River Cypress Creek BCSD SDBC BCSA BCSD SDBC BCSA

18 Frequency of daily streamflow events # of events per year # of events per year Alafia River Obs. Cal. BCSD_GCMs SDBC_GCMs BCSA_GCMs Streamflow (m3/s) Cypress Creek Obs. Cal. BCSD_GCMs SDBC_GCMs BCSA_GCMs Streamflow (m3/s)

19 Step 2: Evaluate uncertainty of CMIP5 P, T and ET projections What we did Evaluated uncertainty in P, T, and ET projections using Global Sensitivity Analysis and Monte Carlo Filtering Why we did it To develop an appropriate ensemble of GCMs, ET methods and RCP trajectories for evaluating future climate change What we found Choice of ET method matters! Evaluate impacts of future projections over an ensemble of GCMs and a variety of ET methods and RCP trajectories

20 CMIP5: Mean and Std Dev of Projected Monthly Averages Mean Precipitation (mm/day) P 6 ET Retrospective Future1 Future2 Mean Evapotranspiration (mm/day) Retrospective Future1 Future Month Month Retrospective Future1 Future2 P-ET Mean P - ET (mm/day) Month

21 Drivers of Uncertainty Precipitation Evapotranspiration Florida P-ET SouthWest P-ET Blue: uncertainty due to GCM, Green: uncertainty due to RCP scenario, Red: uncertainty due to ET method. Solid line 23-26, Dashed line 27-21

22 27-21 Change in Annual P-ET by ET method (averaged over GCMs and RCPs) Mean relative change of P-ET with PM (Fut2) Mean relative change of P-ET with Hargreaves (Fut2) Penman- Monteith Latitude Latitude Hargreaves Mean relative change of Lonitude P-ET with Blaney-Criddle (Fut2) Mean relative change Lonitude of P-ET with Irmak-Rn (Fut2) Blaney- Criddle Latitude Latitude Irmak- Rn Mean relative change Lonitude of P-ET with Irmak-Rs (Fut2) Mean relative change Lonitude of P-ET with Hamon (Fut2) Irmak- Rs Latitude Latitude Hamon Mean relative change Lonitude of P-ET with Dalton (Fut2) Mean relative change Lonitude of P-ET with Kharrufa (Fut2) Dalton Latitude Latitude Kharrufa (PET) Mean relative change Lonitude of P-ET with Meyer (Fut2) Mean relative change of Lonitude P-ET with Priestley-Taylor (Fut2) Meyer Latitude Latitude Priestly Taylor (PET) Lonitude Lonitude -.8

23 Current Work! Estimate agricultural and urban irrigation demand projections for retrospective vs future climate for a variety of ET methods in the Tampa Bay Region! Develop future population, land use and water use scenarios for the Tampa Bay Region! Evaluate future impacts of population, land use, water use and climate change on future water supply and water demand in the Tampa Bay region

24 Group Learning to Date! Choice of downscaling technique matters for water supply planning in Florida! Choice of ET method matters for hydrologic model predictions! Precipitation and ET differences propagate nonlinearly through hydrologic system! Must have local/regional hydrologic models to understand changes in hydrology due to climate! Regional actionable information is difficult!

25 Community Building Lessons Participants à Engagement à Outcomes à Process q Support a Shared Interest q Useable climate information at locally relevant space and time scales q Passion and real need for the information q Get the right people at the table q Energized core group and internal leadership q Different stakeholders engaged q Outreach to new participants q Manage Diversity, Enhance Communication q Value diversity of individuals and institutions Community q Use variety of activities to challenges and opportunities Domain q Respect evolving agendas and learning/communication styles Practice

26 Community Building Lessons Participants à Engagement à Outcomes à Process q Provide Rigorous Science q Reliable predictive tools and evaluate practical applicability q Forecast skill: can we trust the climate information? q Frame the science for various publics q Understand User Perspective and Context q Consider organizational context to understand how climate information adds value to decision making q Sharing case studies about the systems of others contributes to understanding how decisions are made q Ensure Sustainability q Provide added value- balance goals to keep people interested q Build Identity/ownership q Get individual & institutional commitment for time and funding

27 Questions. Comments?