1. Needs of water resources decision-makers for decadal climate predictions Andrea J. Ray NOAA Earth Systems Research Lab & NOAA- CIRES Western Water Assessment CLIVAR Science Symposium Irvine, CA 14 July 2008

2. Overview Rich arena for applications: Decadal scale of many water & natural resource decisions Who studies these A couple of examples: Decision and planning processes and use of information USBR reservoir management Powell-Mead “Shortage Sharing” EIS Front Range Municipal Water planning Types of uses and how this relates to needs Potential for use in climate risk management ….and adaptation? Importance of boundary organizations, e.g. RISA’s, IRI, etc for ongoing interactions with users Uses of decadal information: needs, users, and uses

3. Societal impacts in several sectors water (reservoir management, municipal water supply), fire management, public health, agriculture -- esp permaculture, drought mitigation/planning Management communities who can take advantage Targets for user-oriented experiments, training/education Planning, their scenarios, hedging Interest in climate change, not familiar with decadal variability Skill… Not as simple as threshold “level required to be societally useful” Shifts in risks of extreme events vs specific events forecasted Rich arena for applications

4. Decadal scale of many water & natural resource decisions Water -- UCBR/reservoirs & municipal water supply Glen Canyon Dam Adaptive Management Program Forest Management Plans, National Parks Endangered species recovery plans Upper Colorado endangered fish Multi-species conservation plan for the Lower Colorado Salmon Public health Heat waves; Air quality; risk of temperature-related diseases Many different ongoing planning processes Opportunities for 2-way learning and new information in each new process

5. Who are the Water Resources “Users” of information? Municipal/residential and industrial water suppliers & organizations Agricultural water users and organizations Government managers, regulators, policymakers, planners (local, state, federal) Professional organizations & networks of all of these Scientists and engineers Providers of products and services (govt, pvt, media) NGOs (e.g., biodiversity interests) Recreation interests, individual and companies Boundary organizations, which work between scientists and users where do these water managers talk to each other

6. Regional Integrated Science & Assessments (RISA): one set of boundary organizations Eight regional projects, US and border focus, earliest began ~1995; primarily empirical studies Mechanisms to elicit and understand user needs Perception, cognitive, communications studies Integrate and synthesize needs across groups Determine what services should be: part of a dialogue about risks Take advantage of social science studies of cognition, adoption and diffusion of innovations, and methodologies Focus on users’ problem orientations: drought, hydropower, multi-purpose reservoir management; long-term planning; annual planning Decision studies of water management and agriculture Characterize decisions and decisionmakers Institutional/legal Organizational/behavioral Experiments in communicating with stakeholders and in creating and sustaining partnerships over time Reservoir management, drought task forces, climate change and state water supply planning

7. Example of long term planning and climate: Drought impacts on Lake Powell 2007 water year runoff into Lake Powell was 51%average, part of long term drought Demands on the river are increasing Risk of call on the river? Upper Basin States (CO, WY, UT, NM) may soon be required to cease water diversions that are junior to the 1922 Colorado Compact in order to meet obligations to downstream users Years-decades to fill/re-fill: concern about multi-year drought What if climate change reduces flows on the River? From Harding, 2006, www.hydrosphere.com

8. - USBR Long term planning, evolution of operating criteria Policy landscape provides opportunities to incorporate climate information - “Shortage sharing agreement,” Environmental Impact Statement and EIS for the Aspinall Unit on the Gunnison River - Ongoing implementation of endangered species recovery plans (MSCP, CRRIP) and Glen Canyon Adaptive Management Program Opportunities

9. Reservoir Management uses of information Powell-Mead “Shortage Sharing” EIS Latest in a series of management innovations Most extensive use yet of climate information Paleo record to represent richer range of droughts “Appendix U” coordinated thru RISAs Decadal information is critical to management scale Climate risk management ….and adaptation? More than just the right products, and occurs in a dialogue about risks Understanding the nature of risk and information/knowledge needed to manage risk Managing water in the context of changing climate -- adaptation strategies

10. Front Range Municipalities Study Uses of climate information & forecasts Factors affecting the use of climate information and forecasts Six Front Range water providers: Northern Colorado Water Conservancy District, Boulder, Westminster, Denver, Colorado Springs, and Aurora Serve about 60% of Colorado’s population Context: Interactions with WWA and other climate information providers since 1998 Drought in 2002 Different contexts for growth, water supply reliability; use change, etc Northern Boulder Denver Westminster Colorado Springs Aurora

11. Perspectives from user studies: What do they want? Historical data & projections of these at a basin/sub-basin scale Snowpack/SWE Soil moisture Streamflow current/forecasted Timing of spring peak; “holes” in a river (low flows) Reservoir levels Ground water Surface water supply index (SWSI) Palmer Drought Index Temperature Evapotranspiration, evaporative losses Demand metrics, water and hydropower Outlooks of these, and how ppt and temp outlooks relate, e.g., ppt needed to raise levels to near-average or other thresholds

12. Longer term municipal water planning Drought as part of longer term planning, beyond drought of record -- paleo Assess the potential for future systems to cope with drought: streamflows from the historic record Planning for projects to “firm-up” yield Windy Gap surpluses from early 90’s, but none since -- decadal variation Other supply options Demand projections: primarily population based Temperature trend not considered Several agencies now using paleoclimate reconstructions to expand the types of drought they evaluate Interest in assessments: range of potential climate change scenarios, droughts that have occurred outside the instrumental record

13. Perspectives from across user studies: Users needs:Longer-range questions Increasing requests for information on interannual and decadal time scales (+ 5, 10, 15 years) Reservoir inflows over several years (at least 2) Drought outlooks over the next decade Are the historical droughts of record still a valid planning tool? Are return periods for flooding still valid? “Can we produce reliable baselines for planning to give the large amount of year to year and decade to decade variations?” “Are the assumptions of planning borne out under projections of varying and changing climates?” e.g. 1906 Rio Grande treaty definition of “extra-ordinary drought” invoked 14 times over the last 50 years “Do present simulations of change adequately represent modes of variations (ENSO, NAO, PDO etc.) ?” How might ENSO change under climate change

14. Types of “use” Consult: the product is consulted, e.g., looked up on a web page or received as a briefing or from other source (type1) Consider: after consulting the product, it is considered in management deliberations as a factor potentially influencing decisions, but not necessarily in operational models (type 2) Mental models, judgement, experience. Projections/forecasts may be used in this way when they are not in appropriate forms for use in operational models Incorporate: some form of the forecast is incorporated into an operational model that is utilized in operational decisions (type 3) May be objective Dialogue about risks: communication of risk, i.e. the forecast is used to communicate with other managers and stakeholders about the risk of certain conditions and about the need to take actions, or to justify actions (type 4) Water resource decisionmakers use climate information (e.g. during drought) in a dialogue with their stakeholders about the risk of low inflows, flooding, e.g., and the need to take actions, or to justify actions

15. What’s skill got to do with it?* Depends on…. The decisions, what’s being forecast often concern is about risk of events, and planning to mitigate, avoid, skill of other factors in a decision or planning process The level of climate literacy of the user, and how familiarity with forecasting () High “threshold,” e.g. 75% from Pulwarty & Redmond ‘97 -- true -- but an early stage in understanding for most users Better than climatology Improve on historic record as a planning tool for extremes?

16. Long range planning by agencies at multiple space and time scales Federal: USBR (reservoir plans), USFS (forest management plans), Drought/NIDIS (mitigation planning) State water planning Regional and local govts (e.g. Denver Water) Support water managers and planners’ dialogues with their own stakeholders Dialogue about climate-related risks with policy and planning for 20- 50 year horizons long-lived policies likely to encounter multi-year droughts and impacts of observed trends Synthesis of research into products & analysis that connect climate impacts to water management impacts: Temperature --> evaporation, rain/snow mix, urban demand, length of growing season Timing of spring runoff (Dettinger, Cayan) --> water rights, reservoir reliability Synchroneity (Hamlet, Jain) --> diversity of supply sources Interest in these for the Shortage Sharing EIS for Lakes Powell and Mead Uses of decadal information I: who, for what, how

17. Uses of decadal information II: who, for what, how Reducing vulnerability to climate variations requires consideration of a range of climate scenarios in planning and policy development multi-year droughts and the impacts increased temperatures Support for 20-50 yr planning horizons Capability to view and compare information from multiple sources Need user-oriented metadata, descriptions GIS widely used by resource agencies, state/local planners, but climate information has not generally provided by NOAA in these formats Connect types of data and projections Work with Integrated Assessment groups, including RISAs connect with specific user groups elicit and understand detailed user needs and common needs have long-term partnerships and experiment with communicating information and aboutrisk Understand the pathways that are used for information

18. Decision Support III: Long-range planning and policy Policy-relevant Science questions changes in snowpack, accumulation season, timing of spring runoff, increases in water demand from temperature increases changes in the length of the growing season Impact of temperature trend alone (most skill): ET, drought indices, soil moisture, Changes in ENSO with climate change Changes in the risk of extreme events: Drought, runs of dry years, food risk, severe storms Cold air outbreaks, heat waves Disconnect between the scientific literature and information for managers: Need for synthesis: Panel of experts to contribute to the Lower Colorado office of Reclamation contribute to EIS(other RISAs participating) Model for use of decadal information: engaging users, information publically available as a peer-reviewed report, education of water resources users Potential for PI-based, peer-reviewed process to produce information for water management?

19. Thank you Andrea J. Ray, Ph.D NOAA Earth Systems Research Lab andrea.ray@noaa.gov

20. What is potentially predictable Spectrum of User Needs Current Forecast Products Relationships among current products, potentially predictable, and needed climate information

21. How drought information might be used: WWA observations “Conversation” within water management groups and with their stakeholders, and with scientists Mental models of managers for their systems are important as well as hydrologic and management models Relationship of information to their triggers, thresholds As interested in the information behind the Drought Monitor as the DM itself, in order to make their own assessments Synthesis of research into products & analysis that connect climate impacts to water management impacts: Temperature --> evaporation, rain/snow mix, urban demand, length of growing season Timing of spring runoff (Dettinger, Cayan) --> water rights, reservoir reliability Synchroneity (Hamlet, Jain) --> diversity of supply sources

22. Issues for a policy-relevant “Mountain Hydroclimate” science enterprise Data needed for management, calibrate, verify, initialize models Support observations and data management, and a coherent policy for the multi-agency system that supports hydroclimate data collection NWS data, USGS streamflow data, and National Resource Conservation Service snow and soil moisture data among many others Metadata too! Policy-relevant science questions require synthesis and “assessment” Individual PI-driven science only partly meets needs – need interconnected projects Unfortunately, to date, scientific assessments like the IPCC have focused on the global and continental scale effects of climate change and hence are of limited use to regionally focused decision-making. Regional, regional, regional….. Actually communicating with policymakers is a whole different talk….

23. Concluding thoughts Changing context has introduced criticality for water management of the Colorado Basin Sensitivity to “drought” increasing even if no change in hydrology Reducing vulnerability requires consideration of a range of climate scenarios in planning and policy development multi-year droughts and the impacts increased temperatures Support for 20-50 yr planning horizons Water Providers, Environmentalists, NGOs need to be partners in creation of science Embedded researchers; better job on educating these partners Beyond forecasts to an additional category of services: connecting climate impacts to water management impacts in a dialogue with managers to support decisions “assessments” on science by multiple PIs with policy focus/parter How does a process-oriented, PI-driven program participate and contribute to the larger service?? Synthesis projects and X-RISA efforts are a start

24. Thank you Andrea J. Ray, Ph.D NOAA Earth Systems Research Lab andrea.ray@noaa.gov

25. Decadal “needs” Integrated  Implications of climate change temperature trend  Support use of “scenarios” in water resource planning -- what ifs, and how resilient is the system, vs prediction  Importance of boundary organizations

26. Decadal “services” ***

27. Forecast, hydrological assessment Resource Assessment AOP Annual Operating Plan Water and other Agencies, State and Federal Multi-Resource Coordination & Management Working Groups Scales Hydrological decision support model with climate forecasts Individual agency responsibilities Interaction of primary focus with other issues Hydrologic Scenarios 1. Likely April-July inflow volume 2. Monthly release schedules for hydropower and irrigation planning 3. 4. Flood control 5. Minimum flows 6. Other resource conditions [Issues] 1. Recreational trout spawning 2. Recreation management, on and around reservoir 3. Irrigation planning 4. In stream flow conditions 5. Maintenance scheduling 6. Legal obligations, I.e., interstate compacts 7. Other resource conditions [Scenarios] 1. Peak flow enhancement opportunities 2. Time since last target peak flows 3. Effects of hydrologic scenario on other resource conditions 4. 5. Political issues Method: use HDSS and climate info to discover how diff climate scenarios are stressors on the evolving system Current method: hydrologic decision support model Next steps: 1) process of negotiation to develop better operational products; Method: Institutional analysis: who’s making decisions about resource allocation

28. New Water Managers’ criteria  Operational Forecast criteria:  Value  Consistency  Quality  Murphy, 1993  Decision making criteria:  Credibility of provider and reliability of information  Accessibility of information, compatibility, complexity  Legitimacy  Participation of water managers’ stakeholders in plans and decisions about operations, and decisions influenced by acknowledgement of interdependence

30. Thank you! Andrea J. Ray NOAA Earth Systems Research Lab; Western Water Assessment, Andrea.ray @ noaa.gov

31. What’s needed for decision support Beyond forecasts --> “Services” Dialogue about climate-related risks with policy and planning for 20- 50 year horizons Not forecasting for these horizons, but long-lived policies likely to encounter multi-year droughts and impacts of observed trends Synthesis of research into products & analysis that connect climate impacts to water management impacts: Temperature --> evaporation, rain/snow mix, urban demand, length of growing season Timing of spring runoff (Dettinger, Cayan) --> water rights, reservoir reliability Synchroneity (Hamlet, Jain) --> diversity of supply sources Interest in these for the Shortage Sharing EIS

32. Decision Support II Support year to year decisions on efficient management of storage and releases More important for Lower Colorado below Lake Mead and reservoirs above Lake Powell Seasonal and sub-seasonal forecasts Office of Hydrology (Schaake talk Thurs. p.m.) U.Washington group (Lettenmaier, Hamlet, Wood) Potential improved subseasonal streamflow forecasting Educate and support USBR and other water managers about these new products as they become available Partner to ensure that these products are compatible with USBR decision support models and frameworks

33. Conclusions Changing context has introduced criticality for water management of the Colorado Basin Increased risk of shortages due to increasing demand Shortages related to climate variability and change likely to have greater impact -- increased vulnerability in the system Points of vulnerability similar to SSD conclusions Reducing vulnerability requires consideration of a range of climate scenarios in planning and policy development multi-year droughts and the impacts increased temperatures Support for 20-50 yr planning horizons Services beyond forecasts as an additional category of services: connecting climate impacts to water management impacts in a dialogue with managers to support decisions

34. What do they want? Longer-range questions Increasing requests for information on interannual and decadal time scales (5, 10, 15 years into the future) Reservoir inflows over several years (at least 2) Drought outlooks over the next decade “Can we produce reliable baselines for planning to give the large amount of year to year and decade to decade variations?” “Are the assumptions of planning borne out under projections of varying and changing climates?” e.g. 1906 Rio Grande treaty definition of “extra-ordinary drought” invoked 14 times over the last 50 years “Do present simulations of change adequately represent modes of variations (ENSO, NAO, PDO etc.) ?”

35. Blank slide e Spatial National Crop Regional Requesting

37. WWA timeline 1997-98 2005-6 1999 Climate context El NinoLa Nina2002 Drought Climate Change Interest 1st WWA Workshop Reservoir mgr workshops WY05 Outlook Briefing SLC Drought rapid response 2000-12002

38. Differences in perspective: scientists and managers FactorScientist’s perspectiveWater Manager’s Perspective Identifying a critical issue Based on a broad understanding of the nature of water managementBased on experience of a particular system Time frameVariableImmediate (operations) Long-term (infrastructure) Spatial resolutionDefined by data availability or fundingDefined by institutional boundaries or authorities GoalsPrediction Explanation Understanding of natural system Optimization of multiple conditions and minimization of risk Basis for DecisionsGeneralizing multiple facts and observations Use of scientific procedures and methods Availability of research funding Disciplinary perspective Tradition; Procedure Professional judgment; Training Economics; Politics Job risks ExpectationUnderstanding Prediction Ongoing improvement (project is never actually complete) Statistical significance of results Innovations in methods/theory Accuracy of information Appropriate methodology Save money and time; Protect the public; Protect their jobs, agendas or institutions Product Characteristics Complex Scientifically defensible As simple as possible without losing accuracy Importance of context FramePhysical (atmospheric, hydrologic, etc.) conditions as drivers Dependent on scientific discipline Safety and well being Profit Consistency with institutional culture, policy, etc. Nature of UseConceptualApplied

39. Some conclusions from across user-studies projects: needs for water-related decisionmaking? Scientists need to collaborate with these sophisticated, but non-climate experts in a common language Variables and indices flexible formats, areas, time scales tools to relate observations, historical data, and forecasts to water managers perspectives, e.g. to their problems Ways to evaluate climate scenarios in their management scenarios Tools for managers to talk to their stakeholders Benchmarks beyond “idealized value” Partnerships Interactions maintained over time Influence of scientists on the drought planning process and of water managers on science done Innovation in both science and management from interaction Fora for communication, learning, bringing perspectives together

40. Current uses of climate information in municipal water management Use of the instrumental record of hydro-climate variables in planning and operations models The use of climate influenced hydro-climate parameters to generate projections of streamflow, reservoir contents, or water supply SWE, historic records of streamflow, water year precipitation Use of paleoclimate data, e.g. reconstructions of SWE or streamflow Use of forecasts of climate variables, e.g., precipitation or temperature, such as the NOAA/CPC Monthly and Seasonal Climate Forecasts, or medium-range weather forecasts  Climate variability reflected in annual and longer term operations in ways other than use of forecasts

41. USBR Long term planning, evolution of operating criteria However, in the event of a shortage, vague and often contradictory laws and policy mandates Non-allocated uses, e.g. recreational and environmental are particularly sensitive, yet these have increased in economic importance 2007 policy landscape provides opportunities to incorporate climate information: Ongoing implementation of endangered species recovery plans (MSCP, CRRIP) and Glen Canyon Adaptive Management Program Shortage sharing agreement,” EIS and the Environmental Impact Statement for the Aspinall Unit on the Gunnison River Opportunities