This theme is designed to provide resource managers with the tools, data, and information about climate that are necessary to support management decisions. Projects within this theme would incorporate collaborative work with decision makers to develop, interpret, and apply climate information. Addressing topics related to understanding, explaining, and resolving issues related to uncertainty are important. In most cases, water will be the ‘natural resource’ of note. This theme is an extension of work that WWA has been pursuing for many years, especially in the Colorado River Basin. We anticipate many continuing projects will continue to evolve within this framework.
BACKGROUND INFORMATION ON THE COLORADO RIVER
COMPLETED PROJECTS
CURRENT RESEARCH & ASSESSMENTS
Reclamation Review of Stochastic Streamflow Simulation at Interannual and Interdecadal Time
Impacts of Coupled Climate Change and Dust Deposition on Water Resources in the Colorado River
Basin
Paleohydrology of the Lower Colorado River Basin
Identifying Lessons from the Murray–Darling Basin Potentially Applicable to the Colorado River
Reconciling Colorado River Flow Projections
Colorado River inflows between Lakes Mead and Powell: Past, Present and Future
Reclamation Review of Stochastic Streamflow Simulation at Interannual and Interdecadal Time
Scales and Implications to Water Resources Management
Primary Investigators: B. Rajagopolan, E. Zagona, K. Nowak, C. Bracken
Contributors: B. Udall
The main work on the forecasting component involves expanding the seasonal forecasting framework to produce "second–year" forecasts to drive the Probabilistic Midterm Model. Currently the Colorado Basin River Forecast Center provides seasonal forecasts for the first year of the 24– Month Study. The second year of the 24–Month Study uses climatology to drive the model. The "second–year" forecasts will use a new disaggregation method (Nowak et al., 2009) and knowledge of past observed flows (including paleo–reconstructed flows) that intelligently predict second–year behavior, to provide better operational guidelines to water managers. We are working on a couple of different ideas involving Markov Chains to better simulate the ‘second year’. We will validate and test these methods.
We will also test the forecasting methodology in Bracken et al. (2010) with the new disaggregation method (Nowak et al. 2010; see below) to generate streamflow forecasts at all the Upper Basin locations. Bracken et al. (2010) demonstrated the methodology at four locations.
Development of the Probabilistic Midterm Model will continue with a goal of completion during the summer of 2010. Investigation into improved operating strategies that take advantage of improved second year forecasts will follow the completion of the model.
We have developed a method that simulates daily streamflow at multiple locations simultaneously from a single annual flow at an aggregate location – via K–NN resampling of daily flow proportion vectors. The procedure is simple, data–driven and captures observed statistics quite well. A paper entitled "A Non–parametric Stochastic Approach for Multisite Disaggregation of Annual to Daily Streamflow" has been accepted for publication in Water Resources Research. This paper contains the details of this technique as applied to the San Juan River Basin in southwest Colorado.
Traditional streamflow simulation techniques are not well suited to simulate low frequency behavior, such as long wet and dry periods, which are critical for robust evaluation of system risk and consequently for long–term planning. To this end, we have developed an improved wavelet–based simulation technique that has the ability to capture the spectral properties and thus, the long wet and dry periods much more faithfully. We have coupled this to the disaggregation method described above, resulting in a robust multi–site streamflow simulation approach. This approach has been validated and we are now drafting a submission to Water Resources Research.
Generating realistic streamflow scenarios at interdecadal time scales involves two key steps – (1) understanding the variability and (2) using this information into a simulation tool. We have addressed (1) by performing a systematic time domain (PCA–based) and spectral domain (wavelet– based) analysis of basin–wide natural streamflow and paleo–reconstructed streamflow at Lees Ferry. We find that the dominant modes of streamflow variability are strongly linked to large–scale climate features such as AMO, PDO and ENSO. Furthermore, we find that the paleo–reconstructed flows exhibit strong variability at interdecadal time scales that appear to be modulated at multi–decadal time scales. The results from these analyses are being documented for submission to a climate journal such as Journal of Climate.
The two papers mentioned above will be completed and submitted to journals.
We will develop a stochastic simulation tool that will use the spectral variability of the paleo– reconstructed streamflows. The idea here is to find the ‘current state’ of the streamflow system in the spectral domain and use ‘spectral analogs’ from long paleo–reconstructed streamflows to generate 1–2 decades of streamflow scenarios into the future. The steps for this approach have been formulated and we are in the midst of implementing and validating this. Once complete, we will write this up for submission to Water Resources Research.
The most important component of the project is the evaluation of these simulation capabilities in water resources management in the Colorado River Basin. To accomplish this, we plan to drive the Reclamation CRSS model with a suite of streamflow simulations capturing various regimes (e.g., persistent wet, persistent dry, climate change, etc.) and evaluate the impact on a suite of key management variables. Based on these outputs, we will develop potential management options that can mitigate the risk impact of the flow variability. We hope this will be of use to water managers as they devise long–term operational and management strategies.
Back to Top
Impacts of Coupled Climate Change and Dust Deposition on Water Resources in the Colorado River
Basin
Primary Investigators: J. Deems, T. Painter
Contributors:
C. Landry (Center for Snow and Avalanche Studies)
In this project, we examine the range of changes to water yield in the upper CRB by coupling potential climate change scenarios and dust deposition scenarios. We are using the Variable Infiltration Capacity (VIC) macroscale hydrologic model to simulate the snow cover and melt across the basin, and the naturalized hydrograph and annual CRB water yield at various gages in the CRB culminating at Lees Ferry, AZ, for the historic period of 1915–2008, and for three future time periods: 2010–2039, 2040–2069, and 2070–2100. Our initial modeling indicates that the increased dust loading (represented by a modified snow albedo parameterization) has shifted the hydrograph earlier by about 30 days, and decreased annual water yield in the upper basin (above Lees Ferry, AZ) by an average of 5%, or 0.8 million acre–feet (MAF), since significant dust deposition began (Painter et al., 2010), with a total loss of yield of 70 MAF across 1915–2003. This year, we will begin to convolve the future climate scenarios with all dust scenarios, by permuting BDL, ADL, and EDL dust scenarios with the six future climate forcing datasets. Of particular interest will be the complex interactions of greater dust loading earlier in and throughout the spring, less snow water equivalent upon which dust can effect change, and enhanced ET and soil moisture loss during dust–forced snow–free periods subject to a warmer climate.
Back to Top
Paleohydrology of the Lower Colorado River Basin
Primary Investigators: B. Rajagopalan, J. Lukas
Contributors:
C. Woodhouse and D. Meko (Univ. of Arizona), J. Salas (Colorado State Univ.), D. Kanzer (Colorado River District)
The scope of this project, then, is the generation of paleohydrologic reconstructions of annual flows for the intervening flow between Lees Ferry and Imperial Dam ("Imperial") and the Gila at its mouth. The overall project objectives are:
- Develop robust paleohydrologic reconstructions (>400 years long) of the total Lower Colorado River Basin streamflow, commensurate with existing reconstructions of Upper Basin streamflow (e.g., Meko et al. 2007)
- Compare multiple reconstruction approaches (both parametric and non–parametric) to assess the robustness of each approach, and the sensitivity of the results to the chosen approach
- Use the reconstructions in basinwide water–balance modeling to assess the risks of the paleo– derived variability to CRB water supplies (sensu Rajagopalan et al. 2009)
Back to Top
Identifying Lessons from the Murray–Darling Basin Potentially Applicable to the Colorado River
Primary Investigators: D. Kenney
Contributors:
R. Pulwarty (NIDIS)
Several elements of the MDB institutional environment are potentially relevant as a basis for comparison and insight, including: the use of basinwide planning processes and/or river basin commissions; the relationship between States and the national (federal) government (especially as it involves decisions about interstate allocation); the specification of water rights/entitlements among users, sectors, and jurisdictions (including indigenous/native communities) at both the interstate and sub–state scale; the use (and limits) of markets for water reallocation; the mechanisms used to ensure adequate flows for environmental purposes; the role of the government is supporting/shaping the role of the agricultural sector (both historically and looking forward); and the roles of key climate information providers. In this project, these elements of the MDB institution will be reviewed and compared to the Colorado River Basin, with the aim of identifying transferrable lessons.
Back to Top
Reconciling Colorado River Flow Projections
Primary Investigators: D. Cayan, T. Das, D. Lettenmaier, J. Overpeck, H. Hartmann, R. Webb, M. Hoerling, B. Udall
Contributors:
J. Barsugli, J. Eischeid
- Evaluate alternative data sets; extend to current year and quality control.
- Diagnose reasons for differences in temperature sensitivities among models.
R
- Reconciliation of climate scenarios.
- Evaluate Colorado River discharge sensitivities from NARCCAP runs.
- Continued diagnosis of the role of high elevation feedback on projected future sensitivities. Preliminary analysis of AR5 GCM scenarios.
Colorado River inflows between Lakes Mead and Powell: Past, Present and Future
Primary Investigators: J. Barsugli
I propose to coordinate a workshop that focuses on the water supply aspect of these flows and that brings together scientists working on this issue with the goal of reconciling the data and modeling assumptions used to estimate the historic flows, and to discuss the potential for climate change impacts. The goal of the workshop is to lead to a publication that summarizes the state of knowledge on these flows so that this information. In addition to coordinating, I would like to research the relation of historical climate variability to the observed flows.
Back to Top
Current Projects home
