Sensitivity of Hydrologic Impacts Assessment to Downscaling Methodology and Spatial Resolution
Project ID: 1646
Principal Investigator: Tom Pruitt
Research Topic: Water Resource Data Analysis
Priority Area Assignments: 2011 (Climate Change and Variability Research), 2012 (Climate Change and Variability Research), 2013 (Climate Change and Variability Research)
Funded Fiscal Years: 2011, 2012 and 2013
Questions relate to Climate Change Research Need "2. Obtain Climate Projection Data - Improved understanding on the strengths and weaknesses of downscaled data and data-development methods."
1. Is the portrayal of hydrologic impacts under climate change dependent on the chosen downscaling method (i.e., dynamical downscaling using regional climate simulation versus nondynamical downscaling using statistical or empirical methods)?
2. At what space and time scales do the impacts portrayals begin to be sensitive to methods class, and for what types of hydrologic metrics?
3. How do the responses to these questions vary with the spatial resolution of dynamical downscaling and the adequacy of process representation in atmospheric and hydrologic models?
For impacts assessments under climate change, the planning community has access to multiple methods for spatially downscaling outputs from global climate models (GCM). Methods are often divided into two classes: dynamical and nondynamical. Both classes of methods have been applied to develop downscaled climate projections datasets, which are available online to support planning activities. When choosing a method class, planners have two competing objectives: (1) choose a method that provides better representation of physics and (2) use a method that permits downscaling of many global climate projections over a large geographic domain during a long future period supporting time-flexible adaptation planning (e.g., period spanning 20th to 21st centuries). For (1), dynamical methods are preferred because nondynamical methods feature questionable assumptions about the stationarity of relationships between local climate and large-scale climate predicted by a GCM. For (2), nondynamical techniques are much more feasible given the computational costs associated with (1).
Need and Benefit
Reclamation, the U.S. Army Corps of Engineers (Corps), and other water management agencies need to understand the best methods for incorporating climate change information into longer-term water resources planning (U.S. Geological Survey [USGS] Circular 1331). In such planning assessments, assumptions must be established about future climate and hydrology. Common practice is to initially base assumptions on current climate projections over the assessment region. Then, given a view that the spatial resolution of global climate projections is not adequate for supporting local to regional hydrologic assessments, some form of spatial downscaling is applied, producing downscaled weather information to drive hydrologic and other subsequent analyses.
For Reclamation, Corps, and other water management agencies, recent choices typically relied on downscaled climate projections stemming from nondynamical methods (e.g., Bias Corrected and Spatially Downscaled WCRP CMIP3 Climate Projections archive or the University of Washington's new HB2860 regional dataset). However, even as nondynamical techniques have become widely used, application of dynamical downscaling is becoming more commonplace (e.g., NARCCAP, CEC-funded modeling centers focused on California/Nevada region and also regionally focused research teams at DRI, University of Washington, University of Arizona, and elsewhere).
One merit of these nondynamically generated datasets is that they are often rich with scenarios, which permits planners to characterize climate projection uncertainty and planning assumptions within that uncertainty (USGS Circular 1331). However, questions remain about the veracity of downscaled data produced by such nondynamical methods, particularly where land-atmosphere interactions are important for defining local climate and where nondynamical methods assume such interactions will remain static even as larger scale climate changes might influence such interactions (USGS Circular 1331). Although such dynamical downscaling activities are limited in being able to characterize climate projection uncertainty or time-flexible planning (e.g., NARCCAP focuses on a single, 30-year, mid-21st century period and represents four GCM projections among the 100+ CMIP3 projections available), they offer the unique capability of providing insight on how larger scale climate changes might trigger changes in local land-atmospheric interactions and associated hydrologic conditions relevant to water management.
Moving forward, water agencies will continue to face decisions about which downscaled data to use and implicitly which downscaling method and spatial resolution to rely upon. As decisions are made, there will be interest in understanding the assessment uncertainties introduced by these choices. However, quantitative information on such uncertainties is limited, setting up the need for evaluations such as this proposed effort.
Earlier evaluations have featured intercomparison experiments involving dynamical and nondynamical downscaling methods (e.g., Wood et al. 2004; Salathe et al. 2007). However, these efforts have typically featured coarser resolution dynamical downscaling and fall short of addressing the questions about where and when the stationarity assumption of nondynamical downscaling may not hold. In contrast, by featuring a suite of methodologies applied at a collection of resolutions, this proposed effort is well suited to build from these earlier intercomparisons, separately addressing how downscaling methods and resolution affect the portrayal of weather or hydrologic impacts under historical and climate-changed conditions. As a result, this research has the potential to provide information that might guide the scoping decisions about the downscaling methods and data to use for different types of scale-dependent planning assessments (e.g., flood risk assessments at local scales versus water supply assessments at regional scales).
Contact the Principal Investigator for information about these documents.