Planning for the effects of climate change on natural resources often requires detailed projections of future climate at finer spatial scales consistent with the processes managers typically consider. While it is numerically possible to produce downscaled climate at very fine scales (< 5km), accurate estimation at these scales is difficult and less certain without very detailed local information. Both the absence of a sufficiently dense network of long-term climate observations and the presence of local factors such as topography and land surface feedbacks from vegetation and snowpack contribute to the uncertainties of localized projections. To meet the needs of managers for developing adaptation strategies, vulnerability assessments, climate impacts assessments, and specific resource modeling at landscape scales, we downscaled projections from the coarser scales of global climate models (GCMs) to more local scales. The R1/R6 project was designed primarily to provide climate information consistent with manager requests and to create a basis for more detailed work or for a more comprehensive approach to downscaling and regional climate modeling. The objectives of this project were to:
- Develop consistent historical and future downscaled climate and hydrologic data and projections using the same methodology for several major river basins in the western United States, Columbia, upper Missouri, upper Colorado, and Great Basins (Figure 1), and
- Summarize that information with the needs of the funding agencies in mind, including descriptions of where to get the information, methods, results, uses of and uncertainties associated with the data and projections.
We analyzed global climate models (GCMs) available from the IPCC AR4 assessment to better understand the projected future climate by region and individual model sensitivities within regions. We then developed an ensemble of climate models that have the best capability in the 5 basins and projected downscaled climate and hydrology based on an ensemble delta method and two bracketing scenarios. Applying the downscaled climate data to the historical (1916-2006) and two future timeframes (2030-2059, "2040s"; 2070-2099, "2080s") at 1/16th degree (~6km), we estimated hydrologic output tailored for impacts assessments (e.g., snow water equivalent, soil moisture, potential evapotranspiration, actual evapotranspiration, and runoff). The result is a consistent set of downscaled climate and hydrologic projections at ~6km for the entire Columbia, upper Missouri, and upper Colorado basins and 12km for the Great and lower Colorado basins. The data are summarized at monthly time scales for Bailey's Ecosections (Figure 2), Omernik Level III Ecoregions (Figure 3), and 8-digit Hydrologic Unit Code (HUC 4) basins; but are also available in raw form on a grid-cell basis at daily time steps and in ascii grid (ArcGIS) format for observed and future climatologies.