Models that simulate ecosystems at local to regional scales require relatively fine resolution climate data. Many methods exist that downscale the native resolution output from global climate models (GCM) to finer resolutions. NASA NEX-DCP30 is a statistically downscaled 30 arcsecond resolution climate dataset widely used for climate change impact studies in the conterminous USA (CONUS), but it did not include vapor pressure data which is essential for many types of models. We downscaled vapor pressure data from 28 global climate models included in the Coupled Model Intercomparison Project Phase 5 (CMIP5) to 30 arcsecond resolution for CONUS to augment the NEX-DCP30 dataset. Monthly vapor pressure values were calculated from raw GCM output for the conterminous USA from 1950 to 2100, representing RCP4.5 and RCP8.5 climate change scenarios. Vapor pressure data were then downscaled from the GCM's native spatial resolutions to 30 arcsecond using the Bias Correction-Spatial Disaggregation (BCSD) statistical downscaling method, which had been used to create the original NEX-DCP30 dataset. PRISM LT71m gridded climate data for 1970-1999 served as the reference data. The newly created downscaled vapor pressure dataset may be used in conjunction with the existing NEX-DCP30 data as input for vegetation, fire, drought, or earth system models. The data is available at the Forest Service Research Data Archive.
Keywords: BCSD; Bias correction spatial disaggregation; Climate change; NEX-DCP30; PRISM; Statistical downscaling, humidity; Vapor pressure.