MACAv2-LIVNEH: Gridded Daily Statistical Downscaling of CMIP5 using LIVNEH training dataset

This archive contains meteorological variables at the daily timestep and 1/16-degree resolution for the contiguous United States produced by the MACAv2-LIVNEH (Multivariate Adaptive Constructed Analogs) statistical downscaling method (Abatzoglou and Brown, 2012) performed on daily outputs from global climate model simulations of the historical (1950-2005) and future RCP4.5/8.5 (2006-2099) scenarios of the Coupled Model Intercomparison Project Phase 5 (CMIP5) utilizing the Livneh et al (2013) training dataset. The MACA method removes biases in the global model variables and spatially downscales them using a variation of the constructed analogs method. To be consistent with the observed calendar, leap days have been added to the datasets by averaging values between Feb 28 and Mar 1.