Data for 'Refined climatologies of future precipitation over High Mountain Asia using probabilistic ensemble learning'

This data set consists of outputs from an ensemble learning model. The model, a mixture of experts (MoE), combines precipitation predictions from the regional climate models (RCMs) of the CORDEX-WAS ensemble. The MoE is validated and used to refined future predictions under RCP4.5 and RCP8.5 over High Mountain Asia (20°–40°N 60°–100°E). Further details are provided in the linked preprint.

For the model validation (1981–2005), the continuous rank probability score (CRPS) of the MoE, an equally weighted (EW) mixture model and statistical RCM surrogates are recorded for each location and month of the study area. For the historical and future prediction experiments, the 5th, 50th, and 95th percentiles of the MoE and EW distribution are reported for each location and month of the study area.