NetCDF4 files containing the seasonal output of empirically downscaled GCM data under the RCP8.5 trajectory, has been produced in the Norwegian Research Council Project RCubed, work package 3, using a hybrid dynamical-statistical downscling approach. The data covers most of South Norway, and is available in a lat-lon projection with a resolution of 0.1° in longitude and 0.05° in latitude. The data is produced using the R software package esd. The downscaled variables are seasonal mean 2-meter temperature (T2), seasonal wet-day mean (mu - the precipitation intensity on rainy (>1 mm) days), and seasonal wet-day frequency (fw - the frequency of days where precipitation is 1 mm or more). The data here is the common ensemble, that is GCM model runs which were used for downscaling T2, mu, and fw. From mu and fw seasonal mean precipitation (excluding days where precipitation is less than 1 mm) is derived. The data is also available from MET Norway's thredds server with OPeNDAP http://thredds.met.no/thredds/catalog/metusers/helenebe/catalog.html The data is downscaled using ouput from a dynamical downscaling (see Pontoppidan, 2018) as reference data. This means that the data may contain biases during historical time. The precipitation fields inter-annual variation is not fully captured at the fine grid scale for which the data is produced, so we recommend applying some upscaling/areal averaging if the inter-annual correlation is of importance. The long-term trend is likely captured for the fields at the resolution of the data. An article describing the work behind the data is underway, and will be added here once published. Pontoppidan, M., Kolstad, E. W., Sobolowski, S., & King, M. P. ( 2018). Improving the reliability and added value of dynamical downscaling via correction of large‐scale errors: A Norwegian perspective. Journal of Geophysical Research: Atmospheres, 123, 11,875– 11,888. https://doi.org/10.1029/2018JD028372 The GCM runs downscaled: "1" "ACCESS13_r1i1p1" "2" "bcccsm11_r1i1p1" "3" "CanESM2_r1i1p1" "4" "CanESM2_r2i1p1" "5" "CanESM2_r3i1p1" "6" "CanESM2_r4i1p1" "7" "CanESM2_r5i1p1" "8" "CCSM4_r1i1p1" "9" "CCSM4_r2i1p1" "10" "CCSM4_r3i1p1" "11" "CCSM4_r4i1p1" "12" "CCSM4_r5i1p1" "13" "CCSM4_r6i1p1" "14" "CESM1BGC_r1i1p1" "15" "CESM1CAM5_r1i1p1" "16" "CNRMCM5_r10i1p1" "17" "CNRMCM5_r1i1p1" "18" "CNRMCM5_r2i1p1" "19" "CNRMCM5_r4i1p1" "20" "CNRMCM5_r6i1p1" "21" "CSIROMk360_r10i1p1" "22" "CSIROMk360_r1i1p1" "23" "CSIROMk360_r3i1p1" "24" "CSIROMk360_r5i1p1" "25" "CSIROMk360_r6i1p1" "26" "CSIROMk360_r8i1p1" "27" "GFDLCM3_r1i1p1" "28" "GFDLESM2M_r1i1p1" "29" "GISSE2H_r1i1p1" "30" "GISSE2H_r1i1p2" "31" "GISSE2H_r1i1p3" "32" "GISSE2R_r1i1p1" "33" "GISSE2R_r1i1p2" "34" "GISSE2R_r1i1p3" "35" "HadGEM2CC_r1i1p1" "36" "HadGEM2ES_r1i1p1" "37" "HadGEM2ES_r3i1p1" "38" "HadGEM2ES_r4i1p1" "39" "inmcm4_r1i1p1" "40" "IPSLCM5ALR_r1i1p1" "41" "IPSLCM5ALR_r2i1p1" "42" "IPSLCM5ALR_r3i1p1" "43" "IPSLCM5ALR_r4i1p1" "44" "IPSLCM5AMR_r1i1p1" "45" "IPSLCM5BLR_r1i1p1" "46" "MIROC5_r1i1p1" "47" "MIROC5_r2i1p1" "48" "MIROC5_r3i1p1" "49" "MIROCESM_r1i1p1" "50" "MIROCESMCHEM_r1i1p1" "51" "MRICGCM3_r1i1p1" "52" "NorESM1M_r1i1p1" "53" "NorESM1ME_r1i1p1"

This work was supported by the Research Council of Norway (RCN) through R3 (grant 255397)