High-resolution future climate data for species distribution models in Europe

Description

This dataset contains a set of 13 climatological variables (Variable, VariableName) at a spatial resolution of 1x1km for Europe (nx = 13147, ny = 6071) for historical (ClimatePeriod) and future climate conditions. These variables are a subset of the so-called bioclimatic variables that are often part of global gridded datasets (e.g. WorldClim, CHELSA) that have been specifically developed for species distribution modelling and ecological applications.

The climatological data correspond to 35-year (Startyear_Endyear = 1971_2005) and 30-year (Startyear_Endyear = 2041_2070) mean values representing respectively historical and future climate conditions. To account for the future climate conditions, three possible emission scenarios of greenhouse gases as defined by the Intergovernmental Panel on Climate Change (IPCC) are used (ClimatePeriod = rcp26, rcp45, rcp85).

The complete set of variables (var[1-13]) for which historical and future climate data layers are produced are given below.

The source data for the climate layers were assembled from the EURO-CORDEX archive (Kotlarski et al., 2014). More specifically, we have used the regional climate model simulations for Europe at a spatial resolution of 12.5x12.5km on which a three-step statistical downscaling approach has been applied:

1. Processing (averaging, totals, …) of all available time series of the EURO-CORDEX model experiments (ClimatePeriod = evaluation, historical, rcp) for the climatological variables.
2. Interpolation of the data layers from the 12.5x12.5km EURO-CORDEX grid to a 1x1km spatial CHELSA (Karger et al., 2017) reference grid (see files lat_1km.csv and lon_1km.csv).
3. Calculate differences between the 1x1km-interpolated variables (Variable = only for var[1-9]) from the evaluation model experiments (or ClimatePeriod) and the corresponding reference bioclimatic CHELSA variables. In order to account for possible biases present in the EURO-CORDEX climate models, these differences (or biases) are then subtracted from the respective 1x1-km-interpolated variables for the historical and rcp model experiments (ClimatePeriod).

The dimensions of the 1x1km grid (excl. the first row and column):

• y-dimension = number of columns = 6071
• x-dimension = number of rows = 13147

The longitudes and latitudes of respectively the southwest and northeast corner of the grid are:

• longitude -44.592; latitude 21.991 (southwest corner)
• longitude 64.967; latitude 72.583 (northeast corner)

The climatological variables are used as input data for the species distribution modelling of Invasive Alien Species for the Tracking Invasive Alien Species (TrIAS) project.

Variables

• Variable (VariableName): Unit
• var1 (AnnualMeanTemperature): °C
• var2 (AnnualAmountPrecipitation): mm year^-1
• var3 (AnnualVariationPrecipitation): coefficient of variation
• var4 (AnnualVariationTemperature): stdev
• var5 (MaximumTemperatureWarmestMonth): °C
• var6 (MinimumTemperatureColdestMonth): °C
• var7 (TemperatureAnnualRange): °C
• var8 (PrecipitationWettestMonth): mm
• var9 (PrecipitationDriestMonth): mm
• var10 (30yrMeanAnnualCumulatedGDDAbove5degreesC): °C days
• var11 (AnnualMeanPotentialEvapotranspiration): mm day^-1
• var12 (AnnualMeanSolarRadiation): W m^-2
• var13 (AnnualVariationSolarRadiation): stdev

Files

• varX_VariableName_ClimatePeriod_Startyear_Endyear.csv: climatological data layers for the 13 variables listed above
• lon_1km.csv: longitudes for the 1x1km grid
• lat_1km.csv: latitudes for the 1x1km grid