Bioclimatic dataset of Metropolitan France under current conditions derived from the WorldClim model

Several studies have shown that adequate bioclimatic information is of major importance for mapping ecological niches or for modelling the distribution ranges of species and communities, particularly from a climate change perspective [1,2]. However, in France, there are few data sources that provide consistent information, available data being produced at low spatial resolution and based on classification systems that are not suitable for mapping French ecological systems. This paper presents bioclimatic maps produced on Metropolitan France and based on the Worldwide Bioclimatic Classification System, which are called Global Bioclimatics [3]. This data paper documents a set of variables that includes 23 bioclimatic maps generated according to the Worldwide Bioclimatic Classification System. These maps describe current bioclimatic conditions in Metropolitan France at a resolution of 30 arc-seconds. Climatic parameters and bioclimatic indices usually used for the analysis or modelling of species and communities’ distribution, and bioclimatic typological units, were calculated using the temperature and precipitation data derived from the WorldClim 2 model. These maps can be used in GIS or models by researchers for mapping ecological conditions, but can also provide natural resource managers with analytical tools to assess Nature conservation policies.


a b s t r a c t
Several studies have shown that adequate bioclimatic information is of major importance for mapping ecological niches or for modelling the distribution ranges of species and communities, particularly from a climate change perspective [ 1 , 2 ]. However, in France, there are few data sources that provide consistent information, available data being produced at low spatial resolution and based on classification systems that are not suitable for mapping French ecological systems. This paper presents bioclimatic maps produced on Metropolitan France and based on the Worldwide Bioclimatic Classification System, which are called Global Bioclimatics [3] . This data paper documents a set of variables that includes 23 bioclimatic maps generated according to the Worldwide Bioclimatic Classification System. These maps describe current bioclimatic conditions in Metropolitan France at a resolution of 30 arc-seconds. Climatic parameters and bioclimatic indices usually used for the analysis or modelling of species and communities' distribution, and bioclimatic typological units, were calculated using the temperature and precipitation data derived from the WorldClim 2 model. These maps can be used in GIS or models by researchers for mapping ecological conditions, but can also provide natural re-source managers with analytical tools to assess Nature conservation policies.
© 2020 The Author(s Value of the Data • The dataset provides bioclimatic maps that can be used in GIS or models for mapping ecological conditions • The dataset can be of interest in many fields of research (Ecology, Biogeography, Forestry, Agronomy…) to build models of species, plant communities or vegetation series distribution • The generated maps can also provide natural resource managers with analytical tools to assess Nature conservation policies or agricultural practices

Data
Several studies have shown that adequate bioclimatic information is of major importance for mapping ecological niches or for modelling the distribution ranges of species and communities, particularly from a climate change perspective [ 1 , 2 ]. However, in France, there are few data sources that provide consistent information, available data being produced at low spatial resolution and based on classification systems that are not suitable for mapping French ecological systems. This paper presents biocimatic maps produced on Metropolitan France and based on the Worldwide Bioclimatic Classification System, which are called Global Bioclimatics [3] .
The Worldwide Bioclimatic Classification System offers a quantifiable bioclimatic typology that shows a close relationship between climate and vegetation models [4] . It recognizes 28 bioclimates and about 400 isobioclimates (i.e. aggregation of bioclimates, thermotypic horizons and ombric horizons), highlighting slight climatic variations. Compared to other widely used bioclimatic classification systems such as that of Köppen [5] , the Worldwide Bioclimatic Classification System is adapted to the context of the French ecosystems: it discriminates between subtropical and mediterranean climates and considers mountains belts like altitudinal thermic variations of the global surrounding bioclimate -i.e. as a part of the zonation going from low to high altitudes -rather than like a distinct orobioclimate. Moreover, it includes several levels of submediterraneity ( Ios i and Isbm indices in Table 1 ), which is an important factor in temperate ecosystems of southwest Europe where Mediterraneo-Atlantic species are found. The WorldClim 2 model [6] was used for the current period (1970-20 0 0) to derive the entire dataset. The overall model accuracy is considered by its authors to be very high for temperatures, while precipitation modelling is a bit poorer due to a more heterogeneous regime in time and space than that of temperatures.
Bioclimatic maps that cover Metropolitan France were generated with the original coordinate reference system (WGS-84) at the resolution of 30 arc-seconds ( i.e. about 0.6 km 2 ).
An overview of the results is given in Fig. 1 .

Experimental design, materials and methods
The bioclimatic dataset was derived from the WorldClim v2.1 dataset [6] . Temperature ( i.e. annual average, minimum and maximum temperatures) and annual amount of precipitation variables were obtained at a spatial resolution of 30 arc-seconds. The digital elevation model with a grid resolution of 250m (source: BD ALTI® IGN, French Geographic National Institute), which was used to calculate some of the bioclimatic variables, was aggregated to a 30 arcseconds spatial resolution.
The procedure for calculating bioclimatic variables ( i.e. generation of the climatic parameters and bioclimatic indices first and then the bioclimatic units) was first carried out as defined by [ 3 , 7 ]. Then, following the recommendations of the authors, some of these bioclimatic variables were compensated. For example, temperatures variables ( i.e. T, M, m, Itc and Tp ) were corrected by the altitude to produce maps of the macrobioclimates (see note 1 of table 25 in [3] ).
Data processing was performed using two softwares [8] . Most bioclimatic variables were calculated using simple arithmetic operations or conditional statements with QGIS 3. The other bioclimatic variables were calculated with the raster functions of GRASS: r.latlong for latitude definition; r.slope.aspect for topographic parameters; r.series for statistic operations; r.reclass for segmentation; r.sun to determine the day length ( L parameter) of the Thornthwaite equation [9] that is required for the annual ombro-evaporation index ( Ioe ) [7] . Calculation of the day length was made on the median day of each month and the step parameter was set to 0.025. Appendix A. Supplementary data

Declaration of Competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.