Skip to main content
. 2017 Jul 6;15(7):e04881. doi: 10.2903/j.efsa.2017.4881

Figure 2.

Figure 2

Relative probability of presence of the genus Picea in the European Union territory (based on data from the species: P. abies, P. sitchensis, P. glauca, P. engelmannii, P. pungens, P. omorika, P. orientalis). Left panel: Relative probability of presence (RPP) of species/genera from the European Atlas of Forest Tree Species (de Rigo et al., 2016; San‐Miguel‐Ayanz et al., 2016), aggregated at 100 km² pixel resolution. RPP is defined as the probability of finding species/taxon in a given area, irrespective of the probability of finding other taxa (de Rigo et al., 2017). As a consequence, the sum of all RPPs for different taxa in the same area need not be 100%. The estimates are based on constrained spatial multi‐scale frequency analysis (C‐SMFA) (de Rigo et al., 2014, 2016, 2017): this is a spatial multi‐scale frequency analysis of field observations (de Rigo et al., 2014, 2016), constrained to enhance the estimates’ consistency with the frequency of broadleaved and coniferous taxa derived from Corine Land Cover (Bossard et al., 2000; Büttner et al., 2012). Right panel: Trustability of RPP. This qualitative measure is based on the multi‐scale aggregation of the number of field observations (i.e. the local density of data) for each pixel and taxon. The colour scale of the trustability map is based on the quantiles of this data density (de Rigo et al., 2014, 2016)