Detailed tree species maps boost European forestry policy

Published on
December 13, 2011

More than 35% of Europe's land area is covered by forests


These offer a wide range of 'ecosystem services', which is to say, products and amenities that societies need. Forests not only supply wood and biomass, they also protect soil and water quality, conserve biodiversity and offer areas for recreation. To anticipate when and where societies will require various ecosystem services, policymakers need information on what forests are located where and how they are developing. Up to now, such information has been available only in a very generalised form. That is now changing.

Maps providing detailed Europe-wide information on tree species coverage have been developed by Dick Burs and a group of colleagues at Alterra (part of Wageningen UR) and the European Forest Institute (EFI). Creation of the maps was part of a special issue on sustainability of forestry-wood chains. The story is published in the European Journal of Forest Research.

"The European tree species map contains distribution maps for 18 species groups and a map of dominant tree species per square kilometre", says Dick Brus. "This enlargement of detail and resolution is a big step forward for research on forest development, spatial planning and land use. These maps give us more information on what ecosystem services can be made available where."

To produce the maps, national forest inventories were used from 18 EU countries. These countries offered very high density tree species plot data. Data was also provided by the ICP Forest Level I measurement network, which covers all of Europe but at a much lower density. According to Brus, "We used statistical methods to create the maps. For the areas with national forest inventories, we chose a variant of spatial interpolation with kriging to map the proportions of land covered by the 18 tree species. This ensured that the chances of occurrence were between 0 and 1 and sum to 1. For the rest of Europe, we used generalized linear regression. Supplementary variables were climatic attributes, such as temperature and rainfall, as well as soil type and biogeographic region." ('Biogeographic regions' are areas with similar environmental conditions.) The map with the predicted dominant tree species was validated using plot data gathered separately from the data used to calibrate the statistical models.

More information: