The application of the High Nature Value concept in Estonia in the production of a new map of farmland as a basis for national policy formulation.

The concept of High Nature Value (HNV) farmland has been widely accepted in Europe as a useful index correlated with biodiversity at the landscape scale since the mid 1990’s but there are still problems in its detailed definition. The present work utilized the definition and concept as described in the review paper by Lomba et al (2014). The method described below is consistent with the idea that the current approach for the whole of Europe lacks the spatial and temporal detail that is needed at the regional scale.
Farmland is land used for agriculture and therefore includes linear features and grazed forests, although in Estonia most forests are not used by domestic stock. The concept is mainly mapped at a 1 km square scale. Kikas et al (2015) have reviewed the relevant literature to the background and concept of HNV and its links to biodiversity in Estonia because an HNV map was being developed in order to help the Ministry of Agriculture to identify important environmental areas using appropriate national environmental datasets. The current European map of potential HNV (Parrachini et al (2008)) was shown to have problems because of limitations of European databases, eg the CORINE Land Cover Map. Many bogs were also incorrectly included as farmland. The review concluded that the use of national data bases would improve the accuracy of mapping. The current HNV map will be used for policy formulation in Estonia and will promote conservation of biodiversity and traditional farming practices.
Kikas et al (2017) describe 20 appropriate indicator parameters that were mainly selected by the Ministry of Agriculture, with each one being divided into classes to produce indicator values according to expert judgement. These were added together for each 1 km square to give a score, ie an expert system, to define HNV. In general there are few high correlations between the indicators showing that the selection was sound. The frequency histogram of the scores for all 1 km squares with agricultural land in Estonia showed a normal distribution. The top and bottom 10% from the histogram were identified in order to investigate the structure of the data further. The former are termed Exceptionally High HNV (EHNV) and the latter Relatively Low HNV (RLNV). The distribution of these groups in Estonia show identifiable patterns linked to biodiversity “hot spots”. Cluster analysis within the EHNV squares produced three classes whose distribution and composition are described to demonstrate the character of Estonian landscapes with high HNV, as shown in Figure 1.