Environmental stratifications provide the framework for efficient surveillance and monitoring of biodiversity and ecological resources, as well as modelling exercises. An obstacle for agricultural landscape monitoring in Estonia has been the lack of a framework for the objective selection of monitoring sites. This paper describes the construction and testing of the Environmental Stratification of Estonia (ESE). Principal components analysis was used to select the variables that capture the most amount of variation. Seven climate variables and topography were selected and subsequently subjected to the ISODATA clustering routine in order to produce relatively homogeneous environmental strata. The ESE contains eight strata, which have been described in terms of soil, land cover and climatic parameters. In order to assess the reliability of the stratification procedure for the selection of monitoring sites, the ESE was compared with the previous map of Landscape Regions of Estonia and correlated with five environmental data sets. All correlations were significant. The stratification has therefore already been used to extend the current series of samples in agricultural landscapes into a more statistically robust series of monitoring sites. The potential for applying climate change scenarios to assess the shifts in the strata and associated ecological impacts is also examined.
- Clustering algorithm
- Stratified random sampling