Analysing the spatial structure of the Estonian landscapes: which landscape metrics are the most suitable for comparing different landscapes?

We calculated 15 landscape metrics on 35 Estonian landscapes and performed factor and principal component analyses in order to determine which landscape metrics work on Estonian Basic Map and which do not. The results showed that there are four main components that describe landscape structure: dominance, contrast, shape complexity, and composition. We suggest the following landscape metrics for measuring these aspects respectively: ED or SIDI, TECI or ECON_MN, SHAPE_MN, and PRD. However, the selection of the metrics always depends on the purpose of the study and the character of the ecological process. Principal component analysis also showed that heights and plains have more heterogeneous landscape than lowlands. Moreover, although the heights (Otepaa and Haanja) and the plains with urban areas (Tallinn and Tartu) have high fragmentation and heterogeneity, they have different contrast values: urban areas have very low contrast while in the case of heights contrast is high.