Statistical modeling of land-cover changes based on key socio-economic indicators

Abstract

Landscapes are complex human–environment systems operating at spatio-temporal scales. Time is just as important as space when researching landscape changes. These changes are influenced by both environmental and socio-economic factors. However, correlations between environmental landscape attributes and land-cover patterns/changes are weakened by human activities such as intensification of agriculture eliminating the constraints of water and nutrient availability. Relations between changes in socio-economic organisation and land cover become apparent only over a longer period of time. Thus, in our study, we focused on socio-economic factors and their long-term effects on land cover. We present a method to (i) differentiate types of land-cover changes at district level, (ii) model correlations between socio-economic factors and land cover changes and (iii) identify key socio-economic indicators of land-cover changes between 1945 and 1999 in a German marginal rural landscape. We employed agricultural land-cover data gained from the interpretation of multi-temporal aerial photographs. Based on these data, we differentiated types of land-cover changes, characterising different directions of agricultural land-cover changes in the observation time period. Various socio-economic aspects were considered by introducing data representing factors of demography, employment, economy, infrastructure, agricultural structure and policy. The relations between time series of land-cover data and of socio-economic data were modeled with the help of redundancy analysis. Correlation coefficients were used to identify key socio-economic indicators of land-cover changes. The results showed that a relatively high percentage of variance in land-cover data can be explained by socio-economic factors. The types of land-cover changes can be characterised by combinations of key socio-economic indicators. The indicators can be helpful to reconstruct land-cover changes in other regions. Thus, they provide a basis for the development of sustainable land-cover management systems.