The Caucasus region is one of the global biodiversity hotspots which further comprises highly diverse mountain grasslands. These grassland ecosystems were shaped from a long tradition of human land use and provide multiple ecosystem services such as food supply for grazing animals, recreational sites and erosion control. However, changes of land use practices have induced grassland degradation in the Georgian Caucasus regions. Overgrazing during Soviet period and recent increases in recreational activities resulted in a reduced grass cover, an increased abundance of unpalatable plant species and soil erosion. Due to an expansion of grassland degradation, the loss of services provided by healthy ecosystems can be expected. To protect Georgian mountain grasslands, detailed information about ecological relationships within the ecosystem and methods to monitor grassland conditions are urgently needed.This thesis investigated grassland degradation within the upper Aragvi valley of the Greater Caucasus, in the Republic of Georgia. Field studies were conducted in the vicinity of the village Mleta, in a landscape which is frequented by overgrazing, erosion and mass wasting events. The aim of this thesis was to develop site-specific methods to prevent further degradation in the Caucasus region. Therefore we implemented the commonly used feature of vegetation cover to assess the extent of grassland degradation by remote sensing imagery. We used random-forest regression to estimate vegetation cover from the Normalized Difference Vegetation Index (NDVI) derived from multispectral WorldView-2 data. The good model fit of R2 = 0.79 indicates the great potential of a remote-sensing approach for the observation of grassland cover. The presented vegetation cover map shows grassland degradation on steep slopes close to human settlements and along hiking trails. Further, we investigated the relationships between plant diversity, site conditions and vegetation cover on overgrazed and eroded sites. We used non-metric dimensional scaling ordination and cluster comparison of functional plant groups to describe a gradient of grassland vegetation cover. For our study region, we identified four major vegetation types and performed an indicator species analysis. On abandoned hay meadows of the subalpine zone we identified tall herb vegetation with increasing occurrence of ruderal pasture weeds. Within high-montane grassland a decline of plant diversity can be observed on sites of reduced vegetation cover. Based on the results of the indicator species analysis, a list of 22 recommended native plant species to revegetate beginning small scale damage patches was elaborated and is presented in this thesis. In the last chapter, we improved the detection of grassland degradation by multispectral satellite sensors as we implemented vegetation cover and vegetation types into a classification model. Therefore, we used a hand-held field spectrometer to simulate the multispectral World View 2 sensor. A selection of predictors (vegetatation indices, spectral bands and environmental variables) was implemented into random forest models to predict the vegetation cover and vegetation types. The outcomes were further combined to estimate the grassland condition of our research area. Our results showed an overall accuracy of 75% and were displayed in an NMDS ordination graph.To prevent further large scale erosion events and the loss of precious mountain grasslands we conclude that the presented remote sensing methods are promising tools for the early detection of beginning vegetation damage spots. Previous reforestation efforts for slope protection have failed due to the lack of an appropriate grazing management. Due to a low potential of the grassland ecosystem to balance further vegetation cover damage, the long-term loss of diverse habitats can be expected. Consequently, to conserve precious Georgian mountain grasslands a sustainable landscape management for the collective mountain grasslands is mandatory. The results of this thesis serve for the implementation into sustainable agricultural and touristic development plans of mountain regions which suffer from grassland degradation.
Verknüpfung zu Publikationen oder weiteren Datensätzen
