Comparison of Two Approaches for Pasture Inventory in Albania

This study presents the evaluation of two approaches (field and remote sensing - GIS) applied for inventorying pasture condition in Albania. The field inventory was carried out by trained experts in the context of the latest Albanian National Forest Inventory (ANFI), based on 162 sample plots chosen randomly over the pasture areas. At each sample plotit was collected detailed data about pasture condition which were made available for this study. The second approachwe applied, consisted of a combination of Remote Sensing and GIS algorithms to derive several variable maps(classified into a number of classes): cover components of pastures,slope, aspect, elevation, erosion, distance from the hydrologic network, road network and the location of villages. The results indicated significant differences between both methodologies, especially for the cover components of pastures, where a difference of 14% was found regarding the evaluation of herbaceous species. Environmental and pasture use variables appeared to be evaluated differently too, especially the erosion level and distances of pastures from thevillages, networks of roads and water sources.We conclude that the main reason for such discrepancy should bethe low coverage of pasture area in terms of number of sample plots as well as their distribution.Based on the findings of this study, it is recommended that in the future, inventory of pasture condition should be done as much as possible using Remote Sensing and GIS technology because it guarantees full coverage of the study area, reduces the field work which is expensive and time consuming, and provides more reliable evaluation of pasture condition.