Land-use and vegetation change detection in Plastira artificial lake catchment (Greece) by using remote-sensing and GIS techniques

Detecting changes in the land-use and vegetation conditions by using remote-sensing techniques is a common approach nowadays to assessing human-induced impacts in a specific area. For this purpose, a series of vegetation indices and change detection algorithms such as the normalized difference vegetation index (NDVI), Iterative Self-Organizing Data Analysis Technique (ISODATA), and k-means have been efficiently developed and used in many studies worldwide. However, identifying the driving forces for the estimated changes in land use and vegetation has always been a difficult and challenging task. In this study, Landsat and Système Pour l'Observation de la Terre (SPOT) images have been used to estimate the NDVI and land-use changes at the Plastira artificial lake catchment in the period 1984–2009. The recorded vegetation changes were correlated with a series of environmental and human-related parameters such as precipitation, temperature, specific land uses, and topography to identify the dominant factors of the aforementioned changes. This was done using both linear and geographically weighted regression methods. The results indicate that the precipitation and temperature fluctuations are strongly correlated with the vegetation conditions, whereas, as far as the topographic parameters are concerned, the aspect and slope affect mostly the particular vegetation index (the NDVI) of the study area.