Human factors explain the majority of MODIS-derived trends in vegetation cover in Israel: a densely populated country in the eastern Mediterranean

Abstract

Land cover and land use changes can result from climatic variability and climate changes, as well as from direct and indirect human drivers, such as growth in population and consumption. In this study, we aimed to examine whether major factors driving landscape changes (expressed in vegetation cover) in Israel, a densely populated country in the eastern Mediterranean Basin, are related to physical drivers or to human causes. To this end, we calculated statistical trends in the Normalized Difference Vegetation Index (NDVI—a spectral index representing vegetation cover) from a 14-year MODIS time series, between 2000 and 2014, to identify areas where vegetation cover has either increased or decreased. We chose 125 study areas where statistically significant changes in NDVI were found and used time series of monthly rainfall, Landsat images, Google Earth images and environmental GIS layers to identify the type and cause of landscape changes. The two most common general classes driving land cover changes were agricultural (56 of 125; expansion of agricultural areas or change in agricultural crops) and urban (28 of 125; urban expansion or urban greening). Other important drivers of landscape changes included forestry, woody encroachment, wildfire dynamics and water management. Climate variability was found to explain landscape changes in only 3 of the 125 study areas, all located in the transition zone between the desert and the Mediterranean climate regions of Israel, where a decrease in rainfall led to a decrease in NDVI values. NDVI as an indicator of landscape changes is not effective to detect changes in non-photosynthetic vegetation or to monitor changes in forests where leaf area index values are high. However, we show here that even in a highly heterogeneous and densely populated country, MODIS-derived time series of NDVI are informative to identify landscape change processes.