Spatial modeling to understand the dynamics of land cover MODIS satellite data and Markov Chain

Abstract

Abstract The ever-increasing economic and population growth closely relates to the change in land cover. As a region grows, the land cover shifts from forests to agriculture and further to developed areas. Such a gradual change occurs slowly; however, understanding the land cover dynamics is essential to regional planning, especially in developing regions, including Indonesia. Despite its importance, studies on land cover changes in developing regions lack a coherent dataset, as many studies are based on ad-hoc approaches, resulting in embedded uncertainties. An example is the land cover classification using LANDSAT data in many parts of the developing countries lacking consistency. As an alternative, the MODIS dataset is a relatively new dataset produced by NASA, presenting a standardized land cover dataset with a 500×500 m resolution at global coverage. Such a standardized dataset gives consistency in land cover dynamics and future projections. This study presents the potential of MODIS and its use case in understanding land cover dynamics in an urbanizing watershed, i.e., the Majalaya Watershed in West Java, Indonesia. The study presents the use of Markov Chain Analysis in understanding the tendency of changes between land cover classes. This study also indicates future research directions to enrich the topic further.