Land use changes analysis and prediction using remote sensing and QGIS MOLUSCE Plugin in the Siahkal County

Quantifying land use change dynamics is critical in tackling environmental and socio-economic challenges such as climate change in recent years. This study takes Siahkal County in Guilan Province as the research subject and analyzes the land use changes in two different years: 2000 and 2021, and predicts the change in 2031. We carried out land use change analysis using LANDSAT-7 ETM+ and LANDSAT-8 OLI multitemporal data pertaining to the years 2000 and 2021. For land use data extraction, a pixel-based digital classification using an ISODATA algorithm with a high Kappa index of 0.97 was applied to Landsat images. In addition, the MOLUSCE plugin in QGIS software was used to model land use change for 2000-2021, produce a transition probability matrix, and detect the future of land use for 2042 by using Cellular Automata Simulation. During 2000-2021, the result showed that the areas under agricultural areas, bare, built-up, plantation, shrubland, and woodland classes are found to increase with growth rates of about 31%, 39%, 67%, 88%, 20%, and 5%, respectively. The area under forest, grassland, and water bodies is found to decrease with decrease rates of about 22%, 16%, and 18%, respectively. Mountainous meadows did not change significantly. In terms of projected land use, the result also indicates fluctuations in land use change, especially for built-up land, showing a steady increase over time. We saw evidence of the local expansion of forest plantations, but the continuous decrease of natural forests may negatively impact the natural environment and landscape patterns regionally. In conclusion, the results of monitoring and modeling land use changes can be seen as a warning to managers, policy-makers, and planners. It is also advisable to use more data to analyze the impacts of land use changes on landscape patterns in future studies.