Land Change Science has increasingly relied on spatial analysis methods to monitor, understand, and predict land-use and land-cover change (LULCC). Over the past decade, technological advancements such as high-resolution satellite imagery, machine learning algorithms, and robust GIS platforms have significantly transformed how spatial patterns and environmental transformations are studied. However, there is a lack of a synthesized understanding of how these geospatial methodologies have evolved and been applied across different contexts and regions. This review aims to systematically examine the evolution and application of spatial analysis techniques in land change science, focusing on the tools, models, and analytical approaches used in geospatial studies over the past decade. A systematic literature review (SLR) was conducted using a dataset of 62 peer-reviewed research articles published between 2015 and 2025. The articles were analyzed based on key parameters, including geographic context, spatial analysis methods, software used (e.g., ArcGIS, ERDAS, Google Earth Engine), types of classification models (e.g., CA-Markov, Random Forest, SVM), and theoretical frameworks. The review also considered novelty, limitations, and future research directions highlighted by each study. The review found that CA-Markov modeling, supervised classification, and Random Forest are the most frequently applied spatial analysis techniques. A notable trend is integrating machine learning with remote sensing, particularly through platforms like Google Earth Engine. While ArcGIS remains dominant, open-source tools like QGIS and Python-based APIs are gaining traction. Data availability, spatial resolution, and lack of socio-economic integration often limit studies. Theoretical frameworks, such as Human–Environment Interaction Theory and urban ecological theory, were commonly employed to interpret the findings. Geospatial methodologies in land change science have advanced significantly, enabling more dynamic, scalable, and accurate assessments of environmental change. Future research should focus on integrating socio-economic variables, enhancing ground validation, and developing hybrid models that leverage AI and big data to achieve a more holistic understanding of land system science.
Supiyandi et al. (Thu,) studied this question.