Literature review on land cover change and land use

Abstract

Objetive. A systematic literature review was conducted to identify advances in applications based on Machine Learning and Deep Learning algorithms for the analysis of land cover change and land use for prospective purposes, their covariates and the names of land cover and land use classifications. Methodology. A bibliographic retrieval was carried out consulting different digital libraries based on a search equation limited to the study period from 2000 to 2024. Subsequently, a layered reading was applied to extract information relevant to the study, which consisted in reviewing the introduction, methods, figures, tables, results and conclusions, in order to generate annotations relevant to the research. Results. A total of 55 studies were obtained, including articles, books and thesis, from which the information pertinent to the research was summarized and extracted, then related to discrimination by land coverage and land use types, drivers of change, and frequently used machine learning techniques. Additionally, a citation of the key concepts of land cover and land use with cellular automata was provided. Conclusions. When analyzing land cover and land use change toward prospective objectives, the use of cellular automata is remarkable, for the possibility of generating simulations of possible future lines, enabling scenario analysis, as well as deducing and recreating the transition rules between different land cover types or land uses, in an automated and statistically evaluable way.