Wetlands, which provide multiple functions and ecosystem services, have decreased and been degraded worldwide for several decades due to human activities and climate change. Managers and scientists need tools to characterize and monitor wetland areas, structure, and functions in the long term and at regional and global scales and assess the effects of planning policies on their conservation status. The Landsat earth observation program has collected satellite images since 1972, which makes it the longest global earth observation record with respect to remote sensing. In this review, we describe how Landsat data have been used for long-term (≥20 years) wetland monitoring. A total of 351 articles were analyzed based on 5 topics and 22 attributes that address long-term wetland monitoring and Landsat data analysis issues. Results showed that (1) the open access Landsat archive successfully highlights changes in wetland areas, structure, and functions worldwide; (2) recent progress in artificial intelligence (AI) and machine learning opens new prospects for analyzing the Landsat archive; (3) most unexplored wetlands can be investigated using the Landsat archive; (4) new cloud-computing tools enable dense Landsat times-series to be processed over large areas. We recommend that future studies focus on changes in wetland functions using AI methods along with cloud computing. This review did not include reports and articles that do not mention the use of Landsat imagery.

While wetland ecosystem services are widely recognized, the lack of fine-scale national inventories prevents successful implementation of conservation policies. Wetlands are difficult to map due to their complex fine-grained spatial pattern and fuzzy boundaries. However, the increasing amount of open high-spatial-resolution remote sensing data and accurately georeferenced field data archives, as well as progress in artificial intelligence (AI), provide opportunities for fine-scale national wetland mapping. The objective of this study was to map wetlands over mainland France (ca. 550,000 km²) by applying AI to environmental variables derived from remote sensing and archive field data. A random forest model was calibrated using spatial cross-validation according to the precision-recall area under the curve (PR-AUC) index using ca. 135,000 soil or flora plots from archive databases, as well as 5 m topographical variables derived from an airborne DTM and a geological map. The model was validated using an experimentally designed sampling strategy with ca. 3000 plots collected during a ground survey in 2021 along non-wetland/wetland transects. Map accuracy was then compared to those of nine existing wetland maps with global, European, or national coverage. The model-derived suitability map (PR-AUC 0.76) highlights the gradual boundaries and fine-grained pattern of wetlands. The binary map is significantly more accurate (F1-score 0.75, overall accuracy 0.67) than existing wetland maps. The approach and end-results are of important value for spatial planning and environmental management since the high-resolution suitability and binary maps enable more targeted conservation measures to support biodiversity conservation, water resources maintenance, and carbon storage.