Progress and prospects in global land cover mapping and change monitoring

Land cover and its changes are an important part of global change research, and play an irreplaceable role in climate change research, biodiversity conservation, the global carbon and water cycles monitoring, as well as other major needs. As an essential descriptor of the Earth’s surface, land cover provides a fundamental basis for understanding human-environment interactions and assessing ecosystem functions. Over the past few decades, with the continuous improvement of satellite remote sensing technology, the rapid growth of data storage capacity, and the development of high-performance computing platforms, global land cover mapping and change monitoring have made remarkable progress. The spatial resolution of available products has increased from the early 1 km scale to today’s 10-30 m range, enabling a more detailed characterization of heterogeneous landscapes. Meanwhile, the temporal frequency has evolved from single-period mapping toward multi-year and annual land-cover change monitoring, greatly enhancing our ability to capture dynamic land processes and long-term environmental transitions. In this study, we collected 21 publicly released global land cover products at home and abroad, and then comprehensively summarize their characteristics from three aspects: land cover classification system, methodology and dataset/products. First, we systematically summarized the land-cover classification system of current global land cover products into three groups, and further clarified their characteristics, advantages and disadvantages. Then, we sorted out the development history of the global land-cover algorithms from single land-cover mapping and time series land cover change monitoring, and further evaluated the different development stages and overall development trends of the global land cover research. Finally, the main problems existing in the current global land cover research are described, and some thoughts on the future development of this direction are put forward, particularly regarding the need for harmonized standards, higher spatiotemporal consistency, and better validation frameworks.