Today, land ecosystem restoration actions are among the proven solutions to reverse anthropogenic and climate-driven land degradation and desertification. Such actions have multiple advantages such as supporting ecosystem resilience (Ellison & Ifejika Speranza, 2020; Sacande et al., 2021), increasing carbon sequestration (Dixon et al., 2016), restoring hydrological catchment characteristics (Anderson et al., 2010; Carrick et al., 2019), etc. Thus, ecosystem restoration actions are a global priority (Robinson et al., 2023). Indicative of this is that UN named years 2021-2030 as the Decade on Ecosystem Restoration (Waltham et al., 2020), while the EU incorporates such policies in the Green Deal (Gann et al., 2019) and has announced action like the 3 billion additional trees pledge. Although restoration actions do take place, too few of these are monitored thereafter (Nadal-Romero et al., 2023). Monitoring methods based on remote sensing could be a viable alternative to high-cost and labor-intensive conventional ones (de Almeida et al., 2020). Meanwhile, Machine Learning has a great advantage in dealing with the nonlinear ecological relationships (Guo et al., 2023). In this session we aim to explore: State of the art and advances in land ecosystem monitoring using RS; Artificial intelligence and automated monitoring for assisting conservation and restoration of land ecosystems; Decision support tools for land ecosystem restoration; Bridging space-time scale gaps and tackling uncertainties; Novel datasets and indicators from RS and/or for training AI systems; Ground truthing methods for RS