Semi-natural habitats, red-listed plants and abandonment in Norway: patterns and a screening approach for prioritization

• 112,160 ha of farmland were continuously abandoned in Norway (2020–2024). • Only 3.7 % overlapped with mapped semi-natural ecosystems, mostly grasslands. • 3257 records of red-listed vascular plants found in abandoned semi-natural land. • 13.2 % of Norwegian municipalities showed significant NDVI greening trends over 7 years. • Mapping gaps and regional variation point to key areas for targeted management. Agricultural land abandonment is increasingly affecting rural and low-intensity farming regions across Europe, raising concerns about its impact on biodiversity. While some species may benefit from reduced human disturbance, many species in semi-natural ecosystem types depend on traditional agricultural management to maintain their ecological integrity. This study examines whether abandoned agricultural land in Norway contains semi-natural ecosystems that may hold important remnant populations of red-listed plant species and where continued cessation of farming may further threaten these biodiverse ecosystems. Using spatial data on abandoned farmland, semi-natural ecosystem types and species observations, we identify areas of conservation interest and assess the extent to which these areas support endangered species. In addition, we conducted a time-series analysis of vegetation change using NDVI data (2017–2024) to evaluate whether abandonment led to detectable ecological succession. We also analyzed the spatial distribution of abandonment and its correlation with proximity to active farms to understand regional patterns of abandonment. Our results show that only a small percentage (3.7 %) of the abandoned agricultural land considered in this study overlaps with known semi-natural ecosystem types, yet these areas support a significant number of red-listed plant species. The NDVI analysis revealed generally weak but positive greening trends, suggesting early successional changes that are not yet statistically significant across most habitat types. Our method thus suggests a potential approach to allocate limited management resources to key locations. At present, the amount of semi-natural ecosystems is probably underestimated, however, because of limited and time-consuming mapping activity. These findings emphasize the need for more extensive mapping and targeted conservation efforts and highlight the risks posed by abandonment in biodiversity rich semi-natural ecosystem types.