A multiscale analysis of grassland degradation. A case study from Northeastern Romania

Based on the general assumption that global climate change and anthropogenic interventions have alarmingly affected grasslands worldwide, this study aims to investigate the current state of grassland degradation by closely examining how degradation processes are perceived across different spatial and temporal scales.Climate change and human impact are not the sole causes of grassland degradation. The climatic factor is often separated from other influences, such as hydrology or soil properties. Vegetation dynamics are closely tied to land-surface hydrology, with positive effects when water availability is adequate and adverse effects when water is scarce or excessive. Several studies have also emphasised the influence of landform on vegetation quality and distribution. The geomorphological factor is often linked to high rates of degradation, caused by landslides and gully erosion. In terms of both geological structure and soil properties, geomorphological features are challenging to define.Even if most researchers choose to assess them together, several studies have highlighted the need to distinguish between triggers to ensure appropriate mitigation. By supporting this statement, our analysis focuses on identifying and separating the main drivers of grassland degradation.Some of the most qualitative and widely applied methods are based on the Normalised Difference Vegetation Index (NDVI), which is considered a proxy for grassland degradation. Thus, to determine the current status of grassland degradation in the Moldavian Plateau, a method for analysing vegetation dynamics is proposed, using NDVI Landsat 8 OLI data from 2013 to 2020 (30m); MODIS data from 2000 to 2023 (250m), and AVHRR data merged with MODIS at 9.5 km spatial resolution, materialised through the PKU GIMMS NDVI dataset available from 1982 to 2022.The method applied separates the multiannual NDVI trend from the seasonal component. The NDVI trend analysis is essential because it provides the information needed to investigate and identify the leading degradation agents.The high-resolution analyses captured fine-scale features, while the medium- and low-resolution analyses provided a clear picture of the primary drivers of grassland degradation. The proper association of local (e.g., overgrazing) and regional (e.g., drought) factors contributes to a better understanding of the degradation phenomenon and supports sustainable measures.Although the analysis is more qualitative than quantitative, it emphasises the importance of local analysis in the global process assessment. Moving from one level of spatial analysis to another, we found considerable differences that can affect perceptions of the impact induced by a specific phenomenon, in this case, global climate change. At the scale of the Moldavian Plateau, grasslands remain stable from a climatic perspective, while the primary problem is associated with anthropogenic interventions.