Projected effects of land use change on future soil erosion in southern Italy

Anthropogenic activities are rapidly transforming landscapes, altering ecosystem services across spatial and temporal scale. This study develops an integrated framework combining Dyna-CLUE and InVEST models to evaluate soil erosion and sediment delivery under projected land use/land cover change (LUCC) and rainfall erosivity in Campania region, southern Italy. The novelty lies in merging spatially explicit scenario modeling with ecosystem service assessment, starting from a dense network of field data, at territorial scale. The study area, characterized by diverse land uses (forests, 34%, agriculture, 46%, and urban settlements, 13%) and marked altitudinal and climatic gradients, provides an ideal setting to assess erosion regulation. The Sediment Delivery Ratio (SDR) allowed estimating the fraction of eroded soil reaching water bodies as a proxy for erosion regulation services. Historical LUCC was derived from CORINE maps (1960, 1990, 2000, 2018). Future scenarios (2030–2060) were simulated with Dyna-CLUE and combined with rainfall erosivity data from 259 meteorological stations and soil erodibility from 3316 samples. Results indicate consistent improvement in erosion regulation, driven by forest expansion and farmland reduction. Land at severe erosion risk (>11 t ha −1 yr −1 ) declined from 35.1% (1960) to 16.3% (2018), with a projected 15.2% by 2060. Conversely, areas with tolerable erosion (<2 t ha −1 yr −1 ) increased from 44.7% (1960) to 64.2% (2018), reaching 67.4% in 2060. Rainfall erosivity shows a slight regional decrease, though local vulnerabilities persist. The proposed approach provides a reproducible tool for scenario-based planning, supporting policies to mitigate erosion risk and strengthen land resilience. • Land use, together with rainfall dynamics, significantly influences soil erosion trends. • An integrated Dyna-CLUE/InVEST framework links land change with erosion services. • Extensive soil and rainfall datasets enhance model reliability and spatial accuracy. • Forest expansion and farmland decline substantially reduce severe erosion risk. • Scenario-based modeling supports resilient land planning and erosion mitigation.