Grasslands in Flux: A Multi-Decadal Analysis of Land Cover Dynamics in the Riverine Dibru-Saikhowa National Park Nested Within the Brahmaputra Floodplains

In recent years, remote sensing and geographic information systems (GISs) have become essential tools for effective landscape management. This study utilizes these technologies to analyze land use and land cover (LULC) changes in Dibru-Saikhowa National Park, a riverine ecosystem in Assam, India, from its designation as a national park in 2000 through 2024. The satellite imagery was used to classify LULC types and track landscape changes over time. In 2000, grasslands were the dominant land cover (28.78%), followed by semi-evergreen forests (25.58%). By 2013, shrubland became the most prominent class (81.31 km2), and degraded forest expanded to 75.56 km2. During this period, substantial areas of grassland (29.94 km2), degraded forest (10.87 km2), semi-evergreen forest (12.33 km2), and bareland (10.50 km2) were converted to shrubland. In 2024, degraded forest further increased, covering 80.52 km2 (23.47%). This change resulted since numerous areas of shrubland (11.46 km2) and semi-evergreen forest (27.48 km2) were converted into degraded forest. Furthermore, significant shifts were observed in grassland, shrubland, and degraded forest, indicating a substantial and consistent decline in grassland. These changes are largely attributed to recurring Brahmaputra River floods and increasing anthropogenic pressures. This study recommends a targeted Grassland Recovery Project, control of invasive species, improved surveillance, increased staffing, and the relocation of forest villages to reduce human impact and support community-based conservation efforts. Hence, protecting the landscape through informed LULC-based management can help maintain critical habitat patches, mitigate anthropogenic degradation, and enhance the survival prospects of native floral and faunal assemblages in DSNP.