Regional scale analysis of land cover dynamics in Kerala over last two decades through MODIS data and statistical techniques

The impact of climate change and the effect of rapid urbanisation in a region can be best identified by detecting land use and land cover changes. The present research attempts to statistically evaluate the land cover changes in the state of Kerala over two decades (2001-2019), using land cover derived by the supervised classification of Moderate Resolution Imaging Spectroradiometer (MODIS) datasets. Land cover classification, based on the International Geosphere-Biosphere Programme (IGBP), differentiated nine land cover types in the region such as forests (F), shrublands/scrublands (SL), savannas (S), croplands (CL), grasslands (GL), urban and built-up lands (UBL), cropland/natural vegetation mosaics (CLNVM), water body (WB) and other types (OT). Among the land cover types, savannas cover more than 41% of the total area, followed by forests (20%) and cropland/natural vegetation mosaics (14%). During the analysis period, forests, cropland/natural vegetation mosaics and built-up lands showed increases in their area coverage. Furthermore, the decadal (D1 and D2) analysis of land cover dynamics reveals that maximum spatial variation in area coverage of individual land cover occurred in the second decade, i.e. D2 (2011-2019). Accurate examination of the distribution of land cover types indicates the elevation dependant gradational land cover pattern in the study area. Trend statistics of land cover assessed through Mann-Kendall and linear regression analysis also showed similar results with a statistically significant increase in the area covered by forests, shrublands, urban and built-up lands and cropland/natural vegetation mosaics. In contrast, savannas and grasslands showed a significant decreasing trend. Together, the results indicate an increasing pattern of urbanisation and the development of cropland/natural vegetation mosaics, irrespective of elevation and land cover types point towards the degradation and fragmentation of strong and healthy vegetation cover in the region. Furthermore, the present research finding will help researchers and policy makers to concentrate on areas that showed maximum changes in land cover to derive accurate and sustainable environmental protection and development plans.