Biodiversity characterization at landscape level using geospatial modelling technique Original Research Article

Geospatial presentation of habitat status has become a key issue for planning conservation. Biodiversity characters of the habitat provide the basis of prioritizing the sites in conservation effort. Since India is identified as one of the 12-mega biodiversity countries in the world, this task is of great significance. In India, biogeographers have long since recognized North-East India as one of the most diverse regions with respect to flora and fauna. Since historic times, human interventions had started influencing the landuse and landcover of this region. Deforestation is mainly attributed to shifting cultivation and commercial logging of timber. This has resulted in fragmentation of the landscape. Landscape fragmentation and degradation have been identified as fundamental reasons for biodiversity loss. The paper uses landscape ecological principles for biodiversity characterization. Satellite remote sensing data have been used for characterization of the landscape and stratification for ground inventory. A geographic information system has been used to spatially model the disturbance regimes and to integrate the ground based nonspatial data with the spatial characters of the landscape. The vegetation types and landuse of Meghalaya have been mapped using Indian remote sensing (IRS) time series data. The various parameters (viz. patch shape, patch size, number of patches, porosity, fragmentation and juxtaposition) have been analyzed on the most recent landcover map to spatially present the disturbance regimes. A spatial model incorporating ground based biodiversity attributes of the landscape elements, landuse change patterns, disturbance regimes of the landscape and terrain complexity have been used to delineate the spatial pattern of biological richness. This has been achieved using the landscape analysis package (LAP) developed at the Indian Institute of Remote Sensing. Habitat evaluation using ground based data and their spatial organization have been found to provide reliable information on the biodiversity distribution pattern. The present approach for prioritizing the biodiversity rich sites have the advantage of integrating spatial, non-spatial information and horizontal relationships. The information is finally presented in space and time. This approach will facilitate conservation prioritization, systematic inventory and continuous monitoring.