Semi-Automated Methods for Wetland Mapping Using Landsat ETM+: A Case Study from Tsunami Affected Panchyats of Alappad & Arattupuzha, South Kerala

The global extent of wetlands are under continuous loss and degradation which have resulted in diminished ecosystem services, and also are facing several challenges. So systematic mapping and inventorying is essential for the assessment of their ecological health and integrity. The traditional methods of monitoring wetlands by using field survey and labor force are time and labour intensive. Modern and advanced technologies such as remote sensing and GIS offer the opportunity to map and inventory wetlands rapidly and consistently, irrespective of the geographic location. The present study utilizes a semi-automated method for mapping wetlands using LANDSAT ETM+ and ASTER DEM data. Various spectral and terrain indices, viz. , Normalized Difference Water Index (NDWI), Normalized Difference Vegetation Index (NDVI), Tasseled Cap Wetness Index (TCWI) and terrain slope were used for the extraction of wetland areas. The thresholds for the indices were finalized based on trial and error method in order to avoid overestimation or underestimation of wetland areas. Nearly 80 % of accuracy was observed for the present wetland delineation methodology by comparing the data with Google Earth. The methodology was evaluated for the Tsunami affected Panchayats of Alappad & Arattupuzha, Kerala, India. As the villages lie in the coastal areas, the landscape consists of almost lowland plains. The present methodology can be used to understand the spatial distribution of different wetlands and association with disaster risk reduction initiatives.