Mississippi's coastal wetland Land Use/ Land Cover (LULC) change analysis: Quantitative assessments using remote sensing

Link To Share This Poster: https://cdmcd.co/3wqGQ6

Ranjani Kulawardhana, Dept. of Biology, Jackson State University, Jackson, MS

Background/Question/Methods
Coastal wetland ecosystems serve many important ecosystem products and services. However, their spatial extents are decreasing worldwide at alarming rates and are attributed largely to LULC conversions resulting from natural and human-induced disturbances. Thus, quantitative assessments on the loss of wetland land extent and the evaluation of the causes of LULC change (LULCC) is of significant importance. Remote sensing data and techniques provide effective means for mapping and characterizing LULCC that are applicable even at larger spatial scales. This study was carried out to map and assess Mississippi's coastal wetland LULCCs using remote sensing. Specific objectives were to: 1) develop remote sensing-based LULC classification scheme applicable at local to regional scale for monitoring and assessment of wetland LULC change; 2) evaluate the applicability of medium resolution remote sensing data (Landsat) to map and characterize wetland land use/ land cover (LULC); and 3) map and quantify wetland LULCC using multi-temporal RS data. This study focused on the coastal wetlands and adjacent landscapes of Grand Bay National Estuarine Research Reserve (GB NERR) of Mississippi, USA. LULC maps were derived using medium and very high-resolution RS data and were evaluated for their accuracies.
Results/Conclusions
Our findings indicate LULC mapping accuracies of medium resolution RS data as comparable to that of very high-resolution RS data with overall classification accuracies of 68% & 82% and kappa statistics of 0.58 & 0.76 for Landsat- and Worldview 2-derived LULC maps, respectively. These findings suggest that RS based LULC classification using Landsat data as a promising approach for quantification of wetland LULC change. Wetland LULC change analysis based on Landsat data indicate gradual decrease in percent area of wetlands within the GB NERR site from 1992 to 2015, however this reduction was not significant across different years (1992, 2005 & 2015).