Vulnerability and risk assessment mapping of Bhitarkanika national park, Odisha, India using machine-based embedded decision support system
Shantakar Mohanty,Sk. Mustak,Dharmaveer Singh,Thanh Van Hoang,Manishree Mondal,Chun-Tse Wang
DOI: https://doi.org/10.3389/fenvs.2023.1176547
IF: 5.411
2023-09-29
Frontiers in Environmental Science
Abstract:The vulnerability and flood risk assessment of Bhitarkanika National Park in Odisha, India, was conducted using a data-driven approach and a machine-based embedded decision support system. The park, located in the estuaries of the Brahmani, Baitarani, Dharma, and Mahanadi river systems, is home to India's second-largest mangrove environment and the world's most active and diverse saline wetland. To evaluate its vulnerability and risk, various threats were considered, with a focus on floods. Satellite imageries, such as Landsat 8 OLI, SRTM digital elevation model, open street map, Google pro image, reference map, field survey, and other ancillary data, were utilized to develop vulnerability and risk indicators. These indicators were then reclassified into 'Cost' and 'Benefit' categories for better understanding. The factors were standardized using the max-min standardization method before being fed into the vulnerability and risk model. Initially, an analytical hierarchy approach was used to develop the model, which was later compared with machine learning algorithms (e.g., SVM) and uncertainty analysis indices (e.g., overall accuracy, kappa, map quality, etc. ). The results showed that the SVM-RBF machine learning algorithm outperformed the traditional geostatistical model (AHP), with an overall accuracy of 99.54% for flood risk mapping compared to AHP's 91.12%. The final output reveals that a large area of Bhitarkanika National park falls under high flood risk zone. The Eastern coastal regions of Govindapur, Kanhupur, Chinchri, Gobardhanpur and Barunei fall under high risk zone of tidal floods, The Northern and western regions of Ramachandrapur, Jaganathpur, Kamalpur, Subarnapur, Paramanandapur, etc. , Fall under high risk region of riverine floods. The study also revealed that the areas covered with mangroves have a higher elevation and hence are repellent to any kind of flood. In the event of a flood high priority conservation measures should be taken along all high flood risk areas. This study is helpful for decision-making and carrying out programs for the conservation of natural resources and flood management in the national park and reserve forest for ecological sustainability to support sustainable development goals (e.g., SDGs-14, 15).
environmental sciences