{"title":"Modeling Wetland Habitat Quality in the Rarh Tract of Eastern India","authors":"Rumki Khatun, Somen Das","doi":"10.1007/s13157-024-01849-w","DOIUrl":null,"url":null,"abstract":"<p>Along with wetland loss, wetland habitat quality degradation is a growing concern that requires immediate attention. The current study aimed to assess the Wetland Habitat Quality State (WHQS) of Rarh region, Murshidabad, West Bengal. WHQS used a total of seventeen metrics, including water quality, hydrology, and landscape composition. Machine learning techniques such as ANN, SVM, RF, BAGGING, and REP-TREE were used to model WHQS. The effectiveness of the models was evaluated using statistical techniques such as the Receiver operating characteristics (ROC) curve. According to machine learning models, 6% of the area fall under very weak habitat quality zones in 1990 which increased by 15%, 26%, 41% in 2000, 2010 and 2020, respectively. Very strong portions of wetland area have been decreased from 32.74% in 1990 to 20.72% in 2020. The current study's findings could provide comprehensive research on the monitoring of habitat quality in wetlands, which will serve as the foundation for developing water resource management plans for the conservation, management, and restoration of wetlands.</p>","PeriodicalId":23640,"journal":{"name":"Wetlands","volume":"72 1","pages":""},"PeriodicalIF":1.8000,"publicationDate":"2024-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Wetlands","FirstCategoryId":"93","ListUrlMain":"https://doi.org/10.1007/s13157-024-01849-w","RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ECOLOGY","Score":null,"Total":0}
引用次数: 0
Abstract
Along with wetland loss, wetland habitat quality degradation is a growing concern that requires immediate attention. The current study aimed to assess the Wetland Habitat Quality State (WHQS) of Rarh region, Murshidabad, West Bengal. WHQS used a total of seventeen metrics, including water quality, hydrology, and landscape composition. Machine learning techniques such as ANN, SVM, RF, BAGGING, and REP-TREE were used to model WHQS. The effectiveness of the models was evaluated using statistical techniques such as the Receiver operating characteristics (ROC) curve. According to machine learning models, 6% of the area fall under very weak habitat quality zones in 1990 which increased by 15%, 26%, 41% in 2000, 2010 and 2020, respectively. Very strong portions of wetland area have been decreased from 32.74% in 1990 to 20.72% in 2020. The current study's findings could provide comprehensive research on the monitoring of habitat quality in wetlands, which will serve as the foundation for developing water resource management plans for the conservation, management, and restoration of wetlands.
期刊介绍:
Wetlands is an international journal concerned with all aspects of wetlands biology, ecology, hydrology, water chemistry, soil and sediment characteristics, management, and laws and regulations. The journal is published 6 times per year, with the goal of centralizing the publication of pioneering wetlands work that has otherwise been spread among a myriad of journals. Since wetlands research usually requires an interdisciplinary approach, the journal in not limited to specific disciplines but seeks manuscripts reporting research results from all relevant disciplines. Manuscripts focusing on management topics and regulatory considerations relevant to wetlands are also suitable. Submissions may be in the form of articles or short notes. Timely review articles will also be considered, but the subject and content should be discussed with the Editor-in-Chief (NDSU.wetlands.editor@ndsu.edu) prior to submission. All papers published in Wetlands are reviewed by two qualified peers, an Associate Editor, and the Editor-in-Chief prior to acceptance and publication. All papers must present new information, must be factual and original, and must not have been published elsewhere.