Abstract：
The East Kolkata Wetlands (EKW) in Kolkata, India, span 12,500 ha and are a vital ecological zone providing several benefits, including water purification, flood control, and biodiversity support. This study investigated land use and land cover (LULC) alterations in the EKW from 1991 to 2023, using a random forest (RF) machine learning model. Significant LULC changes were observed over the 32 years, with wetland areas decreasing from 91.2 km2 in 1991 to 33.4 km2 in 2023, reflecting substantial habitat loss and reduced ecosystem services. Conversely, agricultural land expanded from 27.8 km2 to 58.7 km2, driven by economic and food production needs, and built-up areas increased dramatically from 0.2 km2 to 10.5 km2, indicating rapid urbanization. This study evaluated the health, resilience, and ecosystem functionality of EKW by analysing human-induced land use changes and using ecological indicators and landscape metrics. Landscape and class level metrics such as PLAND, largest patch index (LPI), total edge (TE), edge density (ED), number of patches (NP), and patch density (PD) were used to analyse the spatiotemporal dynamics of the wetlands. This study revealed a significant increase in fragmentation, with the number of patches increasing from 2689 in 1991 to 4532 in 2023, despite a consistent decrease in core wetland areas. Ecosystem health indicators, such as the ecosystem structure index (ESI) and landscape deviation degree (LDD), were used to assess landscape metrics and fragmentation changes. The ESI and other metrics revealed significant temporal fluctuations, providing insights into landscape structure, connectivity, and heterogeneity. The ESI improved from 0.87 in 1991 to 1.03 in 2023, indicating enhanced connectivity and diversity. Conversely, the LDD increased from 20.6% to 56.85%, indicating a shift towards impervious surfaces. The vegetation productivity and ecosystem health index (EHI) decreased, indicating biodiversity loss and reduced carbon sequestration. The EHI also dropped from 0.67 to 0.55, signifying ongoing environmental stress. This study emphasizes the need for conservation efforts to maintain the ecological integrity of the EKW amidst urbanization and land use changes and recommends a balanced approach for sustainable urban development and enhanced wetland resilience.
摘要：
加尔各答的东加尔各答湿地（EKW），印度,跨度12500公顷，是一个重要的生态区，提供了几个好处，包括水净化，防洪,生物多样性支持。这项研究使用随机森林（RF）机器学习模型，调查了1991年至2023年EKW中的土地利用和土地覆盖（LULC）变化。在32年中观察到了显著的LULC变化，湿地面积从1991年的91.2平方公里减少到2023年的33.4平方公里，反映出栖息地的大量丧失和生态系统服务的减少。相反，在经济和粮食生产需求的推动下，农业用地从27.8平方公里扩大到58.7平方公里，建成区面积从0.2平方公里急剧增加到10.5平方公里，表明城市化进程迅速。这项研究评估了健康，弹性，通过分析人类引起的土地利用变化并使用生态指标和景观指标，实现EKW的生态系统功能。横向和班级级别指标，如PLAND、最大补丁指数(LPI)，总边缘(TE)，边缘密度(ED)，补丁数量(NP)，和斑块密度（PD）用于分析湿地的时空动态。这项研究表明，碎片化显著增加，尽管核心湿地面积持续减少，但斑块数量从1991年的2689个增加到2023年的4532个。生态系统健康指标，如生态系统结构指数(ESI)和景观偏离度(LDD)，用于评估景观指标和碎片化变化。ESI和其他指标显示出显著的时间波动，提供对景观结构的见解，连通性,和异质性。ESI从1991年的0.87提高到2023年的1.03，表明连通性和多样性增强。相反，LDD从20.6%增加到56.85%，表明向不透水表面转移。植被生产力和生态系统健康指数（EHI）下降，表明生物多样性丧失和碳固存减少。EHI也从0.67降至0.55，表明持续的环境压力。这项研究强调了在城市化和土地利用变化中保持EKW生态完整性的必要性，并建议采取一种平衡的方法来实现可持续城市发展和增强湿地复原力。
