Remote sensing monitoring and analysis based on multi-sourced data of mangroves in Pearl Bay, Guangxi

To understand the spatiotemporal evolution of mangroves in Pearl Bay, Guangxi over the past 30 years, we used Landsat TM/OLI satellite remote sensing images as the data source and combined them with Sentinel-2 satellite remote sensing images. Based on the spectral characteristics of the ground objects, we constructed a decision tree classification rule. Through the decision tree classification method based on tasseled cap transformation, we extracted the spatial distribution information of mangroves in 1995, 2000, 2005, 2010, 2015 and 2023, calculated the dynamic degree and landscape pattern index, and analyzed the dynamic spatiotemporal evolution of mangroves in the study area from 1995 to 2023. Results indicate that over the past nearly 30 years, mangrove forests in Pearl Bay have changed significantly in terms of area composition and spatial location, presenting an overall trend of decreasing first and then increasing. The period with the highest dynamic degree is 2005-2010, during which the mangrove area increased to 836.21 hectares, with a dynamic degree of 3.59%. The calculation of the landscape pattern index shows that the fragmentation and landscape differentiation degree generally exhibit a trend of increasing first and then decreasing, while the aggregation degree is relatively stable. The spatial evolution of the mangroves are related to natural factors such as precipitation, temperature, sediment deposition, and pest damage, and are closely related to human activities such as pond culture.