Wetland Monitoring Application in Panjin City Based on Remote Sensing Data of Wide-Band Imaging Spectrometer of Tiangong-2

Abstract

The Liaohe Delta wetland is important in China for its abundant resources. In addition, the current situation of Panjin city wetland is significant for protecting and restoring the Liaohe wetland ecology. The suitability of Tiangong-2 remote sensing imagery and non-remote sensing data to coastal wetland monitoring was studied in this paper. Based on spectral features of various areas, a method of partition classification is proposed. Information about Panjin wetland resources, including four types of natural wetland and three types of artificial wetland was extracted by spectral analysis combining feature shape, patch area and texture characteristics, and classification accuracy was evaluated with Landsat data. The overall accuracy of the partition classification method reached 89.89%, and the kappa coefficient was 0.8748, The feasibility of remote sensing extraction of coastal wetland information in Panjin city based on Tiangong-2 data is validated. The research show that the wetland area is 3184.89 km², of which the natural wetland area is 1129.98 km², accounting for 40.53% of the total area; and the artificial wetland area is 1893.91 km², accounting for 59.47% of the total area. Among these, paddy fields occupy the largest area, accounting for 55.29% of the total wetland area. In order of decreasing area, the others are reed swamps, mudflat wetlands, Suaeda salsa swamps, aquaculture farms, reservoirs/ponds, and river wetland.