Poyang Lake is the largest freshwater lake in China, an internationally important wetland and the largest migratory bird habitat in Asia. There are numerous seasonal lakes in the Poyang Lake region, and these lakes create a complex and diverse wetland landscape and serve as the main activity area for migratory birds. Seasonal lakes have significant effects on hydrological and ecological processes and are highly susceptible to various changes, but research on seasonal lakes is relatively limited. Based on long-term remote sensing images, this study analyzed the spatiotemporal variation characteristics of seasonal lakes and their influencing factors and ecological effects. The conclusions were as follows: the average water area of seasonal lakes showed a unimodal change during the year, reaching a maximum in July and a minimum in January. The interannual water area fluctuated greatly, and the overall water area showed a small downward trend. In terms of spatial variation, the seasonal lakes were connected with the main lake in the wet season and separated from the main lake in the dry season. Among the natural factors, the influences of temperature, evaporation, and rainfall on the water area of the seasonal lakes were successively reduced. In addition, the incoming water from the Yangtze River had an important influence. Among the human factors, the traditional fishing method “enclosing sublakes in autumn” had the greatest impact on the water area in a specific period, followed by the indirect impact of the Three Gorges Project.

From analysis of the factors affecting the service quality of Web GIS, the key factor in the efficiency of online vector data interactive editing is to control the size of the relevant data set. This paper puts forward an online vector data editing method based on the multi-scale representation data structure DBLG-tree. It designs WindowingQuery for large geometric features. Combined with SimplifyQuery, it can effectively control the vertex scale of the geometric features. At the same time, the sub-feature resulting from WindowingQuery retains the connection with the original feature and can be merged and updated. It provides DownsizeQuery which can directly control the volume of query result from the data set, so as to adapt to the performance of different browsers and mobile devices. Research shows that the execution time of DownsizeQuery under the support of multi-scale representation data structure is stable, and the size of the queried and updated data set is also stable. The upper bounded query time and query results ensure the response time of online interactive visualization and updating of the vector features.