Comparative analysis of land use change prediction models for land and fine wetland types: Taking the wetland cities Changshu and Haikou as examples

With the development of social economy and the influence of human activities, land use has undergone great changes. Land use prediction models have gradually emerged for more rational planning and development of the region. This study took Changshu and Haikou, two wetland cities, as examples to explore the prediction effects of land and fine wetland types in PLUS, FLUS and CLUES models, and combined CA-Markov model to make a detailed summary of the composition, use requirements, advantages and disadvantages of each model. The results showed that: (1) From 2015 to 2020, the urban area in Changshu and Haikou showed a significant increase, with growth rates of 16.61% and 21.37%, respectively. However, the total area of wetland remained above 10%, which was not greatly affected; (2) The average overall accuracy of the PLUS, FLUS and CLUES models in the prediction of Changshu and Haikou was 0.85, 0.86, and 0.84, respectively, while Kappa was 0.79, 0.81, and 0.78, respectively. Especially, the prediction in the fine wetland types also achieved high accuracy. (3) FLUS and CLUES had more stable results in the comparison of spatial results, deviation on the prediction of some land use categories was easily born in PLUS. Each model has its own advantages and disadvantages in practical application and operation. The research results of this paper are intended to let more people understand different land use prediction models and provide reference basis for the improvement and application of the models.