A multi-module with a two-way feedback method for Ulva drift-diffusion

Abstract

The outbreak of Ulva in the Yellow Sea has seriously affected marine ecology and economic activities. Therefore, effective prediction of the distribution of Ulva is of great significance for disaster prevention and reduction. However, the prediction method of Ulva is mainly based on numerical simulation. There are two problems with these methods. First is that the initial distribution of Ulva is simulated using independent pixel-level particles. Besides, the influence of Ulva growth and diffusion on the drift is not considered. Therefore, this paper proposes a multi-module with a two-way feedback method (MTF) to solve the above problems. The main contributions of our approach are summarized as follows. First, the initialization module, the generation and elimination module, and the drive module are composed in our way. Second, we proposed an initialization method using rectangle objects to simulate the Ulva distribution extracted from remote sensing images. Thirdly, the drift and diffusion mechanism of the Ulva is considered to realize the two-way feedback between the generation and elimination module and the drive module. The results of our experiments show that the MTF performs better than the traditional method in predicting the drift and diffusion of Ulva. The code is already publicly available at https://github.com/UPCGIT/A-multi-module-with-a-two-way-feedback-method-for-Ulva-drift-diffusion.