A modified cyanobacteria prediction model based on cellular automata model using N and P concentration reverse data: a case study in Taihu Lake

Abstract

The problem of algal bloom caused by eutrophication has attracted global attention. Many scholars have studied the problem associated with algae bloom, but few have carried out dynamic monitoring, instead focusing on the formation mechanism of cyanobacteria. For our study of the Taihu Lake in China, we used Moderate-Resolution Imaging Spectroradiometer (MODIS) and Landsat remote sensing image data from 2017 to establish a prediction model. First, we used MODIS data to retrieve the concentration of N, P, and chlorophyll a in water. Then, we applied the analytic hierarchy process (AHP) model to the inversion results to construct the diffusion potential index. Finally, we used C# to compile the cellular automata (CA) model. We found that the distribution of cyanobacteria predicted by our method was consistent with the algal bloom situation of Taihu Lake in 2017. The results showed that the method effectively predicts the dynamic transfer of cyanobacteria from outbreak to diffusion in a short period of time, which can help decision-makers monitor lake health.

Identification of driving factors for chlorophyll‐a in multi‐stable shallow lakes of China employing machine learning methods

“…Recently, the algal blooms in Taihu Lake show certain spatial differences and are affected by environmental factors, human activities near the lake and hydrodynamic conditions (Zhang, Hu, et al, 2020). For example, previous studies have reported that due to significant differences in nutrient (especially N and P) concentrations in different regimes of Taihu Lake, the impact of sediment release on algal blooms at various stretches of the lake is stated to be different (Zhao et al, 2022). Therefore, real‐time monitoring (in situ) and long‐term prediction of algae blooms are vital for freshwater conservation and restoration (Rolim et al, 2023).…”