A Comparative Analysis of Machine Learning Methods for Algal Bloom Detection Using Remote Sensing Images

Algal blooms are a major environmental challenge for lakes and reservoirs and pose severe threats to water on both aquatic and human health. Conventional algorithms used for algal bloom detection based on remote sensing reflectance proved to be effective in some lakes. However, it is still difficult to obtain high accuracy for multiple lakes using single-threshold-based detection. Currently, machine learning (ML) algorithms have been applied to pinpoint algal bloom locations with excellent results, but the ability of different ML models to be applied in different lakes is still unknown. This article presents the performance of algal bloom detection with commonly used ML algorithms in Chinese eutrophic inland lakes based on Sentinel-2 images. A series of comprehensive tests for accuracy, stability, and robustness was designed for four ML models, including random forest (RF), extreme gradient boosting, artificial neural network, and support vector machine, which were tested in Lake Taihu, Lake Chaohu, and Lake Dianchi. In addition, the index-based methods, including floating algae index and adjusted floating algae index, were also calculated for comparison with ML methods. The results showed that the RF model outperformed other ML models. The comparison results between the RF model and algal indexes revealed that the overall accuracy of RF remained above 0.90. Even with a single lake dataset used as training samples, the RF still maintained a fairly high accuracy of 0.88 for other lakes. To summarize, the four ML models demonstrate promising potential for algal bloom detection across different lakes and provide a practical reference for further applications.

Automated summary

This article presents the performance of multiple machine learning (ML) algorithms in detecting algal blooms in Chinese eutrophic inland lakes. The random forest (RF) model stands out among the four tested ML models, achieving an overall accuracy above 0.90. Even with data from a single lake used as training samples, the RF model maintains a fairly high accuracy of 0.88 for other lakes. These ML models show promising potential for algal bloom detection across different lakes and provide practical references for further applications.