Unveiling Harmful Algal Bloom Dynamics with Sparse Identification of Nonlinear Dynamics and the Model-based Web Platform for Lakes

The proliferation of harmful algal blooms (HABs) is a significant environmental issue exacerbated by climate change. They have many detrimental impacts on public health, recreational activities, ecological equilibrium, animals, fisheries, microbiota, water quality, and the economy. HABs can be attributed to various factors, including water pollution resulting from agricultural activities, discharges from wastewater treatment plants, leakages from sewer systems, natural factors such as pH and light levels, and the impacts of climate change, such as increased water temperature, reduced water flows due to droughts, and water-related disasters like flooding and inundation. Although numerous causes of HABs are acknowledged, the mechanisms by which toxin-producing algae develop, as well as the crucial processes and components that contribute to their proliferation, remain unknown. This lack of understanding is primarily due to the unique dynamics of each lake's algal population and the nonlinear nature of the conditions that influence these dynamics. Modeling HABs in a changing climate is critical to meeting sustainable development targets for clean water and sanitation. Nevertheless, the absence of adequate and sufficient data on HABs poses a substantial obstacle for research endeavors. This study utilized the sparse identification nonlinear dynamics (SINDy) technique to model microcystin, which is a toxin produced by algae, using dissolved oxygen as a variable for water quality and evaporation as a meteorological parameter. SINDy is a state-of-the-art approach that integrates sparse regression and machine learning methodologies to reconstruct the analytical representation of a dynamic system. Furthermore, a web platform was developed that utilizes models and is accessible through the web. This tool aims to promote and enhance environmental education, increase the public's awareness of these events, and generate more efficient solutions to them by using what-if scenarios. This web platform allows tracking the status of HABs in lakes and observing the impact of specific parameters on harmful algae formation. On an interactive and user-friendly platform, users may effortlessly share photographs of HABs in lakes, enabling others to monitor the lakes' status.