Prediction of chlorophyll-a as an index of harmful algal blooms using machine learning models

Authors: BUSARI, IBRAHIM - Clemson University; SAHOO, DEB - Clemson University; Harmel, Daren; HAGGARD, BRIAN - University Of Arkansas

Submitted to: American Society of Agricultural and Biological Engineers
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/12/2024
Publication Date: 3/1/2024
Citation: Busari, I., Sahoo, D., Harmel, R.D., Haggard, B. 2024. Prediction of chlorophyll-a as an index of harmful algal blooms using machine learning models. American Society of Agricultural and Biological Engineers. 2(2):53-61. https://doi.org/10.13031/jnrae.15812.
DOI: https://doi.org/10.13031/jnrae.15812

Interpretive Summary: The complex dynamics of freshwater harmful algal blooms (HABs) necessitate proactive monitoring approaches to mitigate their impacts. The rapid breakthrough in computing prowess and statistical advances is triggering the development of data-driven techniques such as machine learning (ML) models. ML models have been shown in different fields to be instrumental in finding patterns in data usable for explaining relationships in observed data. The findings reveal the potential with ML models, thereby shedding light on crucial factors that necessitate careful deliberation by researchers and policymakers in determining the most suitable approaches to embrace for monitoring HABs.

Technical Abstract: The complex dynamics of freshwater harmful algal blooms (HABs) necessitate proactive monitoring approaches to mitigate their impacts. The rapid breakthrough in computing prowess and statistical advances is triggering the development of data-driven techniques such as machine learning (ML) models. ML models have been shown in different fields to be instrumental in finding patterns in data usable for explaining relationships in observed data. The selected models for this study are regression tree, random forest (RF), multilayer perceptron (MLP), support vector regression (SVR), long short-term memory (LSTM), and gated recurrent unit (GRU) models, with the last two models developed to consider the temporal sequence of obtained water quality datasets. The results showed varying performance with different data splitting ratios, although the RF model outperformed the SVR, MLP and regression tree models. LSTM model predicted temporal dynamics of chlorophyll-a better than GRU, although with more runtime, and showed the potential for developing real-time monitoring of HABs and development of early warning systems. The findings reveal the robustness of the chosen ML models, thereby shedding light on crucial factors that necessitate careful deliberation by researchers and policymakers in determining the most suitable approaches to embrace for monitoring HABs.