A Remote Sensing and Machine Learning-Based Approach to Forecast the Onset of Harmful Algal Bloom

Abstract

In the last few decades, harmful algal blooms (HABs, also known as “red tides”) have become one of the most detrimental natural phenomena in Florida’s coastal areas. Karenia brevis produces toxins that have harmful effects on humans, fisheries, and ecosystems. In this study, we developed and compared the efficiency of state-of-the-art machine learning models (e.g., XGBoost, Random Forest, and Support Vector Machine) in predicting the occurrence of HABs. In the proposed models the K. brevis abundance is used as the target, and 10 level-02 ocean color products extracted from daily archival MODIS satellite data are used as controlling factors. The adopted approach addresses two main shortcomings of earlier models: (1) the paucity of satellite data due to cloudy scenes and (2) the lag time between the period at which a variable reaches its highest correlation with the target and the time the bloom occurs. Eleven spatio-temporal models were generated, each from 3 consecutive day satellite datasets, with a forecasting span from 1 to 11 days. The 3-day models addressed the potential variations in lag time for some of the temporal variables. One or more of the generated 11 models could be used to predict HAB occurrences depending on availability of the cloud-free consecutive days. Findings indicate that XGBoost outperformed the other methods, and the forecasting models of 5–9 days achieved the best results. The most reliable model can forecast eight days ahead of time with balanced overall accuracy, Kappa coefficient, F-Score, and AUC of 96%, 0.93, 0.97, and 0.98 respectively. The euphotic depth, sea surface temperature, and chlorophyll-a are always among the most significant controlling factors. The proposed models could potentially be used to develop an “early warning system” for HABs in southwest Florida.

Machine-Learning Classification of SAR Remotely-Sensed Sea-Surface Petroleum Signatures—Part 1: Training and Testing Cross Validation

“…Others have proposed different divisions for various ML methods: linear versus non-linear, tree-based versus non-tree-based, etc. [70].…”

Know Before You Go: Data-Driven Beach Water Quality Forecasting

“…Data-driven techniqueswhich take advantage of statistical and algorithmical relationships among data sets and are the primary tools of the emerging field of environmental data science can be effective in extracting meaning from complex data . Particularly, environmental data science is well-equipped to leverage the increasingly large and more real-time environmental data sets that are available from sources such as remote sensing platforms, , ecosystem monitoring stations, , individual sampling campaigns, and model output. , As a result, data-driven hazard forecasting applications previously documented include predicting flooding, air pollution, foreign species invasion, and harmful algal blooms. , …”

“…Both studies focus on lacustrine beaches with forecast lead times (the hours or days between when a prediction is issued and when it is valid) of 1 day or less. While there are examples of forecasting other beach parameters days in advance (such as tide level, harmful algal blooms, and air quality), ,, to our knowledge no work exists testing FIB forecasts at marine beaches.…”

Red Tide Detection Based on Improved DenseNet Network—Example of Red Tide Detection from Geostationary Ocean Color Imager Data in Bohai Sea

“…(10)(11)(12) By correcting, synthesizing, analyzing, and interpreting satellite imagery, the results of processing can be analyzed to understand red tide characteristics and to estimate red tide information such as the extent, range, and general trend of development of red tide, providing an important means for rapidly synchronized, spatially extensive, and high-frequency continuous monitoring of red tide. (13) There are three main classes of methods using satellites to detect red tide. The first class involves the identification of red tide through parameters such as chlorophyll content and temperature.…”