Machine learning application in modelling marine and coastal phenomena: a critical review

Abstract

This study provides an extensive review of over 200 journal papers focusing on Machine Learning (ML) algorithms’ use for promoting a sustainable management of the marine and coastal environments. The research covers various facets of ML algorithms, including data preprocessing and handling, modeling algorithms for distinct phenomena, model evaluation, and use of dynamic and integrated models. Given that machine learning modeling relies on experience or trial-and-error, examining previous applications in marine and coastal modeling is proven to be beneficial. The performance of different ML methods used to predict wave heights was analyzed to ascertain which method was superior with various datasets. The analysis of these papers revealed that properly developed ML methods could successfully be applied to multiple aspects. Areas of application include data collection and analysis, pollutant and sediment transport, image processing and deep learning, and identification of potential regions for aquaculture and wave energy activities. Additionally, ML methods aid in structural design and optimization and in the prediction and classification of oceanographic parameters. However, despite their potential advantages, dynamic and integrated ML models remain underutilized in marine projects. This research provides insights into ML’s application and invites future investigations to exploit ML’s untapped potential in marine and coastal sustainability.

Remote Measurement of Tide and Surge Using a Deep Learning System with Surveillance Camera Images

“…Machine learning and deep learning [22], two branches of artificial intelligence [23], are proving to be versatile tools in the examination of coastal environments [24]. This, in turn, can facilitate the informed and sustainable planning and management of coastal regions [25].…”

Comparative study for the performance of pure artificial intelligence software sensor and self-organizing map assisted software sensor in predicting 5-day biochemical oxygen demand for Kauma Sewage Treatment Plant effluent in Malawi

“…AI algorithms can be broadly categorized into supervised learning or unsupervised learning paradigms (El Alaoui El Fels et al, 2023). Supervised learning involves training algorithms on labeled data, wherein every input-output pair is explicitly supplied during the training process (Pourzangbar et al, 2023). As a result, the algorithm can learn a mapping from inputs to outputs and use that knowledge to make judgments or predictions on new, unobserved data.…”

“…As a result, the algorithm can learn a mapping from inputs to outputs and use that knowledge to make judgments or predictions on new, unobserved data. Conversely, unsupervised learning involves training algorithms on unlabeled data, where the algorithm's task is to figure out the data's underlying structure or patterns without direct supervision (El Alaoui El Fels et al, 2023;Pourzangbar et al, 2023). This frequently entails using dimensionality reduction techniques or grouping comparable data elements.…”