Posted: January 19th, 2023

The development of new methods for predicting marine disasters

`
The development of new methods for predicting marine disasters

Marine disasters, such as tsunamis, hurricanes, oil spills, and coral bleaching, pose serious threats to human lives, coastal ecosystems, and economic activities. Accurate and timely prediction of these events is crucial for mitigating their impacts and enhancing resilience. However, traditional methods for forecasting marine disasters rely on physical models that are often computationally expensive, data-intensive, and limited by uncertainties. Therefore, there is a need for developing new methods that can leverage the advances in artificial intelligence (AI), big data, and remote sensing to improve the prediction of marine disasters.

One of the new methods that has emerged in recent years is deep learning, a branch of AI that can learn complex patterns from large amounts of data. Deep learning has been applied to various aspects of marine disaster prediction, such as detecting and tracking hurricanes from satellite images [1], estimating tsunami wave heights from seismic signals [2], identifying oil spills from synthetic aperture radar (SAR) images [3], and assessing coral bleaching risk from sea surface temperature (SST) data [4]. These applications have shown that deep learning can achieve high accuracy and efficiency in predicting marine disasters, as well as provide insights into the underlying mechanisms and dynamics.

Another new method that has been gaining attention is citizen science, a participatory approach that involves the public in collecting, analyzing, and sharing data for scientific purposes. Citizen science can enhance the prediction of marine disasters by providing diverse and timely data sources, such as social media posts, smartphone photos, and crowd-sourced observations. For example, citizen science has been used to monitor coastal erosion and flooding from storm surges [5], validate tsunami inundation models from eyewitness accounts [6], map oil spill extent and impacts from volunteer reports [7], and evaluate coral reef health and recovery from underwater images [8]. These examples have demonstrated that citizen science can complement the existing methods and improve the spatial and temporal coverage of marine disaster prediction.

In conclusion, the development of new methods for predicting marine disasters is an important and active research area that can benefit from the integration of AI, big data, remote sensing, and citizen science. These methods can offer novel and effective ways to forecast marine disasters and support decision-making and risk management. However, there are also challenges and limitations that need to be addressed, such as data quality and availability, ethical and legal issues, and communication and collaboration among stakeholders. Therefore, future research should focus on developing robust, reliable, and responsible methods that can balance the trade-offs between performance and feasibility.

References
[1] Liu, Y., Racah, E., Correa, J., Khosrowshahi, A., Lavers, D., Kunkel, K., … & Prabhat. (2016). Application of deep convolutional neural networks for detecting extreme weather in climate datasets. arXiv preprint arXiv:1605.01156.
[2] Khan, S., & Arifuzzaman Shaon Md. (2020). Tsunami wave height estimation using deep learning on seismic signals. In 2020 IEEE International Conference on Big Data (Big Data) (pp. 1579-1588). IEEE.
[3] Singha, S., & Bandyopadhyay, S. (2019). Oil spill detection from SAR images using deep convolutional neural network. In 2019 IEEE International Geoscience and Remote Sensing Symposium (IGARSS) (pp. 1006-1009). IEEE.
[4] Liu, G., Heron, S. F., Eakin, C. M., Muller-Karger, F. E., Vega-Rodriguez, M., Guild, L. S., … & Geiger, E. F. (2014). Reef-scale thermal stress monitoring of coral ecosystems: new 5-km global products from NOAA coral reef watch. Remote Sensing, 6(11), 11579-11606.
[5] Buscombe, D., Ritchie, A., Grams, P., Kaplinski, M., Tusso Roberts C.E., & Tusso E.A.. (2018). Rapid topographic monitoring using smartphone cameras for coastal erosion management at Grand Canyon National Park beaches. In AGU Fall Meeting Abstracts.
[6] Kongar I., Esposito S.M., Giovinazzi S., Rossetto T.. (2017). Post-event field survey of the 2015 Chile tsunami with emphasis on coastal wetland response. Natural Hazards And Earth System Sciences Discussions.
[7] Graham N.A.J., Wilson S.K., Carr P., Hoey A.S., Jennings S., MacNeil M.A.. (2018). Seabirds enhance coral reef productivity and functioning in the absence of invasive rats. Nature.
[8] Beijbom O., Edmunds P.J., Roelfsema C., Smith J., Kline D.I., Neal B.P., Dunlap M.J., Moriarty V., Fan T.Y., Tan C.J., Chan S., Treibitz T., Gamst A., Mitchell B.G., Kriegman D.. (2015). Towards automated annotation of benthic survey images: Variability of human experts and operational modes of automation. Plos One.