An efficient artificial intelligence model for prediction of tropical storm surge
Process-based models have been widely used for storm surge predictions, but their high computational demand is a major drawback in some applications such as rapid forecasting. Few efforts have been made to employ previous databases of synthetic/real storms and provide more efficient surge predictions (e.g. using storm similarity of an individual storm to those in the database). Here, we develop an alternative efficient and robust artificial intelligent model, which predicts the peak storm surge using the tropical storm parameters: central pressure, radius to maximum winds, forward velocity, and storm track. The US Army Corp of Engineers, North Atlantic Comprehensive Coastal Study, has recently performed numerical simulations of 1050 synthetic tropical storms, which statistically represent tropical storms, using a coupled high resolution wave–surge modeling system for the east coast of the US, from Cape Hatteras to the Canadian border. This study has provided an unprecedented dataset which can be used to train artificial intelligence models for surge prediction in those areas. While numerical simulation of a storm surge at this scale and resolution (over 6 million elements scaling from 20 m to more than 100 km) is extremely expensive, the artificial intelligence takes the advantage of the previous simulations, and effectively learns the relationship between storm parameters representing storm forcing and surge. The artificial neural network method which was used for this study, was shown to outperform support vector machine for extreme storms. ANN model, which is based on a neurobiological analogy, can be conveniently developed, retrained by new data, and is nonparametric. The AI model, which was developed for Rhode Island, was validated using a set of randomly selected synthetic storms as well as real tropical storms in this region. The model performance was found satisfactory with root-mean-square error of <35 cm for observed and synthetic storms. It was also shown that it is not possible to develop a reliable artificial intelligence model for this region using a limited number of data (e.g. 200 storms), which is usually available in historical records.
KeywordsStorm surge Artificial neural networks Hurricane Rhode Island NACCS
Thanks to US Army Corps of Engineers (USACE) for sharing the North Atlantic Coastal Comprehensive Studies’ dataset. In particular, we appreciate the time and effort of Norberto C. Nadal-Caraballo and Debra Green, from USACE, for replying to our enquires. Thanks to National Hurricane Center (NHC) and Center for Operational Oceanographic Products and Services (CO-OPS) from NOAA for supplying hurricane and water level data. Also, thanks to Tatsu Isaj for his comments about NACCS’ data. This work was undertaken with funding support from a Rhode Island Community Development Block Grant (4712) from the US Department of Housing and Urban Development and the state of Rhode Island Division of Planning Office of Housing and Community Development. Matt Lewis is funded by the Welsh Government Ser Cymru QUOTIENT project.
- Cialone MA, Massey TC, Anderson ME, Grzegorzewski AS, Jensen RE, Cialone A, Mark DJ, Pevey KC, Gunkel BL, McAlpin TO, Nadal-Caraballo NC, Melby JA, Ratcliff JJ (2015) North Atlantic Coast Comprehensive Study (NACCS) coastal storm model simulations: waves and water levels. Technical report, Coastal and Hydraulics Laboratory, US Army Corps of Engineers, Engineer Research and Development Center, VicksburgGoogle Scholar
- Das HS, Jung H, Ebersole B, Wamsley T, Whalin RW (2011) An efficient storm surge forecasting tool for coastal mississippi. Coast Eng Proc 1(32):21Google Scholar
- Fausett LV, Hall P (1994) Fundamentals of neural networks: architectures, algorithms, and applications. Prentice-Hall, Englewood CliffsGoogle Scholar
- Govindaraju RS, Rao AR (2010) Artificial neural networks in hydrology. Springer, BerlinGoogle Scholar
- Haykin S (2004) Neural networks: a comprehensive foundation. Pearson Prentice HallGoogle Scholar
- Ho FP, Myers VA (1975) Joint probability method of tide frequency analysis applied to Apalachicola Bay and St. George Sound, FloridaGoogle Scholar
- Luettich Jr R, Westerink J, Scheffner NW (1992) ADCIRC: an advanced three-dimensional circulation model for Shelves, Coasts, and Estuaries. Report 1. Theory and methodology of ADCIRC-2DDI and ADCIRC-3DL. Technical report, DTIC documentGoogle Scholar
- Smith JM, Resio DT, Zundel AK (1999) STWAVE: steady-state spectral wave model. Report 1: user’s manual for STWAVE version 2.0. Technical report, DTIC documentGoogle Scholar
- Spaulding MLL (2014) STORMTOOLS: web tools to support coastal resilience analysis and planning for storms and sea level rise. Technical report, Coastal Resources Center. University of Rhode Island. www.beachsamp.org/maps/stormtools
- Spaulding ML, Isaji T, Damon C, Fugate G (2015) Applications of STORMTOOLS simplified flood inundation model, with and without sea level rise, to RI coastal waters. In: ASCE Solutions to Coastal Disasters, Boston, MAGoogle Scholar
- Wamdi-Group (1988) The WAM model-a third generation ocean wave prediction model. J Phys Oceanogr 18(12):1775–1810Google Scholar
- Yegnanarayana B (2009) Artificial Neural Networks. Prentice-Hall of IndiaGoogle Scholar