Storm Surge Forecasting Research Articles

Sensitivity analysis of the physics options in the Weather Research and Forecasting model for typhoon forecasting in Japan and its impacts on storm surge simulations

ABSTRACT Weather Research and Forecasting (WRF) model is useful for forecasting typhoons as an external force of storm surge forecasts. This study examines the variation in typhoon forecasts caused by different choices of arbitrary physics options in WRF and their influence on storm surge forecasts. Eight frequently used combinations of cloud microphysics and planetary boundary layers were extracted via a review of previous studies. Subsequently, sensitivity analyses of these physics options were performed, targeting nine typhoons that landed in Japan during 2015–2019. Additionally, we conducted case studies of storm surge ensemble forecasts in Tokyo Bay and Osaka Bay using WRF-simulated typhoons generated in the sensitivity analysis. As a result, the ensemble mean of the forecasts was comparable to the storm surge reanalysis simulation results obtained using an empirical typhoon model wherein the best track data is integrated to reproduce atmospheric fields. This may be attributed to the fact that the typhoon parameters (intensity, size, approaching angle, and velocity) obtained from the best track at landfall were generally within the range of the parameters that were simulated using WRF.

Ensemble forecast for storm tide and resurgence from Tropical Cyclone Isaias

Ensemble-based probabilistic forecasting of storm surge is increasingly being used to provide metrics for emergency management decisions such as the near-worst case scenario. The Stevens Flood Advisory System is an ensemble prediction system used to forecast total water levels over a broad coastal region and street-scale flood levels for several New York Harbor (NYH) critical infrastructure sites. As a part of our continuous assessment of this system’s performance, we evaluate its prediction of storm tide and resurgence during Tropical Cyclone Isaias (2020), which tracked northward along the Pennsylvania/New Jersey border and caused the largest storm surge in NYH since Hurricane Sandy. Isaias specific track and speed generated an unusual flood event consisting of a storm surge, a blowout, then a significant resurgence that caused minor flooding. The analysis shows that the super-ensemble spread provided an equal or better estimate of uncertainties than sub-ensembles based only on any single meteorological forcing system. Because of ensemble averaging, the central forecast under-predicted peak water levels and the resurgence peak though these were predicted by some of the ensemble members. The impacts of errors in forecast storm arrival time and resolution-related biases in coarse global atmospheric models on the predictions are noted. A limited comparison for this single storm with the National Hurricane Center’s forecast show SFAS providing better accuracy and spread. Advantages and challenges of SFAS and other similar mid-latitude flood forecast systems are identified along with recommendations for analysis and improvement.

Storm surge forecast calculus on the Mexican Pacific coast using SWASH numerical model

Storm surge is a phenomenon that occurs on coasts with a higher incidence in the rainy season. The factors that promote its appearance are the wind and the tide whose force has been sometimes catastrophic. The objective of the study is to provide a storm surge forecast using the SWASH numerical model for some points on the Mexican Pacific coast. The approach of the experiment refers to previous works that use the infrastructure and data available from NOAA and SEMAR; the use of unstructured meshes used in a wave model to conclude in a beach wave model. In the validation of the results, the tide gauges of the study locations will be used; with the feedback obtained, the necessary adjustments are expected until the model has at least 60% correct measurements. Some authors agree that global warming affects the power with which the phenomenon has been manifesting itself, so that the development of technology could help minimize the consequences.

Storm Surge Forecasting along Korea Strait Using Artificial Neural Network

Typhoon attacks on the Korean Peninsula have recently become more frequent, and the strength of these typhoons is also gradually increasing because of climate change. Typhoon attacks cause storm surges in coastal regions; therefore, forecasts that enable advanced preparation for these storm surges are important. Because storm surge forecasts require both accuracy and speed, this study uses an artificial neural network algorithm suitable for nonlinear modeling and rapid computation. A storm surge forecast model was created for five tidal stations on the Korea Strait (southern coast of the Korean Peninsula), and the accuracy of its forecasts was verified. The model consisted of a deep neural network and convolutional neural network that represent the two-dimensional spatial characteristics. Data from the Global Forecast System numerical weather model were used as input to represent the spatial characteristics. The verification of the forecast accuracy revealed an absolute relative error of ≤5% for the five tidal stations. Therefore, it appears that the proposed method can be used for forecasts for other locations in the Korea Strait. Furthermore, because accurate forecasts can be computed quickly, the method is expected to provide rapid information for use in the field to support advance preparation for storm surges.

Uncertainty of storm surge forecast using integrated atmospheric and storm surge model: a case study on Typhoon Haishen 2020

ABSTRACT Hindcast experiments and pseudo-forecast experiments considering Typhoon Haishen (2020) were conducted using an atmospheric (WRF)-storm surge (GeoClaw) coupled model and a storm surge model with a parametric typhoon model. A series of simulations of the coupled model were used to quantify the error sources of the typhoon track and intensity in the forecast errors of storm surges. The results revealed that the typhoon track forecast had a larger error source for the storm surge forecast for the maximum surge height than the typhoon intensity. Furthermore, the parametric Holland typhoon model used in practice has an overestimation trend compared to the coupled model, and the parametric Holland typhoon model using WRF output was able to forecast the storm surge height near the typhoon (western Kyushu area) and its peak occurrence time accurately. However, the forecast accuracy tended to decrease as the distance from the typhoon to the target location increased. The pseudo-ensemble simulation of the storm surge forecast using forecast error information was conducted considering the uncertainty of the typhoon track forecast. The 20 ensemble forecast simulations revealed that the perturbed typhoon track simulation can increase the possibility of capturing the peak time of the storm surge.

The prediction of floods in Venice: methods, models and uncertainty (review article)

Abstract. This paper reviews the state of the art in storm surge forecasting and its particular application in the northern Adriatic Sea. The city of Venice already depends on operational storm surge forecasting systems to warn the population and economy of imminent flood threats, as well as help to protect the extensive cultural heritage. This will be more important in the future, with the new mobile barriers called MOSE (MOdulo Sperimentale Elettromeccanico, Experimental Electromechanical Module) that will be completed by 2021. The barriers will depend on accurate storm surge forecasting to control their operation. In this paper, the physics behind the flooding of Venice is discussed, and the state of the art of storm surge forecasting in Europe is reviewed. The challenges for the surge forecasting systems are analyzed, especially in view of uncertainty. This includes consideration of selected historic extreme events that were particularly difficult to forecast. Four potential improvements are identified: (1) improve meteorological forecasts, (2) develop ensemble forecasting, (3) assimilation of water level measurements and (4) develop a multimodel approach.

Skew Surge and Storm Tides of Tropical Cyclones in the Delaware and Chesapeake Bays for 1980–2019

Coastal flooding poses the greatest threat to human life and is often the most common source of damage from coastal storms. From 1980 to 2020, the top 6, and 17 of the top 25, costliest natural disasters in the U.S. were caused by coastal storms, most of these tropical systems. The Delaware and Chesapeake Bays, two of the largest and most densely populated estuaries in the U.S. located in the Mid-Atlantic coastal region, have been significantly impacted by strong tropical cyclones in recent decades, notably Hurricanes Isabel (2003), Irene (2011), and Sandy (2012). Current scenarios of future climate project an increase in major hurricanes and the continued rise of sea levels, amplifying coastal flooding threat. We look at all North Atlantic tropical cyclones (TC) in the International Best Track Archive for Climate Stewardship (IBTrACS) database that came within 750 km of the Delmarva Peninsula from 1980 to 2019. For each TC, skew surge and storm tide are computed at 12 NOAA tide gauges throughout the two bays. Spatial variability of the detrended and normalized skew surge is investigated through cross-correlations, regional storm rankings, and comparison to storm tracks. We find Hurricanes Sandy (2012) and Isabel (2003) had the largest surge impact on the Delaware and Chesapeake Bay, respectively. Surge response to TCs in upper and lower bay regions are more similar across bays than to the opposing region in their own bay. TCs that impacted lower bay more than upper bay regions tended to stay offshore east of Delmarva, whereas TCs that impacted upper bay regions tended to stay to the west of Delmarva. Although tropical cyclones are multi-hazard weather events, there continues to be a need to improve storm surge forecasting and implement strategies to minimize the damage of coastal flooding. Results from this analysis can provide insight on the potential regional impacts of coastal flooding from tropical cyclones in the Mid-Atlantic.

Evaluating storm surge predictability on subseasonal timescales for flood forecasting applications: A case study for Hurricane Isabel and Katrina

Coastal flooding operational forecasting in the US is limited to short-range temporal scales (3–7 days), which limits the response time for emergency preparation and planning. The sub-seasonal prediction project (SubX), which produces weather forecasts with a lead time of up to four weeks, provides an opportunity to assess the potential for creating probabilistic flood forecasts with longer lead times. Using the ADCIRC hydrodynamic model for coastal storm surge, two major hurricanes, Isabel (2003) and Katrina (2005), were used as case studies to test coastal flood predictions induced by wind and pressure fields generated from five global weather models within SubX. The storm surges simulations are forced by Sea Level Pressure (SLP) and 10 m winds fields from SubX models for a lead-time of up to 30 days before storm landfall. The subseasonal surge forecasts are evaluated temporally and spatially at 1–4 weeks lead-time against the NOAA tide gages observations and a verification dataset derived by forcing the storm surge model with wind and pressure fields from the NCEP-Reanalysis. The results are evaluated in terms of lead-time and forecast skill metrics. The storm surge forecast skill is measured using the mean square error skill score (MSESS) relative to the verification dataset and an approximate of the climatology. A skill score greater than 0.55 is considered here useful for flood forecasting. The multi-model ensemble (MME) mean surge forecasts driven by several members of SubX models demonstrate skill greater than 0.55 up to a 4-day and 10-day lead for Katrina and Isabel, respectively. A sharper decrease in MSESS was noted from week 1 to week 3 lead-times for Katrina, in comparison to Isabel. Some ensemble members forecasted hurricanes and storm surges as early as 3–4 weeks lead-time. However, due to the offsets developed in the timing and magnitude of the peak at these lead-times, and based on a sample size of only two events, it is hard to establish the significance of these longer lead-time results. While a follow up study involving flood reforecasts over the entire SubX reforecast period (1999–2015) is needed to support more robust statistics of the forecast skill, our case studies demonstrate the feasibility of probabilistic flood forecasting at subseasonal timescales using the SubX models.

Effects of Wave–Current Interaction on Storm Surge in the Pearl River Estuary: A Case Study of Super Typhoon Mangkhut

This study explored the effects of interactions between waves and current on storm surge in the Pearl River Estuary (PRE) using a fully coupled wave–current model. The model was validated based on in situ observations during the traverse of super typhoon Mangkhut. The results indicated that the model could reproduce the storm surge and wave setup processes. Numerical experiments showed that simulations of storm surge are minimally affected by wave setup. The wave setup during super typhoon Mangkhut reached up to 0.23 m and contributed to the total near shore storm surge by up to 8%. The simulations of the coupled model showed a better correlation with observations compared to those of an uncoupled model. The storm surge increased with transport upstream in a tidal-dominated outlet, whereas it decreased in a river-dominated outlet. The storm surge and wave setup increased and decreased, respectively, during spring tide as compared to that during a neap tide. The storm surge increased with increasing runoff in the upper river reaches, whereas there was little change in the tidal-dominated lower river reaches. This research emphasizes the importance of integrating the effects of multiple dynamic factors in the forecasting of storm surge and provides a reference for similar studies in other estuaries with multiple outlets and a complex river network.

Fast Storm Surge Ensemble Prediction using Searching Optimization of a Numerical Scenario Database

AbstractAccurate storm surge forecasts provided rapidly could support timely decision-making with consideration of tropical cyclone (TC) forecasting error. This study developed a fast storm surge ensemble prediction method based on TC track probability forecasting and searching optimization of a numerical scenario database (SONSD). In a case study of the Fujian Province coast (China), a storm surge scenario database was established using numerical simulations generated by 93,150 hypothetical TCs. In a GIS-based visualization system, a single surge forecast representing 2562 distinct typhoon tracks and the occurrence probability of overflow of seawalls along the coast could be achieved in 1–2 min. Application to the cases of Typhoon Soudelor (2015) and Typhoon Maria (2018) demonstrated that the proposed method is feasible and effective. Storm surge calculated by SONSD had excellent agreement with numerical model results (i.e., mean MAE/RMSE: 7.1/10.7 cm, correlation coefficient: >0.9). Tide prediction also performed well with MAE/RMSE of 9.7/11.6 cm versus the harmonic tide, and MAE/RMSE of phase prediction for all high waters of 0.25/0.31 h versus observations. The predicted high-water level was satisfactory (MAE of 10.8 cm versus observations) when the forecasted and actual positions of the typhoon were close. When the forecasted typhoon position error was large, the ensemble surge prediction effectively reduced prediction error (i.e., the negative bias of −58.5 cm reduced to −5.2 cm versus observations), which helped avoid missed alert warnings. The proposed method could be applied in other regions to provide rapid and accurate decision-making support for government departments.

Implementation of an Artificial Neural Network for Storm Surge Forecasting

AbstractAccurate and timely storm surge forecasts are essential during tropical cyclone events in order to assess the magnitude and location of the impacts. Coupled ocean‐atmosphere dynamical models provide accurate measures of storm surge but remain too computationally expensive to run for real‐time forecasting purposes. Therefore, it is common to utilize a parametric vortex model, implemented within a hydrodynamic model, which decreases computational time at the expense of forecast accuracy. Recently, data‐driven neural networks are being implemented as an alternative due to their combined efficiency and high accuracy. This work seeks to examine how an artificial neural network (ANN) can be used to make accurate storm surge predictions, and explores the added value of using a recurrent neural network (RNN). In particular, it is concerned with determining the parameters needed to successfully implement a neural network model for the Mid‐Atlantic Bight region. The neural network models were trained with modeled data resulting from coupling of the Hybrid Weather Research and Forecasting cyclone model (HWCM) and the Advanced Circulation Model. An ensemble of synthetic, but physically plausible, cyclones were simulated using the HWCM and used as input for the hydrodynamic model. Tests of the ANN were conducted to investigate the optimal lead‐time configuration of the input data and the neural network architecture needed to minimize storm surge forecast errors. Results highlight the accuracy of the ANN in forecasting moderate storm surge levels, while indicating a deficiency in capturing the magnitude of the peak values, which is improved in the implementation of the RNN.

Verification of forecasts for extreme rainfall, tropical cyclones, flood and storm surge over Myanmar and the Philippines

Within the humanitarian sector, there is a pressing need to scale up anticipatory action as climate change-related disasters increase. This article evaluates forecasts relating to extreme weather events – extreme rainfall, tropical cyclones, river flooding and storm surge – in Myanmar and the Philippines to assess the feasibility of using such forecasts to develop early warning systems and responses. To make best use of limited extant data, a variety of methods (reliability diagrams, hit rates, false alarm ratios, correlations) are employed. We review the skill of the European Centre for Medium-Range Weather Forecasts (ECMWF) tropical cyclone forecasts and find that whilst errors in cyclone position are relatively small, forecasting intensity is more difficult. When a tropical cyclone has formed, the probabilities provided in the ECMWF track forecast are highly reliable and only slightly overconfident. A tropical cyclone activity product is relatively reliable for forecasts less than a week ahead for North Indian Ocean cyclones affecting Myanmar, but becomes very overconfident beyond this. Looking at flood forecasting models, a comparison of the Global Flood Awareness System (GloFAS, produced by the ECMWF and the European Commission as part of the Copernicus Emergency Management Services) with the Global Flood Forecasting Information System (GLOFFIS, produced by Deltares) demonstrates that both GloFAS and GLOFFIS have difficulty simulating 1 in 2 year return period flows or higher, although GloFAS performance is better than GLOFFIS. GloFAS reforecasts show significantly overconfident probabilities over Myanmar where discharge observations are available, whilst the lack of a GLOFFIS reforecast prevents evaluation of this forecast system directly. Evaluation of the ten-day operational storm surge forecast (the Global Storm Surge Information System, GLOSSIS) produced by Deltares was attempted but lack of any data prevented assessment. These findings present valuable insights into how well forecasts perform, which is crucial information for establishing effective humanitarian action mechanisms.

Adjoint-based sensitivity analysis for a numerical storm surge model

Numerical storm surge models are essential to forecasting coastal flood hazard and informing the design of coastal defences. However, such models rely on a variety of inputs, many of which carry uncertainty. An awareness and understanding of the sensitivity of model outputs with respect to those uncertain inputs is therefore essential when interpreting model results. Here, we use an unstructured-mesh numerical coastal ocean model, Thetis, and its adjoint, to perform a sensitivity analysis for a hindcast of the 5th/6th December 2013 North Sea surge event, with respect to the bottom friction coefficient, bathymetry and wind stress forcing. The results reveal spatial and temporal patterns of sensitivity, providing physical insight into the mechanisms of surge generation and propagation. For example, the sensitivity of the skew surge to the bathymetry reveals the protective effect of a sand bank off the UK east coast. The results can also be used to propagate uncertainties through the numerical model; based on estimates of model input uncertainties, we estimate that modelled skew surges carry uncertainties of around 5 cm and 15 cm due to bathymetry and bottom friction, respectively. While these uncertainties are small compared with the typical spread in an ensemble storm surge forecast due to uncertain meteorological inputs, the adjoint-derived model sensitivities can nevertheless be used to inform future model calibration and data acquisition efforts in order to reduce uncertainty. Our results demonstrate the power of adjoint methods to gain insight into a storm surge model, providing information complementary to traditional ensemble uncertainty quantification methods.

Development and Verification of NEMO based Regional Storm Surge Forecasting System

본 연구에서는 한반도 중심 해역을 포함하는 북서태평양 영역에서의 폭풍해일 예측을 위해 NEMO(Nucleus for European Modelling of the Ocean) 모형을 이용하여 지역규모의 폭풍해일 ì˜ˆì¸¡ì‹œìŠ¤í œì„ 구축하였다. 이 ì‹œìŠ¤í œì€ 조석과 해일 예측으로 구성되어 있으며보다 ì •í™•í•œ 폭풍해일을 예측하기 위해 수심자료와 대기-해양 경계에서의 모수화(parameterization) ìµœì í™” ê³¼ì •ì„ 수행하였다. 이를 통해 2018ë „ 8~10월과 태풍 솔릭 사례에 대하여 국립해양 조사원 조위 관측자료를 이용한 통계 방법을 ì ìš©í•˜ì—¬ 검증을 ìˆ˜í–‰í•˜ê³ , 이를 POM(Princeton Ocean Model) 기반의 예측모델 결과와 비교하였다. 수행결과 NEMO 기반의 폭풍해일 ì˜ˆì¸¡ì‹œìŠ¤í œì´ POM 기반의 예측결과에 비해 í‰ê· ì˜¤ì°¨ì™€ RMSE가 각각 약 29%와 약 20 % 감소한 것으로 나타났으며, 태풍 시기에도 NEMO 기반 예측결과에서 ì „ë°˜ì ìœ¼ë¡œ 오차가 낮게 나타났다. 핵심용어: 조석, 폭풍해일, NEMO, 지역모델

WAVE EFFECTS ON STORM SURGE IN A SMALL TWO-INLET BAY

Storm surge resulting from oceanic extreme events, commonly tropical cyclones, is a major contributor to coastal flooding and property damage. Thus, there is significant investment in accurate predictions. However, forecasts of storm surge often are focused on regional scales, and are unable to resolve complex nearshore bathymetry and small tidal inlets (Yin et al. 2016) that can be critical to local surge magnitudes and timing. Here, model simulations with a regional wave-flow coupled model (NACCS), a high bathymetric resolution uncoupled flow model (ADCIRC), and a high resolution coupled model (CSTORM) are compared with observations of storm surge during Hurricane Irene (Atlantic Storm 09, 2011) within Katama Bay, Martha's Vineyard, Massachusetts.Recorded Presentation from the vICCE (YouTube Link): https://youtu.be/hKdA2zYWI2Y

STORM SURGE FORECAST MODEL USING GENETIC PROGRAMMING

Stormsurge is a typical genuine fiasco coming from the ocean. Therefore, an accurate forecast of surges is a vital assignment to dodge property misfortunes and decrease the chance of tropical storm surges. Genetic Programming (GP) is an evolution-based model learning technique that can simultaneously find the functional form and the numeric coefficients for the model. Moreover, GP has been widely applied to build models for predictive problems. However, GP has seldom been applied to the problem of storm surge forecasting. In this paper, a new method to use GP for evolving models for storm surge forecasting is proposed. Experimental results on data-sets collected from the Tottori coast of Japan show that GP can become more accurate storm surge forecasting models than other standard machine learning methods. Moreover, GP can automatically select relevant features when evolving storm surge forecasting models, and the models developed by GP are interpretable.

The contribution of short-wave breaking to storm surges: The case Klaus in the Southern Bay of Biscay

This study investigates the contribution of short-wave breaking to storm surges through a high-resolution hindcast of the sea state and storm surge associated with the extra-tropical storm Klaus. This storm made landfall in January 2009 in the Southern Bay of Biscay and produced the largest storm surges observed in this region over the last 20 years, with 1.70m in the Arcachon Lagoon and 1.10m in the Adour Estuary. A fully-coupled 3D modelling system, which uses a vortex force formalism to represent wave-current interactions, is applied with a spatial resolution down to 35m in the surf zones in order to properly compute the wave-induced setup. Modelling results reveal that the wave setup contributes by up to 40% and 23% to the storm surge peak in the Adour Estuary and the Arcachon Lagoon respectively. Accounting for wave forces in the circulation model improves storm surge predictions by 50 to 60%. This is explained by the dominant role played by wave forces in the momentum balance at the inlets under storm waves. Numerical experiments further reveal that the wave-induced setup can be tidally-modulated, although this phenomenon seems to be site-specific. Finally, a sensitivity analysis highlights the importance of the model grid resolution in the surf zones to correctly resolve the wave setup along open-ocean coasts. Inside the lagoon, the storm surge and wave setup are less sensitive to the grid resolution while tidal propagation cannot be accurately represented with a resolution of 1000m, which is typically used in operational storm surge forecast.

Study on the applicability of a stochastic typhoon model for probabilistic forecasting of storm surge induced by a typhoon

ABSTRACT This study investigates the applicability of a stochastic typhoon model, STM, for probabilistic forecasting of storm surge induced by a typhoon. When a typhoon forms and approaches the coast, the present STM generates a number of virtual typhoons with initial conditions specified by the recent history of the actual typhoon. Storm surge is then computed for each of the generated typhoons, and the probabilistic characteristics of the estimated storm surges are investigated. The present STM is based on the higher-order autoregressive model, which can account for the history of the characteristics of individual typhoons. The concept of minimal pressure was introduced to improve the predictive skills of the statistical characteristics of extreme typhoons. The model was applied to the cases of Typhoons Jebi and Trami, which hit Japan in 2018. The present system reasonably explained the contrasting characteristics of Jebi and Trami in terms of their peak storm surge levels in the inner part of Osaka Bay. Furthermore, the sensitivity analysis of the present probabilistic forecasting system suggested that a relatively large number of virtual typhoons are needed for forecasting of the worst storm surge conditions induced by the arbitrary actual typhoons approaching the target coast.

Numerical Simulation of Typhoon Storm Surge in Wenzhou Coastal Areas

Focusing on the numerical simulation of storm surge in the coastal waters of Wenzhou, this paper uses the atmospheric Mesoscale Model5 (MM5) model to provide the wind field of typhoons, and the advanced circulation (ADCIRC) model for oceanic, coastal, and estuarine waters to simulate the process of storm surges, which adopts the finite element method with a dry-wet grid. The simulated results agree well with the observed data from marine observation stations (Longwan, Aojiang, Ruian), indicating that the simulation of storm surge is accurate. The good agreement between the simulated and observed results of four typhoons (Nos. 200216, 200414, 200509, 200608) shows that the combination of the two numerical models has the capacity to calculate storm surge elevations along the coast of Wenzhou. Using this method to simulate the 23 typhoon storm surge processes of the Wenzhou sea area of the last 23 years, we find that the majority of storm surge is between 0.5 to 2.5 m, and about one-fifth have an extreme value of more than 2 m. The maximum storm surge value recorded in Wenzhou coastal waters is 3.02 m, which occurred during Typhoon Wipha (No. 200713) at 03:00 on September 19, 2007 when landed. Water decreasing processes occurred 10 h after the main water increasing process with the maximum negative storm surge of -0.5 m. This paper can be used as a reference for storm surge forecasting, coastal engineering design and damage minimization in the future.

Genetic Programming for storm surge forecasting

Storm surge is a genuine common fiasco coming from the ocean. Therefore, an exact forecast of surges is a vital assignment to dodge property misfortunes and to decrease a chance caused by tropical storm surge. Genetic Programming (GP) is an evolution-based model learning technique that can simultaneously find the functional form and the numeric coefficients for the model. Therefore, GP has been widely applied to build models for predictive problems. However, GP has seldom been applied to the problem of storm surge forecasting. In this paper, we propose a new method to use GP for evolving models for storm surge forecasting. Experimental results on datasets collected from the Tottori coast of Japan show that GP can evolve accurate storm surge forecasting models. Moreover, GP can automatically select relevant features when evolving storm surge forecasting models, and the models evolved by GP are interpretable.