1. Introduction
Recently, frequent flash floods due to climate change and abnormal climate patterns have severely damaged the urban area. Specifically, unlike inland urban areas, coastal urban areas, which are affected by a combination of precipitation and sea-level patterns, have more severe damage even with the same amount of precipitation [
1,
2,
3,
4]. In coastal areas with high population density and high asset values, the damages caused by flooding are fatal and critical, so practical strategies are required to prevent flooding disasters in coastal areas. South Korea is vulnerable to flood disasters in coastal areas because many country regions are composed of the ocean [
5,
6]. Moreover, considering the sea-level rise and the extreme precipitation patterns due to climate change, it is necessary to analyze the vulnerability of flooding through a complex inundation simulation for coastal areas in South Korea.
In South Korea, the damage caused by typhoons in 2003 was 531 people (133 dead, 15 missing, 383 injured), and property damage was approximately 4.4 billion dollars. For example, Gyeongsangnam-do, which has many coastal urban areas in South Korea, suffered from fatal flooding damages in 2003. The flooding damage cost was approximately 2 trillion dollars, accounting for approximately 50% of the damage that occurred nationwide, and flood damage was in an area of 27,000 ha. Therefore, it is essential to study the mitigation of flood damage by developing an early warning system in coastal areas vulnerable to rising sea levels and extreme precipitation patterns. For this, research on the flooding simulation model that can simulate and predict the inundation in urban areas considering the characteristics of precipitation and sea-level rise according to climate change should be accompanied first.
Research on urban inundation analysis caused by extreme precipitation and sea-level rise has been widely conducted worldwide, including in the US [
7,
8,
9], Asia [
10,
11,
12,
13,
14], Europe [
15,
16,
17], and South Korea [
5,
18,
19,
20]. Specifically, inundation simulation and analysis have been performed using 1D and 2D numerical models and hydraulic models such as the Storm Water Management Model (SWMM) and MIKE models. For example, Hsu et al. (2000) [
21] evaluated the rainwater treatment capacity of the sewer system and the pump station installed in Taipei by performing a flood simulation using the SWMM model. Hsu et al. (2002) [
22] analyzed the correlation between the urban sewer system and the surface flow using a 2D non-inertia overland flow model and SWMM’s EXTRAN module. In addition, Leandro et al. (2009) [
23] analyzed the flood inundation in urban areas using a combination of the 1D sewer model and 2D surface network system. These studies have shown that physical-based models have sufficient power to simulate and analyze the inundation in urban areas.
In South Korea, where major cities are located along the coast, many studies on inundation analysis using urban flooding simulation models for coastal cities have been conducted [
24,
25,
26]. For example, Jeong et al. (2011) [
27] evaluated the impact of extreme precipitation and sea-level patterns enhanced by typhoons on inundation in urban areas using a 2D numerical model. They suggested using a hydraulic structure such as installing a drainage system in urban areas and emphasized the importance of a flood analysis computational model to prevent flood damage in advance. Kim et al. (2009) [
28] conducted a flood simulation by considering the dual-drainage method that considers the outflow and inflow from the manhole simultaneously and the effect of the tidal patterns on the coast. Moon et al. (2006) [
29] used the MIKE21 model to simulate the flooding of nine typhoons that caused large tidal waves over the past ten years. They showed how coastal urban areas are vulnerable to sea levels rising when a storm is coming. In addition, Kang et al. (2013) [
30] performed an urban floods simulation by applying radar precipitation to SWMM to predict flash floods accurately and rapidly. However, since most studies have mainly investigated flood damage based on past conditions, studies on potential flood damage in the future are lacking.
Climate change scenarios are data that evaluate and predict the scientific, technological, and socio-economic impacts of climate change due to complex factors such as global future social, economic, and human activities and the resulting greenhouse gas emissions. The Intergovernmental Panel on Climate Change (IPCC) has published a climate change assessment report every 5 to 6 years and has consistently reported scientific evidence for climate change and its effects. One of the main tasks of the IPCC is to develop potential future climate change scenarios according to greenhouse gas emissions and evaluate the climate change response strategies.
Many researchers have evaluated the impact of future climate patterns on inundation in urban areas by considering climate change scenarios reported by the IPCC [
31,
32,
33,
34,
35,
36]. Specifically, in the case of studies targeting coastal urban areas, inundation analysis has been performed according to complex factors caused by both the extreme precipitation patterns and the sea-level rise at the same time. For example, Wang et al. (2012) [
1] studied the effects of climate change on urban inundation and complex factors combining sea-level rise, storm surge, and subsidence in Shanghai, China using MIKE21 software. They found that approximately 46% of areas in Shanghai would be flooded by 2100. They suggested the need for a study on the effects of complex factors such as precipitation and sea level on coastal urban areas. In addition, there are notable studies that conducted hydrological impact assessments of climate change using the latest scenario data (i.e., AR6 of IPCC) reflecting the latest information released [
37,
38].
Changwon city, one of the typical coastal cities in South Korea, has suffered inundation damage when a typhoon has hit, but disaster-prevention strategies and policies are insufficient. In particular, the typhoon in 2012, named Sanba, caused severe damage to property and humans. At that time, the maximum hourly precipitation was 25 mm, which was lower than the designed precipitation rate of the sewer pipeline system. Still, the sea-level rise in the ocean exceeded the warning level, and flood damage occurred. This is one of the examples of why we need to consider both the precipitation patterns and sea-level rise for inundation analysis in coastal urban areas, unlike inland regions. Therefore, this study aims to simulate inundation in Changwon City, South Korea, considering extreme precipitation and sea-level rise and predicting the potential vulnerability to flooding damage enhanced by climate change scenarios.
4. Discussions
This study analyzed the inundation damage of coastal urban areas under climate change scenarios. The proposed SWMM model used potential probable precipitation and sea-level rise in the future. Even though the simulated results have shown that both the flooding area and the inundation depth increase as the future target period increases, some limitations should be discussed. The first limitation is the uncertainty regarding the inputs of the model (i.e., pipeline features), which is the main hydraulic input data of the SWMM model. For example, only the main pipelines of the sewage system were considered, and the effect on branch lines was ignored in this study. Although the effect of branch lines on water flow in the sewage system was negligible, it may lead to lower model performance when simulating inundation damage.
The second limitation of this study is uncertainty in climate change scenarios. This is a common limitation of most studies considering climate change scenarios. Since the future precipitation and sea-level data come from GCMs, the uncertainty inherent in the results should be investigated. Therefore, since the prediction results based on climate change scenarios may also have some uncertainties, it is suggested to consider the results to understand the overall potential trends of the increase in the inundation depth or area in the future.
The final weakness of this study is the verification processes of the SWMM model. In this study, as mentioned in
Section 2.3, due to the lack of observational data in the area for model verification, the model was verified by comparing it with the actual flooded area caused by typhoon “Sanba” on 16–17 September 2012. Although the verification results confirmed that the modeled flood area was approximately 90% consistent with the actual flooding area, it is suggested that an additional process for model verification should be performed in future studies.
Considering these limitations comprehensively, it is suggested that future studies should consider the process for improving the accuracy of modeling results by taking into account the uncertainty in climate change scenarios, strengthening the model validation process, and improving the reliability of input data.
In addition, in this study, precipitation and sea-level patterns according to the AR4 scenario were used, but it is necessary to consider the latest scenario (i.e., AR6) in the future study [
37,
51,
52,
53]. It is expected that more accurate flood vulnerability simulation and prediction will be possible because the latest global conditions and information such as land management, greenhouse gas fluxes, and terrestrial ecosystems are comprehensively considered in the latest scenarios.
5. Conclusions
In this study, the two-dimensional (2D) urban runoff model (i.e., SWMM) was used to simulate the inundation of the coastal metropolitan area of Changwon City according to both the precipitation and sea-level rise in consideration of climate change. The probable precipitation was applied with different precipitation frequencies and durations to evaluate the vulnerability drainage system in the area to flood damage. The frequencies of precipitation divided into the short- (10 years), mid- (30 years), and long- (50 years) term and durations of 1 h–12 h were applied as input data of the model to conduct an inundation analysis. In addition, we used the potential increase rate of precipitation and sea level in Changwon City provided by climate change scenarios to investigate the potential vulnerability to flooding in the future.
As a result of analyzing the future vulnerability to flooding damage in Changwon city by predicting and investigating via the application of climate change scenarios, we determined the local drainage system in the area is dependent on the complex factors of precipitation and sea-level pattern together. It was found that the damage caused by heavy rains and typhoons, etc., which is very vulnerable to flood damage caused by climate change, will be further aggravated.
For the 10-year, 30-year, and 50-year precipitation-frequency scenarios, as the target period increases, both the flooding area and the maximum inundation depth increase compared to the results of the inundation simulation according to the current precipitation and sea-level conditions. The inundation area increased by 2.6–16.2% compared to the current state, and the number of cells with flooded depths of 1 m or more increased. In the case of urban areas with high population density, the flooding damage, with a larger flooded area and a high inundation depth, will be enormous. The vulnerability of coastal cities to flooding damage is expected to increase rapidly due to climate change.
Based on the results of this study, it is essential to consider the comprehensive impact of precipitation and sea-level features on flooding damage in the drainage system of coastal urban areas. Suppose a modeling simulation is performed that considers the actual land use of the watershed and the location or characteristics of the buildings in the metropolitan area. In that case, it is expected that the results of this study will be utilized as valuable primary data for the development of a flood warning system in the future.
Considering the topographical characteristics of South Korea, where three sides of the country are in contact with the ocean and Changwon City, the study area can be expanded to the entire coastal urban area. Future flooding scenarios in the country can be analyzed using future prediction data such as climate change scenarios. It is considered that continuous research is necessary to predict the vulnerability to flooding damage and to come up with a comprehensive strategy.