A Study on the Pre-Survey and Plan for the Establishment of the Korean Typhoon Impact-Based Forecast

Abstract

The intensity of typhoons affecting the Korean Peninsula has been rapidly increasing, resulting in significant damage. Notably, this intensification correlates with the rise in Sea Surface Temperature (SST) in the western Pacific Ocean and surrounding sea areas, where typhoons that impact the Korean Peninsula originate and develop. The SST in these regions is increasing at a faster rate than the global average. Typhoon-related meteorological disasters are not isolated events, such as strong winds, heavy rains, or storm surges, but rather multi-hazard occurrences that can affect different areas simultaneously. As a result, preparation and evaluation must address multi-hazard disasters, rather than focusing on individual weather phenomena. This study develops the Typhoon Ready System (TRS) to improve impact-based forecasting in Korea, in response to the growing threat of multi-hazard weather disasters. By providing region-specific pre-disaster information, the TRS enables local governments and individuals to better prepare for and mitigate the impacts of typhoons. The system will be continuously refined in collaboration with the U.S. Weather-Ready Nation (WRN), which possesses advanced impact forecasting capabilities. The findings of this study offer a crucial framework for enhancing Korea’s ability to forecast and respond to the escalating threats posed by typhoons. By utilizing the TRS, it will be possible to assess the risks of various multi-hazard weather disasters specific to each region during the typhoon forecast period, and the relevant data can be efficiently applied at both the individual and local government levels for typhoon prevention efforts. The system will be continuously improved through cooperation with the U.S. WRN, leveraging their advanced impact forecasting systems. It is expected that the TRS will enhance the accuracy of typhoon impact forecasts, which have been responsible for significant damage in Korea.

1. Introduction

1.1. Background of Study

Climate change refers to alterations in the climate system that occur in addition to the natural climate variations observed over extended periods, primarily due to the increased concentration of greenhouse gases from human activities. Climate change poses irreparable risks to human civilization, including water shortages, food shortages, rising sea levels, and extreme meteorological events. The global climate has reached a crisis point, surpassing the concept of mere “change”. There is growing advocacy for terms such as “climate crisis”, “climate emergency”, and “climate failure” to more effectively convey the severity of these risks, as highlighted in recent reports by the World Meteorological Organization (WMO) [1]. These terms underscore the urgent need for adaptation and mitigation strategies in response to extreme weather events, including the intensifying threat of typhoons.

There has been a call to shift from meteorological phenomenon-based forecasting, which provides information based solely on meteorological phenomena, to impact-based forecasting, which considers the socio-economic impact of vulnerabilities to meteorological hazards at the individual level, despite steadily improving meteorological forecast accuracy [2]. As a result, countries around the world are actively promoting the introduction of impact-based meteorological forecasts and utilizing them in the field. In the case of the Korean Peninsula, several studies have been conducted on the necessity and planning of meteorological impact-based forecasting [3,4], and some studies have addressed impact-based forecasting for heat waves and heavy rains [5,6,7]. However, research on typhoon impact-based forecasting remains insufficient. Regarding typhoons affecting the Korean Peninsula, there have been basic scientific studies related to climatology [8,9] and efforts to improve prediction accuracy [10,11], but there is no research on typhoon-related meteorological disasters based on impact-based forecasting. Typhoons affecting Korea are becoming stronger and more frequent, and their characteristics are rapidly evolving [12,13]. Therefore, the introduction of typhoon impact-based forecasts for the Korean Peninsula is essential.

According to WMO, four major climate change indicators—greenhouse gases, sea levels, SST, and ocean acidity—reached record highs in 2021, warning that human activities are having a detrimental impact on sustainable development and ecosystems. This is just the beginning. Furthermore, rapid changes in meteorological disasters, such as heat waves and floods, which cause enormous human and property damage, are occurring globally. The WMO stated that among multi-hazard and large-scale natural disasters, tropical cyclones, especially typhoons, result in the greatest economic losses due to their associated multi-hazard meteorological impacts, such as floods and strong winds. According to the Intergovernmental Panel on Climate Change (IPCC), extreme events and disaster risk management for climate change adaptation suggest that global damage will increasingly result from multi-meteorological events, rather than isolated ones. Based on this, many countries treat tropical cyclones, including typhoons, as “multi-hazard” events rather than single meteorological disasters. Additionally, an analysis of global natural disaster damage data from the past 30 years (1970–2019) showed that typhoons ranked second among all natural disasters, with 577,232 human casualties, following droughts (650,000). Typhoons also caused the most property damage, with an estimated USD 521 billion. Among various weather-related disasters, tropical cyclones, including typhoons, caused the greatest economic losses due to their complexity, involving floods, strong winds, and other simultaneous meteorological disasters. Therefore, it is crucial to conduct research on advanced forecasting systems to reduce typhoon-related damage.

1.2. Typhoon Forecast

1.2.1. Existing Global Study on Typhoon Impact-Based Forecasting

Although the concept of an impact-based forecast is not clearly defined, it is generally referred to in Korea as a forecast [14] that considers the possibility of dangerous meteorological events occurring, and as a forecast [15] that conveys the socio-economic impacts expected from meteorological phenomena by providing detailed meteorological information while accounting for risks and vulnerabilities. The Met Office defines an impact-based forecast as one that estimates the social and economic impacts that may occur during a meteorological disaster, considering the intensity of the disaster, the degree of exposure, and regional vulnerability.

In 2003, the WMO established the Disaster Risk Reduction (DRR) program, with the vision of increasing the contribution of meteorological and hydrological services to cost-efficiency, automation, sustainability, and the protection of life and property. The goal is to provide meteorological and hydrological services in an efficient and systematic manner to ensure stable protection of life and property, which includes impact-based forecasts under the DRR program. The WMO has long promoted programs aimed at reducing the damage caused by meteorological disasters, with a strong emphasis on impact-based forecasting. In 2011, the National Oceanic and Atmospheric Administration (NOAA) and the National Weather Service (NWS) introduced a paradigm shift with the declaration of the “Weather-Ready Nation” (WRN), a national strategy designed to minimize damage by providing advance information about the risks associated with extreme meteorological disasters. This approach focuses on proactive response and preemptive action to mitigate the effects of extreme weather events, such as record snowfall, intense tornadoes, destructive hurricanes, widespread flooding, and droughts.

With the paradigm shift between the WMO and the U.S., several countries around the world (including the United Kingdom, France, Japan, and China) adopted the concepts of “readiness” and “preparedness” in relation to tropical cyclones, such as typhoons and hurricanes, to establish proactive prevention measures and adjust national policy frameworks accordingly. Additionally, with support from the WMO, several developing countries in East Asia (such as the Philippines, Vietnam, and Thailand) that are affected by typhoons have also actively adopted impact-based forecasting systems [16].

1.2.2. Korea Typhoon Forecast Status

Following the declaration of the WRN paradigm shift in the United States, many countries around the world have been establishing pre-disaster strategies for meteorological disasters, particularly for tropical cyclones like typhoons, hurricanes, and storms. This has led to a change in national policies, with an active focus on introducing impact-based forecasts.

In the case of Korea, analyzed the economic value of impact-based forecasts for various meteorological disasters and concluded that “Typhoon Impact-Based Forecast” has the potential to generate the highest economic value among all types of impact-based forecasts. The Korea Meteorological Administration began developing technology for impact-based meteorological forecasting in 2015, establishing both the necessary evidence and an implementation plan. As a result, the Korea Meteorological Administration has been providing impact-based forecasting services for heat waves and cold waves since 2019 and announced plans to begin typhoon impact-based forecasting in 2021. However, as of 2022, the current system—referred to as “Detailed Typhoon Risk Point Information”, which is provided after a special typhoon announcement—is still simply an extension of the existing typhoon forecasting system. In other words, there has been no significant practical progress in implementing a true impact-based forecast for typhoons, and the current system is inadequate for typhoon prevention activities in the field.

While the global community is making strides in building a foundation for impact-based forecasting in response to the climate crisis, Korea’s efforts have not yet reached a functional or operational level. To address this, researchers signed the WRN Ambassador Convention with NOAA’s NWS in the United States, which declared a paradigm shift aimed at reducing the damage caused by meteorological disasters using impact-based forecasts. This collaboration marked the first attempt to establish a foundation for a typhoon impact-based forecasting system in Korea, aiming to develop a system that can be used in practical applications. The goal is to calculate the meteorological scale of multi-hazard disasters related to typhoons and produce risk indices for these disasters by using region-specific data for each administrative district. This would allow local governments and individuals to take regional vulnerabilities into account when preparing for typhoons.

In this study, we aim to analyze the changes in typhoons affecting the Korean Peninsula and conduct a preliminary study to introduce a typhoon impact-based forecasting system. The objective is to establish a solid foundation for such a system, which is essential for ensuring public safety and mitigating property damage in the context of the climate crisis. This system is designed to provide information that can be utilized by both individuals and local governments, thereby improving preparedness and response efforts across the Korean Peninsula.

2. Data and Methods

2.1. Data

To examine the changes in typhoons affecting Korea, we utilized data from the Automated Synoptic Observing System (ASOS), ocean buoys (BUOY), and isobaric observation stations (BEACON), all operated by the Korea Meteorological Administration. This study incorporates long-term meteorological observations from both inland ASOS stations and coastal areas, using BUOYS and BEACON, ensuring comprehensive coverage of both inland and coastal regions of the Korean Peninsula. Based on several previous studies [17,18,19] that analyzed changes in typhoons affecting the Korean Peninsula, the Korea Meteorological Administration has been providing data on typhoon occurrence and frequency since 1951. However, subjective analyses, such as detailed evaluations of these data, have only been conducted since 1954, and most studies have rarely performed long-term analyses to determine changes in typhoon characteristics. Additionally, analyses using observational data recorded during actual typhoon events are uncommon. Therefore, this study aims to examine the long-term changes in the characteristics of typhoons affecting the Korean Peninsula using meteorological observation data provided by the Korea Meteorological Administration. The analysis covers 119 years, from 1904 to 2022, marking the beginning of modern meteorological observation on the Korean Peninsula. This study specifically analyzes trends in the frequency and intensity of typhoon impacts over this period. After applying quality control (QC FLAG) to verify whether the observed values were within normal ranges, only the data labeled as “0” (normal) at 1 h intervals were included in this study.

Typhoons from 1904 to 2010 were analyzed using the typhoon list from the Typhoon White Paper published by the Korea Meteorological Administration in 2011. Typhoons from 2011 to 2021 were analyzed using the Typhoon Analysis Report (2011), and the officially designated “typhoon impact period” announced by the Korea Meteorological Administration was used in this study.

2.2. Methods

In this study, meteorological numerical modeling was conducted to quantify the hazards of strong winds, heavy rains, and storm surges during the typhoon impact period. The results from this modeling were then used as input data for atmospheric diffusion modeling to assess the quantitative hazards of high-concentration air pollution.

Numerical Modeling

In this study, the Weather Research and Forecasting Model (WRF) was used as the numerical model. WRF is a numerical weather prediction and simulation system developed by the National Centers for Environmental Prediction (NCEP), a division of NOAA, and is suitable for both industrial and research applications. It is a next-generation mesoscale numerical weather prediction system designed for both atmospheric research and operational forecasting. WRF is characterized by its two dynamical cores, data assimilation system, support for parallel computation, and system scalability. The version used in this study is WRF version 4.1.2.

The atmospheric diffusion modeling used in this study is the Models-3/Community Multi-scale Air Quality Model (CMAQ), developed by the U.S. Environmental Protection Agency (EPA). The first official version of CMAQ was released in June 1998, and it has continued to undergo active improvements. CMAQ can perform simulations across a range of scales, from local to regional, by adjusting the modeling area, and it is capable of simultaneously considering multiple pollutants. The model can calculate aerosols in addition to sulfur compounds and ozone compounds. For this study, we employed CMAQ version 4.7.1 due to its compatibility with our data and simulation requirements, ensuring reliable outputs for our specific analysis. The CMAQ Chemical Transport Model (CCTM), which is the core component of CMAQ, features a modular structure that allows for the easy modification or removal of individual subprograms and the main program. Generalized coordinates are used, enabling the integration of meteorological data directly into the air pollution model. This ensures consistency between the meteorological model, the air pollution model, and all input data.

Additionally, vulnerability factors were considered for each risk index in the system, including old buildings, impervious surfaces, river density, water depth, urbanization level, and the number of vulnerable people. These factors were based on the typhoon impact period data provided by the government for each administrative region.

3. Results

3.1. Recent Trends and Changes in Typhoons Affecting the Korean Peninsula

3.1.1. Typhoon Impact Frequency

Typhoon Impact Frequency refers to the annual number of typhoons that impact the Korean Peninsula. The frequency of typhoon impacts since 1904, when meteorological observations on the Korean Peninsula began, is illustrated in Figure 1. The number of typhoons each year shows considerable variability, and there appears to be a cyclical pattern. Typhoon frequency increased during the late 1960s and early 2010s. The slope of the linear regression line is 0.0025, with a p-value of 0.7648, which is greater than 0.05, indicating no statistically significant trend in increasing frequency. In other words, the observed increase in typhoon frequency is not statistically significant.

3.1.2. Meteorology Data Accompanied by Typhoons

In this study, extreme values observed during the typhoon impact period were analyzed to examine changes in the intensity of typhoons affecting the Korean Peninsula. Figure 2 presents the results of meteorological data recorded during this period.

The analysis of maximum wind speed and maximum instantaneous wind speed shows an increasing trend in both parameters, suggesting that wind speeds associated with typhoons have been rising over time. In particular, the trend line for maximum instantaneous wind speed shows a statistically significant increase (p-value = 0.0002), indicating that the intensity of peak wind speeds during typhoons is escalating. Similarly, the accumulated precipitation demonstrates a significant upward trend (slope = 1.1328, p-value < 0.0001), indicating that the amount of rainfall associated with typhoons has been increasing. This increase in precipitation further exacerbates the risk of flooding and other weather-related disasters. On the other hand, the analysis of central minimum pressure shows a decreasing trend (slope = −0.0640, p-value < 0.0001), which suggests that typhoons impacting the Korean Peninsula are becoming more intense. A lower central minimum pressure is typically associated with stronger, more destructive typhoons, which could lead to greater damage from wind and storm surge.

While the frequency of typhoons impacting the Korean Peninsula shows an upward trend, this change is not statistically significant (p > 0.05). However, the clear and statistically significant increases in both wind speed and accumulated precipitation, combined with the decrease in central minimum pressure, highlight the growing intensity of typhoons and the increased risk of multi-hazard meteorological disasters, including stronger winds, heavier rainfall, and more severe storm surges.

3.2. Typhoon Ready System

Recent analyses of typhoons affecting the Korean Peninsula show a significant increase in both wind speed and accumulated precipitation, as well as a decrease in central minimum pressure, indicating that typhoons are becoming more intense. Although the frequency of typhoons has not shown a statistically significant upward trend, the increasing intensity of these events suggests a growing risk of multi-hazard meteorological disasters. These changes reflect broader global trends, with meteorological disasters becoming more frequent and severe due to climate change. In response to these changes, the WMO has recommended a shift from phenomenon-based forecasting to impact-based forecasting, emphasizing the need for proactive disaster response. To address this, the NOAA and the NWS in the United States introduced the WRN paradigm shift in 2011 as part of a 10-year strategic plan to minimize the damage from extreme weather events, such as hurricanes, tornadoes, floods, and droughts. In line with the WRN paradigm, many countries have adopted impact-based forecasting systems that allow for early response to tropical cyclones like typhoons. However, while Korea has made progress since 2016 by developing technology, gathering evidence, and creating promotion plans for an impact-based typhoon forecasting system, there have been few practical implementations of these forecasts. The current system remains insufficient for direct use in field operations. To address these limitations, this study proposes the establishment of the Typhoon Ready System (TRS), a decision-making support system designed to enhance the accuracy and utility of typhoon impact forecasts on the Korean Peninsula. TRS will provide more detailed, impact-based meteorological information, enabling local governments and emergency services to take proactive measures to mitigate the risks associated with increasingly severe typhoons in the context of the climate crisis.

3.2.1. Configuration

In the U.S. WRN system, risk data for tropical cyclones are provided, focusing on multi-hazards that may occur within U.S. territories due to hurricanes. Similarly, Japan’s typhoon impact-based forecast system provides information on the risks of all regional meteorological disasters associated with typhoons (Figure 3).

In this study, the TRS was developed to address multi-hazards that can occur on the Korean Peninsula during typhoon impact periods. The system (Figure 4) is designed to provide efficient information for disaster prevention activities related to typhoons, which are often accompanied by multi-hazards. This approach is particularly well suited for Korea, where multi-hazards can occur simultaneously due to the country’s diverse topographical features.

3.2.2. Risk Index

An impact-based forecast goes beyond simply warning of disaster occurrence; it provides regionally differentiated impact assessments that consider the vulnerability of each area. Therefore, in the TRS developed in this study, the meteorological risks associated with strong winds, heavy rains, storm surges, and air pollution, each of which represents a multi-hazard meteorological disaster brought by typhoons—were defined as disaster factors. Regional characteristics and vulnerabilities were also identified, and a risk index was calculated by integrating both the disaster factors and the vulnerability factors.

As shown in Figure 5, the algorithm for calculating the risk index incorporates various input data, such as wind speed, precipitation, storm surge height, and air quality measurements (PM10 concentration), during typhoon periods. These inputs are used to assess specific hazards, like wind gusts, probable maximum precipitation, storm surge height, and high air pollution. The system then calculates the vulnerability of a region by considering factors such as building age, urbanization level, and proximity to rivers. Finally, the risk index is derived for each hazard—strong wind, heavy rain, storm surge, and air quality—by combining the hazard magnitude with the region’s vulnerability characteristics. This risk index helps to provide a comprehensive assessment of the potential impact of typhoons on the Korean Peninsula.

Strong Wind

A strong wind is defined as a wind with a speed of 14 m/s or higher, and typhoons, which are tropical cyclones, are typically accompanied by strong winds. The strong winds generated by typhoons last for a relatively long time, from the time the typhoon approaches until it passes. These winds are not referred to as gusts, and they do not usually cause significant damage. However, when the wind speed and direction fluctuate suddenly during a typhoon, the rapid and intense gusts that occur are often responsible for substantial damage. Therefore, in this study, the 3-second gust (3s gust) was used as the hazard factor for calculating the Strong Wind Risk Index (SWI). The Risk Assessment Prediction Model [20,21,23], which was used to estimate the 3s gust, calculates the average wind speed over a 3 s interval at a height of 10 m (33 feet) above ground level, based on a 50-year return period during typhoons. This model assesses the potential damage to buildings caused by the 3s gust, which represents the maximum instantaneous wind speed during a typhoon [23]. The model was developed with reference to the Florida Public Hurricane Loss Model (FPHLM), created by experts in meteorology, wind and structural engineering, computer science, geographic information systems (GIS), statistics, finance, and insurance accounting.

In this study, data on old buildings by administrative districts were applied to account for vulnerability factors to strong winds in each region. “Old buildings” were defined as buildings over 35 years old, and the ratio of old buildings to the total number of buildings in each administrative district was calculated. As shown in Figure 6, which illustrates the number of old buildings by administrative district in 2002 (a) and 2021 (b), a significant increase in the number of old buildings over time can be observed, particularly in certain regions of the Korean Peninsula. This increase highlights the growing vulnerability of these regions to strong winds due to aging infrastructure. Given the significant variation in the data from year to year, annual data were used to accurately capture the changes in building age distribution. The old building data were integrated into the calculation of the SWI by considering both the proportion of old buildings and the total number of buildings in each administrative district. To calculate the SWI, the proportion of old buildings was weighted based on the total number of buildings and old buildings in each administrative district, the minimum spatial unit used for analysis. The Strong Wind Risk Index was calculated by applying this weighted ratio to the 3s gust, as shown in Equation (1).

Figure 6. Number of old buildings by administrative district in 2002 and 2021.

Figure 6. Number of old buildings by administrative district in 2002 and 2021.

$$SWI=3s gust\times {\displaystyle \frac{B}{\overline{B}}}$$

(1)

$3s gust$: Maximum possible instantaneous wind speed.

$B$: Number of old buildings by administrative district.

$\overline{B}$: Total number of buildings by administrative district.

Heavy Rain

In this study, the probable maximum precipitation (PMP), as suggested by [24], was used as an index to assess the risk of heavy rain associated with typhoons. PMP represents the maximum precipitation that can occur under the most extreme meteorological conditions in a specific region. It is widely used in civil engineering and water resource management as a standard index. In this study, PMP was applied as a disaster factor for heavy rain, in a similar way to how the 3s gust was used as a disaster factor for strong winds. The calculation of PMP followed the formula in Equation (2), as outlined by [25], which is the most widely used statistical method globally for PMP estimation.

$$h_m=\overline{h_n}+k_n{s}_n$$

(2)

$h_m$: PMP.

$\overline{h_n}$: Average of the maximum time series observed at the point.

$s_n$: Average deviation of the maximum time series observed at the point.

$k_n$: Frequency coefficient.

As urbanization continues in low-lying areas prone to flooding, particularly in urban floodplains, the risk of flood damage caused by torrential rain in urban watersheds is increasing compared to other regions. There is a strong correlation between the extent of impermeable surfaces, water quality, ecosystem health in rivers, runoff volume, and the scale of natural disaster damage [26,27]. The increase in impermeable surfaces due to urbanization leads to higher surface runoff and places additional strain on drainage systems, thereby raising the risk of flood damage in urban areas. In the report Development of an Evaluation System Based on the Flood Vulnerability Index, the impermeability rate (as shown in

Table 1. The impermeability rate by administrative district.

Administrative	Percentage of Impermeable	Administrative	Percentage of Impermeable
Seoul	53.605	Gangwon	2.847
Busan	83.587	Gyeong-gi	11.844
Ulsan	21.019	Gyeongnam	15.844
Daegu	89.631	Gyeongbuk	4.488
Daejeon	25.027	Jeonnam	6.664
Gwangju	41.091	Jeonbuk	7.344
Incheon	19.273	Chungnam	8.084
Sejong	12.510	Chungbuk	6.510
		Jeju	8.801

) was used as a topographical factor in assessing flood exposure vulnerability. In this study, the impermeability rate at the administrative district level was considered a key vulnerability factor for heavy rain associated with typhoons. Data provided by the Korean Statistical Information Service on land use in each administrative district were utilized to calculate the area of various land types, such as residential, industrial, transportation, and park areas. The impermeability rate was determined by calculating the proportion of impermeable surfaces within the total area of each administrative district.

Table 1 presents the impermeability rate by administrative district. For instance, Seoul and Busan show the highest impermeability rates, with 53.605% and 83.587%, respectively, indicating significant urbanization and a higher risk of flood damage. Conversely, Gangwon and Jeju show much lower impermeability rates, with 2.847% and 8.801%, respectively, suggesting lower vulnerability to flooding. This regional variation highlights the importance of considering impermeability rates when assessing the vulnerability to heavy rainfall. Since the analysis data were only available from 2007 onward, data from 2002 to 2007 were applied in the analysis, and data from the relevant years were used for other periods [28].

After calculating the risk of heavy rain accompanying the typhoon by administrative district, the Heavy Rain Risk Index (HRI) accompanying the typhoon was calculated by considering the weight of each administrative district for the impermeability rate data and river density data calculated through the above results. Among the factors affecting flood damage in the Flood Damage Index (FDI), which can easily identify the potential risk of flood damage proposed by [29], the weight and weighting ratio for each factor corresponding to natural factors were used and applied to the heavy rain risk index in this study. In this study, the impermeability rate and river density data were obtained as data for each administrative district of the vulnerable factors for heavy rains accompanied by typhoons and used to calculate (3).

$$HRI=Rainfall\times I\times RD$$

(3)

$Rainfall$: PMP, $I$: Permeability weight, $RD$: River density.

Storm Surge

In this study, to estimate the risk of storm surges along the coast caused by tropical cyclones, various meteorological factors such as the typhoon’s movement speed and central atmospheric pressure were considered. The Storm Surge Hazard Potential Index (SSHPI) [30], which accounts for the lack of limitations on data availability, was used. SSHPI was developed to estimate the height of the highest tidal wave caused by a tropical cyclone, and the calculation formula is shown in Equation (4).

$$SSHPI=\left({\displaystyle \frac{V_{max}}{V_{ref}}}\right)\left({\displaystyle \frac{R_{33}}{R_{ref}}}\right)\left[\left(1013-P_c\right)\right]\left({\displaystyle \frac{L_{30}}{L_{*}}}\right)$$

(4)

$V_{max}$: Maximum wind speed of typhoon, $V_{ref}$: 33 m/s.

$R_{33}$: 33 m/s strong wind radius, $R_{ref}$: 96.6 km, $P_c$: Central atmospheric pressure.

$L_{30}$: 30 m deep separation distance, $L_{*}$: 40 km.

As a storm surge approaches the shore, friction between the water and the seabed reduces its speed. Consequently, the wave’s speed decreases in shallower waters and increases in deeper waters. Higher speeds result in less energy loss as the wave travels, leading to greater destructive power upon landfall. In this study, the depth separation distance was considered as a disaster factor in SSHPI. The coastal administrative districts were categorized into three regions—West Sea, South Sea, and East Sea—based on the depth separation characteristics, which were applied in the analysis. When evaluating the risk to coastal areas, the degree of urbanization was included as a vulnerability factor. The Korean Peninsula, the subject of this study, has 55 coastal administrative districts managed by relevant central government agencies. These districts were classified into metropolitan, urban, and rural areas by [31], and this classification was used in this study (Figure 7). Figure 7 visually depicts the degree of urbanization in these coastal districts, illustrating those metropolitan areas, which tend to have higher levels of urbanization, may face greater vulnerability to storm surges due to increased impermeable surfaces and higher population density.

Additionally, the most significant factor affecting the size of the storm surge, the astronomical tide level, was applied to the Storm Surge Risk Index (SSI). For this study, the tidal observation data from the National Oceanographic Research Institute were used, and the tidal level from the nearest observation station was selected as the representative value for each administrative district. The data used were from 2002 onward, when the analysis began, and the annual average tidal level was set to 1. The SSI was calculated by applying a weight to the tidal level for each administrative district accordingly.

$$SSI=\left({\displaystyle \frac{V_{max}}{V_{ref}}}\right)\left({\displaystyle \frac{R_{33}}{R_{ref}}}\right)\left[\left(1013-P_c\right)\right]\left({\displaystyle \frac{L_{30}}{L_{*}}}\right)\times urbanization\times tide$$

(5)

$urbanization$: degree of urbanization, $tide$: tide value.

Air Quality

The Air Quality Risk Index (AQI) focuses on air quality, which may not be an obvious concern in the context of multi-hazard meteorological disasters associated with typhoons. It is believed that pollutant concentrations in the air decrease during a typhoon due to strong winds and precipitation, which help disperse pollutants. However, contrary to this perception, high concentrations of air pollution have frequently been observed during the typhoon impact period on the Korean Peninsula. To assess this, PM10 concentration values were calculated before, during, and after the official typhoon impact period, as announced by the Korea Meteorological Administration. The period was divided into three-day intervals based on meteorological criteria [32,33,34]. The risk of high-concentration fine dust during typhoons was calculated by determining the probability of PM10 concentrations reaching 81 μg/m³ or higher, classified as ‘bad’ air quality according to forecasting standards.

In this study, the number of vulnerable populations by administrative district was applied to evaluate the risk of high-concentration fine dust during typhoons (Table 2). Vulnerable populations include children (aged 0 to 17) and the elderly (aged 65 or older). These groups are particularly sensitive to air pollution, and their exposure to high concentrations of fine dust during a typhoon may result in serious health impacts.

Table 2. Total population by administrative district and vulnerable population (elderly, children).

	Over 60 Years of Age	Children (0–17 Years Old)	Vulnerable Population	Total Population	Rate
Seoul	2,124,510	1,227,792	3,352,302	9,586,195	0.35
Busan	892,277	442,776	1,335,053	3,349,016	0.4
Deagu	559,877	352,737	912,614	2,410,700	0.38
Incheon	597,761	443,334	1,041,095	2,945,454	0.35
Gwangju	290,547	241,381	531,928	1,477,573	0.36
Deajeon	299,964	228,182	528,146	1,488,435	0.35
Ulsan	218,134	186,950	405,084	1,135,423	0.36
Sejong	48,720	83,214	131,934	353,933	0.37
Gyeong-gi	2,529,512	2,190,957	4,720,469	13,511,676	0.35
Gangwon	431,276	211,536	642,812	1,521,763	0.42
Chungbuk	390,720	239,792	630,512	1,632,088	0.39
Chungnam	527,371	332,379	859,750	2,176,636	0.39
Jeonbuk	492,736	263,774	756,510	1,802,766	0.42
Jeon-nam	537,046	261,470	798,516	1,788,807	0.45
Gyeongbuk	749,388	365,104	1,114,492	2,644,757	0.42
Gyeongnam	790,629	523,311	1,313,940	3,333,056	0.39
Jeju	142,796	116,257	259,053	670,858	0.39

Table 2. Total population by administrative district and vulnerable population (elderly, children).

	Over 60 Years of Age	Children (0–17 Years Old)	Vulnerable Population	Total Population	Rate
Seoul	2,124,510	1,227,792	3,352,302	9,586,195	0.35
Busan	892,277	442,776	1,335,053	3,349,016	0.4
Deagu	559,877	352,737	912,614	2,410,700	0.38
Incheon	597,761	443,334	1,041,095	2,945,454	0.35
Gwangju	290,547	241,381	531,928	1,477,573	0.36
Deajeon	299,964	228,182	528,146	1,488,435	0.35
Ulsan	218,134	186,950	405,084	1,135,423	0.36
Sejong	48,720	83,214	131,934	353,933	0.37
Gyeong-gi	2,529,512	2,190,957	4,720,469	13,511,676	0.35
Gangwon	431,276	211,536	642,812	1,521,763	0.42
Chungbuk	390,720	239,792	630,512	1,632,088	0.39
Chungnam	527,371	332,379	859,750	2,176,636	0.39
Jeonbuk	492,736	263,774	756,510	1,802,766	0.42
Jeon-nam	537,046	261,470	798,516	1,788,807	0.45
Gyeongbuk	749,388	365,104	1,114,492	2,644,757	0.42
Gyeongnam	790,629	523,311	1,313,940	3,333,056	0.39
Jeju	142,796	116,257	259,053	670,858	0.39

$$AQI={\displaystyle \frac{M_h}{\overline{M}}}\times {\displaystyle \frac{O_h}{\overline{O}}}\times {\displaystyle \frac{P}{\overline{P}}}$$

(6)

$M_h$: The number of hours of high concentration.

$\overline{M}$: Total number of hours.

$O_h$: High concentration of hours observed by administrative district.

$\overline{O}$: Total number of hours observed by administrative district.

$P$: Number of vulnerable populations by administrative district.

$\overline{P}$: Total population by administrative district.

Case: Typhoon Rusa (0215)

Typhoon Rusa made landfall on the Korean Peninsula on 31 August 2002, leading to 246 people either dead or missing, and caused over KRW 5 trillion in property damage. This typhoon, known for its heavy rainfall, provides a representative example for analyzing the Heavy Rain Risk Index (HRI). Figure 8 illustrates the various components of the HRI calculation, showing the hazard index (a), vulnerability factors such as river density and impermeability rate (b), and the resulting Heavy Rain Risk Index (c).

In Figure 8a, the hazard index, represented by the probable maximum precipitation (PMP), indicates that the highest rainfall was expected in Gangwon-do, Gyeongsang-do, I modified it.and Jeju. These areas recorded significantly higher PMP values, meaning that these regions were at a higher risk of extreme rainfall during Typhoon Rusa. Figure 8b shows the vulnerability factors, including river density and impermeability rate, which are essential for calculating the overall risk. Regions with a higher river density, such as Gangwon-do and parts of the metropolitan area, are more vulnerable to flooding due to the increased risk of overflowing rivers. The impermeability rate, which measures the percentage of land that cannot absorb water (e.g., urbanized areas), also plays a crucial role. Areas with high impermeability rates, such as the metropolitan area and parts of Gyeonggi-do, have a greater risk of surface runoff and flooding, even if the total rainfall is moderate. Figure 8c combines these factors to calculate the HRI. While areas like Gangwon-do show high HRI values due to their high PMP, other regions such as the metropolitan area and Gyeonggi-do, where the PMP was moderate, still exhibit a high HRI due to their higher vulnerability factors. This demonstrates how regional vulnerabilities, such as urbanization and river density, can amplify the risk of flooding even when the rainfall itself is not the most extreme. The example of Typhoon Rusa underscores the importance of considering both meteorological hazards and regional vulnerabilities when assessing the risk of heavy rain.

The slow movement of Typhoon Rusa, combined with the unique topographical features of Korea, vulnerability factors such as the Sobaeksan Mountain range, led to significant rainfall in metropolitan area and the YeongDong region. The actual damage and the KMA’s post-typhoon analysis, including the collapse of reservoirs and widespread flooding, aligns with the high HRI values calculated for these areas. By integrating these factors, the HRI successfully identified the areas where Typhoon Rusa had the most severe impact, demonstrating its reliability as a predictive tool for HRI. In summary, the real-world damage caused by Typhoon Rusa, on the vulnerability of regions with high impermeability rates and river density, supports the accuracy of the HRI model used in this study. The results of the HRI align closely with the actual impact observed during Typhoon Rusa, demonstrating the index’s effectiveness in identifying regions at high risk of flood damage during extreme rainfall events.

The actual damage is also consistent with the analysis results of this paper due to the collapse of reservoirs and flooding of several facilities. Examples and accuracy analysis of all risk index will be introduced in a future paper.

4. Summary

The increasing severity of meteorological disasters, driven by climate change, urbanization, and demographic shifts, highlights the urgent need for more advanced forecasting systems. Traditional phenomenon-based forecasting methods, which focus solely on the occurrence and timing of events, are no longer sufficient to address the socio-economic impacts of modern disasters. This study confirmed, through long-term data analysis, the rising intensity of typhoon impacts on the Korean Peninsula and the need for a specialized forecasting system.

In response, the TRS was developed, representing a significant advancement by combining meteorological hazard data with vulnerability factors to produce a comprehensive risk index. This multi-dimensional approach enables more accurate predictions of typhoon impacts and supports better disaster preparedness. The TRS allows local governments to prioritize resources and tailor disaster response strategies more effectively, and its modular design ensures flexibility for future enhancements such as real-time data integration and machine learning algorithms. Although the TRS has been successfully implemented for the Korean Peninsula, efforts are currently underway to expand its application to other regions through pilot projects in various countries. These projects involve customizing the system to account for different geographic and socio-economic conditions.

In conclusion, this study presents a pioneering approach to meteorological disaster forecasting through the TRS. By addressing both physical issues and vulnerabilities, the TRS offers a more comprehensive framework for disaster preparedness. As climate-related risks escalate, systems like the TRS will play a vital role in mitigating the impacts of meteorological disasters globally.