A method for estimating the risk of fishery ports against typhoon: a case study on Dongsha fishery port

After standard seawalls have been built successfully, fishery ports become the structures most easily damaged during a typhoon. Estimating the risk of fishery ports against typhoons would be useful for identifying weaknesses and implementing corrective measures to protect fishing boats from a typhoon. This study describes a versatile methodology for conducting this type of quantitative assessment at fishery ports. The Dongsha fishery port in Zhejiang Province was selected as a case study to test the results derived from a high-precision Hydrodynamic Flexible Mesh model coupled with the Spectral Wave model. First, typhoon characteristics were assessed based on historical typhoons in the study area, and then, the wind, tide, storm surge, and waves were modeled and tide-surge interactions were investigated. Through comparisons of the destructive parameters from the typhoon assessment with the design and structural parameters of the fishery port, the level of the Dongsha fishery port against typhoons was determined to be 12, and the main weaknesses of the port’s defenses were found to be located near feature points T2, T3, T8, and T15. The results obtained demonstrate that the proposed methodology can be used to acquire valuable information on the risk of fishery ports against typhoons.


Introduction
As one of the countries with the largest fishery resources around the world, China ranks first in the world in terms of the output of aquatic products, number of fishing boats, and number of fisheries employees (NDRC and MARA 2018). However, as China lies on the west coast of the Pacific Ocean, its coastal areas are susceptible to various marine disasters, especially typhoons and storm surges (Ministry of Natural Resources of the People's Republic of China, 2019). Within China, Zhejiang Province near the East China Sea is well-known for fishing. The total marine fishery production output here was ranked first nationally at 3,200,000 t. Coastal areas within Zhejiang, especially the city of Wenzhou, are vulnerable to typhoon-related damage (Du et al. 2020;Shi et al. 2020b). Almost every year, more than one typhoon strikes the coast of Zhejiang Province, and these typhoons frequently cause damage to the breakwater structures, wharfs, and fishing boats. According to the Zhejiang Marine Disaster Bulletin (2019), a total of 2064 fishing boats were damaged by typhoons, and the direct economic losses due to typhoons amounted to 87.25 hundred million yuan (Department of Natural Resources of Zhejiang Province 2019). Since record keeping began, the largest storm surge event near the Dongsha fishery port occurred in 1997 (210 cm at the Kanmen tide gauge station). Significant fluctuations in the sea level are caused by the strong winds in the low-pressure storm systems that cross over the Dongsha fishery port. As storms pass over the sea, the conditions create storm surges. Low atmospheric pressure and winds cause an increase in water levels at nearby coastal areas, which often leads to flooding (Wang et al. 2017). After standard seawalls are successfully built, disaster prevention and mitigation efforts at fishery ports become particularly important. Knowledge of the degree of fishery ports against typhoons would be of great benefit to disaster prevention plans and coordination of mitigation activities within a region.
Most research on fishery ports has focused on the biology and ecology of a port and its geomorphic stability. However, the ability of fishery ports to resist the damage caused by typhoons has not received much research attention yet. Notably, Premwadee et al. (2006) studied the trends in marine fish catches at the Pattani fishery port, and Kawaguchi et al. (1995) presented construction recommendations for an offshore fishery port to prevent coastal erosion following hydraulic model tests and numerical simulations of wave induced currents near the port. Additionally, there have been numerous studies about the risks of hurricanes or typhoons at home and abroad. In America, the National Weather Service storm surge model, named SLOSH (Sea, Lake, and Overland Surge from Hurricanes), has been used to delineate coastal areas susceptible to hurricane storm surge flooding (Glahn et al. 2009). A computer simulation of super typhoon Haiyanin with the resulting wave heights and storm surge levels was made using the MIKE21 model in Tacloban city (Prelligera et al. 2014). Li et al. (2019) examined the dependence of typhoon-induced storm surge and wave setup effects on the typhoon intensity and size. MIKE21 was also used to evaluate the overtopping risk of seawalls and levees from the combined effects of the storm tide, sea level rise, and land subsidence in Shanghai (Wang et al. 2011). A methodology for storm surge risk assessments in coastal counties was established following research in Jinshan District, Shanghai city (Shi et al. 2020a).
Estimating the risk of fishery ports against typhoons is a difficult task. In particular, because each fishery port is different in terms of its geographical location, topography, anchoring water, and shape, we cannot carry out one assessment under the same typhoon conditions or even for completely different typhoons. Additionally, the storm surge can be influenced significantly by the landfall location of a typhoon with the same pressure (Sun et al. 2015a, b). Abeshima et al. (2017) clarified the mechanism of port disturbance generation at the Kumaishi fishery port and concluded that the quantitative indicator H 1/3 (over 2.0 m) can be introduced as a decision indicator for evacuations by observational statistics. Some exploratory work has been conducted in China on fishery ports' resistance to damage caused by typhoons. Notably, one study used an analytical hierarchy process for indexes of wind and wave features, the number of sheltering boats, anchoring methods, emergency measures, and the local management system to assess the relative preparedness of fishery ports in Xiamen against typhoons (Dongshui and Qionglin 2019). Based on the nested model of Delft3D, the fishery ports were first evaluated in terms of the following three aspects: the level of shoreline facilities, anchorage areas, and breakwaters to lessen typhoon impacts (Sun et al. 2017). And the wind direction of the maximum observed frequency was roughly regarded as the typhoon pathway, which was the key factor in that study. Moreover, the interaction of the tide and surge was not taken into account.
Since the modern seawall and wharf are solid and rarely breakdown, the infrastructure stability is not considered in this paper. So the risk of a fishery port during a typhoon is consist of two aspects in this study. One is that when the water level during typhoon exceed the design height of coastal, it will cause severe economic losses and casualties (Wu et al. 2019). And the other is that the ships drag anchor during typhoon, causing collisions, groundings, and other serious accidents (Yang 2017). This study describes a systematic and quantitative method for estimating the risk of fishery ports against typhoons. The method can be used to conduct comparisons among different fishery ports, and the proposed method also relies on basic data for the three aspects described above. Additionally, the typhoon resistance capability of a fishery port is indicated by the sustainable maximum wind scale of the port. Meanwhile, typhoon pathways and tide-surge interactions, the key factors of the assessment, are studied in detail. After deriving a quantitative value for the resistance level of a fishery port against typhoons, effective countermeasures for typhoons can be proposed, and such data should also be useful for making judgments as to the need for evacuations by administrators.

Study area
The Dongsha fishery port is located on the east side of Dongtou Island (121°10 0 -121°11 0 E and 27°50 0 -27°51 0 N) in the city of Wenzhou, China (Fig. 1). It is C-shaped and surrounded on three sides by mountains, which makes it a natural sheltered harbor for fishing boats. Presently, it is the best sheltered harbor in Wenzhou. The length of the fishery port coast is 5.17 km, and there is a 0.35 km long breakwater, which was built at the entrance of the fishery port. The water area of the port is approximately 750,000 m 2 with a depth of 3-9 m.

Data
To model a specific area of a vulnerable fishery port, accurate topographic, meteorological, and other types of basic data are required. Here, multisource data were classified into four types (Table 1) and were used to run and validate the Hydrodynamic Flexible Mesh (HD FM) model coupled with the Spectral Wave (SW) model for the Dongsha fishery port. The topographic data (Fig. 2), which were at the same datum plane, were collected to construct the numerical model, and the meteorological data and hydrologic data were used as dynamic data and validation data for the numerical model. The design and structural parameters of the Dongsha fishery port were compared with the numerical simulation results, and then, these results were used to judge the resistance of the port.

Methods
In this study, a framework ( Fig. 3) is proposed for estimating the risk of fishery ports against typhoons related damage. The framework is composed of the following five parts: typhoon building, model configuration and verification, hazard simulation, individual assessment, and comprehensive assessment.
For typhoon building, for the convenience of reading, there are some terms that need to be explained first. The influential typhoons refer to historical typhoons that have had an impact on the study area within a certain distance. The typical typhoons are typhoon categories classified from the influential typhoons by certain rules. The alternative assessment typhoons refer to alternative typhoon prototypes that are representative in each typical typhoon category. The final assessment typhoon is the typhoon with the maximum destructive parameters among the alternative assessment typhoons.
Using MIKE21 software, the current, storm surge, and waves under various typhoon scenarios were simulated. These scenarios provided the information required for the assessment and were chosen so that the data would cover future typhoon events anticipated to have significant impacts on the Dongsha fishery port. The wind data and current data were used to calculate the stresses on fishing boats, which were compared with the holding power of anchors. The level of the anchorage against typhoon was represented by the minimum typhoon intensity when the force from the wind and currents was larger than the holding power of the anchor. In a similar manner, the level of the shoreline facilities against typhoon was represented by the minimum typhoon intensity when the water level of  Stochastic Environmental Research and Risk Assessment (2022) 36:1993-20131995 storm surge adding to 1/2 the significant wave height (Yang et al. 2021) was higher than the coastline elevation. Similarly, the level of the seawall against typhoon was represented by the minimum typhoon intensity when the significant wave height or the sheltered area of the typhoon was higher than the design wave of the seawall.

Numerical model configuration
The 2D shallow water model has been shown to reproduce storm surges well (Bertin et al. 2012). Notably, MIKE21 was used successfully for the simulation of tidal waves during a storm surge in the north part of Liaodong Bay (Kong 2014). In this study, the MIKE21 model was used to construct the hydrodynamic module, storm surge, and where Nðx; r; h; tÞ is the action density; x ¼ ðx; yÞ is the Cartesian co-ordinates; t is the time; m ¼ ðc x ; c y ; c r ; c h Þ is the propagation velocity of a wave group in the four-dimensional phase space x, r and h; and S is the source term for the energy balance equation. r is the four dimensional differential operation in the x; r; h space. MIKE21 FM uses the finite volume method to solve the Navier-Stokes equations. The shallow water equations are as follows: where x, y are the Cartesian co-ordinates; g is the surface elevation; h is the total water depth; u, v are the velocity components in the x, y direction; f = 2Xsin/ is the Coriolis parameter; g is the gravitational acceleration; q is the density of water; S xx , S xy , S yx and S yy are components of the radiation stress tensor; q 0 is the reference density of water; S is the magnitude of the discharge due to point sources. Unstructured meshes were used in the model, along with atmospheric pressure and wind. Detailed information for MIKE21 can be found in the scientific documentation and user guide for the model (DHI 2012).
The inset of Fig. 4 shows the computational domain and the mesh grid. It covered a large area that ranged from 106°F ig. 3 Framework for assessing the level of fishery ports against typhoons to 135°E and 12°to 41°N; a large area was used to properly reproduce storm surges and waves generated at a greater distance from the Dongsha fishery port. The grid used was fine near the area of interest and decreased in resolution in the deepwater area where minute details were not as important. There were 67,549 grid cells and 35,899 nodes, which became denser closer to the Dongsha fishery port. The minimum resolution of the grid size was 20 m, which could embody the seawall, wharf, and other structures. The bathymetry data were obtained from several charts from the Maritime Safety Administration of the People's Republic of China and actual measurements, which were unified at the same datum plane of the 1985 national height datum. The boundaries of tidal elevation and storm surge elevation model are the tidal elevation data extracted by Tidal Model Driver.

Numerical model verification
The typhoons of 9711, 0509, 0713, 0716, and 1509 were selected when the observed values were the maximum or the typhoon caused relatively extensive damage. The numbers published by the China Meteorological Administration are indicative of the year and order of typhoons that have impacted China, for example, 9711 means the 11th typhoon that occurred during 1997. There were some ocean gauge stations near the Dongsha fishery port, and each station observed different oceanographic elements. The storm surge model was validated with the data from the Kanmen and Wenzhou tide gauge stations. The wave model was validated with the data from the Nanji and Wenzhou wave gauge stations. The gauge stations are shown in Fig. 5. The whole hourly storm surge was  Modeling with good results very close to the observed data was very difficult to achieve, as the wind, rain, current, and wave interactions were complex during a typhoon. The preliminary results showed that it was possible to forecast the effects of storm surges and waves by several days in advance.

Typhoon prototype selection
Influence typhoons were chosen by a method of distance screening from the history of typhoons, which amounted to 1841 typhoons in total for China during the period 1949-2017. The method of distance screening involved drawing a circle with the fishery port at the center and a radius of 40 km. This radius was set because the geometric mean radius of maximum wind is 47.5 km in the Atlantic and eastern Pacific (Willoughby and Rahn 2004) and concentrated at 40 km in the western North Pacific . The influential typhoons were classified in order to determine typical typhoon conditions. Assessment of the typhoons was carried out with the maximum risk for alternative assessment typhoons according to the results of simulations. First, 1841 historical typhoon pathways were collected from the tropical cyclone information center of the China Meteorological Administration, and these typhoons all occurred from 1949 to 2017. Next 123 influential typhoons were chosen by the method of distance screening from the abovementioned historical typhoon pathways (Fig. 8).
Because the influential typhoons occurred in all directions, the influential typhoons were categorized into four typical typhoon patterns according to their pathways as shown in Fig. 9. At the same time, by considering the opening direction of the Dongsha fishery port where the seawall gap faces toward the southeast, the typhoon pathway toward the northwest was selected as the fifth typical typhoon pattern. Then, five representatives were selected from each typical typhoon pattern. Next, five representatives were moved to a radius of 40 km around the Dongsha fishery port, and these represented the alternative assessment typhoons ( Fig. 10; Table 4).
Five scenarios of different alternative assessment pathways under a level 17 typhoon were calculated, including the storm surge and typhoon waves. Seven feature points, as shown in Fig. 11, were extracted from the results to reflect the area of the seawall (B1), entrance (B2), anchorage water (B3, B4), wharf (B5, B6), and Dawangdian Bay (B7). The south side toward the west scenario was selected as the final assessment typhoon pathway (Fig. 12), during which the storm surge and waves were at the maximum values at the feature point (Table 5).
In this study, the main approaches used for the typhoon wind field modeling were described by the Fujita Model, which has been employed in this area Fujita 1952).

Parameter setting
The tide of the open boundary was determined by using the tide obtained from the Tide Model Driver (TMD) package with its harmonic components (M2, S2, N2, K2, K1, O1, P1, Q1, and M4). The resulting forcing had a time step of 1 h. The input parameters for the wind model were the radius of maximum winds, traveling speed, and pressure difference between the storm's central pressure and the ambient (or peripheral) pressure. The radius of maximum winds was estimated from available observations by using a previously published empirical formula (Zhu and Huang 2002): where R k is the empirical parameter (usually a value of 40 km was used), R is the radius of maximum winds, and P 0 is the central pressure.
The traveling speed for the forward velocity of the storm was obtained from the observed value of the prototype typhoon, for which data were collected from the China Meteorological Administration. The pressure difference of the typhoon was derived from the wind information provided in the typhoon history. According to experience and norms (China Meteorological Administration 2006), 10 different values for the central pressure were used, namely, 995, 991, 985, 975, 965, 955, 945, 935, 925, and 915 (Table 6).

Tide-surge interaction calculation
The water level is presumed to be a superposition of the tide and surge. The impacts of typhoon parameters on the storm were studied . Storm surges are known to have some potential interactions with tides (Flather 2001). Idier et al. (2012) concluded that the instantaneous tide-surge interaction is non-negligible in the eastern half of the English Channel, where it reaches values of 74 cm in the Dover Strait. From an operational perspective, an understanding of this interaction is of value in order to choose relevant strategies in the risk analysis. Thus, to better assess the resistance level of the fishery port against typhoon damage, tide-surge interactions were investigated. The coupling processes of storm surges and tides were simulated in the following way. The surges were computed by gradually adding 2 h tide interactions under the level 17 typhoon. Considering the tide period in this area, there were seven scenarios. ''ST-2'' represented 2 h after the ''ST'' scenarios, and ''ST?2'' represented 2 h before the ''ST'' scenarios. In this study, as shown in Table 7, the maximum storm surge occurred during the ''ST-6'' scenarios, that is, most of the largest practical storm surges occurred around low tide, which is similar to results of the other study (Idier et al. 2012). Then, ''ST-6'' scenarios as tide-surge interaction conditions were used for further simulation.

Force calculation
The forces exerted on fishing boats from wind were divided into lateral and vertical directions as follows: where F xw and F yw are the component forces from wind in the lateral and vertical directions (kN), respectively; A xw and A yw are the above water force area in the lateral and vertical directions (m 2 ), respectively; V x and V y are the wind speed in the lateral and vertical directions (m s -1 ), respectively; f 1 is a nonuniform coefficient that was set to the recommended value of 1 in this study; and f 2 is the altitude correction factor that was set to the recommend value of 1 in this study (Ministry of Transport of the People's Republic of China 2006). The forces exerted on fishing boats from currents were calculated by the following formulas: where F xsc and F ysc are the component forces from currents in the lateral and vertical directions (kN), respectively; C xw and C yw are the coefficients of the fore and aft, which were obtained from a look-up table (Ministry of Transport of the People's Republic of China 2006) as 0.09 and 0.04, respectively; V is the current speed (m s -1 ); q q is the water density (kg m -3 ); and B 0 is the underwater area of the lateral direction (m 2 ). The force exerted on the ships was the resultant force of the wind and current: The anchor holding power of fishing boats was calculated by the following formula: where P is the resultant force of anchor holding (kN); P a is the force of anchor holding (kN); P c is the force of anchor chain holding (kN); k a is the coefficient of the anchor, which was set to 3.5 in accordance with the clayey silt bottom material; k c is the coefficient of the anchor chain, which was set to 0.6 in accordance with the clayey silt bottom material; W a is the anchor weight, which was set to 0.15 t, 0.5 t, and 0.7 t for large, medium, and small types of fishing boats; W c is the anchor chain weight per meter; and l is the length of the anchor chain underground. In Dongsha fishery port, each boat is anchored by two anchors, main anchor and auxiliary anchor, respectively on the fore and the aft. The forces acting on the fore and the aft are considered. According to local knowledge, the resultant force of the fore and the aft is 1.3 times the resultant force P.

Results
Two types of runs were implemented with the HD model, namely, one with the forcing (tide, wind, atmospheric pressure) and the other with the tide only. Based on historical storms and in collaboration with constructive typhoon characteristics, a suit of typhoon scenarios under level 8-17 typhoons were created for surge and wave modeling using HD and SW. These scenarios provided the information required for the assessment and were chosen so that the data would cover future typhoon events anticipated to have significant impacts on the Dongsha fishery port.
Next the results will be analyzed considering the following three aspects: seawall, berth waters, and shoreline.

Seawall
The design and construction data for the seawall shows that the design wave elements H 1/3 of a 50-year return period is 6.5 m at the seawall head, and the H 1/3 was 6.7 m at the seawall toe. The data extracted from typhoon scenario calculations were compared with the design wave elements (Tables 8,9). The design wave elements are smaller than the calculated elements at both the seawall head and seawall toe under a level 13 typhoon. Additionally, to resist a typhoon, the design wave elements should be larger than the calculated elements. Thus, from the design wave point, the resistance level of the Dongsha fishing against typhoon damage is 12.
According to the design data, the sheltered areas for large, medium, and small types of fishing boats are 70,000 m 2 , 280,000 m 2 , and 180,000 m 2 , respectively. Anchoring wave conditions of large, medium, and small types of fishing boats are 1.2 m, 1.0 m, and 0.5 m, respectively. A distribution map of the wave amplification that propagated into the port is shown in Fig. 13. Because it is shielded by Dongtou Island, inrushing waves at the fishery port are small. The sheltered areas of level 8-17 typhoon scenarios are presented in Table 9. The areas where H 1/3 is smaller than 0.5 m, 1.0 m, and 1.2 m are compared between the design and simulation. For instance, the design area where H 1/3 \ 0.5 m is 18 9 10 4 m 2 , which is much smaller than the simulated sheltered area 65.1 9 10 4 m 2 under the level 8 typhoon. Under the same typhoon level, the design area where H 1/3 \ 1.0 m is (18 ? 28) 9 10 4 m 2 , which is still smaller than the simulated sheltered area (65.1 ? 3.1) 9 10 4 m 2 . Similarly, the design area where H 1/3 \ 1.2 m is (18 ? 28 ? 7) 9 10 4 Fig. 11 Feature points of the five alternative assessment typhoon scenarios (from Google Earth) Stochastic Environmental Research and Risk Assessment (2022) 36:1993-20132005 m 2 , which is also smaller than the simulated sheltered area (65.1 ? 3.1 ? 0.2) 9 10 4 m 2 under the level 8 typhoon.
Thus, we could conclude that the Dongsha fishery port can resist the level 8 typhoon from the aspect of the sheltered Fig. 12 Pathway of the assessment typhoon area. In a similar manner, the comparisons were carried out at the remaining typhoon levels. The results showed that the maximum resistance level of the Dongsha fishery port against typhoon damage is 16.
According to the principle of high not low, the resistance level of Dongsha fishery port against typhoon damage is 12.

Berth waters
A total of 23 feature points were selected for fishing boats anchored in water in accordance with information from the fishery port's administration department (Fig. 14). In Fig. 14, the rectangles represent berth waters and the feature points are at the centers of the rectangles. Considering the long period force on fishing boats, the data for the wind and currents at those points were extracted from a suit of typhoon scenarios under level 8-17 typhoons (Table 10).
In the Dongsha fishery port, each boat is anchored by two anchors on the fore and the aft. The forces of fore and aft are considered. By comparing the force exerted on the ship with the resultant force of the fore and aft (Table 11), it could be concluded that the resistance level of the Dongsha fishing against typhoons damage is 12.

Shoreline
In consideration of the features of the Dongsha fishery port, 20 points were selected to represent the different types of shoreline (Fig. 15). Regarding the typhoon rating assessment for the shoreline, knowledge on the elevation of the coastline and the water was required. The water elevation was the height of the storm surge adding to 1/2 H s . The results for the shoreline against typhoons are shown in Design wave elements 6.5 6.7  Table 12 and Fig. 13. The elevation of the shoreline should be higher than that of the water. Therefore, it could be concluded that the resistance level of Dongsha fishing against typhoons damage is 12.

Comprehensive assessment
According to the ''Regulation for typhoon prevention assessment of fishery ports,'' there are two types of typhoon damage resistance levels for fishery ports. One represents the lowest level, while the other represents the comprehensive level. The lowest level of a fishery port represents the lowest values for the seawall, berth waters, and shoreline level, and the Dongsha fishery port was found to have a value of 12. The comprehensive level represents the weighted average of the seawall, berth waters, and shoreline level. The weighting factors of the seawall, berth waters, and shoreline are 0.25, 0.45, and 0.3, respectively. Hence, the calculated comprehensive level of the Dongsha fishery port is 12.

Discussion
The risk assessment of fishery ports against typhoons is a significant issue. Previously, some studies tried to give quantitative assessments using the analytic hierarchy process (AHP) and expert-scoring method. Although these studies considered various factors, their methods are somewhat subjective (Yang 2017;Dongshui and Qionglin 2019). The motivation of this study is to figure out a quantitative and objective resistance level of a fishery port against typhoons by using the methodology proposed in this manuscript. Then this quantitative resistance level can help the administrative department conduct rational and effective countermeasures before the arrival of typhoons, such as whether the evacuations are needed. In the numerical model verification part, to ensure accuracy, observed data from two wave gauge stations during five typhoons, which caused relatively significant b Fig. 13 Distribution maps of the waves under level 8-17 typhoons (from Google Earth) Fig. 14 Feature points of the fishing boat anchoring water Stochastic Environmental Research and Risk Assessment (2022) 36:1993-20132009 impacts on the project's vicinity, were selected for verification. It should be noted that there are biases in the observed data from gauge stations, which can sometimes be considerable, especially under severe weather and ocean conditions. Moreover, as the modeling of wind, current, and wave also contains deviations, the verification errors may be high under typhoon scenarios. In addition, the complex interactions among wind, rain, current, and wave make it challenging to separate and explain the errors. At present, there is no specific requirement for the accuracy of typhoon simulation verification. Therefore when the trends of simulated results are similar to the observed results and the average deviations are not bigger than 30%, the simulated results are acceptable. This loose error limit enables the model to work well under typhoon conditions. We proposed a typhoon prototype selection method for typhoon building, which is rarely systematically analyzed in other similar studies. Through this method, the south side toward the west scenario was selected as the final assessment typhoon pathway among 1841 historical typhoon tracks, which turned out the same conclusion with a previous study carried out at the Dongsha fishery port (Ji et al. 2017). Therefore the typhoon prototype selection method is practical.
The tide-surge interaction is essential for the modeling. The calculation results in this study indicated that the majority of the largest practical storm surges occurred around low tide, which was similar to the results in a study that specifically researched the tide-surge interaction (Idier et al. 2012). The agreement indicates that the tide-surge interaction calculation method is effective. At the same time, it proves that the two-hourly tide interactions were enough to reflect the tide-surge interactions in this study case, although the tide-surge interactions will be more accurate with higher time accuracy.  The assessment of the resistance level of a fishery port against typhoon damage can reveal weaknesses in the port's defenses and allow for the optimization of shelter spaces for fishing boats. The analysis carried out here had several caveats, which are vital to highlight when considering these results. Notably, the level 12 for the Dongsha fishery port does not indicate that boats should be evacuated when a level 12 level typhoon is coming. Instead, boats should consider taking shelter when a level 12 typhoon slams into the Dongsha fishery port at the radius of maximum winds. The feature points of T2, T3, T8, and T15 are the weaknesses of the Dongsha fishery port, which is consistent with the actual situation of the Dongsha fishery port, and the port could enhance its defenses by increasing the elevation at these weakness points.
There are some improvements to be made in this study. Considering the uniform standard, we treated the distance from the fishery port to the storm track roughly, the geometric mean radius of maximum wind. Other distances were not taken into account in the assessment. In this analysis, all other impacts (sea level rise, rain, stability of infrastructure) were disregarded; the proposed methodology does not assess the total conditions of the fishery port.
In addition to estimating the risk of fishery ports against typhoons, the weather forecasting and warning systems established in Wenzhou have proven to be efficient at preventing human and economic losses from typhoons. Further, evacuation plans and disaster response and preparedness solutions should be employed.

Conclusion
Some of the damage to fishery ports from typhoons may be preventable. This study described a systematic and quantitative method for estimating the risk of fishery ports against typhoons, and a case study was carried out on the Dongsha fishery port in Zhejiang Province, China. Historical typhoons were studied to identify the most useful typhoon pathways (south side toward the west) and scenarios (level 8-17 typhoons) for the assessment. Importantly, the Dongsha fishery port was found to have a resistance level of 12, and several points of weakness were identified where improvements in elevation could lessen impacts from future typhoons. In conclusion, the findings of this study demonstrated that this is a versatile framework for assessing fishing ports and developing disaster prevention plans. Though there remain a few constraints in its application (such as with regard to sea level rise, rain, and the stability of infrastructure), the proposed method should be readily applicable to other locations.