1. Introduction
Typhoon-force winds, a significant meteorological threat in the mid-northern region of the Taiwan Strait, have substantial implications for maritime safety and port operations in the area [
1,
2]. Accurate wind forecasts are essential for ensuring the safety of navigation and the efficiency of port operations [
3,
4]. Globally, an average of 80 tropical cyclones form annually, frequently leading to strong winds, heavy rainfall, and storm surges [
5,
6,
7]. The northwest Pacific Ocean is the most active region for tropical cyclones, with about 33% of typhoons being formed there each year [
8,
9,
10]. Statistics show that, on average, 6.9 typhoons make landfall along China’s coastline annually, constituting about 25.5% of the total number of generated typhoons [
11,
12]. Severe typhoon disasters frequently occur during the peak summer to early autumn period, predominantly impacting the southeastern coastal provinces, with a rising impact [
13]. Among the provinces and regions, Guangdong, Taiwan, Hainan, Fujian, and Zhejiang experience the highest frequency of landings [
14]. In Fujian, typhoons can be generally categorized into three types: direct landfall, passage through Taiwan into Fujian, and passage through Guangdong into Fujian, with the “passage through Taiwan into Fujian” being the most common [
15]. The central mountain range of Taiwan serves as a natural barrier, disrupting the vertical structure of typhoons of the “passage through Taiwan into Fujian” type. The high-level circulation persistently flows over the mountains, causing an increasing disparity between the mid-level and low-level circulations. Consequently, the low-level pressure field becomes filled, leading to a substantial reduction in the typhoon’s intensity. Upon making landfall in Taiwan, irrespective of its initial strength, a typhoon tends to diminish to a tropical storm within approximately 24 h [
16].
Deviation in typhoon forecasts may result in decreased shipping efficiency and energy wastage [
17,
18,
19,
20,
21]. The handover process, including pilot boarding and berthing, is significantly influenced by maritime meteorological conditions, often resulting in port entry failure or personal injury accidents [
22]. The entry and berthing of large cargo ships are complex and intricate processes. Any failure in either process can lead to overall failure. Therefore, pilots are held to extremely high time management standards [
23]. Ship pilots experience significant impact from powerful winds, particularly in the context of typhoons [
24]. The best approach is to avoid unfavorable sea conditions, necessitating sophisticated meteorological services for the shipping channel [
25]. In the event of a typhoon or strong winds, the ladder-climbing pilot experiences violent shaking, which poses a significant safety risk. Safety accidents occur annually during boarding operations, and if safety conditions are not met, the operation must be aborted. If the cargo ship is unable to enter the port due to adverse weather conditions, the resulting delay could vary from one to several days. Incoming goods mainly consist of high-value bulk commodities. Any delays would result in substantial economic losses, significantly hindering sustainable development.
Fujian, situated on the southeastern coast of China and adjacent to Taiwan across the sea, serves as the main landfall area for the majority of typhoons heading northwest. Fujian province experiences an average of around 1.6 typhoons annually, although in exceptional years such as 1990 and 2010, this number has escalated to five. These typhoons, known as “Taiwan entry to Fujian” typhoons, usually lead to a decrease in their intensity. Satellite remote sensing studies have revealed that strong winds precede typhoons in the Taiwan Strait [
26], and downwind flows develop at the northern and southern ends of the Central Mountain Range, creating what is commonly referred to as “corner flows”. Yang et al. found through numerical simulations that a typhoon passing over Taiwan may cause its eyewall to disintegrate, which could then reconstitute under the influence of convective heating towers [
27]. Taiwan’s topography influences the northeast-southwest asymmetry of the typhoon, resulting in a change in its trajectory [
28]. Weaker and slower-moving typhoons demonstrate more significant deviations in their trajectory [
29]. Some studies use physical constraints, surface meteorological variables, and preliminary estimates as predictors in a multivariate linear regression model to compute wind products. The findings show that the errors in wind speed magnitude and wind direction are both less than 1.5 m/s and 30 degrees, respectively [
30]. Some research has emphasized the importance of large-scale variables as vital predictors for forecasting near-surface wind fields [
31], while other research has examined the attributes of wind patterns affected by topography [
32]. Previous studies have predominantly relied on basic statistical analysis and a single typhoon case, failing to conduct a comprehensive examination of wind field characteristics in the central and northern regions of the Taiwan Strait during typhoon events.
The central and northern regions of the Taiwan Strait are situated between the southeastern coast of the Chinese Mainland and the island of Taiwan [
33]. This maritime passage is of significance for connecting the eastern coast of China with the South China Sea islands. The irregular coastline has resulted in the formation of numerous bays and ports [
34]. Meizhou Bay Port, situated on the western shores of the Taiwan Strait in Fujian Province, marks the confluence of Putian and Quanzhou cities. It serves as a natural deep-water harbor with multiple berths, a rarity in China and uncommon globally. Pingtan is strategically positioned in the prominent southeastern coastal area of the mainland, and it acts as a crucial intersection for north-south routes in the western Pacific. Historically, it has functioned as a gateway to the East China Sea, South China Sea, and the Northwest Pacific Ocean, providing access to the Indian Ocean. It is widely recognized as the “Maritime Corridor” and plays a crucial role in the “Maritime Silk Road”. The location is commonly considered one of the “Three Eyes”, alongside Nanao in Guangdong Province and Penghu in Taiwan [
35]. The establishment of a 10,000-ton passenger and cargo roll-on/roll-off terminal has positioned it as the primary conduit for maritime interactions and communication between the two sides of the Taiwan Strait [
36]. Ports in the central and northern regions of the Taiwan Strait are significantly influenced by the near-surface wind patterns, particularly during typhoon occurrences. High winds directly affect the operability of port terminals, subsequently impacting cargo turnover on ships [
37]. For this reason, there is a pressing demand for accurate ground wind field forecasts in this area to reduce transportation costs and ensure vessel safety. Hence, conducting a detailed investigation of the characteristics and corrective measures associated with the surface wind field at port terminals in the central and northern sectors of the Taiwan Strait during typhoon events is of significant practical importance and societal value for cargo transport operations at these terminals.
This study focuses on typhoon events and analyzes wind field characteristics in the middle-northern coastal areas of the Taiwan Strait under the influence of typhoons with varying paths. The researchers used polynomial fitting to correct and evaluate the ECMWF wind field data. This study aims to analyze the influence of typhoon-generated winds on port terminals in the central and northern regions of the Taiwan Strait, using data from ground automatic station observations and numerical model forecast products along different trajectories. In addition, this study delves into the attributes of numerical model forecast deviations and their subsequent correction. The objective of this study is to provide precise forecasts of typhoon winds at port terminals in the central and northern regions of the Taiwan Strait during typhoon events. This will offer a scientific basis for making decisions regarding the suspension of port terminal activities affected by typhoon winds.
3. Introduction and Tracks of Selected Typhoons
Typhoon Mekkhala (202006) was designated on August 10, 2020 at approximately 11:00 AM in the northeastern area of the South China Sea (see
Figure 2). Afterwards, it proceeded northward, consistently gaining in intensity. At approximately 7:30 AM on August 11th, it reached the coast of Zhangpu County in Fujian Province, China, at nearly its maximum intensity. At landfall, the maximum near-center wind speed peaked at 33 m/s, corresponding to a grade 12 storm [
44]. Following its landfall, Mekkhala proceeded to drift northward, progressively losing strength. Ultimately, the China Meteorological Administration stopped monitoring Typhoon Mekkhala at 5:00 PM on August 11.
Typhoon Maria (201808) formed in the eastern part of the Pacific Ocean, east of Guam, on July 4, 2018 at 20:00. At 21:00, the Japan Meteorological Agency upgraded the weather disturbance to a tropical storm, naming it Maria. This was followed by a similar upgrade by the China Meteorological Administration (see
Figure 2). The Joint Typhoon Warning Center upgraded it to a tropical storm at 23:00. At 5:00 on July 8th, Typhoon Maria intensified into a super typhoon in the northwest Pacific Ocean, approximately 1930 km southeast of Yilan County, Taiwan. On July 9 at 10:00, the China Meteorological Administration issued a yellow warning for the typhoon, prompting the National Meteorological Center to initiate a Level III emergency response for significant meteorological disasters caused by the typhoon. On July 10th, at 10:00, the China Meteorological Administration issued a red typhoon warning. Typhoon Maria struck the Huangqi Peninsula in Lian-jiang, Fujian Province at 09:10 on July 11th. Upon landfall, the peak wind speed near the center measured at 42 m/s (grade 14), accompanied by a minimum central pressure of 960 hPa [
45]. The China Meteorological Administration officially terminated the tracking of Typhoon Maria on the evening of July 11th.
The precursor disturbance of Typhoon Nesat (201709) originated on the sea surface near Palau on July 21st 2017. It progressed into a tropical depression on July 25th, and subsequently intensified into a tropical storm on July 26th, being named Nesat. Subsequently, Nesat strengthened, transforming into a severe tropical storm on July 27th and escalating into a typhoon on July 28th. It reached its peak intensity as a Category 3 typhoon, with wind speeds of 40 m/s and a central pressure of 960 hPa, before making landfall in the eastern coastal area of Yilan County, Taiwan, around 19:40 on July 29th. Nesat briefly entered Taiwan and then exited near Hsinchu City, moving into the Taiwan Strait, as shown in
Figure 2. On July 30th at approximately 6:00, the typhoon made landfall once again in the coastal region of Fuqing City, located in Fujian Province. The typhoon’s intensity was at a lower limit, categorized as a Category 2 with wind speeds of 33 m/s and air pressure of 975 hPa. After making landfall, Nesat started to weaken and eventually weakened into a tropical depression by 14:00 on the same day, with the Central Meteorological Observatory ceasing its numbering of the storm at 20:00, and its remnant circulation eventually merged and moved northward with Typhoon Haitang, which also made landfall shortly afterward on the same day.
Haitang (201710) originated in the central South China Sea on July 26th 2017. It developed into a tropical depression on July 27th and further intensified into a tropical storm on July 28th. On July 31st at 2:50 a.m., it made landfall once more in the coastal region of Fuqing City, Fujian Province [
46]. By 23:00, it had weakened to a tropical depression. The next day at 08:00, the Central Meteorological Observatory stopped numbering for Haitang (see
Figure 2). At 17:30, Typhoon Haitang made landfall on the coast of Pingtung County, Taiwan. By 22:35, the typhoon had moved into the Taiwan Strait from Taichung City. On July 31st, 02:50, Typhoon Haitang made landfall in the coastal area of Fuqing City, Fujian Province, marking a historic occurrence of two typhoons hitting the same location within 24 h. This happened in conjunction with Typhoon Nesat, the ninth typhoon of that year.
7. Conclusions
Coastal islands near Putian experience higher maximum wind speeds during typhoon events compared to those near Pingtan. This is due to the topographic influence of the strait, leading to higher wind speeds in the central Fujian region. When typhoons impact the islands in the middle to northern parts of the Taiwan Strait, the initial wind direction is northeast, followed by the convergence of two airflows, resulting in a southerly wind pattern. The wind strength and direction at each station are determined by the typhoon’s path. Pressure difference can be a reliable indicator of the magnitude and timing of typhoon winds.
After applying polynomial fitting corrections, the u component experiences a reduction in forecast deviation to 1.33, and the v component to 1.87. This represents a 37% reduction in forecast deviation for the u component and 40% for the v component. Due to the model’s strong performance in forecasting the u component, its forecast score increases from 41.60 to 63.17, marking a 51.85% increase. Similarly, the v component’s score rises from 29.33 to 48.89, reflecting a 66.70% increase. Thus, polynomial fitting is demonstrated to be an effective corrective measure for the 10 m wind field products of the EC model.
The applicability and correction of ECMWF wind field data is beneficial for accurately forecasting the wind field in the central-northern part of the Taiwan Strait. During the revision of the ground-level 10 m wind field, the study considered the physical significance of the relationship between typhoon paths and Taiwan’s position, which is significant for statistical research methods considering physical significance. This methodology is more refined and precise in comparison to purely statistical corrections, lacking consideration of the physical implications. Targeted automatic observation facilities can be deployed to produce meteorological services for safety in ports, in order to ensure the sustainable development of port shipping.
This study can provide references for management authorities to appropriately determine navigation restrictions, maritime operations to efficiently unload cargo in ports, and make well-informed decisions. Furthermore, improving the accuracy of typhoon wind speed forecasts will help facilitate sustainable shipping, avoiding energy wastage and contributing to achieving carbon peaking and carbon neutrality.