The effects of offshore wind farms upon the atmosphere over Taiwan Strait during an Extreme Weather Typhoon Event

Model validation

To verify the WRF model’s performance, we compared the simulation track and MSLP of Tropical Storm Haitang to observations. The Japan Meteorological Agency (JMA), typhoon archive, provided the cyclone best track data (and central minimum sea-level pressure (MSLP).http://www.jma.go.jp/jma/jma-eng/jma-center/rsmc-hp-pub-eg/besttrack.html). The hourly precipitation was measured by Central Weather Bureau, Taiwan (CWB), from 522 rain gauge station. Figure1a,b shows the track and MSLP comparisons for July 3031, 2017. The simulated track was similar to the observed track (Fig.1a), but the intensity was underestimated during study period (Fig.1b). The simulated track was comparable to the observed track during a period of typhoons over Taiwan. The bias of average track error was lower than 24km on July 30, 2017, at 06:0018:00 UTC. Although the trend was slightly different from the observed, the simulated central MSLP in Figure.1b was slightly underestimated. After the tropical storm in west Taiwan, the bias was generally less than 5hPa. The observed accumulated precipitation (06:0018:00 UTC 30 July) was 30200mm in southwest Taiwan. It could have been more than 300mm over the mountain ranges of southern Taiwan (Fig.1c). The simulation reproduced and accurately captured the spatial distributions of precipitation over Taiwan (Fig.1c). Between the observed and simulated data, there was strong spatial correlation with a correlation coefficient of as high as 0.73.

Effects of OWFs during the study period

While our simulation was focused on the Tropical Strom Haitang on 29 July, another typhoon Nesat came from the northeastern Pacific on 29 July. This means that Typhoon Nest occurred prior to Typhoon Haitang’s approaching Taiwan (Fig.2), and Taiwan was affected by two tropical storms on 29 July (Fig.2). A Fujiwhara effect was actually observed. These two tropical storms were very close to each other, rotating cyclonically around one center. Their tracks eventually became intertwined. At 11:00 UTC on Tuesday, July 29, 2017, Typhoon Nesat hit the northeastern coast Taiwan (Fig. At the same time, Tropical Storm Haitang was moving toward southwestern Taiwan. It was located in South China Sea. Its intensity dramatically decreased as Typhoon Nesat struck China (at 00:00 UTC 30 Jul in Fig.2). On the other side, it caused a southwesterly stream over the Taiwan Strait (00:0003.00 UTC 30 juillet in Fig.3). These two cyclonic systems interact to create a coupled low-pressure and counter-clockwise rotation area around Taiwan (Figs.2, 3,). The location of the typhoon, as well the direction of the ambient winds, was a key factor in determining whether the OWFs affected the atmosphere over western Taiwan. In Taiwan, the average height of the Central Mountain Range in Taiwan (Fig.1a), is more than 2 000m.²⁸. The mountains can block and uplift rainfall systems, depending on where the cyclone is located and the direction that rain bands movement is moving. The strength of the effect was therefore dependent on the track variation and interaction of wind field and complex geographic structure. The blocking effect of CMR was a factor in the wind speed being weak over western Taiwan, according to Typhoon Nest’s track on 29 July (Fig. 2). This study focused on the passage and impact of Tropical Storm Haitang on Taiwan between 3031 and July 2017.

Sea level pressure (shaded hPa) & 10-m horizontal wind. (vector ms).¹) in the WRF model domain 1 (resolution is 6km) during 2931 July, 2017. The NCAR Command Language (NCL version 6.6.2) was used to create maps and plots.³⁷.

Hub-height (110m), horizontal Wind Speed (shaded.ms¹) and horizontal wind (vector, ms¹) in the WRF model domain 2 (resolution is 2km) during 00:0021:00 UTC on July 30, 2017. The contour lines are black and indicate Taiwan’s elevation (m). The contour interval ranges from 500 to 3500m and is 1000m. The NCAR Command Language (NCL Version 6.6.2) was used to create maps and plots.³⁷.

We conducted simulations with and without the addition of OWFs to examine the impact of OWFs on the atmosphere during extreme weather events. Data showed that Tropical Storm Haitang struck Pingtung County in southern Taiwan on July 30, 2017, at 08:40 UTC. Tropical Storm Haitang moved northwards after 09:00 UTC. Its low-pressure center was in western Taiwan. However, strong cyclonic winds and typhoons caused a weak-wind area (less than 10ms) in the eastern portion of the island.¹) on the lee side of the mountains over western Taiwan. Two different directions of flow were observed near the OWFs. They were northeasterly and northwesterly, respectively, between 09:0012:00 UTC and July 30 (Fig.3). As Tropical Storm Haitang was already moving to the north of OWFs, the flow over Taiwan Strait was replaced between 15:00 and 21:00 UTC. The wake effects^3,29,30,31This could be due to a decrease in wind speed downwind of the OWFs on Jul 30. Siedersleben et al.⁷The WRF model and aircraft measurements were used to study all planned and existing wind farms in the North Sea. They reported that the effects on temperature and water vapour could propagate more 100km downwind when there is strong stable atmospheric conditions. The distances of the proposed OWFs to Taiwan’s west coast are approximately 4070 km. Wind turbine wakes are expected to have an effect on the atmosphere in western Taiwan. The differences in wind field near OWFs was examined to determine the impact of wind farms on other meteorological parameters, such as hub-height wind speed, 10-m divergence and atmospheric stability over southern Taiwan.

Figure4a shows the 12-h accumulation of precipitation for case CTRL over Taiwan. Precipitation occurs primarily in southern Taiwan. (See the supplementary in Figure. S1). S1. The results of simulations showed that the average reduction of 12-h accumulated precipitation was about 1% (Table 1). This is due to the values for both positive and negative compensatory. The maximum reduction was as high as 60mm in the mountain ranges, and 47mm in Taiwan’s central plain. The atmospheric stability could also have a strong impact on the vertical or lateral spread of the wake. Magnusson & Smedman⁹The wake effect of atmospheric stability in terms of atmospheric stability determined the strength and extent of the velocity deficit. Port-Agel and Abkar³²The results indicated that the recovery from the wake was slower when stable conditions were less turbulent because there was less turbulent mixing. To estimate atmospheric stability across rotor disk, we used data from the two model levels 1 (17m and 6 (198m).