Wind Speed Interannual Variability with Measured Data Validations and its Impact on Energy Yield in the Southwest Sea of Korea

doi:10.21203/rs.3.rs-4923514/v1

Download PDF

Article

Wind Speed Interannual Variability with Measured Data Validations and its Impact on Energy Yield in the Southwest Sea of Korea

https://doi.org/10.21203/rs.3.rs-4923514/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

The Republic of Korea (ROK) has set an ambitious goal of 40.7 GW of onshore and offshore wind farms by 2038, as outlined in the "11th Basic Plan for Electricity Supply and Demand" in May 2024. To achieve this target, both wind farm developers and the government are crafting policies for "Project Site Planning" and "Offshore Wind Farm Clusters." A major concern for stakeholders is the variability in annual wind resource driven by climate phenomena and climate change, which can lead to increased uncertainty in energy yield predictions. This study investigates the interannual wind speed variability (IAV) of wind speed in the Southwest Sea of Korea, a critical factor influencing energy yield predictions and the Levelized Cost of Energy (LCOE) for offshore wind farms. To achieve this, we used reanalysis datasets, validated against measurement data, to calculate a ROK-specific IAV. Our findings indicate a possibility of reduction in LCOE by accurately accounting for regional IAV, particularly in Jeonnam, where 14 GW of offshore wind capacity is projected. This research provides essential data for improving energy yield predictions, thereby enhancing financial confidence and supporting the development of expertise in ROK’s growing wind energy sector.

Physical sciences/Energy science and technology/Renewable energy/Wind energy

Physical sciences/Energy science and technology/Renewable energy

As climate change caused by indiscriminate greenhouse gas emissions became more severe, natural disasters such as droughts, storm surges, floods, and heat waves have become more frequent, and countries have had to actively participate in measures to reduce greenhouse gases. Accordingly, in countries and cities where energy consumption is concentrated, a lot of budget investment and policy collaboration efforts are being made for energy conservation and carbon emission reduction measures, such as Carbon-Zero and RE100.

In December 2017, Republic of Korea (ROK) set a goal to build 12 GW of offshore wind power facilities by 2030 under the “Renewable Energy 3020 Policy,” and raised the target to 16.8 GW in December 2020, in accordance with the “5th Basic Plan for the Development and Use of New and Renewable Energy Technology.” ^1–3. Subsequently, the target was adjusted to 14.3 GW by the “10th Basic Plan for Electricity Supply and Demand” established in January 2023, and the Ministry of Trade, Industry, and Energy recently announced that it will promote the deployment of renewable energy power generation facilities with an average annual capacity of 6 GW by 2030. However, as of the first half of 2024, the installed capacity of offshore wind power facilities currently in operation is less than 1% of the target of the 10th Basic Plan for Electricity Supply and Demand. GWEC estimates that the growth potential of the ROK offshore wind market will increase by 35% over the next five years, following a revision in 2023⁴ that updated the market capacity target to 2.3 GW compared to the projection in 2022⁵. This updated target is based on the acquisition of EBLs by 90 domestic offshore wind projects, representing approximately 27.7 GW of total rated capacity, most of which are still in the preliminary stages⁶. This suggests significant growth potential for the domestic offshore wind power market in the coming years. Expectations include improvements in industrial infrastructure, a reduction in the LCOE, and an increase in the skilled workforce and expertise within the sector ^7–10.

The uncertainty of offshore wind projects is evident from the initial stage of a project, which is partly due to the wind resource data used to determine the project feasibility and the Annual Energy Production (AEP) estimations. By following industry best practices and guidelines, the use of reliable wind resource data and AEP analysis methods can significantly mitigate this uncertainty. Meanwhile, several studies have recently shown that climate change can lead to localized variability in wind resources in the long term^11–13. This is due to mid-long-term changes in atmospheric stability, abnormal increase in sea surface temperature compared to the average year, changes in radiation and emission (flux), and changes in jet flow patterns, resulting in a trend of decreasing or increasing wind speeds in different regions ², ^13–18. This will cause an increase or decrease in the amount of wind power generation in the region, which in turn will determine the general and economic feasibility of wind power projects.

Wind speed (WS) and the resulting power generation of wind turbine generators (WTGs) varies between regions depending on the temporal and spatial scale. Short-term or mid-long-term prediction of wind speeds near WTG hub heights can have a key impact on site selection, grid management and business economics evaluation. Based on improved prediction accuracy through numerical prediction models and statistical post-processing, monthly and seasonal information has been increasingly used in wind farm design and planning. However, short-term, or mid-long-term predictions of wind speeds are significantly less predictable in regions with more complex terrain and more changes and interactions of meteorological factors¹⁹. In addition, due to intensifying climate change in recent years, the annual volatility of wind resources has become more prominent than the average year. Interannual variability (IAV) of wind speed refers to the changes in wind speed from one year to the next within a specific region. It represents the year-to-year fluctuations in wind speed, as indicated in Eq. 1 (Eqs. 1). In particular, the IAV is an indicator of variability in terms of the standard deviation of the average wind speed per year from the long-term average value, which helps to understand the pattern of fluctuations in wind speed by analysing how the average wind speed varies year on year.

$$\:IAV=\:{\sigma\:}_{annual\:wind\:speed}\:\left[\%\right]$$

In particular, IAVs can have a significant impact on onshore and offshore wind power generation, as wind speed variability affects the farm’s power output and operational efficiency. This is because the expected variability of wind power plays a key role in determining project financing. The impact of IAV on wind power generation is as follows:

Fluctuations in annual energy production: IAV leads to variations in the annual energy generation, introducing challenges for accurate yearly AEP assessment required for project financing, as well as servicing the debt throughout the financing period after construction.
Uncertainty in power generation forecasting and usage planning: In the regional context, the variability in wind resources introduces uncertainty in forecasting capacity and planning for power usage within the grid.
Increased operating and maintenance costs: The unpredictable nature of wind speeds due to IAV may lead to higher operating and maintenance expenses, as equipment may need more frequent adjustments and repairs to accommodate varying conditions.

Ultimately, IAV can introduce business risks for wind projects, especially when the full extent of the risk associated with IAV is not fully understood through short-term wind resource measurements from meteorological masts or lidars. This is equivalent to asking what errors might arise if a year of measurement data is obtained from a potential wind farm site, and assuming that this data represents the farm’s wind resources for the next 20–30 years of operation. To address this issue and enhance the understanding of the associated uncertainty, a detailed local analysis of wind resources is necessary. The results of this analysis should be incorporated into the uncertainty assessment to improve the accuracy of annual power generation predictions.

In this regard, Garrad Hassan, the predecessor of DNV, revealed that the IAV of UK land was about 6% based on the 1997 UK land measurement data, mesoscale modelling and reanalysis data ²⁰. The findings have been used as the standard IAV of the global wind industry for many years and have played a critical role in the planning and optimization of wind power projects. However, It may be overestimated by assuming annual mean wind speeds are Gaussian distributed with a standard deviation (σ) of 6% IAV²¹. More recently, DNV calculated IAVs for the UK waters in 2016, based on marine measurements (meteorological mast, LiDAR), reanalysis data, and mesoscale modelling data, presuming that Garrad Hassan’s 6% in IAV may be an overestimation for offshore. As a result, DNV found that the IAV in the UK waters has a range of 4 to 5.5% ²⁰. The IAV in UK offshore areas of around 4% has been shown in similar studies ^22,23. Furthermore, the study found that this improved IAV result could lead to a 0.3% reduction in offshore wind LCOE. However, IAV in the UK waters cannot be generalized to IAV in other countries or regions, because as indicated above, IAV can depend on topographic conditions, atmospheric stability, and changes in weather patterns.

Lee et al. ²⁴ stated that monthly and annual variability of wind speed is a key source of uncertainty in the overall wind resource assessment process, and emphasized the importance of using rigorous methods to estimate monthly and annual variability. And it was assessed the annual variability of wind resources over the entire continental United States and oceans under current and future climate conditions and predicted that it could vary by up to 10% (increasing or decreasing regionally) based on reanalysis data and regional climate model simulations ¹¹. Bastin et al. ²⁵ analysed the IAV over onshore and offshore in India using 30 years of ERA5 reanalysis data, and calculated it to be 3.5% and 3.8%, respectively. Pullinger et al. ²⁶ calculated the IAV over onshore in Ireland based on ground observations and reanalysis data, and found that the range of site-specific differences was 4.4–6.9%, but the average IAV of 5.4% was a robust estimate and could be used for energy yield assessment. Yu et al. ²⁷ calculated the long-term IAV of summer wind speed in the entire inland of China using various types of reanalysis data, and although it showed a considerable level of difference by region, it was said that the IAV of the reanalysis data tended to slightly underestimate the IAV compared to the observation data. In this way, IAV on onshore and offshore has been evaluated using various types of data for business evaluation and research of onshore and offshore wind farms around the world. Therefore, the need for advanced measurement and analysis technology to grasp detailed wind speed patterns in various regions is emphasized, and naturally, the same is true for ROK.

Although ROK is surrounded by the sea on three sides, the East Sea is characterised by deep waters, strong wind speeds, and a relatively uniform coastline. In contrast, the South Sea, and the West Sea, which feature shallow waters, experience relatively lower wind speeds, high turbulence intensity, and contain numerous islands, each with distinct wind resource characteristics. Therefore, this study calculates the long-term IAV of the Southwest Sea of Korea and evaluates the characteristics of offshore wind resource. The objective is to improve the economic feasibility of projects and increase the LCOE by assessing the potential to reduce uncertainty in future offshore wind projects in the area. The thesis consists of the following: First, it introduces the status of the Southwest Sea of Korea, and the meteorological data used that was analysed. Then, an IAV based on reanalysis data is calculated, which presents results separately for coastal and distant seas and calculates the differences in IAV for different periods of climate data used. Finally, instead of the 6% IAV that has been used as the industry standard, the P90/50 ratio shows how much the uncertainty of offshore wind projects can be quantitatively improved by using the IAV of this study.

Analysis area

This analysis focuses on the Southwest Sea, specifically in the regions of South Jeolla (Jeonnam) and North Jeolla (Jeonbuk) provinces in ROK. As of the first half of 2024, more than half of the 90 offshore wind projects that acquired an Energy Business License (EBL) are in this region, and it is still the area with the largest number of offshore measurements in progress or planned in ROK for offshore wind projects (6). Therefore, the southern area of Gunsan and the northern area of Jindo, the areas where the most attention of offshore wind project developers is concentrated, were selected as target areas of the analysis as shown in Fig. 1.

Reanalysis data

As part of the analysis, ERA5 and MERRA2 reanalysis data were obtained for each grid point near the Exclusive Economic Zone (EEZ) boundary off the coast of Jeonnam and Jeonbuk of ROK. ERA5 and MERRA2 are climate reanalysis data provided by the European Center for Medium-Range Weather Forecasts (ECMWF) and NASA’s Global Modelling and Assimilation Office (GMAO) and are datasets that reconstruct climate conditions through numerical prediction modelling based on global weather measurement data. MERRA2 has a horizontal resolution of 0.5 degrees x 0.625 degrees (about 50 km) by latitude and longitude, respectively, and ERA5 has a horizontal resolution by a grid of about 31 km. In this study, 1-hour average wind speed data for each grid were obtained from a total of 40 years of reanalysis data from 1983 to 2022 (Table 1, Fig. 2).

Offshore measurement data

To validate the spatial trend of IAV using actual measurements to verify the correlation with reanalysis data, measurement data were collected from offshore buoys and light beacons operated by the Korea Meteorological Administration (KMA) as well as two offshore meteorological masts provided by KEPRI (Korea Electric Power Research Institute) research institute and Mokpo National University (MNU). The location of the offshore buoys, light beacons, and meteorological masts can be seen in Fig. 2, and the measurement period, height, and detailed information of the measurement data are shown in Table 1.

Table 1

The detailed information of measurement wind datasets used in this study.
No	Name	Latitude [°]	Longitude [°]	Type	Period	Measurement Parameters	Source
1	Galmaeyeo	35.61	126.24	Light Beacon	2004–2022	Wind speed (WS), wind direction (WD), Temperature (T), relative humidity (RH), pressure (P) (15 m)	KMA
2	Haesuseo	34.26	126.03	Light Beacon	2004–2022	WS, WD, T, RH, P (12 m)	KMA
3	Buan	35.66	125.81	Buoy	2016–2022	WS, WD, T, RH, P (2 m)	KMA
4	Chilbaldo	34.79	125.78	Buoy	2003–2022	WS, WD, T, RH, P (2 m)	KMA
5	Shinan	34.73	126.24	Buoy	2014–2022	WS, WD, T, RH, P (2 m)	KMA
6	Offshore met mast 1	35.47	126.13	Met mast	2014–2016	WS (97, 95, 86, 76, 66, 56, 46, 26 m), WD (95, 76, 56, 46 m), T, RH, P (94, 13 m)	KEPRI
7	Offshore met mast 2	34.48	125.91	Met mast	2016–2018	WS (107, 100, 75, 35 m), WD (107, 96, 35 m), T, RH, P (98, 10 m)	MNU

Methodology

The sequence of analysis methods of this study is as follows:

Identify the spatial average wind speed distribution and IAV over the past 20 years (2003–2022) and 40 years (1983–2022) for each of the ERA5 and MERRA2 reanalysis datasets.

Compare with the entire analysis target area by analysing IAVs off the coast to analyse the difference in the characteristics of the spatial trend.

Analyse the quantitative and spatial distribution differences of IAVs throughout 20 and 40 years for each reanalysis data.

Analyse the correlation between offshore measurement data and reanalysis data and analyse IAV trends based on measurement data.

Simulate a hypothetical offshore wind farm and perform a sensitivity analysis of the P90/P50 ratio according to each IAV scenario.

Firstly, the average wind speed based on reanalysis data and the spatial distribution of IAVs within the analysis area were identified. This was calculated based on the data for the past 20 and 40 years, respectively. Due to the differences in the modelled calculation heights of the available data, MERRA2 reanalysis data extracted wind speed information at a height of 50 m, and ERA5 used wind speed data at 100 m height. The IAV was defined as the standard deviation of the annual average wind speed over the entire observed period, which was calculated for each node of the reanalysis data.

Second, the spatial distribution characteristics of IAVs were analysed by defining the sea area as up to 100 km off the coastline offshore, to compare with the entire analysis target area up to the EEZ. This is because most bottom-fixed offshore wind projects in the Southwest Sea of Korea are located within 100 km of the coastline, within a water depth of 60 m. Therefore, it was necessary to determine what differences the characteristics of offshore IAVs exhibit from the perspective of bottom-fixed offshore wind projects.

Third, the study attempted to quantitatively and spatially determine the distribution characteristics of the IAVs over the past 20 and 40 years for each reanalysis dataset. The aim of this analysis into the potential indirect impacts of recent climate change on the variability of annual wind speeds in the Southwest Sea of Korea.

Although the characteristics of IAVs in the Southwest Sea of Korea based on reanalysis data can be identified through steps 1 to 3, their validity is assessed through linear regression correlation analysis with the actual measurement data, given the selected measured datasets were not assimilated into the numerical prediction model of the reanalysis datasets. Therefore, the correlation of the reanalysis data against measurement data was analysed using the measurement data of two offshore meteorological masts provided by KEPRI and MNU along with the measurement data of offshore buoys and light beacons operated by the KMA. Particularly in the case of offshore buoys and light beacons, the measurement period was as long as over 10 years at certain locations. Therefore, the study aimed to analyse the characteristics of IAV distribution using measurement data to determine if the IAV trends are consistent with those observed in the reanalysis datasets.

Finally, an annual energy production (AEP) assessment is conducted by creating a hypothetical offshore wind farm within the analysis area. Based on several IAV scenarios that include IAV values calculated in the Southwest Sea of Korea, simulations on how the P90/P50 ratio might vary were performed. The P90/50 ratio in AEP is an important indicator of uncertainty in predicting the energy production of wind farms. In this context, P50 and P90 refer to the following:

P50: The probability of exceedance of 50% is the median expected annual AEP, meaning that there is a 50% probability that the annual energy production will be higher than this value for any year across the operational lifetime of the wind farm.
P90: The probability of exceedance of 90%, is a conservative estimate, meaning that there is a 90% probability that annual energy production will be higher than this value for any year across the operational lifetime of the wind farm.

The P90/ P50 ratio is an essential tool for evaluating the predictive reliability of wind power projects. This allows investors and financial institutions to better understand the risks and potential returns of the project, with a ratio closer to 1 indicating that the uncertainty associated with the energy yield prediction is lower. The ultimate objective of this study is to demonstrate how long-term IAV in the Southwest Sea of Korea, calculated from reanalysis data and validated by measurement data, can reduce the uncertainty in energy yield assessments for offshore wind projects, offering a more accurate assessment compared to generic assumptions like such as the historical 6% IAV.

Wind speed change trend

Figure 3 shows the changing trend in ERA5 and MERRA2 wind speeds based on long-term reanalysis data from random nodes located in the middle of the analysed offshore area. Examining the ERA5 long-term trend of wind speed changes in the Southwest Sea of Korea over the past 40 years reveals a slightly positive average slope, suggesting a marginal increase in average wind speed over time. Conversely, the MERRA2 reanalysis data indicates a weak negative slope, implying that wind speed at a height of 50 metres has decreased compared to the past. Meanwhile Fig. 4 shows the mean wind speeds across the more recent 20 years of data. It indicates the spatial variation in mean wind speeds across the observed region.

IAV spatial distribution

Based on long-term climate data from the past 20 and 40 years of ERA5 and MERRA2 reanalysis data, the spatial distribution of IAVs in the Southwest Sea of Korea was calculated (Fig. 5). Both data sources showed relatively higher IAVs in coastal areas than in the distant sea. However, for the ERA5 data, high IAVs were observed along the Jeonbuk coast in the northeast of the analysis area, while lower IAVs were observed along the southeast coast of Jeonnam, specifically in Sinan-gun. As shown in Fig. 5, this may be because the wind speed of ERA5 is higher in Jeonnam than in Jeonbuk, and in general, the turbulence intensity is higher when the wind speed is low. From the point of view that both turbulence intensity and IAV show a standard deviation from the mean, it can be explained that the variability is greater in Jeonbuk, where the wind speed is lower.

Since most of the bottom-fixed offshore wind projects in the analysed area are either underway or planned within 100 km of the coastline, the IAV of this offshore region was closely examined. The IAV near the coast was slightly higher compared to the IAV for the entire analysis area (Table 2).

Table 2

IAV differences between analysis periods in near coastal and offshore areas.
IAV	ERA5 [%]	MERRA2 [%]	ERA5 (near offshore) [%]	MERRA2 (near offshore) [%]
IAV (20 yrs) (2003–2022)	3.17	3.32	3.29	3.50
IAV (40 yrs) (1983–2022)	2.78	2.79	2.82	2.94

Differences in IAV spatial distribution by analysis period

The IAVs for the last 20 years were calculated to be much higher than for the last 40 years (Fig. 6). These characteristics were consistent for both MERRA2 and ERA5 datasets, suggesting the possibility that the meteorological or marine environment may have undergone long-term changes due to climate change. On the other hand, the results of the MERRA2-based analysis suggest that IAVs in Sinan-gun, southeast of the analysis area, have increased significantly over the past 20 years. Unlike ERA5, MERRA2 utilizes wind speed data based on a height of 50 m, and it is necessary to consider that it is a complex wind climate region with a considerable number of islands distributed compared to the northeast.

Verifications with offshore measurement data

Since the reanalysis data is produced by a numerical prediction model, variations from the actual measurement data occur. Therefore, to understand the reliability of the IAV calculated from the reanalysis data, it is essential to verify the correlation of the reanalysis data against the measurement data and observe similarities in patterns.

Measurement data that had years with less than 90% data recovery rate were excluded from the analysis, and the IAVs for each measurement site is shown in Fig. 7. The wind speed correlation between the offshore measurement data and the reanalysis data showed that the correlation (coefficient of determination, R²) of ERA5 against the measurement data ranged from 0.60 to 0.85 with the highest and very good correlation at Offshore met mast 1 (No 6) and the lowest correlation at Chilbaldo (No 4). The spatial pattern of IAV was also quantitatively similar to that obtained from ERA5. This shows that the agreement of ERA5 with the measurement data is significant even when taking into consideration that the offshore measurement data was measured at a height of about 10 m above sea level (Table 3), as opposed to the model height of 100 m data by ERA5. Meanwhile the correlation of MERRA2 against the measurement data ranged from 0.57 to 0.78, notably lower than that observed by ERA5. For this reason, higher confidence is associated with the ERA5 results.

On the other hand, in the case of the offshore meteorological masts, the measurement period of the two equipment was less than three years, so the resulting IAV figures are not considered representative. Nevertheless, the offshore met mast datasets have proven to be useful for validation as it was possible to confirm the correlation with ERA5 to be higher than 0.8, after checking the sensor measurement data at heights of about 100 m and 50 m.

Table 3

The correlation and IAV are based on reanalysis and offshore measurement datasets.
No	Name	Measurement height [m]	R² with reanalysis data		IAV			Available period
No	Name	Measurement height [m]	ERA5 [-]	MERRA2 [-]	Meas. [%]	ERA5 [%]	MERRA2 [%]	Available period
1	Galmaeyeo	15	0.714	0.690	5.58	3.88	3.54	13.4 yrs
2	Haesuseo	12	0.676	0.667	2.92	3.07	3.59	16.2 yrs
3	Buan	2	0.713	0.695	3.16	2.65	1.97	6.9 yrs
4	Chilbaldo	2	0.606	0.619	4.41	2.71	2.95	13.6 yrs
5	Shinan	2	0.625	0.574	7.90	2.43	2.09	8.8 yrs
6	Offshore met mast 1	95	0.829	0.773	2.15	3.73	3.56	2.8 yrs
7	Offshore met mast 2	100	0.811	0.753	1.18	2.99	3.55	2.7 yrs

P90/P50 ratio based on energy yield

A hypothetical offshore wind farm with a capacity of 510 MW was simulated within the analysis area to evaluate the annual energy production. The NREL 15 MW 240 RD WTG model was used, with a hub height of 150 m and 34 WTGs. The WTGs were spaced with 8 rotor diameters (RD) (1920 m) and 6 RD (1440 m) in the prevailing wind direction and the cross-wind direction, respectively. The wake losses were estimated with the Jensen wake model ^28,29, wake decay constant of 0.04. Secondary losses were applied to derive the P50 value. The uncertainties were then applied based on generic industry standards. However, the uncertainty associated with long-term climate representation was assessed using four IAV scenarios for sensitivity analysis (Table 4, Eqs. 2).

$\:Historical\:period\:uncertainty=\:\frac{IAV}{\sqrt{N}}\:\left[\%\right]$ $\:N:hindcast\:years$ (2)

Assuming the traditional 6% IAV as the base case, the P90/P50 ratio improved by 0.1%, 0.3%, and 0.4% when IAV of 5.5%, 4%, and 3% were applied, respectively. Given that the IAV in the Southwest Sea of Korea, as estimated in this study, averages approximately 3%, there is a potential for a substantial reduction in uncertainty, which warrants further investigation.

Table 4

The IAV sensitivity analysis summary result for calculating the P90/P50 ratio.
No	Scenario	Uncertainty of long-term representation [%]			Change in P90/P50 compared to IAV case A [%]
No	Scenario	1 yr	10 yrs	20 yrs	Change in P90/P50 compared to IAV case A [%]
A	IAV 6%	6.0	1.9	1.3	-
B	IAV 5.5%	5.5	1.7	1.2	+ 0.1
C	IAV 4.0%	4.0	1.3	0.9	+ 0.3
D	IAV 3.0%	3.0	0.9	0.7	+ 0.4

The IAV in the Southwest Sea of Korea was analysed using reanalysis data, which was verified with offshore measurement data. Based on these results, it was confirmed how much the uncertainty of offshore wind projects in the region could be improved, and the results can be summarized as follows:

IAVs based on MERRA2 and ERA5 reanalysis data were 3.32% and 3.17% in the Southwest Sea of Korea over the past 20 years, while IAVs in the 40-year period were 2.79% and 2.78%, respectively.

Relatively higher IAVs were observed in the coastal areas, with 3.50% (MERRA2) and 3.29% (ERA5) for 20-year IAVs and 2.94% (MERRA2) and 2.82% (ERA5) for 40-year IAVs.

Although the height of each reanalysis data must be taken into account, MERRA2-based IAV has higher values overall near the coast compared to ERA5.

The correlation analysis between reanalysis data and offshore measurement data indicates that ERA5 performs well, showing a higher correlation with the measurements compared to MERRA2. While the general spatial trend of IAV based on offshore measurement datasets is similar to that of ERA5, measurements from Chilbaldo and Shinan suggest that ERA5 may not fully capture the complexity of coastal regions due to its coarse resolution.

A site-specific IAV assumption in the Southwest of Korea might be more applicable rather than the historical 6% IAV assumption.

With the hypothetical offshore wind farm, the P90/P50 ratio improved by 0.4%, as a result of applying an IAV of 3% in this study, compared to the traditional 6% IAV or 4-5.5% in the UK offshore.

Reducing the uncertainty of wind power has positive economic, technological, and environmental benefits, which can play a vital role in increasing the success and sustainability of wind power projects. This study can be an important basis to confirm whether the trend of wind speed variations based on reanalysis data can be used for actual uncertainty analysis when long-term measurement data cannot be obtained at sea.

In addition, it can contribute to the improvement of the quality of basic research related to ROK’s offshore wind resources and climatological characteristics to reduce the uncertainty of ROK’s offshore wind projects and the developers’ project risks.

Competing interests

The authors declare no competing interests.

Author Contribution

All authors played significant roles in analysis and writing of this paper. G.R. conceptualized the topic and contributed to the writing, including methodological assumptions, analysis, and validation. H.K. conducted a preliminary literature review and validated the measurement data. D.L. was responsible for acquiring reanalysis data and assessing energy yield. A.P. contributed to the development of methodological assumptions and the analysis of results. H.-G.K. and C.G.K. provided advice on the comparison and validation methods for reanalysis and offshore measurement data. O.S. and J.-Y.K. contributed to the writing and provided overall supervision.

Acknowledgement

This work was supported by the Korea Institute of Energy Technology Evaluation and Planning (KETEP) and the Ministry of Trade, Industry & Energy (MOTIE) of the Republic of Korea (No. RS-2023-00301792).

Data Availability

The MERRA-2 reanalysis dataset is available at MDISC, managed by the NASA Goddard Earth Sciences (GES) Data and Information Services Center (DISC) and can be downloaded through https://disc.gsfc.nasa.gov/datasets?project=MERRA-2. The ERA5 reanalysis dataset is available at the Copernicus Climate Data Store at https://cds.climate.copernicus.eu/cdsapp#!/dataset/reanalysis-era5-single-levels?tab=overview. The offshore measurement data (buoys, light beacons) are publicly available by the Korea Meteorological Administration (KMA) and can be downloaded at https://data.kma.go.kr/data/sea/selectBuoyRltmList.do?pgmNo=52. The offshore meteorological mast measurement data is not publicly available since those are the property of the Mokpo National University and Korea Electric Power Research Institute but are permitted to be used for this paper by Professor Chae-Joo Moon, and KEPRI.

Choi, Y., Park, S., Choi, J., Lee, G. & Lee, M. Evaluating offshore wind power potential in the context of climate change and technological advancement: Insights from Republic of Korea. Renewable and Sustainable Energy Reviews 183, 113497 (2023).
Ryu, G. et al. Atmospheric Stability Effects on Offshore and Coastal Wind Resource Characteristics in South Korea for Developing Offshore Wind Farms. Energies 15, 1305 (2022).
Park, S., Yun, S.-J. & Cho, K. Energy justice: Lessons from offshore wind farm siting conflicts in South Korea. Energy Policy 185, 113972 (2024).
GWEC. Global Wind Report 2023 - Global Wind Energy Council. https://gwec.net/globalwindreport2023/.
GWEC. Global Wind Report 2022 - Global Wind Energy Council. https://gwec.net/wp-content/uploads/2022/03/GWEC-GLOBAL-WIND-REPORT-2022.pdf.
Ryu, G., Kim, H. & Lee, D. LCR Policy Trend Analysis and Implications for Global Offshore Wind Industry: Focusing on the Taiwanese LCR Policy. Journal of Machinery Industry 3, 65–103 (2024).
Johnston, B., Foley, A., Doran, J. & Littler, T. Levelised cost of energy, A challenge for offshore wind. Renewable Energy 160, 876–885 (2020).
Martinez, A. & Iglesias, G. Mapping of the levelised cost of energy for floating offshore wind in the European Atlantic. Renewable and Sustainable Energy Reviews 154, 111889 (2022).
J Barthelmie, R., C. Larsen, G. & C. Pryor, S. Modeling Annual Electricity Production and Levelized Cost of Energy from the US East Coast Offshore Wind Energy Lease Areas. Energies 16, 4550 (2023).
Martinez, A. & Iglesias, G. Site selection of floating offshore wind through the levelised cost of energy: A case study in Ireland. Energy Conversion and Management 266, 115802 (2022).
Haupt, S. E. et al. A Method to Assess the Wind and Solar Resource and to Quantify Interannual Variability over the United States under Current and Projected Future Climate. Journal of Applied Meteorology and Climatology 55, 345–363 (2016).
Sherman, P., Chen, X. & McElroy, M. Wind-generated Electricity in China: Decreasing Potential, Inter-annual Variability and Association with Changing Climate. Scientific Reports 7, 16294 (2017).
Ryu, G. et al. The influence of sea surface temperature for vertical extreme wind shear change and its relation to the atmospheric stability at coastal area. Wind and Structures 36, 201–213 (2023).
Radünz, W. C. et al. The variability of wind resources in complex terrain and its relationship with atmospheric stability. Energy Conversion and Management 222, 113249 (2020).
Gualtieri, G. & Secci, S. Comparing methods to calculate atmospheric stability-dependent wind speed profiles: A case study on coastal location. Renewable Energy 36, 2189–2204 (2011).
Zhan, L., Letizia, S. & Valerio lungo, G. LiDAR measurements for an onshore wind farm: Wake variability for different incoming wind speeds and atmospheric stability regimes - Zhan – 2020 - Wind Energy - Wiley Online Library. Wind Energy 23, 501–527 (2020).
Kim, D., Ryu, G., Moon, C. & Kim, B. Accuracy of a short-term wind power forecasting model based on deep learning using LiDAR-SCADA integration: A case study of the 400-MW Anholt offshore wind farm. Applied Energy 373, 123882 (2024).
Gerkema, T. & Duran-Matute, M. Interannual variability of mean sea level and its sensitivity to wind climate in an inter-tidal basin. Earth System Dynamics 8, 1223–1235 (2017).
Ryu, G. et al. Analysis of Vertical Wind Shear Effects on Offshore Wind Energy Prediction Accuracy Applying Rotor Equivalent Wind Speed and the Relationship with Atmospheric Stability. Applied Sciences 12, 6949 (2022).
STUDY ON UK OFFSHORE WIND VARIABILITY Study on UK Offshore Wind Variability The Crown Estate Commissioners. https://www.thecrownestate.co.uk/media/1772/uk-offshore-wind-variability.pdf (2016).
Pryor, S. C., Shepherd, T. J. & Barthelmie, R. J. Interannual variability of wind climates and wind turbine annual energy production. Wind Energy Science 3, 651–665 (2018).
Potisomporn, P. & R. Vogel, C. Spatial and temporal variability characteristics of offshore wind energy in the United Kingdom. Wind Energy 25, 537–552 (2022).
Omrani, H. et al. Spatial and temporal variability of wind energy resource and production over the North Western Mediterranean Sea: Sensitivity to air-sea interactions. Renewable Energy 101, 680–689 (2017).
Lee, J. C. Y., Fields, M. J. & Lundquist, J. K. Assessing variability of wind speed: comparison and validation of 27 methodologies. Wind Energy Science 3, 845–868 (2018).
Bastin, J., Katyal, R., Vinod Kumar, R. & Yuvasri Lakshmi, P. Inter Annual Variability of wind speed in India. International Journal of Ambient Energy 43, 5232–5246 (2022).
Pullinger, D., Zhang, M., Hill, N. & Crutchley, T. Improving uncertainty estimates: Inter-annual variability in Ireland. J. Phys.: Conf. Ser. 926, 012006 (2017).
Yu, J., Zhou, T. & Jiang, Z. Interannual variability of the summer wind energy over China: A comparison of multiple datasets. Wind Energy 23, 1726–1738 (2020).
Peña, A., Réthoré, P.-E. & van der Laan, M. P. On the application of the Jensen wake model using a turbulence-dependent wake decay coefficient: the Sexbierum case. Wind Energy 19, 763–776 (2016).
Stevens, R. J. A. M., Gayme, D. F. & Meneveau, C. Generalized coupled wake boundary layer model: applications and comparisons with field and LES data for two wind farms. Wind Energy 19, 2023–2040 (2016).

No competing interests reported.

Download PDF

Reviewers invited by journal
11 Sep, 2024
Editor assigned by journal
11 Sep, 2024
Editor invited by journal
26 Aug, 2024
Submission checks completed at journal
26 Aug, 2024
First submitted to journal
16 Aug, 2024

You are reading this latest preprint version

Wind Speed Interannual Variability with Measured Data Validations and its Impact on Energy Yield in the Southwest Sea of Korea

Status:

Version 1

Abstract

Figures

Introduction

Methods

Analysis area

Reanalysis data

Offshore measurement data

Methodology

Results

Wind speed change trend

IAV spatial distribution

Differences in IAV spatial distribution by analysis period

Verifications with offshore measurement data

P90/P50 ratio based on energy yield

Conclusions

Declarations

Competing interests

Author Contribution

Acknowledgement

Data Availability

References

Additional Declarations

Status:

Version 1