[H-CG23] Adaptation to climate change and its social implementation

[HCG23-P01] Assessing the Reliability of Reanalysis Products for Wind Energy Development: A Comparison with In-Situ Observations

*Muhammad Abid KHAN¹, KOJI DAIRAKU¹, Saurabh Kelkar¹ (1.University of Tsukuba)

Keywords:Wind, Surface observation, Reanalysis metadata, Renewable Energy

Reanalysis products play a critical role in a variety of fields for analyzing historical weather patterns, particularly in wind energy resource evaluations. Investigations of wind speed measurements reveal a decrease in land surface wind speed starting from the 1973s, referred to as global terrestrial stilling, which then experienced a reversal around 2004. The current state of knowledge is uncertain regarding the representation of the declining trend and turning point in reanalysis products. To fill this knowledge gap, a comprehensive analysis of the climatic wind patterns and trends in three different reanalysis products, namely ERA5, JRA-55, and NCEP/NCAR was carried out. This was done by comparing the 10-m wind speed time series at grid cells with observational data collected from 32 on-site meteorological stations over the period of 1973 to 2005. A set of statistical metrics were utilized to assess the accuracy of the reanalysis products, including standard deviation (STD), Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Bias (MB), correlation coefficient (R), and coefficient of determination (R²). Based on the evaluation of climatological winds, JRA55 was found to be the closest to the observations. However, significant differences were discovered between actual wind speeds and those simulated. It was noted that the JRA55 reanalysis product mirrored the changes in the Subcontinent observations and was followed by the ERA5 product. The results of JRA55 show exceptional agreement with wind speed observations, as evidenced by its high correlation coefficient of 0.933, a mean bias of 0.142, and a low Root Mean Square Error of 0.343. In contrast, NCEP/NCAR displays lower levels of standard deviation at 0.545. The wind energy sector in the Subcontinent is undergoing growth and expansion, with the potential for the development of both offshore wind farms and the expansion of existing land wind farms. This study aims to serve as a difference between actual and simulated land surface wind speeds highlighting the need for caution when utilizing reanalysis products to assess and forecast winds in the Subcontinent area.