Abstract In the fundamental stage of resource assessment, high-quality wind speed measurements are required to estimate power production. However, this high-quality data is not always available, and therefore the analysis of alternative sources becomes essential. In this work, we analyze the ability of MERRA-2 to represent wind speed characteristics at 24 anemometric stations in Mexico. The assessment was carried out using the Pearson correlation coefficient between the observed time series, and the obtained by interpolating bias-corrected reanalysis-estimated wind speed to all locations for different time-averaging periods. Results showed that the reanalysis' performance is not uniform throughout the country; it depends on the time resolution, local orographic conditions, and the relationship between the local flow and the large-scale circulation. Based on these results, the country was subdivided into eight regions. The best-represented region was the Chivela Pass, where the winds are tightly linked to the interaction between the large-scale circulation and the local orography. The worst performing regions were located where the land sea-mask and orography at the reanalysis' resolution may not be accurate enough to reproduce the station's wind speeds. Reanalysis-estimated capacity factors exhibit large interannual variability in some stations, which can have significant consequences for the operation of individual wind farms and the power grid. The results show that, while caution should be exercised when applying reanalyses to wind resource assessment in Mexico, reanalysis wind power estimates can be a valuable tool to investigate the feasibility and installed capacity requirements for Mexico to meet its renewable energy targets.

中文翻译：

---

使用再分析数据估算墨西哥的风速和容量因子

摘要 在资源评估的基础阶段，需要高质量的风速测量来估算发电量。然而，这种高质量的数据并不总是可用的，因此对替代来源的分析变得必不可少。在这项工作中，我们分析了 MERRA-2 在墨西哥 24 个风速测量站表示风速特征的能力。评估是使用观察到的时间序列之间的 Pearson 相关系数进行的，并且通过将偏差校正的再分析估计风速插入到不同时间平均周期的所有位置而获得。结果表明，再分析在全国各地的表现并不统一；这取决于时间分辨率，当地地形条件，以及局部流与大环流的关系。根据这些结果，该国被细分为八个地区。表现最好的地区是奇维拉山口，那里的风与大规模环流和当地地形之间的相互作用密切相关。表现最差的区域位于再分析分辨率下的陆地海面掩膜和地形可能不够准确，无法再现台站的风速。再分析估计的容量因子在某些站点中表现出较大的年际变化，这可能对单个风电场和电网的运行产生重大影响。结果表明，虽然在将再分析应用于墨西哥风资源评估时应谨慎行事，