Analysis of Numerical Mesoscale Model Data for Wind Integration Studies in the United States

The Western Wind and Solar Integration Study (WWSIS) and the Eastern Wind Integration and Transmission Study (EWITS) are enhancing energy security by defining operating impacts due to large penetrations of renewable energy. The backbones of these studies are the large and consistent wind speed and power production data sets valid at 80 m and/or 100 m above ground derived from numerical mesoscale models for the years 2004-2006 and aggregated into wind power plants. The horizontal and temporal resolution of the data is 2 km and 10 minutes, respectively. The WWSIS data set was produced by 3TIER and the EWITS data set was produced by AWS Truewind under contract to the National Renewable Energy Laboratory (NREL). These data sets, which are available at http://www.nrel.gov/wind/integrationdatasets/, were designed for spatial and temporal comparison of sites for geographic diversity and load correlation and to provide estimates of power production from hypothetical wind plants. These data sets do not depict all possible wind plant sites nor should the data be used as the sole basis of project investment. NREL has performed a quality control check on the annual wind speed and power parameters and will conduct a detailed validation of the seasonal, diurnal, and geographic distribution patterns of the model data. The purposes of the analysis are to identify any anomalies in the data, to assess the regional accuracy of the data, and if warranted, to modify the data sets. One conclusion from the quality control exercise is that there are many details such as spatial and temporal discontinuities in the model output produced during post simulation processing that need to be examined in addition to the overall accuracy of the data. In this paper, we will present the results of the analysis of the mesoscale model data used for the Western and Eastern United States integration studies. We will discuss the validation of the data sets, including comparisons with validated wind maps and ground measurements and what types of conditions are challenging for the models. We will also discuss more broadly, what these data sets can or can’t tell us and appropriate uses and cautions for uses of the data.