The Second Wind Forecast Improvement Project (WFIP2): An Overview of Progress

Wind energy now accounts for more than 6% of electrical power produced in the U.S. and is continuing to increase. As the fraction of electricity provided by wind grows, it becomes increasingly important for power markets and for stability of the electrical grid to have accurate local and regional wind forecasts. There are several challenges associated with forecasting for wind power. One is that the forecasts are required for the hub height of wind turbines, typically 80 m above the surface on land. This is an altitude for which there are relatively sparse observations, and thus forecast models are not well validated at these heights. A second challenge is that wind power plants are commonly built in complex terrain, where many of the physical assumptions on which subgrid-scale parameterizations rest are violated. These challenges motivated the second Wind Forecast Improvement Project (WFIP2).

WFIP2 is a major research project supported by the U.S. Department of Energy (DOE) and by the National Atmospheric and Oceanographic Administration (NOAA) that has engaged a team led by Vaisala, Inc. as well as DOE and NOAA laboratories over the past four years, concluding in September, 2018. The regional focus for WFIP2 was the moderately complex terrain along the Columbia River in Washington and Oregon between the Cascade Mountains to the west and the Rocky Mountains to the east. The two primary objectives of the study were to increase understanding of the physical processes in complex terrain that determine hub-height winds and to incorporate consequent parameterization improvements into operational forecast models.

WFIP2 is an integrated modeling and observational study. Modeling studies centered on NOAA’s WRF-based High Resolution Rapid Refresh (HRRR) Model, which was nested down to 750 m for research purposes in the study. An extensive set of in situ and remote sensing instrumentation was deployed within and near the Columbia Basin. The instruments were analogously nested to provide detail regarding subgrid-scale physical processes within the inner model domains. The field phase of WFIP2 extended from 1 October 2015 through 31 March 2017 with the intention of covering a full annual cycle with continual observations. This approach allowed both an assessment of modeling improvements over the full range of atmospheric conditions and focused analysis of atmospheric phenomena, such as wintertime cold pool erosion, that pose particularly severe challenges for wind energy forecasting.

While other presentations will address atmospheric processes in more detail, this presentation will note significant successes of WFIP2, including new parameterizations for vertical mixing and scale-aware plume modeling that are now part of the NWS operational HRRR. During the field phase of WFIP2, participants maintained an event log to facilitate case studies of challenging events. This event log, all field observations, and key model runs (or initial and boundary conditions) constitute a rich data set for further research by the atmospheric sciences community. Data are publicly available through DOE’s Data Archive and Portal (https://a2e.energy.gov/data).