Solar and Wind Resource Assessments from Two Years of Short-Range High-Resolution Rapid Refresh Forecasts

The High-Resolution Rapid Refresh (HRRR) model is being run hourly in real-time at the Global Systems Division (GSD) of the Earth System Research Laboratory (ESRL). The model is run out to fifteen forecast hours over a domain covering the entire conterminous United States (CONUS) at a spatial resolution of three kilometers, allowing the use of explicit convection. Initial and boundary conditions are obtained from the operational Rapid Refresh (RAP), and three-dimensional variational data assimilation including radar reflectivity observations was implemented in April 2013 within the 3-km HRRR.

The three kilometer scale and hourly updating of the HRRR lends itself to forecasts of small-scale weather variability. Renewable energy forecasting is one of many meteorological applications in which accurate depiction of small-scale spatial and temporal variability of sensible weather elements is critical. Low-level winds and incoming solar radiation are both highly dependent on small-scale orographic and coastal effects, especially in regions of complex terrain, and the HRRR can provide high-resolution forecasts in these regions. The high resolution also permits an accurate depiction of convective-scale structures and their local impacts on cloud cover and wind speeds.

In this study, we analyze long-term averages of HRRR two-hour forecasts of renewable energy fields such as 80-m wind speed and shortwave radiation at the surface in order to estimate renewable energy resource potential in different regions of the country. The short-term (two hour) forecasts allow sufficient time for model spin-up at the smallest resolvable scales while minimizing forecast error inherent in longer forecast lengths.

We will present the methodology and some preliminary results including various measures of the model low-level wind field averages and variability during 2012-13, with a focus on thresholding winds speeds to identify regions of high potential for wind energy development. We will also describe the diurnal cycle of winds, both within the central plains wind corridor and in some regions under consideration for offshore wind farm development. A similar analysis of downwelling solar radiation around the CONUS will be presented. The wind and solar resource availability in each season will be compared between the two years in the context of the relatively dry 2012 and wet 2013 over much of the CONUS.