getting to large-scale solar integration
figure 2a
figure 2b
Any effort to improve forecasts of solar resources will benefit from more observations of solar
irradiance and improved modeling techniques
based on these observations. (Left top) figure
2a shows solar irradiance traveling through the
atmosphere, being absorbed, scattered, reflected,
and some of it landing on the tilted, flat surface of
a PV panel. (Left bottom) figure 2b shows solar
irradiance traveling through the atmosphere,
being absorbed, scattered, reflected, and some
of it landing on a CSP trough.
Copyright © 2011 by the American Solar Energy Society Inc. All rights reserved.
Administration (NOAA), uses the Rapid Update
Cycle (RUC, ruc.noaa.gov) weather model.
RUC provides an 18-hour forecast, updated
hourly, at a 13-km horizontal grid resolution.
An improved weather model that gives more
accurate short-term forecasts, called the Rapid
Refresh ( rapidrefresh.noaa.gov), is scheduled to
become operational at the NWS later this year.
NOAA is working to develop a High-Resolution
Rapid Refresh ( ruc.noaa.gov/hrrr) model, with
a 3-km hourly updated nest inside of the Rapid
Refresh model. NOAA is developing the HRRR
for many sectors and uses, including forecasts
for aviation, agriculture, severe weather and fire
weather. The HRRR also shows promise for
providing forecasts that will help the renewable
energy community, including operators of solar
energy systems, because of its emphasis on high
geographic resolution and short-term forecasts
using near-real-time data.
Co Ur TESY oF To M SToFFEL, DoE/nr EL
However, NWP models are limited by the
“spin-up” problem that decreases the accuracy
of their predictions in the first several hours of
a forecast, limiting their utility for solar forecasting at the needed short time-scales. Though
researchers are overcoming the spin-up problem by assimilating timely observations, especially radar data on water vapor in the form of
haze, rain, mist, fog or snow, many in the solar
industry consider NWP models to have little
usefulness in hour-ahead and minute-ahead
solar forecasting.
Various efforts are under way to improve
short-term forecasts of solar radiation. Currently,
cloud observations are not assimilated into global weather models, but a recent study showed
that doing so would yield improvements in the
6-hour forecast (Pincus et al. 2011). Some non-global models that do assimilate cloud observations include the RUC and the Local Analysis
