Variations in local 3D cloud structure impact the interpretation of NASA ceres and modis data for top-of-atmosphere radiation studies from clouds. Currently, cloud radiative feedback mechanisms are…

This $1,102,083.01 cooperative agreement was awarded by the National Aeronautics and Space Administration (NASA) to the University of Colorado to study variations in 3D cloud structure and its impact on the interpretation of NASA satellite data for top-of-atmosphere radiation studies. The goal is to obtain ground-based data on 3D cloud structure and phase to improve climate model parameterizations of cloud radiative feedback mechanisms, which are a major source of uncertainty. The project will also examine the use of low-cost ground-based camera networks to provide high-resolution data for forecasting solar insolation and cloud cover to support renewable energy grid management. This award was not set aside and is part of NASA's broader cloud forecasting and 3D radiative transfer model validation program.

Variations in local 3D cloud structure impact the interpretation of NASA ceres and modis data for top-of-atmosphere radiation studies from clouds. Currently, cloud radiative feedback mechanisms are one of the largest sources of uncertainty in global climate models. Much of this uncertainty results from lack of knowledge of cloud particle phase and vertical and horizontal structure. However, the ceres spatial resolution of 20 KM does not permit radiation variations by most clouds to be resolved, and both ceres and modis are insensitive to the vertical structure and phase distribution in many clouds. Surface-based data on the 3-d cloud structure and phase for an ensemble of clouds is thus needed during satellite overpasses to be able to intercompare radiative transfer models based on modis and other satellite data with ceres data to improve the 3-d parameterization of GCM cloud models. Such data can be obtained using a multi-sensor array of ground-based visible-light cameras. Closely related, the forecasting of solar insolation and associated cloud cover on time scales out to ~1 hour ahead and with a spatial resolution of ~100 meters is recognized to be valuable for stabilizing power grids with high penetrations of solar photovoltaic energy. Data for cloud-advection based solar insolation forecasting can conceivably be obtained using geostationary sensors, but not with the requisite spatial resolution nor latency of seconds needed to adequately predict motion during high ramp rate events. Moreover, the bottom-up perspective view of clouds provides data that is more strongly correlated with cloud-induced fluctuations in solar power than data obtained from top-down. The development of advanced electric grid management practices for improved integration of renewable solar energy would benefit from a complementary means of observing clouds using a multi-sensor array of low-latency ground-based cameras. The above data needs for both 3D cloud radiation model validation and solar forecasting can be addressed using a network of lowcost upward-looking visible light CCD sky cameras positioned at ~2 KM spacing over an area of ~30-60 KM in size and synchronized to acquire and report imagery on 30 second intervals. Using modern electronic integration and fabrication methods such cameras can be manufactured in quantity and deployed at a marginal cost of ~$200-$400 each, and operated unattended for years using existing communications infrastructure to transmit imagery with extremely low (