Ensemble Cloud Forecasting to Enable Free-Space Optical Communications

Modern society is increasingly reliant on high-performance communications. Ever-increasing amounts of data are being generated and collected by the general public as well as by scientists. As users continue to demand more data, the existing communications infrastructure will have to expand to meet the demands. There are limitations on traditional ways of transmitting data (RF frequencies and underground cable/fiber) that may inhibit these technologies from meeting future requirements. These technological, regulatory, and financial limitations may be alleviated, in part, by Free-Space Optical Communications (FSOC) systems. There are several key advantages to using FSOC to meet future communications requirements. Data can be transmitted through free-space via lasers at very high data rates of multi-Gb/s over long distances. Optical beams are very narrow, and are much less susceptible to interference or interception than RF signals. Unlike RF, the optical spectrum is unregulated. Finally, optical communications systems are relatively small and potentially much less expensive than comparable RF systems, particularly for space missions. However, FSOC needs an unobstructed path between the transmitter and receiver in order to work. As such, one of the major impediments to successful FSOC systems is the presence of clouds in the atmosphere. Given all the advantages of FSOC, NASA is conducting several space-to-ground FSOC demonstrations over the next few years, and is investigating how to optimize performance in the presence of clouds and other atmospheric effects. The missions will require short- and mid-range cloud forecasting in order to optimize mission scheduling and data collection, and to minimize unanticipated cloud outages. In support of this work, NWP ensemble forecasting is being leveraged to make cloud forecasts out to several days for the NASA optical ground stations. The total cloud amount from SREF and GENS ensemble members is being collected for several locations and used to provide realtime guidance to the NASA Lunar Laser Communication Demonstration (LLCD) during the Fall of 2013. Therefore it is important to understand the accuracy of the cloud forecasts. In this paper, the SREF and GENS cloud amount forecasts will be compared to observations in the form of cloud amounts derived from Geostationary satellite imagery. The accuracy of several months of ensemble cloud forecasts will be computed and shown as a function of forecast lead-time. Future near-Earth and deep-space missions will use the knowledge from LLCD and future demonstrations to make determinations such as the amount of data that can be collected and transmitted, the number and locations of ground stations, and on-board data storage required to make operational FSOC systems a reality. Good cloud forecasting will play an important role in these decisions.