Abstract

Optical turbulence models are compared to determine which offers the most utility for producing 4- to 6-h forecasts of C_n² over 1-km horizontal paths. The models tested include a statistical regression model (after Sadot and Kopeika, 1992), a direct estimate of C_n² from similarity theory (Miller and Ricklin, 1988), a bulk model (Andreas, 1988), and a more comprehensive physical model developed for use at White Sands Missile Range (Kunkel and Walters, 1988). One important measure of success in this comparison is the ability to forecast, with reasonable accuracy, the input parameters for the selected model. The statistical model and the Miller and Ricklin model use commonly measured meteorological parameters, which are also commonly forecast. The other models require more elaborate instrumentation and computation for parameters such as soil moisture and heat and moisture fluxes, which are more difficult to forecast. Therefore, these models may not be as convenient to use. In order to assess the relative utility of the models, the paper also includes a brief look at the sensitivity of the models to the input parameters.

© 1992 Optical Society of America