Refraction measurements and modeling over the Chesapeake Bay during the NATO (TG-51) SAPPHIRE trials, June 2006

Optical refraction tends to occur frequently in the atmospheric boundary layer. Due to a gradient in the temperature as function of height, rays are bending down towards the earth (super-refraction), or up towards to the sky (sub-refraction). As a consequence, images of targets at long range may be distorted and mirages may occur, while the maximum detection range may be affected. In addition the irradiance, received from a point target at a sensor pupil, may increase or decrease due to atmospheric focusing effects. Sub-refraction tends to occur more frequently over sea-paths, as the air is normally cooler than the water. Currently available propagation models, like IRBLEM [1] and EOSTAR [2] adequately describe the mentioned effects for this condition. However sparse data are available for model validation in super-refractive conditions, when the air is warmer than the water. For this reason, the SAPPHIRE trial, (Ship and Atmospheric Propagation Phenomena Infrared Experiment), was organized in the Chesapeake Bay near Washington DC by NATO Task Group TG51 in June 2006. At this location and in this time of the year, the probability for having positive ASTD (Air to Sea Temperature Difference) conditions is high. In the bay a buoy was positioned, on which a set of precision temperature sensors was mounted. They provided as well ASTD values as temperature gradients at a height of 3.7 m. By means of a geodetic theodolite, absolute AOA's (Angle Of Arrival) were measured for a set of lights at various altitudes, located on the other side of the bay at a distance of 16.2 km. The buoy was located at about mid-path position. Positive ASTD conditions did occur on a number of days allowing validation of the values of the AOA, predicted by the models, based upon the meteorological data, simultaneously collected at the buoy. It was found, that the temperature profile, generated by the bulk model for the marine surface layer [3] is incorrect, resulting in deviations in the AOA, compared to the data from the theodolite. A set of empirical temperature profiles has been created, using the ASTD and the gradient in the temperature as function of height as input. One of them provides of predicted AOA values, which fit well with the measurements during the trials period. Errors in the prediction are resulting from the variability of the meteorological conditions, causing inhomogeneities along the path. Furthermore the measurement accuracy was limited due to image blur by atmospheric turbulence in a number of occasions