Evaluation of optimization of lidar temperature analysis algorithms using simulated data

Abstract

Temperature lidar data have been simulated in order to test the Jet Propulsion Laboratory (JPL) SO3ANL version 3.2 and Service d'Aeronomie du Centre National de la Recherche Scientifique (CNRS/SA) TEMPER version 2.1 lidar temperature analysis software. Assuming known atmospheric temperature‐pressure‐density profiles, theoretical raw‐photons lidar profiles have been calculated using the actual characteristics of two JPL lidar instruments, located at the Table Mountain Facility (TMF) in California and the Mauna Loa Observatory (MLO), Hawaii, and the CNRS/SA Rayleigh lidar, located at the Observatoire de Haute‐Provence (OHP) in France. The simulations were performed for an initial climatological profile taken from the CIRA‐86 model and for various profiles derived from this model including realistic atmospheric disturbances. Comparisons between the original and retrieved temperature profiles revealed errors of several kelvins for both the JPL and the CNRS/SA programs. By varying parameters in the simulation, it was possible to determine both the source and the magnitude of these errors. Once identified, the errors were corrected, and the analysis programs were optimized, leading to new operational versions of these programs (SO3ANL version 3.6 and TEMPER version 2.2). An accurate accounting of the temperature lidar analysis errors, before and after this work, is presented.