Comparison among error calculations in differential absorption lidar measurements

Error determination is a critical issue in differential absorption lidar (DIAL). In this paper, some common error calculations are compared by applying them to a set of artificial data. First, three methods for the evaluation of the signal-to-noise ratio (SNR) of the lidar return have been examined. Then, the effect of the application to DIAL measurements of correction algorithms is investigated. The mean value and the standard deviation of the set of artificial data have been obtained from experimental determinations accomplished with our system. This was made possible by the shot-per-shot recording of the lidar return. The results of this study allows us to recommend a new method for the data analysis used in DIAL experiments: the SNR should be directly measured and all sources of statistical fluctuation have to be taken into account in the calculation of the final error. Otherwise such error could be underestimated, as suggested also by the disagreement of simultaneous measurements carried out by different systems.