Numerical simulations with photochemical transport models were independently performed for two domains situated in the Iberian Peninsula covering the Lisbon and Barcelona airsheds. Although the days chosen for simulation of the two cities are not the same, the synoptic situations in both cases, known as typical summertime situations, were similar, which allowed the development of typical mesoscale circulations, such as sea breezes and mountain and valley winds dominated by the Azores anticyclone. Emission inventories for the two areas were developed. The O₃ concentrations recorded in both cities have a similar level. Nevertheless, O_x values in Barcelona are higher than in Lisbon, which may, at a first glance, indicate an apparently more oxidant atmosphere in Barcelona. Photochemical modeling for the two cities has shown that the behavior of the circulatory patterns in both urban areas is rather different, which mainly has to do with the different strengths of the sea breeze and the topography, inducing an important offshore vertical layered dimension of pollutant transport in Barcelona versus an important inland horizontal transport in Lisbon.

A first inversion of the backscatter profile and extinction-to-backscatter ratio from pulsed elastic-backscatter lidar returns is treated by means of an extended Kalman filter (EKF). The EKF approach enables one to overcome the intrinsic limitations of standard straightforward nonmemory procedures such as the slope method, exponential curve fitting, and the backward inversion algorithm. Whereas those procedures are inherently not adaptable because independent inversions are performed for each return signal and neither the statistics of the signals nor a priori uncertainties (e.g., boundary calibrations) are taken into account, in the case of the Kalman filter the filter updates itself because it is weighted by the imbalance between the a priori estimates of the optical parameters (i.e., past inversions) and the new estimates based on a minimum-variance criterion, as long as there are different lidar returns. Calibration errors and initialization uncertainties can be assimilated also. The study begins with the formulation of the inversion problem and an appropriate atmospheric stochastic model. Based on extensive simulation and realistic conditions, it is shown that the EKF approach enables one to retrieve the optical parameters as time-range-dependent functions and hence to track the atmospheric evolution; the performance of this approach is limited only by the quality and availability of the a priori information and the accuracy of the atmospheric model used. The study ends with an encouraging practical inversion of a live scene measured at the Nd:YAG elastic-backscatter lidar station at our premises at the Polytechnic University of Catalonia, Barcelona.