LIDAR measurements: Determination of the aerosol extinction coeffcient and comparison of incomplete overlap correction methods

Kurzfassung

The energy demand that accompanies the economic stability of the industrialized countries is steadily growing. As the stock of fossil fuels becomes extinct, the need of renewable energy resources emerges. The Concentrated Solar Power technology offers a climate-friendly alternative for the production of electricity in the renewable energy market. A key factor of the e�ciency of a CSP plant is the extinction of the direct normal irradiance, due to scattering and absorption of light by molecules and particles. In this thesis elastic LIDAR measurements, corresponding to 33 h, have been performed at the Plataforma Solar de Almer��a and analyzed, using reference data of in-situ sun photometer and ceilometer measurements. With the application of the Klett-Fernald method the aerosol extinction coe�cient and the aerosol optical depth (AOD) were determined for the 355nm and 532nm lidar wavelengths and analogous for the 1064nm ceilometer wavelength. Sensitivity studies show that the choice of the reference height, the lidar ratio and the application of the background correction are of major importance for the Klett-Fernald method. Furthermore, the AOD data sets of the LIDAR system, the sun photometer and the ceilometer were intercompared. The deviation between the LIDAR and sun photometer AODs indicates that the LIDAR system su�ers from the overlap problem due to a lack of coincidence between the laser's and the receiver's �fields of view at altitudes below 120 m. A comparison of ceilometer and LIDAR data reveals that the aerosol extinction coe�cient and the according AOD determined with the ceilometer are smaller than the results obtained with the LIDAR. This was thought to be primarily due to the difference in operating wavelengths of both instruments. Therefore, a comparison of ceilometer data to sun photometer data was accomplished. To retrieve the AODs at the speci�c wavelength of 1064nm, the �Angstr�om approach has been applied to the sun photometer data. The AODs obtained with the ceilometer are also smaller than for sun photometer measurements leading to the assumption that a calibration of the ceilometer system is needed. To correct the loss of signal of the LIDAR system in the incomplete overlap region, three different correction functions have been applied to the lidar backscatter signal. The aerosol density in the lower troposphere is expected to be higher than for greater altitudes, due to stirred-up dust and urban pollution. A correction function, therefore, must cause an ampli�cation of the raw lidar backscatter signal in the lowest 120m. The analytical correction function by Stelmaszczyk et al. (2010) is evolved from geometric considerations and has a small impact on the AOD for the 355nm wavelength and no impact for the 532nm wavelength and is therefore not suitable for the correction of the lidar signal. The correction function proposed by Biavati et al. (2011), derived from on an iterative approach from angular measurements, results in an "over-correction" of the signal. The third correction method proposed by Guerrero-Rascado et al. (2010) is iteratively achieved by comparison to the ceilometer attenuation coe�cient at low altitudes and suits for the correction of the lidar signal. Finally, the lidar extinction coe�cient at 90m altitude, obtained �rstly from tilted measurements and secondly from the overlap corrected vertical signal, is compared to a Vaisala FS11 scattermeter data set. The results prove that the correction function by Guerrero-Rascado et al. (2010) is best suited for the LIDAR system.