Christophe Hoareau

Methodology for Water Monitoring in the Upper Troposphere with Raman Lidar at the Haute-Provence Observatory

A Raman water vapor lidar has been developed at the Haute-Provence Observatory to study the distribution of water in the upper troposphere and its long-term evolution. Some investigations have been proposed and described to ensure a pertinent monitoring of water vapor in the upper troposphere. A new method to take into account the geophysical variability for time integration processes has been developed based on the stationarity of water vapor. Successive measurements, considered as independent, have been used to retrieve H2O profiles that were recorded during the same nighttimes over a few hours. Various calibration methods, including zenith clear-sky observation, standard meteorological radiosondes, and total water vapor column, have been investigated. A method to evaluate these calibration techniques has been proposed based on the variance weakening. For the lidar at the Haute-Provence Observatory, the calibration based on the total water vapor column appears to be the optimum method. Radiosondes also give comparable results, but do not allow lidar to be independent. The clear-sky zenith observation is an original technique, and seems to accurately identify discontinuities. However, it appears to be less reliable, based on the variance investigation, than the two others. It is also sensitive to aerosol loading, which is also expected to vary with time.

An EarthCARE/ATLID simulator to evaluate cloud description in climate models

Clouds still remain the largest source of uncertainty in model-based predictions of future climate, thus the description of the clouds in climate models needs to be evaluated. In particular, the cloud detailed vertical distribution that impacts directly the cloud radiative effect needs to be evaluated. Active satellite sensors directly measure the cloud vertical distribution with high accuracy; their observations should be used for model evaluation together with a satellite simulator in order to allow fair comparison between models and observations. The next cloud lidar in space, EarthCARE/ATLID, is planned for launch in 2018, while the current spaceborne cloud lidar CALIPSO/CALIOP is expected to stop collecting data within the next coming years. Here we describe the characteristics of the ATLID lidar onboard the EarthCARE satellite (spatial resolution, SNR, wavelength, field of view, PRF, orbit, HSRL) that need to be taken into account to build a COSP/ATLID simulator. We then present the COSP/ATLID simulator, and the Low, Mid, High level cloud covers it produces, as well as the zonal mean cloud fraction profiles and the Height-intensity histograms that are simulated by COSP/ATLID when overflying an atmosphere predicted by LMDZ5 GCM. Finally, we compare the clouds simulated by COSP/ATLID with those simulated by COSP/CALIPSO when overflying the same atmosphere. As the main differences between ATLID and CALIOP are taken into account in the simulators, the differences between COSP/ATLID and COSP/CALIPSO clouds covers are less than 1% in night time conditions

Subgrid-scale cirrus observed by lidar at mid-latitude: variability effects of the cloud optical depth

Abstract The temporal variability of the 532-nm optical depth of cirrus clouds observed with a lidar at Observatory of Haute-Provence (43.9°N, 5.7°E, and 683-m altitude), has been analyzed. While advection dominates at the first order, variability of the optical depth on timescales of minutes can be related to spatial fluctuations of cloud properties on typical scales of a few kilometers. Log-normal distributions of the optical depth have been used to model the variability of the cirrus optical depth as observed by lidars. These investigations have been performed for three independent classes of cirrus. The log-normal distribution of the optical depth is applicable to the classes of thin clouds; however, for thick clouds, likely due to successive freezing/defreezing effects, the distribution is rather bimodal. This work compares the effects of visible solar light scattered by inhomogeneous cirrus to effects generated by homogeneous clouds having a constant geometrical thickness using the short-scale lidar observations of optical depth distribution and an analytical approach. In the case of thin cirrus, the scattering of solar light reaching the ground is stronger for inhomogeneous than homogeneous cirrus. In case of thick cirrus, multiple-scattering processes need to be considered. The conclusion is that log-normal distribution of the cirrus optical depth should be considered in any radiative calculation in case of model grids larger than a few kilometers whatever the cirrus type is.