Fig.1 Intercomparison of cloud optical depth retrievals for June 16, 2006

Fig. 2 Average lidar ratios in clouds on March 15, 2006

   Low altitude clouds play an important role in global climate forcings, weather and precipitation etc. Unfortunately, measurement of low altitude clouds from space is very difficult due to their low optical depth and the fact that they are relatively warm. and it  is still a big challenge to accurately measure and model their optical and microphysical properties in order to assimilate them to global climate models. In particular, low clouds often have large liquid water path (LWP), and are involved in interactions with anthropogenic aerosols and the atmospheric boundary layer. Even though a single layer of low cloud usually simplifies modeling, intercomparisons among different retrievals and instruments indicate large discrepancies on LWP and optical depth. For satellite sensors with visible and near-infrared channels, low and optically thin clouds are also difficult to detect due to their partial transparency and land surface emission. On the other hand, lidar has been extensively demonstrated for observing clouds. However, most previous works with lidar concentrate on high and thin cirrus clouds at  night. 

This paper focuses on exploring the accuracy and limits of measuring low altitude optically thin cloud measurements by comparing different methods of retrieval. In the first method, the extinction coefficient of particulates (aerosol or cloud) can be directly derived from the N₂-Raman return. without the need for any calibration values. Integrating the Raman-retrieved extinction profile from cloud base z_b to top z_t, provides the first direct approach for the determination of the cloud optical depth.   In the second approach to measure thin cloud optical depth, Young¹ presents a method based solely on the elastic lidar returns above and below the cloud layer. In that method, the actual lidar elastic returns below and above clouds are fitted to theoretical molecular scattering returns which work well for high cirrus because any residual aerosols can be ignored at high altitudes both above and below the cloud. However, for low clouds, corrections for aerosol influences have to be carefully treated due to high aerosol loading in the lower atmosphere. To this end, it is most useful to examine cases where some time intervals are cloud free so that estimates of the aerosol ratio can be calculated. Such a calculation can be done either with an elastic lidar alone in which case an aerosol S-ratio must be assumed or with a combined elastic-raman lidar which calculates the extinction directly and can be used in the elastic channel retrieval. The results of such a comparison are shown in figure 1. In the figure, "raman" is simply the integration of the aerosol extinction. "Mie-uncor" describes the regression method where it is assumed no aerosol is lies below or above the cloud. "Mie-cor-1" describes a correction for aerosols obtained by considering only the elastic signal and therefore makes use of an assumed S ratio. "Mie-cor-2"  We note that when the aerosol loading both above and below the cloud is characterized using the combined Mie-Raman approach, very good agreement is found. What is most interesting is the significant errors when an assumed S-ratio is used. On the other hand, the Raman-Mie approach degrades faster for higher COD values since in this case, both the  N₂-Raman and elastic returns must pass through the complete cloud layer. Therefore, the method will tend to overestimate the extinction at the top edge of the cloud resulting in overestimated of optical depth.

Once the COD measurements are confirmed, providing confidence in the extinction profile within the cloud, estimates of extinction to backscatter ratio can be made within the cloud. An illustration is given in figure 2. We find that when the lidar ratio in cloud is averaged over the vertical extent, an S ratio on the order of 20 sr^-1 is found which is consistent with conventional water phase cloud droplet models. 

Using a reasonable water phase droplet size model, we use the S ratio measurements within the cloud to  explore the droplet sizes.  We find that although the S ratio is near the critical value of 20, in the interior of most clouds we investingate, there are significant regions on the cloud perimeter where  lidar ratio increases markedly and this implies the presence of smaller droplets such as would be associated either with new condensation or evaporation of the droplets near the cloud edge.  Applications to aerosol-cloud interaction and improvements using data at 1064nm will also be discussed.

Acknowledgement

This work was partially supported by the NOAA  Interdisciplinary Scientific Environmental Technology (ISET) Cooperative SCIENCE Center  under grant # NA06OAR4810187 and and NASA #NCC-1-03009

References

1.        S. A. Young, Analysis of lidar backscatter profiles in optical thin clouds, Appl. Opt.,30, 7019~7030 (1995).

2.        E. O'Connor , "A Technique for Autocalibration of Cloud Lidar", Journal of Atmospheric and Oceanic Technology V21, .777-781 (2004).