Characteristics of CALIOP attenuated backscatter noise: implication for cloud/aerosol detection
1Jet Propulsion Laboratory, California Institute of Technology, Pasadena, California, USA
2Joint Institute for Regional Earth System Science and Engineering, University of California, Los Angeles, California, USA
3Harvard University, Cambridge, Massachusetts, USA
Abstract. To study cloud/aerosol features in the upper troposphere and lower stratosphere (UT/LS) with the NASA's A-Train sensors, a research algorithm is developed for a re-gridded CALIOP (Cloud-Aerosol Lidar with Orthogonal Polarization) Level 1 (L1) backscatter dataset. This paper provides a detailed analysis of the measurement noise of this re-gridded dataset in order to compare the lidar measurements with other collocated measurements (e.g., CloudSat, Microwave Limb Sounder). The re-gridded dataset has a manageable data volume for multi-year analysis. It has a fixed (5 km) horizontal resolution, and the measurement error is derived empirically from the background-corrected backscatter profile on a profile-by-profile basis. The 532-nm and 1064-nm measurement noises, determined from the data at altitudes above 19 km, are analyzed and characterized in terms of the mean (μ), standard deviation (σ), and normalized probability density function (PDF). These noises show a larger variance over landmasses and bright surfaces during day, and in regions with enhanced flux of energetic particles during night, where the instrument's ability for feature detection is slightly degraded. An increasing trend in the nighttime 1064-nm σ appears to be significant, which likely causes the increasing differences in cloud occurrence frequency between the 532-nm and 1064-nm channels. Most of the CALIOP backscatter noise distributions exhibit a Gaussian-like behavior but the nighttime 532-nm perpendicular measurements show multi-Gaussian characteristics. We apply σ – based thresholds to detect cloud/aerosol features in the UT/LS from the subset L1 data. The observed morphology is similar to that from the Level 2 (L2) 05km_CLAY+05km_ALAY product, but the occurrence frequency obtained in this study is slightly lower than the L2 product due to differences in spatial averaging and detection threshold. In the case where the measurement noises of two data sets are different, the normalized PDF has proven useful for quantifying the day-night difference of the CALIOP backscatters, showing higher daytime cloud occurrence frequency in the tropical UT/LS. Other cloud/aerosol properties, such as depolarization ratio and color ratio, can be also evaluated with the PDF method.