11:00 〜 13:00
[ACG38-P06] Characterizing the cloud thermodynamic phase using SWIR and LIDAR measurements
The cloud thermodynamic phase (whether a cloud is composed of liquid water droplets ice crystals, or a mixture of them) is a fundamental cloud property that characterizes the radiative effect of clouds. Furthermore, representations of cloud phase in global climate models are a source of uncertainty in model estimates of the climate sensitivity. Global information on cloud phase from satellite observations therefore provides a critical observational constraint on climate prediction by global models.
Cloud phase information estimated from the CALIPSO lidar (CALIOP) measurements in a temperature-independent manner has often been used to evaluate the cloud phase representation in the models. However, the CALIOP-derived cloud phase information is limited to a relatively shallow depth from the cloud top, and therefore can only be used to evaluate a limited aspect of the cloud phase representations in the models. In contrast, cloud phase information derived from the shortwave-infrared (SWIR) channels of passive sensors such as Aqua/MODIS and GCOM-C/SGLI, is a viable option for exploring cloud phase at optical depths deeper than CALIOP. Therefore, combining the two pieces of cloud phase information from CALIOP and SWIR is expected to characterize vertical inhomogeneity of cloud thermodynamic phase, which cannot be obtained by using each of the two sensors separately.
In this study, we employed two cloud phase products as obtained from collocated observations by CALIPSO/CALIOP and Aqua/MODIS, part of the A-Train satellite constellation. For the former, the CALIOP cloud particle type product distributed by the JAXA A-Train Product Monitor was used. For the latter, we used a new cloud phase product derived by applying a cloud phase retrieval algorithm developed by our recent studies to MODIS SWIR measurements provided in the MODIS-AUX product distributed by the CloudSat Data Processing Center.
First, we compared temperature dependences of ice-phase fractions obtained by analyzing the two cloud phase products to find a significant difference, which appears to be large enough to affect model evaluations of cloud phase representations. Next, we combined the cloud phases binarized into liquid (LIQ) or ice (ICE) obtained from the two sensors to classify the vertical heterogeneity of the cloud phases into four categories: LIQ/LIQ, LIQ/ICE, ICE/LIQ, and ICE/ICE, denoted in the form of CALIOP/MODIS. Furthermore, we interpreted the cloud phase vertical heterogeneity characterized by CALIOP and SWIR through comparisons with MODIS thermal infrared (TIR) measurements and CloudSat/CPR radar profiles to validate the four classifications and to investigate how different combinations of the cloud phases correspond to different cloud vertical structures. The results suggest that a combined use of the complementary information from the four sensors (lidar, SWIR, TIR, and radar) can better characterize the vertical structure of the cloud phase, which could then provide more detailed observational constraints on model representations of cloud phase.
Cloud phase information estimated from the CALIPSO lidar (CALIOP) measurements in a temperature-independent manner has often been used to evaluate the cloud phase representation in the models. However, the CALIOP-derived cloud phase information is limited to a relatively shallow depth from the cloud top, and therefore can only be used to evaluate a limited aspect of the cloud phase representations in the models. In contrast, cloud phase information derived from the shortwave-infrared (SWIR) channels of passive sensors such as Aqua/MODIS and GCOM-C/SGLI, is a viable option for exploring cloud phase at optical depths deeper than CALIOP. Therefore, combining the two pieces of cloud phase information from CALIOP and SWIR is expected to characterize vertical inhomogeneity of cloud thermodynamic phase, which cannot be obtained by using each of the two sensors separately.
In this study, we employed two cloud phase products as obtained from collocated observations by CALIPSO/CALIOP and Aqua/MODIS, part of the A-Train satellite constellation. For the former, the CALIOP cloud particle type product distributed by the JAXA A-Train Product Monitor was used. For the latter, we used a new cloud phase product derived by applying a cloud phase retrieval algorithm developed by our recent studies to MODIS SWIR measurements provided in the MODIS-AUX product distributed by the CloudSat Data Processing Center.
First, we compared temperature dependences of ice-phase fractions obtained by analyzing the two cloud phase products to find a significant difference, which appears to be large enough to affect model evaluations of cloud phase representations. Next, we combined the cloud phases binarized into liquid (LIQ) or ice (ICE) obtained from the two sensors to classify the vertical heterogeneity of the cloud phases into four categories: LIQ/LIQ, LIQ/ICE, ICE/LIQ, and ICE/ICE, denoted in the form of CALIOP/MODIS. Furthermore, we interpreted the cloud phase vertical heterogeneity characterized by CALIOP and SWIR through comparisons with MODIS thermal infrared (TIR) measurements and CloudSat/CPR radar profiles to validate the four classifications and to investigate how different combinations of the cloud phases correspond to different cloud vertical structures. The results suggest that a combined use of the complementary information from the four sensors (lidar, SWIR, TIR, and radar) can better characterize the vertical structure of the cloud phase, which could then provide more detailed observational constraints on model representations of cloud phase.