Relationship of cloud top to the tropopause and jet structure from CALIPSO data

Cloud top and tropopause relationships are examined using cloud top observations from the Cloud-Aerosol Lidar Infrared Pathfinder Satellite Observations (CALIPSO) cloud data and National Centers for Environmental Prediction Global Forecast System (GFS) tropopause data. Statistical analyses of cloud top occurrence in tropopause and jet referenced relative altitude coordinates are performed on a global scale using 4 years (2006–2010) of CALIPSO 5 km resolution cloud layer data. The results show that the thermal tropopause appears to be a significant constraint for the cloud top. The zonal vertical distribution of cloud tops in tropopause-relative coordinates shows a maximum at the tropopause level for both the tropics (20°S–20°N) and midlatitudes (40°–60°S, 40°–60°N) for all four seasons. Occurrence of cloud tops above the tropopause is examined and quantified. The results show that with the consideration of tropopause height uncertainty, the data do not provide sufficient evidence of significant presence of cloud tops above the tropopause in the midlatitudes. In the tropics, the significant occurrences of cloud top above the thermal tropopause are found in regions known for seasonal deep convection. In most cases, the occurrence is up to 24% in 2° × 3° latitude-longitude bins with isolated higher frequencies in the western Pacific during the northern hemispheric winter season. The vertical distributions show that these events are mostly up to 2.5 km above the lapse rate tropopause, which is comparable to the differences between the lapse rate and the cold point tropopause in regions of active convection. We speculate that this separation may be responsible for a significant fraction of the cloud tops that do occur above the lapse rate tropopause in our analyses. It is also important to note that our results are limited by the CALIPSO twice-daily sampling with local equator crossing times of 0130 and 1330. The data therefore do not provide a good representation of convection over land, which is known to have maxima in afternoon local times. The tropopause determination is a significant component of this type of studies, and errors in the tropopause height may lead to significantly different conclusions. Our analyses show that the tropopause product from the GFS model is in better agreement with radiosonde measurements. The Goddard Earth Observing System Model Version 5 tropopause product, given as the ancillary data in the CALIPSO data file, shows a much larger uncertainty, primarily because the tropopause is identified at the model grid levels.