Intercomparison of height assignment methods for opaque clouds over the tropics

Abstract

Several methods of determining the height of opaque clouds over the tropics were compared using geostationary satellite measurements. The possible use of ozone channel measurements around the 9.7-μm ozone absorption band was examined in conjunction with the infrared window (IRW; 10.8 μm), H2O (6.3 μm), and CO2 (13.4 μm) channels, which are generally used for the assignment of cloud heights. Cloud top heights were retrieved from Meteosat-8 measurements with the aid of radiative transfer calculations using reanalysis data from the National Centers for Environmental Prediction/National Center for Atmospheric Research (NCEP/NCAR) as inputs. By using cloud top heights from collocated CloudSat observations as a reference, cloud top heights were determined from the one-channel radiance, two-channel brightness temperature difference (BTD), and two-channel radiance ratio methods, and the respective results were then compared for clouds with geometrical thicknesses of > 4 km. Overall, the retrievals from the CO2-IRW ratio and O3-CO2 ratio methods are in substantial agreement with CloudSat observations, while the other methods either underestimate cloud top heights or demonstrate a lower ratio of successful height assignment. The O3-CO2 ratio method appears to be less practical than the CO2-IRW ratio method because it requires two absorption channels. Our comparison also shows that the BTD between the ozone and IRW channels yields information that is similar to that of the IRW channel alone. It further shows that the O3-IRW combination is not appropriate for the two-channel radiance ratio method. These results suggest that the inclusion of the ozone channel in BTD and ratio methods may not offer any significant improvement in convective cloud height retrieval over the tropics. In conclusion, the CO2-IRW ratio method appears to provide the most accurate retrievals for opaque clouds.