Direct Assessment of Water Vapor and Cloud Feedbacks From Observations

Abstract

Cloud and water vapor feedbacks in climate theory will be briefly reviewed, as will the relevant physical processes associated with both cloud cover and water vapor. We will focus on the tropics in this talk. Measurements of these quantities relevant to climate must take account of a number of factors that have often been ignored: 1. Water vapor in the tropics is extremely spatially heterogeneous, and one dimensional treatments using spatially averaged values will almost always be misleading. 2. Precipitation from cirrus detrained from cumulus towers is a major source of tropospheric humidity. However, the evolution of cirrus detrainment is a process which takes on the order of 6 hours. Given the nature of satellite orbits, this leads to statistical issues that must be handled with care. 3. For climate purposes, there is a major difference between changes in cirrus associated with the concentration of convection, and changes normalized by the amount of convection. It is the latter that is relevant to climate sensitivity. These factors will be explained, and in the light of these factors, we will describe new results from CERES, TRMM, and MODIS instruments (as well as the Kwajalein ground based radar) on temperature dependence of precipitation efficiency and cloud coverage and the radiative properties of cirrus (from the recent work of Choi and Ho).