A23K-01
Relationships Among Top-of-atmosphere Radiation and Atmospheric State Variables in Observations and CESM

Tuesday, 15 December 2015: 13:40
3004 (Moscone West)
Kevin E Trenberth, National Center for Atmospheric Research, Boulder, CO, United States
Abstract:
A detailed examination is made in both observations and the Community Earth System Model (CESM) of relationships among top-of-atmosphere (TOA) radiation and surface air temperatures, as well as water vapor, tropospheric temperatures and precipitation for 2000-2014 to assess the origins of radiative perturbations and climate feedbacks empirically. The 30-member CESM large ensemble coupled runs are analyzed. Both global and local relationships are examined. There is a lot more high frequency variability in radiative fluxes than in temperature, highlighting the role of clouds and transient weather systems in the radiation statistics. Surface temperatures respond to a radiative imbalance and also greatly affect the outgoing longwave radiation OLR), especially over land. However, tropospheric temperatures are much more influenced by clouds, which affect both absorbed solar radiation (ASR) and OLR, and with large compensation. The vertical structure of the CESM temperature profile tends to be top-heavy in the model, with too much deep convection and not enough lower stratospheric cooling as part of the response to tropospheric heating. There is too much ASR over the southern oceans and not enough in the tropics, and ENSO is too large in amplitude in this version of the model. However, the co-variability of monthly mean anomalies produces remarkably good replication of most of the observed relationships. Over the Warm Pool in the tropical western Pacific and Indian oceans, where non-local effects from the Walker circulation driven by the ENSO events are important, several related biases emerge: in response to high SST anomalies there is more precipitation, water vapor and cloud, and less ASR and OLR in the model than observed. Different model global mean trends are evident, however, and hint at too much positive cloud feedback in the model.