Outgoing longwave spectra obtained by the Atmospheric Infrared Sounder (AIRS) and GPS radio occultation data obtained by the CHAMP (Challenging Mini-satellite Payload) and COSMIC (Constellation Observing System for Meteorology, Ionosphere and Climate) missions will be analyzed to provide observational constraints on longwave feedbacks and radiative forcing in the climate system. The high-spectral-resolution radiance measurements of AIRS contain the fingerprints of radiative forcing by a variety of atmospheric thermodynamic variables and constituent concentrations, and these fingerprints lead directly to longwave greenhouse forcing and atmospheric feedbacks. Because clouds and surface temperature contribute similar spectral signatures, radio occultation will provide the information to distinguish between the two.
The investigators found, under their previous NSF support, that radio occultation trends constrain the surface air temperature response of climate models while infrared spectral radiance trends unambiguously constrain radiative forcing of the climate and the longwave feedbacks in climate models. Now they will apply these findings to the aforementioned satellite data to obtain estimates of radiative forcing by well-mixed greenhouse gases globally to at least 10% uncertainty, to estimate tropical water vapor and cloud longwave feedbacks to ~7% uncertainty by anomaly analysis, and to begin the estimation of global longwave feedbacks by trend analysis, albeit with large uncertainties due to natural variability.
Broader impacts of this work lie in its use of optimal fingerprinting using multiple data types to test climate models against satellite data, and in the field of climate prediction. Highly accurate and reliable data will be used to provide strong observational constraints on climate models. The ultimate goal of this research program is to improve our skill at decadal scale climate projection. In addition, graduated students from the joint Harvard and University of Michigan efforts will be supported.
We have utilized the AIRS data to derive the spectral flux and cloud radiative effect (CRE) at 10cm-1 interval over the entire longwave spectrum. The algorithm was further validated against collocated CERES measurements as long as using synthetic spectral flux and AIRS spectra. We then used such spectral flux and CRE to evaluate the climate model simulations. The advantage of such spectral-band resolved flux and CRE is that it avoids the compensating biases that inevitably exist in the broadband flux and CRE comparison. By diagnosing simulations from three state-of-the-art climate models (i.e., GFDL AM2, Canada CCCma CanAM4, and NASA GEOS-5), we quantitatively disclose the band-by-band details of the biases in broadband longwave flux and CRE, explore their underlying physical causes, and therefore, demonstrate the superiority of such band-by-band quantities in evaluating climate models, particularly its top-of-atmosphere radiation budget. This line of study of AIRS and its synergy with CERES observations has resulted in three peer-reviewed publications. In addition, we have studied the interannual variability of ice cloud water path and upper tropospheric humidity by using satellite observations, most recent reanalysis data sets, and large amount of climate model simulations prepared for the IPCC AR4 and AR5 assessments. We also studied how the ice cloud water path would change with the circulation change in the future climate. Furthermore, using a simple radiative-convective equilibrium model, we provided a physical explanation why the Walker circulation should weaken in response to the global warming. By this explanation, we are then able to assess the relative importance of the changes of clear-sky radiative cooling and the change of lapse rate on affecting the tropical Walker circulation. It turns that at different vertical levels in the troposphere, the roles played by two competing factors (clear-sky radiative cooling and lapse-rate change) are different. Two peer-reviewed publications have incurred from this line of study.
