3D Shortwave Radiative Kernels of Marine Boundary-layer Clouds Using Scanning Radar/Lidar and Array Spectroradiometer

Abstract

The response of global mean surface temperature to emissions of greenhouse gases from human activities remains highly uncertain. One of the primary causes for this uncertainty is cloud feedback, to what extent changes in low-altitude cloud cover and properties will amplify or dampen climate change. To better quantify cloud feedback, the radiative kernel approach has been increasingly used to compare model simulations across a range of climate change scenarios.
Cloud radiative kernels are primarily calculated from climate model simulations of broadband fluxes at the top of the atmosphere, using plane-parallel clouds and one-dimensional (1D) radiative transfer. These kernels, however, have not been evaluated by observations and cannot provide proper constraints on surface radiation that is crucial for understanding the radiative and hydrological energy balance of the climate system. Broadband fluxes alone are also insufficient to detect compensating errors in climate models. Importantly, the cloud feedback uncertainty is most evident for the region of marine subtropical low-level clouds that are highly heterogeneous; thus, the assumption of plane-parallel clouds with 1D radiative transfer is not appropriate.
The objective of this proposal is for the first time to develop fully three-dimensional (3D), observationally-based and spectrally-resolved radiative kernels at both the surface and the top of the atmosphere, which is necessary in tackling the uncertainty in cloud feedback. To do this, we propose to develop a novel 3D cloud retrieval method that uses ARM measurements synergistically by adapting an ensemble Kalman Filter approach. The retrieved cloud fields will be used to compute 3D kernels with help from a newly developed radiative transfer scheme. Finally, we will exploit this new knowledge of the observed 3D cloud fields to assess the inaccuracies in model-based 1D kernel calculations and impacts upon the estimated cloud feedback magnitude.
Results from this project will be of considerable interest to the climate modelling community, in particular those interested in systematic errors associated with cloud properties and in understanding and quantifying cloud feedbacks. Extending the radiative kernel approach beyond the state of the art, to consider the 3D and spectral nature of cloud feedback, will make a novel contribution to calculating cloud feedback more precisely.