The nonlinear relationship between albedo and cloud fraction on near‐global, monthly mean scale in observations and in the CMIP5 model ensemble

Abstract

We study the relation between monthly mean albedo and cloud fraction over ocean, 60°S–60°N. Satellite observations indicate that these clouds all fall on the same near‐exponential curve, with a monotonic distribution over the ranges of cloud fractions and albedo. Using these observational data as a reference, we examine the degree to which 26 climate models capture this feature of the near‐global marine cloud population. Models show a general increase in albedo with increasing cloud fraction, but none of them display a relation that is as well defined as that characterizing the observations. Models typically display larger albedo variability at a given cloud fraction, larger sensitivity in albedo to changes in cloud fraction, and lower cloud fractions. Several models also show branched distributions, contrasting with the smooth observational relation. In the models the present‐day cloud scenes are more reflective than the preindustrial, demonstrating the simulated impact of anthropogenic aerosols on planetary albedo.

Analysis of albedo versus cloud fraction relationships in liquid water clouds using heuristic models and large eddy simulation

“…Points are color coded by the relative cloud radiative effect, rCRE, a quantity of high relevance to clouds and climate. Most likely by chance, the results for f c based on a (visible) cloud optical depth τ c > 5 are remarkably similar to the best fit of the data by Engström et al [] derived from space‐based data for all clouds. Rather than focus on the similarities, we use these as illustrations of observed and modeled ‐ f c relationships and probe into factors (methodological, meteorological, macrophysical, microphysical, and radiative) that affect the relationship [see also Feingold et al , ].…”

“…Unless otherwise stated, an overhead Sun is assumed ( θ o =0). Note that this relationship is for visible albedo, in contrast to the broadband albedo measurements considered in Engström et al []. The ‐ f c relationship is then constructed from the Considine model using equations and .…”

“…The relationship between A and b is exponential: A = 1.49 × 10 −5 e (12.3 b ) . The Engström et al [] curve (dashed line) is superimposed for reference. (b) As in Figure a but for adiabatic clouds and considering the influence of solar zenith angle in the calculation of .…”

“…The endpoints of this relationship are defined by for f c =0 and for f c =1. Satellite remote sensing of a number of marine stratocumulus basins by the Moderate Resolution Imaging Spectroradiometer (MODIS) and Clouds and the Earth's Radiant Energy System spectrometers suggests a linear relationship between and f c [ Bender et al , ] for 1° × 1° monthly averages, while a more recent analysis of all clouds between 60°S and 60°N from the same instruments and averaging scales exhibits a characteristic curvature between the endpoints, which has been described by Engström et al [] as “exponential.” (We will refer to it as superlinear. )…”

Deeper, Precipitating PBLs Associated With Optically Thin Veil Clouds in the Sc‐Cu Transition

“…The dissipation of overlying stratocumulus and the development of puffy cumulus clouds beneath stratocumulus are often accompanied by the decoupling of the MBL. Several studies have shown that compared with observations, the model low-cloud fraction is not extensive enough, and the clouds that are represented have higher reflectivity, a problem known as the "too few, too bright problem" (Engström et al, 2015;Nam et al, 2012). At some point, overlying stratocumulus completely dissipates partly caused by strong entrainment of overlying dry air by the coupled cumulus, and active cumulus clouds become the dominant feature of the MBL, which completes the SCT (Jones et al, 2011;Krueger et al, 1995).…”