Albedo bias and the horizontal variability of clouds in subtropical marine boundary layers: observations from ships and satellites
Robert Pincus, Sally A. McFarlane, and Stephen A. Klein
Journal of Geophysical Research, March 1999.
Volume 104, pages 6183-6192.

Abstract

Cloud optical properties vary dramatically at spatial scales smaller than typical grid cells in large-scale models. If left untreated this variability can lead to a significant overestimate of cloud albedo. Computational techniques for reducing this plane parallel homogeneous (PPH) albedo bias exist if the mean cloud optical thickness and the amount of variability are available, but very little is known about how much variability is to be expected, and to what factors it is sensitive.

The authors combine 1105 observations made by volunteer surface observers with moderate resolution satellite imagery to assess the relationships between cloud fraction, cloud optical properties, and cloud type. The study focuses on marine boundary layer clouds off the coast of California during summer. Estimates of cloud fraction from the two datasets are compared, and are in best agreement when a reflectance threshold between 0.09 and 0.10 is used. Satellite-derived cloud fraction increases slowly with sensor resolution at spatial scales from 1 to 32 km. Cloud fraction in scenes dominated by cumulus is much more sensitive to the reflectance threshold used for cloud detection than are scenes containing stratiform clouds.

The mean magnitude of the PPH bias found here, 0.025, is considerably smaller than those found in other recent studies. Reasons for this difference are discussed. When fit to the observed distributions of optical thickness both log-normal and gamma distributions substantially reduce the PPH bias, although the fits are not statistically significant. The mean and dispersion of log optical thickness are related to cloud type: optical thickness increases as cloud type changes from cumuliform to stratiform, while the relative amount of variability decreases. The authors suggest that this relationship could provide the basis for the parameterization of unresolved variability in large scale models.

Download a reprint (PDF format); request a copy via email.
Copyright 1999 American Geophysical Union. Further reproduction or electronic distribution is not permitted.

We wrote a non-technical description of this paper.