Understanding Global Model Systematic Shortwave Radiation Errors in Subtropical Marine Boundary Layer Cloud Regimes

• Published in: Journal of advances in modeling earth systems Vol. 10; no. 8; pp. 2042 - 2060
• Main Authors: Ahlgrimm, Maike, Forbes, Richard M., Hogan, Robin J., Sandu, Irina
• Format: Journal Article
• Language: English
• Published: Washington John Wiley & Sons, Inc 01.08.2018; American Geophysical Union (AGU)
• Subjects: Boundary layers; Climate models; Cloud cover; Cloud properties; Clouds; Cumulus clouds; Diurnal cycle; ENVIRONMENTAL SCIENCES; Errors; Heterogeneity; marine boundary layer clouds; Meteorological satellites; Oceans; Radiation; Radiation calculations; Satellites; Ships; shortwave bias; Stratocumulus clouds; Temperature; Weather forecasting
• ISSN: 1942-2466; 1942-2466
• DOI: 10.1029/2018MS001346

Global numerical weather prediction and climate models are subject to long‐standing systematic shortwave radiation errors due to deficiencies in the representation of boundary layer clouds over the ocean. In the subtropics, clouds are typically too reflective in the cumulus regime and not reflective enough in the stratocumulus regime. Potential sources of error include cloud cover, liquid water path, effective radius, and subgrid heterogeneity, but diagnosing the absolute contributions of each to the radiation bias is hampered by uncertainties and sometimes contradictory information from different observational products. This paper draws on a set of ship‐based observations of boundary layer clouds obtained during the ARM MAGIC campaign along a northeast Pacific Ocean transect, crossing both stratocumulus and shallow cumulus cloud regimes. The surface‐based observations of cloud properties are compared with various satellite products, taking account of the diurnal cycle, to provide an improved quantitative assessment of the deficiencies in the European Centre for Medium‐Range Weather Forecasts global numerical weather prediction model. A series of off‐line radiation calculations are then performed to assess the impact on the shortwave radiation bias of correcting each of the model's deficiencies in cloud characteristics along the transect. A reduction in the bias is achieved by improving the agreement between modeled and observed in‐cloud liquid water path frequency distributions. In the cumulus regime, this is accomplished primarily by reducing the all‐sky water path, while for the stratocumulus regime, an underestimate of cloud cover and liquid water and an overestimate in effective radius and subgrid heterogeneity all contribute to a lack of reflected shortwave radiation. Key Points A methodology to attribute the shortwave radiation bias to specific errors in cloud properties is applied to the global ECMWF model In trade cumulus cloud, the shortwave bias is almost entirely due to errors in the model's liquid water path In stratocumulus clouds, errors in cloud cover, water path, effective radius, and subgrid heterogeneity all contribute to the shortwave bias