Drizzling low clouds at Graciosa Island - how does the ECMWF model perform?

Poster PDF

Description

Comprehensive observations of clouds, precipitation, and surface radiation were gathered over a 19-month long period during the ARM Mobile Facility (AMF) deployment on Graciosa Island. During this time, boundary-layer clouds were the most frequently observed cloud type, often associated with drizzling conditions.

A common problem in global models is an overestimate of light precipitation occurrence. The observations from Graciosa provide a ready opportunity to evaluate and improve the representation of low clouds and precipitation in models. We evaluate the occurrence of clouds, surface precipitation, and precipitation re-evaporation in the European Centre for Medium-Range Weather Forecasts (ECMWF) model, as well as cloud properties and biases in surface radiation.

The ECMWF model’s monthly statistics of cloud occurrence and type are generally good and in reasonable agreement with the observations. However, systematic but compensating surface radiation errors exist and can be linked with broken and overcast low cloud conditions. This is a feature of the model previously identified at the continental ARM Southern Great Plains site.

As anticipated, the model’s precipitation occurrence does not agree well with observations. Precipitation is produced too frequently at cloud base, and only a small fraction of it re-evaporates before reaching the surface, leading to an overestimate of surface precipitation frequency as well. The single-column model is used to test combined changes to the autoconversion/accretion and precipitation re-evaporation processes in the convection and cloud parameterizations to improve the model cloud-radiation impacts and precipitation statistics at Graciosa.