Application of CM-SAF cloud and radiation products to verify surface cloud radiative effects in climate models over ocean

Clouds play a major role in the climate system due to their impact on the energy budget, both at the surface and at the top-of-atmosphere. It is therefore important to understand cloud processes and cloud-radiation interactions, and to represent them correctly in climate models. The Satellite Application Facility on climate monitoring (CM-SAF) provides global datasets of cloud properties and radiative fluxes for climate studies and model evaluation. However, these datasets also require validation by independent measurements to determine their accuracies. To date, only limited measurements are available over ocean for this purpose. Through the OCEANET project, the atmospheric conditions for several Atlantic ship transects between the northern and southern hemisphere are available with high temporal resolution cloud and radiation measurements. In this study, the CM-SAF radiation products are compared to OCEANET measurements, as well as with the output of the shortwave radiation scheme of a state-of-art climate model. The radiation transport scheme is taken from the ECHAM-5 global circulation model, and is used as a single column model. As model input, satellite-retrieved cloud properties such as cloud liquid water path, cloud top temperature or effective radius, are complemented by OCEANET measurements such as microwave-retrieved profiles of temperature and humidity, cloud liquid water path and cloud cover. The resulting fluxes at the surface are compared against OCEANET measured and CM-SAF retrieved fluxes.

The deviation between the model predictions and the measurements are discussed for several cases. Therefore three experiments are defined with different input parameter. All cases show partly large deviation between modelled and measured fluxes. Also overcast conditions lead to large depending of the liquid water path distribution on the downward shortwave radiation. Different reasons for deviation are discussed such as problems of comparing temporal averaged ship based LWP and spatial averaged CM-SAF LWP. Also the cloud cover can have a large effect through its definition of diffuse and direct parts of surface irradiance. In general the comparison of CM-SAF solar fluxes at the surface to modelled fluxes base on the CM-SAF cloud properties shows higher agreement over the day in all cases but still show a significant offset sometimes. Also the shortwave cloud radiative effect shows positive values in the measured data that are not observed in the modelled data.