Spatial Aggregation of Satellite Observations Leads to an Overestimation of the Radiative Forcing Due To Aerosol-Cloud Interactions

The estimation of cloud radiative forcing due to aerosol-cloud interactions, RFaci (also known as the first indirect effect), relies on approximating the cloud albedo susceptibility to changes in droplet concentration, & beta;. & beta; depends on the cloud albedo and droplet concentration, both of which can be observed by satellites. Satellite observations are often spatially aggregated to coarser resolutions, typically 1 x 1 & DEG; scenes. However, on such spatial scales, the cloud albedo tends to be heterogeneous, whereas the & beta; approximation assumes homogeneity. Here, we demonstrate that the common practice of aggregating satellite data and neglecting cloud albedo heterogeneity results in an average overestimation of 10% in previous estimates of the RFaci. Additionally, we establish a relationship between the magnitude of the bias in & beta; and Stratocumulus morphologies, providing a physical context for cloud heterogeneity and the associated bias. Lastly, we propose a correction method that can be applied to cloud albedo gridded data. This paper explores the effect of cloud albedo morphology, which is a reflection of cloud heterogeneity, on radiative forcing due to aerosol-cloud interactions (RFaci). The RFaci is estimated from satellite observations based on the assumption that clouds are homogeneous within a given scene. However, when satellite data is spatially aggregated to reduce the amount of data to a user-friendly gridded format-a common practice-this assumption is no longer valid. Consequently, an overestimation of the RFaci occurs, particularly in heterogeneous scenes, where the overestimation can reach up to 50%. This means that the RFaci is lower than previously estimated. Our results also suggest that cloud albedo enhancement due to an increase in droplet concentrations would be most effective in homogeneous scenes. Therefore, marine cloud brightening strategies should take cloud albedo homogeneity into account to achieve the most effective albedo enhancement. The common practice of spatial aggregation of satellite data into 1 x 1 & DEG; scenes leads to an average 10% overestimation of the RFaciThe overestimation is due to neglecting cloud albedo heterogeneity, and is associated to different types of Stratocumulus morphologiesA correction is proposed, which calls for the incorporation of cloud reflectance statistics in Level 3 data