Global and regional estimates of warm cloud droplet number concentration based on 13 years of AQUA-MODIS observations
Abstract. We present and evaluate a climatology of cloud droplet number concentration (CDNC) based on 13 years of Aqua-MODIS observations. The climatology provides monthly mean 1 × 1 degree CDNC values plus associated uncertainties. All values are in-cloud values, that is, if the grid box is covered to 10 % with clouds, then the reported CDNC value will be valid for the cloudy part of the grid-box. Here, we provide an overview on how the climatology was generated and assess and quantify potential systematic error sources including effects of broken clouds, and remaining artefacts caused by the retrieval process or related to observation geometry. Retrievals and evaluations were performed at the scale of initial MODIS observations (in contrast to some earlier climatologies, which were created based on already gridded data). This allowed us to implement additional screening criteria, so that observations inconsistent with key assumptions made in the CDNC retrieval could be rejected. Application of these additional screening criteria led to significant changes in the annual cycle of CDNC both in terms of its phase and magnitude. After an optimal screening was established a final CDNC climatology was generated. Resulting CDNC uncertainties for the climatology are in the order of 30 % in the stratocumulus regions and 60 % to 80 % elsewhere. The climatology is available in Network Common Data Format (netCDF) and adheres to the Climate and Forecast (CF) convention. The climatology is available via Digital Object Identifier (Bennartz and Rausch, 2016, doi:10.15695/vudata.ees.1).