10-minute resolution BACADA-derived cloud amount estimates at the Baseline Surface Radiation Network (1994-2014), link to NetCDF files

• Main Authors: Bojanowski, Jedrzej S, Stöckli, Reto
• Format: Dataset
• Language: English
• Published: PANGAEA 2017
• Subjects: Antarctica; Australia; BER; Bermuda; BOU; Boulder; Brasilia; Brasilia City; Distrito Federal; Brazil; BRB; CAB; Cabauw; Canada; CAR; Carpentras; Cener; CNR; Colorado; United States of America; DAA; DAR; Darwin; De Aar; Dronning Maud Land; E13; Elevation of event; Event label; File format; File name; File size; France; Georg von Neumayer; Germany; GOB; Gobabeb; GVN; Israel; IZA; Izaña; KWA; Kwajalein; Latitude of event; LIN; Lindenberg; Longitude of event; MAN; Momote; Neumayer; Antarc*
• Online Access: https://doi.pangaea.de/10.1594/PANGAEA.876005

Automated high-temporal resolution cloud cover measurements can contribute to the weak understanding of the net radiative cloud effect and its evolution with climate change. They can further serve as a reference for satellite-based cloud retrievals. A dataset of 10-minute cloud amount estimates at 24 sites of the Baseline Surface Radiation Network is presented. These sites are located worldwide covering a wide range of climatic zones. The length of cloud amount time series vary among sites from 3 to 22 years (until 2014). Cloud amount was calculated from ground measurements of long-wave incoming radiation, air temperature and relative humidity by means of the Bayesian Automatic Cloud Detection Algorithm (BACADA), which builds on the Automatic Partial Cloud Amount Detection Algorithm (APCADA, Dürr and Philipona, JGR, 2004). Evaluation of cloud fraction (0-100%) was carried out based on comparison with synoptic and total-sky imager cloud observations. It is demonstrated that BACADA improves the performance of partial cloud amount estimates (MBE=1.55%, MAE=15.35%) as compared to the existing APCADA algorithm (MBE=7.18%, MAE=17.89%). Yet, the aim of BACADA is to provide total cloud amount. These estimates are of MBE=-1.53% and MAE=17.86%. Although the study focuses on the need of cloud amount estimates for evaluation of satellite-based retrievals, the dataset demonstrated here may potentially be valuable for other disciplines.