Improving the accuracy in particle concentration measurements of a balloon-borne optical particle counter UCASS

Abstract. For balloon-borne detection of aerosols and cloud droplets (diameter 0.4 < D_p< 40 µm), a passive-flow Universal Cloud and Aerosol Sounding System (UCASS) was used, whose sample flow rate is conventionally derived from GPS-based balloon’s ascent rates. Improvements are achieved by implementing thermal flow sensors (TFS) 94 mm downstream of the UCASS detection region for continuously measuring true UCASS sample flow velocities. UCASS-mounted TFS were calibrated during wind tunnel experiments at up to 10 m s^-1 also under various angles-of-attack (AOA), as these vary during actual balloon ascents. It was found that the TFS-calibration is determined with sufficient precision using three calibration points at tunnel flows of ~ 2, 5, and 8 m s^-1, simplifying efficient TFS-upgrades of numerous UCASS. In iso-axial alignment, UCASS flows are accelerated (by ~ 11.3 %) compared to tunnel flows (at 2 – 8 m s^-1). In-flight comparisons revealed that UCASS sample flows rarely match the balloon’s ascent rate, instead, equality (v_GPS = v_TFS) is achieved only at AOA ≠ 0°, potentially affecting the UCASS-internal flow pattern and particle transmission efficiency. To minimise errors on calculated UCASS-based particle number concentrations, real-time measurements of the true UCASS flow velocity are recommended.