Numerical simulations and Arctic observations of surface wind effects on Multi-Angle Snowflake Camera measurements

Abstract

Ground-based measurements of frozen precipitation are heavily influenced by interactions of surface winds with gauge-shield geometry. The Multi-Angle Snowflake Camera (MASC), which photographs hydrometeors in free-fall from three different angles while simultaneously measuring their fall speed, has been used in the field at multiple mid-latitude and polar locations both with and without wind shielding. Here we show results of computational fluid dynamics (CFD) simulations of the airflow and corresponding particle trajectories around the unshielded MASC and compare these results to Arctic field observations with and without a Belfort double Alter shield. Simulations in the absence of a wind shield show a separation of flow at the upstream side of the instrument, with an upward velocity component just above the aperture, which decreases the mean particle fall speed by 55%(74%) for a wind speed of 5 m/s(10 m/s). MASC-measured fall speeds compare well with Ka-band Atmospheric Radiation Measurement (ARM) Zenith Radar (KAZR) mean Doppler velocities only when winds are light (<5 m/s) and the MASC is shielded. MASC-measured fall speeds that do not match KAZR measured velocities tend to fall below a threshold value that increases approximately linearly with wind speed but is generally <0.5 m/s. For those events with wind speeds <1.5 m/s, hydrometeors fall with an orientation angle mode of 12 degrees from the horizontal plane, and large, low-density aggregates are as much as five times more likely to be observed. We conclude that accurate MASC observations of the microphysical, orientation, and fall speed characteristics of snow particles require shielding by a double wind fence and restriction of analysis to events where winds are light (<5 m/s). Hydrometeors do not generally fall in still air, so adjustments to these properties' distributions within natural turbulence remain to be determined.