Optoelectronic Simulation of the PMS 260X Optical Array Probe and Application to Drizzle in a Marine Stratocumulus

Abstract

Measurement of drizzle drop sizes and concentrations are often made using optical probes with linear arrays. Their accuracy is affected by both diffraction of light and response times of the electronics connected to the sensor photodiodes. In this study the behavior of the Particle Measuring Systems, Inc. (PMS), 260X probe (a 62-photodiode system) is studied by calculating the two-dimensional Fresnel diffraction patterns for opaque disks and averaging this optical signal onto areas corresponding to the sizes of the photodiodes. The two-dimensional optical field is then transformed to a one-dimensional optical time-varying field; this simulates the passage of a drop past the photodiode array when the probe is mounted outside an aircraft. The optical field is convolved with an electronic response function to simulate the variation of the electronic signal as a drop passes through the diode array. For aircraft speeds less than 120 m s?1, one component of the probe sample volume, the depth of field, is found to exceed the manufacturer's values by up to 50% for larger drops (drop radius R > 35 ?m), whereas for smaller drops (20?35 ?m) the depth of field may be a small fraction of that suggested by the manufacturer. For higher airspeeds, the depth of field tends to be smaller than the manufacturer's values. The present results also show depth of field values that may be more than three times larger than those found in a previous theoretical study using Fresnel diffraction. Examination of observations from a marine stratocumulus flight with high drizzle rates reveals a problem with the probe. The optoelectronic model predicts that the 260X probe should not be able to register counts in the two smallest-sized bins. Yet the counts in bins 1 and 2 constitute 96%?99% of all the drop counts for 33 1?min segments in cloud or drizzle below cloud. The high number of counts correlates with the mean-volume radius of drizzle drops. One possible explanation that is consistent with the measurements is that counts in bins 1 and 2 occur when large drizzle drops impact on the probe tips, followed by small droplet fragments moving through the laser beam at reduced speed. Measured drop spectra in the 19?23.5-?m range, using the new calibration, do not match well with another sensor that measures cloud droplets. This suggests that drizzle drop breakup may also contribute artificial counts to bins 3 and larger.