Precipitation Imaging Package (PIP): A Potential Instrument for Measuring Raindrops

Precipitation Imaging Package (PIP): A Potential Instrument for Measuring Raindrops

Sloane Poppei1, Ali Tokay2,3, Charles Helms2,4

1Department of Climate and Space Sciences and Engineering, University of Michigan, Ann Arbor, MI, USA

2Mesoscale Atmospheric Processes Laboratory, NASA, Goddard Space Flight Center, Greenbelt, MD, USA

3Goddard Earth Sciences Technology and Research II, University of Maryland Baltimore County, Baltimore, MD, USA

4Earth System Science Interdisciplinary Center, University of Maryland, College Park, College Park, MD, USA

The importance of raindrop size distribution in remote sensing precipitation retrieval algorithms and regional and global weather forecasting models is well recognized. However, despite advances in video disdrometers, the accurate measurement of raindrop size distribution from drizzle drops (0.2 mm) to ultra large raindrops (9.3 mm) by a single instrument is still an active area of research. The Precipitation Imaging Package (PIP), a camera-based disdrometer developed by NASA, has traditionally been used for snowflake measurements but is now gaining recognition for its potential in raindrop analysis. The PIP is a reliable, cost-effective, and low maintenance instrument that makes it a promising tool not only for research but also for operational use.

This study evaluates the performance of PIP in rainfall measurements leveraging data from NASA’s Global Precipitation Measurement (GPM) Ground Validation Program, collected during the winters of 2023 and 2024 in Storrs, Connecticut. Special attention is given to the comparison between two PIP processing algorithms and their handling of spurious raindrops—particles inaccurately identified as raindrops due to measurement anomalies. The PIP standard algorithm results in approximately 25% of the particles tracked twice or more. This low percentage of tracking forces PIP algorithm to process size and fall velocity matrices independently. Thus, the spurious raindrops could not be removed with fall velocity-based screening as it has been done for other optical disdrometers (e.g. PARSIVEL). The standard algorithm also creates a subset of particles known as ap60, where particles are only recorded from every 25th frame to avoid multiple measurements of the same particle. The newly constructed PIP algorithm by co-author, Dr. Charles N. Helms (CNH), tracks two-thirds of the total particles twice or more. Thus, most of the spurious particles are screened out based on thresholding where particle fall speed is outside the ±50% of its terminal fall speed. This is a significant improvement from the standard PIP algorithm.

Despite these advancements, the PIP tends to overestimate total rainfall, as indicated by comparisons with reference gauges. Th overestimation is likely due to the accumulation of spurious particles on the PIP camera lens, as observed through 10-minute average heat maps of particle occurrence. These anomalies, which lead to erroneously high rain totals, underscore the need for ongoing refinement in PIP’s data processing techniques. The CNH terminal velocity method was found to be the most consistent with reference measurements, but unable to screen all spurious particles. This discrepancy highlights the need for further statistical analysis and processing enhancements to mitigate the impact of spurious data on overall rainfall estimates.