Our research group owns, operates, and delivers meteorological data from our University of Utah MesoNet (UUNET) to the National Weather Service. We test and experiment different sensor types, manufacturers, and methodologies to stay current with commercial technology and research applications.
We started testing an alternative radar precipitation sensor at our Mountain Meteorology site located at the UofU campus in early 2023, and have had promising results.
An article was recently published in the American Avalanche Association's (A3) industry journal, The Avalanche Review (TAR), about this sensor and some early results. A3 members and subscribers, check your mailbox. For non-members, a digital version of TAR will be available later this year (https://www.americanavalancheassociation.org/tar). For everyone else, enjoy this blog post highlighting a storm event 07-Feb-2024.
What's this all about? A new radar-precipitation gauge has an onboard heater to keep the sensing surface snow-free. A recent storm dumped over 20mm of snow-water-equivalent (SWE) in a 12-hour period, overwhelmed the heater, and resulted in a 15mm 'missed' precipitation event. Offset sensors helped us catch the event, and we took an IR gun up the next day to investigate:
Top Left: IR image (next day) showing the active heat surface of the sensor.
Top Right: snow capped sensor during storm period
Bottom: 30 day cumulative totals from three precipitation gauges.
Where are we?
Our Mountain Meteorology Lab is located on the eastern-edge of campus at the mouth of Red Butte Canyon:
The outdoor laboratory includes a variety of air quality, remote sensing, and meteorological measurements operated by University and industry partners. Realtime data available here: MesoWest - MTMET
This post will focus on the following sensors used to measure liquid precipitation and snow-water-equivalent (SWE) of solid precipitation.
- A heated tipping bucket owned and operated by Salt Lake County
- Antifreeze/bucket weigh-gauge (UUNET)
- Radar precipitation device owned and operated by our group, the WS100
The radar device was plugged into shore power (110 VAC) with 24 VDC regulation for the heater system. It communicates via SDI-12 into a datalogger enclosure (bottom right).
Video showing the heater surface is clear, unfortunately with a doughnut "cap" that skews the measurement. The video was recorded near the tail-end of the storm, 07-Feb-2024 16:00 local time:
The 12-hour plot below ends at 3pm local, about an hour before the video was taken. The plot shows cumulative precipitation for each of the sensor types.
Note the lower chart after 08:00 when the black line (WS100) starts to under report relative to the Pluvio and tipping bucket. The storm started warmer with sleet and mixed precipitation particles, before a transition to solid (snow) and cooler temperatures. This could have formed an ice cap on the device or snowfall intensity + duration could have formed a snow pillow on the device, masking the signal, and under reporting through the remainder of the storm.
T
These graphics are generated in real-time and available here: MTMET - Precip Comparison
Visual analysis (printing, drawing straight lines, and calculating slopes) yielded 2-4 mm/hr of precipitation during the two pulses of this storm cycle (based on the Pluvio gauge). 8am to 3pm is 7 hours, 7 hours * 2 mm/hr is 14 mm total. 7 hours * 3 mm/hr is 21 mm total, matching the total cumulative as of 3pm local. Micro-intensities would have exceeded 3 mm/hr as a quiet period exists around lunchtime (11am to 1pm):
The chart below show air temperature, relative humidity and dew point. The A and B storm pulses are shown in the lower precipitation chart (Pluvio data). The dew point depression narrows around 8am as the storm system hit the station.
The winds shifted from 'calm', north-easterlies, to light westerlies. Our anemometer also rimed and was frozen during the 10am "A" pulse and the 2pm "B" pulse. Graphics via Synoptic Data Viewer - MTMET
The next day had similar storm conditions, and we noted similar under reporting (missed precipitation) from the radar device while only 2mm of SWE occurred. This time, we brought an infrared (IR) gun to investigate what part of the radar-gauge is heated and tried to catch it in the act.
As we arrived around 11am local time 08-Feb-2024, any ice cap or snow doming on the sensor had melted and evaporated by the heater. It's interesting to see the IR gun shows graupel particles that have fallen on the sensor:
Ultimately, we continue to investigate measurement, reporting, and data visualization techniques to understand these events.
Please contact the author or the Horel research group if you have any questions, ideas, or comments. Happy to chat more.
Thank you,
-Colin
.jpg)
