Date of Award

January 2015

Document Type


Degree Name

Master of Science (MS)


Atmospheric Sciences

First Advisor

Matthew Gilmore


Snowbands can produce locally larger snowfall accumulations as well as reductions in visibility thereby being hazardous to vehicles and aircraft. The study herein is the first to combine multi-Doppler retrieved winds, in situ snow crystal size distributions, and polarimetric radar variables within snowbands for two radar wavelengths. Data includes two polarimetric radars: Doppler on Wheels (DOW) – a mobile X-band polarimetric Doppler weather radar – and the University of North Dakota (UND) polarimetric C-band radar (hereafter: “UND radar”). Also used in this study are data from the two-dimensional cloud (2DC) probe attached to the UND’s Citation II weather research aircraft. Retrieved wind velocities, from dual-Doppler analysis, and dual polarization radar variables, are matched to the aircraft’s transect location and 2DC probe images inside and outside the snowband.

Regarding kinematics, upward motion in both the retrieved vertical wind and aircraft-measured winds is seen generally west of the DOW location with downward motion generally east. The dual-Doppler retrieved horizontal winds also show easterly flow at lower altitudes and westerly at higher altitudes, consistent with a sounding from Bismarck, ND. These wind patterns are generally persistent in the local environment regardless of the snowband’s presence.

Ice hydrometeors, measured by the 2DC probe, are more numerous and larger inside the snowband, compared to a weaker-reflectivity snow-filled region outside the snowband. These differences in number concentrations are present at all altitudes sampled but are most distinct at higher altitudes. Along the aircraft transects, both radars observe larger average KDP values (most altitudes) and larger average HV values (all altitudes) inside the snowband.

Differences exist between the same radar variable for near-simultaneous dual radar measurements. These differences are: greater reflectivity (regardless of altitude and location) for DOW compared UND, greater average HV for the DOW radar compared to UND, closer-to-0 dB average ZDR values for DOW inside the snowband, and closer-to-0 dB average ZDR values for UND outside the snowband. These radar variable differences could be related to calibration and wavelength differences between the DOW and UND, slight differences in the sampling area, and small scale variability within the snowband.

Radar reflectivity (greater values inside the snowband) and ZDR (values closer to 0 dB inside the snowband) are consistent with the original hypothesis. However vertical velocity has similar values both inside and outside the snowband. Stronger radar reflectivity and ZDR closer to 0 dB do not coincide with stronger updrafts inside the snowband.

This snowband had unique polarimetric and hydrometeor size distribution characteristics compared to its surroundings. The characteristics inside and outside the snowbands determined from this study, could be used to improve the microphysical parameterization within forecasting models of cold season events. Better microphysical parameterization could improve the forecasted timing, duration, and snowfall amounts from snowbands, improving transportation safety and efficiency. Also, because retrieved vertical velocity does not differ significantly inside versus outside the snowband, another process is responsible for larger aggregate hydrometeors within the snowband. Another atmospheric process, such slantwise convection, could be the reason the snowbands in the study formed.

To improve upon this study, more information on the precipitation size hydrometeor characteristics is needed, in addition to surface conditions both inside and outside snowbands. To make these critical observations, future field experiments should include the following aircraft and surface-based instruments. Adding measurements from a High Volume Precipitation Spectrometer probe, the full size spectrum of precipitation-size hydrometeors could be sampled. Surface snowfall and visibility measurements both inside and outside the snowband could be used to better quantify snowband impacts at and near ground level.