Quantifying suspended frazil ice using multi‐frequency underwater acoustic devices

Abstract

An intensive frazil ice field sampling campaign was undertaken at the Port of Quebec on the St. Lawrence River from February-March 2009. Two underwater acoustic instruments set at different frequencies of 420 and 1228.8 kHz were used to detect frazil ice in the water column. In this paper, frequency inversion methods are presented and subsequently applied to the observations to estimate frazil ice characteristics and concentration. Using inversion methods, most of the detected ice crystals had estimated radii of 0.06-0.18 mm. This range compares favourably to the estimated value of 0.20 mm obtained by analysing the Rouse number related to the vertical distribution of the frazil crystals. The results were in contrast to a previous study of frazil ice at another site in the St. Lawrence, which reported radii tens of times larger . The tiny crystals observed here were of similar size to those observed in laboratories (e.g. 0.09 mm), suggesting that the particles formed locally. Frequency analyses were also used to estimate the volumetric suspended frazil concentration, which appeared to be on the order of 6 ppm. Based on evidence suggested by the data and the volume backscattering coefficient at the two frequencies, this study also presents the complex sequence of processes that occurs during a typical supercooling frazil event. This paper concludes with future directions for research using acoustic instrumentation for further understanding of frazil ice dynamics.

Modeling frazil ice growth in the St. Lawrence River

“…As for previous calculations, the same size distribution for the frazil ice particles was used (i.e., log-normal) along with the same assumed shape (perfect discs of constant diameter to thickness ratio). Using this size distribution, most of the particles have radii measuring between 0.02 and 0.2 mm, which is similar to size classes the sonar is likely to have detected (i.e., 0.06 to 0.18 mm; Richard et al 2011). Figure 8 shows only data during supercooling periods and during which time anchor ice did not blind the transducer.…”

“…The instrument was setup to measure vertical profiles (646 layers) of the reflected acoustic energy. The measured echoes were used to calculate volume backscattering coefficients using the procedure in Richard et al (2011). Each vertical profile was recorded at an interval varying between 1 and 20 min, and over 116 000 profiles have been measured.…”

“…5E presents acoustic measurements made by the acoustic profiler. Measurements are shown by means of the depthaveraged and time-averaged backscattering coefficient ( v , which is proportional to concentration -see Richard et al 2011). The measurements are completely independent from the calculated results shown in other figure parts.…”

“…The measurements are completely independent from the calculated results shown in other figure parts. It is believed that most of the reflected acoustic energy measured by the sonar is caused by frazil ice for the following reasons: (1) other suspended particles, such as sediment or biomass, are probably too small to be detected by the sonar (i.e., particles with radii smaller than about 0.2 mm are believed to be undetected by the sonar -see discussions below and Richard et al 2011); and (2) the reflected acoustic energy is virtually zero when the water is not supercooled (i.e., in between frazil ice events, when there is no anchor ice buildup on the transducer). Figure 6 shows the time series of some of the same variables, over a shortened period of time, for the event #1.…”

Laboratory measurements of acoustic backscattering from polystyrene pseudo-ice particles as a basis for quantitative characterization of frazil ice

“…Recent river ice research efforts have included several studies (Jasek et al, 2005;Morse and Richard, 2009;Marko and Jasek, 2010a,b,c;Richard et al, 2011and Ghobrial et al, 2012, 2013 directed at quantifying frazil suspensions from acoustic backscattering (ABS) data. Although the reported results offered previously unavailable insights into frazil dynamics and properties, underlying simplifications and approximations left considerable interpretative uncertainties.…”