Key points are not available for this paper at this time.
The formation and evolution of frazil ice during river freeze-up is a key source for anchor ice and surface ice formation, both of which significantly impact river hydrology. Anchor ice formation on the riverbed may lead to flooding and accumulations on water intake trash racks may completely block inflows to water treatment plants. Numerous previous laboratory studies have investigated frazil ice generation during classic supercooling events which occur when the upward air-water heat flux remains constant. However, frazil generation when the heat flux varies during a supercooling event, which occurs commonly in the field, has not been explored in laboratory studies. To investigate this phenomenon, a series of controlled laboratory experiments were conducted in which variations in the air-water heat flux were induced by controlling the air temperature during the experiments. Images of frazil particles and flocs were captured while the air temperature was increased or decreased by 10 °C at different times during supercooling events. Varying the heat flux during different supercooling phases led to different responses in the time series of water temperature and frazil ice properties. Increasing the heat flux raised the mean particle number concentration by 25–33 %. Decreasing the heat flux only produced a measurable effect when the change occurred early in the supercooling event, prior to significant ice formation, reducing mean particle and floc number concentrations by 10 and 22 %, respectively. Particle and floc production rates varied by approximately a factor of two when the heat flux was increased or decreased prior to significant ice formation. • Varying the air-water heat flux during supercooling affects frazil ice generation. • Increasing the heat flux increases mean particle number concentrations by 25–33 %. • Frazil production does not always cause dramatic changes in the supercooling curve. • Particle and floc production rates varied by approximately a factor of two.
Pei et al. (Tue,) studied this question.