Answers are grounded in this paper and cite their sources.
An ice cover does not form everywhere a river is cold; it forms where the river is slow. This paper evaluates the two classical formation criteria for a wide reach of the upper St. Lawrence type — the floe-scale Froude condition for pans to juxtapose rather than submerge, and the lateral growth of border ice from the banks — and maps them onto a stated cross-section. For pans 0.30 metres thick, the underturning criterion — a densimetric Froude number of 1.2 on the floe's submerged thickness — gives a critical surface velocity of 0.59 metres per second. The reach's centreline runs at 0.96 metres per second, a floe Froude number of 1.93, comfortably above the threshold, so mid-channel pans arriving there cannot rest; they underturn, dive and either deposit downstream or thicken the cover's rough underside. But the cross-section's velocity distribution leaves 22 percent of the width — about 130 metres against each bank — slower than the critical value, and it is there that the cover is born, as border ice that then encroaches at a couple of metres a day and narrows the open lane to 941 metres, a lane far too wide for the banks to close unaided in one season, which in practice is finished by a single arriving field of pans once the lane's own velocity falls, or by an ice boom that manufactures slow water.
No takes yet. Share an insight, caveat, or question.
William Alexander Aldred (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: