On a northern river or lake, the load that decides whether bank protection survives is often not the design flood. It is ice, and it does things flowing water does not.
The four mechanisms
1. Ice lifting
Ice forms in contact with the bank and freezes onto whatever it touches. When the water level then rises — a thaw, a rain event, a reservoir operation — the ice sheet lifts, and it takes what it is frozen to with it.
Units are then carried away when the sheet moves or breaks up, and they are deposited somewhere else entirely.
This mechanism removes armor that no hydraulic condition at the site would move, because the force is vertical and it is applied by buoyancy across a large bonded area.
2. Ice shove
Wind and thermal expansion drive an ice sheet shoreward. The forces are large and applied horizontally at the ice surface level, and they can push material up the bank, pile it, and ride over structures.
Ice shove is capable of moving substantial objects and of pushing ridges of material well above the normal shoreline. On large lakes it is a recognised design load.
3. Abrasion
Ice moving against a surface wears it. Repeated across many freeze-thaw cycles and many seasons, abrasion degrades:
- Wire, in gabion baskets, which is the mechanism that makes them a poor choice in ice environments
- Coatings of all kinds
- Fabric and mesh, where exposed
- Concrete surfaces
4. Impact
Ice floes during breakup arrive with mass and velocity. Impact damages what it strikes and, on a river, breakup can be the most severe event of the year.
Ice jams add a further problem: an accumulation that blocks the channel, raising water levels sharply upstream and, when it releases, producing a surge downstream with velocities well beyond the normal regime.
Why the damage sits in a band
Ice damage is concentrated at the winter water level, which is frequently not the level a summer site visit shows and not the level the treatment was designed around.
That produces a characteristic signature: a band of damage part way up the bank, with sound protection above and below it. If you are inspecting a treatment and find the damage in a horizontal band rather than at the toe, ice is the likely agent rather than flow.
On a reservoir with managed levels the band can be wide, because the winter level itself moves.
What a design can do
Assume ice is a design load where it occurs, rather than treating it as an occasional nuisance. It frequently governs.
Prefer units that resist being carried. Heavier individual units bonded to a smaller proportion of the ice sheet are harder to lift than a light, continuous, high-surface-area system.
Avoid systems whose weak point is thin metal. Gabion wire is attacked directly by abrasion and by the mechanical work of ice movement, which is why the case against baskets is specific to wet, abrasive conditions — see when a gabion is still the right answer for where they do suit.
Design for repairability. Ice will remove some units in a bad year. A system where the repair is replacing individual units is in a much better position than one where the repair is a section of continuous structure.
Extend the treatment above the ice band. Protection that stops below the winter level leaves the ice working on unprotected bank.
Consider a flatter slope, which gives ice shove less to push against and lets it ride up rather than load a face.
Expect an inspection and maintenance cycle. Annual spring inspection after breakup, with a small allowance for replacing displaced units, is a realistic maintenance model in an ice environment and a more honest one than assuming no losses.
Where bagged armor sits
Reasonably well, with a caveat.
In its favour: individual units, so damage is local and repair is incremental; no wire to abrade or corrode; conforming placement so units re-seat as the bed adjusts; units can be added where losses occur.
The caveat: the mesh is a fabric, and ice abrasion works on fabric. We publish which test methods back which claims rather than measured values, and we do not hold citable ice-abrasion data. What the answer bank does address is what changes at sites exposed to heavy ice or floating debris, and it is a question worth asking directly rather than assuming.
Where ice is severe and persistent, that is a conversation to have before specifying, and when not to use rock bags is the standing statement of what our data does not cover.
Inspection after breakup
The most informative inspection of the year:
- Missing units, and where they went — displaced units downstream tell you the direction and force
- Damage in a band at the winter level
- Abrasion, particularly on the exposed faces
- Ice ridges and pushed material above the shoreline, which record how far the shove reached
- Bank damage above the treatment, indicating the protection stopped too low
- Level survey, since ice can rework the bed as well as the bank
Where to go next
- How to read a failing bank for mechanical damage as a mechanism
- Lakefront erosion for the lake case
- When a gabion is still the right answer for why wire suffers here
- Why scour protection fails for the other failure modes
- When not to use rock bags
- The erosion control hub for the rest
If ice is a factor on your site, say so early. Photographs after breakup and a note on the winter water level are the useful things to send through the quote form.