Every spring we get the same call. The ice has gone out, the owner has walked down to the water for the first time since October, and something is wrong. The dock has a slope it did not have before. One piling stands proud of its neighbours. A crib has moved a foot toward shore. Boards have lifted at the fasteners.
Almost none of this is storm damage. Almost all of it is ice, and the mechanisms are specific and predictable enough that you can design against them.
What ice actually does to a structure
Start with the physical fact underneath everything else: water expands by roughly 9% when it freezes. A bay that skims over at three inches thick is not simply water that has gone solid — it is water that has become meaningfully larger and has nowhere to put the extra volume except sideways, into whatever is at the edge.
On Lake George that edge is your shoreline, your cribs, and your pilings. And the lake freezes in most winters: full ice cover has occurred in roughly 70% of years since 1980, typically arriving around 19 January, and lasting about 76 days when it happens. Design as though it will freeze, because in most years it does.
From there, three distinct mechanisms do the damage.
1. Sheet push (static expansion)
The simplest one. The sheet forms, it is larger than the water it replaced, and it presses outward against every fixed object at its perimeter. Loads are substantial and sustained across weeks. Structures that resist by strength alone — a slender piling, a lightly built crib — lose. Structures that resist by mass and friction generally win.
2. Thermal ratcheting (the cycle that actually moves things)
This is the mechanism most owners have never heard of, and it is the one that walks structures across a winter.
An ice sheet is not static. On a warm afternoon it expands and pushes shoreward. On a cold night it contracts — but it does not pull the structure back with it, because ice is far better in compression than in tension, and because water freezes into the gap that opens up. The next warm day, the sheet is a little larger again and pushes from a new starting position.
Repeat that for sixty or seventy days and you have a ratchet. Each cycle moves the structure a few millimetres, none of which is observable, and by ice-out a crib has travelled several inches or a foot. This is why the damage accumulates rather than appearing in one dramatic event, and why owners so often say "it was fine last year" about a structure that has been moving for five.
3. Ice jacking (the piling killer)
The single most common failure we are called to on Lake George.
Ice freezes onto a piling — bonds to it, all the way round. Then the water level rises, or the sheet lifts on a thaw, and the ice takes the piling with it. The bond between ice and timber is strong enough to overcome the friction holding the piling in the lake bed, and the pile is drawn upward out of its embedment.
When the ice melts, the piling does not go back down. It stays where the ice left it, now with less embedment than it had, supporting a structure that is no longer level. And because it has less embedment, next winter it jacks more easily. The failure is self-accelerating.
Why Lake George is harder than most lakes
Three local factors compound the general problem.
The winter drawdown. Lake George is drawn down over the winter, lowering the surface by several feet from summer level. That means the ice sheet forms at one elevation and the water underneath it changes — producing exactly the vertical movement that drives ice jacking, and leaving structures partially unsupported at moments in the cycle.
The bottom. Much of this shoreline is ledge, glacial boulder and hardpan under a thin veneer of sediment. A pile driven into that will often refuse early, achieving nominal embedment rather than real embedment. It looks fine in August. It jacks in February.
Fetch. The broad water south of Bolton Landing gives wind a long uninterrupted run, which affects how the sheet forms, where it cracks, and how much it moves. Exposed shorelines take a genuinely different loading regime from protected coves, and they should not be built the same way.
How to build for it
The good news is that this is a solved problem. Structures on this lake have survived a century of winters. The solutions are not exotic; they are just consistently applied.
Resist with mass, not with strength
A timber crib works because it is heavy, broad and full of stone. It does not try to be stronger than the ice — it is simply too much bother to move. Generous plan dimensions and well-graded stone fill do more for winter survival than any amount of over-specified hardware.
This is the core reason cribs dominate Lake George while pile docks dominate softer-bottomed lakes elsewhere. It is not tradition. It is the bottom and the ice.
Chase real embedment
Where pilings are the right answer, they need genuine depth in material that will hold them. Stopping at refusal on a boulder is not embedment — it is a pile resting on a rock, waiting for the ice to lift it. This is why we sound and probe before committing to a pile design, and why we sometimes recommend cribs on a site where the owner expected piles.
Detail the shore transition deliberately
The point where the dock meets the land takes the worst of the sheet push and the worst of the spring runoff. It is where a very high proportion of the failures we repair actually began. It needs to accommodate movement rather than resist it absolutely, and it needs to shed water rather than trap it.
Make components replaceable
Decking fastened so that individual boards can be lifted and replaced without dismantling a bay. Hardware that is hot-dip galvanised or stainless throughout, because a fastener that fails at the waterline takes the frame with it. Connections that can be inspected. A dock is a structure you will maintain for decades; build it so maintenance is possible.
What you can do each autumn
Design decisions are made once. Seasonal decisions are made every year, and they matter almost as much.
- Remove everything removable. Floating sections, swim platforms, ladders, boat lifts and cradles, canopies, furniture, and any hardware that unbolts. Anything left in a shallow slip is at the mercy of the sheet.
- Have the structure inspected before ice-in. A marginal piling or a crib that has started to spread will not survive being loaded by ice. Finding it in October is a repair; finding it in April is a rebuild.
- Keep the shore transition clear of debris and vegetation that will freeze into a solid mass and transfer even more load.
- Clear snow from boathouse roofs. Different mechanism, same season — heavy snow and ice accumulation causes collapses and leaks every winter on this lake.
A word on de-icers and bubblers
De-icers keep water moving so ice cannot form and grip a structure, and they do work. They also create open water and dangerously thin ice on a lake with substantial winter recreation traffic — ice fishing, skating, snowmobiles. That is a decision affecting more people than you. If you are considering one, talk to us and to your neighbours, mark the area clearly, and treat it as a serious choice rather than a purchase.
Spotting damage before it compounds
At ice-out, walk the structure and look for:
- A piling standing proud of its neighbours — the signature of jacking
- A dock out of level, or twisted along its length
- Decking opened up, cupped, or split at the fasteners
- The whole structure shifted shoreward, with a gap at the water end
- Cracked framing at connections, particularly in a bay restrained at both ends
- Stone that used to be inside a crib now lying on the bottom beside it
The economic argument for catching these early is stark. A piling jacked three inches in one winter can be reset in a day. The same piling after three winters has lost most of its embedment, has pulled its neighbours out of alignment, and has been transferring load into framing that is now cracked. We have rebuilt entire docks that would have been a two-day repair if someone had looked at them thirty months earlier.
The short version
Ice damages docks in three ways: it pushes because it is bigger than the water it replaced, it ratchets structures out of position through daily thermal cycling, and it jacks pilings vertically out of the lake bed by freezing onto them and lifting. On Lake George, the winter drawdown and a rocky bottom make all three worse.
You build against it with mass rather than strength, real embedment rather than nominal embedment, a shore transition designed to move, and components you can replace. You manage it by removing everything removable each autumn and inspecting twice a year.
Done properly, a Lake George dock is a thirty-to-fifty-year structure. Done as though this were a warm-water lake, it is a ten-year structure that spends most of that decade being repaired.
Lake George Docks
We design, permit and build docks, boathouses, cribs and shoreline structures across all 32 miles of Lake George, New York — and repair what the lake does to them. Questions about your own shoreline? Call (518) 260-9878 or send us the details.
