Clear ice about 10 cm thick bears one person on foot, 13-18 cm bears a snowmobile and 23-30 cm bears a small car, but a working crew needs at least 20 cm, because harvesting puts several people and 27 kg blocks on the sheet for hours and cuts an open channel through it. White or snow-formed ice needs double the clear-ice figure, so judge the colour as well as the thickness, test in many places rather than at the shore, and spread the crew out. Gold’s formula, P = A x h squared with P in kilograms and h in centimetres, puts about 1,400 kg of static load on a 20 cm sheet.
Part of Ice Houses and Cold Storage
Ice thickness requirements are the minimum depths of sound ice at which people, sleds and equipment can safely work on a frozen water body, along with the conditions that make those depths unreliable.
Thickness Alone Does Not Make Ice Safe
The ice-houses parent article sets a floor of 20 cm for harvesting and says to test thickness by chopping a hole, to check several spots and to work in pairs. This page explains where such figures come from and what can make a 20 cm reading wrong.
Safe thickness depends on four things together: how thick the ice is, what kind of ice it is, what load it carries, and what has happened to it since it formed.
What You Need
- An ice chisel or auger and a tape — to drill or chop test holes and read the thickness at each.
- A pole — to probe ahead and feel soft or thin ice.
- The rescue kit — see Ice Harvesting.
Method: Judge The Ice In Five Steps
Step 1: Look at the colour. Clear, hard, blue-black ice formed from freezing water is the strongest. White or opaque ice, formed from slush or snow soaking into the surface, is much weaker. Grey, honeycombed or slushy ice is weaker still. Minnesota’s ice-safety guidance gives white ice as roughly half as strong as clear ice. Cloudy ice also melts faster in storage.
Step 2: Measure the thickness at test holes. Test many points, not only at the shore. Ice is thinner over currents, springs, inlets, around rocks, logs and reeds, and under snow drifts.
Step 3: Read the thickness against the load. Use the figures below, and remember they apply to new, clear ice.
| Load | Minimum clear ice | On white or opaque ice |
|---|---|---|
| One person on foot | about 10 cm | about 20 cm |
| One person standing in place over 2 hours | about 15 cm | about 30 cm |
| A working crew harvesting ice | at least 20 cm | at least 40 cm |
| Snowmobile or similar light vehicle | about 13-18 cm | about 26-36 cm |
| Small car or light pickup | about 23-30 cm | about 46-60 cm |
Sources for the table. The clear-ice figures are the Minnesota Department of Natural Resources ice-thickness poster (ENF_0006_2023), which gives 4 in on foot, 5-7 in for a snowmobile and 9-12 in for a car, for new, clear ice only. Ontario’s guidance gives 20 cm of clear blue ice for a snowmobile and 30 cm or more for a light vehicle — noticeably more conservative than Minnesota for the same loads, which is itself worth knowing.
No authority publishes a figure for ice harvesting, so the 20 cm working floor is derived rather than quoted. Health Canada gives 15 cm for skating alone and 20 cm for groups; Alberta’s guide for workers raises walking from 10 cm to 15 cm for anyone standing in one place over two hours. Harvesting is a group, standing for hours, hauling 27 kg blocks, and cutting an open channel that removes the sheet’s continuity. Every one of those pushes above the lone-walker figure, so the group figure is the right starting point and 20 cm is the floor. 25 cm is better, and is what Ice Cellar asks for, because ice thinner than that also cracks while being cut.
Step 4: Judge the conditions. Do not rely on the number if the ice has cracked, been thawed and refrozen, been covered with water, or has a layer of slush. A warm spell or direct sun cuts strength quickly, and ice that has been weakened looks the same on the surface.
Step 5: Re-test through the job. Ice changes. Test again after any mild day or rain, and stop if it deteriorates.
Worked Example: Gold’s Formula And A Bunched Crew
A rule given in the ice-engineering literature (L. W. Gold, 1971, Canadian Geotechnical Journal) is that the load a floating sheet of good clear ice can bear scales with the square of its thickness: P = A x h squared, with P in kilograms, h in centimetres and A about 3.5 for clear ice in good condition. This is recalled, not checked against the paper, so treat it as a guide to comparison rather than a design value.
- 10 cm: 3.5 x 10^2 = 350 kg.
- 15 cm: 3.5 x 15^2 = 3.5 x 225 = 788 kg.
- 20 cm: 3.5 x 400 = 1,400 kg.
- 25 cm: 3.5 x 625 = 2,188 kg.
Doubling the thickness quadruples the load (1,400 / 350 = 4).
Compare with the loads on the job. A worker with tools is about 100 kg. A sled with six blocks of 27 kg each is 6 x 27 = 162 kg, plus the puller (80 kg) and the sled (say 20 kg), a total of 262 kg. Five workers bunched together are 5 x 100 = 500 kg.
- A bunched crew of 500 kg on the 20 cm floor (1,400 kg) has a margin of 1,400 / 500 = 2.8.
- The same crew bunched on 15 cm (788 kg) would have only 788 / 500 = 1.6 — which is the single clearest reason the floor is 20 cm and not 15.
- The same crew spread across the sheet loads each patch far less. Spreading out raises the margin at no cost.
White ice has about half the strength constant. Halving A to 1.75, the thickness carrying the same load rises by the square root of 2, about 1.4 times — so the algebra says 20 cm of clear ice becomes 28 cm of white ice.
Do not use that number. Double the thickness instead: 40 cm. Both the Minnesota DNR poster and Ontario’s guidance say to double the clear-ice figure for white or opaque ice, and the published rule is deliberately more conservative than the square-root result. The reason is that the constant itself is uncertain: Barrette’s review for the National Research Council of Canada records that Gold’s 1971 assumption about white-ice strength was never experimentally verified, and that the various national guidelines disagree enough to produce large differences in the loads they permit. Barrette’s own laboratory work found flexural strength falling to roughly half near melting.
When the strength constant is the uncertain quantity, the margin belongs on the safe side of it. The algebra is here to show why white ice is weaker, not to set the working figure.
Common Mistakes
- Testing only near the shore — shore ice is not representative.
- Trusting the number over the ice — slush, cracks and thaw cycles defeat a thickness reading.
- Working in a bunch — the load concentrates. Spread out.
- Forgetting the cut — each saw channel leaves a weak line and open water.
- Reusing yesterday’s reading — a mild afternoon changes the ice.
Safety
Cold water disables your hands before it cools your body
After falling through, the first minute is a gasp reflex that can pull water into the lungs. After that, you have roughly ten minutes of useful movement, and your hands lose grip sooner, which is why self-rescue fails so often. Each worker must carry two ice picks on a cord, there must be a person on the shore with a throw-line, a ladder or long plank, and a warm shelter with dry clothes, and no one walks out to a hole to help. After rescue, rewarm the person gently in shelter. See Hypothermia.
What’s Next
- Learn next: Cutting Tools — the tools that go on the ice
- Combine with: Ice Harvesting — the full harvest procedure