Lake Ice Out and Ice In Dates, State by State

It is the day in spring when the winter ice cover finished leaving a lake. The exact test differs from lake to lake and always has. Maine defines it for its official list as the point where a boat can navigate unimpeded from one end of the water to the other. Lake Winnipesaukee in New Hampshire defines it as the moment the M/S Mount Washington could run between its five ports. Minnesota DNR uses the lake being completely free of ice. Those are genuinely different thresholds, which is why comparing a lake against its own past is far more meaningful than comparing two lakes against each other.

No. Every figure here is the date ice finished leaving a lake, or froze over it, in a past year. Ice strength varies across a single lake on a single day with depth, current, springs and snow cover, and no date in this data measures any of that. For anything involving going out on ice, contact your state natural resources agency and look at the water yourself.

Because eight states are all that have a lake ice record deep enough. These pages cover every United States water body in the international ice phenology database with at least 40 recorded years, 116 of them across nine states. The ninth is Alaska, whose single record is the Tanana River at Nenana rather than a lake, so it is named on this page rather than given a page of its own. The rest of the country either does not reliably grow lake ice or has never had anyone write the date down.

The oldest observation in this data is Sebago Lake in 1807. Of the 116 water bodies here, 50 have a record spanning a century or more, and the whole set carries 14,221 individual dates. Almost none of it was collected for science. It was kept by water companies, power companies, newspapers, town officials, lake associations and individuals, for entirely practical reasons, and only became a climate record afterwards.

In this data, overwhelmingly yes. Of the 78 lakes whose record spans at least 80 years, 73 show ice out moving earlier, with a median shift of -6 days per century. Freeze up shows the same thing from the other side: of the 23 lakes with a long enough freeze record, 22 show the lake closing over later. These are least squares lines through observations, which is the standard description in the ice phenology literature. They describe what the record did. They are not projections.

It means exactly that, and the good datasets record it rather than leaving a blank. Grand Traverse Bay has 44 winters on record in which it never froze across, the most of any water body here. Those years matter: if you drop them and average only the years the lake did freeze, you quietly bias the whole record toward an earlier thaw, because the winters warm enough to prevent freezing are exactly the ones that would have produced the earliest ice out.

From the state agency. Maine's Bureau of Parks and Lands posts ice out for more than 150 waters as reports come in. Minnesota DNR maps ice out and ice in reports statewide. The Wisconsin State Climatology Office publishes the Madison lakes every winter. The National Weather Service in Burlington records the day Lake Champlain closes over. Each state page links its own source; these pages hold the long record.

The record is historical, ending in 2020 for most lakes, so it describes past winters rather than predicting the next one. Each state page shows how the current date has played out across the recorded years, and the state agency linked on every page publishes the current season as reports arrive. Nothing on these pages describes ice condition or ice strength; for anything involving going out on ice, the state agency is the source.