What Goes In, What Comes Out
Eighteen gauged rivers, a few hundred ungauged brooks and the rain that falls on the surface go in. One river comes out — the Richelieu, north into Quebec, past a gauge at Fryers Rapids that has been counting since 1937. Between those two numbers you can weigh the lake, close the budget, and say how long a drop of water stays. None of it is looked up here; all of it is computed, and the arithmetic checks itself.
The budget closes itself
That is the check. Nothing was tuned to make it come out: the eighteen rivers are measured, the outlet is measured, and the difference between them lands on a number that a hydrologist would have guessed for the country it drains. If the residual had come out at twice the gauged runoff, or negative, something here would be wrong.
Two point one years, not three
The figure usually quoted for Lake Champlain is three years. It is not obvious where that comes from; three years would need an outflow of about 275 cubic meters a second, and the gauge has averaged 362 over eighty-eight years and 395 over the last thirty. Both numbers here are shown with their inputs so you can see which you prefer.
The seasons
Mean outflow by month across eighty-eight years. Four fifths of a normal year's variation is snow: the lake empties in April and May and barely moves in September.
It is not one bucket
A residence time for the whole lake hides the thing that matters. Sixty per cent of the water is in one deep trough down the middle; the bays at the north-east corner hold almost none of it. Segments here are the monitoring program's own — the same names its cyanobacteria reports use — grown out from its shoreline stations by a walk through the water, so a label cannot cross a headland the way a straight-line partition would.
| segment | km² | km³ | % of the water | mean depth |
|---|