How this instrument works
The same amount of liquid water produces very different amounts of snow depending on how cold it is when it falls. Near freezing, snowflakes are wet, dense, and compact tightly, so an inch of liquid-equivalent precipitation might only produce a few inches of snow; in deep cold, snowflakes form as light, fluffy, air-filled crystals that pile up far higher for the same amount of water. Meteorologists capture this with a snow-to-liquid ratio — how many inches of snow one inch of melted liquid water produces — and that ratio rises sharply as temperature drops.
This calculator applies a temperature-banded lookup table of these ratios rather than one fixed number, since a single average ratio (the old rule of thumb of 10:1 is a commonly cited example) glosses over how much colder conditions genuinely inflate snow totals. The bands run from a coefficient of just 0.1 for wet, near-freezing precipitation (34-45°F) up to 100 for extreme cold (below -20°F) — a thousand-fold range across the temperature spectrum, reflecting just how differently the same rainfall behaves depending on how cold the air is.
Real snow-to-liquid ratios genuinely vary by more than just surface temperature — the depth and structure of the cloud layer where the snow actually forms matters too, and any single lookup table is a simplification of a more complex atmospheric process. Treat the result as a solid ballpark estimate for planning, not a precise weather forecast.
- Enter Rainfall equivalent in inches — the liquid-equivalent precipitation amount (what a rain gauge would measure if it fell as rain).
- Enter Air temperature in °F — the temperature during the precipitation event.
- Read Expected snowfall in inches — the estimated snow accumulation, scaled by the temperature-appropriate snow-to-liquid ratio.
Worked example — 1 inch of rain at 25°F
1 inch of rainfall equivalent falling at 25°F lands in the 20-27°F band, which carries a coefficient of 15: expected snowfall = 1 x 15 = 15 inches. That's a substantial snow total from what would only be a modest 1-inch rain event — exactly the kind of jump that surprises people who assume a rough 1-to-1 or 10-to-1 relationship between rain and snow.
Deep cold changes the picture even more dramatically: 4 inches of rainfall equivalent at 3°F (the 0-9°F band, coefficient 40) produces 4 x 40 = 160 inches of expected snow — over 13 feet. And at the opposite end, right at the edge of freezing, 0.5 inches of rain at 40°F (the 34-45°F band, coefficient 0.1) produces only 0.5 x 0.1 = 0.05 inches of snow, correctly reflecting that precipitation this close to freezing barely accumulates as snow at all before melting or compacting away.
Questions
Why does the same rainfall produce so much more snow when it's colder?
Because snowflake structure changes with temperature. Near freezing, snowflakes are wet and dense and pack together tightly as they fall and land, so a given amount of water produces a relatively shallow, dense snow layer. In much colder air, snowflakes form as light, highly branched, air-filled crystals that don't compact as readily, so the same amount of water spreads out into a far deeper, fluffier layer. That's the entire reason a snow-to-liquid ratio has to vary by temperature rather than being one fixed number.
What's the old "10 inches of snow per inch of rain" rule of thumb?
It's a widely repeated average approximation — some regional climate studies suggest 12:1 may be more representative on average for parts of the US — but it's exactly that: an average across many different temperature and cloud conditions, not a fixed physical constant. Actual ratios genuinely range from close to 0.1:1 in wet, near-freezing conditions up to 100:1 or more in extreme cold, which is why a single flat ratio can be badly wrong for any specific, unusually warm or unusually cold snow event.
Does anything besides temperature affect the snow-to-liquid ratio?
Yes — the depth and temperature profile of the cloud layer where the snow crystals actually form matters as much as the surface temperature does, along with wind and how the snow settles once it lands. Surface air temperature is the single most practical, easy-to-measure proxy for estimating the ratio, which is why it's the standard basis for simplified snow-to-liquid tables like this one, but it's a simplification of more complex atmospheric physics, not the full picture.
Can I use this to predict an actual snowstorm's totals?
Treat it as a rough planning estimate, not a substitute for an actual weather forecast. Real snowfall forecasting accounts for the full vertical temperature profile of the atmosphere, not just surface temperature, plus storm dynamics this simplified lookup doesn't model. For an actual storm, a National Weather Service or equivalent forecast will be far more reliable than scaling a rainfall-equivalent estimate by a fixed temperature-band ratio.
Why is the ratio so low (0.1) right near freezing?
Because precipitation falling at 34-45°F is barely cold enough to be snow at all — it's often a wet, heavy, half-melted mix that compacts almost as densely as rain itself once it lands, so very little vertical snow depth accumulates per inch of liquid. A ratio of 0.1 means it takes 10 inches of liquid-equivalent precipitation to produce just 1 inch of measurable snow depth in these marginal, near-freezing conditions.
Where do these specific temperature-band ratio numbers come from?
They're a widely circulated banded version of the same temperature-to-ratio relationship meteorologists describe qualitatively — colder air produces fluffier, higher-ratio snow, and near-freezing precipitation produces a dense, low-ratio one. Worth knowing: the National Weather Service's own overview of snow ratios (cited below) is explicit that a single fixed rule of thumb is often inaccurate, since cloud structure and other factors matter too — which is exactly why this tool's own guidance is to treat its output as a ballpark planning estimate, not a precise forecast, rather than presenting these bands as an exact physical law.