{"id":"us-snow-hydrology","slug":"us-snow-hydrology","name":"United States Snow Hydrology Dataset","description":"What each snow season did at 847 automated American snow stations: how much water the snowpack held at its fullest, which month that was, and which month the pack had gone. Water years 1964 to 2026, each station placed in the named subbasin its publisher assigns it to.","methodology":"Derived by Hydrionis from monthly snow water equivalent published by the NRCS Air and Water Database, and organised on the water year — 1 October to 30 September, named for the year it ends — because a calendar year cuts a snow season in half and makes every statistic from it meaningless. Peak is the largest monthly mean in the year, which is the fullest month rather than the fullest day. Peak month and melt-out month are counted from October, so that a December peak and a March peak can be subtracted from one another without crossing a year boundary. Melt-out is the first month after the peak at or below a tenth of that year’s own peak, a share of the year rather than a fixed depth, so that a thin year is measured against its own snowpack instead of being called melted-out before it ever accumulated. Each station is placed in a subbasin by the twelve-digit hydrologic unit code on its own station record, truncated to the first eight digits — a containment given by the coding hierarchy, not inferred by Hydrionis — and named from the USGS Watershed Boundary Dataset.","domains":["snow-ice"],"metricIds":["snow-water-equivalent-peak","snow-peak-month","snow-melt-out-month"],"subjectScope":"United States automated snow stations, chiefly the mountain West, in 331 named subbasins","derived":true,"context":{"hydrionisClass":"water-driver","relationshipConfidence":"direct","eligibility":{"api":true,"download":true,"overlay":true,"publicPage":true},"whyHeld":"In the American West more water arrives as snow than as rain, and it arrives months after it fell. A snowpack is a reservoir that fills all winter and empties in one season, so how full it got and when it went are the two facts that shape a river’s year. These are measurements of water standing above a river, not of water in it.","waterRelevance":["Spring and summer river flow in snowmelt-fed basins","The timing of the melt, and therefore of the seasonal peak","Summer low flow, which is largely what the snowpack did not deliver early","Reservoir inflow through the filling season","Drought onset in basins where the winter, not the summer, decides it"],"suitableUses":["Describing how full the snowpack got in a given water year at a given station","Comparing one season with the same station’s own history","Reading peak snowpack against the water-year flow of a gauge in the same subbasin, where the pairing layer allows it","Melt timing as a description of when the water left the pack"],"unsuitableUses":["Forecasting how much water will reach a particular river. What arrives depends on melt rate, soil moisture and sublimation, none of which are measured here.","Treating a station as the snowpack of its catchment. It is one point on one mountain, and it can miss drifting and wind scour.","Assuming a station is upstream of a gauge in the same subbasin. A subbasin is an area, and snow on one tributary may join the river below a gauge on another.","Reading these values as snow depth, which is a different quantity and a much larger number.","Comparing a station against a different station as though the two measured the same snowpack. Elevation, aspect and exposure differ, and so does what the pillow sees."],"joinDimensions":[{"kind":"publisher-stated","on":"The eight-digit hydrologic unit the NRCS records on each station, which is the same subbasin level the USGS resolves its gauges to. Both publishers state the code; Hydrionis does not compute it.","limitation":"A subbasin is an area, not a channel. Sharing one with a gauge does not place a station upstream of it, and snow on one tributary may join the river below a gauge on another."},{"kind":"time-only","on":"The water year, 1 October to 30 September, which the American flow record also uses","limitation":"A shared water year allows a snowpack to be read against the year it fed. It does not reach the melt itself: US flow is held here as an annual mean, so the month the water arrived in cannot be matched to the month the pack went."}]},"cadence":"annual","version":"1.0.0","released":"2026-09-02","changelog":[{"version":"1.0.0","date":"2026-09-02","summary":"First release. The seasonal statistics are new; the monthly record they are computed from was already held and is unchanged.","additions":["31,222 water years of peak snow water equivalent, peak month and melt-out month across 847 stations.","Every station placed in a named subbasin on its publisher’s own hydrologic unit code, which is what makes the snowpack readable against the rivers in the same basin."]}],"limitations":["A snow pillow is a point on a mountain, not a catchment average. One instrument stands for a great deal of ground, and it can miss drifting and wind scour.","Snow water equivalent is not snow depth. A metre of fresh powder and a metre of settled spring snow hold very different amounts of water, and only the water is measured here.","This is water standing in the snowpack, not water that reaches a river. How much arrives depends on melt rate, how dry the ground beneath is, and how much sublimates without ever becoming liquid.","Peak is the fullest month, not the fullest day. A pack that peaked sharply mid-month reads lower here than a manual course visited at the right moment would record.","Being in the same subbasin as a gauge does not put a station upstream of it. A subbasin is an area, and snow on one tributary may join the river below a gauge on another.","A water year is published only where it holds nine of its twelve months and every month from February to July, and a year whose maximum falls in the first month held is refused outright — a maximum at the edge of a record is a maximum of the record rather than of the year. Six station-years peaking below one millimetre were also dropped, because melt-out is defined as a tenth of the peak and a tenth of nothing is nothing.","Melt-out is withheld where a month is missing between the peak and the crossing, or where the pack never falls that far before the water year ends. Neither case arose in this release: every one of the 31,222 published water years contained its own crossing, because these packs melt out completely and the zeros are recorded. The rule is exercised by test rather than by the corpus."],"observationCount":93728,"seriesCount":2553,"subjectCount":851,"from":1963,"to":2025,"sourceIds":["us-nrcs-snotel"],"units":["mm","months"],"redistributable":true,"commerciallyReusable":true,"citation":"Hydrionis (2026), United States Snow Hydrology Dataset, version 1.0.0. Retrieved from https://hydrionis.com/data/datasets/us-snow-hydrology/","downloads":[{"file":"/downloads/observations-snow-ice-us-nrcs-snotel.csv","format":"CSV","label":"snow-ice observations from NRCS (470,842 rows)"}],"apiPath":"/api/v1/datasets/us-snow-hydrology.json","inputsLastRead":"2026-10-01","fields":[{"id":"snow-water-equivalent-peak","name":"Peak snow water equivalent","definition":"The largest monthly mean snow water equivalent recorded at a station in one water year — the fullest the snowpack got that year.","unit":"mm","nativeGeography":["monitoring-station"],"comparabilityNotes":"One value per water year, running October to September, named for the year it ends. It is the peak of a monthly series, so it is the fullest month rather than the fullest day: a pack that peaked sharply mid-month reads lower here than a snow course visited at the right moment would record."},{"id":"snow-peak-month","name":"Month of peak snowpack","definition":"How many months into the water year the snowpack reached its largest monthly mean — 1 is October, 6 is March.","unit":"months","nativeGeography":["monitoring-station"],"comparabilityNotes":"Counted from the start of the water year rather than as a calendar month, so a peak in December and a peak in March are 3 and 6 and can be compared arithmetically without crossing a year boundary."},{"id":"snow-melt-out-month","name":"Month the snowpack went","definition":"How many months into the water year the snowpack first fell to a tenth or less of that year’s peak — the point the melt was essentially over.","unit":"months","nativeGeography":["monitoring-station"],"comparabilityNotes":"A tenth of that year’s own peak rather than a fixed depth, so a thin year and a deep year are both measured against their own snowpack. It is a timing, not a quantity: two years can melt out in the same month having held very different amounts of water."}]}