ANALYSIS
The University of Vermont will lead a $5.9 million project to install automated snow monitoring equipment on mountains across Vermont, New Hampshire and Maine — the first serious money aimed at a gap that has left the fastest-warming season in the eastern United States almost entirely unmeasured where it matters most.
The award came from the National Science Foundation on Aug. 24, part of $39.9 million spread across eight projects across 16 states and territories. UVM receives $3.3 million and the University of New Hampshire $2.6 million as a collaborator, with subawards to Vermont State Colleges, the University of Maine, the University of Maine Farmington, and the Mount Washington Observatory. The project runs four years and is called Snow STORE.
The principal investigator is Keith Jennings, director of research at UVM’s Water Resources Institute, whose own research specialty is rain-on-snow flooding.
What you can check today
Mount Mansfield snow depth, updated daily through the ski season: matthewparrilla.com/mansfield-stake, which republishes the National Weather Service reading.
NWS Burlington’s Winter/Spring Flood Outlook for northern New York and all of Vermont, issued through late winter and spring: weather.gov/btv.
Your town’s flood risk, including which buildings, roads and critical facilities sit in a flood hazard area: the state’s Vermont Flood Ready Atlas. For a plain-language summary of where your own town stands — whether it participates in the National Flood Insurance Program, whether its emergency operations plan is current, and whether its flood hazard maps have been digitized — pull your Community Report.
What Vermont measures now
Vermont’s best-known snow measurement is a wooden two-by-four.
The Mount Mansfield snow stake is a 12-foot post marked off in feet, standing at 3,900 feet of elevation. It was installed in 1954 and is read remotely by camera by the National Weather Service office in Burlington. It has produced a nearly unbroken seventy-two-year record, and skiers and forecasters watch it closely — when it passed 100 inches in March 2025, it made news.
The stake measures how deep the snow is. It does not measure how much water is in it.
That distinction is the reason this award exists. The number hydrologists actually need is snow water equivalent — the depth of water you would be left with if the whole snowpack melted at once. Depth and water content are not the same thing and can diverge sharply: a dense, wet spring pack holds several times the water of the same depth of cold, dry powder. Two feet of snow can be a nuisance or a flood, and depth alone will not tell you which.
Vermont has some of the equipment that measures water content. UVM’s Summit to Shore network has 22 weather and snow stations running from Lake Champlain up Mansfield and across to the Northeast Kingdom, funded through the Army Corps of Engineers’ Cold Regions Research and Engineering Laboratory. What it does not have is regional coverage, and neither does anywhere else in the Northeast.
The number that defines the gap
The Northeast Snow Survey feasibility study — a UNH-led effort with UVM and the Appalachian Mountain Club, funded through the U.S. Department of Agriculture’s Natural Resources Conservation Service after Sen. Jeanne Shaheen secured $1 million for it in the FY2023 federal funding bill — put the problem in a single statistic. Across the Northeast, fewer than 5 percent of all snow measurement stations, manual or automated, sit above about 2,700 feet. Almost everything the region knows about its own winters comes from low-elevation stations in populated valleys.
The contrast with the West is stark:
SNOTEL, the federal automated snow network built by the Natural Resources Conservation Service, runs more than 900 stations across the western states.
It has been operating since the 1970s.
It reports snow water equivalent, depth, precipitation and temperature, and it feeds the water supply forecasts that western states use to allocate water for the coming year.
It also feeds flood forecasting. NRCS’s own published assessment of the network describes high-elevation readings as crucial to building, testing and running flood-forecast models — because mountain conditions cannot be reliably guessed from valley-floor stations. Temperature inversions and terrain break the arithmetic.
The Northeast has no equivalent network. The stations that exist are disconnected and uncoordinated.
Jennings put the imbalance plainly in UVM’s announcement, saying western states have dominated American snow research for decades.
Meanwhile the region is changing faster than the one with all the instruments. NESS researchers found that winter is the fastest-warming season east of the Mississippi, and that the Northeast has lost two to three weeks of sustained winter conditions over the past century, at low elevations and in some mountain areas alike.
The place warming fastest in the season changing fastest is the place with the thinnest measurement.
Why this is a flood story
Both the NSF announcement and UVM’s release lead with the winter economy, and the economic case is real. Vermont’s alpine ski areas logged 4.36 million skier visits in 2025-26, the best season since 2014-15, and tourism overall accounts for roughly 9 percent of state GDP. Resorts that know their snowmaking window three days earlier open more terrain.
But the larger public use of these instruments is flood forecasting, and it gets a subordinate mention.
Every winter and spring, the National Weather Service office in Burlington issues flood outlooks for Vermont and northern New York. Snow water equivalent is one of the listed inputs, alongside soil moisture, groundwater and river levels. Better readings from more places, more often, mean better outlooks — and more lead time for the emergency managers, road crews and town officials who act on them.
That connection is built into the project rather than inferred from it. Both the NWS Burlington forecast office and the Northeast River Forecast Center — the office that issues river and flood forecasts for all of New England — are listed partners on Snow STORE, alongside the ski associations and the state outdoor recreation offices. The forecasters who would use the data helped shape what gets collected.
The Vermont sites are named. UVM plans to add equipment at existing weather stations at the Jericho Research Forest, Mount Mansfield, and the Sleepers River Research Watershed in Danville, along with Stowe Mountain Resort. Sleepers River is the telling one: it is a long-running federal hydrology research basin in the Northeast Kingdom, studied for decades precisely because of what its snowpack does to streamflow in spring.
Vermont’s flood exposure is concentrated in narrow river valleys where most of the state’s residents live, and the specific hazard is rain falling on an existing snowpack. Rain warms the snow, the snow releases water it has been storing for months, and the runoff can far exceed the rain that triggered it. Compass examined that risk on April 9, in the context of warmer winters producing more rain events and more marginal snowpacks.
Knowing how much water is sitting on the Greens in March is not a ski question. It is a question about whether the town below gets warned in time.
What the record does not settle
Two things are worth watching, neither a reason to discount the award.
Who runs it in year five. SNOTEL is a standing federal program with a permanent operating budget. Snow STORE is a four-year research grant. Instruments installed under a grant need someone to maintain, calibrate and pay for them after the grant closes, and the announcement does not address it.
There is a bill that would answer the question, and it has been sitting for three years. The Snow Survey Northeast Expansion Act would direct the Natural Resources Conservation Service — the agency that runs SNOTEL — to extend its Snow Survey and Water Supply Forecasting Program to the Northeast, making a permanent network a federal operating responsibility rather than a grant cycle. Sen. Jeanne Shaheen of New Hampshire first introduced it in July 2023 and reintroduced it as S.1029 on March 13, 2025. It was referred to committee and has not moved since.
The bill has exactly two cosponsors: Susan Collins and Angus King, both of Maine. Vermont is the lead state on the research award, and no Vermont senator has signed on to the legislation that would turn it into something permanent.
Who gets the data. Elizabeth Burakowski, a research associate professor of Earth sciences at UNH, said the new monitors will provide data and a dashboard. SNOTEL’s data is free and open to anyone, which is a large part of why it is useful. The access terms for this one are not yet in the record.
For seventy-two years, Vermont’s headline winter instrument has been a wooden post that someone carried up a mountain and planted in the ground. It has done honest work, and the record it produced is genuinely valuable. It was never going to tell a town in a river valley how much water was waiting above it.