Nonprofit, nonpartisan journalism. Supported by readers.


We’re pouring millions of tons of salt on roads each winter. Here’s why that’s a problem.

U.S. road maintenance departments have been spreading salt on streets and highways to melt snow and ice since the 1940s.

This article was originally published by Ensia.

If you live — and drive — in a northern or mountainous climate, you’ve seen highway trucks spreading loads of rock salt on snowy highways to melt the ice. But where does the salt go?

A lot of it ends up in our lakes and streams. A recent study of 371 lakes in North America — most in the northern states and southern Canada — showed chloride concentrations rising in more than a third. More than two dozen were nudging toward levels harmful to aquatic life. Extrapolated to all lakes in the U.S. northern Great Lakes and Northeast regions, about “7,770 lakes may be experiencing elevated chloride concentrations, likely due to road salt runoff,” the study concludes.

U.S. road maintenance departments have been spreading salt on streets and highways to melt snow and ice since the 1940s, but the use of salt skyrocketed over time — from 0.15 metric tons (0.16 tons) per year during the 1940s to about 18 million metric tons (19.8 million tons) per year today. Road salt use is common and growing throughout Canada, Europe, Japan, China and even South America. As much as 60 million metric tons (66 million tons) may be applied worldwide each year. Unlike chemicals that break down into less harmful compounds, road salt persists and may remain in water and soil for years, until it eventually is diluted and carried away by moving water.

Despite the ever-greater use, road salt’s effects on streams, lakes and groundwater have been largely ignored until recently. As recently as 2014, when biologist Rick Relyea began studying the effects of salt-laden runoff at Rensselaer Polytechnic Institute, “the world of science didn’t pay very much attention to the impacts of road salt on water,” he says. “Now we’re paying much more attention.”

Recent research is showing that in many waterways, chloride is on a persistent upward trend, with potential to harm aquatic communities and even impair drinking water.

Neither Relyea nor other researchers suggest highway salting crews should sacrifice public safety for the sake of healthy streams and lakes, but they say there are ways to cut salt use without impairing winter road maintenance.

Same old salt

To melt ice and prevent the accumulation of new ice on winter roads, highway crews apply salt. In the U.S., salt use is heaviest in the Midwest, Great Lakes region, New England, Alaska and the northern Appalachians. Road salt is mostly sodium chloride, the same stuff you sprinkle on food, but in coarse granular form. When it dissolves in slush it lowers the freezing point, causing ice to melt. For the same reason, salt is spread on sidewalks and parking lots.

Chloride is the component of salt of greatest concern for aquatic life. Chloride has been shown to be benign at low concentrations, but as concentrations increase salt can kill plankton, disrupt aquatic communities, increase algae blooms and stunt fish. The U.S. Environmental Protection Agency has set a long-term threshold for aquatic life of 230 milligrams per liter. Canada’s guideline for long-term exposure is 120 mg/liter. (For comparison, seawater has a chloride concentration of nearly 20,000 mg/liter.)

road salts
Wikimedia Commons
Calcium chloride, center, and magnesium chloride, right, improve traction at lower temperatures than sodium chloride, left, but they are more expensive and still cause environmental harm.

“Just that contrast makes you realize we don’t have a good idea of what concentrations are really harming our environments,” says Hilary Dugan, assistant professor at the University of Wisconsin–Madison Center for Limnology and lead author of the North American lakes study, published this spring in the Proceedings of the National Academy of Sciences.

In many cases, the U.S. and Canadian thresholds are already being exceeded. Keeping freshwater “fresh,” according to Dugan’s paper, “is critically important for protecting the ecosystem services freshwater lakes provide, such as drinking water, fisheries, recreation, irrigation, and aquatic habitat.”

Toll on waterways

Dugan’s paper tracks long-term chloride concentrations in North American lakes with detailed available records. Most were in what researchers called the “North American Lakes Region,” which includes Connecticut, Maine, Massachusetts, Michigan, Minnesota, New Hampshire, New York, Ontario, Rhode Island, Vermont and Wisconsin.

Mean chloride levels ranged from hardly any at all to 240 mg/liter, above both U.S. and Canadian standards. About 10 percent in the lakes region exceeded 100 mg/liter. And perhaps most concerning, slightly more than a third of the lakes overall showed persistent upward trends in chloride concentrations.

Dugan attributes the increased levels to factors such as more roads, bigger roads, more traffic, and more parking lots. The lakes with the greatest long-term concentration of chloride were those with the greatest proportion of impervious surfaces, such as roads and parking lots, in their watersheds. But it didn’t take a lot — as little as 1 percent road surface within a half-kilometer (third of a mile) of the water body. “It was a surprisingly small percentage of impervious surface that led to long-term increases in chloride,” says Dugan. “I’m not sure that anyone expected that percentage to be so low.” According to her study, 27 percent of large lakes in the United States have more than 1 percent impervious surfaces nearby.

Much of the salt runs off these surfaces shortly after it’s applied or with spring melt. But some of it seeps into soil, creating a “reservoir for chloride,” Dugan says. “Even if we stopped applying road salt today, there’s a high likelihood that chloride levels [in lakes] would continue to increase for awhile as some of those chlorides flush out of soils.”

Dugan’s big-picture look at North American lakes squares with Asleson’s finer-grained analysis of Minnesota’s Twin Cities, in which 19 lakes currently exceed the water-quality standard for chloride. And chloride concentrations were increasing in most Twin Cities lakes.

“When you have a watershed area that has a road density of 18 percent or greater [in the entire watershed], that’s where you’re most likely to see water quality problems because of winter deicing salt,” says Asleson.

Trophic cascades

While chloride is not yet poisoning our waterways, chloride does have the potential to change aquatic communities, stunt fish growth, aid exotic species and even affect tourism.

Rick Relyea is director of Rensselaer’s Jefferson Project at Lake George, a deep, clear 32-mile (51.5-kilometer)-long finger of water in northern New York. Relyea and colleagues monitor the lake and conduct experiments in artificial habitats to determine the effects of chloride and other components of salts on aquatic life.

“When some activity like road salt harms one species, it’s usually not the end of the story,” says Relyea. “It indirectly affects a lot of other species.”

In Relyea’s study, high road salt concentrations induced a “trophic cascade,” reducing zooplankton and producing an upsurge in their food, phytoplankton, which seemed to thrive in the high salinity.

Relyea’s team also found that exposure to salt drove zooplankton evolution toward salt tolerance. “Those zooplankton populations that were knocked down by a lot of salt actually bounced back and started doing really well,” he says. That tolerance was passed on to subsequent generations. “That’s really the hopeful message,” he says. “It’s not that we should ignore the issue. It’s hopeful that we could buy some time until we solve the issue.”

In another study, tadpoles raised in salty water became male rather than female frogs at a 10 percent greater rate than expected. The team doesn’t understand the underlying mechanism, says Relyea, but “the explanation is clearly that we have converted some of the females into anatomical males while they are tadpoles.”

Relyea found that salt levels in Lake George are rising but are still far too low to impair aquatic life. Streams in the watershed are a different story. Chloride concentrations spike to levels 100 time greater that those found in lakes, and remain high through the year as chloride leaches from soils. “That’s probably true throughout the northern U.S. and Canada,” Relyea says.

Other research has shown that salt can affect trout growth.Calcium chloride had the greatest effect of common road salts, at chloride concentrations of 860 to 3,000 mg/liter. The effect was greatest at the highest concentration, reducing weight of rainbow trout by more than 30 percent. “If you grow more slowly, you can be more susceptible to predators, it will take you longer to be reproductive, you will lay fewer eggs,” says Relyea. “Growth for a fish is everything.”

Courtesy of the MPCA Chloride Project
In the Minneapolis–St. Paul area, 19 lakes (red dots) and miles of streams (red lines) are considered impaired by chloride.

 Relyea says the sodium in salt can trigger the release of other metals from soil that run into waterways. Released calcium can favor some species over others. “Now you make it easier for some invasive species, like say Asian clams, zebra mussels, various snails — you make it easier for them to get a foothold if they ever arrived in your lake,” he says.

Road salt also damages and kills vegetation, though the effects are concentrated within 200 feet of roadways.

High salt use can cause problems for humans, too. Salt seeps into groundwater, raising the salinity of drinking water. In Madison, Wisconsin, where Dugan lives, “That’s a huge concern for municipalities and water treatment plants,” she says. And according to research by the EPA and U.S. Geological Survey, high chloride increases the corrosion of poisonous lead from old water pipes.

Low-salt solutions

Researchers have experimented with salt substitutes such as beet juice, which lowers freezing temperature and melts ice as the sugar it contains dissolves on the road. But the sugar is a fertilizer that feeds algae growth.

“In most lakes, we already have enough nutrients going in, particularly in clear, infertile lakes. They’ll be more green and less transparent and less aesthetically pleasing to most people,” says Relyea. “The less transparent the water becomes, the less valuable the attraction to tourists and the less income that comes into communities.”

Many states regulate road salt storage. But many do not. And none specifically regulates the application of road salt, says Asleson. Instead, road maintenance departments are encouraged to use best management practices. New Hampshire offers a voluntary certification and training program for private applicators maintaining large surfaces such as parking lots. Likewise, Canada has developed a “code of practice” for road salt use.

The MPCA has created a web-based tool for public works departments and other winter maintenance pros to help evaluate their own programs, from small details (Do they overfill their salt and sand trucks?) to big issues (Do they stockpile road salt outside?). “We’ve looked at every aspect that we could with this core group of winter maintenance experts to find every opportunity possible to reduce salt use,” Asleson says.

Asleson thinks the biggest single change to use less salt is switching to liquid solutions. The brine spreads more evenly, stays put and begins working immediately because the salt is already in solution. As a result, spraying liquid brine is more effective while using less salt. Asleson says cities that have switched to tanker trucks have reduced salt use by up to 70 percent and paid back their equipment investment in a year or two.

In northern New York, Relyea says, local governments have been adopting so-called live-edge plows. The plow blade, rather than being solid, is divided into short independently moving sections that follow the contours of the road and better remove snow and ice. That leaves less ice to be removed by chemicals, reducing salt usage. “You still salt, but you don’t salt as much,” he says.

“The salt issue is biologically very complex, but I think it has motivated people to think about how we can simultaneously have safe roads and healthy ecosystems,” says Relyea. “If communities could have the ability through technology to purchase less salt, to salt fewer times, pay less truck driver time and help their lakes that are big tourist attractions, it really can be a win-win for everybody involved. It’s not really about posing the health of ecosystems against public safety.” 

Comments (9)

  1. Submitted by Brian Simon on 11/20/2017 - 10:35 am.

    Private sector?

    One challenge is in encouraging reduced private sector usage. One place i visited this weekend had a parking lot thick with salt. The surface was det, with no ice anywhere in site. Yet, someone had recently spread several bags worth.

    Is there a way to disincent such over-application? From their perspective, they’re likely far more concerned about liability of a slip and fall, than of wasting money on salt, or where their runoff goes.

  2. Submitted by David Markle on 11/20/2017 - 11:51 am.


    Very prompt and thorough clearing away of snow can help minimize the need for salt. Then,sunlight and sublimation can often and quickly result in the pavement drying. Any remaining significant film of snow will likely change into ice. I wonder if a second-pass use of big rotary brushes could offer an economical alternative.

    Rocky Mountain states tend to get by with little or no use of salt, but they have drier climates than ours.

    • Submitted by Ray Schoch on 11/20/2017 - 01:28 pm.

      Yes, but…

      I can’t speak for other states, but I spent a dozen years on the edge of Colorado’s Front Range foothills, and while it uses far less than Minnesota, Colorado (or at least Jefferson County and metro Denver) still uses **some** road salt. In place of much of that salt, in Colorado, CDOT uses “sand.” I put the word in quotes because what passes for “sand” in Colorado is what I grew up in Missouri calling “aquarium gravel.” The pieces are large enough that they can (and regularly do) chip automotive paint and crack windshields when slush containing the “sand” is splashed on to vehicles in adjacent lanes on multi-lane roadways. Cars with cracked windshields are as common in metro Denver and along the Front Range as are rusted vehicles here in the “salt belt.” In addition, while “sand” doesn’t usually have the same chemical aftereffects on water, the excess and residue wash off the roads with spring and summer rains, and end up clogging local streams, which sometimes has significant negative effects on local fish and other aquatic populations.

      There ain’t no free lunch. Unless you’re independently wealthy and have a very large supplemental freezer in the basement, there’s probably no way to stay off the roads through the whole of a Minnesota winter, so some form of treatment of roads to make them safe(r) during the winter seems likely to be necessary for a long time to come.

      • Submitted by Paul Udstrand on 11/21/2017 - 09:15 am.

        Back to sand I say

        We used to use sand around here as well, and I think in many places we should return to it. The main problem with ice and snow is that it reduces traction, sand restores traction even if it doesn’t “melt” the snow. Furthermore, frankly, I don’t like driving in all the slush, I’d rather drive on snow or compacted snow. In addition to the salt on the road look at that wiper fluid everyone is blasting all winder long in order to cope with windshields fouled by slushy roads, and I’m not sure slush is really THAT much less slippery than snow, is there any data that all this slush is actually reducing the number accidents during a snow storm?

        The other problem with that salt slush is that it cakes up in wheel wells, and it’s hard on residential driveways.

        The problem with sand in the cities is it ends up clogging the storm sewers, but a combination of sand, nothing at all, and brine would make that problem manageable. Out on rural highways there are no storm sewers so sand away, and plow. You can even scoop up some of the sand and re-use it.

        In St. Louis Park most of the residential streets don’t actually need any treatment beyond plowing for the most part except for intersections. You can identify problem areas and higher traffic streets, but the majority of “road” in a city is low traffic residential.

        Not to be the old man curmudgeon but when I was kid and a teenager (60’s, 70’s, and 80’s) the city used road graders to plow the streets, and the streets were all plowed by 10:00 in the morning at the latest. Many a morning it was I has to suit up and move my car parked in the street (at 6:00 am) so it wouldn’t plowed in by the road grader… and if I got out there in time the driver would actually wait a few minutes for me to clear off the windshield! The graders plowed, dump trucks came by later with sand. You could say combination truck-plows are more efficient, but my street doesn’t plowed now until 5 or 6 in afternoon after some storms. That means people leaving for work AND coming home from work on un-plowed streets. I think it’s having to go refill the salt that slows the plowing down, or may fewer trucks because of the increased cost (I imagine those fancy trucks with all the attachments are more expensive than a road grader). Whatever, what I can tell you is this: My lived experience of 38 years as a driver is that the roads actually are no easier or safer to drive on now than they were before in the winter. So if all that salt is harming the environment, it’s not worth it.

        • Submitted by Elizabeth Henry on 12/10/2017 - 11:57 am.

          Regarding the slush issue

          Thank you Paul for you insight. I agree with everything you wrote. The chloride compounds being mixed with sand create a myriad of problems. There are three commonly used – sodium chloride, calcium chloride, and magnesium chloride. They all melt snow but not without a price. They each react a little differently to temperature too. The chemical properties can greatly affect water tables, stream runoff, aquifers, water wells, and soil chemistry over the long haul of heavy use. Then there is the safety factor- the sand is great for helping with traction and the chloride compounds do melt the snow and ice but combined the two create a mucky mess that becomes very slippery, can actually refreeze creating a situation similar to black ice, adheres to mirrors and headlights and is very difficult to remove – all detrimental to safe driving. I would rather see campaigns for safer driving specifying winter driving, drivers slowing down, good plowing and grading with some sanding at critical areas, and eliminate the salt except maybe on highways.

  3. Submitted by Carrie Preston on 11/20/2017 - 02:14 pm.

    Less corosive

    Our garage is very close to where they are clearing snow and dropping the salt/sand mixture. It has eaten away the corner of our garage. Whatever is in that mix can’t be good.

  4. Submitted by David Markle on 11/20/2017 - 06:52 pm.

    A failed experiment

    Many years ago–in the 1960’s, I think–Minn DOT made a deal with Archer Daniels Midland or some comparable purveyor of agricultural products. A mixture of salt and linseed oil got applied to busy highways around the metro area, in an effort to reduce the corrosive effect on vehicles. The results were truly dangerous. Windshield wipers lost their effectiveness; even in combination with washer fluid they only smeared the oil, dramatically reducing visibility. Minn Dot never tried that again.

    I suspect Carrie Preston may have a masonry foundation or masonry garage. Concrete and mortar harden as products of hydration, and salt dehydrates them.

  5. Submitted by Peter Pesheck on 11/26/2017 - 11:19 am.

    Thanks, helpful info

    I’m a retired chemist/biochemist working to understand (there’s a big word!) water quality dynamics in small lakes. This study (your article and the original paper you reference) will be very useful in my work. / p

  6. Submitted by Jim Spensley on 11/27/2017 - 09:20 am.

    Undersground Contamination

    Much of Minneapolis and Hennepin Cpunty sits on a limestone layer. In many place the limestone has been tunneled out for storm sewers and other reasons. Some from acid rain and a lot from street and highway drainage, acidic water seeps in to natural cavities and tunnels dissolving calciom and enlarging voids in the limestone.

    This process has caused collapes (sink holes), polluted wells and lowered water tables and aquifers at an increasing rate since the 1840’s.

Leave a Reply