The Hudson's highest recorded flood at Albany, 21.71 feet in February 1857, came from ice jamming on sandbars south of the city. March 1913 set the river's record from Glens Falls to Troy. Dredging after the 1930s eased the jamming nearer Albany.
Key takeaways
- NWS Albany lists February 1857 as the highest flood ever recorded at Albany, 21.71 feet, with ice jammed on sandbars south of the city.
- March 1913 set the Hudson record from Glens Falls to Troy: 29.7 feet at Troy and 21.45 feet at Albany.
- March 1936 ran close behind, 29.5 feet at Troy and 17.9 feet at Albany.
- NWS Albany says ice jamming is not much of a problem between Troy and Waterford because the Champlain Canal dams are fixed, and that problems near Albany and south eased after the 1930s dredging for shipping.
- Ice has not stopped moving water about: on 25 January 2019 a surge of water and ice tore seven vessels from their moorings near Troy, and some struck bridges including the Patroon Island Bridge.
- No Hudson flood-gauge figures after 1936 were available for this guide, so nothing is claimed about crest heights since.
What is an ice jam, and why does the Hudson get them?
An ice jam is a blockage of broken ice in a channel. River ice breaks up in a thaw, travels downstream, and piles against a shallow, a bend or a structure. Water backs up behind the pile, often faster than rain would raise it.
The mechanism explains the season. Jams belong to late winter and early spring, when a thaw or a warm rain breaks a frozen surface into plates and sends them downstream together. The pile stops where the channel can no longer pass it. Behind the pile the water has nowhere to go but up and out, and when the pile finally releases, the same water arrives downstream in a surge rather than a rise. That is why ice floods are remembered as sudden in a way that summer floods are not.
The historical record at Albany reflects exactly that. NWS Albany lists February 1857 as the highest flood ever recorded at Albany, at 21.71 feet, with ice jammed on sandbars south of the city; the geography did the damage as much as the volume of water. A separate account from a river museum puts a different mark on a flood above 1857, and because the two have never been reconciled this guide does not print that second figure.
What do the Albany and Troy flood records actually say?
Three events dominate. February 1857 remains the highest at Albany at 21.71 feet. March 1913 set the record for the river from Glens Falls to Troy, with 29.7 feet at Troy and 21.45 feet at Albany. March 1936 reached 29.5 feet at Troy and 17.9 feet at Albany.
The gap between the Troy and Albany numbers is worth sitting with, because it is the most useful thing in the list for a homeowner. The same river in the same week stood roughly eight feet higher at Troy than at Albany in 1913, and about twelve feet higher in 1936. River stage is a local measurement against a local datum; it is not a depth of water over your street, and it is not transferable from one city to the next. A Troy figure quoted in Albany, or the reverse, tells you almost nothing about your own ground.
These are also the only Hudson gauge figures this guide can offer. No crest heights after 1936 were available when it was written, so there is no table here of recent high water on the river, and no claim that any later event did or did not approach these marks. Where a number is missing the honest page says so, and then explains what can be said instead, which is what the rest of this guide does.
Has the ice problem changed since then?
According to NWS Albany, yes. It reports that ice jamming is not much of a problem on the Hudson between Troy and Waterford because the Champlain Canal dams are fixed, and that problems near Albany and to the south eased after the dredging of the 1930s for shipping.
Both halves of that are engineering, not climate. A fixed dam holds a pool at a known level, so the ice above it does not get the chance to stack against a shifting shallow in the same way. Downstream of Troy, the channel that was dredged to bring shipping up to the Port of Albany is deeper than the bar-and-shoal reach that caught the ice in 1857, and a deeper channel passes broken ice instead of collecting it. The reach that produced the city's record flood is not, physically, the reach that exists now.
What that does not mean is that ice stopped mattering. It means the old failure mode was reduced and the river still carries plenty of ice every winter. It is also a reminder of how to read a historical flood: the number belongs to the channel of its day. Our ice jam and spring thaw flooding page deals with what current conditions put into buildings.
What happened in January 2019, and why does it still matter?
On 25 January 2019 a surge of water and ice tore seven vessels from their moorings near Troy, and some of them struck bridges, including the Patroon Island Bridge. No houses are claimed here; the lesson is about force rather than depth.
Moving ice is a battering ram with the river behind it. Vessels that were secured for winter were taken off their moorings and carried down into fixed structures, and that is a plain measure of load - not of how high the river stood. For anyone with property on the bank, a dock, a bulkhead, a boathouse or a low outbuilding, the January case is more relevant than the historical crests, because things get destroyed at that level by impact and not by inundation.
It also illustrates how quickly conditions change in a thaw. Nothing about an ice surge waits for a forecast to be read, and the things it moves are the things left in its way. For a household, that argues for knowing in advance what sits lowest and what is anchored, and for not storing anything that matters at ground level on the riverside in February.
Is there a local record of ice and thaw flooding besides the gauges?
There is one, and it comes with a caveat. An Albany Argus article of March 1919 described Troy and Green Island flooded in a combined ice, storm and thaw event. It survives in this guide only as quoted on a blog; no weather service record of it was found.
That is stated as attribution rather than as a hedge. The 1919 event may well be exactly as the newspaper described, but a nineteenth- or twentieth-century press account repeated on a blog is a different grade of evidence from a weather service gauge reading, and a reader is entitled to know which one they are being given. Where the two grades appear on the same page, the right thing is to label them, not to blend them.
What the account does usefully show is the combination. Ice, a storm and a thaw arriving together is the pattern that put water into these riverside places historically, and the places named - Troy on the east bank and Green Island on its island in the river - are low, built-up and old. Nothing about that geography has changed, even where the channel has.
Which places along this stretch sit lowest?
Green Island is a village on an island in the Hudson opposite Troy. Albany's South End lies mostly on the river's flood plain and is level. Watervliet, Rensselaer and Troy itself front the river directly, and the 1857 jam formed on sandbars south of Albany.
Height above the river is the first thing worth knowing about your own address, and it is not something a flood story from another town supplies. The South End sits on the flood plain, where ground-floor level is close to grade; much of its older housing was cleared when the state took 98.5 acres by eminent domain in 1962, so what stands there now is a mix of eras rather than a single one. Green Island, by contrast, is surrounded by the river on all sides.
Troy's own exposure runs along the riverfront while the Central Troy Historic District behind it, listed in 1986 with nearly seven hundred properties from the early nineteenth to the mid-twentieth century, climbs away from the water. Along Fifth Avenue there are modest two-story brick worker houses; cellars in stock of that age are stone or soft brick and they take up water from outside as readily as from a pipe. Our Hudson River flood cleanup page covers what river water leaves in a building like that.
What should a riverside household do before a thaw?
Know the height of your lowest floor relative to the river, get anything irreplaceable off that floor, make sure the sump and any backwater valve work, and keep the heat on so pipes in cold corners do not add their own flood.
A thaw week is a double exposure, and that is the part people miss. The river is rising and breaking up outside while the freeze that preceded it has been working on the plumbing inside, so the same week can bring both river water and a burst supply line in a cold crawl space. Checking a sump pump before the ground thaws is ten minutes of work; checking it afterwards is a conversation about a wet basement. The sump pump and backwater valve failure page covers the mechanical side.
The second thing is documentation ahead of time. Photographs of a dry basement and its contents, taken in January, are worth a great deal more than photographs of a wet one, because they establish what was there and in what condition. Insurers deal with records, and treatment of flood water and of plumbing failure differs by policy; New York's insurance regulator is the Department of Financial Services. Build the record before you need it.
Frequently asked questions
Does the 1857 record mean the river could reach 21.71 feet at Albany again?
It means that is the highest stage ever recorded there, in a specific set of conditions: ice jammed on sandbars south of the city. NWS Albany notes that problems near Albany and to the south eased after the 1930s dredging for shipping, so the channel that produced that crest is not the channel that exists today. Treat the figure as history, not as a forecast.
Why was Troy so much higher than Albany in 1913 and 1936?
River stage is measured against a local datum at each gauge, so the numbers are not comparable as depths. In March 1913 the Hudson reached 29.7 feet at Troy and 21.45 feet at Albany, and in March 1936 it was 29.5 feet at Troy and 17.9 feet at Albany. The practical point is that a crest figure from a neighboring city tells you nothing reliable about your own property.
Are ice jams still a danger between Troy and Waterford?
NWS Albany reports that ice jamming is not much of a problem on that stretch, because the Champlain Canal dams there are fixed and hold a known pool. That is about the stacking of ice in the channel. It does not mean the river carries no ice, and January 2019 showed that moving water and ice can still tear vessels from moorings near Troy.
What is the difference between river flood water and a burst pipe indoors?
Category, and therefore scope. Water that arrived over the ground from a river is handled as contaminated, which generally means removing carpet, pad, insulation and the lower band of wall finishes, then cleaning and disinfecting before drying. A burst supply line starts clean, so more can often be dried in place if it is caught quickly. The same volume of water produces two different jobs.
Should I pump a flooded cellar out as fast as possible?
Not while the ground outside is still saturated or the river is still high. Water in the ground presses on walls and floors, and removing the water inside too quickly can leave an old masonry cellar carrying that pressure unbalanced. Pump it down in stages, keep an eye on the walls, and if there is any sign of movement or bulging stop and get someone to look at it.
Is this stretch of the river the same one as the Mohawk at Cohoes?
No, and it is worth keeping them apart. Cohoes is where the Mohawk drops about seventy feet over Cohoes Falls into the Hudson, which is its own geography. This guide is about the Hudson at Albany and Troy: the gauge records at those two cities, the 1857 jam on the sandbars south of Albany, and the ice that still moves on the river each winter.
Sources
- National Weather Service, Albany - major floods
- National Weather Service, Albany - ice jam information
This guide is general information for Albany County homeowners, not a substitute for an on-site inspection. Prices are estimates. See our disclaimer.