Water Damage Restoration in Somerset County, NJ
Four rivers meet inside a few miles here, so the flood arrives late, slowly and with warning. Extraction, drying, mold remediation and the rebuild, by the crews who do all four.
Somerset does not flood the way a small-stream county floods. The North Branch and the South Branch of the Raritan, the Millstone and the main stem all come together within a few miles of one another, and a river that big responds to how much rain fell over the whole basin rather than to how hard it rained on your street. The county's claim file shows it: of 6,502 flood claims on record here, 3,019 name a stream, river or lake overflowing and 2,956 name accumulated rainfall or snowmelt — almost dead level, which is a signature you will not find in a flash-flood county.
That has one enormous practical consequence. Here, the water usually announces itself. Gauges upstream are rising while your basement is still dry, the crest is forecast in hours rather than minutes, and the difference between a bad night and a ruined one is often what got carried up the stairs before the water arrived.
What is common to every job in the state sits on one page rather than on twenty-one: the contract you sign at the kitchen table, the three business days you have to cancel it, and the permit rules the rebuild runs under are all on the New Jersey water damage page. What follows is Somerset only.
What we do
- Water extraction and structural drying — our own crews, not a dispatch network
- Mold remediation — remediation of what the water left behind
- Sewage and flood cleanup — water that came up through a sewer or across the ground
- Rebuild after the damage — the same company puts the building back
- All 21 New Jersey counties — the shop and the crews are in Garfield, Bergen County
- Paid by insurance or by you — we document the loss and the work; we do not handle the claim
The crest arrives after the rain stops, and that is time you can use
The North Branch of the Raritan near Raritan, in Branchburg, reaches its action stage at 8 feet and floods at 10. The published impacts then step up in a way that reads like a countdown: at 12 feet Vanderveer Avenue in Branchburg begins to flood, at 14 feet the first evacuations begin near Route 28, at 15 feet Route 202 starts going under in the northbound lanes and Chambers Brook backs up along Chambers Brook Road, at 18 feet Helfreds Landing in Bridgewater floods, and at 19 the Old York Road bridge between Branchburg and Bridgewater closes.
On the Millstone at Blackwells Mills in Franklin Township, the sequence is the same idea at different heights. Flood stage is 9 feet, and at 9 the text names River Road in Blackwells Mills, South River Street in Millstone, the base of the Route 533 bridge and Canal Road near Suydam Road. At 10 feet, one sentence: the Delaware and Raritan Canal is flooded by the Millstone River. Upstream at Griggstown, the causeway closes at 9.5 feet and River Road in Montgomery Township starts flooding at 10.
A homeowner who knows which of those numbers is theirs has something a homeowner in a flash-flood county does not: a few hours. Use them on the things that cannot be done afterwards — getting contents off the floor and up the stairs, photographing rooms while they are still dry, moving what is stored in the basement that cannot be replaced. What cannot be usefully done in those hours is pumping, because the river has not crested yet.
The county's north-western corner is a separate system and behaves differently again. Bernards Township drains north to the Passaic rather than west to the Raritan, and the gauge that watches it — the Passaic near Millington — publishes impacts that are almost all backwater: water pushed up side channels and into back yards rather than a wave down a valley. At 12 feet the Somerset County Environmental Education Center is surrounded by water. Backwater flooding rises slowly, comes in through the lowest opening it can find, and is easy to underestimate right up until it is in the mechanical room.
Which record — the height, or the amount of water?
Somerset is the best place in the state to see a trap that catches contractors, insurers and news reports alike. At Blackwells Mills, the USGS has measured 102 annual peaks since 1921. The highest STAGE in that record is 21.24 feet, on 28 August 2011, when the river was carrying 20,700 cubic feet per second. But the highest DISCHARGE in the same record is 22,200 cubic feet per second, on 28 August 1971 — more water, at a stage of 18.68 feet, two and a half feet lower.
Both are records. They are records of different things, and a channel that has been dredged, bridged or built up over fifty years does not turn the same flow into the same height. So when somebody tells you this was the biggest flood ever recorded on the Millstone, the only useful reply is: biggest by which measure, and in which series?
There is a smaller version of the same problem a mile up the road. The National Weather Service crest list puts the September 2021 crest on the North Branch at Raritan at 19.52 feet; the USGS annual peak series for the same gauge and the same crest says 19.38 feet, with a discharge of 31,400 cubic feet per second — the largest of 102 annual peaks going back to 1896. We quote the USGS figure on this page and name it, because a number without its series attached is not checkable. And at Griggstown, where the record is 24.21 feet from August 2011 out of 32 annual peaks, the series is measured on a different vertical datum from the gauge downstream — so that 24.21 and the 21.24 above are not two heights of the same thing and should never be put side by side.
None of this is trivia. Flood heights are what decide whether a building is treated as substantially damaged, what a floodplain administrator uses, and what a neighbour repeats to you as fact. Knowing that two true numbers can disagree by several feet is what stops you accepting the wrong one.
Newer houses, deeper basements, and a claim file that shows both
Somerset has the youngest housing stock of any county in New Jersey. Of 133,639 units, 61,983 — 46.4 percent — were built before 1980, against 62.8 percent statewide, and only 12,493, 9.3 percent, date from 1939 or earlier. This is a county of subdivisions, not of pre-war frame houses.
Now put that next to the claim file. Of the 6,502 flood claims here, 5,245 carry a basement or enclosure code, and 5,013 of those — 95.6 percent — are on a building with a basement or enclosure. That is not a coincidence: post-1970 construction on this side of the Raritan is full-basement construction, and a great many of those basements were finished after the fact into family rooms, home offices and spare bedrooms.
A finished basement is the worst possible room to flood and the most expensive one to put back. Carpet and pad, laminate or engineered flooring, insulated stud walls with paper-faced drywall, base trim, built-in cabinetry, a bathroom, sometimes a whole second kitchen. The water reaches all of it at once, and the drying problem is not the puddle — it is the wall cavity behind the drywall and the subfloor under the finish, neither of which can be assessed by looking at them.
One consequence of the younger stock is worth stating plainly, since the hazard question dominates older counties: fewer Somerset buildings sit on the wrong side of the 1978 lead-paint line, though 46.4 percent still predate 1980 and the rules that apply before a tear-out in an older house are set out in full on the statewide page.
A flood record with no single villain
Most New Jersey counties have one storm that dominates their claim file. Somerset does not. Irene accounts for 1,330 claims, Floyd 1,066, the April 2007 northeaster 989, Ida 826 and the March 2010 northeaster 700 — five separate events, over twenty-two years, each of them large. By worst year it is the same story: 1,383 claims in 2011, 1,099 in 1999, 1,021 in 2007, 858 in 2021.
That flatness is the most useful thing in the table. A county with one catastrophe has homeowners who think of flooding as a once-in-a-lifetime event. A county with five has a different truth: on these rivers, a serious flood is roughly a once-a-decade occurrence, and the building that took water in 1999 and 2007 and 2011 is the same building.
The insurance picture does not match the hydrology at all. There are 1,407 residential flood policies in force in Somerset against 104,223 residential structures — 1.35 percent. Inside the mapped high-risk zone the rate is 53.1 percent, which is ordinary; outside it, 705 policies cover 102,900 structures, 0.69 percent. Six of the county's 21 municipalities hold a Community Rating System class and every one of the six earns a real discount: Manville at class 5 for 25 percent off policies inside the mapped zone, Bedminster and Franklin at 6 for 20, Bernards at 7 for 15, and Warren and North Plainfield at 8 for 10. That is genuinely good floodplain administration in a county where almost nobody buys the policy it discounts.
| Measure | Somerset County | Where it comes from |
|---|---|---|
| Flood claims on record, 1978 to 2026 | 6,502 | OpenFEMA NfipClaims v3, county code 34035 |
| Closed with a payment | 5,767 | Paid on the building, on contents, or both |
| Total paid | About $239.8 million | Buildings $200.8m; contents $38.7m; compliance $0.34m |
| Claims naming a river, stream or lake overflowing | 3,019 | Cause code 2 |
| Claims naming rainfall or snowmelt piling up | 2,956 | Cause code 4 |
| Claims with a basement or enclosure code | 5,245 of 6,502 (80.7%) | The field is blank on the rest |
| On a building with a basement or enclosure, of the coded rows | 5,013 (95.6%) | Basement codes 1 and 2 |
| Five largest events, by claim count | Irene 1,330; Floyd 1,066; April 2007 nor'easter 989; Ida 826; March 2010 nor'easter 700 | FEMA's own event names |
| Flood policies in force | 1,407 against 104,223 residential structures | 1.35% — from FEMA's penetration data, 3 August 2026 |
| Inside the mapped high-risk zone | 702 of 1,323 (53.1%) | Same file |
| Outside the mapped zone | 705 of 102,900 (0.69%) | Subtraction on the same row |
| Municipalities in the flood insurance programme | 21 of 21 | Community Status Book, refreshed 2026-08-25 |
| Holding a Community Rating System class | 6, and all six earn a discount | No Somerset community is parked at class 10 |
| Housing units | 133,639 | ACS 2020-2024 five-year, table B25034 |
| Built before 1980 | 61,983 (46.4%) | Categories 007 to 011; 006 is the 1980s and is excluded |
| Built in 1939 or earlier | 12,493 (9.3%) | Same table |
| Peak of record, North Branch Raritan near Raritan | 19.38 ft, 31,400 cfs, on 2 September 2021 | USGS 01400000; the NWS crest list says 19.52 ft |
| Peak of record, Millstone at Blackwells Mills | 21.24 ft, 28 August 2011 | USGS 01402000 — but the biggest flow was 1971 |
Why two of these rows carry a caveat rather than a single number. A river gauge produces two independent records — how high the water stood and how much was moving past — and they do not always rank the same events, because a channel changes over a century. Blackwells Mills is the clearest case in New Jersey: its highest stage is 2011 and its highest discharge is 1971. Where the National Weather Service and the USGS publish different heights for the very same crest, as they do at Raritan, this page prints the USGS figure and says so.
Opening up a finished basement, and in what order
River water has run across ground before it reached the building, and that is the fact that shapes the first day. Material that has absorbed it and cannot be cleaned and dried to a known standard comes out rather than being dried in place — usually carpet and pad, usually the lower band of drywall, usually the insulation behind it, and anything made of particle board. What stays is cleaned, and then the structure itself is dried to a documented endpoint rather than to the touch.
In a finished basement the sequence has an extra step that people do not expect: the finish has to come off before the structure can be dried at all. A stud wall with wet insulation behind intact drywall will not dry through the drywall in any useful time, so the choice is not between saving the wall and removing it — it is between removing the lower part of it now, with the cut line chosen, or removing more of it later because the cavity never dried.
The rebuild is the same company. That matters more here than in a county of unfinished cellars, because in Somerset the rebuild is most of the job: the room that flooded was a finished room, and the finished room is what has to come back. Slow-arriving river floods also tend to leave silt, which means cleaning is a real line of work rather than an afterthought.
And if the same basement has now taken water twice in a decade — which this county's five-event record makes ordinary rather than unlucky — the conversation worth having during the rebuild is what goes back in and how. That is a design question, and it is a much better time to ask it than after the next crest.
- Contaminated or ground-contact water means porous material comes out, not dried in place.
- In a finished basement, the finish comes off before the cavity behind it can be dried.
- Silt from a river flood makes cleaning a separate stage of the work.
Common questions
The river is forecast to crest tomorrow morning. Is there anything worth doing tonight?
Yes, and this county is one of the few where there is time to do it. Get everything you can off the basement floor and up the stairs, starting with what cannot be replaced rather than what is heaviest. Photograph and video every room while it is still dry — a before picture is worth more than any amount of argument later. Find the gauge that matters to your address and learn which number puts water on your road. What is not worth doing tonight is pumping: the river has not crested, and anything you remove will be replaced by the crest.
Our basement has flooded twice in ten years. Is that unusual here?
No, and the claim file says so. Somerset's five biggest flood events — Floyd in 1999, the April 2007 northeaster, the March 2010 northeaster, Irene in 2011 and Ida in 2021 — each produced between 700 and 1,330 claims in this county alone. On these rivers a serious flood is closer to a once-a-decade event than a once-in-a-lifetime one, and the building that took water in one of them is usually the building that takes it in the next. That is worth factoring into what goes back into the room during the rebuild, which is a conversation we are happy to have before anything is ordered.
Sources
- FEMA, OpenFEMA NFIP Redacted Claims (v3), Somerset County rows (county code 34035) — retrieved
- FEMA, NFIP Residential Penetration Rates, Somerset County row (as of 2026-08-03) — retrieved
- FEMA, NFIP Community Status Book, the 21 Somerset County communities (refreshed 2026-08-25) — retrieved
- US Census Bureau, American Community Survey 2020-2024 5-year, table B25034, Somerset County — retrieved
- NOAA National Water Prediction Service, North Branch Raritan River near Raritan (RRTN4) stages and impact statements — retrieved
- NOAA National Water Prediction Service, Millstone River at Blackwells Mills (BKWN4) and at Griggstown (GTNN4) — retrieved
- NOAA National Water Prediction Service, Passaic River near Millington (MILN4) backwater impact statements — retrieved
- USGS, annual peak streamflow, North Branch Raritan River near Raritan (01400000), 102 annual peaks from 1896 — retrieved
- USGS, annual peak streamflow, Millstone River at Blackwells Mills (01402000), 102 annual peaks from 1921 — retrieved
- USGS, annual peak streamflow, Millstone River at Griggstown (01401750), 32 annual peaks, datum NGVD29 — retrieved
Water in the building?
Tell us what happened and what you can see. Pick “Water Damage” and we will come back to you about extraction, drying and what the rebuild involves.