Ant Control in Park Slope

Where is the wet wood in a brownstone, and how did it get wet?

In five predictable places, and it got wet because a detail that used to shed water stopped doing it.

Carpenter ants do not eat wood. They excavate it to make galleries, and they excavate wood that is already soft — water-damaged, decaying, or held at an elevated moisture content long enough for fungi to have started work. That is why an ant finding is nearly always a water finding first, and why treating the insect without correcting the water buys a season.

On a nineteenth-century Brooklyn rowhouse the water gets in at these points, in rough order of how often we would expect to find it.

The cornice. A projecting cornice exists to throw rain and snowmelt clear of the face of the building. Most of the ones here are pressed sheet metal on a light timber framework, cantilevered out from the top of the front wall. When a seam opens, a fixing corrodes or the back edge against the roof fails, water goes into a hollow framed void full of timber and stays there. Nobody sees it. The framing softens over years, and the first indication is frass appearing on a top-floor floor.

The window and door sills. The Landmarks rowhouse manual is explicit that sills and lintel tops should be pitched so water runs off. A century and a half of repairs, repointing and replacement has flattened a great many of them. A flat sill holds water against the joinery and against the masonry beneath it.

The rear extension seam. Where a later extension was built against the original back wall, there is a full-height junction between two structures of different ages that move differently. When it opens, and it does open, water enters at every level at once and reaches framing on both sides.

The roof, its drainage and anything sitting on it. Blocked internal drains, scuppers that no longer clear, a parapet flashing that has lifted, and — increasingly — a deck with planters on it. The planter is worth its own sentence: it holds wet soil permanently against a membrane above a top-floor ceiling, it never dries underneath, and it is almost never lifted.

Plumbing in a chase. In a subdivided house the retrofitted waste stack and supply risers run inside a boxed chase through every floor. A weeping joint in that chase wets structural timber continuously, in a place nobody can see, for as long as it takes somebody to notice a stain two rooms away.

The brownstone itself contributes. The Landmarks manual identifies water as the eventual cause of most masonry deterioration, and explains that brownstone veneer was commonly set with the grain running vertically, so water between the bedding layers freezes and forces the layers apart. Spalled stone exposes fresh, absorbent surface, the wall stops shedding water properly, and whatever timber is behind it stays damp. Ants do not cause that. They report it.

What has the street tree got to do with the ants inside?

Frequently everything — because on these blocks a mature tree stands within a few feet of a facade with its branches on the cornice.

This is the point that changes the shape of a great many jobs in this neighborhood, and it is missed constantly. A carpenter ant colony’s main nest is very often outdoors. Its preferred sites are decayed wood in contact with moisture, and the best examples of that within fifty feet of a Park Slope house are usually trees: a street tree with an old pruning wound or a decayed limb, a yard tree with a rotten root crown, or a stump left in a rear yard when something was taken out decades ago.

From an outdoor main nest, a colony sends foragers out along whatever route is available, and it establishes satellite nests in drier, warmer voids nearer the food. Those satellites can be in a building even when the main nest never is. So the classic Park Slope pattern is a colony in a tree, satellites in the cornice framing or a top-floor wall void, and a foraging trail running along a branch onto the cornice and in through a seam.

That has three practical consequences.

The building can be treated thoroughly and the problem persists, because the main colony is not in the building.

Cutting the physical connection matters more than product. A branch that touches or overhangs a cornice, a cable, a trellis, a downpipe, a fence running to a wall. Clearing a gap around the roof edge and the facade removes the bridge. Where street trees are involved, that is the city’s decision rather than the owner’s, but garden trees usually are not.

The trail is worth following outdoors before anything else happens. Carpenter ants forage mainly at night, and a flashlight at ten in the evening on a warm night in late spring will show you, in twenty minutes, something that no amount of daytime inspection produces: which direction they are coming from, and whether the flow is into the building or out of it.

The other outdoor sites are worth listing because they repeat on this housing stock. A rear extension with a timber deck on sleepers. Sleepers or railway ties used to build a raised bed. An old fence post let into soil at a boundary. Firewood stacked against a wall. Timber left buried when a yard was regraded. Each of those is decaying wood in soil contact within a few yards of the house.

What is happening at the top of the building?

More than most owners realize, because the roof deck era has put wet soil directly over the highest timber in the house.

Roof decks are now a standard feature of this housing stock, and they change the moisture picture at the top of a building in a way that is genuinely new.

A deck is built on sleepers or pedestals sitting on the membrane. Under it there is a void that holds debris, retains moisture and is not accessible for inspection without lifting boards. The membrane below it cannot be seen, which means a lifted seam, a failed upstand at the parapet or a blocked drain goes unnoticed for years rather than weeks. And the planters — which are the point of the deck for most people — hold soil that is watered through the summer, sitting on the deck or directly on the membrane, never drying underneath.

That combination puts sustained moisture over the top-floor ceiling structure. It also puts soil, organic matter and decaying timber at the top of the building, which is a habitat that would otherwise only exist in the yard.

What we find, in order:

Damp framing at the parapet and the top plate. The junction where the roof structure meets the top of the wall is where water that gets past the membrane ends up, and it is structural timber.

Colonies in the deck itself. Sleepers in constant contact with a damp membrane, and pressure-treated timber that has been down for fifteen years, are both perfectly good nesting sites.

Blocked drainage. Leaves and soil from planters wash into internal drains and scuppers. A roof that ponds keeps everything below it wet.

Cornice framing wet from behind. Water that gets into the roof edge reaches the back of the cornice void, which is the same void discussed above.

The corrective work is unglamorous. Lift the deck, or enough of it, once every few years. Get planters off the membrane and onto feet with a tray, or better, off the roof structure over habitable space entirely. Keep the drains and scuppers clear. Have the membrane and the parapet upstands looked at by somebody who does roofs. None of that is pest control and all of it is the reason the ants are there.

Why does the downhill house get them after rain?

Because the slope moves water fast, and everything it moves has to arrive somewhere.

Park Slope falls continuously from Prospect Park West down across the avenues to Fourth Avenue and the Gowanus lowland at sea level. Water sheds off the upper slope quickly, and it concentrates wherever that gradient is interrupted — a rear extension, a raised patio, a retaining wall between two terraced gardens, a paved yard regraded decades ago by somebody thinking about seating rather than drainage.

Small ants respond to that directly. A colony living in soil, under paving, in an areaway joint or beneath a yard slab does fine until its site floods. When it does, foragers and sometimes the whole colony move to the nearest dry sheltered space, which is usually a building. The pattern residents describe is very consistent: nothing for weeks, then a heavy storm, then ants in a kitchen or a bathroom for a few days, then nothing again.

That is a drainage report, not an ant problem. The useful response is to go outside during the next storm and watch. Where does the water come from, where does it pond, and what is it running against? On a terraced garden with a retaining wall partway down, water frequently sits behind the wall and saturates the ground on the upper level. On a lower-slope property it may simply be that everything from four gardens uphill arrives at your rear wall.

The corrections are the same ones that serve every other pest problem on this stock: get the downspouts discharging away from walls rather than at their base, clear and test the areaway and rear yard drains, lower beds that have been built up against masonry, and fix the interrupted gradient rather than mopping up what it produces.

There is a second post-rain pattern worth naming, and it belongs to the bottom of the slope. Where drainage is under stress — and the Gowanus system has fourteen combined sewer overflow points on a waterway where heavy rain floods streets and causes sewage lines to overflow — a storm loads the network and things back up. That is usually a rodent and drain-fly conversation rather than an ant one, but it is the same underlying event, and it is dealt with under rodent control in Park Slope.

Which ants are actually in a Park Slope kitchen?

Four, mostly, and the identification changes the work rather than just the label.

Carpenter ants are the large ones, and the ones that matter structurally. They are a moisture and decay finding. Nest sites are wet timber: the cornice void, the sill, the extension seam, a chase around a leaking waste line, a deck sleeper, a tree. Frass is the diagnostic — coarse excavated material with insect fragments and bits of insulation through it, accumulating below a gallery. Activity peaks at night. Winged forms indoors in spring mean a mature colony somewhere in or against the building.

Odorous house ants are small, dark and very common in city housing. They follow moisture indoors, they will nest in wall voids, under sinks and behind trim, and their defining behavior for this work is that they bud. A stressed colony splits and disperses into satellites rather than dying. That is the mechanism behind the most common complaint in this trade — that spraying made it worse — and in an attached house it can push a satellite through a party wall into the building next door.

Pavement ants work the joints in areaway paving, garden slabs, stoop treads and the gaps around a cellar entrance. They nest in soil beneath hard surfaces and push out small craters of excavated grit at the joints, which is the easiest identification in this list. Indoors they are a foraging problem rather than a structural one, and they arrive from the areaway or the yard at ground level.

Pharaoh ants are the one that changes the method completely. They are very small, pale, and they are a multi-unit building problem: they nest in wall voids, in warm service runs and around plumbing, they have many queens, and they disperse aggressively when disturbed. In a subdivided rowhouse a pharaoh ant problem is a whole-building problem from the first day, and it is one of the few situations where getting the identification wrong and applying the wrong approach reliably makes the situation materially worse.

Identification is cheap and worth doing properly. A specimen in a sealed container, or taped to a card, settles in seconds a question that otherwise gets argued about for two visits.

Why does treating the trail extend the problem?

Because a trail is the least important part of a colony, and hitting it with the wrong thing scatters what you were trying to concentrate.

The visible line of ants crossing a worktop is foragers. They are a small fraction of the colony, they are the most replaceable part of it, and they are the part that is currently telling you where the colony is. Killing them removes the information and leaves the population.

With the budding species — odorous house ants especially — it does something worse than nothing. A colony that experiences a repellent chemical barrier or a sudden mortality event at its foraging front responds by fragmenting. Satellites establish in new voids, often further into the building, and what was one trail through one room becomes three trails in three rooms. In an attached house with continuous party wall voids and a retrofitted plumbing chase, “further into the building” can mean the apartment upstairs or the house next door.

The alternative is not complicated. Follow the trail rather than erasing it: it leads to an entry point, and often to a nest. Use materials the foragers will carry back rather than ones that stop them at the boundary, so that the effect reaches the part of the colony that matters. Place them where the ants are already traveling, because ants that have been disturbed will route around anything new in an obvious position. And correct the moisture, which is what decides whether the site is attractive again next year.

There is a specific instruction for residents that follows from all of this, and it is the same one we give for every ant call: do not spray before the survey. A treated, scattered, disrupted population is genuinely harder to work with than an undisturbed one, and the trail you erased on Sunday was the map.

How do you find a colony in a house you can only half access?

By reading the building rather than searching it, and by accepting that on this stock some of the answer is behind a neighbor’s wall.

The physical search is straightforward where access exists. Follow the trail, at night, with a flashlight, in both directions. Look for frass below every suspect void — under a cornice line, at the base of a wall on a top floor, below a chase, under a window with a flat sill. Sound the timber where it is reachable: tapping framing and listening for a hollow note finds excavated wood that looks intact. Look at the moisture sources first and the ant sightings second, because the two lists overlap heavily and the moisture list is shorter.

What limits the search here is the same thing that limits every job in this housing band.

Half the boundary is somebody else’s. The party walls carry the floor joists of both attached buildings, and the joist pockets were never sealed. A satellite nest in wet timber inside a party wall may be being fed by water entering the neighbor’s building.

The house is in several tenancies. In an owner-occupied two- or three-family, the top-floor tenant’s ceiling may be the only place the frass is visible, and getting in there requires notice and arrangement. Where obligations and access rights are unclear, landlord and tenant pest responsibilities sets out how they usually divide, and the general duty on an owner of a multiple dwelling sits at the NYC Housing Maintenance Code and pests. The scheduled investigation and integrated pest management requirements at Local Law 55 include eliminating water sources, which for this insect is the entire job.

Everything is boxed in. Chases, bath panels without hatches, boarded cellar ceilings, retrofitted insulation over the band joist and the sill line. Each of those covers a surface a survey needs to read, and opening a small area is often the only honest way to answer the question.

The roof is a separate expedition. Cornice voids and deck substructures are not visible from anywhere inside, and the survey has to go up.

Where the search genuinely cannot reach — a sealed cornice void, a wall shared with an uncooperative neighbor, a chase behind finished joinery — the correct output is a written statement of what was examined, what was not, and what would have to be opened to answer the rest. That is a more useful document than a confident guess.

What has to be repaired before treatment is worth buying?

The water. All of it, and specifically the item that is wetting the timber the colony is in.

The list is short and it is nearly always the same on this housing stock:

Make the cornice throw water clear again, which means metalwork, fixings and, where the framing behind has decayed, carpentry. Closing a hole in soft timber achieves nothing.

Re-pitch or replace flat sills, and repoint the masonry beneath them.

Open, rake out and repack the rear extension seam in appropriate mortar, and remake the flashing where the extension roof meets the original wall.

Clear the roof drainage, lift the deck and get the planters off the structure, and have the membrane and parapet upstands looked at.

Fix the leak in the chase, which usually means a plumber and usually means opening a boxing that has not been opened since it was built.

Correct the ground: downspouts discharging away from the wall, areaway and rear yard drains clear and tested, beds lowered off the masonry, and the interrupted gradient in a terraced garden dealt with rather than tolerated.

Remove the outdoor colony sites you own: the stump, the sleepers, the buried timber, the firewood against the wall, the dead limb over the roof edge.

That is a building repair list rather than a pest control list, and it is why this work overlaps so heavily with structural exclusion in Park Slope. It is also why we would rather quote it honestly than sell an annual visit against a defect that a roofer could fix once.

When in the year is any of this findable?

Spring for the evidence, summer for the search, autumn and winter for the repair.

March into May is when it declares itself. Colonies resume activity with the warmth, winged reproductives appear indoors at windows, and frass starts accumulating again below active galleries. This is also when the confusion with termite swarmers happens, and it is worth settling on sight: a winged ant has a pinched waist, elbowed antennae, and forewings longer than the hind pair. Keeping a couple of specimens costs nothing and answers it. In a brownstone the answer nearly always comes back carpenter ant, which is the work on this page; if it comes back the other way, the eastern subterranean termite profile identifies it, and Graduate does not take termite work.

April into June is also the small-ant window, particularly after the first heavy storms of the season, when flooded outdoor sites push pavement ants and odorous house ants indoors.

Late spring through late summer is when foraging is at its most readable. Warm nights are the right conditions for following a trail outdoors with a flashlight, which remains the single most productive twenty minutes anybody spends on one of these jobs.

Late summer into autumn is the right window for the moisture repairs, because the weather allows roofing, metalwork and masonry, and because the corrections can be tested against the same autumn’s rain rather than waiting a year to find out whether they worked.

Winter is diagnostic in a different way. A cold still morning shows where warm air is escaping a building, and those points are frequently the same openings and the same failed junctions that let water in. It is also when frass that has been quietly accumulating in an unused top-floor room finally gets noticed.

For the discipline in general, see ant control; the moisture-driven insects share their causes with everything under structural exclusion in Park Slope, and wet cornice voids are frequently also wildlife management in Park Slope territory. For the neighborhood picture across every service, see pest control in Park Slope, with the New York City hub for how city work is organized. Brooklyn Heights is the same problem in older rowhouse fabric; on Long Island, Garden City shows what the same insect does with deep Victorian cornices on detached houses.

If you have found coarse shavings under a ceiling or a trail you cannot follow, tell us where it is and what time of day you see it.

Related

• Ant Control — how we approach it
• All pest control in Park Slope
• Rodent Control in Park Slope
• Cockroach Control in Park Slope
• Bed Bug Treatment in Park Slope