Building a Temporary Containment Berm

Why this matters

A containment berm's job is not to hold water in place - it's to redirect flow to one controlled low point where a recovery unit can pick it up. That distinction matters because most berm failures on a real job site don't happen through the middle of a long run, where the material is doing exactly what it looks like it should. They happen at the ends, where a berm meets a wall, a garage door threshold, or a landscape edge, because that transition is where the perimeter actually has to seal against something other than itself. This is the skill the recovering-wash-water standard hands off to; here is how to build one that actually holds under a running wand rather than one that looks right sitting dry.

Each step below is framed by what you lose if you skip it, because on this particular job the ends and the low point are where the real cost sits, not the parts of the build that look the most like work.

Site the low point before you touch any berm material, or you lose the ability to recover anything at all

Walk the work zone and identify the single lowest point where a recovery unit's intake can sit - the point the entire perimeter will direct flow toward. This has to happen before material goes down, because the berm's shape follows from this point, not the other way around. Skip this and you end up building a perimeter first and discovering afterward that no single point inside it is actually low enough to collect against, which means re-building the whole run rather than adjusting one corner.

On a sloped surface, the low point is usually obvious by eye; confirm it anyway with a small pour of plain water and watch where it pools. On a surface that reads as flat, the low point is real but subtle - pour water at two or three candidate spots and compare, because "flat" surfaces virtually never are.

Match the berm material to the surface, or you lose the seal before you ever apply chemical

An absorbent containment berm (a flexible tube or sock, typically filled with an absorbent core) seals well against a rough or slightly uneven hard surface because the material conforms to small gaps. A rigid barrier does not conform the same way and needs a smoother, more consistent surface to seal against. Check the surface you're actually working on - broom-swept concrete, exposed aggregate, asphalt with texture - and confirm the material you're deploying is rated to conform to it, not just rated for "hard surfaces" generically on its packaging.

Skip this check and the failure shows up exactly where you'd expect: a berm that looks continuous but has a thin gap under it every few feet on a textured surface, invisible until water is moving and finds every one of them at once.

Lay the perimeter to a closed loop with one deliberate low-point gap, or you lose control of where the water actually goes

Run the berm material in a continuous closed shape around the work zone, with the only intentional break in the loop at the low point identified in step 1, where the recovery unit's intake sits. Every other point on the loop should be continuous material, seam to seam, with no gaps.

A loop with an accidental second low spot - a dip in the surface the crew didn't check for - creates a second place water wants to go, and that second place has no recovery unit waiting for it. Walk the completed loop once, end to end, checking for any point where the material doesn't sit flush against the surface, before moving to the next step.

Seal every end and every wall transition by hand, or you lose containment exactly where the job needs it most

This is the step most berm failures actually trace back to. Where the berm meets a vertical surface - a building wall, a garage door frame, a fence line - press the material firmly into that corner and check for a visible gap at the base. A berm that runs perfectly along open pavement and then leaves even a half-inch gap where it meets a wall has built a functioning perimeter with an open door in it.

Do the same at any seam where two lengths of berm material join each other. A seam is inherently weaker than the material's own middle, and pressure from moving water finds a weak seam quickly. Overlap seams rather than butting them end to end, and press the overlap down firmly.

Size the loop to the volume the job will actually produce, or you lose the fight to overtopping mid-wash

A berm has a maximum depth of water it can hold before flow simply goes over the top rather than through a gap. Match that depth against the pump's flow rate and how long a section will run before the recovery unit catches up - a rig in the 4 to 8 GPM class run steadily against a berm with a small footprint fills faster than the same rig against a larger enclosed area, because the same volume of water has less surface to spread across. Where the enclosed area is small relative to the flow rate, either enlarge the loop or plan to run the recovery unit continuously from the first pass rather than intermittently, so the water level inside the berm never has the chance to build toward the material's limit.

Flow-test with plain water before any chemical goes down, or you lose the chance to catch a leak while the stakes are still low

Once the loop is complete and sealed, run plain water into the enclosed area - a hose is enough, you don't need the pressure washer running yet - and watch every foot of the perimeter for a leak, paying closest attention to the wall transitions and seams from the previous steps. A leak found now costs a few minutes of re-sealing. The same leak found after chemical is in the water costs a breach response and, depending on what crossed the line, possibly an escalation call.

Confirm the recovery unit's intake, positioned at the low point, is actually pulling water down rather than the loop simply holding a shallow pool with nowhere to drain to. A berm that holds water but doesn't route it to the intake has solved only half the problem.

Monitor and reinforce during the job, or you lose the seal you already proved

A flow test proves the berm holds under a garden hose's volume; the actual wash, especially with a wider fan tip or a section run longer than planned, can move more water than the test did. Watch the perimeter, especially the wall transitions, through the first several minutes of active washing and reinforce any point that starts to weep before it becomes a breach. A berm that passed its flow test is not guaranteed to hold at full working volume, only proven to hold at the volume it was actually tested with.

Break down and dispose of collected material correctly, or you lose the paper trail that shows the job was done right

Once the wash is finished and the recovery unit has emptied the enclosed area, remove the berm material and check it for sediment or residue caught in its own seams and folds. Bag any visibly soiled material and dispose of it as solid waste through your normal route, unless it visibly carries a significant chemical residue - a strong odor, discoloration - in which case it follows the same disposal path as the recovered water itself, not the ordinary trash. Rinse and dry reusable berm material before it goes back on the truck; storing wet material against itself is how a berm degrades early and fails a flow test on a job where it matters more.

Worked example: a driveway apron job

A commercial driveway apron, roughly 20 feet by 15 feet, sealed concrete with a broom-swept texture, bordered on one side by the building wall and open on the other three sides to the parking lot.

Low point sited at the building-wall corner nearest a floor drain the property manager had confirmed was authorized for this job. Berm material: absorbent containment tubes rated for a textured hard surface, laid in a closed loop with the single gap at that corner for the recovery unit's intake.

The wall-side run needed hand-sealing at both ends where the tube met the building - the first attempt left a visible quarter-inch gap at one corner, caught and corrected before the flow test. The three open-side runs sealed cleanly against the pavement texture with no additional work.

Sizing: at a 5 GPM pump run in sections roughly 5 minutes each before the recovery unit fully caught up, the loop's footprint held comfortably below the material's rated depth throughout, confirmed by watching the water line during the flow test rather than trusting the rating alone.

The flow test, run with a garden hose for about two minutes, found the corrected wall corner holding and no other leaks; the recovery unit's intake was confirmed pulling water down rather than letting it pool. During the actual wash, one seam on the parking-lot side began to weep faintly around the halfway point of the job - reinforced with the overlap pressed down again, holding for the remainder.

At breakdown, the tubes carried visible sediment but no chemical odor or discoloration - this job used a low-strength surfactant only - so the material was bagged as solid waste after a rinse, and the tubes themselves were rinsed and dried before returning to the truck for the next job.

How to verify you got this right

Before calling the containment complete, confirm three things independently: the flow test found no leak at any wall transition or seam, the recovery unit's intake demonstrably pulled test water down rather than letting it pool, and the loop's footprint comfortably holds the expected volume at the pump's actual flow rate rather than at a rated capacity you haven't tested against. A berm that passes all three has earned the confidence to run chemical through it; one that's only been eyeballed has not.

References

• See related: Recovering Wash Water on a Hard Surface Job Standard, the procedure this build hands off to.
• See related: Vacuum Recovery Versus Containment Berm Reference, for deciding whether a job needs this build at all.
• See related: Wash Water Containment and Storm Drain Protection Standard, for the lighter-weight drain-sock approach used when a full berm isn't required.