Odor Monitoring for Wastewater Treatment Plants
This is ambient air and fence-line monitoring, not a dissolved sulfide probe in your wet well. We put continuous nodes around the plant and along the property line, where the neighbors actually smell you. Each node reads odorants (H2S, ammonia, methane, mercaptans, TVOC) plus particulate in four size fractions, and a weather station on your property logs wind on the same clock. You get a live dispersion map, an alert before the plume reaches a downwind receptor, and a time-stamped record you can hand an inspector. It answers the one question your corrosion sensors never will: what left the property, when, and from which unit process.
Ambient air at the property line, not sulfide inside the pipe
Vendors pitch two very different products as H2S monitoring. Know which one you are shopping for before you sit through the demo.
- Odorant gases and particulate in outdoor ambient air. H2S, ammonia, methane, mercaptans, TMA, DMS and DMDS, TVOC and CO2, plus PM1, PM2.5, PM4 and PM10.
- Property line, plant access road, and right next to the unit processes that make the smell. Add points downwind when the houses all sit on one side.
- No dissolved sulfide probe in a wet well. No liquid-phase sensor in a force main. No corrosion rate for a concrete interceptor. Different problem, different vendor.
- The plant superintendent, the EHS manager, and whoever takes the call from a council member. Not the collections crew chasing crown corrosion.
- A dissolved reading tells you what's in the water. A fence-line reading tells you what a resident smelled at 5 a.m. and whether it came off your site.
Where plant odor actually starts
Complaints almost never trace back to the whole plant. They trace back to one process, one shift, one wind direction. These are the spots we cover first.
Headworks and preliminary treatment
Screening, grit removal and the drop into the channel strip sulfide out of solution fast. Turbulence does the work. A node here catches the release before the plume crosses the yard, and it flags a septic slug the moment it hits the plant from the collection system.
Primary clarifiers
Scum blankets, weirs and skimmings give off steady low-grade odor across a lot of open surface. Nothing spikes. It just sits there and drifts on a light wind. Continuous readings show when the baseline crept up over a few weeks, and that usually points to scum removal or blanket depth.
Thickening
Gravity thickeners and DAF units hold solids long enough to go anaerobic in warm weather. Overflow and float removal both give off odor. Watch the deck and the downwind side and you can tie a jump to one pump-out, or to a thickener that sat too long.
Dewatering and biosolids storage
The belt press or centrifuge room is usually the strongest single source on the site. Shear drives off sulfide and organic sulfur compounds, and polymer chemistry stacks amines on top. Cake keeps off-gassing in the bunker and in the roll-off box. Cover the loadout pad and the nearest property line.
Lagoons and equalization basins
Big open surfaces, low velocity, seasonal turnover. Spring and fall are the bad weeks. The area is far too big to chase with a handheld, so one fixed node upwind and one downwind give you the real gradient across the basin.
Biofilter and scrubber outlet
Your odor control unit is only as good as its last media change. A node at the outlet and one downwind of it catch breakthrough as it starts, not after the phone rings. Trend the outlet against the plant baseline and you see removal in service, not on the submittal.
Residual ammonia off biosolids behaves nothing like H2S
Plants that solved their sulfide problem still get calls. Ammonia and the amines riding along with it are usually why.
It shows up after stabilization
Digestion and lime addition knock sulfide down and leave ammonia behind. The cake smells different after treatment, and the wording in your odor log changes with it.
Neighbors describe it differently
Rotten egg is sulfide. Sharp, fishy or cleaning-product is ammonia and amines. If the complaint language shifted this quarter, the dominant source shifted too.
It peaks in storage, not at the machine
Cake in a bunker or a container keeps giving off odor for days. The strongest reading often lands on the pad Monday morning, well after the dewatering shift ended.
Polymer and lime dose drive it
Change the cationic polymer or the lime dose and the signature moves within a day or two. Continuous ammonia data gives operations a feedback loop instead of a guess.
One node reads both
You don't need a separate ammonia program. The same node reports ammonia next to sulfide, methane and particulate, on one clock and one screen.
The complaint hits while your night shift is the only crew on site
Daytime sun mixes the air and carries odor up and away. After sundown the ground cools, the air stratifies, and whatever you release stays low and travels. That's why the calls land before breakfast.
- Wind drops off. The plume stops diluting and starts hugging the ground. Property-line readings climb while the plant runs exactly the way it ran at noon.
- The inversion sets up. Cool air traps odor under a warm layer and pushes it out along the terrain. A quarter mile downwind can read higher than the plant road.
- Peak exposure at the receptor, and nobody is out there with a nose. Without continuous data you have no record of the worst hours of the night.
- The sun breaks the inversion and the air clears in half an hour. By 8 a.m. the plant smells fine, the operator finds nothing wrong, and the complaint looks unfounded.
- A 24/7 record with wind on the same timestamp shows the overnight event plainly. You stop arguing about whether it happened and start working on why.
What to do when the call comes in
Most plants answer complaints from memory and a paper log. Here's the sequence once you have a record.
Log the time and the cross street
Get the location and the time as close to the minute as the caller can give you. Everything after this step rides on those two numbers.
Pull the window
Open the same window on the nodes. You want the hour before the call, not the moment you sat down at the screen.
Check the wind
Your on-site weather station gives speed and direction for that hour. If the wind was blowing away from the caller, say so plainly and go look somewhere else.
Trace it back to a process
Line the timing up against the plant log. Dewatering run, thickener pump-out, a hauler on the pad, a blower down. The source node closest to it usually rose first.
Answer with a record, not an opinion
Send the caller, the county and the state agency the same chart. Time, concentration, wind direction, what you found, what you changed.
Watch the fix hold
Keep the window open for the next two weeks. If the change worked, overnight readings at the property line drop and stay down.
How a plant gets covered
Most plants land between three and eight monitoring points. Where your property line runs and where the neighbors sit drive the count far more than MGD ever will.
Smart.Odor at the perimeter
Fixed nodes on the property line and at downwind receptors. This is the layer that answers the complaint and builds the record you hand over.
Smart.Compost at the sources
Fixed multi-gas nodes at headworks, the dewatering building and the cake pad. These tell you which process moved before the perimeter reading moved.
Smart.Mini for surveys
Portable and battery powered. Walk the fence line before you commit to a layout, chase a hot spot, or cover a short construction job.
Weather station on your site
Wind speed and direction measured at the plant, on the same clock as the gas and particulate data. The airport sits too far away to explain a plume at 3 a.m.
Gases and particulate together
Odorants and PM1, PM2.5, PM4 and PM10 come off the same node, in the same enclosure, at the same timestamp. No second network to maintain.
What lands on your desk
A live dispersion map, threshold alerts by text or email, exportable history you can audit, and reports written to survive a public meeting.
What we bring and what we don't claim
You'll ask these on the first call, so here they are up front.
- We have run continuous environmental monitoring at operating plants since 2019, with nodes reporting around the clock rather than on a sampling visit.
- We set thresholds and write reports on an established odor assessment approach, applied to a continuous record instead of a one-off sampling event.
- No EPA, MCERTS or SCAQMD approval, and we do not claim any. This is operational and community evidence. If your permit names a reference method, keep using it.
- We are new to this market and our installed base is in Brazil. If a domestic reference site is a hard requirement for you, say so on the first call.
- No published price list. Point count, site layout and reporting scope set the number, so we quote after we have walked your plant.
- Send us a site plan and your complaint log. We come back with a proposed node layout, a point count and a straight answer on fit.
Frequently asked questions
Is this the same as a dissolved sulfide sensor in the collection system?
No. Those sensors sit in the liquid, in a wet well or a force main, and they exist to manage corrosion and chemical dosing. Ours sit in outdoor air on poles around the plant. Different measurement, different buyer, different problem. Plenty of utilities run both. One protects the concrete, the other answers the neighbor.
Do I still need periodic odor sampling and lab analysis?
Probably, yes. Sampling gives you a defensible snapshot at one moment. Continuous monitoring gives you every hour in between, including the 3 a.m. hours nobody samples. The two work together. We set thresholds on an established odor assessment approach, so the continuous record lines up with what your consultant reports.
How many monitoring points does a wastewater plant usually need?
Three to eight for most sites. The count comes from property line length, how many directions have receptors, and how many separate sources you want to tell apart. A compact plant with one neighborhood to the east needs fewer points than a long site with houses on two sides. We propose a layout from your site plan before we quote.
Can it tell me which unit process caused a complaint?
It gets you close. With a node at each major source, a node at the property line, and wind measured on site, the sequence usually points at one process. Dewatering ran, the source node rose, the wind was out of the southwest, the perimeter node rose a few minutes later. That's a working answer for an operations meeting. It is not a legal determination and we don't sell it as one.
Is the system certified or approved for permit compliance in the United States?
No. We hold no EPA, MCERTS or SCAQMD approval and we claim none. Use it the way plants already use continuous operational data. Catch problems early. Document what happened overnight. Show a county inspector or a state agency that somebody is watching the property line. If a permit condition names a specific method, that method stays in place.
Why measure particulate on the same node as the gases?
Because not every complaint is odor. Lime silo venting, grit and cake loadout, and truck traffic on a gravel drive all throw dust, and residents put dust and odor in the same phone call. Reading PM1, PM2.5, PM4 and PM10 next to the gas data lets you separate them. Sometimes the honest finding is that the odor trace was flat and the dust trace wasn't.
What do you need from us to get a system installed?
A site plan, your complaint history, and a walk of the property line. Nodes need a mount and either power or a solar setup, and they report over cellular, so nothing touches your SCADA network. From there we come back with a layout and a point count. Start at the contact page and we'll put you on a call with an engineer.
Have the log before the next call
Tell us about your unit processes, your biosolids handling and which direction the neighbors sit. We will propose where to measure.
Talk to our team →