Reviving a biofilter at a pump station for odour control

Pumping stations tucked into suburban streets are unsung workhorses of any sewerage network, but when hydrogen sulphide escapes from the wet well, the local community notices quickly. A failing biofilter is often the culprit, and bringing one back into service is something operators around the country face more often than utility managers would like to admit. The work is weather-dependent and rarely documented, which is exactly why a written case study is useful for the next crew that gets the call.

This case study walks through the rehabilitation of a 30-cubic-metre biological filter serving a sewage pump station in Melbourne's western fringe, where growth from a new housing estate and an expanded sporting oval turned minor odours into regular resident complaints. Within eight months, the team had diagnosed the underlying issues, rebuilt the media bed and returned hydrogen sulphide readings to well below the limits most residents would notice, with a treatment approach other utilities can adapt to their own sites.

The complaints no one could ignore

The first six months after commissioning, the biofilter was barely mentioned during toolbox talks. Then a run of calls came in from a cul-de-sac about 80 metres downwind of the station, and the owner of the local café said her outdoor tables had lost trade every arvo the wind swung the wrong way. A site walk later that week confirmed what the complaints suggested: the air coming off the exhaust stack carried a clear rotten-egg character that residents no longer wanted to put up with.

The catchment feeding the pump station had grown faster than originally modelled, with connections from roughly 800 new dwellings over the previous three years. Septicity increases as retention time stretches in rising mains, particularly through summer when ambient temperatures around the western suburbs regularly push above 35°C. The biology of the existing biofilter could not keep pace with the loading, and a compacted, uneven media surface visible from the manhole told the rest of the story.

Inspecting the bed and finding the root causes

Once the complaint log was filed, the maintenance team scheduled a planned shutdown over a long weekend to lift the lid and inspect the bed up close. The media, originally specified as a 50/40/10 blend of bark, hardwood chip and loam, had settled into a compacted, waterlogged mass almost half its original depth. Irrigation nozzles were partially blocked by calcium build-up, and the irrigation pump was cycling far more than designed because the pressure-switch settings had drifted.

Gas readings downstream of the exhaust confirmed the visual story. Baseline levels at the boundary fence were hovering around 1.5 ppm during peak pumping, well above the 0.1 ppm threshold most Australian utilities use as a community nuisance trigger. A side-by-side comparison with a sister station running a healthy biofilter made the gap obvious: that well-performing site sat comfortably under 0.05 ppm regardless of pump cycling.

Rebuilding the media and the irrigation network

The rebuild was sequenced so the new media could go in while the existing bed was being stripped, cutting the outage to four days rather than two weeks. The crew removed the compacted layer down to the original plenum, jetted the underdrain laterals to clear silt and biofilm that had crept back into the pipework, and replaced two failed irrigation laterals entirely.

New media was installed to a revised specification agreed with the original supplier: a 60/30/10 blend of composted pine bark, coarse hardwood chip and structured loam, with a lower target bulk density than what was originally specified. The tradies installed a drip-style header with larger-orifice nozzles spaced at 300-millimetre centres, mounted on a frame that lifts out for cleaning without emptying the bed. A small VSD was added to the irrigation pump so flow can be dialled back during cooler months when biological activity outside the bed slows noticeably.

Commissioning and the first round of performance data

Initial commissioning ran for 10 days, with irrigation on a 60-second-on / four-minute-off duty cycle and the biofilter ramped up to full load in steps rather than dumping the full H₂S stream through on day one. By day six, gas sampling downstream of the stack showed H₂S readings under 0.1 ppm during pump cycling, and by day nine they were tracking close to the 0.04 ppm seen at the healthier sister station.

Three months in, the team has a credible performance picture. They compare readings against a simple dashboard on the SCADA page using data from the new side-stream gas monitor, and they have started layering in ambient temperature and pump run-time data so they can predict bed stress before it shows up as another complaint. A cost reconciliation put the rebuild at roughly AUD 38,000, about 40 per cent less than a full replacement vessel and blower package would have cost.

Where this fits alongside other treatment upgrades

The team kept records in a format that lines up with how they document other treatment retrofits across the network, and referenced a related uv disinfection case study in their internal knowledge base. Both projects involve retrofitting existing assets rather than buying new ones, and the cost-per-cubic-metre framework they borrowed here started as a spreadsheet on that earlier project. Comparing the two side by side has helped justify future biofilter work on the same kind of whole-of-life grounds.

What worked particularly well was treating the biofilter as one element within an integrated odour and corrosion management plan rather than a standalone project. Reducing sulphide at the pump station reduces concrete corrosion downstream, which reduces the chance of upstream complaints during the next big wet weather event. That wider framing is exactly the kind of argument that resonates with finance teams who tend to weigh capital projects on maintenance cost avoidance rather than raw capital spend alone.

Making the rebuild part of a routine maintenance program

The rebuilt biofilter will only stay healthy if it gets the same attention as the rest of the pump station, so the operators have folded it into their broader preventative maintenance plan. Routine tasks now include quarterly inspection of the irrigation laterals, six-monthly H₂S profiling, and an annual partial media refresh on the top 100 millimetres of the bed to keep it from compacting back into the original problem.

Crew members rotate through biofilter checks alongside other site work, which keeps the skills embedded in the team rather than tied to one specialist. A short visual job-card hangs in the switchroom, summarising what healthy media should look like and what the crew should immediately flag. When the next major service comes around, the contact details of the original suppliers, the media specification and the commissioning checklist are all stored in the same folder, ready to hand.

Scaling the approach across the network

The experience is now feeding into a network-wide screening exercise, prioritising other stations with similar load profiles and age of biofilter media. Five of the highest-risk sites are scheduled for partial rebuilds over the next two financial years, using much of the same pricing structure and the same commissioning playbook. Internal trade pricing for the media blend has already been negotiated to drop the per-cubic-metre cost, which should make the case easier to put to the board for medium-sized stations.

There is also a small but growing conversation with regional councils about whether shared procurement would suit biofilters in towns like Warrnambool and Mildura, where each utility typically has only a handful of these units on the books. Sharing a contractor and a media specification between several rural councils would reduce the unit cost of rebuilds, and a couple of the regional Water Corporations have already put their hands up. The lessons from this Melbourne site may travel further than the team expected.

The real test of any biofilter rebuild sits not in the commissioning report but in what residents notice twelve months down the track. Quiet operation, predictable dosing and clean handover paperwork are the three things that separate a project that lasts from one that ends up back on the team's desk within a year.

Operators across Australia are quietly building up a body of practical knowledge around biofilters, and the case notes from projects like this sit alongside other working examples at industry events and inside local operator networks. The bench strength for the next unexpected rebuild lives in those shared folders.