Filter Breakthrough After a Storm
A 4.0 MGD conventional surface-water plant - alum coagulation, flocculation, sedimentation, four dual-media filters - serves a mountain community. A thunderstorm drops 1.8 inches in two hours. Raw turbidity climbs from 6 to 90 NTU and raw alkalinity falls from 110 to 55 mg/L as CaCO3. The operator holds the alum dose at its normal 28 mg/L and keeps all four filters in service at 2.0 gpm/ft². Within six hours, settled turbidity rises from 1.5 to 8 NTU, filter runs collapse from 40 hours to 6 hours, and finished-water turbidity reads 0.18 NTU and rising. Filter effluent particle counts are up eightfold. Clearwell chlorine residual is steady at 1.0 mg/L, and the clearwell holds about 14 hours of system storage.
1. Settled turbidity and filter effluent turbidity are both climbing. What is the most probable mechanism, and what would you verify before touching the chemical feed settings?
Read the worked answer
Most probable mechanism: coagulant demand has outrun the fixed dose, so floc is small and weak (pin floc) and passes the filters.
Evidence:
- Alkalinity fell roughly 50% (110 → 55 mg/L). Alum consumes roughly 0.45–0.5 mg/L of alkalinity per mg/L of alum, so 28 mg/L of alum consumes about 13 mg/L as CaCO3. That leaves buffer, but it also means the coagulant is now being dosed into water whose pH and particle character (storm colloids, organics, clay) are completely different.
- Settled turbidity of 8 NTU means the problem starts upstream of the filters. Filters are doing what they can with a bad feed.
- Short runs plus high particle counts is classic surface blinding followed by breakthrough, not simple media loss.
Verify before changing feed:
- Run a jar test on the current raw water across a dose and pH sweep - do not guess a number.
- Confirm the alkalinity and pH results are real (check reagents, probe calibration).
- Walk the filters: headloss trend, filter-to-waste valve position, media condition (boils, cracks, media loss), turbidimeter calibration.
- Confirm the sedimentation basin is actually settling - sludge blanket depth and hydraulic loading.
If you raise the dose without knowing the pH optimum for the new water, you can make floc worse and drive pH down further.
2. List, in priority order, the operational moves you would make in the next two hours to keep combined filter effluent below 0.3 NTU. Justify each.
Read the worked answer
Priority order - protect the finished water first, then fix the chemistry.
- Cut the hydraulic loading rate. Take one filter out of service or throttle plant flow. At 2.0 gpm/ft² with blindable floc, you are asking the filters to do the clarifier's job. Lower rate = more contact time and a survivable run.
- Start or extend filter-to-waste on any filter with rising effluent turbidity. Keep the dirty water out of the clearwell and out of the distribution system; it buys you hours.
- Run the jar test and adjust coagulant/pH. Likely items: raise alum dose, add a polymer as flocculant aid, or add caustic/lime to hold pH near the optimum range for alum (typically about 6.5–7.0, confirmed by jar test, not assumed). Report your actual numbers.
- Verify clarifier performance. Adjust sludge withdrawal, check flocculator speed, and confirm the settled water is the best you can get before it hits the media.
- Backwash on headloss or turbidity trigger, not on the clock. Do not run a filter past breakthrough to save water.
- Protect residual and pressure. Maintain clearwell chlorine residual and confirm the distribution system is not seeing a turbidity slug. 14 hours of storage gives you a cushion, but only if you are honest about the trend.
Justification theme: every action either reduces loading, keeps bad water out of the clearwell, or fixes the coagulant chemistry. Everything else is secondary.
3. What reporting, notification, and recordkeeping issues should you be thinking about while you work this event?
Read the worked answer
This is both an operational event and a compliance event. Handle them in parallel.
- Turbidity limits. Conventional filtration has a combined filter effluent limit (commonly 0.3 NTU in at least 95% of monthly measurements, with individual filter benchmarks). Do not wait for the monthly math - if you have exceeded the individual filter trigger, find out now what your state's rule requires.
- State notification. Call the primacy agency (in New Mexico, the Drinking Water Bureau) as soon as you know you have an exceedance or a treatment-process problem the system can't correct quickly. Ask them the notification clock rather than assuming.
- Tier assignment. Public notice tier depends on whether this is a violation, a treatment technique problem, or a situation with potential health risk. Get the classification from the regulator and in writing if possible.
- Records. Log: raw and settled turbidity trends, jar test sheets, dose changes with times, backwash records, filter-to-waste periods, clearwell residual, and every phone call. Contemporaneous notes are your defense.
- Customer contact. If turbidity reached the distribution system, expect discolored water complaints - and a credible explanation of what you did and when.
- Post-event. Review why the fixed alum dose was carried into a 15-fold turbidity swing. That is a procedure gap, not just a storm.