Turbidity Spike at a Two-Intake River Plant
A 4.0 MGD conventional surface water plant draws from a river through two intakes: a fixed low-level intake 6 ft below normal pool and a floating intake that follows the surface. The plant is producing 3.2 MGD. A 2.5-inch rain fell over the watershed in six hours; the river is rising 0.8 ft/hr and raw turbidity jumped from 9 NTU to 140 NTU in 90 minutes, with the floating intake reading 210 NTU. Finished water turbidity is 0.14 NTU and creeping up; alum feed is at its normal 18 mg/L and the jar tester is not set up. Raw alkalinity is 55 mg/L as CaCO3 at pH 7.4. The clearwell holds about six hours of production at current flow, and the operator on duty is alone until relief arrives in four hours.
1. Rank your first hour of actions and explain what makes each one urgent.
Read the worked answer
Order the actions by irreversibility of harm and by lead time, not by convenience.
- Move to the low-level intake (or blend). The surface intake is pulling 210 NTU and rising; at depth the river is typically carrying a fraction of that. Take a quick turbidity reading at the low intake to confirm, but the cost of waiting is a plant full of solids.
- Start jar tests immediately. Each round takes 30-45 minutes, and the coagulant dose is the controlling variable. Delaying the jar test means you will be guessing doses during the worst part of the event.
- Make a modest, safe dose increase now (e.g., 18 to 24 mg/L alum) rather than waiting for results. Finished water is already drifting and the clearwell gives you roughly six hours of buffer, not six hours of comfort.
- Watch alkalinity and pH at the higher dose (alum consumes roughly 0.5 mg/L alkalinity per mg/L of alum), and watch each filter's effluent, not just the combined value.
- Verify the raw water screens and pumps are clear of debris, call in help, and log times, doses, and readings as you go.
- Consider trimming plant output to 2.4-2.8 MGD to lower filter loading and extend detention - but only after you confirm storage can cover the difference.
The governing idea: protect the finished water first, then the decisions with long lead times.
2. The jar test returns: 18 mg/L gives a pin floc with settled turbidity of 14 NTU; 34 mg/L gives good floc but settled turbidity is still 6 NTU and settled pH is 6.9. What are your options, and what limits them?
Read the worked answer
Options, roughly in order of cost and speed:
- Raise the dose further - but the dose-response curve is flattening, and 34 mg/L of alum has already consumed about 17 mg/L of your 55 mg/L alkalinity. Pushing much past 40-45 mg/L risks driving pH below 6.5, where alum floc performs poorly and you create a corrosion problem downstream.
- Correct pH back up with caustic or lime to roughly 7.2-7.4. This is often the cheapest way to make a higher alum dose actually work, and it protects the distribution system.
- Add a coagulant aid polymer at 0.1-0.3 mg/L to build larger, tougher floc that settles and does not shear through the filters.
- Change coagulant (PACl or ferric) if the raw water character has genuinely shifted.
- Reduce hydraulic loading so sedimentation has more time; settled water at 6 NTU will blind the filters quickly at design rate.
- Keep using the better intake and consider roughing/pre-settling if you have any available basin volume.
The limit that governs: whatever you choose, the filters have to be able to polish it. Settled turbidity of 6 NTU against a 0.3 NTU combined filter effluent limit is a short runway, so pair the chemistry change with a flow reduction.
3. What must you record during this event, and when does the Illinois EPA need to hear from you?
Read the worked answer
Record continuously and contemporaneously: raw turbidity at both intakes, settled turbidity, chemical doses and any changes with timestamps, jar test sheets (dose, pH, alkalinity, settled turbidity), plant flow, individual filter and combined filter effluent turbidity, backwash times and durations, and every operational decision with its rationale. These records are the difference between a defensible event and a violation you cannot explain.
Notify the Illinois EPA as soon as you know, or reasonably should know, that a turbidity limit was not met - the monthly report is not a substitute for prompt notification. The limits that matter for a conventional plant: combined filter effluent at or below 0.3 NTU in at least 95 percent of monthly measurements and never above 1 NTU, and individual filter effluent excursions above 1.0 NTU, which carry their own follow-up obligations.
Practical test: if you would not want to explain the gap in a sanitary survey, pick up the phone. Also notify your superintendent and, if the situation may force reduced pressure or a precautionary advisory, coordinate that message early rather than at the end.