11.4 Stage 1 & Stage 2 Disinfection Byproduct Rules
Key Takeaways
- The Disinfection By-Product Rule is a named sub-topic in the SWRCB distribution Expected Range of Knowledge, and Disinfectant By-Products is separately named under Disinfection.
- Stage 1 sets MCLs of 0.080 mg/L for total trihalomethanes, 0.060 mg/L for the five haloacetic acids, 0.010 mg/L for bromate, and 1.0 mg/L for chlorite.
- Maximum residual disinfectant levels are 4.0 mg/L for chlorine and chloramines as an annual average and 0.8 mg/L for chlorine dioxide.
- Stage 2 changed compliance from a system-wide running annual average to a locational running annual average calculated separately at each monitoring site.
- Byproduct formation increases with higher total organic carbon, higher chlorine dose, longer contact time, higher temperature, and higher pH for trihalomethanes.
The Fundamental Trade-Off
Chlorine reacts with natural organic matter (NOM) - humic and fulvic acids from decaying vegetation - to form halogenated byproducts. Where bromide is present in the source, brominated species form as well, and California's coastal and Delta-influenced sources carry significant bromide. The regulatory answer has never been to stop disinfecting; it is to remove the precursors, minimize the dose and the contact time, and use disinfectants that form fewer byproducts.
The Regulated Families
| Group | Members | MCL |
|---|---|---|
| Total trihalomethanes (TTHM) | Chloroform, bromodichloromethane, dibromochloromethane, bromoform | 0.080 mg/L (80 µg/L) |
| Haloacetic acids (HAA5) | Monochloroacetic, dichloroacetic, trichloroacetic, monobromoacetic, dibromoacetic | 0.060 mg/L (60 µg/L) |
| Bromate (from ozone + bromide) | - | 0.010 mg/L |
| Chlorite (from chlorine dioxide) | - | 1.0 mg/L |
Maximum Residual Disinfectant Levels (MRDLs)
MRDLs limit the disinfectant itself, not its byproducts:
| Disinfectant | MRDL | Basis |
|---|---|---|
| Chlorine | 4.0 mg/L as Cl₂ | Running annual average |
| Chloramines | 4.0 mg/L as Cl₂ | Running annual average |
| Chlorine dioxide | 0.8 mg/L | Daily, at the entrance to the distribution system |
[!NOTE] MRDLG vs MRDL vs MCLG vs MCL. The maximum residual disinfectant level goal (MRDLG) and the maximum contaminant level goal (MCLG) are non-enforceable health targets. The MRDL and MCL are the enforceable numbers. Chlorine dioxide is the odd one out - it is measured daily at the entry point rather than as an annual average, because its acute health endpoint differs.
Stage 1 (1998/2002)
Stage 1 established the MCLs and MRDLs above and added precursor removal requirements. Compliance was based on a running annual average (RAA) of all samples system-wide - which meant a system could average away a chronically high site with several low ones.
Enhanced Coagulation and Enhanced Softening
Conventional filtration plants must remove a specified percentage of total organic carbon (TOC) between the source and the combined filter effluent, based on a matrix of source TOC and source alkalinity:
| Source TOC (mg/L) | Source alkalinity 0-60 mg/L | 60-120 mg/L | > 120 mg/L |
|---|---|---|---|
| > 2.0 to 4.0 | 35.0% | 25.0% | 15.0% |
| > 4.0 to 8.0 | 45.0% | 35.0% | 25.0% |
| > 8.0 | 50.0% | 40.0% | 30.0% |
The pattern is intuitive once you see it: higher TOC means more removal required; higher alkalinity means less removal required, because high-alkalinity water resists the pH depression that makes enhanced coagulation work. Plants with source TOC of 2.0 mg/L or less are not subject to the requirement.
Systems can also comply through alternative compliance criteria, such as source TOC below 2.0 mg/L, treated TOC below 2.0 mg/L, low SUVA, or already-low TTHM and HAA5 levels.
Enhanced coagulation in practice means increasing coagulant dose and depressing pH into roughly the 5.5 to 6.5 range, where metal-humate complexation and adsorption onto the metal hydroxide floc are strongest. The costs are real: more coagulant, more sludge, more caustic downstream to bring pH back up for corrosion control, and potential effects on filter performance.
Stage 2 (2006) - The Change That Mattered
Stage 2 kept the same MCLs but changed where and how compliance is calculated:
| Stage 1 | Stage 2 | |
|---|---|---|
| Compliance metric | Running annual average (RAA) of all sites combined | Locational running annual average (LRAA) - each site averaged separately |
| Site selection | Representative sites | Sites selected from an Initial Distribution System Evaluation (IDSE) targeting high TTHM and high HAA5 locations |
| Effect | High sites could be offset by low sites | Every site must comply on its own |
Worked example. A monitoring site returns quarterly TTHM results of 62, 78, 104, and 88 µg/L. LRAA = (62 + 78 + 104 + 88) / 4 = 332 / 4 = 83 µg/L - which exceeds the 80 µg/L MCL at that location, and it is a violation even if every other site in the system averages 40 µg/L.
[!IMPORTANT] Stage 2 is why distribution system water age became an operational priority. Under Stage 1, a system could tolerate a stagnant tank at the end of a long transmission main. Under Stage 2, that tank's monitoring site has to comply by itself, and TTHM keeps forming as long as free chlorine and organic precursors are in contact. Reducing water age is a DBP control strategy.
Stage 2 also added the operational evaluation level (OEL): if the sum of the two previous quarters plus twice the current quarter, divided by four, exceeds the MCL at any site, the system must conduct an operational evaluation and submit it - an early warning that fires before the LRAA itself goes over.
Operational Control Levers
| Lever | Mechanism | Trade-off |
|---|---|---|
| Remove precursors - enhanced coagulation, GAC, membranes | Less NOM to react with | Chemical cost, sludge, pH management |
| Move the point of chlorination downstream (post-filter rather than pre-coagulant) | Chlorine contacts far less NOM | Lose pre-oxidation benefits and some CT |
| Convert to chloramines for the distribution residual | Chloramine forms far less TTHM/HAA5 | Nitrification risk; weaker disinfectant; dialysis and aquatic notifications |
| Lower the chlorine dose to the minimum that meets CT | Less reactant | Must not compromise inactivation or residual |
| Reduce water age - tank turnover, unidirectional flushing, looping dead ends | Less contact time | Water loss during flushing |
| Aeration of finished water | Strips volatile THMs | Does not remove HAA5, which are non-volatile |
| Use ozone or UV for primary disinfection | Little or no halogenated byproduct | Ozone forms bromate with bromide present; UV gives no residual |
[!TIP] THMs and HAA5 respond differently, and that is heavily tested. THMs are volatile and rise with higher pH; HAA5 are non-volatile and form better at lower pH. Aeration and spray systems strip THMs but do nothing for HAA5. HAA5 are also biodegradable in the distribution system, so HAA5 can actually decline with water age while TTHM climbs. If a system is failing HAA5 near the plant and TTHM at the far end, that pattern is chemistry, not a sampling error.
Quarterly TTHM results at one Stage 2 monitoring location are 70, 85, 96, and 81 micrograms per liter. Is this location in compliance?
A plant installs finished-water aeration to strip trihalomethanes. What effect will this have on haloacetic acids?
A conventional plant has source water with a total organic carbon of 6.0 mg/L and alkalinity of 45 mg/L as calcium carbonate. What percentage of TOC must enhanced coagulation remove?