Fairview Water Works: Lead Action Level Exceedance and a Corrosive Water Problem
Fairview Water Works serves 2,400 people and collects 20 standard lead and copper tap samples every six months. The most recent round, sorted from lowest to highest lead concentration in mg/L, is: 0.002, 0.003, 0.004, 0.005, 0.006, 0.007, 0.008, 0.009, 0.010, 0.011, 0.012, 0.013, 0.014, 0.015, 0.016, 0.018, 0.021, 0.024, 0.031, 0.045. The highest copper result is 1.6 mg/L. The 0.018 mg/L sample came from a site on the sampling pool, but the resident reported running the kitchen tap about five minutes before filling the bottle, and the site is a 1990s apartment building with no lead service line and no lead plumbing. Finished water pH is 7.3 and total alkalinity is 22 mg/L; the calculated Langelier saturation index is −1.2. Nothing in the treatment process changed since the last round, when the 90th percentile was 0.011 mg/L and pH averaged 7.8 - soda ash feed was later shut down to reduce scaling complaints.
1. Calculate the 90th percentile lead result and state what it means for the system.
Read the worked answer
Method: order the results and take the value at the 90th percentile position. With 20 samples, 0.90 × 20 = 18, so the 90th percentile is the 18th value in ascending order.
- 18th value = 0.024 mg/L
- Action level = 0.015 mg/L
- 0.024 > 0.015, so the lead action level is exceeded. Note that two samples (0.031 and 0.045) sit above the action level and three exceed the 0.015 line overall - this is not a single outlier driven result.
What that triggers, broadly:
- Public education on lead within the rule's deadline (typically 60 days), delivered to all bill-paying customers, with the required health-effects language and steps consumers can take.
- Continued six-month monitoring and submission of water quality parameter results.
- Evaluation of whether corrosion control treatment is optimal; if it is not, the system must conduct or complete corrosion control treatment work and demonstrate it works.
- A lead service line replacement obligation if the action level is exceeded again in later rounds, plus the inventory work that establishes where the lines actually are.
Confirm the specific deadlines in your state's adoption of the rule before you commit them to a letter. Copper: a 1.6 mg/L single high value does not by itself prove an exceedance of the 1.3 mg/L action level - you must run the same 90th percentile calculation on the full copper dataset.
2. What do you do with the flushed sample, and how does removing it change the result?
Read the worked answer
That sample is invalid for compliance purposes. A lead and copper tap sample must be a first-draw sample after the water has stood undisturbed in the plumbing for at least six hours. Five minutes of running the kitchen tap flushes the service line and the building plumbing, so the bottle measures water that is not representative of what the resident actually drinks. The effect is to make the water look better than it is - flushing lowers the reported lead.
Document the invalidation (site, date, what the resident said), collect a replacement sample from a qualifying site, and do not quietly drop the number without a record.
Recompute with 19 valid results: 0.90 × 19 = 17.1, which rounds up to the 18th value. Removing the 0.018 result leaves the ascending order ending 0.016, 0.021, 0.024, 0.031, 0.045, so the 18th value is 0.031 mg/L - the exceedance gets worse, not better.
The lesson to carry into the field: the credibility of the whole dataset rests on sampling instruction and site eligibility. Train the resident, leave written first-draw instructions, and check the sampling pool for tier eligibility so you are not defending a number you cannot stand behind.
3. Explain how the treatment change caused this, and what corrosion control approach you would propose.
Read the worked answer
The water chemistry tells the story. pH fell from about 7.8 to 7.3 after the soda ash feed was shut down, alkalinity is only 22 mg/L, and the Langelier saturation index of −1.2 says the water is aggressive toward metal and cement surfaces. Low pH and low alkalinity mean the protective scale inside old lead service lines and lead solder stops forming and starts dissolving. The jump in the 90th percentile from 0.011 to 0.024 mg/L in one round, with no other process change, points at the soda ash decision.
Options, roughly in order of cost and speed:
- Restore pH to the previous range with soda ash or caustic, and consider raising alkalinity as well. This is the fastest reversibility and is what the historical data supports.
- Add a phosphate-based corrosion inhibitor (orthophosphate or blended phosphate) if pH adjustment alone is not enough or creates scaling problems elsewhere.
- Verify with water quality parameter monitoring at the tap and at entry points, and confirm the change through the next monitoring round, not just a day of bench readings.
Sequencing matters: a corrosion control change needs state review and a demonstration that it works, so treat the soda ash shutdown as an operational decision to reverse now and a treatment change to document carefully, not a quick tweak. Also finish the lead service line inventory - you cannot plan replacement without knowing where the lead is.