Storm Pollution Needs a Measure of Water as Well as Chemistry

Storm pollution becomes easier to misread when more water enters the river. A sample may show dilution while the storm carries a larger mass of pollutants downstream. San Diego State University’s September 16 account of new watershed research puts that distinction at the center of an observed relationship between storm flow and pollutant transport.

The researchers examined storms sampled between 2016 and 2024 in the San Diego River and Alvarado Creek, a highly urbanized tributary. They combined discharge and water-quality observations to investigate what moved through the waterways during an event. The result directs attention beyond the chemical strength of an individual bottle of water.

Concentration describes the sample; load describes the movement

Concentration expresses how much of a substance is present in a given volume. A load concerns the mass transported over a period. To move from the first to the second, the amount of water flowing through the channel has to remain part of the calculation.

A simple hypothetical illustrates the difference. If concentration falls by half while the volume passing a point triples, the transported mass can rise. Neither observation contradicts the other: they describe different aspects of the event. The arithmetic does not establish what happened in any particular storm, but it explains why dilution alone is an inadequate account of downstream exposure.

USGS storm-runoff work has long combined constituent measurements with flow to estimate event loads. The San Diego findings apply that distinction to a local watershed and show why a monitoring program built around concentration alone could miss an important contribution. A cleaner-looking sample is not sufficient evidence that the storm exported less material.

The tributary can matter beyond its size

The study found a statistically significant association between maximum storm discharge and cumulative pollutant transport. It also identified high dissolved organic carbon and nitrogen fluxes per unit area in Alvarado Creek, compared with the larger river watershed. The university’s account describes the tributary’s catchment as much more heavily urbanized.

That result changes the question asked of a small creek. Its area is relevant, but area alone does not determine its contribution. A small, intensely urban catchment can carry a disproportionate share of particular pollutants during storms. Monitoring only the main river would show material passing downstream without necessarily identifying the tributary contribution that helped produce it.

The finding is bounded to the monitored watersheds, events and constituents. It does not provide a general multiplier for every urban stream, or a forecast for an unobserved future storm. Its practical value is to justify closer attention to tributary loads, supported by measurements at the appropriate places and times.

Why an early sample may miss the continuing input

The reported observations did not consistently follow a conventional first-flush pattern in which concentrations peak early and then decline. Elevated concentrations persisted during portions of events, and pollutant flux continued rising with larger storms. That leaves less support for treating all contamination as a finite stock quickly washed away at the start.

The university connects the findings with earlier watershed evidence pointing toward continuous sources, including leakage from sanitary sewers. The latest study’s storm relationship is not a pipe-by-pipe attribution. It strengthens a diagnostic question about persistent inputs; it does not identify every defective asset or quantify the benefit of repairing one.

For a monitoring team, the consequence is concrete. Sampling needs to cover enough of the storm to describe transport, and streamflow observations need to accompany the chemistry. For an infrastructure team, a high event load can help identify where further investigation is warranted. The San Diego work makes those two activities speak to each other without pretending that a measured load already specifies the repair.