Why it keeps happening
Three things we can measure in the data. None of them proves a cause on its own, but each points at a different kind of infrastructure problem.
A discharge in dry weather is not an exceptional circumstance, so it is not permitted on any reading of regulation 4.
See the 30 →A strong signal, not a finding. We cannot see whether flow is transferred to another works for treatment, and the permitted figure is not always published.
See the 0 →Rainfall across the catchment does not explain a discharge that an overflow's own treatment works did not share.
See the 68 →Where the indicators concentrate
| Overflow | Hours | Dry | What it points to | Verdict |
|---|---|---|---|---|
| Greenacres, ChagfordCSO | 1,631 | 83 | High spill frequency | No action |
| Compton CastleCSO | 645 | 134 | High spill frequency | On record |
| Dunsford STW, Teign ValleySO | 2,471 | 276 | High spill frequency | On record |
| Forde Park/torquay Rd, Newton AbbotCSO | 39 | 216 | High spill frequency | No action |
| Lemon Rd, Newton AbbotCSO | 0 | 99 | High spill frequency | No action |
| Heathfield STW, Newton AbbotSSO | 2,088 | 137 | Very long spills | On record |
| Christow STW, Teign ValleySSO | 1,054 | 193 | Very long spills | On record |
| Homers Ln, KingsteigntonCSO | 105 | 69 | Dry spilling | No action |
| Summerhill pumping station, IlsingtonPS-CSO/EO | 1,364 | 10 | Very long spills | No action |
| Cheriton Bishop STWSSO | 512 | 5 | High spill frequency | No action |
Hours are the last full year; dry spills are since 2020. Verdict is whether a measure is linked to the overflow — see the gaps.
What this analysis cannot see
These indicators are measured from spill timing and rainfall alone. They cannot see misconnections, the quality of effluent from a works that is running normally, flow transferred between works for treatment, a company's maintenance history, or the periods when a monitor was offline. Because a quiet monitor records nothing, every count here is a floor, not a total.