The Overflow Has a Clock: Drawing Storm-Sewage Disclosures as Episodes in Time

Storm overflows are intermittent by design: they operate when rainfall or snowmelt overwhelms the sewer network, and they discharge diluted sewage into rivers and coastal waters. In 2026, the public record is richer than ever. Event Duration Monitoring gives spill counts and durations at thousands of sites, and water companies publish annual returns. Yet many public-facing maps still reduce this to a single red dot on a beach or a riverbank. The dot is not false, but it is incomplete. It answers where, and leaves the harder questions — how often, for how long, after what weather, and what was not reported — to the imagination.

A dot is a location, not an episode

On a map, a point is a natural symbol for a fixed asset: an outfall, a pumping station, a bathing water. But an overflow is an episode. Its meaning depends on duration and recurrence. A site that spills once for ten minutes in a wet January is not the same as one that spills forty times across the bathing season. A marker that treats both as a spill invites alarm or complacency. For a swimmer checking a beach in August, the useful question is not whether the outfall exists but whether it has discharged recently, for how long, and under what conditions. For a resident, it is whether the same street-end outfall appears every time heavy rain passes. For a councillor, it is whether the pattern is improving and whether the published record is complete.

Three questions, three visual channels

The first design principle is separation. Frequency, duration and weather dependency should not be collapsed into a single risk score unless the method is fully disclosed and easily inspected. Frequency is a count, and counts need a denominator. A spill count without the length of the reporting period, the number of monitors and the weather of that period is hard to interpret. Duration is a distribution, not a single average: a mean of two hours can conceal many short spills and a few long ones. Weather is a covariate, not an excuse. The rainfall that precedes an overflow is part of the story, but it should be shown alongside the spill record rather than used to explain it away.

A site card can carry these as separate small charts. A count bar shows reported spills per season. A duration strip shows the spread of event lengths. A rainfall panel shows the antecedent conditions. The reader can then form a judgement without being handed one.

Draw the episode, not just the alert

On a site card, a horizontal event timeline can show each reported spill as a bar whose length is duration and whose position is date. A calendar heat strip can summarise a year at a glance: one cell per day, shaded by reported hours, with a separate texture for no data and a neutral fill for a day with no reported spill. The distinction between zero and missing is essential. A blank cell that means no spill is a claim; a blank cell that means monitor offline is a different claim. The legend must say which.

On a phone, daily cells may be too fine. Aggregate to weeks, but keep the daily detail one tap away. The overview should still answer the reader’s first question: has this site spilled recently, and for how long?

Put rainfall in the frame

Rainfall context makes the record legible. A small aligned panel can show daily rainfall for the catchment or the nearest gauge, with the spill bars on the same time axis. The aim is not to overlay two series and imply a simple cause. It is to let the reader see the relationship without being told a single story. Where data allow, annotate the antecedent rainfall window used by the company. Where they do not, say so. Dry-day spills, which occur without significant rainfall, deserve their own label because they raise different questions about network condition and operation.

Make the gaps speak

Public maps often imply completeness. In practice, monitors fail, telemetry drops, sites are added, and some discharges are estimated rather than measured. A disclosure graphic should carry a data-completeness panel: the number of days with valid records, the number of missing days and the reason where known. Use grey hatching for missing periods, not the same colour as zero. Label the series reported spills rather than spills. This is not pedantry; it changes what a councillor can responsibly say in a meeting. A site with a low count and a long monitoring gap is not a clean site; it is an uncertain one.

Show uncertainty without paralysis

Sensor timing and duration estimates carry uncertainty. Start and end times may be rounded; a monitor may detect flow at a threshold rather than measure volume. Interval bars, rounded values and plain-language notes can communicate this. Avoid false precision: if the data are reported hourly, do not print 3.7 hours. Use at least for an ongoing event and reported for a completed one. A short methodology note, written in plain English, is part of the graphic, not an appendix.

Keep the map, but give it a job

The map still matters. It shows proximity to bathing waters, rivers, homes and designated sites. It shows catchment context and the relation between an outfall and the places people use. But the map should be an index, not the whole argument. Each marker can open a site card with frequency, duration, rainfall context and completeness. Consistent scales across sites prevent a dramatic but small outfall from dominating a region. League tables should be avoided unless they are normalised and annotated; a raw spill count rewards sites with more monitors and wetter weather.

Design for the glance and the read

Public information design has to work at two speeds. At a glance, a reader should see today’s status, the most recent reported duration and a seasonal pattern. At a longer read, they should be able to inspect the event timeline, the rainfall panel and the data gaps. Use a colour palette that survives colour-vision deficiency; do not rely on red and green alone. Pair colour with pattern, label and position. Provide text alternatives for screen readers and a table view for those who need the numbers. On mobile, lead with the question the reader came to ask, then offer the evidence.

A storm overflow is not a pin. It is a clock, a weather record and a disclosure. The red dot can stay, but it should be the beginning of the answer, not the answer itself. A visual grammar that separates frequency from duration, shows weather dependency and makes reporting gaps visible gives swimmers, residents and councillors a fair basis for judging recurring risk. That is the work of information design: not to make the data look simple, but to make the complexity readable.