Explore the starter data
We used AI-assisted research to investigate the published sources, then created 18 non-sensitive synthetic records shaped by what those sources expose. It is starter data for exploration—not real service data or evidence.
Read the note on this dataset
Eighteen invented weekly effluent readings standing in for the monitoring feed a wastewater operator already holds, with six lettered works, a banded daily flow, and ammonia and phosphorus in milligrams per litre, so anomaly triage can be discussed without operational or network detail.
Every record in the file
Inspect, compare, understand
Use 18 synthetic weekly readings across six works to explore unusual-reading detection without operational network data. Switch views without losing the description, source file, or limitations on this page.
Choose a working view
| SYN-WW-01 | Works A | 2025-W18 | 10000-20000 m3/day | 1.4 | 0.8 |
|---|---|---|---|---|---|
| SYN-WW-02 | Works A | 2025-W19 | 10000-20000 m3/day | 1.9 | 0.9 |
| SYN-WW-03 | Works A | 2025-W20 | over 20000 m3/day | 3.6 | 1.5 |
| SYN-WW-04 | Works B | 2025-W18 | 5000-10000 m3/day | 0.9 | 0.6 |
| SYN-WW-05 | Works B | 2025-W19 | 5000-10000 m3/day | 1.1 | 0.7 |
| SYN-WW-06 | Works B | 2025-W20 | 5000-10000 m3/day | 1 | 0.6 |
| SYN-WW-07 | Works C | 2025-W18 | 2500-5000 m3/day | 2.2 | 1.1 |
| SYN-WW-08 | Works C | 2025-W19 | 2500-5000 m3/day | 5.8 | 2.4 |
| SYN-WW-09 | Works C | 2025-W20 | 2500-5000 m3/day | 2.4 | 1.2 |
| SYN-WW-10 | Works D | 2025-W18 | 1000-2500 m3/day | 0.7 | 0.5 |
| SYN-WW-11 | Works D | 2025-W19 | 1000-2500 m3/day | 0.6 | 0.4 |
| SYN-WW-12 | Works D | 2025-W20 | 1000-2500 m3/day | 0.8 | 0.5 |
| SYN-WW-13 | Works E | 2025-W18 | 250-1000 m3/day | 3.1 | 1.8 |
| SYN-WW-14 | Works E | 2025-W19 | 250-1000 m3/day | 3.4 | 1.9 |
| SYN-WW-15 | Works E | 2025-W20 | 250-1000 m3/day | 3 | 1.7 |
| SYN-WW-16 | Works F | 2025-W18 | under 250 m3/day | 4.2 | 2.1 |
| SYN-WW-17 | Works F | 2025-W19 | under 250 m3/day | 4 | 2 |
| SYN-WW-18 | Works F | 2025-W20 | under 250 m3/day | 9.6 | 3.3 |
AI-assisted research helped identify and interpret the published sources. We then created a small, non-sensitive synthetic dataset shaped by the information those sources expose. It is for exploration—not evidence, training, or operational decisions.
Method note
How it was prepared
AI authored fictional flow, ammonia, and phosphorus readings, including two weeks that differ from their site's pattern.
Conditions for use
What this dataset cannot tell you
- The readings are invented and the works are letters, so nothing here describes any real site, network, or discharge.
- Weekly figures per works hide the rainfall, hydraulics, and sensor faults that decide what an unusual reading actually means.
- Two measures cannot identify what is in the water, and a real feed also carries the gaps, drift, and maintenance events this tidy file does not.
- It is invented. No record describes a real person, place, case, or measurement, and no result computed from it is evidence that anything works in a live service.