Using Sensors to Identify Point Sources of Pollution in the City of Atlanta’s Combined Sewer System
Bibliographic record
Abstract
Agencies often look for efficient and inexpensive ways to lower the risk of unauthorized discharges into their combined sewer system, specifically pertaining to water quality monitoring and finding point sources of pollution. The City of Atlanta needs to identify sources of specific pollutants (heavy metals and total dissolved solids) in the stormwater runoff entering the City’s combined sewer system. The traditional approach would be to deploy manual samplers at multiple sites based on potential pollutant sources, such as proximity to an interstate highway or an industrial site, to try to capture samples after a rain event. Although automation exists to collect grab samples, collection at the correct time requires significant coordination around frequent sample collection. This may lead to an unknown probability of success in understanding if and where there is a problem. The City of Atlanta is taking an innovative approach to identify point sources of specific pollutants in stormwater runoff entering the City’s sewers. The City and Stantec collaborated on a solution to preliminarily characterize sites prior to more expensive and expansive sampler deployment. This lower cost solution uses Internet-of-Things sensors to capture electrical conductivity and total dissolved solids water quality readings. Inexpensive sensors captured water quality every 15 min, and the findings indicated where to deploy samplers at sites that warrant further investigation. The project team analyzed the data using Stantec’s Altitude Operational Services Platform to correlate data with rain events. A multi-pronged approach was used for early identification, sensor logistics, sensor calibration, and to show how the technology behind the platform is helping to improve the combined sewer system that 1.2 million people depend on daily. The next steps in this machine learning journey include how predicting pollutant concentrations for future rain events will help to confirm point sources of pollution, how the system reacts during first flush rain events, and understanding impacts on City systems. This is an example of using machine learning to find the right balance of smart sensors to build on the City’s solution to stormwater and wet weather water quality fluctuations.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".