Interesting! could you post more details on what kind of lead test is used and what your sampling protocol is? thanks! looking forward to learning more.

Hi @nshapiro, We started out using computer vision to analyze simple strip tests like you can get at your local hardware store for their ease of use, immediacy of results and low cost. However, after doing a controlled study comparing the strip tests to lab-tested samples (results published at AWWA ACE17 https://docs.google.com/presentation/d/1tWOEIZ0yhNs9g3sK5bJtTwx7ldODiI5RVTMbwwwxxHw/edit?usp=sharing) we concluded that these tests were insufficiently reliable. So we instead pivoted to utilize a rigorous early AM 1st & 2nd draw sampling with 1-2 minute flush protocol and processing the samples in a lab. By utilizing web-forms on the tester's smart phone, we're automatically capturing the time of day, duration of flush, geolocation, and camera images to assure a high level of data integrity and traceability. In the bigger picture, I see crowdsourced testing/mapping of our public resources as being somewhat independent of any one test type and can imagine it changing as new technologies develop. It may even be possible to think about a consortium of different test types being utilized to optimize resources so that faster/lower-cost tests could help more quickly identify issues in specific locations and points in time that could then be followed up with more detailed testing methods. Hopefully that answers your question, but happy to discuss more. Cheers, Sean

Very cool! it would be great to have more of that info on your sampling and analyzing approach on your website as most project don't have as advanced a method as yours--it would maybe help for buy-in. We have built some colorimetric analysis apps that have been validated in an exposure chamber (results are under review rn) for HCHO with good success. here's the code if its of interest https://github.com/publiclab/SmART-Form . having multiple test types imho would be tricky especially on the low cost side of things, but with your tenacity maybe it could work out! Thanks for the detailed response and god luck as it moves forward!

Hi @smmontgom , I enjoyed reading your poster on your previous sampling protocol. Thank you for sharing! I am a big fan of publishing data in this way, and creating a public map about public water fountains sidesteps the privacy issues faced when trying to share environmental health data collected in people's homes.

Could you please share more information about the sampling method you currently use?

• What actual kit do you send out?
• What lab processes it?

It's good practice to share these details transparently without requiring someone to sign up as a beta tester.

Hi @liz, Thanks for your compliments. Our testing utilizes a standard sample collection kit with 2 bottles [1st draw, flush, 2nd draw] that is then tested by a lab certified by the NYC DEP. I agree with you about adding more info to our website, including nitty gritty details about testing process as well as more big picture info for the uninitiated to understand the issue with lead in drinking water, how it got there, and what the health effects are. We just finished the website, webforms, and backend infrastructure to get the whole thing live updating with data as it comes in. It's just a small group of us working nights and weekends since about this time last year, but we certainly have a host of improvements and goals for the future. Thanks for your helpful comments to improve transparency!