search

NIEHS / beethoven

BEETHOVEN is: Building an Extensible, rEproducible, Test-driven, Harmonized, Open-source, Versioned, ENsemble model for air quality
https://niehs.github.io/beethoven/
Other
2 stars 0 forks source link

Prediction at finer resolution in selected locals #281

Closed sigmafelix closed 3 weeks ago

sigmafelix commented 5 months ago

I read an article on local(ized) difference between air pollution prediction models at moderate and high resolution (Chambliss et al. 2023), whose co-authors worked with many times with @Spatiotemporal-Exposures-and-Toxicology. Although the data they used is based on the campaign in 2015-2017, I think we could collect local data such as PurpleAir to produce predictions at finer resolution (e.g., 100 meters) in the future. Of course this issue should be included on long-term goals and it needs extensive considerations on calibration, data availability, and time/financial costs. Is the mobile sensor measurement campaign going on recently @Spatiotemporal-Exposures-and-Toxicology ?

kyle-messier commented 5 months ago

@sigmafelix Yes, Sarah Chambliss was a PhD student with my postdoctoral advisor, Josh Apte, while I was there. Hyper-local air pollution prediction has predominately been done with monitoring such as a mobile monitoring or dense networks. I think adding purple air would be a great extension - why I think having open source and access code base is so critical. It will make this extensions more feasible. But as you say, definitely another analysis and paper for the future. Perhaps something you could write a K99 around 🚀 😜

kyle-messier commented 3 weeks ago

@sigmafelix Let's discuss this whenever we get to future projects or grant applications. As it pertains to beethoven, I am closing.