Raw mean PM2.5: — µg/m³
Cleaned mean PM2.5: — µg/m³
⚠ Couldn't load the 3D engineThree.js failed to load from the CDN. Check your connection and reload.
Real environmental monitoring networks — the low-cost PM2.5 sensors behind citizen-science air quality maps, IoT water-quality buoys, satellite-fed climate stations — never report clean data. Every reading carries per-device calibration drift, dropouts and occasional spikes, and the trustworthy number the public sees is the output of a quality-assurance pipeline, not the raw stream. This simulator renders a live 3D grid of 64 IoT air-quality sensors over a moving pollution field and runs that pipeline in real time: a rolling z-score test rejects outliers per sensor, a gap-fill window reconstructs rejected or missing samples from recent history, and a calibration slider corrects for systematic sensor bias the way a periodic co-location check against a reference instrument would. Live readouts compare the naive raw network AQI against the cleaned AQI so you can see exactly how much of "the data" is actually the QA/QC pipeline talking.