Thanks jmh2002. I understand this, however, the forecast visualisation provided did not reflect what was actually happening and, thus, was not useful. My question is more about the reliability of windy.com's "forecast visualisations". Sascha

Hey all

I have been using your PM2.5 air quality data and forecasts to be informed about air quality in Sydney during the bushfires.

This has worked really well until about last week, when your model did not represent reality at all. For example, on 8 January 2020, air quality in Sydney (Darlinghurst) was really bad and PM2.5 went up to above 100 microgram per cubic metre (with hovering around 50 other times during the day), whereas your map showed PM2.5 to be at around 10 microgram per cubic metre most time during the day. The model also did/does not forecast or show the poor PM2.5 situation in Sydney yesterday (11 January) or today 12 January (just now at 10:45am, your map shows 9 microgram per cubic metre whereas AirRater shows 51 and it is really smoky outside).

Is there any explanation for this and can your forecast be made more reliable?

You are the only source I can find to do some forecasting on this, which is really valuable.

Sascha