It would be appreciated if the smoke forecasts from the NOAA RAP-Smoke (13.5km grids) (https://rapidrefresh.noaa.gov/RAPsmoke/) and/or HRRR-Smoke (3km grids) (https://rapidrefresh.noaa.gov/hrrr/HRRRsmoke/) were integrated into Windy.
The most useful of both of those two forecasts are the near-surface smoke and the vertically integrated smoke.
Those models do a substantially better job predicting fire related smoke than existing Windy layers for air quality (CO, PM2.5).
@idefix37 The HRRR model is already integrated into various forecast layers for the US despite the limited geographic area that it covers. If there were a "smoke" layer that allowed users to choose relevant (or even just all) accessible models it would be fairly straightforward.
Models with the most worldwide coverage could be the default for the layer and then local layers could be selected and set as user-defaults for those in particular areas.
The current CO and PM2.5 layers do a woeful job of displaying current and forecast wildfire smoke, which is different in behavior, detection, and forecasting.
@idefix37 I'm not familiar enough with all of the various weather models and their different layers and predictions. I know that Windy has a worldwide PM2.5 layer from ECMWF that doesn't allow any forecast selection (but presumably could, to specify smoke plumes?). There are a variety of English language regional and national smoke forecast models I do know about: the two in the title from NOAA, Canada has one called Blue Sky [that also covers the US], Australia has a smoke module in its AQFx model.
Given the current layout of Windy and the presence / availability of models that cover limited time and geographic areas I'm confused by your implication that something has to cover everywhere for consideration for inclusion.
@idefix37 The HRRR model is already integrated into various forecast layers for the US despite the limited geographic area that it covers. If there were a "smoke" layer that allowed users to choose relevant (or even just all) accessible models it would be fairly straightforward.
Models with the most worldwide coverage could be the default for the layer and then local layers could be selected and set as user-defaults for those in particular areas.
The current CO and PM2.5 layers do a woeful job of displaying current and forecast wildfire smoke, which is different in behavior, detection, and forecasting.
It would be appreciated if the smoke forecasts from the NOAA RAP-Smoke (13.5km grids) (https://rapidrefresh.noaa.gov/RAPsmoke/) and/or HRRR-Smoke (3km grids) (https://rapidrefresh.noaa.gov/hrrr/HRRRsmoke/) were integrated into Windy.
The most useful of both of those two forecasts are the near-surface smoke and the vertically integrated smoke.
Those models do a substantially better job predicting fire related smoke than existing Windy layers for air quality (CO, PM2.5).