Real-time North American Ensemble Forecast of Hazardous Air Quality Events
Abstract
Wildfires and dust storms are two major natural emission sources of aerosols in the atmosphere, which degrades the air quality and adversely affects human health. Prediction of the air quality effects of wildfire and dust emissions is challenging due to uncertainties in fire/dust emissions, fire plume rise calculation, and other model inputs/processes. Ensemble forecasting is increasingly used to improve the predictability of wildfire and dust aerosols. This study aims to develop a near real-time North American ensemble forecast of hazardous air quality events. Both regional and global models are used to create the ensemble, including NASA GEOS, NRL NAAPS, NOAA GEFS, NOAA HRRR, and NOAA NACC-CMAQ model. Novel methodology was developed to build a better ensemble forecast compared to the ensemble mean. The random walk technique was used to decide the best consolidation of the ensemble over the whole domain and in different regions. The ridge regression method was used to determine the best weighting factors for each ensemble member (model). The evaluation of particulate matter less than 2.5 µm in diameter (PM2.5) forecasting performance of the new near real-time ensemble was conducted by analyzing statistical metrics using AirNOW ground observations. Moreover, we proposed a way to forecast the probability of air quality exceedances during wildfires. The proposed ensemble exceedance probability forecast can be further applied to early warnings of extreme air pollution episodes during large wildfire events.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2022
- Bibcode:
- 2022AGUFM.A54G..02L