Using Chemical and Dynamical Measurements within the CESM to Delineate the Interplay between Global Wildfires and Dynamical Forcings

Wildfires emit aerosols and gasses which are detrimental to human health. Furthermore, they are difficult to quantify and predict since they exhibit intra-annual and inter-annual variability, with one of the significant controls being meteorology, mostly through precipitation. Meanwhile, the heat and aerosols emitted from the fires further influence the atmospheric optical and physical properties, leading to changes in meteorology. We hope that investigating the bi-directional interaction between global wildfires and dynamical forcings can better account for the co-costs or co-benefits of the joint variability between wildfires and the climate system.

To represent wildfires, we obtain the NO₂column measurement data from the OMI satellite product. Due to the short lifetime of NO₂(a few hours), this approach captures the source of the wildfires precisely. To represent meteorology, we obtain the indices of the most significant dynamical drivers of climate variability: El Niño, the Indian Ocean Dipole, and the North Atlantic Oscillation. All datasets are transformed into weekly duration, allowing us to capture both variability and continuity, an improvement over previous studies using monthly or daily data. First, we adapt a variance maximization filter to obtain the regions undergoing the most intense wildfires, and find an increase of 30% to 50% in the emissions spread over space and time. Second, we employ a t-test between the different datasets, gaining insights into first order wildfire-meteorology interactions, and find that El-Niño is not as significant over regions where others have reported it to be. We also find the IOD is significant in more places than reported. Third, we employ multi-linear regression using lagged and unlagged meteorological indices, which is the most physically reasonable, since it captures both rainfall as well as changes in the land surface moisture and water storage. We determine that the lagged variables in general have a stronger relationship, although the weighting varies from region to region. Finally, we use the observed patterns of NO₂and dynamics to drive CESM specifically with respect to CO and black carbon, in order to better understand the atmospheric impact and the magnitude of the various bi-directional mechanisms between wildfires and future climate.