Computing Influence Function Values for 28-Years of Western North America Tropospheric Ozone Observations Using FLEXPART-ERA5

Recent observations spanning the past few decades have indicated a decline in the occurrence of the lowest tropospheric ozone levels (< 20 ppb) across the northern hemisphere. This trend has raised concerns regarding the increased transport of ozone to surface levels in the western United States. In this study, we analyze the geographic origins of vertically-resolved ozone observations from a range of in-situ measurements conducted from 1994 to 2021. These comprehensive measurements include Alpha Jet Atmospheric eXperiment (AJAX) and In-service Aircraft for a Global Observing System (IAGOS) aircraft samples, ground-based measurements, and ozonesondes. The collected data from 900 hPa to 300 hPa pressure levels have been organized into 0.2 x 0.2 degree, 100 hPa boxes covering the expanse of the eastern North Pacific Ocean and western North America. To further our analysis, we have employed a Lagrangian particle dispersion model (FLEXPART v10.4-ERA5) to calculate the influence functions affiliated with these observations on a global scale. Considering the computational cost, we output the resulting influence functions aggregated into a spatial grid of 1 x 1 degree, encompassing five vertical layers (0-0.3 km, 0.3-3 km, 3-8 km, 8-13 km, and 13-20 km) and hourly time intervals. We convolve the influence functions with various emission sectors to understand the different source sectors, such as anthropogenic, biomass burning, lightning, aircraft, and shipping associated with each ozone observation. The estimation of influence functions is key to unraveling the transport pathways and emissions sources behind the declining trend in the cleanest air (e.g ozone < 5th percentile relative to baseline values).