Lightning NO_x in the 29–30 May 2012 Deep Convective Clouds and Chemistry (DC3) Severe Storm and Its Downwind Chemical Consequences

Abstract

A cloud‐resolved storm and chemistry simulation of a severe convective system in Oklahoma constrained by anvil aircraft observations of NOx was used to estimate the mean production of NOx per flash in this storm. An upward ice flux scheme was used to parameterize flash rates in the model. Model lightning was also constrained by observed lightning flash types and the altitude distribution of flash channel segments. The best estimate of mean NOx production by lightning in this storm was 80–110 mol per flash, which is smaller than many other literature estimates. This result is likely due to the storm having been a high flash rate event in which flash extents were relatively small. Over the evolution of this storm a moderate negative correlation was found between the total flash rate and flash extent and energy per flash. A longer‐term simulation at 36‐km horizontal resolution with parameterized convection was used to simulate the downwind transport and chemistry of the anvil outflow from the same storm. Convective transport of low‐ozone air from the boundary layer decreased ozone in the anvil outflow by up to 20–40 ppbv compared with the initial conditions, which contained stratospheric influence. Photochemical ozone production in the lightning‐NOx enhanced convective plume proceeded at a rate of 10–11 ppbv per day in the 9–11 km outflow layer over the 24‐hr period of downwind transport to the Southern Appalachians. Photochemical production plays a large role in the restoration of upper tropospheric ozone following deep convection.

Evaluation of Lightning Flash Rate Parameterizations in a Cloud‐Resolved WRF‐Chem Simulation of the 29–30 May 2012 Oklahoma Severe Supercell System Observed During DC3

“…The analysis showed that the LMA observations were well represented by two schemes. The ICEFLUX scheme (Finney et al, 2014) most closely replicated the observed flash rate trends during the first 4 hr of convection, which is important for the LNO x simulation reported in the companion paper (Pickering et al, 2024). ICEFLUX also provided the best estimate of the total flashes over the entire simulation.…”

“…The initial 8-hr simulation ran from 18:00 UTC on 29 May to 02:00 UTC on 30 May using 480 processors on the NCAR Yellowstone supercomputer. The simulation did not extend past 02:00 UTC because we were interested in this time period for atmospheric chemistry analyses when lightning would be influencing the NO x mixing ratios sampled by the DC3 aircraft in the anvil outflow (Pickering et al, 2024). For subsequent simulations a restart file was used at 21:00 UTC to reduce computation time since the observed convection began just after 21:00 UTC.…”

“…The total observed and simulated flash counts were similar over the period in which NO x was sampled by the aircraft (ending at 01:00 UTC observed time) and over the entire duration of the convection. Therefore, the ICEFLUX scheme was used in WRF-Chem simulations of LNO x for this storm that are reported in the companion paper (Pickering et al, 2024). However, the UV scheme best represented the lightning flash rate fluctuations of all the 18 schemes after 00:00 UTC model time, specifically the magnitude of the primary peak and subsequent decreasing trend.…”