A cloud-scale model study of lightning-generated NO_x in an individual thunderstorm during STERAO-A

Understanding lightning NO_x (NO + NO₂) production on the cloud scale is key for developing better parameterizations of lightning NO_x for use in regional and global chemical transport models. This paper attempts to further the understanding of lightning NO_x production on the cloud scale using a cloud model simulation of an observed thunderstorm. Objectives are (1) to infer from the model simulations and in situ measurements the relative production rates of NO_x by cloud-to-ground (CG) and intracloud (IC) lightning for the storm; (2) to assess the relative contributions in the storm anvil of convective transport of NO_x from the boundary layer and NO_x production by lightning; and (3) to simulate the effects of the lightning-generated NO_x on subsequent photochemical ozone production. We use a two-dimensional cloud model that includes a parameterized source of lightning-generated NO_x to study the production and advection of NO_x associated with a developing northeast Colorado thunderstorm observed on July 12, 1996, during the Stratosphere-Troposphere Experiment - Radiation, Aerosols, Ozone (STERAO-A) field campaign. Model results are compared with the sum of NO measurements taken by aircraft and photostationary state estimates of NO₂ in and around the anvil of the thunderstorm. The results show that IC lightning was the dominant source of NO_x in this thunderstorm. We estimate from our simulations that the NO_x production per CG flash (P_CG) was of the order of 200 to 500 mol flash^-1. NO_x production per IC flash (P_IC) appeared to be half or more of that for a CG flash, a higher ratio of P_IC/P_CG than is commonly assumed. The results also indicate that the majority of NO_x (greater than 80%) in the anvil region of this storm resulted from lightning as opposed to transport from the boundary layer. The effect of the lightning NO_x on subsequent photochemical ozone production was assessed using a column chemical model initialized with values of NO_x, O₃, and hydrocarbons taken from a horizontally averaged vertical profile through the anvil of the simulated storm. The lightning NO_x increased simulated ozone production rates by a maximum of over 7 ppbv d^-1 in the upper troposphere downwind of this storm.