Evaluating PEM-TROPICS observations: Uncertainty analyses and effects of deep convective mixing on tropospheric chemistry

Rao Kotamarthi and Donald J. Wuebbles

Significant uncertainties currently exist in the upper tropospheric and lower stratospheric NOx and HOx chemistry and budgets. Recent measurements indicate large unexplained sources of HOx in the tropical upper troposphere and of NOx in the mid-latitude northern hemispheric upper troposphere. In-situ measurements of a suite of source gases and their products from aircraft based measurement platforms are expected to provide answers to these questions. However, analysis of data collected from aircraft platforms in the troposphere represent a significant challenge for atmospheric chemistry modelers. Box photochemical models have been widely employed to analyze data collected in the troposphere by the DC-8 and other aircraft platforms with varying degrees of success. Several uncertainties including incomplete mixing and entrainment of ambient air for parcels undergoing convection, errors in measured data and constraints imposed on the box models and their effect on the model calculated results and problems defining the initial conditions for parcel trajectory calculations have not been given sufficient attention. In this proposal we address issues related to improving the reliability of box model and trajectory model calculations in the troposphere in support of the needs for understanding the PEM-Tropics measurements. We will a) build capability of box models to reflect the uncertainty in the data and the sensitivity of model results to the large number of parameters used in the model and b) improve box models to reflect the mixing and entrainment of ambient air experienced by air parcels undergoing convection.

Objectives of this study

• investigate the sensitivity of box model results for radicals, derived quantities such as production and loss of ozone due to errors in initial conditions and reaction rates.
• investigate the effect of mixing and cloud processes, including entrainment of ambient air in to the convective plume using a newly developed approach known as linear eddy mixing.
• perform extensive sets of air parcel calculations for observed and synthetic convective warm cloud forming events to include the uncertainties in the air parcel initial conditions, entrainment and its destination inside the cloud.
• use the sensitivity analysis tools and an adjoint model for the air parcel trajectory model to improve the reliability of specification of initial conditions for photochemical air parcel trajectories.
• provide data and modeling analysis support to the science team as required.