A three‐dimensional global episodic tracer transport model: 1. Evaluation of its transport processes by radon 222 simulations

Abstract

A three‐dimensional, Eulerian, global episodic tracer transport model is developed in this study. This model is designed to represent those episodic processes that are important for the transport and removal of short‐lived tropospheric constituents better than most other global tracer models. It is also designed to be consistent with a regional model so that a nested modeling system in future studies can provide the much‐needed finer resolutions over industrial regions with concentrated anthropogenic emissions. This model subdivides the atmosphere into 18 levels and 128 by 64 horizontal grids and calculates tracer concentrations by numerically solving the mass conservation equation on the basis of high temporal resolution, local meteorological conditions in an event‐based manner. It represents the first attempt to utilize a mesoscale‐type description of cloud effects on tropospheric chemical species for individual clouds in a globe‐scale tracer transport model. The vertical transport by clouds is represented by a one‐dimensional mixing model with subgrid‐scale mass adjustment and entrainment effects. A modified Bott's [1989a, b] scheme for spherical geometry is used for the advection calculations, which are mass conservative, positive definite, efficient, and accurate. Three parameterization schemes are provided for the description of mixing process in the planetary boundary layer (PBL), one of which has a nonlocal submodule for the convective PBL. An evaluation of the model transport processes is conducted by simulating radon 222 using 2 months of the National Center for Atmospheric Research Community Climate Model version 2 meteorology. Model‐simulated surface concentrations fall within the observed ranges over both continents and oceans. The model average vertical profile over midlatitude northern hemisphere and the percentages of radon mass above 2 and 5.5 km are in reasonable agreement with available observations. Much better agreement with observations is achieved when direct, subgrid‐scale cloud transport is included than when it is not. Model results suggest that cloud vertical transport contributes substantially to the radon in the middle and upper troposphere at northern midlatitudes and in the upper troposphere at low latitudes. One implication of these results is that subgrid cloud convection may play a dominant role in bringing surface emissions of reactive chemical compounds into the upper troposphere.