Global modeling of tropospheric chemistry with assimilated meteorology: Model description and evaluation

TLDR

We present a first description and evaluation of GEOS‑CHEM, a global 3‑D tropospheric chemistry model driven by assimilated meteorological observations from GEOS. The model was applied to a 1‑year simulation of ozone‑NOx‑hydrocarbon chemistry for 1994 and evaluated against observations for 1994 and other years. The model reproduces ozone concentrations within ~10 ppb, captures seasonal patterns but underestimates amplitude at northern midlatitudes, reproduces NO and PAN within a factor of 2, overestimates HNO₃ by 2–3×, yields a methylchloroform lifetime of 5.1 yr versus 5.5 yr observed, and shows higher OH, lower CO, and underestimates acetone over the South Pacific.

Abstract

We present a first description and evaluation of GEOS‐CHEM, a global threedimensional (3‐D) model of tropospheric chemistry driven by assimilated meteorological observations from the Goddard Earth Observing System (GEOS) of the NASA Data Assimilation Office (DAO). The model is applied to a 1‐year simulation of tropospheric ozone‐NO x ‐hydrocarbon chemistry for 1994, and is evaluated with observations both for 1994 and for other years. It reproduces usually to within 10 ppb the concentrations of ozone observed from the worldwide ozonesonde data network. It simulates correctly the seasonal phases and amplitudes of ozone concentrations for different regions and altitudes, but tends to underestimate the seasonal amplitude at northern midlatitudes. Observed concentrations of NO and peroxyacetylnitrate (PAN) observed in aircraft campaigns are generally reproduced to within a factor of 2 and often much better. Concentrations of HNO 3 in the remote troposphere are overestimated typically by a factor of 2–3, a common problem in global models that may reflect a combination of insufficient precipitation scavenging and gas‐aerosol partitioning not resolved by the model. The model yields an atmospheric lifetime of methylchloroform (proxy for global OH) of 5.1 years, as compared to a best estimate from observations of 5.5 +/− 0.8 years, and simulates H 2 O 2 concentrations observed from aircraft with significant regional disagreements but no global bias. The OH concentrations are ∼20% higher than in our previous global 3‐D model which included an UV‐absorbing aerosol. Concentrations of CO tend to be underestimated by the model, often by 10–30 ppb, which could reflect a combination of excessive OH (a 20% decrease in model OH could be accommodated by the methylchloroform constraint) and an underestimate of CO sources (particularly biogenic). The model underestimates observed acetone concentrations over the South Pacific in fall by a factor of 3; a missing source from the ocean may be implicated.