UCAR > Communications > Quarterly > Winter 1999 Search

Winter 1999

MOZART: An uncompromising global model

Orchestrating MOZART are, from left, Doug Kinnison, Claire Granier, Guy Brasseur, Louisa Emmons, Stacy Walters, and Larry Horowitz. (Photo by Carlye Calvin.)

A new version of NCAR's global chemical model is about to give atmospheric scientists a better look at the global budget and interactions of ozone and almost 50 other chemical species in the troposphere. From there, the model of ozone and related trace species (MOZART) is moving up all the way to 85 kilometers and expanding to encompass an additional 30 species.

"MOZART has been a major project in ACD for five years or more," says Atmospheric Chemistry Division director Guy Brasseur. "It is a theoretical tool that allows us to understand and quantify the global budget of chemicals and to look at changes in the chemical composition of the atmosphere in response to human activities."

In the troposphere, ozone is a greenhouse gas, harmful to people and animals. Before the 1970s, scientists thought that ozone was in the troposphere simply because it sank down from the stratosphere. Groundbreaking work by Paul Crutzen and William Chameides at that time showed that it is born out of reactions in the troposphere itself. Because some of the gas's precursors, including pollutants like the nitrogen oxides, are increasing, ozone concentrations are rising sharply in many parts of the world. However, scientists don't have all of the measurements they need to understand its reactions and transport, especially from the tropics and Southern Hemisphere. The answer is global modeling.

Atmospheric scientists in this country may not be as familiar with global chemical modeling as they are with global climate modeling; a large number of the world's 15 or so global chemical models are European. The United States is home to only five or six, with strong groups at NASA Goddard Space Flight Center and Goddard Institute for Space Studies, Harvard University, and Lawrence Livermore National Laboratory, besides NCAR.

MOZART is "among the most detailed" of the global models, says Brasseur. "We have made very, very few compromises. We don't make many assumptions or simplifications." Because it requires high resolution (2.8° x 2.8°, with 34 vertical layers), 20-minute time steps, and a large number of species, MOZART is costly to run--ten times as expensive as NCAR's climate system model.

Users have done many types of studies using MOZART. For example, ACD researchers have reproduced the preindustrial chemical composition of the atmosphere by working backward from recent data, and they also have looked forward to 2050 and 2100 using IPCC scenarios of energy consumption, population, etc. "We have seen that the place where real perturbation will happen is the tr0opics," says Brasseur. "Also, in the upper troposphere [about 10 km] we expect a relatively large impact from airplane emissions."

MOZART version 2 incorporates an improved treatment of chemical transport and convection, as well as updated pollution information. It can use wind, temperature, and water vapor from climate models or from observational data to "push the chemicals around in the atmosphere," says Larry Horowitz, an ACD scientist involved in the model development. The ability to use different data sets will make MOZART useful in field campaigns. It will be fed with analyzed winds from the upcoming Tropospheric Ozone Production about the Spring Equinox experiment to help the TOPSE scientists understand their observations. For example, says Brasseur, "If as they fly north they suddenly have a peak of carbon monoxide, is that because they flew near a source, or was there an event that created that? We can answer those questions."

The model will make use of the daily measurements of carbon monoxide from the MOPITT (measurement of pollutants in the troposphere) instrument on Terra, the NASA Earth Observing System satellite launched on 18 December. Combining the instrument's along-track CO measurements with wind and temperature observations, MOZART will create a global three-dimensional distribution of CO with much more vertical resolution than the satellite allows.

Doug Kinnison is leading the effort to prepare an extended version of MOZART that will reach to 85 km altitude. This version includes additional species, such as chlorine and bromine compounds, that are important in the stratosphere and an additional 18 layers. With 250 chemical reactions, the new MOZART "has much more detail in the stratosphere," says Kinnison. "For example, it includes the chemistry of polar stratospheric clouds. We're testing those modules now." This version will be the basis for the model that principal investigators Ray Roble, Byron Boville, and Rolando Garcia are currently creating (see p. 1). Kinnison explains, "What we're doing is testing out the chemistry in MOZART before handing it off to the climate side."

Brasseur is a strong advocate of this kind of collaborative venture and other coupling projects. "We plan to couple MOZART in a few years with a marine biogeochemical model, and we already are working somewhat with the terrestrial biogeochemical system. We have created a surface/atmosphere interface model called biosphere-atmosphere-chemistry (BACH), which uses the land-surface model of [NCAR scientist] Gordon Bonan to help capture emissions from the biosphere. So we're trying to get BACH and MOZART to work together.

"All of these models have a goal, and that's to understand the entire earth system," Brasseur concludes. "CGD is focusing on the physical climate system; we are focusing more on the biogeochemical part of the climate system. The ultimate goal is to build a model of the whole earth system, and MOZART is extremely important to that."

MOZART is available on the Web. Those interested in using it should call Stacy Walters in ACD at 303-497-1476 or Doug Kinnison at 303-497-1469 to obtain a password.

In this issue... Other issues of UCAR Quarterly

Edited by Carol Rasmussen, carolr@ucar.edu
Prepared for the Web by Jacque Marshall
Last revised: Tue Apr 4 15:11:41 MDT 2000