Modeling the flow of carbon in the atmosphere

Science Daily: The global travel logs of greenhouse gases are based on atmospheric sampling locations sprinkled over the Earth and short towers that measure the uptake or release of carbon from a small patch of forest. But those measurements don't agree with current computer models of how plants and soils behave.

A University of Michigan researcher is developing a unique way to reconcile these crucial data. "If we're going to adapt to climate change, we need to be able to predict what the climate will be," said Anna Michalak, assistant professor in the Department of Civil and Environmental Engineering and the Department of Atmospheric, Oceanic and Space Sciences. "We want to know how the sources and sinks of carbon will evolve in the future, and the only way we can manage climate change is with scientific information."

…The problem: Michalak said the data takes such a big-picture approach that it is difficult to isolate carbon being emitted or taken up in specific regions, or even countries. Scientists are left with an understanding of carbon sources that isn't nimble enough to understand the variability, or to be confident about predicting the future.

Michalak has developed a robust way to use available data to understand this variability called "geostatistical inverse modeling." This method breaks the globe into small regions and examines how much CO2 must have been emitted in each region to achieve the concentrations measured at atmospheric sample points. This method also allows her and her collaborators to use information from other existing satellites that measure the Earth's surface to supplement the information from the atmospheric monitoring network. Eventually, this method aims to trace the carbon levels at each sample point to a particular source or sink on the surface….

A diagram of the carbon cycle by Eraxx, modifying a diagram by FischX