Carbon Inverse Modeling

One method utilized to quantify sources of and sinks of carbon around the planet is the inverse approach. In short, this approach finds those sources and sinks, when transported around the planet by a transport model, generate predicted CO2 concentrations that best match those measured at various locations around the globe.

The elements I use for atmospheric carbon inverse estimation are atmospheric tracer transport models, optimization routines and CO2 observations. In the past, I have availed of the Colorado State University general circulation model, BUGS, for tracer transport. Recently, I have migrated to using an off-line tracer transport model called PCTM from Randy Kawa at NASA Goddard. These transport models are used to transport carbon dioxide around the planet given prescribed emissions. Comparison of the predicted CO2 concentration with the measured data forms the core of the inverse approach to understanding the global carbon cycle. The optimization approach I use has been a Bayesian synthesis inverse approach but I am now migrating to a more formal data assimilation technique referred to as Maximum Liklihood Ensemble Filter (MLEF).

A considerable portion of my current effort in this research area involves coordination of an international global carbon cycle inversion intercomparison called “TransCom“. The aim of TransCom is to isolate those components of the inverse process that give rise to the current spread in 3D inverse modeling results.

Some results of this work can be found in the following papers:

Some press reports provide a summary of our early findings.

Some other recent work on the global carbon cycle: