Inverse Modeling of CO2 Sources and Sinks and the Northern Hemisphere sink


Jay Gregg


April 5, 2006


            The goal of inverse modeling is to determine the spatial and temporal distribution of carbon sources and sinks.  This is accomplished by coupling observed carbon dioxide (CO2) concentration and flux data with a planetary transport model, generating predicted pattern of CO2 concentrations for the entire world.  Ideally, the transport model is chosen such that the predicted CO2 concentrations match those measured at various locations around the globe.

            As such, the inverse modeling approach has three basic components: 1) observed atmospheric concentrations of CO2 over terrestrial locations around the globe, 2) observed sea surface concentrations of CO2 and 3) a forward-looking general transport model.  Atmospheric concentrations are based on a global network of on-going flask measurements.  Oceanic observations are predominately transect samples of fractional CO2 concentration of the sea surface water and the atmosphere immediately above it (ΔpCO2).  The ocean concentrations are filled in using a proxy model based on temperature, solubility, turbulence and average wind speed.  The general circulation model (GCM) is formed from available observations and is based on the temporal fluxes of carbon between grid cells on the Earth.  Many various transport models exist, and determining which transport model is most appropriate is the focus of many current studies.  Comparing results across models also permits understanding of the regions and biomes of the Earth that have the greatest uncertainty in the global carbon cycle.

Sources of CO2 include fossil fuel based emissions, cement manufacture, and land use change.  Carbon sinks include photosynthetic productivity, plant response to elevated atmospheric CO2 concentrations, marine uptake.  Of these, the confidence in magnitude, temporal patterns and spatial distribution are currently low.  The only exception is fossil fuels, where data exists to determine the magnitude of emissions on a per country annual scale.  The greatest uncertainty is in the terrestrial uptake, particularly the northern hemisphere. Evidence for the so-called “missing carbon sink” stems from inverse modeling studies; given the known sources and sinks, inverse model results indicate that the predicted CO2 concentrations of the northern hemisphere, and particularly North America, should be much higher than actually observed.  This implies that there must be some terrestrial sink that is not accounted for in the observed flux data.

            Other results from inverse modeling allow understanding of the interactions between various Earth systems.  For instance, there is evidence that the carbon circulation is greatly influenced by the El Nino and La Nina cycles.  Also, the oceanic thermohaline circulation has a large influence on the oceanic sink of carbon.  Volcanic events, such as the 1992 eruption of Mount Pinatubo, affect the radiation the Earth’s surface receives, contributing to a change in photosynthetic net primary productivity.  Temperature and humidity also influence soil respiration, increasing respiration at higher temperatures and humidity.  Finally, inverse modeling allows us to begin to predict the responses in various systems to increased levels of atmospheric CO2 that we are expect to see in the future.