Inferring Ocean and Terrestrial Sinks

Understanding the behavior of the natural carbon cycle requires knowing the fluxes of carbon into and out of land and ocean sinks. Because the terrestrial biosphere’s behavior is hard to observe directly, researchers have turned to both “top-down” and “bottom-up” computer models to infer the size of land-based sources and sinks. One “top-down” technique is atmospheric inversion modeling, which takes the distribution of CO2 in the atmosphere and back-calculates the location and size of sources and sinks that would produce that spatial pattern.

An aspect of inversion models that has provoked controversy is their relatively small predicted flux of carbon to the atmosphere from tropical deforestation. Previous inversion studies have indicated a deforestation flux of 1 billion tons of carbon per year, while forward and process models, along with observations, predict a source nearly twice as large.

A new study by Andy Jacobson and colleagues incorporates new data from the ocean’s interior that improves estimates of regional ocean fluxes. The new approach yields a larger estimate of the ocean sink than previous inversions, with a greater than 70% probability that the flux is greater than 1.5 GtC/yr (versus earlier studies that put the flux nearer 1 GtC/yr). To balance the carbon cycle, this larger ocean sink requires a bigger flux from tropical and southern land to the atmosphere, consistent with estimates of a larger tropical deforestation flux. The new result brings inversion results into agreement with findings from forward and process models, and also matches results from ocean models and analysis of atmospheric oxygen data.


Modelling Terrestrial Sinks

Steve Pacala and colleagues are using another inversion technique to gain more insight into the behavior of terrestrial sinks. Forest inventories record the integrated response of a forest to changes in climate-related parameters, but to model the future, the response of individual species has to be characterized. The group has attacked this problem by gathering new field observations and developing a statistical inversion technique to extract growth functions for hundreds of individual species from forest inventories. The new functions will be used in a model of the terrestrial carbon cycle to diagnose current sources and sinks when forced with today’s conditions, and will be able to predict how hundreds of tree species will respond to changes in precipitation and temperature on long timescales.

In addition to the forest inventory data, data on atmospheric composition, satellite observations, and eddy flux information will also be incorporated into the analysis. The ultimate goal is a terrestrial carbon observing system that will monitor both short and long-timescale changes in the carbon cycle, and provide predictions for the future.


Feasibility of determining Northern hemisphere Carbon sinks using atmospheric data and transport models

Another effort carried out by Manuel Gloor and colleagues analyzed the detection problem for the Eurasian carbon sources and sinks. This research is assessing whether northern hemisphere mid-latitude land carbon sources and sinks can be accurately estimated from atmospheric data, and if so, how high a data density would be needed to obtain estimates to a given precision. The researchers used highly resolving regional atmospheric models and land biosphere fluxes with realistic diurnal cycles to determine the effective signal/to noise ratio for a 20% increase of a land biosphere uptake across Eurasia during June 1998.

Figure 6. Effective signal-to-noise ratio for detection of a 20% increased land sink in June 1998 located in Eurasia. A signal to noise value below 1 means the sink cannot be detected at the one sigma significance level even with hourly afternoon sampling. The effective signal-to-noise ratio has been calculated with two high-resolution regional transport models, HANK and REMO.


Their findings show that over regions with large fossil fuel emissions, a 20% increased biosphere sink in July (approx. 1.5 billion tons of carbon per year) cannot be detected, even with hourly afternoon sampling. In contrast in the Eastern part of the continent where fossil fuel emissions are smaller, the change can be captured with very frequent sampling (between daily to a few days sampling). The results also indicate that signals above the mixed layer are so tiny that, with the currently achievable measurement precision and accuracy, they are likely too small to be helpful on their own. Gloor and colleagues work suggests that detecting even large changes in terrestrial sinks will require frequent high-density sampling in the mixed portion of the planetary boundary layer.