Final Report Summary - COSIRIS (Investigating the terrestrial carbon and water cycles with a multi-tracer approach)
The aim of COSIRIS was to isolate the simultaneous fluxes of photosynthesis and respiration of the terrestrial biosphere. Explicit knowledge of the component fluxes will enable us to test process based models of photosynthesis and respiration, and investigate the sensitivity of each flux to environmental conditions. Specifically, the goal of COSIRIS was to develop a new tracer, carbonyl sulfide (COS) as a tool to separately investigate the carbon fluxes, photosynthesis and respiration, and the water fluxes, transpiration and evaporation, of terrestrial ecosystems. COS flux data have been challenging to obtain because atmospheric COS variations are 600 times smaller than any other molecule measured with eddy covariance so far. We completed a series of field campaigns to measure COS with a sophisticated setup of eddy covariance combined with automated chambers and high-frequency laser spectrometers. We collected extensive COS datasets at a semi-arid Mediterranean oak woodland, a freshwater marsh, and the first COS data from a tropical rainforest, at La Selva Biological Station, Costa Rica. With our field data, we successfully demonstrated COS-based partitioning of net ecosystem carbon fluxes into the component fluxes due to photosynthesis and respiration. We developed a process-based description of leaf COS exchange, and included COS in a depth-resolved litter/soil diffusion-reaction model. The soil model was then combined with field data to obtain optimized soil COS flux parameters through data assimilation. We found that leaf litter plays a more important role in surface COS fluxes than previously thought. Soil fluxes can be sources or sinks, but tend to be small compared to vegetation COS fluxes during the growing season. We also utilized the coupling of COS and photosynthetic carbon isotope discrimination to derive a global vegetation sink of COS in the order of 900±100 Gg S per year.