ARS | Publication request: The Potential of Carbonyl Sulfide as a Tracer for Gross Primary Productivity at Flux Tower Sites

Submitted to: American Geophysical Union
Publication Type: Abstract Only
Publication Acceptance Date: September 30, 2010
Publication Date: December 13, 2010
Citation: Blonquist, J., Montzka, S., Yakir, D., Dragoni, D., Griffis, T., Monson, R., Munger, J., Scott, R.L., Bowling, D. 2010. The Potential of Carbonyl Sulfide as a Tracer for Gross Primary Productivity at Flux Tower Sites. American Geophysical Union 2010 Fall Meeting, [Abtstract]. AGU, San Francisco, Calif., 13-17 Dec.

Interpretive Summary: Regional/continental scale studies of atmospheric carbonyl sulfide (OCS) seasonal dynamics and leaf level studies of plant OCS uptake have shown a close relationship to CO2 dynamics and uptake, suggesting potential for OCS as a tracer for gross primary productivity (GPP). Canopy CO2 and OCS differences (mole fraction within canopy minus that above canopy) at a temperate deciduous forest (Harvard Forest AmeriFlux site) were analyzed relative to net ecosystem exchange (NEE) and GPP, respectively. Canopy CO2 and OCS vertical gradients (CO2 and OCS differences divided by within and above canopy measurement height differences) were used to calculate ecosystem relative uptake (ERU; relative canopy OCS gradient divided by relative canopy CO2 gradient, where relative gradients are gradients normalized by above canopy mole fractions), from which GPP was estimated using an equation that assumes OCS follows the same physical pathway as CO2 into plant leaves and where GPP / NEE was proportional to OCS gradient / CO2 gradient. Additionally, canopy CO2 differences from five other AmeriFlux sites were analyzed, and OCS differences were projected from these differences (via an assumed ERU) to further evaluate OCS as a potential GPP tracer. At Harvard Forest, canopy CO2 differences were related to NEE (y = 0.041x + 0.046, r2 = 0.14, P < 0.025) and OCS differences were related to GPP (y = 0.43x - 2.0, r2 = 0.18, P < 0.1), indicating the influence of canopy uptake on canopy differences. Relative canopy OCS and CO2 gradients were linearly correlated (slope = 4.4, intercept = -0.00028, r2 = 0.69, P < 0.025), indicating CO2 and OCS dynamics were likely controlled by similar mechanisms. Estimates of GPP derived from OCS and from temperature-based NEE partitioning showed a strong linear relationship (slope = 1.2, intercept = 3.1, r2 = 0.99, P < 0.0005), indicating the potential of OCS as a GPP tracer. As with Harvard Forest, canopy CO2 differences at the other AmeriFlux sites were related to NEE, and projected canopy OCS differences were related to GPP. At forest sites, projected OCS differences were similar in magnitude to those at Harvard Forest and were near precision limits of current OCS measurement capabilities (approximately 5-10 pmol mol-1). Projected canopy OCS differences in short, dense canopies (C4 grassland and soybean crop) were much greater than differences in forests, indicating ERU is potentially measurable with current grab-sample-based OCS measurement capabilities and may provide an alternative means of estimating GPP at flux tower sites.

Technical Abstract: Regional/continental scale studies of atmospheric carbonyl sulfide (OCS) seasonal dynamics and leaf level studies of plant OCS uptake have shown a close relationship to CO2 dynamics and uptake, suggesting potential for OCS as a tracer for gross primary productivity (GPP). Canopy CO2 and OCS differences (mole fraction within canopy minus that above canopy) at a temperate deciduous forest (Harvard Forest AmeriFlux site) were analyzed relative to net ecosystem exchange (NEE) and GPP, respectively. Canopy CO2 and OCS vertical gradients (CO2 and OCS differences divided by within and above canopy measurement height differences) were used to calculate ecosystem relative uptake (ERU; relative canopy OCS gradient divided by relative canopy CO2 gradient, where relative gradients are gradients normalized by above canopy mole fractions), from which GPP was estimated using an equation that assumes OCS follows the same physical pathway as CO2 into plant leaves and where GPP / NEE was proportional to OCS gradient / CO2 gradient. Additionally, canopy CO2 differences from five other AmeriFlux sites were analyzed, and OCS differences were projected from these differences (via an assumed ERU) to further evaluate OCS as a potential GPP tracer. At Harvard Forest, canopy CO2 differences were related to NEE (y = 0.041x + 0.046, r2 = 0.14, P < 0.025) and OCS differences were related to GPP (y = 0.43x - 2.0, r2 = 0.18, P < 0.1), indicating the influence of canopy uptake on canopy differences. Relative canopy OCS and CO2 gradients were linearly correlated (slope = 4.4, intercept = -0.00028, r2 = 0.69, P < 0.025), indicating CO2 and OCS dynamics were likely controlled by similar mechanisms. Estimates of GPP derived from OCS and from temperature-based NEE partitioning showed a strong linear relationship (slope = 1.2, intercept = 3.1, r2 = 0.99, P < 0.0005), indicating the potential of OCS as a GPP tracer. As with Harvard Forest, canopy CO2 differences at the other AmeriFlux sites were related to NEE, and projected canopy OCS differences were related to GPP. At forest sites, projected OCS differences were similar in magnitude to those at Harvard Forest and were near precision limits of current OCS measurement capabilities (approximately 5-10 pmol mol-1). Projected canopy OCS differences in short, dense canopies (C4 grassland and soybean crop) were much greater than differences in forests, indicating ERU is potentially measurable with current grab-sample-based OCS measurement capabilities and may provide an alternative means of estimating GPP at flux tower sites.