Carbon isotope dynamics of free-air CO2-enriched cotton and soils

Abstract

A role for soils as global carbon sink or source under increasing atmospheric CO2 concentrations has been speculative. Free-air carbon dioxide enrichment (FACE) experiments with cotton, conducted from 1989 to 1991 at the Maricopa Agricultural Center in Arizona, maintained circular plots at 550 μmol mol−1 CO2 with tank CO2 while adjacent ambient control plots averaged about 370 μmol mol−1 CO2. This provided an exceptional test for entry of carbon into soils because the petrochemically derived tank CO2 used to enrich the air above the FACE plots was depleted in both radiocarbon (14C content was 0% modern carbon (pmC)) and 13C (δ13C≈ −36‰) relative to background air, thus serving as a potent isotopic tracer. Flask air samples, and plant and soil samples were collected in conjunction with the 1991 experiment. Most of the isotopic analyses on the plants were performed on the holocellulose component. Soil organic carbon was obtained by first removing carbonate with HCl, floating off plant fragments with a NaCl solution, and picking out remaining plant fragments under magnification. The δ13C of the air above the FACE plots was approximately −15 to −19‰, i.e. much more 13C depleted than the background air of approximately −7.5‰. The δ13C values of plants and soils in the FACE plots were 10–12‰ and 2‰13C-depleted, respectively, compared with their control counterparts. The 14C content of the FACE cotton plants was approximately 40 pmC lower than tha tof the control cotton, but the 14C results from soils were conflicting and therefore not as revealing as the δ13C of soils. Soil stable-carbon isotope patterns were consistent, and mass balance calculations indicate that about 10% of the present organic carbon content in the FACE soil derived from the 3 year FACE experiment. At a minimum, this is an important quantitative measure of carbon turnover, but the presence of 13C-depleted carbon, even in the recalcitrant 6 N HCl resistant soil organic fraction (average age 2200 years before present (BP)), suggests that at least some portion of this 10% is an actual increase in carbon accumulation. Similar isotopic studies on FACE experiments in different ecosystems could permit more definitive assessment of carbon turnover rates and perhaps provide insight into the extent to which soil organic matter can accommodate the ‘missing’ carbon in the global carbon cycle.