Skip to main content
ARS Home » Plains Area » Temple, Texas » Grassland Soil and Water Research Laboratory » Research » Publications at this Location » Publication #232520
Title: Primary productivity and water balance of grassland vegetation on three soils in a continuous CO2 gradient: initial results from the lysimeter CO2 gradient experiment

Author
item Fay, Philip
item KELLEY, ALEXIA - DUKE UNIVERSITY
item PROCTER, ANDREW - DUKE UNIVERSITY
item Jin, Virginia
item JACKSON, ROBERT - DUKE UNIVERSITY
item Polley, Herbert

Submitted to: Ecosystems
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/11/2009
Publication Date: 5/21/2009
Citation: Fay, P.A., Kelley, A.M., Procter, A.C., Jin, V.L., Jackson, R.B., Polley, H.W. 2009. Primary productivity and water balance of grassland vegetation on three soils in a continuous CO2 gradient: Initial results from the lysimeter CO2 gradient experiment. Ecosystems. 12:699-714.

Interpretive Summary: In order to understand the likely future impacts of rising atmospheric CO2 concentrations on grassland productivity, diversity, and ability to support livestock production, it is crucial to understand how the rate of grassland ecosystem responses to past increases in CO2 compare to future increases in CO2, and to understand how grassland responses to CO2 may vary spatially due to differences in soil type. We constructed the Lysimeter CO2 Gradient (LYCOG) facility to apply a continuous gradient of subambient to enriched air CO2 concentrations (250 to 500 ppm) on grassland plots of controlled species composition on clay, silty clay, and sandy loam soils, which differ in hydrologic properties and soil carbon (C) and nitrogen (N) pools. The facility is an advancement over previous CO2 gradient facilities because it includes weighing lysimeters and systems to collect and measure the volume of soil water drainage, allowing closure of the soil moisture budget. The LYCOG facility was successful at creating and controlling the CO2 gradient with no major variation in other factors such as air and soil temperature that may confound CO2 effects and complicate their interpretation, nor with major confounding effects from the tunnel infrastructure on leaf photosynthesis and soil respiration, two key fluxes of C in grasslands.

Technical Abstract: Field studies of atmospheric CO2 effects on ecosystem processes usually include only a few levels of CO2 and a single soil type, making it difficult to ascertain the shape of ecosystem responses to increasing CO2 or to generalize CO2 effects across ecosystems on varying soil types. The Lysimeter CO2 Gradient (LYCOG) facility consists of two 50 m linear chambers that impose a continuous gradient of atmospheric CO2 concentrations (~ 250 to 500 µL L-1) on grassland communities established on 1.5 m3 intact soil monoliths from three soil series varying in texture, water holding capacity, and C/N pools. The chambers maintained nearly linear daytime and nighttime CO2 gradients. There were daytime temperature increases of 3 to 8 °C within the 5 m sections between cooling coils, but little overall temperature increase along the full chamber length. The within-section gradient caused effects on aboveground biomass and evapotranspiration that were small compared to the effect of varying CO2 concentration. The chamber infrastructure caused little appreciable effect on leaf level carbon uptake in a dominant grass and forb species, or on soil CO2 efflux. Four grass (C4), two forb, and one legume were planted in equal abundances on the three soils (Austin silty clay, Bastrop sandy loam, and Houston clay) in 2002. After the final pretreatment growing season (2005), the established plant communities differed in productivity and species composition, with Austin soils > 40% less productive aboveground and belowground compared to Bastrop and Houston soils. The established plant communities differed in composition and productivity in ways similar to that seen in natural grassland communities and remained directly comparable. These differences likely reflect varying soil N and water limitation of plant growth, and suggest that N limitation may increase with CO2 enrichment, especially on Austin soils. We conclude that LYCOG maintained a satisfactory gradient of CO2 without major changes in factors that could confound the effects of CO2 on grassland ecosystem processes.