Submitted to: American Geophysical Union
Publication Type: Proceedings
Publication Acceptance Date: October 30, 2009
Publication Date: December 12, 2009
Citation: Filley, T.R., Stott, D.E., Boutton, T.W., Creamer, C., Olk, D.C. 2009. Linkages Between Land Cover, Enzymes, and Soil Organic Matter Chemistry Following Encroachment of Leguminous Woody Plant into Grasslands. Eos, Transactions of the American Geophysical Union 90(52): B21B-0338. Technical Abstract: In the Rio Grande Plains of southern Texas, subtropical thorn woodlands dominated by the N-fixing tree Prosopis glandulosa have largely replaced native grasslands over the last 150 years as a result of fire suppression and over grazing. This land cover change has resulted in the increase of belowground stocks of C, N, and P, changes to the amount and chemistry of soil-stabilized plant biopolymers, and the composition and activity of soil microbes. Given that extracellular enzymes produced by plants and microbes are the principle means by which complex compounds are degraded and the production of such enzymes is triggered or suppressed by changes in primary input and nutrient availability we sought to relate how these fundamental changes in this ecosystem are reflected in the activity of soil stabilized extracellular enzymes and soil organic matter (SOM) chemistry in this system. We focused upon a successional chronosequence from C4-dominant grassland to woody patches of up to 86 yrs since mesquite establishment and related the molecular composition and concentration of hydrolysable amino acids and sugars, as well as CuO extractable lignin and substituted fatty acid to the potential activities of five extracellular enzymes (arylamidase, acid phosphatase, b-glucosidase, b-glucosaminidase (NAGase, polyphenoloxidase (PPO)) and a general marker for hydrolytic activity, fluorescein diacetate (FDA). Each of these enzymes, with the exception of PPO, showed higher potential activity in soils from woody clusters than grasslands and have activities generally well correlated to carbon content. PPO, an enzyme often targeted as a proxy for microbial lignin decay, showed no statistical difference between grassland and forest sites and no significant relationship to soil C content. Yields of total amino acids and amino sugars all show increases in content with cluster age when normalized to soil mass, as did the enzyme activities targeted to their decomposition, but only weak trends with age when normalized to carbon content. Although, PPO was nearly invariant across the chronosequence lignin content rose dramatically, both normalized to soil mass and normalized to organic carbon content, indicating oxidase activity was not matched to input and lignin was selectively accruing in the below ground carbon pools. We propose that the dramatic increase in available N in the leguminous system suppresses soil oxidase activity allowing lignin to selectively accrue while microbial activity for the other plant biopolymers was nearly keeping pace with input-consistent with recent findings in other temperate deciduous systems. These results have important implications for the modeling of woodland-grassland conversion and the dynamics of biogeochemical cycles in this globally significant land cover change.