Carbon and Water Cycling in a Texas Hill Country Woodland

Date

2014-12-11

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

Two tree species, Plateau live oak (Quercus fusiformis) and Ashe juniper (Juniperus ashei) survive and thrive in a dense woodland on thin soil overlying massive limestone formations in the Texas Hill Country with recurrent annual summer drought punctuated every few years by intense rain and flooding. Previous research has shown that these species exhibit nearly opposite drought survival strategies at the root, stem and leaf levels. A fundamental question developed as to how these two apparently co-dominant species partition the scarce water resource under varying annual precipitation patterns. Eddy covariance and dendrochronology techniques were used to investigate carbon and water cycling from 2004 to 2012 in this setting. Essential information on the forest canopy age and species composition was obtained from a line-transect survey coupled with the bootstrap statistical method. Interannual change in water storage masked the relationships between annual precipitation and both annual evapotranspiration and annual productivity. A pair of methods were developed to minimize this masking effect caused by the interannual change in water storage using sequential linear regressions of annual precipitation versus ET or GPP by optimizing the start date of the annual timeframe as well as making a lag adjustment to the data for best goodness of fit.

The oaks and junipers were found to be co-dominant in the woodland canopy by number, each composing approximately 50%. Juniper was clearly dominant in the understory at 76%, while oak was clearly dominant in terms of carbon flux (80%) and standing biomass (85%). Evapotranspiration accounted for 72% of the fate of annual precipitation and the oaks are presumed to be the greatest water users due to the link between carbon and water fluxes through stomatal conductance.

Using October 1st (calendar day 274) as the start date for mass balance determination minimized the effect of the change in storage of plant-available water for both evapotranspiration and carbon flux. The optimal lag adjustment for evapotranspiration was 95 days while that of carbon flux was 91 days. These methods increased the ability of annual precipitation to explain the water and carbon budgets to 97% (up from 59%) and 96% (up from 64%) respectively. In this ecosystem, this demonstrated that most of the remaining variation when using the calendar year is a function of storage capacity and an artifact of timing.

Description

Citation