|Yepez, E. - UNIVERSITY OF ARIZONA|
|Williams, D. - UNIVERSITY OF ARIZONA|
|Lin, G. - COLUMBIA UNIVERSITY|
Submitted to: Agricultural and Forest Meteorology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: April 29, 2003
Publication Date: September 15, 2003
Citation: Yepez, E.A., Williams, D.G., Scott, R.L., Lin, G. 2003. Partitioning overstory and understory evapotranspiration in a semiarid savanna woodland from the isotopic composition of water vapor. Agricultural and Forest Meteorology. 119:43-68. Interpretive Summary: This scientific paper reports on an experiment to determine the sources of evaporation in a mesquite-dominated woodland along the San Pedro River in southeastern Arizona. Evaporation can be derived from mesquite trees, understory plants, or from the soil surface. In order to determine the relative proportion of these sources to the total evaporation from the forest, this study used a novel technique that included measurement of the evaporated water's isotopic composition and its concentration. Analysis of this data, revealed that 2.5 mm per day was derived from the mesquite, 0.5 mm per day came from the understory grasses, and 0.5 mm per day came from the drying of the moist soil.
Technical Abstract: The relative contributions of overstory and understory plant transpiration and soil evaporation to total evapotranspiration (ET) in a semiarid savanna woodland were determined from stable isotope measurements of atmospheric water vapor. The savanna overstory was dominated by the deeply rooted, woody legume Prosopis velutina ("mesquite"), and the understory was dominated by a perennial C4 grass, Sporobolus wrightii. "Keeling plots" (turbulent mixing relationships) were generated from isotope ratios (dD and d18O) of atmospheric water vapor collected within the tree (3-14 m) and understory (0.1-1 m) canopies during peak (July) and post monsoon (September) periods of 2001. The unique regression intercepts from upper and lower profiles were used to partition the ET flux from the understory layer separately from that of the whole ecosystem. Although ET partitioning was problematic during the first sampling period in July, our results in September provided support to the validity of this method for measuring and understanding the dynamic behavior of water balance components in this semiarid savanna. During the post-monsoon period (September 22nd), transpiration accounted for 85% of ecosystem ET. Transpiration by the grass layer accounted for 50% of the understory ET over the same period. Ecosystem ET estimated by eddy covariance (EC) on September 22nd was 3.5 mm d-1. Based on partitioning by the isotope method, 2.5 mm d-1 (70 %) was from tree transpiration and 0.5 mm d-1 (15%) was from transpiration by the grass layer. Independent estimations of overstory and understory ET partitioning from distributed understory EC measurements were remarkably consistent with our isotope approach.