Evapotranspiration from a Saltcedar- Dominated Desert Floodplain: a Scaling Approach

The purpose of this study was to investigate evapotranspiration (ET) from a variety of scales (leaf to landscape) in saltcedar-dominated floodplain vegetation along the lower Virgin River of southern Nevada. Leaf-level gas exchange indicated that saltcedar exhibits similar stomatal conductance as the sympatric phreatophytes arrowweed, mesquite, and willow. However, sap flow in saltcedar was higher per unit sapwood area than the other species, suggesting that it maintains higher leaf area per unit sapwood area. At the stand level, saltcedar ET was found to exceed potential ET early in the summer when soils were moist and the water table was near the surface, but by late summer, after floodplain soils had dried and the water table had dropped, saltcedar ET was well below potential rates. Summer irrigation did not result in increased conductance of saltcedar for at least four weeks, suggesting that saltcedar does not utilize summer rainfall under normal conditions in the arid Mojave Desert. The importance of riparian zones in the arid Southwest as wildlife habitat and as recreational resources has recently led to a considerable amount of research and management efforts that seek to determine the instream flow requirements of riparian vegetation (Stromberg and Patten 1990). There have been numerous ecological studies of the floodplain vegetation of many riverine systems in the West'-Much of the literature consists of little more than inventories of dominant plant and wildlife communities, although recent studies have linked ecological processes with various hydrogeological variables. For example, studies of the relationships between the population biology of riparian plants and stream discharge dynamics (Everitt 1968; Stromberg and others 1991) have helped to clarify the essential role of flooding in the recruitment of trees in riparian forest communities. Studies of diverted streams in the western U.S. have shown that curtailment of annual flooding may result in a long-term senescence of riparian vegetation (Smith and others 1991; Stromberg and Patten 1990). I~: Barrow, Jerry R.; McArthur, E. Durant; Sosebee, Ronald E.; Tausch, Robm J., comps. 1996. Proceedings: shrubland ecosystem dynamics in a changing environment; 1995 May 23-25; Las Cruces, NM. Gen. Tech. Rep. INT-GTR-33~. Ogden, UT: U.S. Department of Agriculture, Forest Service, Intermountam Research Station. Stanley D. Smith is Associate Professor, Department of Biological Sci~nces, University of Nevada, Las Vegas, NV 89154-4004. Anna Sala is ~slstant Professor, Division of Biological Sciences, University of Montana, Mls~oula, MT 59812. Dale A. Devitt is Associate Professor, Department of EnVIronmental and Resource Sciences, University of Nevada, Reno, NV 89512. James R. Cleverly is Graduate Student, Department of Biological Sciences, University of Nevada, Las Vegas, NV 89154-4004. 199 The riparian communities of the arid Southwest are dominated by saltcedar (Tamarix ramosissima), a naturalized thicket-forming shrub that is reputed to be capable of losing vast quantiti~s of water through evapotranspiration (ET). From a very localized distribution in the 1920's, saltcedar has spread across the floodplains of rivers in the West, and is continuing to expand its dominance of floodplains and of isolated springs and wetlands today. Most of the large river systems of the Southwest were historically dominated by gallery cottonwood-willow (Populus-Salix) forests, only a small percentage of which are left today. In a study of the distribution of saltcedar, Everitt (1980) concluded that the invasion of saltcedar along southwestern watercourses was related to seed becoming increasingly available as changing physical and hydrological factors associated with development (dam-building, levee construction, etc.) was changing the flow regimes that native forests were historically adapted to. At the same time, the lack of annual flooding was resulting in the senescence of the former cottonwood-willow dominants due to increased water and salinity stresses along aridland floodplains (Busch and Smith 1995). A number of studies have attempted to quantify "consumptive use" (Le., ET) from southwestern riparian zones. Researchers have tended to conduct these studies in floodplains that have been invaded by saltcedar due to the widespread belief that it is a "water spender" relative to native phreatophytes such as cottonwood, willow, and mesquite. Most of these studies have used lysimeters or vertical micrometeorological masts (e.g., Bowen ratio) to estimate ET from individual plants or whole stands. Estimates on the order of 150-210 cm (ca. 60-80 inches) annually, based largely on lysimeter studies in Arizona and New Mexico (van Hylckama 1970; Horton 1972) have been reported and used in management programs throughout the Southwest. Robinson (1965) used such data to estimate that over 1.2 x 10 ma of water are consumed in the Southwest by saltcedar each year. Unfortunately, some of the early water consumptive studies were conducted in open areas, rather than in characteristic dense vegetation, leading to potentially large overestimates of ET due to advective water loss (Anderson and Idso 1987). Potential extrapolation problems from lysimeter studies led to attempts to predict the ET of saltcedar communities from meteorological data and vegetation structural characteristics (Gay and Fritschen 1979). However, such estimates were often obtained without accurate knowledge of the effects of environmental factors on stomatal resistance of the plant surfaces in question. This file was created by scanning the printed publication. Errors identified by the software have been corrected; however, some errors may remain.