Water table decline alters growth and survival of Salix gooddingii and Tamarix chinensis seedlings

Abstract

In the western United States, many native riparian forests are declining downstream from dams and diversions, primarily due to a lack of successful regeneration of native species. Conceptual models have been described that incorporate stream flow characteristics that promote successful regeneration of western riparian Populus seedlings. However, these models need to be calibrated for additional species and specific stream characteristics. In this study, we examined the growth and survival of two dominant Sonoran desert riparian tree species, a native Salix gooddingii, and an exotic, Tamarix chinensis, to simulated water table declines at rates of 0, 1, 2, and 4 cm/day. Salix seedlings performed best in the 0 cm/day treatment in which the soil was saturated and showed decreasing survival and growth with increasing rates of water table decline. Root-to-shoot ratios were high for Salix, but root elongation rates were low with an emphasis on lateral root development. This root growth pattern may be an adaptation to habitat where Salix typically grows, near the stream channel where the water table is shallow but the likelihood of flood scour is high. Lateral root development may help to protect Salix seedlings from late season flood scour. Tamarix seedlings showed consistently high survival across all treatments and had the greatest growth in the 0 and 1 cm/day treatments. Tamarix seedlings had greater root elongation rates in response to water table decline relative to the control, with the greatest elongation in the 1 cm/day treatment. Greater root elongation rates and higher drought tolerance allowed Tamarix seedlings to persist in dry soils where Salix seedlings died. Differences in growth and survival of these two species in response to water table decline will need to be incorporated into management practices to promote successful native riparian tree regeneration.

Introduction

Riparian ecosystems in the American Southwest compose less than 3% of the landscape area (Naiman and Decamps, 1997), however, they are highly valued for wildlife habitat, biodiversity, recreation, flood attenuation, and water quality (Carothers, 1977, Patten, 1998). Despite their importance, demands for water and subsequent development projects, including dams and diversions, have contributed to a decline of southwestern riparian forests (Stromberg, 1993, Busch and Smith, 1995, Stromberg et al., 1996, Patten, 1998). Dam operations alter the hydrology of rivers, often reducing sediment loads and increasing erosive power causing greater channel incision and water table decline (Reily and Johnson, 1982). The abrupt downstream decline of alluvial water tables caused by dams and diversions has been proposed as a mechanism of riparian forest decline (Rood and Heinze-Milne, 1989, Rood and Mahoney, 1990, Mahoney and Rood, 1991, Mahoney and Rood, 1992). Also, flood attenuation and altered timing of peak flows by dams and diversions can limit regeneration by downstream native trees (Fenner et al., 1985, Bradley and Smith, 1986, Rood and Heinze-Milne, 1989, Rood and Mahoney, 1990, Mahoney and Rood, 1998) and promote the establishment of exotics that are more competitive than natives under drier conditions (Smith et al., 1991, Busch and Smith, 1995, Cleverly et al., 1997, Smith et al., 1998). Chronic regeneration failures by native riparian trees can shift community composition towards dominance by exotic and more xeric species (Friedman et al., 1995).

The germination requirements of many western riparian trees are well known (Stromberg, 1993, Braatne et al., 1996, Mahoney and Rood, 1998). High spring flows typically scour competing vegetation and deposit sediment, creating safe sites for germination. Most native riparian trees are adapted to this disturbance regime and produce seeds prolifically in synchrony with flood water recession in early spring (Stromberg, 1993). The presence of significant numbers of seedlings every year suggest that seed availability is not limiting to establishment, although many of these seedlings do not survive the first year (Stromberg, 1993, Scott et al., 1997, Mahoney and Rood, 1998). The primary causes of mortality are typically water stress caused by rapid water table declines (Mahoney and Rood, 1991, Stromberg, 1993, Mahoney and Rood, 1998) and scouring by late summer and fall floods (Stromberg, 1993). A key to seedling establishment is rapid root growth as the water table declines (Mahoney and Rood, 1991, Mahoney and Rood, 1992). Often the rate of water table decline on dam-regulated rivers is very rapid, which can cause extensive water stress and mortality in new seedling populations (Mahoney and Rood, 1998). Populus seedlings are known to have root growth rates of 0.6–1.3 cm/day, resulting in 72–162 cm of root growth by the end of their first season (Fenner et al., 1984, Mahoney and Rood, 1998), however, it has been reported that Populus seedlings can survive ground water decline rates of 2–4 cm/day (Mahoney and Rood, 1991, Mahoney and Rood, 1992, Segelquist et al., 1993). These findings suggest that some riparian tree seedlings can use water from the capillary fringe above the alluvial water table (Mahoney and Rood, 1998). The thickness of the capillary fringe can range from less than 10 cm in coarse cobbles to greater than 100 cm in fine silts (Mahoney and Rood, 1998).

The ’recruitment box’ conceptual model of Populus seedlings based on these ideas combines root growth rates with rates of water table and capillary fringe declines to estimate successful establishment sites (Mahoney and Rood, 1998). Utilizing a 2.5 cm/day rate of water table decline combined with Populus root growth rates and the thickness of the capillary fringe, Mahoney and Rood (1998) determined that for many Populus species, successful establishment would occur at elevations 0.6–2 m above the late summer low flow stream stage. The model was used successfully on the Truckee River in Nevada to promote Populus regeneration by regulating rates of water table decline in a controlled flood. This model should have applications in other semi-arid riparian areas around the world, and may be an important tool in conservation and restoration of riparian habitats. However, the model will need calibration for different species and specific stream conditions (Mahoney and Rood, 1998, Shafroth et al., 1998). The estimated survivable rate of ground water decline of 2.5 cm/day used by Mahoney and Rood (1998) may not be applicable in southwestern riparian ecosystems, which are typically much hotter and drier than the systems reviewed in their study. In addition, Salix gooddingii Ball has been shown to be more sensitive to changes in ground water availability than the co-occurring Populus fremontii S. Wats. (Stromberg, 1993) and grows well in saturated soils (Stromberg, 1997). In contrast, Tamarix chinensis Lour. has been reported to be more drought tolerant than native species (Cleverly et al., 1997, Smith et al., 1998) and is able to persist indefinitely in drier soils (Everitt, 1980).

In the Sonoran desert, low elevation (<1250 m) riparian forests were traditionally dominated by P. fremontii and S. gooddingii (Stromberg, 1993). However, in response to water development, many of these native populations have been invaded and often replaced by the exotic, T. chinensis, (Christensen, 1962, Stromberg, 1993, Brock, 1994, Busch and Smith, 1995). Whereas field observational studies have provided insight into the factors that regulate establishment of these southwestern riparian species (Stromberg, 1993, Stromberg, 1997), more information is needed on the effects of different rates of water table decline on seedling establishment and growth. In this study, we utilized rhizopods (Mahoney and Rood, 1991, Mahoney and Rood, 1992) to investigate the effects of differing rates of ground water decline on growth and survival of one important, native riparian tree species, S. gooddingii, and one invasive exotic, T. chinensis.