Spatial Variation in Abundance Parameters of a Federally Threatened Groundwater Salamander Within and Among Central Texas Headwater Creeks

Abstract

Semi-arid conditions in central Texas relegate Eurycea salamanders to aquatic habitats influenced by groundwater (i.e., springs, spring-fed creeks, and subterranean water in alluvium and aquifers). Many studies have noted that epigean (surface) populations of central Texas Eurycea occur near spring outlets. Consequently, the U.S. Fish and Wildlife Service designated surface critical habitat units for three species as a uniform distance up and downstream of occupied spring outlets. Here, we used data from visual encounter and quadrat surveys to model relative abundance and relative density, respectively, of federally threatened Jollyville Plateau Salamanders (E. tonkawae) in relation to downstream distance from a spring outlet in headwater creeks. We additionally use recapture data to investigate movement within these systems and in relation to the critical habitat units. Consistent with other studies, E. tonkawae relative abundance and relative density decreased with increasing distance from a spring outlet, and E. tonkawae occurred outside of its federally designated surface critical habitat unit at some sites. Importantly, the downstream extent of E. tonkawae and the rate of change in relative abundance and relative density varied among sites, which is incongruent with the uniform federal surface critical habitat distance. We observed limited movement within the headwater creeks, with most salamanders recaptured within 5 m of their previous capture location.

Keywords: conservation; distribution; groundwater; headwater creek; salamander; threatened species.

FIGURE 1

Jollyville Plateau Salamander ( Eurycea tonkawae ) in situ at Hill Marsh Spring in Williamson County, Texas, USA.

FIGURE 2

Jollyville Plateau Salamander ( Eurycea tonkawae ) survey locations and federally designated surface critical habitat units in Travis and Williamson counties, Texas, USA. Each circle represents the location of a surface critical habitat unit but not the actual size which is an 80 m radius around occupied spring outlets (U.S. Fish and Wildlife Service 2013b). The background is a hillshade layer which is a grayscale, three‐dimensional representation of the terrain surface. This layer has a pixel size (spatial resolution) of 1 m. (1) Avery Deer Spring, (2) Avery Springhouse Spring, (3) Hill Marsh Spring, (4) MacDonald Well, and (5) PC Spring.

FIGURE 3

Relative abundance of Jollyville Plateau Salamanders ( Eurycea tonkawae ) compared to distance from the nearest spring outlet in seven headwater creeks in Travis and Williamson counties, Texas, USA. Counts (lines) and 95% confidence intervals (dark shaded regions) are predictions from the top generalized linear mixed model with day‐of‐year held constant at the mean survey day (Day 178), sampling method held constant at “stratified VES”, and a log offset of effort held constant at 100 searched cover objects. Raw salamander counts (points) are not standardized by the number of searched cover objects, and one point at PC 2 exceeds the y‐axis limit. The vertical red line represents the limit of the federally designated critical habitat unit, i.e., 80 m from the spring outlet (U.S. Fish and Wildlife Service 2013b). The vertical dashed line denotes the furthest downstream detection of E. tonkawae in each headwater creek.

FIGURE 4

Relative density (counts per 30 cm × 30 cm quadrat) of Jollyville Plateau Salamanders ( Eurycea tonkawae ) compared to distance from the nearest spring outlet in seven headwater creeks in Travis and Williamson counties, Texas, USA. Predicted counts (lines) and 95% confidence intervals (dark shaded regions) are from the top generalized linear mixed model with day‐of‐year held constant at the mean survey day (Day 180) and sampling method held constant at “stratified”. Points are raw salamander counts. The vertical red line represents the limit of the federally designated critical habitat unit, i.e., 80 m from the spring outlet (U.S. Fish and Wildlife Service 2013b). The vertical dashed line denotes the furthest downstream detection of E. tonkawae in each headwater creek.