Spatial Ecology of the Oregon Gartersnake, Thamnophis atratus hydrophilus, in a Free-Flowing Stream Environment
Spatial patterns of animals have important implications for population dynamics and can reveal other key aspects of a species' ecology. Movements and the resulting spatial arrangements have fitness and genetic consequences for both individuals and populations. We studied the spatial and dispersal patterns of the Oregon Gartersnake, Thamnophis atratus hydrophilus, using capture–recapture techniques. Snakes showed aggregated dispersion. Frequency distributions of movement distances were leptokurtic; the degree of kurtosis was highest for juvenile males and lowest for adult females. Males were more frequently recaptured at locations different from their initial capture sites, regardless of age class. Dispersal of neonates was not biased, whereas juvenile and adult dispersal were male-biased, indicating that the mechanisms that motivate individual movements differed by both age class and sex. Males were recaptured within shorter time intervals than females, and juveniles were recaptured within shorter time intervals than adults. We attribute differences in capture intervals to higher detectability of males and juveniles, a likely consequence of their greater mobility. The aggregated dispersion appears to be the result of multi-scale habitat selection, and is consistent with the prey choices and related foraging strategies exhibited by the different age classes. Inbreeding avoidance in juveniles and mate-searching behavior in adults may explain the observed male-biased dispersal patterns.Abstract
Linear densities of (A) neonate, (B) juvenile, and (C) adult Thamnophis atratus hydrophilus along Hurdygurdy Creek from 1986 to 2001 (initial captures only). The x-axis lists the mesohabitat units of the study reach (downstream end on the left). Labels are provided for some, but not all, mesohabitat units (see text and Table 1). Mesohabitat lengths are not to scale.
Boxplots of (A) neonate, (B) juvenile, and (C) adult densities of Thamnophis atratus hydrophilus in three mesohabitat types based on water flow and turbulence (Hawkins et al., 1993): fast-water non-turbulent (FWNT), fast-water turbulent (FWT), and slow-water scour pool (SWSP). Letters above boxes represent significance in Tukey-Kramer pair-wise comparisons (α = 0.10); groups with the same letter are not significantly different. The center line represents the median value. The top and bottom of the box represent the 25th and 75th percentiles, the notched area represents the 95% confidence interval, and whiskers represent the upper and lower adjacent values, set at 1.5 times the interquartile range. n = number of individuals observed.
Frequency of microhabitats used by three age classes of Thamnophis atratus hydrophilus. Deep fast microhabitats included fast-water habitats (riffles, cascades, and runs) with > 10 cm water depths; deep slow included slow-water habitats (pools, glides, and edgewaters) with > 10 cm water depths; shallow fast included fast-water habitats with < 1–10 cm water depths; and shallow slow included slow-water habitats with < 1–10 cm water depths (Lind and Welsh, 1994). n = number of individuals observed.
Frequency histograms of movement distances for Thamnophis atratus hydrophilus: (A) all recaptures (n = 532), and (B) juvenile male (n = 68) and adult female (n = 36) recaptures with no change in age class between captures. Zero on the x-axis represents observations of recaptures in the same location as the previous capture; positive values indicate upstream movements and negative values indicate downstream movements.
Boxplots of overall movements in meters per month for Thamnophis atratus hydrophilus (A) snakes that did not change age class between subsequent captures; (B) snakes first captured as juveniles and recaptured as adults; and (C) snakes initially captured as neonates and recaptured as juveniles. See caption on Fig. 2 for details on interpreting boxplot letters and symbols.
Contributor Notes
Associate Editor: J. D. Litzgus.