Effects of Temperature, Moon Phase, and Prey on Nocturnal Activity in Ratsnakes: An Automated Telemetry Study

—Nocturnal activity is important for many animals, but difficulty in documenting that activity has hampered efforts to understand factors that influence when animals are active at night. We used automated radiotelemetry to provide the first detailed tests of the hypothesis that the nocturnal activity of free-ranging snakes should be influenced by temperature, moon phase, and prey abundance by using data for Ratsnakes (Pantherophis spp.) from Texas and Illinois. Ratsnakes exhibited some nocturnal behavior throughout their active season in both Texas and Illinois, although snakes were much more active at night in Texas than in Illinois. Texas snakes transitioned from primarily diurnal activity to primarily nocturnal activity over this snake’s active season, whereas Illinois snakes were always most active in the middle of the day. For both populations, nocturnal activity was positively related to temperature but unrelated to moon phase. Ratsnakes in Texas exhibited a stepwise increase in nocturnal activity in mid-summer, independent of temperature and coincident with the shift in their diet to almost exclusively mammals active at night. Given the ability of snakes in both populations to be active at night when temperatures allow, warming climates could lead to an increase in nocturnal activity, with consequences for both the snakes and the species on which they prey. Although nocturnal behavior can be a critical aspect of a species biology (e.g., McNeil et al., 1992), documenting nocturnal behavior poses logistic difficulties for many wildlife species. Commonly available methods such as night-viewing technologies (Allison and Destefano, 2006), nocturnal capture rates, and radiotelemetry using conventional hand-tracking (e.g., Greenwood, 1982; Grinder and Krausman, 2001) usually provide incomplete data on nocturnal behavior and are challenging logistically. Satellite radiotelemetry allows continuous monitoring of behavior, including at night (Tucker, 2010), but is available only for species large enough to carry a GPSenabled transmitter (Cooke et al., 2004). Advances in automated radiotelemetry have allowed detailed examination of the nocturnal behavior of a small number of species (Lambert et al., 2009) but has been largely absent from snake research. Here we document nocturnal activity patterns of Ratsnakes (Elaphe obsoleta) using automated radiotelemetry and test several hypotheses about the factors predicted to affect variation in nocturnal behavior. The limited research to date that has examined nocturnal activity of snakes has relied on capture rates (Shine, 1979; Brown and Shine, 2002; Maciel et al., 2003), observations of captive snakes (Moore, 1978), or occasional nocturnal radio tracking using conventional methods (Durner and Gates, 1993). Although these methods have limitations, this research has shown that nocturnal activity appears important for some snake species and highlights the need for more-detailed examination. Automated telemetry has been used extensively to continuously monitor body temperature of free-living snakes (e.g., Brown and Weatherhead, 2000). The use of automated telemetry to quantify activity patterns of free-ranging snakes, however, has been used rarely (Slip and Shine, 1988a,b; Davis et al. 2008), and identification of the factors affecting nocturnal activity using continuously collected data has yet to be undertaken for any snake species. Many snake species vary diel patterns of activity through the active season (e.g., Moore, 1978; Slip and Shine, 1998a). Snakes may shift to nocturnal activity to take advantage of more suitable nighttime temperatures or to avoid dangerously hot daytime temperatures. Nocturnal capture rates have been shown to vary with temperature for several species (Brown and Shine, 2002; Maciel et al., 2003). Also, videography has documented more-extensive predation on birds’ nests at night by Ratsnakes in Texas (Stake and Cimprich, 2003) than in Missouri (Stake et al., 2005), which may be related to latitudinal differences in nocturnal temperatures. A recent analysis of thermoregulation by Ratsnakes also suggested that nocturnal activity increased with temperature and decreased with latitude (Weatherhead et al., 2012). Here we use data collected from Ratsnakes in Texas and Illinois to test the hypothesis that variation in nocturnal behavior is a function of temperature. This hypothesis predicts that Ratsnake activity at night should increase with daytime and nighttime temperatures which, in turn, should lead to more nighttime activity in mid-summer than early and late in the active season. In addition, Ratsnakes should be more active at night in Texas than in Illinois. Factors other than, or in addition to, temperature could also affect nocturnal activity. For example, thermal profiles suggest that Rubber Boas (Charina bottae) may occasionally be active at night when temperatures are suboptimal (Dorcas and Peterson, 1998). A potential nonthermal reason for snakes to be active at night is to synchronize their activity with that of their prey (Marques and Puorto, 1998) or, alternatively, for snakes to reduce their own risk of predation by being active when their predators are less active (Dorcas and Peterson, 1998). We used two indirect approaches to assess the possibility that predator– prey relationships affect nocturnal activity of Ratsnakes. The first approach took advantage of the seasonal change in Ratsnakes’ diet and focused on our data from Texas, where the snakes’ active season is longer. Ratsnakes prey on small mammals throughout their active season but also prey extensively on songbirds when the birds are nesting (Weatherhead et al., 2003; Carfagno et al., 2006; Sperry and Weatherhead, 2009). Nesting songbirds are active primarily during the day, and nesting season in Texas is primarily concentrated between April and July. The four species detected most commonly at our study sites were the Blue-grey Gnatcatcher (Polioptila caerulea), White-eyed Vireo (Vireo griseus), Northern Cardinal (Cardinalis cardinalis), and Painted Bunting (Passerina ciris) (Sperry and Weatherhead 2009), and all have either ceased or have muchreduced breeding effort by August (Ellison, 1992; Hopp et al., Corresponding Author. E-mail: [email protected] DOI: 10.1670/11-325 1995; Halkin and Linville, 1999; Lowther et al., 1999). The three small mammals on which Ratsnakes prey most extensively in Texas (Peromyscus spp., Sigmodon hispidus and Baiomys taylori; Sperry and Weatherhead, 2009) are active primarily at night (Marten, 1973; Cameron and Spencer, 1981; Eshelman and Cameron, 1987). Because the nesting season ends in the middle of the snakes’ active season in Texas, we tested the hypothesis that, if the timing of prey activity affects when the snakes are active, a change in the timing of activity should coincide with when the snakes transition from a mixed diet of birds and mammals to a diet that is comprised largely of mammals. Specifically, the snakes should become more active at night following the end of the bird nesting season. The second indirect approach we used to assess whether predator or prey activity could affect when snakes are active relied on extensive evidence that many small mammal species are less active on moonlit nights (Wolfe and Summerlin, 1989; Daly et al., 1992), whereas some predators are more active on nights with a full moon (Fernandez-Duque, 2003; Jetz et al., 2003). Given that Ratsnakes prey extensively on small mammals, variation in prey availability could lead to increased activity on moonlit nights, particularly given evidence that Ratsnakes are visual predators (Mullin and Cooper, 1998). Alternatively, snakes could be more vulnerable to visual predators on moonlit nights, favoring reduced activity. Several studies have documented increased (Lillywhite and Brischoux, 2011) or decreased (Madsen and Osterkamp, 1982; Clarke et al., 1996; Weaver, 2011) capture or encounter rates of snakes on nights with more moonlight. Therefore, we sought to determine whether Ratsnake nocturnal activity varies with moonlight.