Selection and validation of release sites for conservation translocations of temperate-zone snakes

Abstract

Poor habitat quality is one of the most important reasons for reintroduction failure of reptiles; therefore, release site suitability ought to be evaluated prior to conducting conservation translocations. In temperate zone snakes, translocations have failed due to high overwinter mortality, so practitioners have recommended that release sites be located near suitable hibernacula. The presence of a Life Zone (LZ), the underground space above the groundwater table and below the frost line, may indicate the presence of suitable hibernation habitat. Identification and validation of sites with LZ, however, is challenged by the dynamic nature of groundwater and frost levels, coupled with the secretive nature of hibernating snakes. In this study, our goals were to 1) identify potential reintroduction sites for the globally imperiled Eastern Massasauga (Sistrurus catenatus) in southern Canada based on the presence of a LZ, 2) to validate the suitability of those reintroduction sites by hibernating a surrogate species (Eastern Gartersnake, Thamnophis sirtalis) in constructed hibernacula, and, 3) to determine if two separate measures of LZ were associated with the survival of individual gartersnakes to the end of hibernation. Four 1-ha study grids, each consisting of multiple groundwater and frost monitoring stations, were surveyed from 2015 to 19. Release sites were identified in each grid where a LZ of ≥ 10 cm was observed consistently for at least 2 full winters, and gartersnakes were successfully hibernated therein and at 2 reference sites over 3 winters. Overall, survival of subadult/adult snakes was very high (100 %; n = 20), regardless of site, whereas juvenile survival was lower (78 %; n = 93). A Kaplan-Meier test indicated that juvenile survival differed significantly among sites, and ranged from 60 % to 100 %. GLMM analysis indicated that mass at ingress had a significant positive effect on snake survival during hibernation. The temporary loss of a LZ did not impact snake survival. Contrary to our expectations, snake survival in hibernacula was negatively associated with both minimum LZ size and LZ frequency (i.e., the % of sampling occasions when LZ size ≥ 10 cm). Hibernating snakes can tolerate periods with a very small or non-existent LZ, but a large LZ may become unsuitable to juveniles, possibly due to a lack of moisture. Our results will guide the selection of release sites for Massasauga reintroductions, and for the development of a rigorous release site selection process for temperate zone snakes.