Population ecology of green snakes (Opheodrys aestivus) revisited

A population of rough green snakes (Opheodrys aestivus) was studied by markrecapture at Ransom Lake (RL) in north-central Arkansas over 7 yr (1986-1992). Results are compared to those obtained previously from a population in similar habitat located 12 km distant (Bald Knob Lake, BKL). Over 2700 captures made along the RL shoreline were nonrandomly distributed within areas of dense vegetation. Green snakes lived within home ranges averaging 67 m in shoreline length and differing in location between years by about 50 m. Individuals were recaptured on average every 19 days, in which time they had moved linearly along the shoreline about 28 m. About 3% of all movements were over water. Each of these spatial parameters at RL was very similar to those obtained at BKL. Growth rates of individual snakes decreased linearly with snout-vent length (SVL) and were similar to those characterizing snakes at BKL. However, greater variance in growth rates at RL resulted in a weaker relationship between body size and age. Females grew to be larger than males. Body size of females was slightly larger at RL than at BKL. Younger age groups were proportionally larger at RL than at BKL. Sex ratio was 1:1 both in adults and in all snakes combined. In 1987 and 1988 when about 60% of the population consisted of adult snakes, sex and age structure of the population were similar between years. Maximum population density was about 800 snakes/ha. Catchability of individual snakes in monthly samples was equal in 1987 but equivocal in 1988. Survivorship was measured directly by agespecific recapture proportions. Adult survivorship (males 28%, females 41%) was less at RL than at BKL. Calculated overall first-year survivorship (21.5%) was similar to that at BKL and was slightly greater than hatchling survivorship measured directly by recapture (15-19%). Survivorship for ages -0 in both sexes showed a linear logarithmic decline with age (type II survivorship curve). As at BKL, overall clutch size at RL averaged 6 eggs. Unlike at BKL, clutch size at RL differed among years concordant with body condition (weight-length ratio) of snakes. As at BKL, the net reproductive rate at RL (RO = 0.60) was insufficient for sustaining the population over the long term. Estimated population size varied fivefold over 7 yr and decreased from approximately 350-500 snakes in the first 3 yr (1986-1988) to 100-200 in the last 4 yr (1989-1992). The rapid population decline apparently was due to unusually high mortality among adults, and the increased mortality, in turn, was correlated with unusually hot and dry weather leading to a decrease in the body condition of snakes. Predation by snakes and birds may have contributed to the decline. Although snake populations are often regarded as being relatively stable, this study demonstrates that populations of a small insectivorous species are not immune to environmental vagaries and may respond by decreasing density as do other vertebrates.