Home Range and Seasonal Movements of the Turtle Graptemys flavimaculata

--An investigation of habitat use, home range, and seasonal movements of adult yellowblotched map turtles (Graptemys flavimaculata) was conducted on the Pascagoula River in southeastern Mississippi. Females were found in deeper, faster water at greater distances from the shoreline in areas with fewer emergent snags than were males. Male distance from the shoreline was inversely correlated with surface current speed during the summer, but female distance was not. Male mean home range area was 1.12 ha; mean home range length was 1.8 km. Female mean home range area was 5.75 ha; mean home range length was 1.5 km. Home range areas and lengths did not differ significantly between the sexes. Graptemys flavimaculata was active in the study area throughout the year. Some turtles were relatively sedentary but others moved seasonally. Seasonal movements appeared to be associated with nesting (females), searching for receptive mates (males), or to avoid increased currents during winter (both sexes). Two females regularly moved into a cypress pond in early spring, presumably to increase their foraging opportunities. The yellow-blotched map turtle, Graptemys flavimaculata, is endemic to the Pascagoula River and its larger tributaries in southeastern Mississippi (Cliburn, 1971; Ernst et al., 1994). Population declines in the Leaf and upper Pascagoula rivers led to its listing as a threatened species in 1991 (U.S. Fish and Wildlife Service, 1993). Little information is available on the biology of G. flavimaculata other than in the original description (Cagle, 1954) or in notes on the distribution of the species (Cliburn, 1971; McCoy and Vogt, unpubl.). The only detailed study of this species was by Brauman and Seigel (in prep.), who investigated reproduction and nesting. Development of effective conservation measures for G. flavimaculata requires an understanding of its habitat requirements and movement patterns. Data on these aspects of its life history are important when evaluating the effects of proposed river modification or other development projects on the species and in determining whether the species could reoccupy areas where populations have declined. The objectives of this study were to acquire information on where G. flavimaculata occurs within its riverine environment, its use of habitat, its movements and spatial requirements, and seasonal changes in these variables. MATERIALS AND METHODS Study Area.-The study area was located on the Pascagoula River within Ward Bayou Wild376 This content downloaded from 207.46.13.150 on Thu, 28 Jul 2016 04:03:36 UTC All use subject to http://about.jstor.org/terms HOME RANGE OF GRAPTEMYS FLAVIMACULATA life Management Area, Jackson County, Mississippi. It included parts of the Pascagoula River, the east and west Pascagoula channels, Parish Lake, and Ward and Parish Lake bayous (Fig. 1). Stream widths ranged from 100-120 m in the Pascagoula River above the east and west channels to 20-50 m in Parish Lake and Ward bayous. Stream depths exceeded 12 m in some locations. Parish Lake was approximately 70 m wide and had depths of up to 3.5 m. Bottom substrates were composed primarily of sand, silty sand, and clays, although the bottom of Parish Lake and its associated channels was a mixture of sand and silt with a large proportion of organic material. The only recording flow gage on the Pascagoula River during the study was near Merrill in George County, approximately 97 river km above the study area. Annual mean water flow at this station for 1931-1992 was 10,090 cfs (Plunkett et al., 1993). Water flow from June 1992 through May 1993 was lowest between June and October and highest between November and May (Fig. 2). Peak flow for the period occurred in January 1993, when discharge exceeded 90000 cfs. Vegetation of the higher banks along the river commonly included Quercus laurifolia, Liquidambar styraciflua, Betula nigra, Acer rubrum, and Carya aquatica. Major understory species were Sabal minor and Arundinaria gigantea. Wetter areas, particularly along Parish Lake Bayou, had mixtures of Nyssa aquatica, N. biflora, Taxodium distichum, and Q. laurifolia. Home Range and Habitat.-I captured 26 adult G. flavimaculata (16 males, 10 females) in basking traps between May 1992 and June 1993. Straight line carapace lengths and midline plastron lengths were measured to the nearest mm with tree and/or dial calipers and weights were measured to the nearest g with a portable field scale. Turtles were fitted with a transmitter package (AVM Instrument Co., Ltd.) consisting of a waterproofed transmitter (150-151 MHz) attached with epoxy to a brass flange. The flange was attached to the carapace with non-galvanized bolts fitted through holes drilled in the marginals. I released turtles at their capture points 24 h after transmitter attachment. The transmitter package, including the bolts, averaged ca. 5% of male body weight and <2% of female body weight. Transmitter packages did not appear to affect either the behavior or health of the turtles, although two telemetered animals were found dead during the study. A male died of unknown causes ca. 1 mo after transmitter attachment, and a female found dead ca. 1 yr after transmitter attachment had been struck by the propeller of a boat. I relocated turtles using an AVM LA12-DS FIG. 1. Map of the study area on the Pascagoula River. The inset shows the location of the study area in southeastern Mississippi. receiver attached to a hand-held three-element Yagi antenna. Turtles were relocated once a day from 1-3 d per week from mid-1992 to November 1993 and at least once every two weeks from November 1993 through August 1994. Transmitter signals were audible at distances of up to 1 km in straight sections of the river if the turtles were basking or near the surface. Turtles were usually relocated to within an area of approximately 2 m in diameter. I recorded two relocation points in a day if the turtle had moved more than 5 m from the point where it had been located earlier. Straight-line distances of relocation points from the river bank and from known reference points along the bank were measured with rangefinders. Relocation points were plotted on scale maps of the study area maintained for each turtle. Water depth, distance from the shoreline, water temperature, number of snags within a 5 m radius, and surface current speed were measured at relocation points for all turtles from August 1992 through September 1993. Current speed was measured with a current velocity meter and water depth was recorded either directly using a measured line or with a battery-operated depth finder. / 377 This content downloaded from 207.46.13.150 on Thu, 28 Jul 2016 04:03:36 UTC All use subject to http://about.jstor.org/terms