A comparative study of environmental factors that affect nesting in Australian and North American freshwater turtles

J. Zool., Lond. (2005) 267, 397 404 C 2005 The Zoological Society of London Printed in the United Kingdom doi:10.1017/s0952836905007533 A comparative study of environmental factors that affect nesting in Australian and North American freshwater turtles Kenneth D. Bowen*, Ricky-John Spencer and Fredric J. Janzen Department of Ecology, Evolution and Organismal Biology, Iowa State University, Ames, Iowa, 50011, U.S.A. (Accepted 16 March 2005) Abstract The timing of reproductive events is critical for fitness, and these events are often linked to weather and climate. Weather components are thought to influence the nesting behaviour of freshwater turtles, but to date there have been few quantitative studies and no comparative studies. We compared the environmental cues used by nesting Australian (EmyduramacquariiandChelodina expansa) and North American (Chrysemys picta) freshwater turtles, and quantified the differences in weather between days with and without nesting activity within the nesting season. We also characterized the diel time of nesting for each species. The results suggest that nesting behaviour is related to warm air and water temperatures in C. picta and to rainfall in E. macquarii and C. expansa. Chrysemys picta primarily nests in the afternoon and evening, E. macquarii is a crepuscular nester, and C. expansa nests diurnally. While changes in life history resulting from climate change are difficult to predict, we suggest that an increase in the number of El Niño events may have adverse effects on the two Australian species, whereas increases in environmental temperature may expand the number of nesting opportunities for C. picta. Key words: Emydura, Chelodina, Chrysemys, weather, life history INTRODUCTION The schedule of reproduction at various temporal scales has substantial fitness ramifications. The relative timing of reproductive events within the appropriate season and even within a single day is often critical to reproductive output and success. For example, reproductive success of male Douglas firs Pseudotsuga menziesii is dependent upon phenological overlap with females during the first two-thirds of the breeding season (Burczyk & Prat, 1997), and male garter snakes Thamnophis sirtalis that mate with females first can enforce chastity by placing a copulatory plug in the female s cloaca (Devine, 1975; Shine, Olson & Mason, 2000). Similarly, the timing of propagule dispersal can reduce parental energy expenditure (Congdon & Gatten, 1989) or direct exposure to predation (Tucker, Filoramo & Janzen, 1999), minimize the probability that oviposition sites are detected by predators (Burger, 1977), swamp dispersers or predators with propagules (Eckrich & Owens, 1995), and increase the probability of post-hatch offspring survival (Warner & Andrews, 2003). Given these consequences for fitness, identifying potential mechanisms underlying temporal patterns of reproduction provides crucial * All correspondence to: K. D. Bowen. E-mail: hideneck@yahoo.com information for understanding this fundamental aspect of living organisms. Weather and climate are often linked to both the initiation of the breeding season and the timing of reproductive events within the breeding season. For instance, courtship and nesting in many birds (Heffelinger et al., 1999; Whitehead & Saalfield, 2000) and turtles (Seabrook, 1989; Burke, Gibbons & Greene, 1994; Wilson, Mushinsky & McCoy, 1999) is stimulated by rainfall. Environmental temperatures affect the onset of the nesting season in some birds (MacInnes et al., 1990), the onset of the breeding season in many amphibians (Gibbs & Breisch, 2001), and the interval between nesting events in sea turtles (Solow, Bjorndal & Bolten, 2002). The links between weather, climate, and reproductive events may be impacted by both direct and indirect effects of global warming (Janzen, 1994; Parmesan & Yohe, 2003; Root et al., 2003). Several authors have suggested a connection between weather conditions and the nesting behaviour of freshwater turtle species. Obbard & Brooks (1987) found that the onset of the snapping turtle Chelydra serpentina nesting season is related to spring water temperatures in Quebec. Christens & Bider (1987) found that the onset of the painted turtle Chrysemys picta nesting season in Quebec is related to ambient temperature in summer of the previous year. Iverson & Smith (1993) and Rowe, Coval & Campbell (2003), however, found that onset of the C. picta nesting season in Nebraska and

398 K. D. BOWEN,R.-J.SPENCER AND F. J. JANZEN Michigan, respectively, is related to ambient temperatures in spring of the current year. Red-eared sliders Trachemys scripta elegans in Illinois seem to nest in response to rainfall and increased temperatures (Tucker, 1997), whereas striped mud turtles Kinosternon baurii in Florida (Wilson et al., 1999), common mud turtles Kinosternon subrubrum in South Carolina (Burke et al., 1994), and the broad-shelled turtle Chelodina expansa in Australia (Georges, 1984; Booth, 2002; McCosker, 2002) emerge in response to rainfall. To date, however, there are no studies that quantify a variety of environmental cues to compare nesting behaviour among multiple species of freshwater turtles. Such studies might yield insight into the lifehistory evolution of the species involved and aid in the prediction of the effects of climate change on turtle populations. In the current study, the nesting behaviour of three species of freshwater turtle was compared to determine how they respond to weather conditions. Specifically, the behaviour within the nesting season was examined in an effort to identify the environmental cues that females use as signals to emerge from the water to nest. Three species were compared: Emydura macquarii and Chelodina expansa are sympatric in the Murray River of Australia and Chrysemys picta is widespread in North America. Several hypotheses were tested using this comparison. First, it was hypothesized that E. macquarii and C. picta would respond similarly to environmental cues because they have similar life histories: both species nest in spring/summer, are omnivorous, and prefer to live in slow-moving or stagnant water (Ernst, Lovich & Barbour, 1994; Cann, 1998). Accordingly, it was hypothesized that C. expansa might respond differently to environmental cues than the other two species because it nests during a cooler time of year (autumn; Cann, 1998). Also, embryos of C. expansa undergo a period of secondary diapause and prolonged incubation (Booth, 1998, 2000, 2002). Finally, based on observations made in the field, it was hypothesized that all three species would use rainfall as a cue for nesting. MATERIALS AND METHODS Study animals and field methods Australia The Murray short-necked turtle Emydura macquarii is a widespread chelid turtle (suborder Pleurodira) inhabiting the Murray Darling drainage system, with several forms distributed throughout eastern flowing rivers of coastal New South Wales and Queensland. Emydura macquarii nests from late spring into summer (October December; Cann, 1998). The broad-shelled turtle Chelodina expansa is the largest freshwater chelid in Australia, and is sympatric with E. macquarii throughout much of its range. Chelodina expansa nests in autumn (March until May; Cann, 1998) at the site used in this study. Study sites were located in the upper Murray River near Albury, New South Wales (36. 0863 S, 146.9223 E; see Spencer, 2002a,b; Spencer & Thompson, 2003 for full description). Nocturnal and diurnal searches for female E. macquarii were conducted from late October to mid- December from 1996 to 1998 (see Spencer, 2002a for more detail). Searches did not occur for up to 7 days in 1996 and 1997 at the beginning of November, as eggs were transported to the University of Sydney for a separate experiment (these days were not included in any analyses). Nocturnal and diurnal searches for female C. expansa were conducted from late March to late April in 1997 and 1998. Water temperatures were measured daily using a maximum minimum thermometer placed at a depth of 20 cm in at least 2 lagoons during the nesting season. Weather data, including daily maximum and minimum air temperatures, long-term average maximum and minimum air temperatures, daily precipitation, weather condition at 09:00, 12:00, and 15:00 (rain, clear overcast, thunderstorm), and air temperature at 09:00, 12:00, and 15:00, were obtained from the Australia Bureau of Meteorology (2004). North America The painted turtle Chrysemys picta is a small to mediumsized freshwater turtle (suborder Cryptodira) that ranges from southern Canada to the Gulf of Mexico and from the Atlantic to the Pacific Oceans (Ernst et al., 1994). Chrysemys picta nests from late spring to mid-summer (May July in most areas; Ernst et al., 1994). From this point onward, seasons rather than months are referred to when comparing species because of the difference in timing of seasons between hemispheres. Data presented here represent a portion of a long-term study of the nesting ecology of the painted turtle (Janzen, 1994) at the Thomson Causeway Recreation Area (TCRA) near Thomson, Illinois, U.S.A. (41.9514 N, 90.1165 W). The Thomson Causeway is a c. 450 900 m island on the eastern bank of the Mississippi River, and it contains a c. 1.5-ha nesting area that is bordered on the east side by a 200-m wide slough from which most turtles emerge to nest. The site is managed and maintained by the United States Army Corp of Engineers (USACE; see Kolbe & Janzen, 2002 for a more complete site description). The nesting beach at TCRA was monitored for nesting turtles from early June to late June in 2001 and from late May to early July in 2002 and 2003. The nesting beach was searched for females once every hour from dawn until dusk. In each year, nesting females were observed leaving the water, constructing nests, and laying eggs. All nests were excavated in autumn to count hatchlings; so false nests were identified and removed from the dataset before analysis. Weather data was obtained for the nesting season from the USACE Lock and Dam 13 c. 12 km south of the study site. Data included daily maximum and minimum air temperatures, long-term average maximum and minimum air temperatures, daily precipitation, and water temperature.

Freshwater turtle nesting behaviour 399 Data analysis The weather conditions of nesting and non-nesting days were compared within the nesting season (maximum and minimum air temperature, water temperature, precipitation, the change in maximum and minimum air temperatures from the previous day, and the change in maximum and minimum air temperatures from the long-term average for a given day; air temperature and day condition at 09:00, 12:00, and 15:00 were included for Australian species only) using non-metric multidimensional scaling (nmds), calculated with the software package PRIMER (Clarke, 1993). nmds yields a map-like output in which the relationships among nesting observations can be compared: the nearer the points on the map are to one another, the more similar the nesting patterns. How well the points in space match the actual distance between data points can be measured by stress statistics: stress values < 0.2 give a useful 2-dimensional representation of the data (Clarke & Warwick, 1994). Differences in weather conditions between nesting and non-nesting days were also looked for using analysis of similarities (ANOSIM), a multivariate non-parametric analogue of ANOVA (Clarke & Greene, 1988). SIMPER, an analysis that calculates the average Bray Curtis distance between all pairs of inter-group samples (i.e. distance between each variable for nesting and non-nesting days), was used for post-hoc comparisons. The SIMPER module (Clarke & Warwick, 1994) identifies environmental factors that make an individual contribution of > 2% to the ANOSIM result. The same statistical methods were used to compare among species. In addition to these comparisons, the diel time of nesting for all species was characterized by plotting the number of turtles observed nesting at different times of day. Turtles are known to retain fully developed eggs for long periods while apparently waiting for suitable nesting conditions (Galbraith, Graesser & Brooks, 1988; Buhlmann et al., 1995), and female E. macquarii have been artificially induced to oviposit up to 2 weeks before the onset of nesting on the Murray River (Spencer, 2001). In our comparisons of nesting vs non-nesting days, weather data for up to 2 weeks before the first turtle observed nesting was therefore included in addition to data for nonnesting days throughout the nesting season. Data were standardized (to a mean of 0 and standard deviation of 1) before analysis to remove any effect of different scales of measurement for the different variables (Legendre & Legendre, 1998). RESULTS Comparison of nesting vs non-nesting days Differences in climate during the nesting period relate to season, but comparing the weather conditions of nesting vs non-nesting days determines environmental preferences (and avoidances) of species within the season. Seventy-six E. macquarii nests from 33 nesting days, 29 C. expansa nests from 22 nesting days, and 504 C. picta nests from Table 1. SIMPER analysis of dissimilarity of weather conditions during nesting and non-nesting days for Emydura macquarii. Change in daily MAX and MIN, change in maximum and minimum temperatures from the previous day, respectively, for a given day during the nesting season; change in average MAX and MIN, change in maximum and minimum temperatures from the long-term average for a given day during the nesting season. Only climate conditions that contributed substantially to differences between nesting and non-nesting days are displayed (see text for more detail) Average value Average Non- contribution Variable Nesting nesting (%) Rainfall (mm) 5.6 0.3 15.3 Change in daily MAX ( C) 1.1 0.6 14.0 Change in daily MIN ( C) 0.6 0 11.7 Change in average MIN ( C) 2.0 0.5 10.1 MIN ( C) 12.2 9.8 9.7 Air temperature at 15:00 ( C) 22.5 24.3 9.5 MAX ( C) 24.3 25.7 8.7 Change in average MAX ( C) 1.2 0.2 8.7 93 nesting days were used in our analyses. Multidimensional scaling of environmental variables produced a plot with good separation of nesting and non-nesting days and a good stress value for all species. Weather conditions differed significantly between nesting and non-nesting days for all species (E. macquarii, nmds stress = 0.09, ANOSIM global R = 0.20, P = 0.001; C. expansa,nmds stress = 0.08, ANOSIM global R = 0.31, P = 0.001; C. picta, nmds stress = 0.1, ANOSIM global R = 0.48, P = 0.001). Rainfall made the greatest contribution to the dissimilarity between nesting and non-nesting days in E. macquarii (15%). Emydura macquarii individuals preferred to nest during or after rainfall. Minimum air temperatures were slightly warmer and maximum air temperatures slightly cooler than the previous day and the long-term average (Table 1). Similarly, C. expansa individuals generally nested during or shortly after rain, although changes in air temperatures contributed greatest to the dissimilarity between nesting and non-nesting days, with turtles avoiding cooler days with below average temperatures (Table 2). In contrast, temperature related variables, rather than rainfall, make the greatest contribution to the dissimilarity between nesting and non-nesting days in C. picta (> 80%; Table 3). The air temperature on nesting days was c. 7 C warmer than non-nesting days, and the water temperature on nesting days was c. 5 C warmer. The air temperature on non-nesting days was generally 5 C below the longterm average (Table 3). Once the nesting season began, C. picta individuals were observed nesting almost every day throughout the nesting season, whereas the two Australian species nested more sporadically (Fig. 1). Comparison among species Multidimensional scaling of environmental variables on days of nesting produced a plot with a separation of species

400 K. D. BOWEN,R.-J.SPENCER AND F. J. JANZEN Table 2. SIMPER analysis of dissimilarity of weather conditions during nesting and non-nesting days for Chelodina expansa. Change in daily MAX and MIN, change in maximum and minimum temperatures from the previous day, respectively, for a given day during the nesting season; change in average MAX, change in maximum temperature from the long-term average for a given day during the nesting season. Only climate conditions that contributed substantially to differences between nesting and non-nesting days are displayed (see text for more detail) Average value Average Non- contribution Variable Nesting nesting (%) Change in average MAX ( C) 0.4 3.1 12.7 Change in daily MAX ( C) 0.6 0.6 11.5 Rainfall (mm) 6.3 0.0 11.4 Temperature at 15:00 ( C) 23.7 20.0 10.8 MAX ( C) 24.7 21.7 10.8 Change in daily MIN ( C) 0.4 0.3 10.7 MIN ( C) 9.3 6.9 10.0 Table 3. SIMPER analysis of dissimilarity of climate conditions during nesting and non-nesting days for Chrysemys picta. Change in daily MAX and MIN, change in maximum and minimum temperatures from the previous day, respectively, for a given day during the nesting season; change in average MAX, change in maximum temperature from the long-term average for a given day during the nesting season. Only climate conditions that contributed substantially to differences between nesting and non-nesting days are displayed (see text for more detail) Average value Average Non- contribution Variable Nesting nesting (%) Change in daily MAX ( C) 0.9 0.2 19.3 Change in daily MIN ( C) 0.4 0.4 18.3 MIN ( C) 15.4 8.6 17.3 MAX ( C) 26.4 18.4 13.3 Change in average MAX ( C) 1.1 3.8 10.8 Water temperature ( C) 21.5 16.7 10.8 Proportion of days 1 0.8 0.6 0.4 0.2 0 Emydura macquarii Chelodina expansa Species Chrysemys picta Fig. 1. Proportion of days that turtles were (white portion of bars) or were not (grey portion of bars) found nesting during the entire nesting season. Days included were from the first turtle observed nesting until the last turtle observed nesting. and a reasonable stress (0.15). There was a statistically significant difference in the weather conditions of nesting days between species (ANOSIM global R = 0.33, P = 0.001). Both of the Australian species were likely to nest under a wider variety of weather conditions than C. picta, and nested under different weather conditions than C. picta overall (ANOSIM global R = 0.30, P = 0.001 for E. macquarii; ANOSIM global R = 0.49, P = 0.001 for C. expansa). There was only a minor difference in the weather conditions of nesting between E. macquarii and C. expansa (ANOSIM global R = 0.07, P = 0.05) despite these species nesting in different seasons. Both E. macquarii and C. expansa individuals tend to nest when it is raining and maximum air temperatures are lower than the day before. Because E. macquarii individuals nest in spring, maximum and minimum air temperatures are greater than for the autumn-nesting C. expansa. The primary difference in nesting weather between the Australian species and C. picta seems to be rainfall: the Australian species generally nest during or after rain (cf. Tables 1 & 2 to Table 3). Diel time of nesting Emydura macquarii has a bimodal daily nesting activity with peaks occurring near dawn and dusk (Fig. 2a). Chelodina expansa individuals were observed nesting only during daylight hours (Fig. 2b). Chrysemys picta individuals nest throughout the day with peak activity several hours before dusk (Fig. 2c). DISCUSSION The timing of reproductive events within the reproductive season has wide-ranging fitness ramifications. Reproduction early or late in the breeding season is related to survival and fecundity of adults (Brinkhof et al., 2002) and the survival of offspring (Einum & Fleming, 2000). The timing of reproduction with respect to unpredictable environmental factors within the reproductive season may affect the probability of reproducing successfully and can have indirect effects on fecundity (Thorhallsdottir, 1998; Satake, Sasaki & Iwasa, 2001). Describing the timing of reproductive events in relation to environmental conditions within the reproductive season is thus a critical component of the study of life-history evolution. Emydura macquarii and C. expansa respond strongly to rain and the air temperature changes associated with rain, usually emerging to nest during or after storms. Contrary to our expectations, this behaviour differs from that of C. picta, which respond to air and water temperature by predominantly nesting when temperatures are at or above average during the nesting season. How do these differences provide insight into life-history evolution of these species? The nesting behaviour of C. picta in response to environmental cues may be explained by the thermal energy needed for successful nesting. Chrysemys picta

Freshwater turtle nesting behaviour 401 10 9 (a) 8 7 6 5 4 3 2 1 0 04.00 06.00 08.00 10.00 12:00 14:00 16:00 18:00 20:00 22:00 00:00 02:00 6 (b) 5 4 Turtles (n) 3 2 1 0 04.00 06:00 08:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 00:00 02:00 120 100 80 60 40 20 0 04:00 06:00 08:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 00:00 02:00 (c) Time Fig. 2. Daily nesting activity of (a) Emydura macquarii, (b) Chelodina expansa and (c) Chrysemys picta. The white portion of the bar underneath the histogram corresponds to daylight hours during the nesting season.

402 K. D. BOWEN,R.-J.SPENCER AND F. J. JANZEN may require warm morning and midday temperatures for basking before nesting in the afternoon. Basking functions in thermoregulation (Boyer, 1965; Crawford, Spotila & Standora, 1983; Schwarzkopf & Brooks, 1985), drying and ectoparasite removal (Cagle, 1950; Neill & Allen, 1954; Boyer, 1965), digestive facilitation (Moll & Legler, 1971), vitamin D synthesis (Pritchard & Greenhood, 1968), and may aid in eggshell development before nesting. Evening temperatures at TCRA can be highly variable and several turtles were observed stranded overnight on the nesting beach during cool evenings. Increased body temperature from basking may hasten nest construction and reduce the amount of time spent on land by a female. It is possible that any day that is warm enough to obtain sufficient thermal energy for nesting is suitable. Once ambient temperatures are warm enough for the nesting season to begin, factors other than weather may be driving the decision to emerge and nest. This hypothesis would explain the fact that very few days lack nesting turtles once the nesting season begins for C. picta. Unlike C. picta, rainfall seems to be a critical determinant of nesting behaviour within the nesting season for E. macquarii and C. expansa. Rain may soften the ground and make nest excavation easier and thus faster (Seabrook, 1989), minimizing the time that turtles spend on land and are exposed to predators. Water may also play an important role in nest construction. During excavation, female C. expansa and E. macquarii release fluid from their cloaca (Goode & Russell, 1968), presumably to soften the substrate. Predators use both visual and olfactory cues to detect nests (Spencer, 2002a) and nesting during rain may have the dual effect of diluting body fluid (i.e. urine and scent trails; Burke et al., 1994) and disguising the nest by compacting loose soil or disturbing the ground surrounding the nest. Rain would also minimize evaporative water loss of nesting turtles (Wilson et al., 1999). Because many of these factors could also benefit C. picta, we are currently unable to explain this difference between the turtle species. However, consistently high levels of nest depredation at the Australian sites (Spencer, 2002a; Spencer & Thompson, 2003) compared to fluctuating levels at TCRA (Kolbe & Janzen, 2002) and different nest predators at the sites (primarily fox Vulpes vulpes in Australia and raccoon Procyon lotor at TCRA) may be responsible. Each species nests during different periods of the day. Nesting of C. expansa in northern Australia is also associated with rainfall (Georges, 1984; Booth, 2002; McCosker, 2002) and occurs during the day (Georges, 1984). Emydura macquarii are associated with nesting during the evening after rain as well (Green, 1996). Peak nesting occurs in the afternoon or evening hour in C. picta (Congdon & Gatten, 1989; Rowe et al., 2003). The diel nesting activity of C. picta is probably linked to basking activity, as discussed earlier. However, the differences in diel activity between the two Australian species may be more difficult to interpret. Spring summer storms usually occur in the late afternoon in south-eastern Australia (Australia Bureau of Meteorology, 2004) and nesting of E. macquarii individuals generally occurs en masse in discrete areas close to shore (Spencer, 2002a) throughout the night and into the following morning. Chelodina expansa nest individually much farther from shore than E. macquarii (Georges, 1984; Spencer & Thompson, 2004) and diurnal nesting may simply embody the process of finding an adequate nesting site and returning to the water. Diel activity may also reflect different strategies of predator avoidance, as E. macquarii is known to avoid predators during nesting (Spencer, 2002a; Spencer & Thompson, 2003). It is difficult to predict the precise effects of climate change on life history evolution, in part because physiological or genetic constraints may prohibit or minimize climate-driven changes in any particular lifehistory trait (Winkler, Dunn & McCulloch, 2002). However, the close coupling of the nesting of Australian turtles with rainfall events suggests that a decrease in rainfall may adversely affect the reproduction of these animals. This finding is important considering that El Niño Southern Oscillation (ENSO) events result in dry conditions in Australia. If the frequency and intensity of ENSO events continue to increase (Trenberth & Hoar, 1997), the decreasing occurrence of storms and rain may adversely affect E. macquarii and C. expansa by delaying nesting, making nest construction more difficult, or making nests more conspicuous to predators. It is well known that global temperatures are increasing (IPCC, 2001), and while this warming may adversely affect the sex determination of C. picta (Janzen, 1994), little effect on the nesting behaviour of this species is anticipated within the nesting season. At this time, some C. picta individuals emerge on almost every day of the nesting season. Rising temperatures are unlikely to alter this pattern because average or above average temperatures characterize nesting days. An interesting avenue for future study would be to determine if temperature increases affect the initiation and duration (i.e. number of clutches) of the nesting season of C. picta. The implications of such a change for reproductive energetics and population dynamics could be substantial. Phylogeny has not been considered in these comparisons, and our interpretations are therefore tentative. For example, the similarities and differences in nesting behaviour discussed here could result from phylogenetic constraint. Future studies should focus on addressing this possibility by adding more species to the comparison and then adopting analytical approaches that specifically incorporate phylogenetic relationships. We also advocate continued monitoring of these and other populations of freshwater turtles and their responses to weather. Such long-term studies may provide valuable insight into the effects of climate change on natural systems. Acknowledgements We thank the U.S. Army Corps of Engineers for continued access to the Illinois field site and the Janzen Lab turtle camp crews from 2001 to 2003 for helping with data collection at this site. The research in Illinois was

Freshwater turtle nesting behaviour 403 conducted under approved animal care protocol 1-1- 4733-J from Iowa State University and by permission of the U.S. Fish & Wildlife Service (2001 permit 32576-01017; 2002 permit 32576-02024; 2003 permit 32576-03014) and the Illinois Department of Natural Resources (2001 permit NH-01.0073; 2002 permit NH-01.0073; 2003 permit NH-01.0073). Funding for this part of the study was provided by NSF grants DEB0089680 and IBN0080194 to FJJ. Research in Australia was supported by a Reserves Advisory Committee Environmental Trust Grant and an Australian Research Council Small Grant, and was conducted under New South Wales National Parks and Wildlife Service permit B1313, New South Wales Fisheries License F86/2050, and University of Sydney Animal Care and Ethic Committee approval L04/12-94/2/2017. We thank M. Thompson for support during this part of the study, as well as the Webb, Griffith, and Delaney families, R. Ruwolt, and Z. Ruwolt for their support and use of their property. KDB acknowledges the support of an Iowa State University Graduate College Fellowship. REFERENCES Australia Bureau of Meteorology (2004). www.bom.gov.au. Booth, D. T. (1998). Nest temperature and respiratory gases during natural incubation in the broad-shelled river turtle, Chelodina expansa (Testudinata: Chelidae). Aust. J. Zool. 32: 592 596. Booth, D. T. (2000). Incubation of eggs of the Australian broadshelled turtle, Chelodina expansa (Testudinata: Chelidae), at different temperatures: effects on pattern of oxygen consumption and hatchling morphology. Aust.J.Zool. 48: 369 378. Booth, D. T. (2002). The breaking of diapause in embryonic broadshelled river turtles (Chelodina expansa). J. Herpetol. 36: 304 307. Boyer, D. R. (1965). Ecology of the basking habit in turtles. Ecology 46: 99 118. Brinkhof, M. W. G., Cave, A. J., Daan, S. & Perdeck, A. C. (2002). Timing of current reproduction directly affects future reproductive output in European coots. Evolution 56: 400 411. Buhlmann, K. A., Lynch, T. K., Gibbons, J. W. & Greene, J. L. (1995). Prolonged egg retention in the turtle Deirochelys reticularia in South Carolina. Herpetologica 51: 457 462. Burczyk, J. & Prat, D. (1997). Male reproductive success in Pseudotsuga menziesii (Mirb.) Franco: the effects of spatial structure and flowering characteristics. Heredity 79: 639 647. Burger, J. (1977). Determinants of hatching success in Diamondback terrapin, Malaclemys terrapin. Am. Midl. Nat. 97: 444 464. Burke, V. J., Gibbons, J. W. & Greene, J. L. (1994). Prolonged nesting forays by common mud turtles (Kinosternon subrubrum). Am. Midl. Nat. 131: 190 195. Cagle, F. R. (1950). The life history of the slider turtle, Pseudemys scripta troostii, (Holbrook). Ecol. Monogr. 20: 31 54. Cann, J. (1998). Australian freshwater turtles. Singapore: Beaumont. Christens, E. & Bider, J. R. (1987). Nesting activity and hatching success of the painted turtle (Chrysemys picta marginata) in southwestern Quebec. Herpetologica 43: 55 65. Clarke, K. R. (1993). Non-parametric multivariate analyses of changes in community structure. Aust. J. Ecol. 18: 117 143. Clarke, K. R. & Greene, R. H. (1988). Statistical design and analysis for a biological effects study. Mar. Ecol. Prog. Ser. 46: 213 226. Clarke, K. R. & Warwick, R. M. (1994). Change in marine communities: an approach to statistical analysis and interpretation. Plymouth: National Environment Research Council. Congdon, J. C. & Gatten, R. E. (1989). Movements and energetics of nesting Chrysemys picta. Herpetologica 45: 94 100. Crawford, K. M., Spotila, J. R. & Standora, E. A. (1983). Operative environmental temperature and basking behavior in the turtle, Pseudemys scripta. Ecology 64: 989 999. Devine, M. C. (1975). Copulatory plugs in snakes: enforced chastity. Science 187: 844 845. Eckrich, C. E. & Owens, D. W. (1995). Solitary versus arribada nesting in the Olive Ridley sea turtles (Lepidochelys olivacea): a test of the predator satiation hypothesis. Herpetologica 51: 349 354. Einum, S. & Fleming, I. A. (2000). Selection against late emergence and small offspring in Atlantic salmon (Salmo salar). Evolution 54: 628 639. Ernst, C. H., Lovich, J. E. & Barbour, R. W. (1994). Turtles of the United States and Canada. Washington, DC: Smithsonian Institution Press. Galbraith, D. A., Graesser, C. J. & Brooks, R. J. (1988). Egg retention by a common snapping turtle (Chelydra serpentina) in northcentral Ontario. Can. Field-Nat. 102: 734. Georges, A. (1984). Observations on the nesting and natural incubation of the long-necked tortoise Chelodina expansa. Herpetofauna 15: 27 31. Gibbs, J. P. & Breisch, A. R. (2001). Climate warming and calling phenology of frogs near Ithaca, New York, 1900 1999. Conserv. Biol. 15: 1175 1178. Goode, H. J. & Russell, J. (1968). Incubation of eggs of three species of chelid tortoise, and notes on their embryological development. Aust. J. Zool. 16: 749 761. Green, D. (1996). The Macquarie river turtle Emydura macquarii. J. Vict. Herpetol. Soc. 8: 73 77. Heffelinger, J. R., Guthery, F. S., Olding, R. J., Cochran, C. L. & McMullen, C. M. (1999). Influence of precipitation timing and summer temperatures on reproduction of Gambel s quail. J. Wildl. Manage. 63: 154 161. IPCC. (2001). Climate change 2001: synthesis report. Watson, R. T. (Ed.). Cambridge: Cambridge University Press. Iverson, J. B. & Smith, G. R. (1993). Reproductive ecology of the painted turtle (Chrysemys picta) in the Nebraska Sandhills and across its range. Copeia 1993: 1 21. Janzen, F. J. (1994). Climate change and temperature-dependent sex determination in reptiles. Proc. natl Acad. Sci. U.S.A. 91: 7487 7490. Kolbe, J. J. & Janzen, F. J. (2002). Spatial and temporal dynamics of turtle nest predation: edge effects. Oikos 99: 538 544. Legendre, P. & Legendre, L. (1998). Numerical ecology. Amsterdam: Elsevier. MacInnes, C. D., Dunn, E. H., Rusch, D. H., Cooke. F. & Cooch, F. G. (1990). Advancement of goose nesting dates in the Hudson Bay region, 1951 1986. Can. Field-Nat. 104: 295 297. McCosker, J. (2002). Chelodina expansa (broad shell river turtle) and Emydura signata (Brisbane shortneck turtle). Reproduction. Herpetol. Rev. 33: 198 199. Moll, E. O. & Legler, J. M. (1971). Life history of a neotropical slider turtle, Pseudemys scripta (Schoepff), in Panama. Bull. Los Angeles Co. Mus. nat. Hist. 11: 1 102. Neill, W. T. & Allen, E. R. (1954). Algae on turtles: some additional considerations. Ecology 35: 581 584. Obbard, M. E. & Brooks, R. J. (1987). Prediction of the onset of the annual nesting season of the common snapping turtle, Chelydra serpentina. Herpetologica 43: 324 328. Parmesan, C. & Yohe, G. (2003). A globally coherent fingerprint of climate change impacts across natural systems. Nature (Lond). 421: 37 42.

404 K. D. BOWEN,R.-J.SPENCER AND F. J. JANZEN Pritchard, P. C. H. & Greenhood, W. F. (1968). The sun and the turtle. Int. Turt. Tort. Soc. J. 2: 20 25. Root, T. L., Price, J. T., Hall, K. R., Schneider, S. H., Rosenzweig, C. & Pounds, J. A. (2003). Fingerprints of global warming on wild animals and plants. Nature (Lond). 421: 57 60. Rowe, J. R., Coval, K. A. & Campbell, K. C. (2003). Reproductive characteristics of female Midland painted turtles (Chrysemys picta marginata) from a population on Beaver Island, Michigan. Copeia 2003: 326 336. Satake, A., Sasaki, A. & Iwasa, Y. (2001). Variable timing of reproduction in unpredictable environments: adaption of flood plain plants. Theor. Pop. Biol. 60: 1 15. Schwarzkopf, L. & Brooks, R. J. (1985). Application of operative environmental temperatures to analysis of basking behavior in Chrysemys picta. Herpetologica 41: 206 212. Seabrook, W. (1989). The seasonal pattern and distribution of green turtle (Chelonia mydas) nesting activity on Aldabra atoll, Indian Ocean. J. Zool. (Lond.) 219: 71 81. Shine, R., Olson, M. M. & Mason, R. T. (2000). Chastity belts in gartersnakes: the functional significance of mating plugs. Biol. J. Linn. Soc. 70: 377 390. Solow, A. R., Bjorndal, K. A. & Bolten, A. B. (2002). Annual variation in nesting numbers of marine turtles: the effect of sea surface temperature on re-migration intervals. Ecol. Lett. 5: 742 746. Spencer, R.-J. (2001). The Murray River turtle, Emydura macquarii: population dynamics, nesting ecology and impact of the introduced red fox, Vulpes vulpes. PhD thesis, University of Sydney. Spencer, R.-J. (2002a). Experimentally testing nest-site selection: fitness trade-offs and predation risk in turtles. Ecology 83: 2136 2144. Spencer, R.-J. (2002b). Growth patterns in two widely distributed freshwater turtles and a comparison of common methods used to estimate age. Aust. J. Zool. 50: 477 490. Spencer, R.-J. & Thompson, M. B. (2003). The significance of predation in nest site selection of turtles: an experimental consideration of macro- and microhabitat preferences. Oikos 102: 592 600. Spencer, R.-J. & Thompson, M. B. (2004). Experimental analysis of the impact of foxes on freshwater turtle populations using large-scale field and modelling techniques: implications for management. Conserv. Biol. 19: 845 854. Thorhallsdottir, T. E. (1998). Flowering phenology in the central highland of Iceland and implications for climatic warming in the arctic. Oecologia (Berl.) 114: 43 49. Trenberth, K. E. & Hoar, T. J. (1997). El Nino and climate change. Geophys. Res. Lett. 24: 3057 3060. Tucker, J. K. (1997). Natural history notes on nesting, nests, and hatchling emergence in the red-eared slider turtle, Trachemys scripta elegans, in West-central Illinois. Ill. Nat. Hist. Surv. Biol. Notes. 140: 1 13. Tucker, J. K., Filoramo, N. I. & Janzen, F. J. (1999). Size-biased mortality due to predation in a freshwater turtle, Trachemys scripta. Am. Midl. Nat. 141: 198 203. Warner, D. A. & Andrews, R. M. (2003). Consequences of extended egg retention in the Eastern fence lizard (Sceloporus undulatus). J. Herpetol. 37: 309 314. Whitehead, P. J. & Saalfield, K. (2000). Nesting phenology of magpie geese (Anseranas semipalmata) in monsoonal northern Australia: responses to antecedent rainfall. J. Zool. (Lond.) 251: 495 508. Wilson, D. S., Mushinsky, H. R. & McCoy, E. D. (1999). Nesting behavior of the striped mud turtle, Kinosternon baurii (Testudines: Kinosternidae). Copeia 1999: 958 968. Winkler, D. W., Dunn, P. O. & McCulloch, C. E. (2002). Predicting the effects of climate change on avian life-history traits. Proc. natl Acad. Sci. U.S.A. 99: 13595 13599.