Use of automated radio telemetry to detect nesting activity in ornate box turtles, Terrapene Ornata.
Behavioral data may be collected in the field by a variety of methods. Focal observations allow researchers to make detailed observations (Ruby and Niblick, 1994) but require significant time investment to collect and may influence the behavior of study animals (Baker and McGuffin, 2007). Video recordings of animal behavior may reduce the impacts of researcher presence (Beringer et al., 2004) and facilitate simultaneous monitoring of multiple individuals (MacNulty et al., 2008) but can also generate more data than can be easily processed (Kleist et al., 2007) and may not be feasible for longterm observations of small animals that do not concentrate their activities at predictable locations. In such instances, researchers may rely on radio telemetry to obtain desired information on animal location, behavior, or physiology (Wikelski et al., 2003; Cooke et al., 2004; Rutz and Hays, 2009). Depending on study population and habitat characteristics, radio tracking can increase search efficiency over other approaches, such as walking transects (Refsnider et al., 2011), but because of the time involved in locating a large number of transmittered animals, a single investigator is still forced to choose between making detailed observations of a single animal or obtaining lower resolution observations of multiple animals (Bradley et al., 2004).
Automated activity monitoring via the signal change method is an extension of radio telemetry that can be used to monitor the activity patterns of many animals simultaneously (Kays et al., 2011). This approach relies on the principle that any movement of a standard radio transmitter can strongly affect the intensity of the transmitter's signal at a stationary receiving station (Cochran and Lord, 1963; Kjos and Cochran, 1970; Kays et al., 2011; Ward et al., in press). Thus, movement or changes in orientation of an animal fitted with a transmitter, with or without changes in location, may result in temporal changes in the strength of the signal reaching a stationary receiver. These changes in signal strength can be used to distinguish activity from inactivity (Cochran and Lord, 1963; Kjos and Cochran, 1970). Many researchers have subsequently used radio signal strength recordings to assess the frequency and duration of animal activity bouts (Kjos and Cochran, 1970; Sunquist and Montgomery, 1973; Kays et al., 2011; Ward et al., in press), seasonal differences in activity patterns (Ables, 1969; Risenhoover, 1986), and effects of weather oil activity (Broekhuizen et al., 1980; Lancia et al., 1982; Sperry et al., 2013). Additionally, when coupled with information about the species' natural history, activity monitoring can be used to make inferences about what an animal is doing when it is active (Nams, 1989). Inferential techniques are especially useful when the number of individuals being monitored is large, when direct visual observation is impaired, or when a researcher's presence may affect an animal's behavior (Plummet, 2003). However, they generally lack the precision that can be achieved by direct observation or video surveillance (Ruby and Niblick, 1994). Attempts to categorize different activities based upon radio signal strengths have proven challenging (Nams, 1989), but simply distinguishing activity from inactivity can be accomplished in near real time (Kays et al., 2011) and is easy to perform by researchers in the field.
In this paper, we describe how automated radio telemetry and the signal change method may be used to detect nesting activity in ornate box turtles (Terrapene ornata). Many studies involving turtles rely on the ability of investigators to find nests. For example researchers often use turtles to study life history and reproductive strategies (Congdon et al., 2001; Valenzuela and Janzen, 2001), but determining reproductive success is difficult when nests are not easily located (Hellgren et al., 2000). Additionally, for many turtle species, finding nesting females is one of the few approaches available for obtaining reproductive data needed to complete life table analyses that are used to guide conservation and management measures (Congdon et al., 1993, 1994).
Ornate box turtles are small and wary animals that are known to nest at night but are otherwise diurnal (Legler, 1960; Converse et al., 2002). Observations of this species have indicated that females may construct preliminary cavities over a period of several nights before ovipositing (Legler, 1960; Doroff and Keith, 1990). This, combined with the fact that nest sites are often located in heterogeneous environments, and the possibility that females may bury themselves while nesting (C. Tucker and T. Radzio, pers. obs.) can make visual detection of nesting females difficult and time consuming, even with the aid of radio tracking. However, since ornate box turtles are usually only active at night when nesting, we investigated whether the signal change method could be used to efficiently identify nesting activity by monitoring individuals for night activity in near real time. We hypothesized that night activity would correspond to nesting, enabling a single researcher to efficiently locate nesting females by selectively tracking only those females which were active at night, thus reducing the effort required to locate nesting females.
MATERIALS AND METHODS
We conducted our study on a remnant sand prairie at the Upper Mississippi River National Wildlife and Fish Refuge in Carroll County, Illinois (Zone 15T in NAD83; 0738846 E, 4657387 N). The study site is 146 ha and is bordered to the west by a slough of the Mississippi River and to the east by a dirt road and railroad tracks. The site, which is bisected by a paved bicycle trail, is dominated by native remnant sand prairie vegetation and has been identified as a high quality vegetation community (Ebinger et al., 2006). Ornate box turtles at this site have been the subject of several past studies (Bowen et al., 2004; Kuo and Janzen, 2004; Refsnider et al., 2011; Refsnider et al., 2012) but their nesting biology remains unaddressed, in part due to the difficulty of finding nests.
STUDY ANIMALS AND RADIO TRANSMITTERS
In spring 2010 and 2011, we hand captured adult ornate box turtles and fitted individuals with standard two-stage radio transmitters (164-165 MHz, ~12-13 g; Advanced Telemetry Systems, Isanti, Minnesota). Transmitters affixed to turtle carapaces using 5 min epoxy were mounted in an offcenter orientation (Bernstein and Black, 2005) with posterior free trailing antennas (22 cm). The combined mass of transmitters and epoxy was 4-6% of turtle body mass. The nominal pulse width, pulse interval, and projected battery life of transmitters were 22 ms, 1935 ms, and 408 d, respectively. Transmitter range using a hand-held Yagi antenna often exceeded 1 km.
AUTOMATED TELEMETRY EQUIPMENT
To achieve complete coverage of the portion of the site most frequented by ornate box turtles (~45 ha), towers were spaced such that at least one tower would record strong signals over this portion of the study site. Five automated telemetry towers were positioned in two staggered parallel rows, such that towers were 390 m from adjacent towers in the same row and 300 m from nearest towers in the opposing row. Each tower consisted of an omni-directional whip antenna and six directional Yagi antennas mounted at 60[degrees] intervals atop a 9.2 m aluminum tower, an Automatic Receiving Unit (ARU; Sparrow Systems, Fisher, Illinois; Kays et al., 2011; Ward et al., in press), a 12 v battery, and a solar charging system. Only the omni-directional antenna of each tower was used to monitor turtle activity patterns. This antenna attached vertically to an aluminum rod that extended upward approximately 1.5 m from the top of each tower.
The ARUs monitored turtle radio transmitter signal strength, pulse width, and frequency-specific noise at 1 min intervals in 2010 and 2 min intervals in 2011. Sequential signal strength values were used to infer animal activity patterns because transmitter signal intensities at receiving towers typically change very little (<1 dBm) between sequential recordings when study subjects (and their transmitters) are stationary but can change greatly when individuals are moving (Kjos and Cochran, 1970; Kays et al., 2011). Importantly, even small changes in turtle orientation or antenna position against vegetation can elicit large changes in signal strengths at receiving towers.
VALIDATION OF AUTOMATED ACTIVITY RECORDINGS: 2010
To evaluate correspondence between transmitter signal strength recordings and turtle activity, we video recorded the activities of 27 free-ranging, radio-tagged turtles on 8 d between 30 May and 31 Jul., 2010. Each turtle was located during daylight hours using radio telemetry and video recorded for 15 min, except when turtles left the area being monitored or could not be observed because they became hidden behind vegetation or buried themselves before video monitoring was completed. Twelve of the 27 turtles were also video monitored on a subsequent day, resulting in a total of 39 recording periods. Turtles were found and monitored in both open and heavily vegetated areas of the study site, but monitoring was restricted to rain-free periods.
Activities of turtles between sequential signal strength recordings (collected at 1 min intervals) were categorized as: no movement = turtle in same orientation and location between automated telemetry recordings, small movement = turtle orientation changed less than 45[degrees] and location changed less than 0.5 body lengths, and large movement = turtle orientation changed more than 45[degrees] or location changed more than 0.5 body lengths. We determined the number of observations in each activity category (no movement, small movement, and large movement) that corresponded to changes in recorded signal strength of <3 dBm and [greater than or equal to] 3 dBm.
PRELIMINARY OBSERVATIONS OF NIGHT ACTIVITY AND NESTING: 2010
During the 2010 nesting season, transmitter signal strength recordings collected from 19 females indicated that individuals were rarely active at night. Importantly, however, when females exhibited substantial nocturnal activity, they generally did so only on one or a few closely spaced nights. These observations, which were consistent with accounts of females sometimes constructing preliminary cavities over multiple nights before oviposition (Legler, 1960), gave rise to the hypothesis that nocturnal activity inferred from the automated recordings of females corresponded to nesting. We subsequently hand tracked four night-active females and confirmed nesting activity in three of the four turtles. Observations of the fourth turtle were inconclusive, possibly because investigator activity near the turtle may have influenced its behavior.
DETECTING NESTING ACTIVITY USING NIGHT ACTIVITY RECORDINGS: 2011
From 15 May to 30 Jun. 2011, a single researcher used automated telemetry to monitor the nocturnal activity patterns of 18 females. To allow sufficient sampling from the period when night activity may occur, radio telemetry data from ARUs were downloaded no earlier than 90 min after legal sunset. Determinations of activity were made in the field by viewing the signal strength plots for each turtle (Fig. 1) on a laptop computer. All females exhibiting substantial activity after sunset were then tracked using conventional hand-held radio telemetry methods. In several instances, animals whose signals were ambiguous with respect to night activity (turtles with weak signals, which tend to fluctuate more, and turtles with infrequently changing signals) were also tracked. Upon locating a night-active female, her activity was classified (engaged in nesting activity or not engaged in nesting activity) and location was documented using a hand held GPS receiver and surveyor's flag. Potential nest sites were revisited the following morning to determine whether or not a nest had been completed.
To further investigate whether ARU indicated nocturnal activity was associated with nesting, we also tracked and visually determined the activities of randomly selected females whose signal recordings suggested inactivity. We refer to these females as "behavioral controls." A behavioral control was tracked every time potential nesting behavior was visually observed as well as in several additional instances. We did not use females as behavioral controls after they were known to have nested.
We present the number of turtle tracking nights (one tracking night represents one turtle located on one night) required to monitor nesting activity of our study population. We only tracked turtles when ARU observations suggested nocturnal activity, which we hypothesized would be primarily limited to nesting. We compared our tracking effort to the number of tracking nights that would be required if we had no knowledge of the nocturnal activity status of our study animals. Under such a scenario, we would have tracked all individuals from the start of the nesting season (15 May) until oviposition or, in the case of animals that did not nest (Doroff and Keith, 1990), until the expected end of the nesting season (30 Jun.).
We compared the frequency of nesting and non nesting during tracking of animals identified as active versus those identified as inactive, based upon field interpretations of radio transmitter signal strength records. Data were not independent because individuals that were treated as behavioral controls were frequently tracked later in the season in response to automated activity monitoring observations that suggested night activity. Therefore, we restricted our analysis to those turtles that both served as controls and later nested (n = 7). Because our dataset was limited and not normally distributed due to its binary nature (i.e., engaged in nesting activity versus not engaged in nesting activity), we performed a randomization test on a repeated-measures ANOVA in SAS version 9.1 (SAS Institute Inc., Cary, North Carolina).
Given that females may construct preliminary cavities over multiple nights (Legler, 1960), we expected greater activity by females on nights preceding oviposition than on nights following oviposition. To determine if activity indices generated by automated telemetry reflected such a pattern, we compared the number of nights with activity (which was defined as signal changes [greater than or equal to] 3 dBm among [greater than or equal to] 10% of sequential recordings) 30-90 min after legal sunset for 10 d preceding oviposition and 10 d following oviposition using females observed to successfully nest in 2011. Because the differences in the number of nights with activity between 10 d preceding and following oviposition were not normally distributed, we used a one-tailed Wilcoxon Signed-Rank Test using Minitab version 16 (Minitab Inc., State College, PA).
VALIDATION OF AUTOMATED ACTIVITY OBSERVATIONS: 2010
We determined that changes in signal strength [greater than or equal to] 3 dBm between sequential readings (which represented a doubling or halving of the signal strength) were indicative of box turtle activity (Table 1). Large movements frequently elicited signal strength changes [greater than or equal to] 3 dBm. Small movements resulted in intermediate signal changes. When turtles did not move, signals remained stable and never changed by more than 3 dBm between 1 min recordings (Table 1). Consequently, we accepted 3 dBm as a conservative threshold for inferring box turtle activity from signal strength observations.
DETECTING NESTING ACTIVITY USING NIGHT ACTIVITY RECORDINGS: 2011
Signal strength recordings suggested that individual females exhibited activity on 1-7 d preceding oviposition during earlyevening monitoring periods (30-90 min after sunset) during the 2011 nesting season (Fig. 2). Some of this activity was confined to the early portion of evening monitoring periods and likely corresponded to night form (i.e., shallow depression used for protection and thermoregulation) construction, rather than nesting activity. Nesting females (n = 12) were located by selectively tracking individuals whose automated telemetry recordings suggested evening activity that extended beyond what would be expected in association with night form construction.
Locating these 12 nests required 28 tracking nights, in large part because females often constructed preliminary cavities before ovipositing (seven tracking nights). On three tracking nights, females who were thought to be active based upon visual analysis of signal strength plots were found to be inactive upon locating them. In six additional instances, animals whose signals were ambiguous with respect to night activity (turtles with weak signals, which tend to fluctuate more, and turtles with infrequently changing signals) were also tracked. Had we located each female nightly during the nesting season (15 May until 30 Jun.), 641 turtle tracking nights would have been required to monitor this population. Thus, our effort in locating 12 nests across 28 turtle tracking nights represents over a 22 fold reduction in search effort as a result of this system.
Nocturnal activity, as indicated by the automated telemetry system, was associated with nesting activity (randomization test on a repeated-measures ANOVA; P < 0.01). We documented nesting behavior (including preliminary cavity construction on nights before oviposition occurred) in 19 of 28 tracking events that were initiated in response to fluctuating signal strength recordings that suggested night activity. In contrast we observed no instances of nesting activity in any of the 21 tracking events involving behavioral control females.
Six females (164.854, 164.564, 164.725, 165.259, 164.664, and 164.488) nested on the first night that they were located in response to automated telemetry observations indicating night activity. However, we also documented five females (164.816, 164.956, 164.583, 165.062, and 164.763) constructing preliminary cavities on one or more nights preceding oviposition. Four females constructed preliminary cavities on only a single night preceding oviposition, but one female (165.062) exhibited this behavior on three nights. Telemetry recordings of an additional female (164.164) indicated substantial activity on a night preceding oviposition, but severe weather precluded tracking. Preliminary cavity construction often occurred on the night immediately preceding oviposition but was also visually observed to occur up to 9 d before ovipositlon, often in different locations on subsequent nights. Automated telemetry observations reflected preliminary cavity construction on nights preceding oviposition. Signal change recordings suggested that, on average, females that oviposited were active on more early evening recording periods in the 10 d preceding oviposition than in the 10 d following oviposition (n = 12, Wilcoxon statistic = 28, P = 0.011; Fig. 2).
We determined that the signal change method provides accurate indices of ornate box turtle activity patterns at a 3 dBm activity threshold. We used this information to locate 12 nests by locating active animals across 28 turtle tracking nights. Although a nest was not produced each time an animal was identified as night-active and tracked, we found turtles nesting or constructing preliminary cavities in 19 of 28 tracking events. Furthermore, six of the tracking events that did not correspond to nesting behavior represented conservative efforts to track turtles, even when their activity signals did not clearly suggest night activity. In contrast animals whose radio signals suggested inactivity but were tracked as behavioral controls, were never found exhibiting any type of nesting behavior. Because of the effort that would have been required without this system (641 tracking nights), it is highly unlikely that a single researcher would have been able to find these 12 nests.
Manual palpation, followed by subsequent relocation of gravid females via radio telemetry, is an alternative means of locating nests in this species (Converse et al., 2002). We opted not to employ this technique however, in part due to concerns that it could disrupt natural activity patterns, which were the subject of a concurrent study. Additionally, manual palpation can cause ornate box turtles to void bladder water, which may function in osmoregulation and as a physiological reserve during dry periods (Peterson and Stone, 2000). Furthermore, palpation would have required repeatedly locating every individual, the very task which this method was intended to obviate, and would not have contributed to actually locating nests. Finally, ornate box turtles can carry shelled eggs for at least 8 d before ovipositing (Legler, 1960). Consequently, by only tracking gravid females when they are active at night, researchers can locate nests while avoiding considerable wasted effort.
Although it is possible that some females managed to nest undetected, we believe it is likely that we found all nests from our sample of 18 intensively monitored turtles in 2011. First, females in many populations of this species do not nest annually and our nesting rate (12 of 18) is consistent with that of Doroff and Keith (1990) who found 21 of 37 females to have nested over a 2 y period at a similar latitude. Second, none of the females tracked as behavioral controls (animals whose radio signals suggested inactivity) were found engaged in nesting activity. Third, a characteristic oscillating pattern of signal strength was often present in radio recordings of nesting females (Fig. 1) but was not observed in the recordings of females not known to nest. These patterns only appeared during evening hours on oviposition days or during a small number of evenings preceding oviposition when turtles may have constructed preliminary cavities before data were downloaded. We hypothesize that these oscillating patterns which were also observed in spring and fall, presumably when turtles dug into and out of their hibernacula, were associated with nesting activity. These patterns were not always apparent in females found to have nested but were never evident in females who were not found to have nested. Importantly, even if this method failed to detect some nesting events, it still produced a number of nests which could be expected from this population (Doroff and Keith, 1990) and greatly reduced search time and disturbance to study animals. Additionally, identifying nesting activity shortly after the behavior begins provides valuable opportunities to collect behavioral observations that span nearly the entire nesting process. This may not be possible for a single researcher monitoring a large population when manually locating every female is required to verify nesting behavior.
Because we wished to locate every nesting female in our transmittered population, we tracked animals in six instances when their radio signals were ambiguous with respect to night activity status. For example we tracked some animals whose telemetry signals included few signal strength changes above our 3 dBm threshold. Some of these animals were several hundred meters from the nearest tower, on the periphery of the distance from which stable signal strengths could reliably be recorded from a motionless animal. Others exhibited only minor activity, likely corresponding to night form construction. Night form construction can be apparent on signal strength plots; however, in contrast to nesting activity, does not continue late into the night.
To identify nesting activity in near real time, we downloaded towers during early evening hours starting 90 min after sunset. Downloading automated telemetry towers interrupts data collection. Therefore, the only uninterrupted set of signal strength plots (Fig. 1) from all towers is during the period before downloading began in the early evening. Nesting activity often continues well into the night, frequently not ending until morning; whereas, non nesting turtles rarely display any significant activity long after sunset (C. Tucker, pers. obs.; Fig. 1). Consequently, if activity patterns were analyzed for the entire night instead of only for the period of 30-90 min after sunset (Fig. 2), it is possible that signal strength plots would convey clearer distinctions between nesting activity and non nesting activity (i.e., night form construction). However, presentation of data from this time frame (Fig. 2) is more indicative of how this method was applied here, because time constraints dictated how late into the night data could be downloaded while still allowing ample time to locate animals before they vacated nest sites in the early morning. The early sampling period may also account for three animals that were tracked due to suspected activity but were found to be inactive upon locating them. These animals may have ceased activity during the time it took to locate them, and it is possible that downloading towers at a later time would have precluded tracking these animals if their activity subsequently subsided. Much of our tracking effort was related to locating behavioral controls, and if this is not required, other users of this method may choose to download data later in the night.
Gravid ornate box turtles may spend up to eight nights searching for suitable nest sites and may construct preliminary cavities multiple times before actually ovipositing (Legler, 1960). We confirmed that females who nested were more active during the ten nights preceding oviposition compared to ten nights following oviposition. This finding, along with our direct observations of preliminary cavity construction, indicates that much of the nocturnal activity detected outside the actual oviposition night was indeed related to nesting activity, rather than false positives or night form construction. Many factors may lead a female to abandon a nesting attempt, including poor nest site conditions, unfavorable weather, predator sign or presence, and researcher presence. Although we avoided approaching turtles early in the evening, we usually visited nesting females prior to midnight before oviposition was initiated. This could have resulted in some nest abandonment. However, motion sensitive cameras that were placed at some nest sites in 2011, as part of a concurrent study, showed no signs of researcher presence leading to nest abandonment.
One drawback of this method is the cost associated with necessary equipment (ARU, tower, antenna, coaxial cable, solar panel, etc.), which can approach $8000 for a single tower Unit, mainly because of the $5200 cost of the ARU. However, the five-tower array we used was more than adequate to cover our area of interest (~45 ha) and much of the data we present here were generated by a single centrally located tower. Our radio transmitters were designed with batteries to last for 1 y. However, stronger transmitter packages with shorter life spans would enable use of fewer towers during a 2 mon nesting study.
We used automated activity monitoring to efficiently identify and subsequently locate nesting ornate box turtles on a sand prairie in Illinois. Locating small cryptically-colored ornate box turtles (Refsnider et al., 2011) at night can take up to an hour (C. Tucker, pers. obs.). This task can be made even more difficult when nesting individuals are completely buried in the substrate (C. Tucker and T. Radzio, pers. obs.). Analyzing signal strength plots before manually tracking individuals can allow a single researcher to prioritize search efforts and, therefore, monitor a much larger population than would otherwise be possible. This method may be applicable to other species, especially when females must be located a priori, and followed to nests (Frazer et al., 1991), but its utility is restricted to species that tend to nest outside of their normal activity periods.
Acknowledgments.--Thanks to the Upper Mississippi River National Wildlife and Fish Refuge and all staff therein, including Eric Tomasovic for assistance with tower maintenance. Thanks to Dr. Fred Janzen's Iowa State University "Turtle Camp" crew for sharing the study site and assistance with capturing animals. Thanks to William Cochran, Tony Borries, and Andrew Walde who provided their expertise at various stages of this project. We are especially appreciative to Bill Cochran for opening up his home and laboratory on multiple occasions and sharing his enthusiasm for automated activity monitoring. We are also very grateful to Dr. Michael Ward for sending us several spare ARUs in summer 2010. This project would not have been possible without a generous equipment loan from the United States Army Construction Engineering Research Laboratory. All procedures for this research were approved by the Missouri State University Institutional Animal Care and Use Committee (protocol 120011), assuring compliance with animal care guidelines (Institute of Laboratory Animal Research, 1996). Research was conducted under Illinois Department of Natural Resources Permit 10-15S. The findings and conclusions in this article are those of the authors and do not necessarily represent the views of the U.S. Fish and Wildlife Service or the U.S. Army Construction Engineering Research Laboratory.
ABLES, E. D. 1969. Activity studies of red foxes in southern Wisconsin. J. of Wildl. Manage., 33:145-153.
BAKER, R. L. AND M. A. MCGUFFIN. 2007. Technique and observer presence affect reporting of behavior of damselfly larvae. J. N. Am. Benthol. Soc., 26:145-151.
BERINGER, J., J. T. MILLSPAUGH, J. SARTWELL, AND R. WOECK. 2004. Real-time video recording of food selection by captive white-tailed deer. Wildl. Soc. Bull., 32:648-654.
BERNSTEIN, N. P. AND R. W. BLACK. 2005. Thermal environment of overwintering ornate box turtles, Terrapene ornate ornata, in Iowa. Am. Midl. Nat., 153:370-377.
BOWEN, K. D., P. L. COLBERT, AND F. J. JANZEN. 2004. Survival and recruitment in a human-impacted population of ornate box turtles, Terrapene ornata, with recommendations for conservation and management. J. Herp., 38:562-568.
BRADLEY, R. W., F. COOKE, L. W. LOUGHEED, AND W. S. BOYD. 2004. Inferring breeding success through radio telemetry in the Marbled Murrelet. J. Wildl. Manage., 68:318-331.
BROEKHUIZEN, S., C. A. VAN'T HOFF, M. B. JANSEN, AND F. J. J. NIEWOLD. 1980. Application of radio tracking in wildlife research in the Netherlands, p. 65-84. In: C. J. Amlaner, Jr. and D. W. Macdonald (eds.). A handbook on biotelemetry and radio tracking. Pergamon Press, Oxford, U.K.
COCHRAN, W. W. AND R. D. LORD, JR. 1963. A radio tracking system for wild animals, J. Wildl. Manage., 27:9-24.
CONGDON, J. D., A. E. DUNHAM, AND R. C. VAN LOBEN SELS. 1993. Delayed sexual maturity and demographics of blanding's turtles (Emydoidea blandingii): implications for conservation and management of long-lived organisms. Cons. Biol., 7:826-833.
--, --, AND R. C. VAN LOBEN SELS. 1994. Demographics of common snapping turtles (Chelydra serpentina): implications for conservation and management of long-lived organisms. Amer. Zool., 34:397-408.
--, --, R. D. NAGLE, O. M. KINNEY, AND R. C. VAN LOBENSELS. 2001. Hypotheses of aging in a long-lived vertebrate, Blanding's turtle (Emydoidea blandingii). Exp. Gerentol., 36:813-827.
CONVERSE, S. J., J. B. IVERSON, AND J. A. SAVIDGE. 2002. Activity, reproduction and overwintering behavior of ornate box turtles (Terrapene ornata ornata) in the Nebraska Sandhills. Am. Midl. Nat., 148:416-422.
COOKE, S. J., S. G. HINCH, M. WIKELSKI, R. D. ANDREWS, L. J. KUCHEL, T. G. WOLCOTT, AND P. J. BULTER. 2004. Biotelemetry: a mechanistic approach to ecology. Trends in Ecology and Evolution, 19:333-343.
DOROFF, A. M. AND L. B. KEITH. 1990. Demography and ecology of an ornate box turtle (Terrapene ornata) population in south-central Wisconsin. Copeia, 1990:387-399.
EBINGER, J. E., L. R. PHILLIPPE, R. W. NYBOER, W. E. MCCLAIN, D. T. BUSEMEYER, K. R. ROBERTSON, AND G. A. LEVIN. 2006. Vegetation and flora of the sand deposits of the Mississippi River Valley in Northwestern Illinois. Il. Nat. Hist. Sur. Bull., 37:191-238.
FRAZER, N. B., J. W. GIBBONS, AND J. L. GREENE. 1991. Life history and demography of the common mud turtle Kinosternon subrubrum in South Carolina, USA. Ecology, 72:2218-2231.
HELLGREN, E. C., R. Z. KAZMAIER, D. C. RUTHVEN, III, AND D. R. SYNATZKE. 2000. Variation in Tortoise Life History: demography of Gopherus berlandieri. Ecology, 81:1297-1310.
INSTITUTE OF LABORATORY ANIMAL RESEARCH, COMMISSION ON THE LIFE SCIENCES, NATIONAL RESEARCH COUNCIL. 1996. Guide for the care and use of laboratory animals. Washington, DC: The National Academies Press., 124 p.
KAYS, R., S. TILAK, M. CROFOOT, T. FOUNTAIN, D. OBANDO, A. ORTEGA, F. KUEMMETH, J. MANDEL, G. SWENSON, T. LAMBERT, B. HIRSCH, AND M. WIKELSKI. 2011. Tracking animal location and activity with an automated radio telemetry system in a tropical rainforest. Comp. J., 54:1931-1948.
KJOS, C. G. AND W. W. COCHRAN. 1970. Activity of migrant thrushes as determined by radio telemetry. Wilson Bull., 82:225-226.
KLEIST, A. M., R. A. LANCIA, AND P. D. DOERR. 2007. Using video surveillance to estimate wildlife use of a highway underpass. J. Wildl. Manage., 71:2792-2800.
KUO, C. H. AND F. J. JANZEN. 2004. Genetic effects of a persistent bottleneck on a natural population of ornate box turtles (Terrapene ornata). Conserv. Genet., 5:425-437.
LANCIA, R. A., W. E. DODGE, AND J. S. LARSON. 1982. Winter activity patterns of two radio marked beaver colonies. J. Mammal., 63:598-606.
LEGLER, J. M. 1960. Natural history of the ornate box turtle, Terrapene ornata ornata Agassiz. Univ. Kansas Publ. Mus. Nat. Hist., 11:527-669.
MACNULTY, D. R., G. E. PLUMB, AND D. W. SMITH. 2008. Validation of a new video and telemetry system for remotely monitoring wildlife. J. Wildl. Manage., 72:1834-1844.
NAMS, V. O. 1989. A technique to determine the behavior of a radio tagged animal. Can. J. Zool., 67:254-258.
PETERSON, C. C. AND P. A. STONE. 2000. Physiological capacity for estivation in the Sonoran mud turtle, Kinosternon sonoriense. Copeia, 2000:684-700.
PLUMMER, M. V. 2003. Activity and thermal ecology of the box turtle, Terrapene ornata, at its southwestern range limit in Arizona. Chel. Conserv. Biol., 4:569-577.
REFSNIDER, J. M., T. S. MITCHELL, H. M. STREBY, J. T. STRICKLAND, D. A. WARNER, AND F. J. JANZEN. 2011. A generalized method to determine detectability of rare and cryptic species using the ornate box turtle as a model. Wildl. Soc. Bull., 35:93-100.
--, STRICKLAND, J., AND JANZEN, F.J. 2012. Home range and site fidelity of imperiled ornate box turtles (Terrapene ornata) in Northwestern Illinois. Chel. Cons. and Biol., 11:78-83.
RISENHOOVER, K. L. 1986. Winter activity patterns of moose in interior Alaska. J. Wildl. Manage., 50:727-734.
RUBY, D. E. AND H. A. NIBLICK. 1994. A behavioral inventory of the desert tortoise: development of an ethogram. Herpetol. Monogr., 8:88-104.
RUTZ, C. AND G. C. HAYS. 2009. New frontiers in biologging science. Biology Letters, 5:289-292.
SPERRY, J. H., M. P. WARD, AND P. J. WEATHERHEAD. 2013. Effects of temperature, moon phase, and prey on nocturnal activity of ratsnakes: an automated telemetry study. J. Herp., 47:105-111.
SUNQUIST, M. E. AND G. G. MONTGOMERY. 1973. Activity patterns and rates of movement of two-toed and three-toed sloths (Choloepus hoffmanni and Bradypus infuscatus). J. Mammal., 54:946-954.
VALENZUELA, N. AND F. JANZEN. 2001. Nest-site philopatry and the evolution of temperature dependent sex determination. Evol. Ecol. Res., 3:779-794.
WARD, M. P., J. H. SPERRY, AND P.J. WEATHERHEAD. In press. Using automated radiotelemetry to quantify activity and movement patterns of snakes. J. Herp.
WIKELSKI, M., E. M. TARLOW, A. RAIM, R. H. DIEHL, R. P. LARKIN, AND G. H. VISSER. 2003. Costs of migration of free-flying songbirds. Nature, 423:704.
SUBMITTED 20 DECEMBER 2012
ACCEPTED 6 JUNE 2013
CHARLES R. TUCKER (1), *
Department of Biology, Missouri State University, Springfield 65897
THOMAS A. RADZIO *
Department of Biodiversity, Earth, and Environmental Science, Drexel University, Philadelphia, Pennsylvania 19104
JERAMIE T. STRICKLAND AND ED BRITTON
Upper Mississippi River National Wildlife and Fish Refuge, Savanna, Illinois 61285
DAVID K. DELANEY
U.S. Army Construction Engineering Research Laboratory, P.O. Box 9005, Champaign, Illinois 61826
DAY B. LIGON
Department of Biology, Missouri State University, Springfield 65897
(1) Corresponding author: e-mail: email@example.com
* Authors contributed equally
TABLE 1.--Relationships between activity of video recorded turtles and changes in transmitter signal strength at telemetry towers. Large movements frequently elicited signal strength changes [greater than or equal to] 3 dBm. Small movements resulted in intermediate signal changes. When turtles did not move, signals remained stable. No Movement = turtle in same orientation and location between ARU recordings; Small Movement = turtle orientation changed less than 45[degrees] and location changed less than 0.5 body lengths; and Large Movement = turtle orientation changed more than 45[degrees] or location changed more than 0.5 body lengths No Small Large movement movement movement Number of Observations 310 71 50 [DELTA] Amplitude <3 dBm 310 (100%) 69 (97%) 17 (34%) [DELTA] Amplitude [greater than or equal to]3 dBm 0 (0%) 2 (3%) 33 (66%) Mean [DELTA] Amplitude (dBm) 0.13 0.78 6.11
|Printer friendly Cite/link Email Feedback|
|Author:||Tucker, Charles R.; Radzio, Thomas A.; Strickland, Jeramie T.; Britton, Ed; Delaney, David K.; Ligon|
|Publication:||The American Midland Naturalist|
|Date:||Jan 1, 2014|
|Previous Article:||Evaluating growth, survival and swimming performance to determine the feasibility of telemetry for age-0 pallid sturgeon (Scaphirhynchus albus).|
|Next Article:||Territory characteristics of Cassin's sparrows in Northwestern Oklahoma.|