1. Individual variation in movement is profoundly important for fitness and offers key insights into the spatial and temporal dynamics of populations and communities. Nonetheless, individual variation in fine-scale movement behaviours are rarely examined even though animal tracking devices offer the long-term, high-resolution, repeatable data in natural conditions that are ideal for studying this variation. Furthermore, of the few studies that consider individual variation in movement, even fewer also consider the internal traits and environmental factors that drive movement behaviour which are necessary for contextualising individual differences in movement patterns.
2. In this study, we GPS-tracked a free-ranging population of sleepy lizards, Tiliqua rugosa, each Austral spring over five years to examine consistent among-individual variation in movement patterns, as well as how these differences were mediated by key internal and ecological factors.
3. We found that individuals consistently differed in a suite of weekly movement traits, and that these traits strongly covaried among-individuals, forming movement syndromes. Lizards fell on a primary movement continuum, from ‘residents’ that spent extended periods of time residing within smaller core areas of their home range, to ‘explorers’ that moved greater distances and explored vaster areas of the environment.
4. Importantly, we also found that these consistent differences in lizard movement were related to two ecologically important animal personality traits (boldness and aggression), their sex, key features of the environment (including food availability, and a key water resource), habitat type, and seasonal variation (cool/moist vs hot/drier) in environmental conditions.
5. Broadly, these movement specialisations likely reflect variation in life-history tactics including foraging and mating tactics that ultimately underlie key differences in space use. Such information can be used to connect phenotypic population structure to key ecological and evolutionary processes, for example social networks and disease-transmission pathways, further highlighting the value of examining individual variation in movement behaviour.

The following description is from Michelangeli et al. (2021) but modified for brevity. Note that citations have been removed, but can be found in Michelangeli et al. (2021)

Study system

The sleepy lizard is a common skink species to South Australia. They are large-bodied (adults are 400 – 950g, snout-vent-length 25 – 35 cm), long-lived (up to ~48 years) and slow-moving. They have been found to maintain relatively stable and consistent home ranges (HR) over multiple years (our tracking covering up to 8 years; mean 95% HR area in hectres [95% CI]; females: 3.53 [3.07, 4.01] and males: 4.91 [4.42, 5.43]). We studied a population within a ~1.2km² field site near Bundey Bore Station (33.888240 S, 139.310718 E). Our field site has semi-arid Mediterranean climate and is primarily chenopod shrubland dominated by bluebush (Maireana sedifolia) and some black oak (Casuarina cristata). Various annual plants, including the lizards’preferred food item, the invasive Ward’s weed, Carrichtera annua , tend to grow around and under these shrubs. The field site has a seasonal dam that retains water and soil moisture for longer than other parts of the area. Lizards at this field site are primarily active from September (early spring) to December (early summer). The beginning of the lizard’s activity period is characterised by moderate temperatures and higher relative abundance of food and water. Early spring is also the breeding season in which lizards form monogamous pairs for up to 10 weeks before they mate. During this time males perform more scent-marking and competition is higher, potentially leading to sex differences in movement behaviour early in the season. On the other hand, the late season is hotter and drier, and food and water resources become increasingly scarce and patchier in distribution.  

GPS data collection

Sleepy lizards were GPS tracked during their activity season between 2013-2017. The GPS data was filtered to remove all errors and points that were outside a lizards primary activity period (between 06:00 and 20:00). The data was also thinned to follow a 20 min resolution. Using the wrangled GPS data, we calculated six movement metrics - daily distance travelled, max daily displacement, straightness index, residence time, home range size and home range use ratio. These metrics are commonly used to represent a range of possible movement and spatial patterns that we predicted might vary among lizards. To enable comparison between all metrics, we averaged the daily scale metrics within weeks. To eliminate correlation between the number of weekly GPS points and corresponding movement metric value, we only included weeks in our analysis with at least 120 GPS locations (which represents ~ 3.5 days of GPS data). we obtained GPS data from 128 lizards and a total of 569341 GPS fixes (after sub-sampling that data to 20 min intervals). From these data, we obtained on average 17 ± 11.6 (mean ± SD; range 2 – 45) weekly replicates for each movement metric over an average of 2.35 ± 1.4 (range 1-5) years per lizard. Each weekly replicate contained on average 189.64 ± 30.00 GPS points

Measurement of lizard internal traits

In addition to the movement data, we also collected data on each lizards size every 2 weeks. Between 2015-2017 seasons, lizard aggression and boldness behaviour was also assayed 2-3 times per year, with each trials occuring at least 2 weeks apart. . To assess a lizard’s boldness, we measured a lizard’s response to a preferred food item (a piece of banana) in the presence of a potential threat (a looming observer). A lizard’s boldness was ranked based on their response, from quickly approaching the food item and the observer (high boldness rank), to showing very little response (intermediate boldness), to quickly fleeing from the observer into refuge (shy). Boldness was ranked on a 1-10 scale in 2015 and on a 1-7 scale in 2016-2017, but higher ranks always referred to bolder animals and thus ranks were standardized within year (by subtracting the mean and dividing by the standard deviation). To assess a lizard’s aggression, we ranked a lizard’s response to a simulated capture attempt by an observer which involved the observer’s hand coming to within ~10 cm of the lizard’s head. Responses were ranked on a scale of 1-11, from fleeing the observer into refuge (passive = 1) to attempting to bite the observer (high aggression = 11). Both aggression and boldness have shown to be repeatable over an 8-year period (r ± SE = 0.43 ± 0.05 and r = 0.27 ± 0.04, respectively), and are only weakly positively correlated suggesting independence between the traits (correlation coefficient ± 95% CI = 0.15 [-0.07, 0.39]). In the present study, we used the average personality score for each individual within each year in our analysis

Measurment of environmental variables

From the movement trajectories within each year, we calculated an individual’s mean weekly distance from the dam (DamDist) and the number of unique visits to the dam (i.e. DamVisits). A unique visit to the dam involved a lizard entering and then leaving a 15m radius around the dam. The dam is the only reliable water resource at our field site, and lizards are often observed in this area, but we also know that approximately only about 50% of our study population access the dam within a given year. We also calculated a lizard 95% weekly HR overlap with ‘forested’ habitat (ForestHR). Forested areas tend to contain denser vegetation and refuge for lizards and may be preferred by lizards seeking shelter. Forested habitats were defined by drawing polygons around forested regions based on satellite images from Google Earth. During 2015 -2017, we also conducted habitat surveys of food quality in 123 quadrats distributed in a grid over the field site. Within each quadrat, we surveyed a 40m radius around a central point, and ranked the availability of moist food (i.e. primarily Ward’s weed) on a 1-5 scale. Since lizard HRs could include multiple quadrats, we averaged the quadrat food scores within a lizard’s 95% HR, weighted by the usage of that quadrat (relative to usage of all quadrats within the focal lizard’s 95% HR), within each week, (FoodQualityHR).