The global trend of urbanization is creating novel challenges to many animal species. Studies investigating behavioral differences between rural and urban populations often report a general increase in risk-taking behaviors in urban populations. According to the most common energy management model (the performance model), behaviors that increase access to resources, such as aggression and boldness, and behaviors that consume net energy, like locomotion and stress responses, are both positively correlated to resting metabolic rate (RMR). Thus, we expect urban populations to not only exhibit a higher level of risk-taking behavior but also a higher RMR. However, these interactions remain poorly investigated. Our main goal was to analyze the relationship between RMR and risk-taking behaviors in the greater white-toothed shrew (Crocidura russula) in rural vs. urban populations. Trapped shrews were brought to captivity where we measured RMR, boldness and exploration rate three times in each individual. Our findings revealed urban shrews were indeed bolder and more exploratory, but contrary to our expectations, their RMR was lower than that of rural shrews. This is likely explained by differences in the environmental conditions of these two habitats, such as higher ambient temperatures and/or lower prey availability in cities. When looking at each population separately, this relationship remained similar: urban shrews with a higher RMR were less bold, and rural shrews with a higher RMR showed a lower exploration rate. We conclude the energetic strategy of C. russula is dependent on the environmental and observational context and cannot be explained by the performance model.

Capture sites and shrews

All shrews were captured between July and October of 2018 with wooden box live traps (16.5×8.0×9.5 cm; PPUH A. Marcinkiewicz, Rajgród, Poland). Traps were set before sunset and checked every three hours until around midnight to prevent casualties and unnecessary stress. Shrews were sexed visually based on their external anatomical features, a reliable method in shrews of the genus Crocidura. Shrews were captured in two different habitats at separate locations. A total of 17 individuals (13 males and 4 females) referred to as “rural shrews” were captured in three sites in Sintra-Cascais Natural Park, central-western Portugal. Trapping sites were located in the park’s mountain range, where human disturbance is minimal. This area is characterized by a wet climate, and typical forest habitats with Mediterranean shrubby-type vegetation and Quercus spp.. Another 17 individuals (12 males and 5 females) referred to as “urban shrews” were captured at four sites in the city of Lisbon. Trapping sites included urban plantations, parks, and vacant lots near buildings and roads.

Trapped C. russula were transported to the animal facility of the Faculty of Sciences, University of Lisbon, while pregnant females and other trapped species were released on site. No more than three shrews were collected each night as this was the maximum number of individuals that could be tested in a single day. Since each individual was experimented upon every two days (see below), trapping was done for two nights every week, resulting in six individuals being tested during each experimental week.

Shrews were housed in white plastic terraria (38×28×17 cm) topped with a metal grid. Terraria were enriched with a mixture of sand and soil as substrate, half of an eggbox and paper sheets as shelter, and some natural vegetation collected at the trapping site. The temperature of the holding room was maintained at 22–24 ºC with a controlled photoperiod with lights on from 0500 to 1700 hours. Shrews were fed upon arrival, and then every night (after experiments) with 1.0 g of mealworms, 1.5 g of blowfly pupae and 3.0 g of minced cow beef. Water was changed daily at the time of feeding.

Metabolic and behavioral tests

We performed three replicates of each metabolic and behavioral test with each individual: a first trial on day one, a second trial on day three and a third trial on day five after capture.

Metabolic measurements

The resting metabolic rate (RMR) of shrews was measured during the day (between 09:00 and 17:00 hours), when they are less active (Genoud and Vogel 1981; Oliveira et al. 2016). For each individual, oxygen consumption was measured in an open flow respirometry system using an oxygen analyzer (Servomex, series 4100). Shrews were taken to the metabolism experimental room where they were placed inside a small acrylic cylindrical chamber (length: 25 cm; diameter: 7 cm). The chamber was placed horizontally inside an incubator (Sanyo 089A) at 30 ºC, a temperature within the species’ thermoneutral zone (Sparti 1990). Atmospheric air was pumped through the chamber with a flow rate of 500 ml min^-1. Before and after each trial, baseline values of atmospheric oxygen were obtained by performing a measurement without an animal inside the chamber that lasted until readings stabilized. Flow rate was controlled and measured continuously at the chamber inlet by a calibrated mass flow controller (Sierra Instruments 840L connected to a Sierra Instruments Cal=Bench for readouts). Air was dried using silica gel columns before entering the chamber and before entering the oxygen analyzer. To minimize error in the conversion of oxygen consumption to energy expenditure when the respiratory quotient (RQ) is unknown, carbon dioxide was not removed from air entering the metabolic chamber (Koteja 1996). Readings from the oxygen analyzer were digitized approximately 35 times per second. The accumulated data was then averaged over a period of 15 seconds, meaning each 15 second measurement was based on a total of about 500 actual readings. RMR was calculated as the average of the ten lowest and consecutive values registered, corresponding to 2.5 minutes of stable sections of each respirometry measurement period. Shrews were checked every few minutes through a plexiglass window until they had settled down inside the chamber, which took approximately 1.5 hours. Oxygen consumption (VO₂) was obtained following Depocas and Hart (1957): VO₂ = V₂(F₁O₂ - F₂O₂) / (1 - F₁O₂), where V₂ is the flow rate measured at the outflow of the chamber, and F₁O₂ and F₂O₂ are the oxygen concentrations measured at the inflow and outflow of the chamber, respectively. All VO₂ measurements were corrected to standard temperature and pressure and converted to energy units assuming 1 ml O₂ = 20.1 J and 1 J min^-1 = 16.667 mW.

Body mass is one of the most important variables explaining RMR variance (Speakman 2013; McNab 2015); thus, shrews’ body mass (g) was recorded before and after entering the metabolic chamber using a pocket balance (PESOLA PPS200, accuracy of 0.01 g), and the average of the two values was used in statistical analysis. After metabolic measurements, shrews were placed back into their home cages and returned to the holding room where they stayed for at least one hour before behavioral tests.

Boldness test

After metabolic runs were over and soon after lights went out (at 1700 hours), individuals were separately tested in a boldness test. Each shrew was taken out of its home cage, placed inside a closed wooden box live trap (inner size 15.0 5.0×6.5 cm), and carried to a contiguous behavior experimental room to be tested. This room was illuminated by dim red light and its temperature was similar to the holding room. The box trap was placed into an empty clean transparent box (44×30×32 cm) and left untouched for five minutes to let the shrews habituate to the trap. Then, the door of the box trap was opened, and the shrew was allowed to leave the box at its own pace. The experiment ended either when the shrew had left the trap completely (all four paws touching the floor outside the trap) or after 90 minutes had passed since the trap was opened. The experimenter stayed inside the room during the first 15 minutes after opening the trap. The experiment was video-recorded from above with a Sony HDR-PJ780 camera with night vision. From the video, the shrew’s latency to fully exit the trap was measured.

We chose the latency to emerge from a shelter as a measure of boldness, as it reflects the natural behavior of small mammals emerging from a safe place to begin foraging or exploring. It is a common metric for boldness in small mammals, with those individuals leaving the shelter rapidly considered bolder than individuals that hesitate for long periods of time (Frynta et al. 2018; von Merten et al. 2020).

Exploration test

Exploration tests were carried out immediately after the boldness tests. Each shrew, by means of a small jar, was gently placed in a random position in an open-field arena (50×50×30 cm). The floor of the arena was divided into four equal squares by two perpendicular PVC barriers. The barriers were 1.5 cm high and 1.5 cm wide, small enough to allow the shrews to easily cross it, but high enough so they would perceive it as an obstacle. The test lasted 10 minutes and was filmed from above using the same camera as in the boldness test. After completion, shrews were returned to their home cage. Exploration rate was quantified following video analysis according to the number of squares visited, including the starting square, with repeated visits to the same square being counted.

We chose the number of crossings of barriers inside an experimental arena to measure exploration rate, as it reflects the challenges of small mammals when exploring unknown territory. It is a common metric for exploration rate in small mammals, with those individuals crossing barriers more often and entering a higher number of different sections in the experimental arena assumed to be more superficial explorers (Gifford et al. 2014; Mazue et al. 2015; Frynta et al. 2018).

To run the R Markdown file, unzip the folder "Oliveira_2020_BehavEcol_dataset_analysis.zip" into any location.
When running RStudio, set that location as your working directory.
Make sure you have all the required libraries installed to run the script.