An improved method of capture and immobilisation for medium to large-size macropods

Study species

The agile wallaby is sexually dimorphic, with males weighing 16–27 kg and females 9–15 kg. It is generally solitary but forms groups when feeding on pastures or open ground. Breeding occurs at any time of the year, and in agricultural environments, agile wallaby populations can double in size every 2–3 years (Stirrat 2008).

Study area

The study was conducted on the Douglas Daly Research Farm (DDRF; 13°50′S, 131°11′E) in the Northern Territory, Australia. The DDRF is a 3100-ha agricultural research facility managed and owned by the Northern Territory Government since the 1960s (Shotton 2011). Currently, the property has approximately 2500 ha of cleared land, mainly for improved pastures, wet season cattle fodder, and some grain crops, focusing primarily on research related to cattle production (Shotton 2011). Agile wallaby densities on the farm are high, i.e. 205 individual per km² (95% CI 170–247), especially on the improved pastures (Bedoya-Pérez et al. 2017). The region is characterised by rainfall seasonality: a wet season from October to April (Mean annual rainfall = 1207 mm) and a dry season from May to September, receiving little or no rain (Shotton 2011). Temperatures range from 13 to 37°C, and mean monthly maximum temperatures range from 31 to 37°C (Shotton 2011).

Capture and collaring

From 25 March 2016 to 30 May 2016, 40 agile wallabies were caught, and 34 were fitted with a very high frequency (VHF) collar (V5C 163E 130–260 mm VHF Std Collar Sirtrack Ltd). Across two areas of the DDRF, we established 17 capture locations, each of which was separated by at least 2.5 km. This gave us eight capture locations around the farm homestead and nine in grazing paddocks (Fig. 1). The longer-term effects of capture on animal condition were monitored daily at midday, dusk, midnight and dawn in radio-collared individuals during four 7-day tracking periods from 13 June 2016 to 21 July 2016.

Fig. 1.

A map of seventeen drop-net locations at the Douglas Daly Research Farm in two different areas: around the homestead on the west part of the farm and on grazing paddocks on the east side of it. Both areas are within 2 km of the Douglas River that fringes the north of the farm.

We used a ‘nylon drop-net’ technique originally used to capture ungulates (Kock et al. 1987; Mdetele et al. 2021). The technique was modified by wildlife management practitioners ABS Scrofa Pty Ltd, partners in this study. To our knowledge, only one study describes a nylon drop-net technique for capturing macropods (Lentle et al. 1997). Lentle et al. (1997) describe a method in which operators are required to lower the net manually to ‘close’ the trap, while our method relies on the animal hitting the net and triggering the fall of the net, as described by Mdetele et al. (2021). In our method, we created a ‘funnel’ to guide the animals towards the net, where only one animal per trapping attempt was captured by entanglement. Additionally, in contrast with Mdetele et al. (2021), we did not use vehicles to ‘herd’ the animals towards the net, but instead used people walking at a constant pace to ‘flush’ animals towards the net.

Further details of our method are as follows: The monofilament nylon drop-nets, approximately 20 m long, with 117 mm mesh size and 2 m high, were suspended between 2 m tall fence pickets driven into the ground on a slight angle (70°) in the direction the wallabies were expected to arrive from (Fig. 2a). The fence pickets were placed every 3–4 m across potential wallaby escape routes, e.g. fence lines and obviously used pads (Fig. 2b). Hessian fabric wings (2 m high) were sometimes added to the sides of some of the nets to funnel animals toward the drop-net (Fig. 2c). The top edge filament of the net was attached to the fence picket by light wire hooks. Before the start of a capture session, the bottom of the drop-net was aligned with the bottom of the fence pickets, folding the mesh to form a ‘pocket’ that opened towards where the wallabies were expected to enter the net. Once an animal hits the net, the impact releases the net from the light wire hooks, trapping the animal and allowing other wallabies to jump over the fallen net. In this way, only one wallaby is trapped at a time.

Fig. 2.

Drop net set up before the catching session. The net in (a), (b) and (d) is completely suspended to prevent animals from getting caught outside a catching session. A hessian fabric wing was used to funnel animals towards the net (c).

Netting was timed for the early morning and late afternoon when ambient temperatures were lower than 28°C to reduce the risk of hyperthermia. At least two people (handlers) were hidden behind the net during the catching session. Wet hessian bags were also hidden close by. Once the handlers were ready, at least four other people (walkers) walked toward the net in a line parallel to the net. Walkers started from up to 1 km away from the net, depending on where the wallabies were expected to be. During the approach, walkers created disturbance (noise) to ‘flush’ the wallabies towards the net.

Once caught, the handlers approached the animal quickly, placing a wet hessian bag on top of it to calm it down. The muscle relaxant and anxiolytic sedative Diazepam (Ilium Diazepam injection 5 mg/mL) was immediately administered to prevent capture myopathy. A first dose of Diazepam was administered at the point of capture based on visual estimation of body weight in a 5 kg body weight range (Table 1) at 1 mg/kg (Jackson 2007; Sadler 2010). The Diazepam was injected intramuscularly into the epaxial muscle (single-use sterile 22–23 gauge, ¾ inch needles, 3 mL syringe).

Following the administration of the first dose of Diazepam, the animal was removed from the net and placed in a light-proof hessian bag. It was then weighed using a hanging weighing scale, discarding the weight of the bag. If the actual weight was higher than the visual estimate, a second dose of Diazepam was administered. At this point, the wallaby also received a dose of Richtasol (1 mL/25 kg) (Richtasol AD₃EC injection Richter – Vitamin A, D₃, E and C) by intramuscular injection (single use of a sterile 22–23 gauge, ¾ inch needles, with 1 mL syringe) into the epaxial muscle as per Diazepam. Richtasol is a multivitamin used to prevent and treat capture myopathy (Pierson et al. 2023).

After weighing and measuring (see Supplementary material), each individual was fitted with a VHF collar with an expandable insert as a self-release mechanism (Fig. 3). Animals were tracked during June and July 2016. In addition, each individual was fitted with a custom-made, uniquely colour-coded ear tag. Ear-tagging individuals for subsequent monitoring prolonged the handling time and may have influenced the outcome of the trials. We provide a detailed description of the ear tags as they are of a novel design. Ear tags were constructed from a plastic transparent UV-resistant cable tie (Fig. 4a). A 3 cm section of clear heat shrink tubing (6 mm diameter) was passed over the identification cable tie and fixed by a heat gun to the base (i.e. at the side closest to the locking head) (Fig. 4b, c). This acted as a cushioned surface to prevent irritation of the pierced ear. One piece (1.5 cm) of coloured heat shrink tubing (1.6 mm diameter) was then fixed by a heat gun over the cable tie’s identification tab (Fig. 4d, e). The colours of heat shrink tubing were selected to make 15 unique combinations (Fig. 4f). Two tags of each colour combination were constructed to double the combinations to 30 by placing the tags in the right (female) or left (male) ear.

Fig. 3.

Fitting of VHF collar to captured agile wallaby. (a) The collar model used; (b) collar fitting process; (c) the elastic insert added to the collar for self-release; and (d) wallaby with the collar fitted.

Fig. 4.

Construction of ear tags for agile wallabies: (a) plastic transparent UV-resistant identification cable tie; (b) 3 cm section of clear heat shrink tubbing (6 mm diameter) next to cable tie; (c) cable tie with clear heat shrink section fixed by heat gun to the base; (d) piece (1.5 cm) of coloured heat shrink tubbing (1.6 mm diameter) next to cable tie; (e) coloured heat shrink tubbing fixed by heat gun, over each side of the cable tie’s identification tab; and (f) each colour combination constructed by pairs.

Statistical analyses

A ‘health index’ was calculated using a series of Generalised Linear Models (GLM). Each model combined a morphological measurement related to size, i.e. tibia length or pes length, and a morphological measurement related to fat reserves, i.e. weight or tail circumference. All models included sex as a fixed factor to account for sexual dimorphism. A log-transformation was applied to the models with ‘weight’ as the response variable due to non-normality and overdispersion of residuals. Adjusted R-squares for each model were used to select the model with the best fit, and the residuals were used as the ‘health index’ for each animal. The ‘health index’ values are centred at 0, i.e. the population average, with negative values indicative of animals with lower than average fat reserves adjusted by their size (lower condition) and positive values indicative of animals with higher than average fat reserves adjusted by their size (lower condition).

The factors influencing recovery times, i.e. time from release to flight, were explored by a separate Generalised Linear Model (GLM). The initial model included recovery time as the response variable and sex (male or female), age group (adult or subadult), health index, catch-to-bag time (the time it took from the moment the animal hit the net to when the animal was placed in the bag), and processing time (the time it took to measure, ear tag and collared the animal) as fixed factors. We compared the fit of models using corrected Akaike information criterion (AICc; ΔAIC) and excluded models with ΔAIC > 2 as having substantially less support (Burnham and Anderson 2010). The most parsimonious model of recovery times included only age group and processing time.

All analyses were performed using R v 4.4.1 (R Core Team 2024). The function glm from the package ‘MASS’ (Venables and Ripley 2002) was used to construct the GLMs. Assumptions were tested by performing: Shapiro-Wilk tests of normality, using the function shapiro.test from the package ‘stats’ (R Core Team 2024); Levene’s tests for homogeneity of variance (homoscedasticity) using the function leveneTest from the package ‘car’ (Fox and Weisberg 2018), and Pearsonʼs dispersion test by custom function. The function stepAIC from the package ‘MASS’ (Venables and Ripley 2002) was used for model refinement. P-values were generated by the type II Wald chi-square test using the function Anova from the package ‘car’ (Fox and Weisberg 2018).

Sex and age ratios

Of the 40 agile wallabies captured, 55% were adults, and 45% were subadults. Of the captured adults, 82% were female, and only 18% were male. Among captured subadults, 67% were male and 33% were female (Table 1).

Capture and recovery times

Agile wallabies received a Diazepam dose based on their expected weight class (1 mL for every 5 kg increment) (Table 2). Total handling time was kept as short as possible (16:24 ± 00:55 min; x¯±s.e.m., n = 40) (Table 3). Animals were released when they were fully recovered. Four animals delayed their flight after release – one adult male took 11 min from release to flight, and two adult females took 11 and 21 min, respectively. One female took 373 min to recover, although no ill effects were observed following release and during subsequent VHF tracking. Heart rate and body temperature were monitored on this female during recovery, and no obvious cause of the recovery delay was detected. Capture-to-bag time was under 5 min for all animals (Table 3). Although recovery times were most affected by age group (X12=3.335, P = 0.068) or processing time (X = 1.607, P = 0.213), these effects were not significant. The models suggest that adults tend to have longer recovery times.

Trapping and chemical restraint

There are many methods for reducing capture myopathy in macropod as well as ungulate (and other) species (Shepherd et al. 1988; Vogelnest and Woods 2008; McMahon et al. 2013; Blumstein et al. 2015; Braud et al. 2019; Breed et al. 2019; Latham et al. 2020). Our method adheres to ‘the three ‘Rs’’ of animal-based research – Replacement, Reduction and Refinement – required to obtain animal ethics clearance. In their strict meaning, ‘replacement’ refers to using another species or captive animals, ‘reduction’ refers to minimising the number of animals handled, and ‘refinement’ refers to building on previously successful methods. Here, we apply these principles, but not always strictly in their original sense.

We replaced the typically used cyclohexamine anaesthetic Ketamine with the benzodiazepine anxiolytic sedative Diazepam. Ketamine is a schedule [S8] controlled substance and not readily available or as safe as Diazepam (Tyner et al. 1997; Posner and Burns 2009; Ferreira et al. 2015). While Ketamine has a wide margin of safety in most species, it does not provide the muscle relaxation needed to prevent capture myopathy in macropods (Fish 2008).

Another alternative used with Macropods including agile wallabies (Stirrat 1997) is a combination of Tiletamine and Zalazepam (Roberts et al. 2010; Mayberry et al. 2014; Thompson et al. 2023). This combination is commonly known as Zoletil 100 or Telazol. Tiletamine is a cyclohexamine related to Ketamine, with dissociative effects, while Zalazepam is a benzodiazepine that can be used as a mild sedative or muscle relaxant, similar to Diazepam; however, it is slightly more potent (Popilskis et al. 2008). Due to its dissociative effects, Tiletamine–Zolazepam is commonly used in darts, where rapid and persistent recumbency is required (Stirrat 1997; Roberts et al. 2010; Thompson et al. 2023). This makes Tiletamine–Zolazepam inappropriate for studies like the one we describe, where quick processing and recovery are paramount.

The sole use of Diazepam as an anxiolytic and light sedative was highly successful. Diazepam provides marked sedation and muscle relaxation but no analgesic effect (Hillyer and Quesenberry 1997; Roberts et al. 2010; Thompson et al. 2023). It has been used to reduce the occurrence of capture myopathy in other macropods, e.g. eastern grey kangaroo (Roberts et al. 2010) and western grey kangaroo (Thompson et al. 2023). Used alone, Diazepam does not provide a true anaesthetic state, and awareness persists with relaxation even at high dosages. Hence, there is a need to use methods of restraint that maintain a quiet capture environment that keeps the animal calm. In our case, animals were placed in a light-proof bag and kept cool by wetting the bag. Midazolam is another benzodiazepine that could replace Diazepam as it is better suited to intramuscular applications (Hillyer and Quesenberry 1997).

While typically, ‘reduction’ in an animal ethics context refers explicitly to reducing the number of animals used, here we show how other relevant ‘reductions’ were applied in this study. We reduced the stress on the animal by keeping handling times short and administering a safe anxiolytic (Diazepam) and a preventative multivitamin (Richtasol). In addition, we conducted captures on cooler days (below 28°C) and in the evenings to reduce the risk of heat stress, which can exacerbate capture myopathy (Meyer et al. 2008a, 2008b; Meyer 2009; Breed et al. 2019; Ali et al. 2023).

Lastly, we refined the drop-net capture method (Lentle et al. 1997; Mdetele et al. 2021) to capture one wallaby at a time, keeping handling times short and allowing for its proper care and restraint with the minimum of stress. These simple modifications to the drop-net method successfully prevented mortalities, particularly deaths due to capture myopathy, and misadventures (Hampton and Arnemo 2023). Lentle et al. (1997) drop-net technique requires the net to be lowered by an operator, with several animals potentially being enclosed at once; this can increase processing times and stress for the animals. Additionally, by using ‘walkers’ instead of vehicles, as described by Mdetele et al. (2021), we ensure the pace at which the animals are funnelled towards the net is steady and produces minimal flight response stress.

We further refined the management of potential capture myopathy by administering the multivitamin Richtasol (Pierson et al. 2023). We injected all study animals with a combination of vitamins A, D₃ E and C (Richtasol® AD ₃EC). The use of vitamin E and selenium combination or an aggregate supplementation is widely recommended for the prevention of capture myopathy in macropods and other species undergoing capture and release (Lapidge 2005; McMahon and Bradshaw 2008; McMahon et al. 2013; Tarszisz et al. 2014).

It is important to acknowledge that the success of the capturing method we describe here, specifically the use of Diazepam and vitamin injections, has limited comparison potential with other methods that do not use these pharmacological interventions. In this study, we did not have control animals that did not receive Diazepam or vitamin injections. However, it is also important to recognise that this kind of study is challenging to perform when we must balance knowledge of current practices while trying to test if new approaches are beneficial. In the light of this, we consider our method to be still a valuable addition to the available repertoire of techniques.