Scats can reveal the presence and habitat use of cryptic rock-dwelling macropods

1. What they did

The rock-dwelling wallabies and wallaroos of tropical northern Australia live in steep, rugged rock country, are wary, and flee to shelter when disturbed, so very little is known about them. Their scats (dung pellets) are much easier to find than the animals. The authors asked whether scats could be used as a survey tool, and tested three things that surveys based on scats take for granted:

• Can the scats be told apart? They measured 35 to 500 scats from each of seven macropod (kangaroo family) species and the rock ringtail possum, collected in places where the source species was known from direct observation, earlier studies or local knowledge. For each scat they recorded length, width and breadth (to 0.1 mm), colour, shape, and whether it had “squeeze points” (thin tapering ends). They built an identification key with a classification tree, first for all species and then for the four rock-dwelling macropods only.
• How long do scats last, and how do they change? At Litchfield National Park they placed 15 fresh short-eared rock-wallaby scats in each 1 m square plot, on bare rock and on leaf litter over sand. Every two months they counted the scats left and scored their condition in decay categories, from shiny black to grey and crumbly, plus insect damage. Experiment A started in September 2002 (before the wet season) with four plots on each substrate at four sites up to 15 km apart; Experiment B started in the dry season of 2003 with eight plots on each substrate at one site. Both ran until August 2004.
• Do scats show where the animals really go? On a 100 ha site where eight short-eared rock-wallabies had been radio-tracked, they counted scats in a 1 m square plot every 100 m and recorded the habitat, rock cover, grass and whether the area had burnt. They compared the habitat ranking from scat density with the ranking from the radio-tracking.

2. What they found

The identification key

• Scats varied a lot in size and shape, both within and between individuals of each species, but there were enough differences to build a key.
• Width was the first and most important split. Rock-wallaby scats were small (generally under 14 mm wide) and black to brown. The rock ringtail possum’s scats, found in the same rocks, were usually more rufous and finer in texture; all macropod scats were fibrous.
• Nabarlek scats were separated from short-eared rock-wallaby scats by their shorter length (generally under 12 mm). Agile wallaby scats were narrower than those of the other larger macropods, common wallaroo scats wider. Black wallaroo scats tended to be heart-shaped and antilopine wallaroo scats rectangular. There were too few northern nailtail wallaby scats to separate them in the model.
• With all species included, 31.9% of single scats were misclassified, mainly because young short-eared rock-wallaby scats looked like nabarlek scats and the larger macropods overlapped. With only the four rock-dwelling species, misclassification fell to 14.8%. Because this rate applies to one scat at a time, measuring many scats from a site gives a more reliable answer.
• “Squeeze points” were common on macropod scats (42 to 58%) but did not help separate species.

How long scats last

• More than half the scats disappeared in the first six months (54.4% in Experiment A and 59.2% in Experiment B), yet some lasted over two years.
• The biggest losses came in the wet season, and dry-season fires also destroyed many scats.
• Scats were lost faster from litter over sand than from rock.
• Fresh scats stayed black or brown and glossy for up to four months. Old, grey, fraying scats could be anywhere from 2 to more than 24 months old. After two years nearly all were grey and crumbly.
• Scats decayed more slowly in the dry season, and aged faster on rock than on sand with litter.

Scats versus radio-tracking

• Habitat was the best model of scat density (an 84% probability of being the best model). Burning and rockiness added little.
• Scat density ranked the habitats in the same order as the radio-tracking: outcrop, then flat next to outcrop, hill slope, swale and hilltop. However, the scats suggested outcrops were used far more than everything else, while the radio-tracking suggested the nearby flats were used about as much as the outcrops.
• Many radio-tracking locations had no scats (21.1%) or few scats (32.9%). But wherever an animal was located by day (resting and den sites), scats were always present.
• Radio-tracked wallabies moved up to 480 m from their shelters, but scats were found only up to 200 m from possible rocky shelters.

3. What they concluded

• Scats are an efficient, cost-effective way to study elusive species such as the rock-dwelling macropods of the monsoon tropics.
• Short-eared rock-wallaby scats are the most distinctive of the seven species, so they can be used to confirm the species’ presence and map its distribution. Surveys should target rocky habitats.
• For reliable identification, measure many scats and know the habitat they came from. Because young animals’ scats overlap with adults of smaller related species, the largest scats found are the most useful.
• Nabarlek scats are the least suitable for surveys, as they look like those of young short-eared rock-wallabies; DNA from scats would be the best way to confirm nabarlek presence.
• For the shy, mostly nocturnal black and common wallaroos, scats give a far better chance of detection than watching or trapping.
• Scats show broad habitat use and core areas, but not fine differences between habitats or the total area used. The authors think the lack of scats beyond 200 m reflects animals spreading out as they forage away from shelter, not an absence of dung-dropping there. The higher scat density on outcrops may come from scats lasting longer on rock or from animals defecating more after resting.
• Habitat use surveys should use only new, dark, glossy scats and be done in the dry season (early, before burning, or late, after it). For monitoring whether sites are still occupied, set up permanent plots in rocky areas with existing scats; fresh glossy scats signal recent use, while weathered grey scats are of unknown age.

4. How tracking was or was not used

The study is about one kind of sign: scats. The authors identified scats to species, aged them by their condition, and used scat density across a grid to infer habitat use. No footprints or trailing were used. They note that scats and other spoor have long been used by indigenous peoples and hunters to find animals. Scats for the identification key came from places where the species was known from direct observation, earlier studies or local knowledge, and the authors acknowledge some circularity in assuming wild scats came from a particular species.

5. Tracker or observer certification

Not reported. The paper does not say who identified, collected or scored the scats, or test their skill. The key itself was built from measured scats, and the paper cites an earlier Tasmanian study in which even experienced observers sometimes misidentified wallaby and pademelon scats.

6. Other technologies, skills or innovations

• A measured identification key built with classification tree modelling in R, using calipers to 0.1 mm.
• A simple scale for scoring scat condition: six stages from glossy black with a mucous coating to grey and crumbly, and a seventh category for insect damage.
• Radio-telemetry, and a Global Positioning System (GPS) to locate grid points.
• Spatial models that correct for spatial autocorrelation, and scat density maps made in a Geographic Information System (ArcMap).
• Comparison of captive and wild scats: captive agile wallabies on vegetable diets produced scats quite different in size and shape from wild ones.

7. Implications for conservation (from the paper)

• Scats give wildlife managers a simple, cost-effective way to monitor rock-dwelling macropods that are hard to see or trap.
• Monitoring plots should be in rocky core areas with existing scats. Monitoring in the wet season, when scats disappear fastest, gives the most precise picture of recent use, but the dry season is the most successful time because of a lower chance of “false absences”.
• Future work should measure the chance of finding scats in each habitat, so scat densities can be corrected before they are used to judge fine-scale habitat use.