Sunday 8 January 2023

Dunnarts in the Sandhills

It's a recurring theme of this blog, especially notable in last year's series on Old World leaf monkeys, that a great many mammalian species are threatened in some way. Population numbers of many species are declining, to the point that their continued existence is in doubt, often due to human encroachment on their habitat, but also due to the habitat itself changing as climate change continues. In this context, the quest of zoologists to understand how mammals (and other animals) behave is not just one of intellectual curiosity but can be of direct benefit to the creatures themselves.

For example, it is useful to conservation efforts to understand not only where a given species lives, and the habitat requirements it may have, but how it makes use of that environment. (Obviously, there's more to it than this, for example, how different species in the same area interact with one another but we'll stick with this one point for today). What particular features of the habitat are important to it? How much land does it need? How is its population distributed across the area? 

For larger and/or charismatic mammals, we usually have a good handle on this. Which isn't to say that more detailed information wouldn't be helpful, but at least we know where to start. For many smaller and lesser-known creatures, however, we may have far less to go on, lacking even some of the basics. How we gather this information has naturally changed over the decades, but let's take a look at a study published last year into one relatively obscure species to see how it made use of the habitat it lives in.

The mammal in question is the sandhill dunnart (Sminthopsis psammophila). You might think, from the picture at the start of this post, that it's some kind of desert mouse, but it isn't. In fact, it's an Australian marsupial, one of many small, superfically shrew-like, carnivorous marsupials that inhabit the continent. While it isn't true to say that they've been entirely ignored, these are among the least studied of all marsupial groups and the sandhill dunnart is no exception.

The sandhill dunnart was first described in 1895, from a specimen collected deep in the Australian interior. It's fair to say that it didn't attract much attention and that's not least because, after that, nobody spotted another one... it just seemed to have vanished and, for all anyone knew, might have gone extinct while we weren't looking. Until that is, 1969 when a second specimen was found living near the south coast, over seven decades later and 750 miles (1,200 km) away from the first one.

Once we'd confirmed that, yes, it was still around, we started finding more, but still not terribly many. It is currently known from just three locations - none of them particularly close to where that first specimen was collected back in the late 19th century. These include the east coast of the Eyre Peninsula and the edge of the Nullabor Plain, both in South Australia, and the southwestern corner of the Great Victoria Desert, in Western Australia. 

The size of these areas is so small that, by virtue of their size alone, this would normally be enough to qualify a formal listing as an "endangered species" under the rules of the International Union of the Conservation of Nature. However, the fact remains that this entire region of the continent is remote, largely unsurveyed, and sparsely inhabited at best. Plus, quite honestly, they're small nocturnal animals that nobody has particularly been looking for. So, when it last evaluated the species, in 2016, the IUCN decided that the animals almost certainly live elsewhere as well but that nobody has actually spotted them yet. Thus, with a total estimated population of no more than 10,000, they are officially a "threatened" or "vulnerable" species, but they are not an endangered one.

Just from looking to see what these areas have in common, we already know the basic habitat requirements of the species. The areas are all semi-desert, the sort of climate that we'd also find in, say, much of central and southern Spain (especially away from the coast) or lowland New Mexico. The primary vegetation here is "spinifex" tussock grass, but with patches of acacia thorn scrub and gum trees. What was less clear was how exactly the dunnarts used this habitat - how they moved about, what they were eating, and so on.

To find out, the researchers first had to trap some sandhill dunnarts. To do this they used pitfall traps, which as the name suggests are essentially canisters buried so that their rim is at ground level - in this case, they were 65 cm (25 inches) deep. They then fitted the captured animals with GPS devices that would log their location every five to ten minutes and fall off naturally after a week or so. The devices have a radio transmitter attached that enables them to be found afterwards, but not one large enough to transmit the detailed data to a ground station, so they do have to be picked up in order to be examined. This keeps the total weight of the device down to less than 2 g (0.07 oz.), which is obviously a lot more comfortable for the 40 g (1.4 oz.) dunnart than a more conventional one would be; the same model is often used in studies of birds and doesn't get in the way of them flying.

Having done that, the next thing to do is examine the area closely to evaluate the different kinds of micro-habitat within it, and which ones the dunnarts had been visiting - for example, the areas where gum trees were found, or where the tussock grass was more common. Then, since dunnarts are carnivorous and unlikely to be eating the grass itself, it's a matter of painstakingly finding as many of their faecal pellets as you can and examining them under a microscope to see precisely what they've been eating.

Looking at how the animals move across their landscape allows one to determine their "home range". This is the specific area within which an individual moves about, which in turn shows you how much land they need in order to be comfortable. Previous estimates for the sandhill dunnart showed that this was probably rather small, but the new study gave a figure more in line with what we'd expect for an animal of its size living in a semi-arid environment where food won't always be easy to come by. According to this new study, however, a sandhill dunnart requires, on average, 70 hectares (175 acres) of land - defined as the area within which, left to its own devices, it spends 90% of its time. 

But the average figure does conceal some variation. In particular, males occupy larger areas than females. This is likely because the males want to encounter as many females as they possibly can, while the females are less concerned about doing the reverse (since they're not sexually promiscuous) and, indeed, may be somewhat territorial, pushing rival females out. The movement records also showed that, as expected, sandhill dunnarts are nocturnal, although they did occasionally venture out in the daytime when the weather was particularly cool.

As both the common and scientific names suggest (psammophilus translates as "sand-loving") sandhill dunnarts are thought to live in areas with numerous sandy dunes. Breaking down the semi-desert habitat of the study site into smaller sub-environments showed that they did, indeed, have a preference for sand dunes, with flatter stretches being less popular, followed by stands of eucalypt gum trees and then patches of acacia. This is probably because of the relatively dense low-lying vegetation here, with the tussock grass that covers the dunes not only providing them cover from potential predators, but also likely giving them a richer selection of their own prey.

Another feature that they noted is that dunnarts avoided land that had experienced a fire within the last three years altogether. This is significant, as it shows how bushfires - becoming increasingly common in Australia in recent years - have an effect on the animals beyond the merely direct one, destroying the habitat they need to thrive. They probably avoid such areas because their prey avoids them as well, leaving them with little to eat, so we'd expect this to be an ecology-wide effect but it emphasises what we already knew about the importance of using fire management to preserve the habitat.

As for what was learned from examining those droppings, it turns out that sandhill dunnarts really like eating ants, with sugar and rainbow ants forming a large part of their diet. This is somewhat unexpected, as ants are not especially easy or nutritious to eat. For example, the little long-tailed dunnart (S. dolichura) lives in the same area but, despite being less than half the size of the sandhill species, feeds on prey that are much larger than ants - moths, dragonflies and even small lizards. This is likely to indicate a specialised adaptation to this particular prey (ants are apparently unusually common in the Great Victoria Desert) and give us some further hints as to what exactly it is that these small marsupials are looking for.

That such requirements are comparatively unusual shows the importance of preserving a wide variety of different habitats where we can, even apparently desolate sandhills.

[Photo from the Landscape Boards of South Australia, available under the CC-BY-3.0 AU license.]

No comments:

Post a Comment