Kingfisher spearing into a tree

If I told you that mangrove kingfishers fly into trees at full speed to make a nesting hole, you would not believe me, so watch the video.

This mangrove kingfisher (Todiramphus chloris) is nesting in a dead Sonneratia tree deep within a tall stilt mangrove swamp. At intervals of approximately one minute, the kingfisher flew into the tree with a run up of approximately 10 m. The bird was in a frenzy to make the nest hollow as quickly as possible. A female was watching. Mangrove kingfishers are quite large at about half the size of a kookaburra, so it is quite surprising that they can survive such an impact, let alone repeat it. In the video, the kingfisher passes through the field of view in a single video frame. I saw about six hard runs at the tree.  Between runs, the kingfisher flew around the tree, landing on stilt roots, calling and displaying to the female.

The tree that the kingfisher was chiseling away at also has a story.  It belongs to a species that only colonises the seaward edge of mangrove swamps.  However as the mangroves prograded, it became stranded and then overshadowed by stilt mangroves (Rhizophora spp.) and died.  Beetles drilled into the tree and spread fungal spores which grow into the white fungal coating that the beetle larval eat.  Tasty beetle larvae attract striped possums and this tree was cratered with striped possum pits when I first recorded the tree in June 2017 for the striped possum post.  From my experience in locating fig parrot nests, I know that there is a window of time between when the tree becomes soft enough for a bird to make a hollow and the tree developing vertical cracks which pipe water down inside the timber or the tree simply falling over.  This window probably lasts for several months only.  Mangrove trees also tend to be quite thin and it is would be hard to find one with sufficient diameter for a nest.  Sometimes a kingfisher will hollow out a tree that is too small and will drill clean through the tree or create cracks that allow chunks of wood to fall away.  Looking at the hole, I suspect the female will reject it.  The mouth of hollow is just too wide and it appears that part of the roof of the hollow has fallen away.

The kingfisher hollow, photo taken at night when kingfisher was away.

Water mouse (Xeromys myoides), a predator of crabs

One of the most mysterious and elusive mangrove animals is the water mouse (Xeromys myoides). They mostly live in mud nests in the mangroves and come out at night to prey on small crabs. They occur in the Northern Terrritory from Darwin to Arnhem Land and in Queensland from Cannonvale, near Mackay down to the Queensland border. The gap between the NT and QLD populations is about 4000 km in round numbers and possibly twice that long for a small rodent that can’t swim across wide rivers. During the previous twenty years, many zoologist have searched long and hard to find water mice in the great gap between the known populations and no water mice were ever found until now.

Records of water mice, Atlas of Living Australia

On 20 June 2017, I was walking through the mangroves in Cairns at night when I saw an interesting rodent with a white underside running around on the floor of a stilt mangrove (Rhizophora) swamp. I saw enough to suggest that the rodent was a water mouse but could not get a photo of the mouse as it disappeared through the dark tangle of roots. Six days later, one of these creatures came right up to me even though I was following it with a powerful torch. This time the camera was ready and as a result water mice have been officially recorded for Cairns, which is 500 km north of the next closest record as the seagull flies.

The first photo of a Cairn's water mouse, about to escape into a crab hole

Perhaps one of the reasons why water mice have been so hard to locate is that it is hard to form a good search image for them. Most of the available information is in text form, which not quite the same value as a photograph, especially for a visual thinker like me. There are also many other creatures that leave similar signs to water mice. Adding to the complication, water mice are apparently not present in every seemingly suitable mangrove swamp either and nobody knows why. In this post, I will try to present a clearer picture of what to look for.

Water mouse, also known as false water rat (photo: wikimedia)

Water mice occur as far south as the Queensland border and occur in very different environments to the places where they have been found in Cairns. In southern areas, they were first studied at Stradbroke Island, which is a giant sand island. At low tide, freshwater seeps from the ground so I thought that water mice were limited by freshwater availability. In these areas, the mice made strange mud nests in clumps of sedges. On the Noosa River, also in southern Queensland, they made mud nests that were referred to as termitaria-type nests. Several years later, water mice were found on Curtis Island, near Rockhampton. Curtis Island is in a dry region where mangroves occur as a band between the salt pans and the sea and there is unlikely to be much groundwater seepage. In this area, they lived in the Ceriops mangroves, which are mangroves that prefer higher and more saline environments.

The Mackay region is believed to be the species stronghold, however the mangroves in Mackay are challenging place to search. Huge tides flush away the relatively high rainfall of Mackay and make the mangrove environment a more saline environment than it ought to be. Stunted mangroves often occur in low woody thickets of robust trees that difficult to push through let alone walk through. Some swamps in river mouths or near grassy plains can even have an understorey of succulent herbs and grass! Given the known habit in southern Queensland, the swamps with grassy understoreys and adjacent grassy flats with signs of freshwater influence were the obvious place to search, however data collected over the last twenty years suggests that water mice do not like that habitat.

Mangroves with grassy understory, Dunrock near Mackay (water mouse central)

In Cairns, the tidal range is smaller than Mackay and the remains of crabs that have been preyed on by water mice are less likely to be swept away. Mangroves thrive on the higher rainfall and are taller and are easy to move through. However, water mice are not the only crab-eating rodent and care has to be made when identifying both rodents and the signs they leave. Rakali (Hydromys chrysogaster), which are approximately six times larger and a formidable predator dominate the swamps. I thought that rakali might also prey on or fight with water mice and limit their distribution. There must be some kind of habitat partitioning however where water mice are present, rakali are also almost certainly present. Introduced black rats are also a common predator on crabs in mangrove areas. Some other native rodents such as Melomys also venture into the mangroves. Water mice apparently do not climb trees to escape, like the rodent below.

These unknown rodents were also foraging in the mangroves 

Water mice prey on smaller grapsid crabs, which tend to have flat rectangular bodies, usually grey in colour. Grapsid crabs often live on the lower part of trees or in simple burrows that go straight down into the ground. Sesarmid are larger crabs that have a squat cylindrical body shape (scone-shaped) and often live in mud igloos which the crabs build on the forest floor. Remains of sesarmid crabs that have been caught by a rakali look as if the crab has been blown up. In contrast, water mice leave a tidier scene. Typically, there will be a bright white upturned carapace which is often close to a scattering of legs and nippers. The best evidence that crab remains are from water mouse predation is an intact segmented breast plate, which is the undersurface of the crab (pers comms Tina Ball). As the carapace is small and can potentially be moved by tides, finding the carapace with the other parts provides stronger evidence that the water mouse consumed the crab at that spot. Empty crab shells with the legs still attached are probably moulted exoskeletons rather than remains of predated crabs.

Small, thumbnail sized empty crab shells are the main sign of water mouse presence

Rakali eat larger crabs and make a mess of them

A cleanly removed breast plate is a good water mouse sign

Water mice are too small to leave trails of footprints, except in places where the ground is quite soft and where it remains undisturbed by other creatures. Such conditions are rare in mangrove swamps and I have only seen footprints where mud had dried hard soon after the prints were made or in soft wet mud that had yet to be disturbed by the traffic of crabs and snails. Rakali leave prodigious numbers of footprints. Rakali have long webbed hind feet with toes of different lengths. Their front feet leave large star-shaped prints with the span of a 50 cent piece. Bandicoots, wallabies and a variety of other terrestrial wildlife including other small rodents also get around in the mangroves and leave prints, so footprints provide unreliable evidence.

Possible water mouse footprints (front and rear)

Water mice live in mud nests which they construct. Depending on the surrounding environment, these nests can be easy to see or they can be almost impossible to distinguish. Mud lobsters, sesarmid crabs and even mangrove ants create large mounds of mud that are shaped like water mouse nests and these other mounds can be so numerous, they even outnumber the mangrove trees. Suitable mud for nest building may be one of the environmental parameters that water mice require. The nests also need to be located near the high tide line as the water mice probably do not like deep or prolonged submersion. Whilst water mice feed in stilt mangrove swamps, they prefer to make nests in areas which are tidally inundated less often.

Possible nest pointed out by Tina Ball

A hollow filled with mud, with crab remains on top is very likely to be a nest

A possible nest showing the horizontal entrances, crab remains were in the tree hollow as well.

The nests have a variety of forms and many other animals make similar mud nests so it is hard to be certain whether the nests belong to water mice. Crab igloos often have vertical chimney style exits whereas water mice tend to make horizontal exits which are supposed to be more oval-shaped than crab holes which are round. As water mice are a listed threatened species (listed as Vulnerable), breaking nests open to see the occupants would be an offense under the Nature Conservation Act. The only alternative is to look for crab dinners which the mice consumed on or in their nest or to set up a camera trap to photograph the water mice as they come and go.

Water mice also make mud ramp nests, but this mud ramp is likely to be a crab house

Sesarmid crab igloos around the base of Ceriops mangrove trees

My working theory is that the most productive feeding areas for water mice are the drier mangrove forests, where grapsid crabs are more abundant than sesarmid crabs. Look for where the ground is flat and full of small crab holes and avoid areas where the mud is covered with crab igloos or the ground is intensely churned by subterranean creatures such as mud lobsters. Proximity to freshwater is not a requirement as the places they have been found are unlikely to have potable freshwater during the dry season. In Cairns, water mice were found deep in mangrove communities that are isolated from the landward fringe and terrestrial vegetation. Water mice seem to be in most of the mangrove swamps I have searched, however water mouse sign is scarce in luxuriant mangroves be they swamps of Ceriops, Rhizophora or some other species. Sign was most abundant in harsh saline back swamps that may be seasonally become quite fresh during the summer wet season. These places tend to be mosquito infested, even by mangrove swamp standards. Some of these places are also subject to deep and violent flooding when creeks within the catchment run deep and water mice apparently climb up high inside hollow trees to survive these conditions. Water mice seem to hunt mainly along the boundary between the Ceriops and Rhizophora mangrove zones where tides are still a regular event and crabs are usually active.

Water mice hunt in stilt mangrove swamps as well as Ceriops swamp

As a threatened species, I will not tell exactly where I found the water mice but those who are interested can contact me via this site. I would also like to thank Tina Ball from the Queensland National Parks and Wildlife Service, for investigated my sites in the field and via camera trapping. She has was able to officially confirm the presence of water mice and to help me sharpen my knowledge of this mysterious creature.