Water Deer Activity - Types of Activity

It is again studies in captivity that have provided some insight into the activities in which water deer can be found engaged. I will summarise some of these studies here because they're the only data we have, but it should be recognised that these are not necessarily representative of wild deer.

At Whipsnade, Endi Zhang found that feeding was the principal daytime activity, accounting for an average of half (50%) of daytime hours across the study as a whole – almost all of it (98%) spent grazing rather than browsing. Resting was the second most common activity, at just shy of 38%, with the remaining 12% devoted to standing, social behaviour, walking and auto-grooming. This overall figure masked considerable seasonal variation, though: quarterly feeding time averaged 50.2% in winter, 47.6% in spring, 46.9% in summer and peaked at 56.8% in autumn. There was no overall difference between the sexes in diurnal feeding time, though males showed slightly greater variation across the year. Females increased their feeding between May and June to around 55%, sustaining a similar level (55–56%) through to September, while males held steady at around 45% over the same period, instead peaking later, at 57% of daytime hours in February once the rut was over. During the rut, which Zhang defined as November–December, feeding time in both sexes fell to its lowest average of the year -- 41% in bucks and 42% in does -- with December itself the single lowest month, when male feeding dropped to 38%. Bucks consistently devoted significantly more time than does to walking and social behaviour, a difference most pronounced during the rut, when it rose to 5–7% of daytime hours in males against just 1–3% in females.

Observations by Zhang and Stefan Stadler at Whipsnade suggest that grooming is a fairly common activity, and my observations of both free-range captive and wild animals supports this. Auto-grooming (i.e., grooming ones self) involves nibbling of the coat between the lower incisors and the gum pad of the upper jaw, and Stadler considered it a "comfort behaviour" along with scratching, stretching and yawning. Based on statistical analysis of 12 deer of each sex, Zhang found that it accounted for 3–4% of daytime activity, with bucks and does equally likely to partake. Stadler also found auto-grooming tended to last for only a few seconds, but in one case an individual spent seven minutes engaged in the activity. In Stadler's sample, grooming was most common and prolonged during May (i.e., the moult), and most often observed in transition from activity to rest or vice versa. In other words, resumption or termination of an activity could be roughly predicted by the onset of grooming. Scratching, which involved standing on three legs and using the fourth to manipulate part of the body, was sometimes associated with grooming and typically used to reach those areas of the head and neck inaccessible to the mouth. Scratches lasted between four and six seconds typically, and were often repeated two to four times. Stadler didn't record any prolonged scratching bouts. Stretching was most often performed after long periods of inactivity, but also occasionally observed during feeding bouts.

We have very few data on the activity budgets of wild deer, although the scant observations we do have appear to align with the behaviour of captive individuals in at least some aspects. Both sexes are "on the move" and interacting socially the most during the winter rut, while females spend much of their time feeding during the summer, when the demands imposed by lactation are high. I know of no direct observation studies of water deer activity during the night, but on a handful of occasions when I have made observations in Bedfordshire late into the evening during mid-December, I have found bucks and does either resting together or feeding.

In common with other species of ruminant, while at rest water deer typically exhibit "sternal recumbency". In other words, they assume a semi-upright position, tilted to one or other side of the body, with their legs tucked up against their stomach. This resting position is largely imposed by their rumination mechanism and the deer's head will be upright, and the ears erect as the animal periodically scans for danger. In captivity, I have seen deer propped up against fences or walls, exhibiting a more upright positioning of the body, the legs almost underneath the animal. In my experience, during sleep the head is brought down and tucked against the legs, giving a curled position similar to that of a cat, with the ears remaining erect. The choice of resting site varies, but this species appears to prefer areas either with vegetation sufficiently tall and dense to hide them completely or, in open landscapes, field centres that presumably provide a clear panorama to scan for approaching danger. One might imagine that lying out in the middle of a field would leave the animals exposed, but water deer can 'disappear' into only a few centimetres of vegetation. This remarkable recession is something I've witnessed on several occasions and the late Gerald Durrell made a similar observation during his time at Whipsnade Zoo in the mid-1940s, writing in his 1973 book Beasts in My Belfry:

"You would think that in a large paddock, where the turf had been cropped down by various herds of antelope and deer, an animal this size would be conspicuous; but a Chinese water deer lying down in grass three inches long just seems to melt away and you do not see it until you have almost stepped on it."

In his 1982 paper, François Feer described a characteristic sequence of behaviours preceding recumbency (lying down). Individuals were observed to sniff the ground before lifting a forefoot and rocking or swinging it forwards and slightly side-to-side, as if "hoofing into the void". This action was repeated alternately with both forefeet, followed by both hind feet, after which the animal again sniffed the ground, turned around, and lay down. Feer noted that this sniffing and turning sequence could be repeated several times before the animal finally settled. He also reported that the same behavioural pattern occurred in both short and tall vegetation, although the feet were lifted higher when vegetation was taller. As this activity appeared to remove or flatten surrounding vegetation, Feer interpreted the behaviour as preparation of a resting site within the dense vegetation typical of the species' native habitat.

Similar behaviour was observed by Stadler at Whipsnade, where the sequence typically consisted of a sniff-paw-sniff pattern, often involving only a small number of foot scrapes before the animal lay down and commenced grooming. On some occasions, however, the sniff-paw cycle continued for more than a minute. The behaviour was less evident in areas of short, open grassland, further supporting the interpretation that it functions to clear or prepare the resting place.

I have also observed forefoot scratching or scraping prior to recumbency on multiple occasions, including instances involving both fore- and hind feet, although I have not observed the turning behaviour described by Feer. In one case, a buck rested briefly before standing, scraping again with both fore- and hind feet, and then lying down once more. The hoofing behaviour seems inherent and Chris Whyke sent me a video of his hand-reared water deer doe doing it before lying down on the sofa with his dogs. It should be noted, however, that scraping does not invariably precede resting: in some instances, I have observed individuals approach a resting site -- possibly scraped/prepared earlier(?) -- and lie down immediately, dropping onto the forelimbs before lowering the chest to the ground and finally settling the hindquarters.

Water deer are readily disturbed and display a strong propensity to flee when alarmed. Both my own observations and those reported by Arnold Cooke in Muntjac and Water Deer indicate that these deer typically respond to disturbance in one of two ways. Individuals may crouch low and attempt to remain inconspicuous, even in short or sparse vegetation; or they may flee abruptly in a manner reminiscent of brown hares. This escape response can include a pronounced leaping/bounding gait, with the hind legs thrown high behind the body in a semi-kicking motion that brings the hooves close to the level of the back. Describing the escape behaviour in his original 1870 account of the species, Robert Swinhoe wrote:

"In running they cock their ears, round their fore legs, bend up their hind legs, hog their rumps, and scurry away with little quick leaps, very much after the manner of a Hare."

On several occasions, I have observed water deer standing and watching an approaching disturbance while holding one foreleg raised -- most often the right -- behaviour superficially similar to the stopping posture sometimes adopted by roe deer (Capreolus capreolus) and muntjac (Muntiacus reevesi). I've not observed water deer perform a true stomping action, however. Cooke further suggests that the bounding escape behaviour of water deer, which he notes is more frequently observed in open landscapes, may be functionally analogous to the pronking behaviour exhibited by other deer species.

While quick to flee, as tends to be the case with so-called "saltatorial runners", water deer do not appear to be built for endurance and they will often drop down suddenly into cover once sufficient distance has been introduced between them and the source of the disturbance. I don't know how water deer decide where to drop, but given that it happens so suddenly I can't help wondering whether they remember couch locations and make a direct flight to them. Writing in 1950, Kenneth Whitehead suggested that the short distance sprint-and-drop method was probably a throwback to "their jungle days, when 'lying doggo' in dense cover was their best mode of escape".

During his analysis of gait in 79 mammals filmed running in slow motion, published in May 1977, University of California, Davis zoologist Milton Hildebrand described how deer, including water deer, have only one suspended phase during a gallop: the "extended suspension". This means that all feet are only off the ground simultaneously during the stretch of the bound. Hildebrand looked at which feet hit the ground first while running (the "lead sequence") and classed water deer as having a rotary gait sequence. In other words, if the right hind foot lands just before the left hind foot, the right forefoot lands before the left forefoot too. To the best of my knowledge, there are no confirmed data on how fast this species is capable of running, nor the maximum distance over which speeds can be maintained.

In his 1998 book Deer of the World, Valerius Geist commented on how the saltation mode of escape employed by water deer puts great physical pressure on the lower hind legs, driving the evolution of massively muscled haunches and a long, muscular back. The haunches are certainly much more muscular and powerful than the front legs, and presumably richly endowed with fast-twitch (white or glycolytic) muscle fibres to allow the rapid propulsion from a standing, crouching or lying start. In her allometric analysis of deer limbs, Rutgers University zoologist Kathleen Scott observed that water deer were the only small cervid with markedly elongated limbs, the greatest elongation being in the forelimbs such as to produce limbs of almost equal length. Scott suggested that, although their choice of reedbed and tall grass habitat provides sufficient height of cover, the relative openness may place a greater premium on speed. Heavy undergrowth is not an environment in which speed is either desirable or necessary, which might explain why similarly sized cervids living in dense scrubby/wooded habitats (e.g., muntjac and pudu) have shortened limbs, allowing them to move more easily through brush and probably making them less noticeable.

The powerful hind quarters don't only generate explosive speed, they also make water deer reasonably good jumpers, although my experience has been that they prefer to go through or under fences where possible. As part of a project to look at reducing deer-vehicle collisions in South Korea, a team of researchers led by Hee-Bok Park, at the National Institute of Ecology, studied the behavioural response of water deer to fences of differing heights. While the deer were generally reluctant to jump obstacles, they could easily clear fences up to a metre (3 ft 4 in.) in height, doing so from a standing start. Fences between 1.2 and 1.5 metres (4–5 ft) were less likely to be crossed but could still be cleared with a running jump. A few deer were even able to jump a 1.8 m (6 ft) fence.

As their name and habitat selection attests, water deer are at home in wet/flooded areas and, in common with most deer, are strong swimmers. Arnold Cooke told me that he has only once observed a water deer swimming at Woodwalton, when he disturbed one beside the Great Raveley Drain on the east side of the reserve and it leapt into the water to escape. In Muntjac and Water Deer, Cooke recounts several observations described to him, however, including of animals swimming between areas of prime habitat on the Norfolk Broads, and a particularly interesting incident where three water deer were swimming in floodwater that had encroached into Dunwich Forest in Suffolk after a storm surge in December 2013; one individual apparently browsing on flooded gorse before swimming away through about 1.5 m (5 ft) of flood water when approached. I have only ever seen photos and video of swimming behaviour in this species. Nonetheless, in their preliminary study on the population of Zhoushan Island and adjacent islets, published in Acta Theriologica Sinica during 1984, Helin Sheng and Houji Lu note that the deer swam several kilometres between the islands in the archipelago looking for food and avoiding hunters. In his detailed review of the ecology of Korea's Demilitarized and Civilian Control zones (DMZ and CCZ, respectively), published in 2013, Kwi-Gon Kim wrote of the water deer:

"[it] likes water. It swims twice a day and drinks often, especially more during summer when it is hot."

In an engaging article to Country Life in May 1953, Kenneth Whitehead described a hand-reared water deer doe who seemed to enjoy water. The animal was subjected to the full force of a watering hose on several occasions and failed even to move and, during one summer, was found laying in the main jet of a hose for more than an hour, only moving when the water was turned off. She was also observed placing her nose within about an inch of a hose nozzle and holding it there. Interestingly, despite Whitehead and Kim's observations, water deer are rarely observed either swimming or drinking here in England, the latter presumably because they obtain most of their hydration from their food as most herbivores do. I have only observed drinking by water deer twice, once in the wild and once in captivity, and have received only one report of the behaviour. In January 2026, Gary Rugless watched and photographed a young buck licking melting snow in a field in Norfolk.

Finally, while I know of no resting metabolism data for water deer, Barkley Hastings and colleagues, in their 1989 paper to the Journal of Zoo and Wildlife Medicine, calculated the normal activity heart rate for water deer as 130 beats per minute, with 28 breaths per minute.

References

Cooke, A. (2019). Muntjac and Chinese Water Deer: Natural history, environmental impact and management. Pelagic Publishing, Exeter.

Durrell, G. (1973). Beasts in my Belfry. Collins, London.

Feer, F. (1982). Quelques observations ethologiques sur l'Hydropote de Chine, Hydropotes inermis (Swinhoe, 1870) en captivite. Zeitschrift für Säugetierkunde. 47: 175-185. [In French]

Geist, V. (1998). Deer of the World: Their Evolution, Behaviour, and Ecology. Stackpole Books, London.

Hastings, B.E. et al. (1989). Reversible immobilization of Chinese water deer (Hydropotes inermis) with ketamine and xylazine. Journal of Zoo and Wildlife Medicine. 20: 427-431.

Hildebrand, M. (1977). Analysis of asymmetrical gaits. Journal of Mammalogy. 58: 131-156.

Kim, K-G. (2013). The Demilitarized Zone (DMZ) of Korea: Protection, Conservation and Restoration of a Unique Ecosystem. Springer Heidelberg, New York.

Park, H-B. et al. (2021). Assessment of the behavioural response of Korean water deer (Hydropotes inermis argyropus) to different fence heights. Animals. 11: 938.

Scott, K.M. (1987). Allometry and habitat-related adaptations in the postcranial skeleton of Cervidae. In: Wemmer, C.M. (ed.) Biology and Management of the Cervidae. Research Symposia of the National Zoological Park. Smithsonian Institution Press, Washington. pp. 65-80.

Sheng, H. & Lu, H. (1984). A preliminary study on the river deer population of Zhoushan Island and adjacent islets. Acta Theriologica Sinica. 4: 161-166. [In Chinese]

Stadler, S.G. (1991). Behaviour and Social Organisation of Chinese Water Deer (Hydropotes inermis) under Semi-natural Conditions. Ph.D. Thesis for Universitat Bielefeld, Germany.

Swinhoe, R. (1870). On a new Deer from China. Proceedings of the Zoological Society of London. 38(1): 89-92.

Whitehead, G.K. (1950). Deer & Their Management in Deer Parks of Great Britain & Ireland. Country Life Ltd., London.

Whitehead, G.K. (1953). A Tame Chinese Water-Deer. Country Life. 113(2941): 1691.

Zhang, E. (1996). Behavioural ecology of the Chinese water deer at Whipsnade Wild Animal Park, England. Ph.D. Thesis to University of Cambridge. 196 pp.