Kawau Research Articles

Using timelapse cameras and machine learning to enhance acoustic monitoring of small boat sound

Small recreational boats are an omnipresent source of sound pollution in shallow coastal habitats, which can impact the behavior and physiology of a wide array of taxa. However, effective monitoring of this stressor is currently limited by a lack of tools. The present study coupled passive acoustic monitoring (PAM) with timelapse imagery to provide a comprehensive analysis of sound pollution at two coastal sites varying in habitat structure: Goat (rocky reef) and Kawau (sandy bay) Islands. A convolutional neural network (CNN) was used to automatically count boats in each image, and the relationship between the soundscape and number of boats present was analysed using power spectral density and adaptive threshold analyses. Small boat activity was positively correlated with third octave level (TOL) root mean squared sound pressure levels (SPLRMS 63 – 5011 Hz), and this effect was frequency dependent, at both Goat (F7,9704 = 5.665, p < 0.001) and Kawau (F7,42488 = 325.33, p < 0.001) Islands. However, at Goat Island this interaction effect was driven by a significant difference between 63 Hz and all other TOLs (p < 0.05), whereas at Kawau Island the interaction effect of TOL and boat number was more variable. Furthermore, low frequency (∼50 – 300 Hz) biophony was found to influence the likelihood of boat sound being detected at Goat Island. Small boat impacts are contextual, likely due to habitat specific propagation conditions and the presence/absence of vocalising animals. As such, monitoring of sound pollution in coastal habitats requires a tailored approach which accounts for the localised nature of shallow coastal soundscapes. These findings demonstrate the potential for timelapse imagery to elucidate variability in boat sound, which may be particularly useful for remote sites which are ecologically rich, yet have no acoustic protections, such as many marine protected areas.

A serological survey of the prevalence of antibodies against enteroviruses in brushtail possums (Trichosurus vulpecula) in New Zealand: Enteroviruses have a limited distribution in brushtail possums

AIM: To establish an immunoassay to detect antibodies against enteroviruses in brushtail possums (Trichosurus vulpecula), and to determine the prevalence of seropositive samples of antibodies to enteroviruses in wild possums in New Zealand. METHODS: A sandwich ELISA was established to detect antibodies to enteroviruses in possums. The IgG fraction of guineapig polyclonal antibodies to enteroviruses of possums was used as the capture antibody to bind the enteroviral antigens prior to incubating with test sera from possums. The bound antibodies from the test sera were detected using a conjugate specific to possum immunoglobulins. An archival serum bank was surveyed for the prevalence of seropositive samples, from possums, to the virus. The sera were collected from possums from areas of Northland, Wanganui, Castlepoint, Paraparaumu, Shannon, Orongorongo, Nelson, Hokitika, Banks Peninsula, Riverton, Kawau Island, Chatham Island and Stewart Island, between 1993 and 1996. Possums were captured from farmland, forest margin, scrub, beech, exotic forest or native forest. REULTS: Antibodies to the enteroviruses were detected in possums from Shannon (56/270; 21%), Wanganui (4/46; 9%), Kawau Island (3/44; 7%), and Orongorongo (1/43; 2%). The proportion of seropositive possums per area was similar for those found on the forest margin or pasture grazed by sheep and cattle compared with those in the forest. Seropositive possums were significantly older than those that were seronegative (p<0.01). CONCLUSIONS: The distribution of enteroviruses in possums was clustered in the Manawatu-Wanganui regions, where the viruses were originally isolated. The limited distribution was likely due to the lack of contact between possum populations due to natural barriers. If a recombinant enterovirus carrying a biocontrol agent against possums was released it would encounter little interference from field strains in the majority of the country, due to the limited distribution of these viruses in the field.

The temporal organisation of the feeding behaviour of four species of wallaby: examining chronobiological and homeostatic influences

We studied the temporal distribution of feeding events of tammar wallabies, parma wallabies, brush-tailed rock-wallabies and swamp wallabies by infrared video-recording of captive animals maintained in a 12 : 12 light : dark cycle with constant temperature, and by radio-microphone surveillance of free-ranging animals on Kawau Island, New Zealand. Analyses of mean time spent feeding per 20-min period for tammar and parma wallabies showed significant circadian variation in both captive and free-ranging animals but not in captive or free-ranging brush-tailed rock wallabies or swamp wallabies. While there was no significant difference in the extent of circadian variation of feeding rate between tammar and parma wallabies there were significant differences according to the type of food being fed. Thus, circadian variation in feeding rate was greater in both species when standing sward was fed. None of the survivorship curves of the interfeed intervals of either captive or free-ranging animals showed a discontinuity suggestive of organisation of feed events into meals. There were bicrepuscular peaks of feed-event density in both captive and free-ranging tammar and parma wallabies. There was a similar bicrepuscular distribution of feed-event density in free-ranging brush-tailed rock-wallabies but a more uniform temporal distribution of this parameter in captive specimens. There was a more uniform circadian distribution of feed-event density in a single captive and a single free-ranging swamp wallaby. Thus the circadian incidence of the feeding events of tammar and parma wallabies were more resistant to changes in the incidence of stochastic environmental variables such as predation whereas those of bush-tailed rock-wallabies and swamp wallabies were more susceptible to changes in these stimuli.

A comparison of the gross gastrointestinal morphology of genetically-similar tammar wallabies (Macropus eugenii) from different nutritional environments

The gross morphology of the body and gastrointestinal tract of genetically-similar tammar wallabies (Macropus eugenii) that had been introduced into two different nutritional environments on the North Island of New Zealand were compared using multivariate analysis. Tammar wallabies from Kawau Island had significantly lower body mass per unit body length than those from the Rotorua district where food was more abundant. Furthermore, wallabies from Kawau Island had longer small intestines and caecae relative to other gut regions than did those from the Rotorua district. This suggests that poor-quality food might reduce the efficiency of primary fermentation in the forestomach; similarly, the efficiency of digestion and absorption in the small intestine might be reduced. In addition, poor-quality food might promote secondary fermentation in the hindgut of grazing tammar wallabies.

Temporal aspects of feeding events in tammar (Macropus eugenii) and parma (Macropus parma) wallabies. I. Food acquisition and oral processing

We studied parameters that influence the efficiency of food acquisition and oral processing in the tammar wallaby (a grazer) and the parma wallaby (a grazer/browser), both in captivity and under free range on Kawau Island, New Zealand.In captivity, both species spent less time feeding per gram of dry matter intake when browsing than when grazing, and there were no significant differences between the species with respect to the rates of feeding per gram of dry matter intake of a given food. However, under free-ranging conditions, tammar wallabies spent longer feeding than did parma wallabies, so it was likely that tammar wallabies spent more time grazing than browsing. Differences in the relationships between feeding event and inter-feed interval duration in captive and free-ranging wallabies indicated that feeding behaviour was influenced by different factors in the two situations.Microtemporal analysis of the chewing sounds of free-ranging tammar and parma wallabies showed that the interval between the first and second sounds in a 'run' of chewing sounds was longer than that between subsequent intervals, indicating that there was a time cost associated with food aquisition. However, as there were no significant differences between the two wallaby species, either in the mean duration of 'runs' of chewing sounds within feeding events or in the mean duration of whole feeding events, this cost was similar for grazing and browsing. Chewing characteristics differed from those of larger (eutherian) herbivores in that the numbers of chews in a run were not randomly distributed, both species having a preponderance of runs with seven chews. Whilst, the intervals between the second and subsequent chewing sounds in a run did not vary in a periodic manner, such as would occur in batch processing of food, they were more prolonged in runs with more chewing sounds and were significantly longer in tammar wallabies than in parma wallabies. Thus, the slower rate of oral processing of grass was likely due to a generally slower rate of chewing when grazing than when browsing.

Differences in renal and alimentary water conservation account for differences in the distribution of tammar and parma wallabies on Kawau Island, New Zealand

Surveys of two wallaby species on Kawau Island, New Zealand, indicated that their distribution was stable so as to permit niche partitioning. Multivariate analysis of environmental factors associated with the relative distribution of tammar and parma wallabies suggested that their distribution may be influenced by the availability of fresh water. Tammar wallabies have greater renal size, mass and relative medullary area than parma wallabies and thus may have greater renal water-conserving capabilities. The tammar colon is significantly longer than that of the parma wallaby and the water content of distal digesta is lower in tammar than in parma wallabies, indicating that the former species may also have greater colonic water-resorption capabilities. A laboratory comparison of the water consumption of tammar and parma wallabies showed that the former drink significantly less than the latter.The superior ability of tammar wallabies to colonise drier areas may have contributed to their survival in the presence of the closely related parma wallaby.

Observations on fresh forage intake, ingesta particle size and nutrient digestibility in four species of macropod

The particle-size distributions of the ingesta of the sacciform forestomach in free-ranging animals of a grazing macropod species [Macropus eugenii (tammar wallaby)], a grazer/browser [Macropus parma (parma wallaby)], a browser/grazer [Petrogale penicillata (brush-tailed rock-wallaby)] and a browser [Wallabia bicolor (swamp wallaby)] from Kawau Island, New Zealand, were compared with those of captive animals maintained on a standing ryegrass (Lolium perenne) sward. Nutrient digestibility was also measured in tammar and parma wallabies fed ryegrass or browse, i.e. fresh mahoe (Melicytus ramiflora) and this was related to particle-size distributions of the ingesta.There were significant differences in the particle size distributions of digesta from tammar and parma wallabies in the wild but not in captivity. In free-ranging animals the ingesta from both browsing species, the brush-tailed rock-wallaby and the parma wallaby, contained consistently greater proportions of coarse particles and smaller proportions of fine particles than did those of the tammar wallaby. These differences may be correlated with reported differences in their tooth morphologies. However, the presence of significant differences in particle-size distributions of the digesta between brush-tailed rock-wallabies and parma wallabies when constrained to grass, despite reported similarities in their tooth morphology, suggests that factors other than tooth morphology contribute to differences in the oral processing of food by browsing and grazing macropods. There were greater proportions of grass fragments in the coarse than in the finer fractions of ingesta from free-ranging brush-tailed rock-wallabies, indicating that this species is less effective at chewing grass.There were no overall differences between tammar and parma wallabies in the digestibilities of organic matter, neutral-detergent fibre (NDF) or acid-detergent fibre (ADF) but the NDF and ADF digestibilites of both species increased significantly with increase in the proportion of fine ingesta particles and with increase in mass of fermentative digesta.These findings indicate the importance of oral processing to digestive efficiency in macropods and the relationship between oral processing and tooth morphology.

The histology and morphometrics of the major salivary glands of four species of wallabies (Marsupialia: Macropodiae) from Kawau Island, New Zealand

AbstractAllometric analyses of the three major salivary glands of 44 tammar wallabies Macropus eugenii and 58 parma wallabies Macropus parma from Kawau Island, New Zealand, indicate that tammar wallabies have larger parotid glands while parma wallabies have larger mandibular glands. The difference can be related to the importance of cooling in tammars, an arid‐zone species, and the need for greater buffering of forestomach digesta in parma wallabies because of their greater daily food intakes and a higher browse component in their natural diet. Comparisons of sublingual gland size of these two species viewed in terms of their capacity for mucus production suggest that this may be influenced by the need to protect mucosa from mechanical and chemical injury from browse. Sample sizes of brush‐tailed rock‐wallabies Petrogale pencilliata and swamp wallabies Wallabia bicolor were too small for similar statistical comparisons, but they confirmed that salivary gland histology was broadly similar among all four species.

Biological impacts of boating at Kawau Island, north-eastern New Zealand

Biological impacts of boating were investigated at Kawau Island, an intensively-used marine anchorage in north-eastern New Zealand. An investigation was made into the impact that the anchoring by leisure boats has on the benthic environment, in particular the fauna. If the damage to the benthic fauna by leisure boat anchoring is leading to the degradation of the local environment, then the expected pattern of use of the harbour for recreational boating is unsustainable. However, if damage could be minimised through the implementation of a management plan at the present time, then the area could continue to be viewed as a recreational resource and be managed for recreational anchoring.The investigation involved the sampling of the benthic fauna to detect disturbance or damage, and to establish the recovery period of the fauna from such damage at present use levels. The pinnid bivalve Atrina zelandica was the most conspicuous macrobenthic organism. Qualitative trends in the intensity of anchoring among sites were generally not reflected in the abundance patterns of organisms. Experimental anchoring of differing intensities damaged Atrina, which were then attacked by whelks and starfish. Anchoring scars persisted for up to 3 months, but had diminished in area and depth after 1 month. Despite protracted usage, popular anchoring sites are currently not clearly distinct from less popular sites. As intense anchoring is localised in a few bays over a short time, and macrobenthos can recover over the remainder of the year, benthic impacts are unlikely to require management at present.

Subtidal benthic marine litter at Kawau Island, north-eastern New Zealand

It is widely perceived that litter diminishes appreciation of environmental quality, and therefore may require management. Litter in coastal-marine environments has not been studied intensively, although these are likely to be encountered by humans. In order to assess the severity of litter impacts, we studied the existing litter fauna of shallow embayments in north-eastern New Zealand, in an area where boating is intense. We sought an understanding of existing patterns of litter distribution, how long litter items might take to break down or disperse, and considered possible management options.The distribution, abundance, and persistence of subtidal benthic litter were studied at Kawau Island, north-eastern New Zealand, where high densities of pleasure craft concentrated in sheltered bays during summer. A survey of the benthic subtidal environment at 37 sites found numerous litter items. Glass beer bottles dominated sites with histories of high boating usage, but litter was sparse elsewhere. Litter items placed at a popular anchorage were displaced and rapidly lost, whereas at an infrequently visited site nearby their abundances declined more slowly. Tin cans and to a lesser extent aluminium cans showed signs of physical decay over a 10-month experimental deployment, while glass and plastic did not.Although litter is locally very abundant, it does not currently affect most users, and attempts to remove it would be a more intense disturbance than any existing impact. We suggest that the appropriate management approach is to reduce the amount of litter that arrives in the shallow coastal area, by providing rubbish collection facilities over the period of intensive usage in summer.

Metastatic mineralisation in brushtail possums (Trichosurus vulpecula) from Kawau Island

Aims. To describe the nature and incidence of metastatic mineralisation in brushtail possums (Trichosurus vulpecula) caught on Kawau Island. Methods. Wild-caught possums were individually housed in captivity over a 24-month period and fed a cereal-based mash diet. Possums that became debilitated were examined post mortem for evidence of metastatic mineralisation of the heart, major blood vessels and kidneys. Serum and aortic segments were collected from affected animals and were analysed for mineral composition. Non-debilitated animals and possums that were experimentally poisoned with cholecal- ciferol were examined as a comparison. Results. Fifteen of 42 possums (36%) captured on Kawau Island and housed in captivity were debilitated as a result of metastatic mineralisation of the walls of the aorta and/or of the kidney tubules, with four further possums having evidence of mineralisation when humanely killed. Phosphate and ammonium deposits were found in the aorta, and calcium deposits in the kidney tubules. Serum phosphorus and calcium levels were elevated in the debilitated animals. Similar lesions were found in only one of 288 possums captured from other North Island areas and housed in captivity over the same time period. Conclusion. Possums from Kawau Island appear to be more predisposed to metastatic mineralisation than possums from other areas of the North Island. The lesions observed in these possums were similar to those caused by cholecal- ciferol poisoning.

Internal parasites of possums (Trichosurus vulpecula) from Kawau Island, Chatham Island and Stewart Island

As part of a search for pathogens that might be useful agents for biological control of possums, the three largest offshore islands of New Zealand that still have possums were surveyed to determine the pathogens present in these isolated populations. Brushtail possums from Kawau Island (n = 158), Chatham Island (n = 214) and Stewart Island (n = 194) were examined for internal parasites. Possums from Kawau Island were infected with Eimeria spp. (16.7%), Bertiella trichosuri (5.2%) and Parastrongyloides trichosuri (15.5%). No Paraustrostrongylus trichosuri or Trichostrongylus spp. were found. Possums from Chatham Island were infected with Eimeria spp. (10.9%). Bertiella trichosuri (3.6%), T. colubriformis (6.6%), T. retortaeformis (1%) and T. vitrinus (0.5%). No Parastrongyloides or Paraustrostrongylus spp. were found. Possums from Stewart Island were infected only with Eimeria spp. (4.6%). Because of their paucity of some parasites, the opportunity exists to use these offshore islands to study the introduction and spread of a parasite into a possum population, and what technology would be required to bring it to hyperendemicity.

The ectoparasites of brushtail possum Trichosurus vulpecula in New Zealand

The ectoparasites of the brushtail possum Trichosurus vulpecula were examined at 15 original release sites in New Zealand. Three species of fur mite—Atellana papilio Domrow, 1958, Petrogalochirus dycei (Domrow, 1960), and Murichirus anabiotus Domrow, 1992—were present in all population samples from the North, South, Stewart, and Kawau Islands. M. anabiotus was absent from the Chatham Islands. A. papilio was isolated from 82 (98%) of the 84 digested pelts, P. dycei from 78 (93%) of 84, and M. anabiotus was found in 67 (93%) of the 72 pelts digested from populations where the mites were present. The follicle mite Marsupiopus trichosuri Fain, 1968 was detected in two (7%) and one (3%) of pelts examined from Kawau Island and the Orongorongo Valley, respectively. Trichosurolaelaps crassipes Womersley, 1956 was present on 99% of possums and absent only from Kawau Island. Nymphs and larvae of the cattle tick Haemaphysalis longicornis were isolated from four (33%) of the 12 Northland pelts. The loss of ectoparasitic mites resulting from the transfer of possums from their native Australia to New Zealand has been minimal in mites with a direct life cycle. Acariasis of the rump and tail was attributed to M. trichosuri, and .A. papilio is implicated in fur breakage.

Sex of Pouch Young Related to Maternal Weight in Macropus eugenii and M. parma (Marsupialia: Macropodidae)

Competing theories of sex allocation in mammals may best be reconciled in the light of data from diverse species. The tammar wallaby (Macropus eugenii) is potentially a particularly interesting study animal because females wean only one young per year, and exhibit extreme synchronicity in the annual onset of breeding. By contrast, reproduction in the closely related parma wallaby (M. parma) is almost asynchronous. These two Australian species are found sympatrically only on Kawau Island, New Zealand, where they were introduced in about 1870. We sampled wallabies on Kawau Island in April of 1996, when both species were breeding. Although the sex ratios in both species were not significantly different from unity, offspring of M. eugenii were very significantly more likely to be male with increasing maternal weight (logistic regression χ2 = 16.8, P &lt; 0.0001), and the fewer M. parma data showed a non-significant trend in the same direction (χ2 = 1.9, P= 0.16). These data, at least for M. eugenii, are consistent with the Trivers–Willard hypothesis, and warrant further investigation in wild and captive populations under different measured or manipulated ecological conditions. We suggest an approach utilising the characteristics of M. eugenii which might help determine whether the sex bias is determined close to conception, or is effected later in the reproductive cycle by differential survival of the sexes.

First records of marsupial lice (Insecta: Phthiraptera: Boopiidae) on a brushtailed rock wallaby from New Zealand

Abstract Two species of boopiid lice are recorded from a wild brushtailed rock wallaby (Petrogale penicillata penicillata (Gray, 1825)) from Kawau Island, New Zealand. They are Heterodoxus ampullatus Keler, 1971 and Boopia notafusca Le Souef, 1902. These are the first identified records of marsupial lice from New Zealand, and P. penicillata penicillata is a new host record for B. notafusca. This new host record is discussed and evaluated.

Spatial and seasonal variation in the abundance of juvenile snapper (Pagrus auratus) in the north‐western Hauraki Gulf

Abstract Spatial and seasonal variability in the abundance of juvenile Pagrus auratus were investigated in the north‐western Hauraki Gulf using trawl tows over soft‐bottom habitats. Small‐scale (1 km) spatial variability was high and consistent over a period of 2.5 years. This variability was attributed to differences in micro‐habitat type, with snapper preferring a mud substrate over one comprising muddy sand with a significant percentage of shell. The two micro‐habitats probably result from differing current regimes, but reasons for the snapper preferences are unknown. Snapper catch rates in the Kawau Island region peaked in spring, near the time of maximum day length, and declined to low levels in autumn‐winter. This seasonal cycle is thought to reflect real changes in abundance rather than variations in snapper catchability. However, no seasonal abundance cycle was found in the adjacent Whangaparaoa region, where abundance was consistently low. There was no consistent depth‐related trend in abundance ...

Conservation genetics of the Parma Wallaby Macropus parma: a case study for Australian marsupials

The Parma Wallaby Macropus parma, native only to Australia, exemplifies a number of issues currently under discussion regarding the conservation of Australian marsupials. Thought to be extinct in the earlier part of this century, an expatriate population was identified on Kawau Island, New Zealand in 1967. These animals were used to supply zoos and captive breeding colonies throughout the world. Subsequently, parma populations were rediscovered in the Great Dividing Range of New South Wales, Australia. The Australian populations are small and inhabit severely restricted localities where they are highly vulnerable to predation and further habitat loss. Strategies for the preservation of parmas in Australia include the reintroduction of parmas either directly from Kawau Island or from established captive colonies. However, the founder number of Parma Wallabies on Kawau Island is unknown, hence it is possible the New Zealand derived parmas have a restricted genetic base compromising their suitability for reintroduction programmes. Additionally, there is a possibility that introgression has occurred between parmas and Black-striped Wallabies Macropus dorsalis on the island. Here we report that the level of genetic variation in New Zealand derived Parma Wallabies is not markedly reduced, and that no detectable introgression has taken place between Parma and Black-striped Wallabies. Indeed, re-examination of records casts doubt upon the suggestion that Black-stripes were introduced to Kawau Island.

Circulation in the Hauraki Gulf, New Zealand

Abstract Results from five current meters moored in Hauraki Gulf are analysed. Tidal flows and the tidal constituents M2, S2, and N2 are deduced. Time series techniques are used to examine the effects of wind stress and the spatial coherence in the flow. It is found that currents of oceanic water, set up under south‐west to southerly directed winds, flow into the Gulf through Jellicoe and Cradock Channels, with those through Cradock Channel probably passing more or less directly out through Colville Channel. A net anti‐clockwise circulation is suggested to occur in the inner Gulf, south of the latitude of Kawau Island. A lack of any particular spectral features and limited spatial coherence suggests that in the Hauraki Gulf, variations in circulation in the 2‐ to 15‐day time band considered are not well defined but that the Gulf responds to external forcing events with transient currents which last typically from 2 to 5 days.

The origin of the feral dama (tammar) wallabiesMacropus eugenii, in the Rotorua area of New Zealand

Abstract Dama (tammar) wallabies (Macropus eugenii) have become established in New Zealand around Rotorua and on Kawau Island. They originated from Australia, but from which of the populations in South Australia and Western Australia is not recorded. Examination has been made of allozymic, serum protein and red cell antigenic markers in the Rotorua population. Comparisons have been made with the same markers in South Australian and Western Australian populations. The results clearly indicate that the Rotorua population is very similar to the Kangaroo Island population (South Australia), and quite distinct from the Garden Island population (Western Australia). The Rotorua population may be derived from Kangaroo Island or the now extinct Adelaide population (South Australia). The Kangaroo Island tammar is widely used in Australia for basic biological research. New Zealand workers could use Rotorua animals for the same purposes and safely compare their results with results obtained from current captive popul...

WIDESPREAD OCCURRENCE IN AUSTRALIAN MARSUPIALS OF NEUTRALIZING ANTIBODIES TO A HERPESVIRUS FROM A PARMA WALLABY

Serum neutralizing antibodies to a parma wallaby herpesvirus (PWHV) have been detected in a wide range of marsupials from different locations across Australia, including several islands. A study of 242 animals (mostly macropods) sampled in the wild showed that 23% had antibodies; a significantly higher frequency (41%) of 116 animals in captivity had antibodies, which were generally at higher levels than those of animals in the wild. Antibodies to PWHV were also detected among parma wallabies from a colony on Kawau Island, New Zealand. The highest antibody levels were found in a group of captive tammar wallabies during recurrent outbreaks of clinical infection. It is suggested that PWHV has evolved along with a marsupial host, and that the high antibody levels among captive animals reflects ease of virus transmission due to crowding, or to conditions of stress leading to expression of latent virus.