Day-Time Roost Patterns of New and Previously Translocated North Island Brown Kiwi (Apteryx mantelli)
Round 1
Reviewer 1 Report
I consider the article very good and highly relevant for the conservation of an endangered species. The data presented is of high quality and is capable of directly helping in the practice of managing the species with a view to its conservation.
I bring here just a few more general issues that I consider important to be discussed in the text in order to better situate the reader about the results presented here and how much these results actually reflect the natural aspects of the biology of the species.
As the central focus of the article is to study the spatial day-time roost range of the North Island brown kiwi, the central question in the introduction should discuss whether the fact that there is a physical limitation (fences) to the dispersion of birds could be influencing the size of the day-time roost range. Could the limitation for dispersion be influencing the size of the area? If there were no limitation for dispersion, would birds have larger areas? I think this has to be a central point in the introduction, where the authors could present available data on the size of the day-time roost range in areas with and without limited dispersal of these birds.
Another point that could be discussed in the introduction is to present a bit of the biology of the North Island brown kiwi in order to situate the reader when there are more ecological requirements and thus base what would be expected regarding the area size of this species. For example, could the high concentration of very close individuals in the same area be influencing the size of the area, etc?
Author Response
The authors would like to thank the reviewer for their comments and have provided the response below:
1. Area available /confinement (fences): The authors agree with this comment and have addressed this in the introduction and discussion.
- Addition in Introduction:
"According to previous studies, home ranges can vary considerably and are likely to be influenced by habitat area or confinement and kiwi density. Zhang et al [12] found that dispersal by southern brown kiwi (Apteryx australia) related to the habitat availability and location in the landscape, particularly the presence of large forest habitat patches. Ziesemann [13] showed that nocturnal roost areas were small for NIBK on the small but high density Ponui Island (18 km2) whereas McLennan et al. [14] found that day/night home ranges of birds in the much larger and less dense Kahurangi National Park translocation were considerably greater. A further example, Toy and Toy [10], in which monitored the spatial use and home ranges of roroa (greater spotted kiwi, A. haastii) translocated to the Flora Stream area in the Kahurangi National Park, New Zealand was monitored for a period of 2-8 years. Translocated birds established stable home ranges in as little as 9 days but extending to almost 2.5 years and these covered areas of 33 – 1,745 ha estimated using the Minimum Convex Polygon approach. Additional pest control was implemented during the dispersal phase outside of the original area to protect these far-dispersing birds (maximum 9.8 km), while 4 others were retrieved and returned to the original area where they set up home ranges. This indicates that available space or containment may limit the range size for translocated kiwi."
b. Addition in discussion:
"The literature [13,14] suggests that day-time roost or day/night-time ranges may be influenced by habitat area and bird density. Restricted range may potentially have been an issue in this study in relation to the proximity of the Kaipara Harbour water boundary (Figures 2 and 3), however, Jamieson et al [5] reported that free-roaming birds used swamp areas to a limited extent. It is suggested that the property perimeter fence had little if any impact on roost range, as there was substantial distance between the edge of the ranges and the property boundary. In relation to bird densities, the roost range areas in this study are greater than that found on the confined high density Ponui Island [13] but less than the low density, day/night range Hawke’s Bay study[14]. This intermediate day-time roost area was expected, due to the lower density of birds at the Mataia property and the large ranges of the birds in the Hawke’s Bay study, due to the measurement of both day and night ranges."
2. Biology of the species: The authors agree with this comment and have added a description of this species as the first paragraph of the introduction
"North Island Brown kiwi (NIBK) (Apteryx mantelli) (Order: Struthioniformes, family: Apterygidae)(Bartlett, 1850) are nocturnal flightless ratites, which are endemic to New Zealand and are classified as ‘Vulnerable’ as per the IUCN Red List [1]. Once widespread in North Island, New Zealand, the decline of this ground-dwelling species has been primarily due to introduced predators, such as dogs, cats, rats and stoats but also the loss and fragmentation of habitat [2]. Kiwi are predominantly found in New Zealand native forest habitats and forage during the night consuming primarily invertebrates, earthworms, beetle larvae and small amounts of fruit and leaves (NZ Birds Online, 2013, Dixon 2015)[3,4]. During day-light hours they retreat to burrows preferably in either living or dead trees or holes in the ground in native forest habitats [5]. "
Reviewer 2 Report
This is an interesting manuscript which uses two measures (MCP and KDEs) to estimate variation in roost size area of translocated North Island Brown kiwis released in 2013 and 2014. Sample sizes in field studies of this nature is always constrained, and although these data are from an older 2013 and 2014 dataset, they do serve as an important case study to further our knowledge of movement behaviours and habitat preferences immediately after post-release. I do have some comments on the manuscript:
Tracking data: There is not enough acknowledgment in the manuscript concerning how representative these data are of the species, particularly given such a short period of post-release monitoring. Research from numerous avian (and other taxa) translocation projects have revealed that some translocated individuals, when unfamiliar with their release site, do move considerably (this is mentioned briefly in their discussion but see my later comment on this). In this instance, translocated kiwis did not move that much, particularly the 2014 released birds, and the authors suspect that this may be attributed to anchoring. But what happened after the five-week period (given that these data were from 2013 and 2014 – surely we know what happened to the birds)? How likely is this simply an artifact of the exceptionally short sampling period (just 12 days over a five week period). Or even perhaps the release strategy? The similarities between roost size of 2013 and 2014 birds is interesting, but did the roost area size and habitat preferences of the 2013 birds change in 2014 following the release of the new birds that year? That seems to have been overlooked and is important to present, given the overlap in roost size area and the possibility of an ‘anchoring effect’.
In addition, the use of two estimators of roost home range size is ok but further details are needed on how these estimators were calculated. For example, I’m assuming that MCP100% was used, but this is not entirely clear to the readership that this is the case. Secondly, what about outlier locations and exclusion of the furthermost locations – did the authors have to apply an outlier restricted edge polygon using an outlier exclusion method (see Kenward et al., 2008 and also the approach taken by Bernardo et al 2011)? A more robust approach would be to reanalyse the data and use a wider consensus of estimators, then selecting one which has the lowest amplitude in roost size area estimates. This approach is used quite nicely again by Bernardo et al 2011 – who select between minimum convex polygon (100% MCP and 95% MCP); fixed kernel (95% FK); adaptive Kernel (95% KA); and Neighbor Linkage (95% NL). The authors of this current study specifically do not address differences in their two estimator performances, which I feel is an error, as it is an essential component for a study of this nature (and for the Diversity readership), and would further bolster the claim that this study represents a case study of much broader interest and would represent a suitable.
Introduction: Not enough background on the importance of understanding the propensity to move by translocated individuals into a novel release site, coupled with more details regarding habitat preferences. Numerous case studies on translocated terrestrial bird species could (and should) be utilised for this.
Methods: Details of the translocation are missing: for example, were the birds released using soft-release protocols or was this a hard release? The type of strategy (and degree of acclimation to the release site) again can affect the birds propensity to move and their patterns of habitat use following release. Was supplementary feeding used to further post-release site establishment – if so, provide details as can also have a direct effect on immediate post-release habitat use. Translocated birds (animals) also typically suffer very high post-release mortality during a critical period of vulnerability – tis can be a matter of weeks or months. Given that the sampling period was only 5 weeks, would this not correspond to such a period (if this is known)?
Discussion: Too much of the discussion is largely kiwi specific. That’s understandable but for Diversity and its readership, I would encourage the authors to broaden the comparison and contrast with studies to other translocated species. This is important since their 2014 birds did not move as expected – e.g. Bernardo et al 2011 found that it was in fact the ‘established’ birds (red-billed curassows) from a previous cohort actually moved on the arrival of a newly released cohort, and not the newly released birds. Other studies have revealed that captive-reared species
Author Response
The authors thank the reviewer for their comments and we have provided their response below. The response is structured in relation to the Introduction, the Methods and the Discussion:
Introduction
We have now noted explicitly that dispersal is an important criteria in determining the success of translocations and that in general translocated birds tend to show wider movement than resident birds. We have included some references in support this that include other terrestrial birds as well as kiwi. We have also included that likely relationship between habitat quality and territory size that is likely to translate into a similar relationship between extent of dispersal and habitat quality in translocated birds.
Additions are:
a. "One important criteria for success is the likelihood of dispersal and whether such dispersal remains confined to areas of suitable habitat which in New Zealand also usually includes the management of introduced terrestrial predators of endangered birds (Jahn et al 2022a)."
b. "Translocations of volant terrestrial birds showed these birds tended to move over larger areas than resident birds with their movements being elevated immediately following release (e.g. Clarke & Schedvin; Kemink & Kesler 2013). Dispersal of translocated and released flightless kiwi is also an important issue influencing population establishment, especially at unfenced mainland sites (Jahn et al. 2022a). The extent of movement is also likely to be related to habitat quality. Territory size of Brown kiwi (Apteryx australis mantelli) decreases with proportion of preferred habitat (Taborsky & Taborsky 1995) and it is likely that the extent of dispersal of translocated birds is similarly impacted."
Method:
We now indicate explicitly that it was a hard release with no supplementary feeding. In respect of the estimators of roost home range we would argue that this is not a methodical paper and have restricted estimators to the two commonly used approaches which should be familiar to readers of Diversity.
The authors agree that the use of the 100% MCP and 95% KDE is not clear and have made edits to address this comment and have added a further couple of examples of use of these measures for the estimation of home range.
Addition: "...., while 100% MCP and 95% KDE has frequently been used to determine the home range of a number of different species [25,26]."
However, the authors disagree with the suggestions that the analysis methodology should be reviewed. The use of 100%MCP and 95%KDE has been used frequently to estimate the home range of a range of species, and, as we say "Use of these methods (100%MCP and 95%KDE) was seen as a means to provide a broader understanding of the spatial distribution and also to address inherent disadvantages in each." In addition, the authors would like to highlight a statement in their article that "This study does not compare these analysis methods but focuses on the difference between roost area for birds released in 2013 and 2014." . As such, the authors deem that the methodology used is appropriate for this study.
Discussion
We have made brief reference to other terrestrial birds in respect of tracking data however we continue to focus on comparisons of tracking data with other kiwi given their very distinctive biology (of which flightlessness is likely to be the biggest restriction to dispersal). Systematic location determination of radiotagged birds was restricted to the periods identified in the manuscript, accordingly we cannot compare the change in roost size area of the 2013 birds from 2013 to 2014 when the second release of birds occurred.
Additions:
"This finding is contrary to expectations as a number of other studies on volant terrestrial birds (Clarke & Schedvin 1997; Kemink & Kesler 2013) due to the higher exploratory behaviour expected in newly translocated birds."
"Monitoring over longer periods would likely give larger home ranges. Post-translocation movement and range coverage of roroa (great spotted kiwi, Apteryx maxima) took some six months to stabilise (Jahn et al. 2022b) with small incremental increases in range occurring for up to two years after translocations."
Reviewer 3 Report
This paper summarizes the findings of 15 translocated North Island Brown Kiwis, a Vulnerable species (IUCN). The paper is well written and highlights that home ranges of most individuals are centered around vegetation, although some individuals may disperse in pasture areas. The paper emphasizes the role of 'anchoring' to aid in limiting post-release dispersal and expediting territory formation. I enjoyed reading this rather important paper where the findings may have an applied aspect that could be used by managers. I think the following could be emphasized more:
1. the conservation status of the species and the reasons for its decline could be briefly mentioned in the Introduction.
2. the habitat characteristics of the release sites should be mentioned for the benefit of the international readers.
3. the dietary and habitat requirements of the species should be explained more in the discussion, to help link the findings to known habitat use patterns, for the readers. Some speculation here is ok.
I have provided these comments in the marked up document.
Aside from these relatively minor comments, this is a neat little paper with great conservation management value.
Comments for author File: 
Comments.pdf
Author Response
The authors thank the reviewer for their comments and we have provided the response to comments below and the markups in the attached document.
- Species information: The authors agree and have added the information on the biology of the species as paragraph one in the introduction.
Additional paragraph:
"North Island Brown kiwi (NIBK) (Apteryx mantelli) (Order: Struthioniformes, family: Apterygidae)(Bartlett, 1850) are nocturnal flightless ratites, which are endemic to New Zealand and are classified as ‘Vulnerable’ as per the IUCN Red List [1]. Once widespread in North Island, New Zealand, the decline of this ground-dwelling species has been primarily due to introduced predators, such as dogs, cats, rats and stoats but also the loss and fragmentation of habitat [2]. Kiwi are predominantly found in New Zealand native forest habitats and forage during the night consuming primarily invertebrates, earthworms, beetle larvae and small amounts of fruit and leaves (NZ Birds Online, 2013, Dixon 2015)[3,4]. During day-light hours they retreat to burrows preferably in either living or dead trees or holes in the ground in native forest habitats [5]."
2. Habitat description: The authors agree and a more in depth description has been provided.
Addition: "The property comprises a 400-hectare New Zealand native mixed coastal broadleaf/podocarp forest and salt marshland restoration area within a 1300-hectare private commercial pastoral sheep and beef farm (Figure 1)."
3. Map resolution: The authors agree, however, the relevant co-author is currently in the Philippines and is not within communication. This will be addressed as soon as possible in the New Year.
4. Kiwiback & influenceof: edits have been made
5. Invertebrates associated with habitat: This has been addressed in the discussion.
Addition: "Dixon [4] determined that Coleoptera were the most common invertebrate species found in faecal samples of brown kiwi (Apteryx mantelli) on Ponui Island and were selected in excess of availability. The author also determined that Coleoptera species were more common in forest and swamp habitats compared with scrub and pasture habitat. This, together with the selection of preferred roost locations in dead/live or ground burrows, supports the use of native forest as preferred habitat for brown kiwi. Dixon [4] also stated that the use of pasture habitats in summer months was related to the availability of black field crickets (Formicidae), which may explain the extensive use of pasture habitat by some kiwi in the current study. "
6. Burrow characteristics: this has been addressed in the discussion.
Addition: "Habitat selection has also been linked with availability of preferred burrow sites,which, according to Jamieson et al [5], are typically those associated with the base of living or dead trees or ground burrows in forest habitats. These authors reported that the preference was for long burrows with smaller entrance holes, which was likely to provide protection from the elements and the creation of a warm microclimate at the end of the burrows."
7. Removal of 'are': has been completed
