Preparing for translocations of a Critically Endangered through targeted monitoring of nest survival and breeding biology

J OHANNES H. FISCHER,HEIKO U. WITTMER,GRAEME A. TAYLOR I GOR D EBSKI and D OUG P. ARMSTRONG

Abstract The population of the recently-described Whenua Keywords Aotearoa , Bayesian inference, Hou Pelecanoides whenuahouensis comprises conservation, Pelecanoides whenuahouensis, phenology,  c.  adults that all breed in a single . km colony in , restoration, Whenua Hou diving petrel a dune system vulnerable to erosion. The species would Supplementary material for this article is available at therefore benefit from the establishment of a second breed- doi.org/./S ing population through a translocation. However, given the small size of the source population, it is essential that trans- locations are informed by carefully targeted monitoring data. We therefore modelled nest survival at the remaining Introduction population in relation to potential drivers (distance to sea and burrow density of conspecifics and a competitor) across eabirds, and in particular, are among the most three breeding seasons with varying climatic conditions as a Sthreatened taxa (Croxall et al., ; Dias et al., ). result of the southern oscillation cycle. We also documented Nearly half of all  petrel species (i.e. families Procellari- breeding phenology and burrow attendance, and measured idae, Oceanitidae, Hydrobatidae and Pelecanoididae) are chicks, to generate growth curves. We estimated egg survival threatened with . Petrel species are affected by a     at . , chick survival at . , overall nest survival wide range of threats (Dias et al., ). On land, petrels   at . , and found no indication that nest survival was are threatened by invasive predators (Jones et al., ; affected by distance to sea or burrow density. Whenua Dias et al., ), extreme weather events (Cole, ; Rodrí- Hou diving petrels laid eggs in mid October, eggs hatched guez et al., ) and light pollution (Rodríguez et al., ). in late November, and chicks fledged in mid January at At-sea threats include changes in oceanic productivity,  c. % of adult weight. Burrow attendance (i.e. feeds) climate patterns, and fisheries impacts such as bycatch     decreased from . to . visits per night as chicks and competition (Anderson et al., ; Zydelis et al., ; approached fledging. Nest survival and breeding biology Grémillet et al., ). Various life-history traits render pet- were largely consistent among years despite variation in rels disproportionally vulnerable: they are extremely wide- climate. Nest survival estimates will facilitate predictions ranging (i.e. they utilize entire ocean basins; Shaffer et al., about future population trends and suitability of prospective ), K-strategists (i.e. low fecundity, delayed sexual ma- translocation sites. Knowledge of breeding phenology will turity, high longevity; Rodríguez et al., ) and placed at inform the timing of collection of live chicks for transloca- high trophic levels (i.e. they are top predators; Einoder, tion, and patterns of burrow attendance combined with ). As petrels provide important ecosystem services growth curves will structure hand-rearing protocols. A tuhin- (e.g. nutrient cycling, bioturbation and seed dispersal; Ellis, ā ā ga whakar popoto (te reo M ori abstract) can be found in the ; Orwin et al., ; Otero et al., ), their conserva- Supplementary material. tion is a priority. Translocations are an increasingly common conserva- tion management strategy (Seddon et al., , ), including for petrels (Miskelly et al., ). A conservation

JOHANNES H. FISCHER (Corresponding author, orcid.org/0000-0003-3527- translocation entails the intentional movement of individ- 1671) School of Biological Sciences, Victoria University of Wellington, PO uals for species recovery or ecosystem restoration (Seddon Box 600, Wellington, 6140, New Zealand. E-mail [email protected] et al., ). Translocations may be effective conservation HEIKO U. WITTMER School of Biological Sciences, Victoria University of interventions if habitat is available outside a species’ current Wellington, Wellington, New Zealand range, if the species is unlikely to naturally colonize that GRAEME A. TAYLOR and IGOR DEBSKI Aquatic Unit, Department of Conservation, Wellington, New Zealand habitat, and if the translocation is unlikely to cause undesir- able impacts. Translocations may involve supporting exist- DOUG P. ARMSTRONG Wildlife Group, Massey University, Palmerston North, New Zealand ing populations (i.e. reinforcement), reinstating populations ’ Received  April . Revision requested  June . within the species indigenous range (i.e. reintroduction), or Accepted  July . creating new populations outside of the species’ indigenous

This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, Downloadeddistribution, from https://www.cambridge.org/core and reproduction in any medium,. IP address: provided 202.89.140.187 the original work, on is 29 properly Apr 2021 cited. at 05:53:28, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/termsOryx, Page 1 of 9 © The Author(s),. https://doi.org/10.1017/S0030605320000794 2021. Published by Cambridge University Press on behalf of Fauna & Flora International doi:10.1017/S0030605320000794 2 J. H. Fischer et al.

range (i.e. assisted colonization; IUCN, ). As petrels are ), but additional threats may still be inhibiting pop- often threatened and facilitate ecosystem functioning, their ulation recovery (Fischer et al., a). Unlike other petrels, translocation can be motivated by both species recovery and the species breeds exclusively in fragile foredunes ,  m restoration goals (Miskelly et al., ; Jones & Kress, ). from the springtide line (Fischer et al., c). Storms and For example, Gould’s petrels Pterodroma leucoptera have storm surges, as well as , may thus be the been translocated to Boondelbah Island, , within main threats to this species (Cole, ;Vousdoukasetal., their indigenous range to strengthen the small existing ). Competition with the more aggressive common diving population (reinforcement; Priddel et al., ). Common petrel for burrow sites may also inhibit population recovery diving petrels Pelecanoides urinatrix have been translocated (Fischer et al., ). An unsuccessful hybridization attempt to Mana Island, Aotearoa New Zealand, to reinstate the between a Whenua Hou diving petrel and a common diving ecosystem functions they once provided (reintroduction; petrel has been recorded (Fischer et al., c), suggesting Miskelly et al., ). additional pressures from this closely related species. As Poor understanding of the agents of decline can cause common diving petrels appear to be also attracted to Whenua translocation failure, including for petrels (Jones & Kress, Hou diving petrel calls, acoustic attraction systems may not ; Osborne & Seddon, ). Insights into the drivers be an option to establish new Whenua Hou diving petrel of nest survival are key for translocations. Many , colonies (Fischer et al., b). Therefore, a translocation including most petrels, are wide-ranging (e.g. Shaffer et al., of Whenua Hou diving petrels to a more suitable site ). Because of these wide foraging ranges, associated could help ensure long-term viability. Detailed information threats at sea are unlikely to be affected by translocation, on the factors affecting nest survival and breeding biology is as has been shown for short-tailed Phoebastria required to meaningfully assess site suitability and design albatrus (Deguchi et al., ; Orben et al., ). Under- translocation protocols. Such information was previously standing drivers of nest survival at source sites may thus unavailable. be key to predicting nest survival at potential translocation To inform future translocations, we monitored Whenua sites and, consequently, translocation success (Osborne & Hou diving petrel burrows across three breeding seasons Seddon, ). As nest survival in seabirds can be subject (–) with a burrowscope, stick palisades, and nest to interannual fluctuations driven by climatic conditions boxes. We aimed to quantify nest survival and its underly- (Chastel et al., ; Quillfeldt et al., ), multi-year stud- ing drivers. In addition, we documented breeding phenology ies of nest survival are critical. and patterns of burrow attendance (as a proxy for feeding Poor understanding of the breeding biology of the target regimes), and measured chicks, to generate growth curves. species is also a potential cause of translocation failure  (Jones & Kress, ). Petrels exhibit high philopatry and Study area their semi-precocial chicks are believed to imprint on their natal colonies prior to fledging (Priddel et al., ; The entire Whenua Hou diving petrel colony is restricted  Miskelly et al., ). Thus, the use of chicks, – weeks to a . km strip of coastal sand dunes on Whenua Hou prior to fledging, is required to successfully translocate (Codfish Island; Fig. ),  km off the west coast of Rakiura these species (Miskelly et al., ; Jones & Kress, ). (Stewart Island), Aotearoa New Zealand. This area holds As these chicks then need to be hand-reared at the trans- c.  Whenua Hou diving petrel burrows (Fischer et al., location site, detailed information on the breeding biology a). A small number of common diving petrels breed of the target species is essential to design protocols. For ex- within the study area (Fischer et al., , c). ample, data on breeding phenology (i.e. timing and duration of courtship, incubation, guard, and post-guard stages) will Methods inform translocation timing. Data on feeding regimes and chick growth curves will inform hand-rearing regimes. Nest survival The Critically Endangered Whenua Hou diving petrel Pelecanoides whenuahouensis is a recently-described bur- To quantify Whenua Hou diving petrel nest survival (i.e. rowing petrel species that could benefit from translocations egg, chick, and overall nest survival), we monitored – (Fischer et al., b,c). These were once widespread burrows (–% of the total population) from early throughout southern Aotearoa New Zealand, but follow- September to late January during the ,  and  ing local caused by invasive predators (e.g. rats breeding seasons (we report the year in which breeding Rattus spp.), the species only survives at a single location: commenced). Monitoring was conducted with a burrow- Whenua Hou (Codfish Island; Worthy, ;Holdawayet scope (Sextant Technologies, Wellington, Aotearoa New al., ; Wood & Briden, ). Here, only – adults Zealand; Plate ). Monitoring included daily checks of the remain in a single colony (Fischer et al., a). Invasive pre- burrows monitored for breeding phenology (see below) dators have been eradicated from Whenua Hou (McClelland, and weekly checks of all other burrows that allowed access.

Oryx, Page 2 of 9 © The Author(s), 2021. Published by Cambridge University Press on behalf of Fauna & Flora International doi:10.1017/S0030605320000794 Downloaded from https://www.cambridge.org/core. IP address: 202.89.140.187, on 29 Apr 2021 at 05:53:28, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605320000794 Preparing for translocations 3

FIG. 1 Distribution of Whenua Hou diving petrel Pelecanoides whenuahouensis (WHDP), Pelecanoides urinatrix (CDP) and mixed burrows, and nest boxes in ,  and .

Eggs were detected in  of the  monitored burrows in survival model allowed for () unknown transition and fail-  (apparent lay rate = .), in  of the  monitored ure dates, () varying lengths of phenological stages among burrows in  (apparent lay rate = .), and in  of nests, () estimation of daily survival rates for two pheno- the  monitored burrows in  (apparent lay rate = .), logical stages (eggs and chicks), and () the estimation of resulting in  Whenua Hou diving petrel burrows used fixed and random effects affecting the daily survival rate. in subsequent analyses. During each nest check, we record- Specifically, we fitted the data to a generalized linear ed the phenological stage (egg, chick or fledged) and fate mixed-effects model (GLMM) with a Bernoulli error term (dead or alive). These data were compiled in three capture and a logit-link function: history matrices showing whether each egg had hatched, = a + b × + b chick had fledged, and was alive ( = yes,  = no, NA = not logit(DSRi,j) DSR hatch hatchi,j sea × + b × + b checked) each day. We assumed nests to be alive on first disti WHDP WHDPi CDP detection but otherwise treated the fate of eggs as unknown × CDP + u (1) until a nest was either abandoned or the egg had hatched. i year,y We estimated nest survival using a multi-stage nest sur- where DSRi,j is the survival probability of nest i on day j, α β vival model within a Bayesian framework (Schmidt et al., DSR is the intercept, hatch is the effect of transitioning     ; Converse et al., ). Our custom, multi-stage nest from egg stage (hatchi,j = ) to chick stage (hatchi,j = ),

Oryx, Page 3 of 9 © The Author(s), 2021. Published by Cambridge University Press on behalf of Fauna & Flora International doi:10.1017/S0030605320000794 Downloaded from https://www.cambridge.org/core. IP address: 202.89.140.187, on 29 Apr 2021 at 05:53:28, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605320000794 4 J. H. Fischer et al.

density of other Whenua Hou diving petrel burrows within  a -m radius of the burrow (burrows per m ). Similarly, we measured CDPi as density of common diving petrel  and mixed burrows within  m (burrows per m ). We z-transformed these three variables. We fitted the model using the Bayesian updating soft- ware OpenBugs .. (Spiegelhalter et al., ). The Markov Chain Monte Carlo (MCMC) algorithms make it possible to account for multiple sources of uncertainty, such as survival probability, timing, and duration of phenology stages, which are propagated into posterior distributions for parameters. PLATE 1 A burrowscope being inserted into a Whenua Hou WepooledtwoMCMCchainsof, iterations, after a diving petrel burrow. burn-in of , iterations that was sufficient to give con- vergence based on Gelman–Rubin statistics (Rˆ , .). β β sea is the effect of distance to sea (disti), WHDP is the effect of Whenua Hou diving petrel burrow density (WHDPi), Breeding phenology β CDP is the effect of common diving petrel burrow density To quantify the timing and duration of Whenua Hou div- (CDPi), and uyear,y is the annual random effect. We then es- ing petrel breeding phenology, we monitored – bur- timated daily survival rate for the egg (DSRegg,y), and chick rows (–% of the total population) daily between early stage (DSRchick,y) per breeding season as: September and late January each breeding season. We logit(DSRegg,y) = aDSR + uyear,y (2) recorded arrival dates of birds based on when burrows = a + b + ( ) were dug out by birds (burrows close in winter because of logit(DSRchick,y) DSR hatch uyear,y 3 the movements of the dunes). We quantified the timing of Subsequently we estimated nest survival during the egg subsequent breeding phenology events, including dates for (Segg,y) and chick (Schick,y) stage per breeding season as: laying, hatching, commencement of post-guard phase (i.e. as the first day a chick was left unattended by an adult), = Tinc ( ) Segg,y DSRegg,y 4 and fledging, using a burrowscope. We monitored burrows = Trear ( ) daily until we recorded a breeding phenology event, after Schick,y DSRchick,y 5 which we ceased monitoring until a week before the next in which Tinc and Trear are the estimated mean durations anticipated event. We initially predicted the timing of these of the incubation and chick-rearing stages, respectively. events using published data on the closely related South Ultimately, we estimated overall nest survival per breeding Georgian diving petrel (Marchant & Higgins, ). We season (Sy) as: used the timing of phenology events to delineate the phe- nological stages of courtship, incubation, chick-rearing Sy = Segg,y × Schick,y (6) (consisting of guard and post-guard stages) and the total As we did not know exact dates of phenology events or breeding season, and to calculate the duration of these duration of stages for all nests, missing values for hatching stages. We assessed the influence of interannual variation and fledging status were inferred by modelling the duration on timing and duration of breeding phenology stages of each stage for each nest (Miller et al., ). We assumed using generalized linear models (GLMs) with a Gaussian these durations were normally distributed among nests with error distribution and an identity-link function. In these σ σ means Tinc and Trear and standard deviations inc and rear. GLMs we treated initiation date or duration per stage as We used mildly informative priors for these parameters: the response variable and year as the explanatory variable.      N[mean = , precision = . ] for Tinc and Trear,U[ , ] for We first transformed initiation dates into a numerical σ   σ  inc, and U[ , ] for rear. We based our priors on the breed- variable (i.e. days since September) and subsequently ing phenology of the closely related South Georgian diving z-transformed initiation dates and durations. petrel Pelecanoides georgicus (Marchant & Higgins ). α   We also used a mildly informative prior for DSR (N[ , ]) Burrow attendance β β β β but used vague priors for hatch, WHDP, CDP, and sea (N [, ]) and for the standard deviation of the random effect As a proxy for Whenua Hou diving petrel feeding regimes,   uyear,y (U[ , ]). we quantified burrow attendance (i.e. visits per day) per We measured disti as the distance from the Whenua phenological stage (i.e. courtship, incubation, guard and Hou diving petrel burrow to the highest springtide line post-guard stages) using stick palisades. We assessed the per breeding season (m). We measured WHDPi as the influence of interannual variation per stage using GLMs

Oryx, Page 4 of 9 © The Author(s), 2021. Published by Cambridge University Press on behalf of Fauna & Flora International doi:10.1017/S0030605320000794 Downloaded from https://www.cambridge.org/core. IP address: 202.89.140.187, on 29 Apr 2021 at 05:53:28, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605320000794 Preparing for translocations 5

with a quasi-binomial error distribution and a logit-link We increased our sample size by taking measurements function. In these GLMs we treated visit per day per stage from all chicks accessible within natural burrows (n = ). as the response variable and year as the explanatory variable. We also took measurements from all chicks caught just In addition, we assessed how burrow attendance during the before fledging (n = ). We compared chick measurements post-guard stage changed over time using GLMs with a with mean adult weight ( g; n = ) and mean adult binomial error distribution and a logit-link function treating length ( mm; n = ). visits per day as the response variable and age (expressed as days-before-fledging) as the explanatory variable. We could not account for double feeds (i.e. both parents feeding the Results chick) using our stick palisade method. Thus, burrow atten- dance should be considered only a proxy for feeding regimes. Nest survival

Chick growth curves The daily survival rate of Whenua Hou diving petrel eggs was estimated to be . (% credible intervals = .– To generate Whenua Hou diving petrel chick growth curves .; Fig. a). The daily survival rate of chicks was esti- (i.e. wing length and weight), we monitored  burrows mated to be . (.–.). The duration of the incu-  between early December and late January each breeding bation stage was estimated to be Tinc = . (.–.) days season. To access chicks inside burrows, we installed  and the duration of the chick-rearing stage was estimated to  custom multi-storey nest boxes in existing burrows in be Trear = . (.–.) days. Egg survival was estimated early September , before birds returned (Fig. , Supple- to be . (.–.). Chick survival was estimated to be mentary Fig. ; Fischer et al., a). We selected burrows for . (.–.; Fig. b). Nest survival from laying to nest box instalment if burrows belonged to successful breed- fledging was estimated to be . (.–.; Fig. c). ers in  and/or , brood chambers had a depth of Mean distance to sea was . m (range = .–.), mean ,  cm, and burrows were .  m from the springtide Whenua Hou diving petrel density was . burrows per  line. We subjected chicks in nest boxes to daily measure- m (.–.) and mean common diving petrel density ments of weight (g) and wing length (i.e. flattened wing was . burrows per m (.–.). Estimates and % b chord; mm) once they reached the post-guard stage until credible intervals for the effects of distance to sea ( sea = they fledged. Only four chicks fledged from a nest box −.; −., .), density of Whenua Hou diving petrel b −  −    (most pairs dug new brood chambers behind nest boxes). burrows ( WHDP = . ; . , . ), and density of

FIG. 2 Estimates and % credible intervals for (a) daily survival rates of Whenua Hou diving petrel eggs and chicks, (b) probabilities of surviving the eggs and chick stages, (c) overall nest survival, and (d) slopes (β)of z-transformed covariates affecting the logit of daily nest survival in the breeding seasons of ,  and . Sea, effect of distance to sea; WHDP, effect of Whenua Hou diving petrel burrow density; CDP, effect of common diving petrel burrow density.

Oryx, Page 5 of 9 © The Author(s), 2021. Published by Cambridge University Press on behalf of Fauna & Flora International doi:10.1017/S0030605320000794 Downloaded from https://www.cambridge.org/core. IP address: 202.89.140.187, on 29 Apr 2021 at 05:53:28, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605320000794 6 J. H. Fischer et al.

The average duration of breeding stages was as follows: courtship: . days; incubation: . days; chick guard stage: . days; post-guard stage: . days; resulting in . days for the total chick-rearing period. The total breeding season lasted . days. Duration of courtship stages, incubation stages, and total breeding periods varied slightly among breeding seasons. The courtship and incubation periods lasted slightly longer in  (estimates ± SE: β = −. ± ., β = . ± . and β = −. ± ., β = . ± ., respectively), resulting in a longer breeding season β   ±   β   ±   FIG. 3 Whenua Hou diving petrel phenology in the breeding ( = . . , = . . ). Durations of the guard seasons of ,  and . and post-guard stages were consistent among breeding seasons (β = . ± ., β = . ± . and β = −. ± ., β = . ± ., respectively). b   −    common diving petrel burrows ( CDP = . ; . , . ) did not indicate a clear impact on nest survival (Fig. d). There was also no apparent annual variation in survival. Burrow attendance Twenty-four abandoned eggs were extracted from burrows Whenua Hou diving petrel burrow attendance was not   to assess fertility, and of these ( %) were infertile. uniform throughout the breeding season. Burrow atten- dance was lower during incubation (. visits per day), com-   Breeding phenology pared to burrow attendance during courtship ( . ), guard (.), and post-guard (.) stages (Fig. a). Burrow atten- On average, Whenua Hou diving petrels arrived at the col- dance per stage varied slightly among seasons. Specifically, ony on  September, eggs were laid on  October, chicks burrow attendance during courtship was higher in  hatched on  November, post-guard stage commenced (estimates ± SE: β = . ± ., β = −. ± .), on  December,andfledgingoccurredon January (Fig. ). burrow attendance during incubation was higher in  Phenology events were protracted and non-synchronous (β = . ± ., β = . ± .), burrow attendance among burrows. Timing of breeding phenology varied during guard was lower in  (β = . ± ., β = slightly among breeding seasons. Specifically, arrival . ± .), and burrow attendance during post-guard was occurred slightly earlier in  (estimates ± SE: β = higher in  (β = −. ± ., β = −. ± .). −. ± ., β = −. ± .), laying and hatching During the post-guard stage, visitation rates decreased occurred slightly later in  (β = −. ± ., β = over time from . visits per day in the early post-guard . ± . and β = −. ± ., β = . ± ., stage (– days-before-fledging) to . visits per day respectively), post-guard commenced slightly earlier in during the last week before fledging (Fig. b). This decrease  (β = −. ± ., β = . ± .), and fledging in visitation rates was more pronounced in  (β = −. ± occurred slightly later in  (β = −. ± ., β = .)and (β = −. ± .)thanin (β = −. ± . ± .). .).

FIG. 4 Whenua Hou diving petrel burrow attendance across the breeding season (a), burrow attendance changes during the post-guard stage illustrated with generalized linear models (b), and chick growth curves of wing length (c) and weight (d) as illustrated by locally estimated scatterplot smoother curves.

Oryx, Page 6 of 9 © The Author(s), 2021. Published by Cambridge University Press on behalf of Fauna & Flora International doi:10.1017/S0030605320000794 Downloaded from https://www.cambridge.org/core. IP address: 202.89.140.187, on 29 Apr 2021 at 05:53:28, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605320000794 Preparing for translocations 7

Chick growth curves translocations should have a wing length of – mm and a weight of .  g, as this combination will ensure the Wing lengths of Whenua Hou diving petrel chicks showed, selection of healthy chicks at – days before fledging.  on average, a gradual and consistent growth from mm at Fourthly, although the nest boxes were unpopular with   days before fledging until approaching a plateau of mm adult Whenua Hou diving petrels, four chicks have fledged    ( %ofadultmean)around days before fledging (Fig. c). successfully from these boxes. Nest boxes did not appear to   Maximum recorded wing length was mm ( %ofadult influence nest survival (/ nest attempts inside nest boxes  mean). Weight of chicks, on average, increased from g were successful). Since these nest boxes are designed specif-    at days before fledging to a maximum of g( %of ically for this species (Fischer et al., a), chicks can fledge   adult mean) between and days before fledging and sub- successfully from them, and as access to chicks is crucial,   sequently decreased to g( %ofadultmean)around the use of these nest boxes appears invaluable at future  fledging (Fig. d). Maximum recorded chick weight was translocation sites.   g( %ofadultmean). Fifthly, our findings on burrow attendance and chick growth can inform post translocation feeding regimes for Discussion Whenua Hou diving petrel chicks. At the translocation site, chicks should be fed daily until – days before fledging Our detailed study of nest survival and breeding biology to ensure chicks remain above mean adult weight. When has the potential to inform future translocations to estab- translocated chicks approach fledging, feeds should be lish a new breeding colony of the Critically Endangered slowly reduced to every second day, provided chicks are Whenua Hou diving petrel. Firstly, by applying a novel at, or above, mean weight (Miskelly et al., ). We did Bayesian multi-stage nest survival model, we provide not obtain species-specific information on meal size or insights into nest survival. Estimates of demographic para- diet. However, common diving petrel chicks have been meters such as nest survival are of vital importance when successfully hand-reared using average meal sizes of  g making structured decisions on conservation management (range –; Miskelly et al., ). As common diving (including translocations) in the face of uncertainty (Panfy- petrels have similar adult weights (– g), these meal lova et al., ). For wide-ranging species such as petrels, sizes may also be appropriate for Whenua Hou diving petrel nest survival, not juvenile or adult survival, will be the likely chicks. Miskelly et al. () and Miskelly & Gummer () driver of translocation success, making nest survival esti- have shown that petrel species will thrive on a diet of pureed mates crucial for projecting future population trajectories sardines, regardless of their natural diet, and thus this diet at translocation sites. In addition, our estimates of the effects may also be suitable for this species. of parameters on nest survival are important for assessing Whenua Hou diving petrel nest survival and breeding the suitability of translocation sites. Although no para- biology was largely unaffected by annual variation and there- meters showed a clear impact on nest survival, these may fore varying climatic conditions. The three breeding sea- have an influence at different exposure levels (e.g. nest sur- sons in our study encompassed a range of climatic (El Niño vival may have been affected at higher common diving Southern Oscillation; ENSO) conditions: Oceanic Niño Index petrel burrow densities). As such, these parameters should was −. (La Niña) in , . in  (El Niño), and . still be considered when assessing translocation site suit- (approaching ENSO neutral) in  (NOAA, ). The ability (Fischer et al., ). slight differences in timing and duration of phenology ap- Secondly, our findings on phenology, chick growth and peared unrelated to ENSO conditions. The breeding season nest survival help determine the ideal timeframe for collect- under neutral conditions () was delayed and prolonged ing live chicks for translocations. As petrel chicks should be compared to other breeding seasons under more extreme collected – weeks prior to fledging to prevent imprinting and varying conditions. The apparent phenological insensi- on the natal colony (Miskelly & Taylor, ; Miskelly et tivity of Whenua Hou diving petrels to climatic conditions al., ), collection of Whenua Hou diving petrel chicks is not surprising as it mirrors insensitivity observed in should occur between early December and early January. many other seabirds (Keogan et al., ). Climatic variables, As chicks reach their maximum weight – days before however, could have influenced burrow attendance (Quill- fledging and may thus be less susceptible to stress associated feldt et al., ). In  (La Niña), burrow attendance was with translocations, the last week of December appears the higher during the post-guard stage. Food supplies can change ideal time to collect chicks of this species. with climatic conditions (Schreiber & Schreiber, ). Thus, Thirdly, our growth curves will help select the best suited adults in  may have spent more time provisioning their Whenua Hou diving petrel chicks for future translocations. chicks during the post-guard stage (Chastel et al., ). Wing length combined with weight facilitates the estimation of The Whenua Hou diving petrel population is extremely age and condition of chicks (in days before fledging; Miskelly small (– adults; Fischer et al., a) and room for &Taylor,; Miskelly et al., ). Chicks selected for error in management is slim. Further research is required

Oryx, Page 7 of 9 © The Author(s), 2021. Published by Cambridge University Press on behalf of Fauna & Flora International doi:10.1017/S0030605320000794 Downloaded from https://www.cambridge.org/core. IP address: 202.89.140.187, on 29 Apr 2021 at 05:53:28, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605320000794 8 J. H. Fischer et al.

prior to attempting translocations of this species. The institutional ethics committee (VUW AEC 23283 and VUW appropriate cohort size and number of cohorts used for AEC 27621), the Department of Conservation (M1718/01, M1819/ ō ū translocations must be estimated to minimize risks to the 01, and M1920/02), the Whenua Hou Komiti, and Kaitiaki R p . source population. The impact of translocations on the source colony could be quantified using population models. References We provided a key parameter, nest survival, for such mod- els, but juvenile and adult survival remain unknown and ANDERSON, O.R.J., SMALL, C.J., CROXALL, J.P., DUNN, E.K., are currently under investigation (Fischer, ). Further- SULLIVAN, B.J., YATES,O.&BLACK,A.() Global more, infertility appeared a prevalent cause of egg failure. bycatch in longline fisheries. Endangered Species Research, , –. Although common diving petrels can have a similar rate of CHASTEL, O., WEIMERSKIRCH,H.&JOUVENTIN,P.() Body infertility (Richdale, ), the Whenua Hou diving petrel condition and seabird reproductive performance: a study of population probably suffered from a population bottleneck three petrel species. Ecology, , –. and represents a fraction of the historic genetic diversity COLE,R.() Summary of South Georgian Diving Petrel Field   (Wood & Briden, ). As such, the selection of chicks Observations for / , Codfish Island/Whenua Hou. Department of Conservation, Invercargill, New Zealand. for future translocations may benefit from including a CONVERSE, S.J., ROYLE, A., ADLER, P.H., URBANEK, R.P. & BARZEN, measure of genetic diversity. J.A. () A hierarchical nest survival model integrating incomplete Translocations are a useful tool to combat the ongoing temporally varying covariates. Ecology and , , –. sixth mass extinction by restoring species and ecosystem CROXALL, J.P., BUTCHART, S.E.M., LASCELLES, B., STATTERSFIELD,  functioning (Seddon et al., , ). Translocations of A.J., SULLIVAN, B., SYMES,A.&TAYLOR,P.( ) Seabird conservation status, threats and priority actions: a global petrels fit both conservation and restoration goals (Miskelly  –  assessment. Conservation International, , . et al., ). We have provided most of the data required DEGUCHI, T., SURYAN, R.M., OZAKI, K., JACOBS, J.F., SATO,F., to inform future translocations. The Whenua Hou diving NAKAMURA,N.&BALGH, G.R. () Translocation and petrel is the only petrel species in Aotearoa New Zealand hand-rearing of the short-tailed Phoebastria albatrus: early that breeds en masse in coastal dunes (Worthy, ) and indicators of success for species conservation and island restoration.  – is thus considered an ecosystem engineer (Fischer et al., Oryx, , .   DIAS, M.P., MARTIN, R., PEARMAIN, E.J., BURFIELD, I.E., SMALL, C., ). The Predator Free programme (Russell et al., PHILLIPS, R.A. et al. () Threats to seabirds: a global assessment. ) aims to eradicate seven species of invasive mammals Biological Conservation, , –. from all of Aotearoa New Zealand by . If this pro- EINODER, L.D. () A review of the use of seabirds as indicators in gramme is successful, more habitat could become available fisheries and ecosystem management. Fisheries Research, , –.  for potential Whenua Hou diving petrel translocations. The ELLIS, J.C. ( ) Marine birds on land: a review of plant biomass, species richness, and community composition in seabird colonies. information provided here may facilitate not only the con- Plant Ecology, , –. servation of a Critically Endangered species, but also the FISCHER, J.H. () Integrated conservation of the Whenua Hou restoration of ecosystem function to a threatened habitat diving petrel. PhD thesis. Victoria University of Wellington, throughout Southern Aotearoa New Zealand. Wellington, New Zealand. FISCHER, J.H., CHAMBON, J., DEBSKI, I., HISCOCK, J.A., COLE, R., Acknowledgements JHF was supported by a Victoria University of TAYLOR, G.A. & WITTMER, H.U. (a) Buffering artificial nest Wellington Doctoral Scholarship. Fieldwork was supported by the boxes for Procellariiformes breeding in exposed habitats: investigating National Geographic Society (WW-249C-17), the Ornithological effects on temperature and humidity. Notornis, , –. Society of New Zealand (Bird NZ Research Fund 2017, 2019), the FISCHER, J.H., DEBSKI, I., MISKELLY, C.M., TENNYSON, A.J.D., Mohamed Bin Zayed Species Conservation Fund (Project 192520234), FROMANT, A., TESSLER, J. et al. (b) Analyses of phenotypic the Encounter Foundation, Forest and Bird (JS Watson Trust 2017), differentiations between South Georgian Diving Petrel (Pelecanoides the Centre for and Restoration Ecology (Student Award georgicus) populations reveal an undescribed and highly- 2017), and the Royal Society of New Zealand (Hutton Fund 2017). endangered species from New Zealand. PLOS ONE, ,e. We thank Ngāi Tahu, the Whenua Hou Komiti, and Kaitiaki Rōpū FISCHER, J.H., DEBSKI, I., TAYLOR, G.A. & WITTMER, H.U. (c) for granting visits to Whenua Hou; the dozens of volunteers for their Nest site selection of South Georgia diving-petrels on Codfish assistance in the field; Brooke Tucker for archaeological insights; Island (Whenua Hou), New Zealand: implications for conservation Johannes Chambon, Jake Tessler and Dominique Filippi for technical management. International, , –. support; Janne de Hoop, David Young and Johannes Chambon for FISCHER, J.H., DEBSKI, I., TAYLOR, G.A. & WITTMER, H.U. () providing artwork; and Ben Dilley, Colin Miskelly, Jason Preble and Assessing the suitability of non-invasive methods to monitor Grace Tocker for their comments. interspecific interactions and breeding biology of the South Georgian diving petrel (Pelecanoides georgicus). Notornis, , –. Author contributions Study design, fieldwork: all authors; data FISCHER, J.H., MCCAULEY, C.F., ARMSTRONG, D.P., DEBSKI,I.& analysis, writing: JHF, HUW, DPA. WITTMER, H.U. () Contrasting responses of lizard occurrences to burrowing by a Critically Endangered seabird. Community Conflicts of interest None. Ecology, , –. FISCHER, J.H., TAYLOR, G.A., COLE, R., DEBSKI, I., ARMSTRONG, D.P. Ethical standards This research abided by the Oryx guidelines on &WITTMER, H.U. (a) Population growth estimates of a ethical standards. All methods and protocols were approved by an threatened seabird indicate necessity for additional management

Oryx, Page 8 of 9 © The Author(s), 2021. Published by Cambridge University Press on behalf of Fauna & Flora International doi:10.1017/S0030605320000794 Downloaded from https://www.cambridge.org/core. IP address: 202.89.140.187, on 29 Apr 2021 at 05:53:28, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605320000794 Preparing for translocations 9

following invasive predator eradications. Animal Conservation, Management (edsJ.G.Even,D.P.Armstrong,K.A.Parker& , –. P.J. Seddon), pp. –. Blackwell Publishing, Oxford, UK. FISCHER, J.H., TAYLOR, G.A, DEBSKI,I.&WITTMER, H.U. (b) OTERO, X.L., DE LA PEÑA-LASTRA, S., PÉREZ-ALBERTI, A., FERREIRA, Acoustic attraction system draws in competing seabird species. T.O. & HUERTA-DIAZ, M.A. () Seabird colonies as important Notornis, , –. drivers in the nitrogen and phosphorus cycles. Nature GRÉMILLET, D., PONCHON, A., PALECZNY, M., PALOMARES, M.L.D., Communications, , . KARPOUZI,V.&PAULY,D.() Persisting worldwide PANFYLOVA, J., EWEN, J.G. & ARMSTRONG D.P. () Making seabird-fishery competition despite seabird community decline. structured decisions for reintroduced populations in the face Current Biology, , –. of uncertainty. Conservation Science and Practice, ,e. HOLDAWAY, R.N., JONES, M.D. & ATHFIELD, N.R.B. () PRIDDEL, D., CARLILE,N.&WHEELER,R.() Establishment of Establishment and extinction of a population of South Georgian a new breeding colony of Gould’s Petrel (Pterodroma leucoptera diving petrel (Pelecanoides georgicus) at Mason Bay, Steward Island, leucoptera) through the creation of artificial nesting habitat New Zealand, during the late Holocene. Journal of the Royal Society and the translocation of nestlings. Biological Conservation, of New Zealand, , –. , –. IUCN () Guidelines for Reintroductions and other Conservation QUILLFELDT, P., STRANGE, I.J. & MASELLO, J.F. () Sea surface Translocations. Version .. Species Survival Commission, Gland, temperatures and behavioural buffering capacity in thin-billed Switzerland. prions belcheri: breeding success, provisioning and JONES, H.P. & KRESS, S.W. () A review of the world’s active seabird chick begging. Journal of Avian Biology, , –. restoration projects. Journal of Wildlife Management, , –. RICHDALE, L.E. () Supplementary notes on the diving petrel. JONES, H.P., TERSHY, B.R., ZAVALETA, E.S., CROLL, D.A., KEITT, B.S., Transactions of the Royal Society of New Zealand, , –. FINKELSTEIN, M.E. & HOWALD, G.R. () Severity of the effects RODRÍGUEZ, A., ARCOS, J.M., BRETAGNOLLE, V., DIAS, M.P., of invasive rats on seabirds: a global review. Conservation Biology, HOLMES, N.D., LOUZAO, M. et al. () Future directions in , –. conservation research on petrels and . Frontiers in KEOGAN, K., DAUNT, F., WANLESS, S., PHILLIPS, R.A., WALLING, C., Marine Science, , . AGNEW, P. et al. () Global phenological insensitivity to RODRÍGUEZ, A., HOLMES, N.D., RYAN, P.G., WILSON, K.J., shifting ocean temperatures among seabirds. Nature Climate FAULQUIER, L., MURILLO, Y. et al. () Seabird mortality Change, , –. induced by land-based artificial lights. Conservation Biology, MARCHANT,S.&HIGGINS, P.J. () Handbook of Australian, New , –. Zealand and Antarctic Birds. Volume , Ratites to , Part A, RUSSELL, J.C., INNES, J.G., BROWN, P.H. & BYROM, A.E. () Ratites to Petrels. Oxford University Press, Melbourne, Australia. Predator-free New Zealand: conservation country. BioScience, MCCLELLAND, P.J. () Eradication of Pacific rats (Rattus exulans) , –. from Whenua Hou Nature Reserve (Codfish Island), Putauhinu and SCHMIDT, J.H., WALKER, J.A., LINDBERG, M.S., JOHNSON, D.S. & Rarotoka Islands, New Zealand. Turning the Tide: The Eradication STEPHENS, S.E. () A general Bayesian hierarchical model for of Invasive Species, , –. estimating survival of nests and young. , , –. MILLER, M.W., L EECH,D.I.,PEARCE-HIGGINS,J.W.&ROBINSON, SCHREIBER, R.W. & SCHREIBER, E.A. () Central Pacific seabirds R.A. () Multi-state, multi-stage modelling of nest success suggests and the El Niño Southern Oscillation:  to  perspectives. interaction between weather and land-use. Ecology, , –. Science, , –. MISKELLY, C.M. & GUMMER,H.() Attempts to anchor pelagic SEDDON, P.J., ARMSTRONG, D.P. & MALONY, R.F. () Developing fairy prions (Pachyptila turtur) to their release site on Mana Island. the science of reintroduction biology. Conservation Biology, Notornis, , –. , –. MISKELLY, C.M. & TAYLOR, G.A. () Establishment of a colony of SEDDON, P.J., GRIFFITHS, C.J., SOORAE, P.S. & ARMSTRONG, D.P. common diving petrels (Pelecanoides urinatrix) by chick transfers () Reversing defaunation: restoring species in a changing world. and acoustic attraction. Emu, , –. Science, , –. MISKELLY, C.M., TAYLOR, G.A., GUMMER,H.&WILLIAMS,R.() SHAFFER, S.A., TREMBLAY, Y., WEIMERSKIRCH, H., SCOTT, D., Translocations of eight species of burrow-nesting seabirds (genera THOMPSON, D.R., SAGAR, P.M. et al. () Migratory shearwaters Pterodroma, Pelecanoides, Pachyptila and : family integrate oceanic resources across the Pacific Ocean in an endless ). Biological Conservation, , –. summer. Proceedings of the National Academy of Sciences USA, NOAA () National Oceanic and Atmospheric Administration , –. Climate Prediction Database Repository. origin.cpc.ncep.noaa.gov/ SPIEGELHALTER, D., THOMAS, A., BEST,N.&LUNN,D.() products/analysis_monitoring/ensostuff/ONI_v.php [accessed OpenBUGS User Manual, Version . Bugs, Cambridge, UK.  April ]. VOUSDOUKAS, M.I., RANASINGHE, R., MENTASCHI, L., PLOMARITIS, ORBEN, R.A., O’CONNOR, A.J., SURYAN, R.M., OZAKI, K., SATO,F.& T.A., ATHANASIOU, P., LUIJENDIJK,A.&FEYEN,L.() Sandy DEGUCHI,T.() Ontogenic changes in at-sea distributions of coastlines under threat of erosion. Nature Climate Change, immature short-tailed albatrosses Phoebastria albatrus. Endangered , –. Species Research, , –. WOOD, J.R. & BRIDEN,S.() South Georgian diving petrel ORWIN, K.H., WARDLE, D.A., TOWNS, D.R., JOHN, M.G., (Pelecanoides georgicus) bones from a Maori midden in Otago BELLINGHAM, P.J. & JONES,C.() Burrowing seabird effects Peninsula, New Zealand. Notornis, , –. on invertebrate communities in soil and litter are dominated by WORTHY, T.H. () indicate Pelecanoides georgicus had ecosystem engineering rather than nutrient addition. Oecologia, large colonies at Mason Bay, Stewart Island, New Zealand. , –. Notornis, , –. OSBORNE, P.E. & SEDDON, P.J. () Selecting suitable habitats for ŽYDELIS, R., SMALL,C.&FRENCH,G.() The incidental catch of reintroductions: variation, change and the role of species distribution seabirds in gillnet fisheries: a global review. Biological Conservation, modelling. In Reintroduction Biology: Integrating Science and , –.

Oryx, Page 9 of 9 © The Author(s), 2021. Published by Cambridge University Press on behalf of Fauna & Flora International doi:10.1017/S0030605320000794 Downloaded from https://www.cambridge.org/core. IP address: 202.89.140.187, on 29 Apr 2021 at 05:53:28, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605320000794