Fecal Glucocorticoid Metabolite Responses of Brown Kiwi (Apteryx mantelli) to Ambassador Program Participation and Translocation: Implications for Captive Management and Welfare

Abstract

The brown kiwi (Apteryx mantelli) is a flightless, nocturnal bird native to New Zealand and is classified as "At Risk" due to predation from domestic dogs and ferrets. In the U.S., brown kiwi have been managed under the AZA Animal Population Management and Studbook Program since 2006, with the Smithsonian National Zoological Park (NZP) and Conservation Biology Institute (SCBI) maintaining the species since 1968. However, because they are nocturnal, kiwi are not good exhibit animals and often are difficult for zoo visitors to observe during the day. To address this, the NZP launched a "Meet-A-Kiwi" ambassador program in 1989 to engage the public and raise awareness. The program ran successfully for 28 years until 2017, when renovations closed the Bird House at the zoo in Washington, DC, and the birds were moved to the Front Royal, VA campus. Over time, concerns about the suitability of kiwi as ambassador birds arose as they are nocturnal and do not imprint on people. This study assessed the effects of outreach events, housing, and relocation on adrenal activity in five kiwi (four males, one female) by measuring fecal glucocorticoid metabolite (fGCM) concentrations, a physiological stress indicator, from March to October 2016. Two males participated in outreach (ambassadors), while two males and one female did not (controls). The results showed no significant differences in fGCM concentrations between ambassador and control birds, suggesting that outreach did not cause undue stress. However, individual factors (age, sex, hatching type, and display status) were associated with differences in fGCM concentrations, highlighting the need for personalized management. Further longitudinal studies are needed to explore the physiological responses of kiwi to captive conditions.

Keywords: corticosterone; education; physiology; welfare; zoo.

Figure 1

Longitudinal fecal glucocorticoid metabolite (fGCM) concentrations for ambassador bird NZP#1. NZP#1 was not translocated from NZP to SCBI on 18 May 2017. The red arrow denotes the fecal sample collected the day after translocation. Color markers denote ambassador events for NZP#1 (red: “Meet-A-Kiwi”; green: “Meet-and-Greet”; black: no event). The dotted line represents the hormone baseline calculated from the R package HormLong [44] using 1.5 standard deviations.

Figure 2

Longitudinal fecal glucocorticoid metabolite (fGCM) concentrations for control bird NZP#2. NZP#2 was translocated from NZP to SCBI on 18 May 2017. The red arrow denotes the fecal sample collected the day after translocation. Color markers denote noise levels noted by keepers at the end of the day for NZP birds (black: level 1, low sounds; blue: level 2, loud sounds; red: level 3, very loud sounds). The dotted line represents the hormone baseline calculated from the R package HormLong [44] using 1.5 standard deviations.

Figure 3

Longitudinal fecal glucocorticoid metabolite (fGCM) concentrations for control bird NZP#3. NZP#3 was translocated from NZP to SCBI on 18 May 2017. The red arrow denotes the fecal sample collected the day after the translocation event. The dotted line represents the hormone baseline calculated from the R package HormLong [44] using 1.5 standard deviations.

Figure 4

Longitudinal fecal glucocorticoid metabolite (fGCM) concentrations for ambassador bird SCBI#1. Color markers denote ambassador events for SCBI#1 (black: no event; red: demo event occurred). There was predator activity at the end of the study period that is noted in the profile. The dotted line represents the hormone baseline calculated from the R package, HormLong [44] using 1.5 standard deviations.

Figure 5

Longitudinal fecal glucocorticoid metabolite (fGCM) concentrations for control bird SCBI#2. SCBI#2 was the only female in the study. The dotted line represents the hormone baseline calculated from the R package HormLong [44] using 1.5 standard deviations.

Figure 6

Effect of the sample type (AM, DEMO, and PM) on the mean fGCM concentrations of ambassador bird NZP#1. Within ambassador birds, the Friedman test revealed significant differences in fGCM concentrations for NZP#1 (p < 0.001) in the sample type. The post hoc analysis showed significantly higher fGCM concentrations during AM collection compared to DEMO and PM samples for NZP#1. a, b and c denote statistical significance at the 0.05 level. * is a high data point.

Figure 7

Number of attendees per event vs. fGCM concentrations for ambassador NZP#1. Meet-and-Greet group sizes ranged from 2 to 10 people, whereas demo audiences ranged from 8 to 100 people. An outlier of 100 people was removed and not shown.

Figure 8

Effect of the number of events on the weekly mean fGCM concentrations for ambassador birds (a) NZP#1 and (b) SCBI#1. A pairwise comparison of fixed effects revealed significant differences between the specific number of events for NZP#1 (0 vs. 4 events, p = 0.021) and SCBI#1 (1 vs. 2 events, p = 0.047), with the mean fCGM concentrations decreasing as the event number increased for both birds.

Figure 9

Effect of the display status on the mean fGCM concentrations for NZP birds. The Mann–Whitney U test revealed that birds on public display (pink bar) had significantly lower fGCM concentrations than off-display birds (green bar) (U = 9101, z = −2.34, p = 0.019). However, this difference was primarily driven by NZP#2, as the removal of this bird eliminated the significant difference. NZP#2 is included in this graphic. a and b denote statistical significance at the 0.05 level.

Figure 10

Effect of the hatching status on fGCM concentrations. Wild hatches (pink bar) included NZP#2 and SCBI#1, whereas captive hatches (green bar) included NZP#1, NZP#3, and SCBI#2. The Mann–Whitney U test indicated significantly higher median fGCM levels in wild-hatched birds (430.99 ng/g dry feces) compared to captive-hatched birds (182.18 ng/g dry feces) (U = 81,737, z = 13.605, p < 0.001). a and b denote statistical significance at the 0.05 level.

Figure 11

Effects of translocation on the fGCM concentrations of NZP birds. The Kruskal–Wallis H test demonstrated that (b) NZP#2 exhibited a significant decrease in fGCM concentrations after moving to SCBI (p < 0.001), while the other relocated bird (a) NZP#3 did not show significant changes (p = 0.547). The translocation control bird (c) NZP#1, which remained at NZP, also showed no significant changes (p = 0.202). a and b denote statistical significance at the 0.05 level. * are high data points.

Figure 12

Effects of noise levels on the mean fGCM concentrations of NZP birds (N = 3). A one-way Welch ANOVA found significant differences in fGCM concentrations across noise levels (F(3, 32.062) = 2.998, p = 0.045). The post hoc analysis showed a significant increase in fGCM concentrations between noise levels 1 and 2 (p = 0.032). a and b denote statistical significance at the 0.05 level.