Breeding Habitat Survey of Blackthroat (Luscinia obscura) in Foping Nature Reserve of Qinling Mountains Author(s): Hongfeng Zhao, Ning Han, Lei Luo, Leigang Zhao and Xuebin Gao Source: Zoological Science, 31(5):279-282. 2014. Published By: Zoological Society of Japan DOI: http://dx.doi.org/10.2108/zs130136 URL: http://www.bioone.org/doi/full/10.2108/zs130136

BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published by nonprofit societies, associations, museums, institutions, and presses. Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use. Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder.

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research.

¤ 2014 Zoological Society of Japan

ZOOLOGICAL SCIENCE 31: 279–282 (2014)

Breeding Habitat Survey of Blackthroat (Luscinia obscura) in Foping Nature Reserve of Qinling Mountains Hongfeng Zhao1, Ning Han2, Lei Luo2, Leigang Zhao3, and Xuebin Gao2* 1

College of Life Sciences, Shaanxi Normal University, Xi’an 710062 2 Shaanxi Institute of Zoology, Xi’an 710032 3 Foping National Nature Reserve, Foping 723400

We carried out a primary survey by quadrat sampling to quantify breeding habitat characteristics of blackthroat (Luscnia obscura), a poorly documented and vulnerable bird species, in Foping nature reserve on the central southern Qinling Mountains. Tree layer information was collected in 10 m × 10 m plots, bamboo and shrub layer information was collected in 2 m × 2 m plots, and grass and ground layer information was collected in 1 m × 1 m plots. Our observations showed that blackthroat lives in coniferous forest and coniferous broadleaved forest with dense bamboos at the elevation ranging from 2130 m to 2600 m. Shrub cover, density and height, and ground cover in sites with blackthroat were significantly higher than those in random sites, while tree density, grass height and cover in habitat sites were significantly lower than those in random sites. These results suggest that shrub cover, shrub density, and canopy cover may be relevant to breeding habitat selection by these birds. Our study suggests that Qinling Mountains may be an important blackthroat breeding site. Key words: Blackthroat, habitat characteristics, breeding site, bamboo, Qinling

INTRODUCTION The blackthroat Luscina obscura is a small robin (13 cm), the male of which is characterized by coal-black throat and breast, with lustrous blue head, nape, and back (Zhao, 2001; Qian and Yi, 2011). However, this species is poorly documented, with only a small number of individuals (about 20) recorded in more than 120 years since its discovery in 1891 (BirdLife International, 2001, 2013). Its species distribution, habitat, and elevation range remain poorly understood, although it has been assumed to breed in southwestern China, including southern Gansu, north-central Sichuan, and southern Shaanxi. The species is assessed as vulnerable, with forest habitat fragmentation and loss being the main threats (BirdLife International, 2013). There was only a single record of blackthroat in Qinling Mountains, in 1905 (Harter, 1907), before it was rediscovered in Foping and Changqing nature reserves in 2011 on the south slope of Qinling. In total, 14 male birds were seen in both Foping and Changqing nature reserves in 2011, and these were found in large, dense expanses of bamboo in open coniferous and mixed coniferous broadleaved forest at 2400–2500 m (Davies, 2011). However, the breeding habitat requirements for this species remain unknown. Here we give a primary survey of breeding habitat of blackthroat in Qinling Mountains. Our study has two objectives: 1) quadrat sampling method was used to quantify habitat characteristics; 2) statistical analysis was employed to * Corresponding author. Tel. : +86-13892833187; Fax : +86-29-83217248; E-mail : [email protected] doi:10.2108/zs130136

determine what environmental factors are relevant to breeding habitat selection. MATERIALS AND METHODS Study area Fieldwork was conducted from early May to early August in 2012 at Foping national nature reserve (Fig. 1). Foping nature reserve is located in the southern portion of Shaanxi province on the central southern slopes of the Qinling Mountains (33°33′– 33°46′N, 107°41′–107°55′E). The area covered was approximately 290 km2, with elevations ranging from approximately 980 to 2900 m (Liu and Zhang, 2003). The vegetation in nature reserve is divided into three vertical zones: 1) deciduous broadleaved forest was dominated by Quercus variabilis, Q. glandulifera var. brevipetiolata, and Q. aliena var. acuteserrata (1000–2000 m), 2) birch forest (broadleaved coniferous forest) was dominated by Betula albo-sinensis and B. albo-sinensis var. septentrionalis (2000–2500 m); and 3) subalpine coniferous forest (2500–2900 m) with Abies fargesii being dominant species (Yue et al., 2000; Liu and Zhang, 2003). The reserves are located in the monsoon climatic zone, with an alpine temperate climate and a mean annual temperature of 2–6°C. Data collection We focused on subalpine coniferous forest and mixed coniferous broadleaved forest (birch forest) in the range from 2000 to 2650 m based on the survey of 2011. However, our survey also covered other habitat types in this nature reserve, such as deciduous broadleaved forest (1800–2000 m). A line transect approach was used to locate birds. Ten line transects were selected in subalpine coniferous forest and birch forest respectively, and three line transects were selected in broadleaved forest with each line transect being about 4 km long. We used male songs recorded in 2011 to play back along each line transect of various habitats. Each line transect was investigated three times throughout the season. From early to late May in 2012, we played back male songs from 6:00 am to 8:00

280

H. Zhao et al. (number of shrubs or culms per square meter), basal diameter per culm (BD) (mean basal diameter of all culms), shrub (culm) height (mean height of all shrub (culms)), and shrub (culm) coverage (%). Grass and ground layer information was collected in each 1 m × 1 m plot including dominant grass species, grass height (mean height of all grass), grass cover (%) and ground cover (%). The arithmetic means of the variables from four of 2 m × 2 m plots and four 1 m × 1 m plots was used to generalize shrub and grass information in each 10 m × 10 m plot. Moreover, terrain information was obtained in each 10 m × 10 m plot including altitude, slope, and orientation.

Data analysis Species habitat characteristics were analyzed in two steps. First, Gaussian distributions were examined using KolmogorovSmirnov normality test, and Mann-Whitney U tests were conFig. 1. Confirmed breeding sites of blackthroat in Foping nature reserve, Shaanxi China (Fpnr, Foping ducted to determine differences nature reserve). in each habitat variable between habitat sites and random sites. am and from 4:00 pm to 6:00 pm every day. Each playback length Secondly, we used KMO and Bartlett’s test to determine if principal was about 5 minutes. When a bird came to the playback speaker, component analysis (PCA) can be conducted considering the samwe tried to record its behavior. If there was not any response to ple sizes (number of cases) are very small. The KMO measure was playback songs, we recorded this as an absence of birds in the site. nearly 0.7, indicating factor analysis can be used, and Bartlett’s test Cases in which a bird was spotted on three consecutive days in the of sphericity was also significant, indicating the variables are related same place were regarded as the same individual. Latitude and lonand suitable for factor analysis (IBM, 2010). PCA was used for mulgitude were recorded by a handheld GPS at the locations of tivariate analysis of habitat variables. encounter. The playback was also conducted occasionally from late May to early August in 2012 to examine whether the birds were RESULTS present in the area. No information on species territory size was 1) General habitat characteristics available, but we assumed the territory size of blackthroat to be < At least 14 individuals consisting of 13 male and 1 100,000 m2 since most small passerines exhibit territory sizes smaller than 100,000 m2 (Zheng, 2012). Therefore, each confirmed female were spotted in four of total 23 line transects from individual was over 500 m distant from its neighbors one to avoid early May to late May 2012 in Foping nature reserve (Fig. repetition count. 1). Male birds exhibited strong territorial behavior and were The following habitat survey was carried out from late May to sensitive and aggressive to playback song since early May. middle June in 2012. At each confirmed site, a 10 m × 10 m samMales responded quickly to the playback song; we heard pling quadrat was set to obtain tree layer information, and this was male birds sing after several seconds of song playback, compared to an equal-sized quadrat randomly located in the same after which they flew near the speaker. The males perched habitat and altitude > 500 m distant from the site with an identified on the top of the shrub near to the speaker and kept singing blackthroat. In each 10 m × 10 m plot, we sampled four 2 m × 2 m for 1–2 minutes. Then the males approached the speaker plots randomly to obtain bamboo and shrub layer information, and gradually, flew back and forth over the speaker and attacked four 1 m × 1 m plots randomly to obtain grass and ground layer information (Liu and Zhang, 2003; Yang et al., 2003). Tree layer the speaker. The only female bird observed did not sing but information was collected in each 10 m × 10 m plot including domperched atop the shrub near the speaker. However, territoinant tree species, tree density (number of stems per 100 square rial behavior became weaker since early June. Male song meter), diameter at breast height (DBH) (mean diameter at breast can be heard occasionally until late July, but none was seen height of all stems), tree height (mean height of all stems), and total or heard since August. Unfortunately, no nest was found canopy coverage (%). Shrub layer information was collected in each during the whole summer in Foping 2012. (We found two 2 m × 2 m plot. Although bamboos are not woody plants, we catenests at the same area in late May 2013, and it appears gorized bamboos and shrubs together as shrub layer since their likely that blackthoat reproduced here in 2012.) We physiognomy and heights are similar (Liu and Zhang, 2003). Moreobserved both male and female birds lived in the dense over, bamboos covered most of 2 m × 2 m plots. The shrub layer bamboo forest and fed on insects in the ground litter. Usuinformation consists of dominant shrub species, shrub density

Breeding habitat of blackthroat

ally, the birds flew nearly 1 m above the ground in the dense bamboo forest. All the birds observed lived in coniferous forest and coniferous broadleaved forest with dense bamboos at the elevation ranging from 2130 to 2600 m. No birds were found in forest without any bamboos. Moreover, no birds was found in both mixed coniferous broadleaved forest and deciduous broadleaved forest with dense bamboos below 2130 m. The dominant tree species in these areas are Acer elegantulum, Alnus cremastogyne, Betula albo-sinensis, and Pinus armandii in mixed coniferous broadleaved forest from 2000 m to 2400 m. Betula albo-sinensis var. septentrionalis and Abies fargesii are dominant in mixed forest over 2400 m. The understory was dominated by dense stands of Fargesia qinlingensis. Dominant shrub species are Lonicera szechuanica and Sorbus koehneana while grass layer was dominated by Carex sp., Duchesnea indica and Elymas dahuricus. The ground was almost covered by falling bamboo and tree leaves. Topography is complex in this area, and most individuals tended to choose south slope exposed to the sun. Moreover, most individuals (10 of 14) were found on the ridge, though some individuals were seen in both

281

ridges and gullies. 2) Vegetative structures of breeding habitats Of 11 environmental variables, seven differed significantly between sites of encounter and random sites (Table 1). Sites with blackthroat seemed to exhibit greater ground cover, denser shrub density, greater shrub cover and higher shrub with less grass cover, sparser tree density and lower grass. On the other hand, canopy cover, basal diameter per culm, tree height and tree DBH did not differ significantly between sites with blackthroat and random sites. 3) Environmental factors that correlated with habitat selection by Blackthroat PCA analysis with 11 variables showed that the first four principal components met the criterion of an eigenvalue > 1 and together explained 83.8% of the variance in breeding microhabitats of the sites with blackthroat (Table 2). The first principal axis ordered sites by shrub density and shrub cover, and the second principal axis ordered sites by canopy cover. The third principal axis arranged sites by culm basal diameter, and the fourth principal axis arranged sites by ground cover.

Table 1. The comparison of habitat characteristics between habitat sites used by blackthroat and random sites. Habitat plots n = 14

Variables Altitude range (m) BD (cm) Canopy cover DBH (cm) Grass cover Grass height (cm) Ground cover Shrub density (number of shrub/m2) Shrub cover shrub height (m) Tree density (number of stems /100 m2) Tree height (m)

2130–2650 7.92 ± 0.17 0.46 ± 0.03 18.35 ± 1.50 0.39 ± 0.03 17.03 ± 0.99 0.85 ± 0.02 78.09 ± 3.98 0.67 ± 0.03 2.14 ± 0.07 5.95 ± 0.58 13.89 ± 1.00

Random plots n = 14 2130–2650 7.34 ± 0.39 0.46 ± 0.04 18.29 ± 1.74 0.68 ± 0.02 32 ± 2.59 0.60 ± 0.06 45.82 ± 5.82 0.28 ± 0.04 1.68 ± 0.08 8.75 ± 1.21 11.87 ± 0.89

Z

p

−1.882 −0.08 −0.51 −6.003 −4.715 −3.736 −2.659 −5.447 −3.867 −1.951 −0.714

0.06 0.937 0.61 P < 0.01** P < 0.01** P < 0.01** P < 0.01** P < 0.01** P < 0.01** P < 0.05** 0.475

Table 2. Correlation of habitat variables with the first four principal components derived from 14 habitat plots used by Blackthroat in Foping nature reserve. Variables BD Canopy cover DBH Grass cover Grass height Ground cover Shrub cover Shrub density Shrub height Tree density Tree height Eigenvalue Percent of variance (%) Percent of cumulative variance (%)

Principal components 1

2

3

4

0.158 −0.34 0.737 −0.48 −0.238 0.211 0.909 0.93 0.765 −0.459 0.726 4.031 36.643 36.643

−0.053 0.896 0.084 −0.435 −0.796 0.198 −0.03 −0.118 0.117 0.798 0.309 2.435 22.14 58.783

−0.644 0.188 0.431 0.458 0.319 0.341 −0.232 −0.227 0.164 −0.063 0.468 1.418 12.89 71.673

0.312 0.009 −0.323 0.132 0.294 0.837 0.09 0.048 0.431 0.251 −0.265 1.334 12.131 83.804

DISCUSSION Our survey showed that blackthroat is associated with bamboo thickets of Qinling Mountains, which is consistent with the historical record in which blackthroat was reported to breed in bamboo thickets on the tops of ridges dividing valleys at 3050– 3350 m in Minshan of the Gansu (BirdLife International, 2001). While our elevation record is much lower than the Gansu record, vegetation structure in Foping nature reserve is different from that in the Baihe nature reserve of Sichuan, which provides the most recent breeding record other than the Qinling records, and in which no bamboos were found in the understory of mixed coniferous broadleaved forest (Anderson, 2007). In Foping nature reserve, Bashania fargesii bamboos are distributed primarily in the elevation range of 1000–1900 m, while Fargesia qinlingensis bamboos can be found at l000–2900 m, but mainly between 2200 and 2800 m (Li et al., 2003; Feng et al., 2006). Blackthroat is likely distributed in high elevation Fargesia qinlingensis bamboo forest, and we have not observed any bird in low elevation Bashania fargesii forest. Though bamboo thickets are related with breeding habitat selection, the altitude seems to be a more important factor affecting habitat selection by blackthroat.

282

H. Zhao et al.

Our survey also demonstrated that blackthroat is associated with denser and higher bamboos and shrubs with sparser trees, less and lower grass. It has been reported that blackthroat nested in natural cavities about 0.75 m above the ground (Zhang and Zhang, 2012), and our observations showed its feeding on the ground. Plausibly, denser and higher bamboos may provide security for nest-building. Sparser trees may provide more exposure to sunlight, or more open upper space. Lower and less grass may provide more open understory space for bird foraging and flight. In addition, greater ground cover and more bamboos and other shrubs may enhance food availability, since a variety of insects feed on bamboo. PCA results showed similar trends. The first and the second principal components accounted for nearly 60% of the variance suggesting that shrub cover, shrub density, and canopy cover may be important factors correlated with breeding habitat of blackthroat. To date, more than 20 blackthroat individuals have been confirmed in Foping nature reserve of Qinling Mountains, which outnumbers the total records in previous studies (BirdLife International, 2013), suggesting that Qinling may be an important breeding site for this species. Although the Qinling Mountains were heavily logged in the 1960s, the ecosystem is thought to be recovering thanks to the establishment of nature reserves in the 1980s (Yang, 1993; Zhang, 2000). More than 25 nature reserves were built to protect endangered species and their habitats such as crested ibis Nipponia nippon, giant panda Ailuropoda melanoleuca, golden takin Budorcas taxicolor bedfordi and golden snub-nosed monkey Rhinopithecus roxellana (Zhang, 2000; Xu et al., 2012). The nature reserves in Qinling Mountains may benefit to blackthroat and its habitat conservation. The establishment of nature reserves and vegetation recovery in Qinling Mountains indicate habitat fragmentation and loss may not be a major issue to species in this region. In conclusion, blackthroat tends to choose dense bamboo forest at high altitude on southern slopes of the Qinling Mountains for its habitat. However, the information of habitat selection is still very limited. A few nests were found only in Changqing nature reserve and Foping nature reserve (Zhang and Zhang, 2012). Moreover, population status and distribution in Qinling mountains are still poorly known. More study will be needed to improve the knowledge of this species for future conservation.

plant species. This work was supported by National Science Foundation to H.F.Z. (No. 31201726) and Youth Science foundation of Shaanxi Academy of Sciences to N.H. (2012K-29).

REFERENCES

ACKNOWLEDGEMENTS

Anderson B (2007) Baihe Nature Reserve, Sichuan, China, 1–4 June 2007. Downloaded from: http://www.club300.se/Files/ TravelReports/Baihe2007BA.pdf on 05/05/2013 BirdLife International (2001) Threatened birds of Asia: the BirdLife International Red Data Book. BirdLife International, Cambridge BirdLife International (2013) Species factsheet: Luscinia obscura. Downloaded from http://www.birdlife.org on 05/07/2013 Davies E (2011) Rare robin breeding sites found. Assessed to http: //www.bbc.co.uk/nature/16112187.htm Feng YH, Feng LT, Yong YG, Dang GD, Ren Y (2006) A taxonomic study on the bamboo as the main food of giant panda from Mt. Qinling (II). J Northwest Univ (Nat Sci Ed) 36: 101–102, 124 Harter E (1907) On some rare species of the genus Larvivora from China. Ibis 9: 621–623 IBM (2010) IBM SPSS Statistics 19 Core System User’s Guide. SPSS, Chicago Li Y, Ren Y, Jia H (2003) The taxonomic studies on the bamboo as the main food of giant panda from Mt. Qinling (I). Acta Bot BorOcci Sin 23: 127–129 Liu SF, Zhang J (2003) Biodiversity and Conservation in Foping National Nature Reserve. Shaanxi Press of Science and Technology, Xi’an Qian W, Yi H (2011) First images in the wild of Blackthroat Luscinia obscura, Asia’s most enigmatic robin. Birding Asia 15: 17–19 Xu WH, Luo C, Ouyang ZY, Zhang L (2010) Designing regional nature reserves group: the case study of Qinling Mountain Range, China. Acta Ecol Sin 30: 1648–1654 Yang QC (1993) Annals of Geography of Hanzhong Shaanxi Province. Shaanxi People’s Press, Xi’an Yang WK, Qiao JF, Cambreau O, Gao XY, Zhong WQ (2003) Breeding habitat selection by the Houbara Bustard Chlamydotis [undulata] macqueenii in Mori, Xinjiang, China. Zool Stud 42: 470–475 Yue M, Dang GD, GU TQ (2000) Vertical zone spectrum of vegetation in Foping national reserve and the comparison with the adjacent areas. J Wuhan Bot Res 18: 375–382 Zhang MX (2000) Biodiversity in the Qinling Mountains nature reserves and its conservation and development. Rural EcoEnviron 16: 15–19 Zhang YW, Zhang XF (2012) Discovery of the propagation site of rare bird Luscinia obscura distributed in Qinling Mountain. Shaanxi For Sci Tech 41: 54–55 Zhao ZJ (2001) A Handbook of the Birds of China (Vol. II: Passerines). Jilin Science and Technology Press, Changchun Zheng GM (2012) Ornithology (2nd edition). Beijing Normal University Press, Beijing

We are grateful to Mr. Huisheng Gong and Mr. Yubao Hou for fieldwork assistance. Professor Xianhua Tian helped to identify

(Received July 1, 2013 / Accepted January 6, 2014)