Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Use of DNA metabarcoding of bird pellets in understanding raptor diet on the Qinghai-Tibetan Plateau of China

Use of DNA metabarcoding of bird pellets in understanding raptor diet on the Qinghai-Tibetan... Background: Diet analysis is essential to understanding the functional role of large bird species in food webs. Mor- phological analysis of regurgitated bird pellet contents is time intensive and may underestimate biodiversity. DNA metabarcoding has the ability to circumvent these issues, but has yet to be done. Methods: We present a pilot study using DNA metabarcoding of MT-RNR1 and MT-CO1 markers to determine the species of origin and prey of 45 pellets collected in Qinghai and Gansu Provinces, China. Results: We detected four raptor species [Eurasian Eagle Owl (Bubo bubo), Saker Falcon (Falco cherrug), Steppe Eagle (Aquila nipalensis), and Upland Buzzard (Buteo hemilasius)] and 11 unique prey species across 10 families and 4 classes. Mammals were the greatest detected prey class with Plateau Pika (Ochotona curzoniae) being the most frequent. Observed Shannon’s and Simpson’s diversity for Upland Buzzard were 1.089 and 0.479, respectively, while expected values were 1.312 ± 0.266 and 0.485 ± 0.086. For Eurasian Eagle Owl, observed values were 1.202 and 0.565, while expected values were 1.502 ± 0.340 and 0.580 ± 0.114. Interspecific dietary niche partitioning between the two spe - cies was not detected. Conclusions: Our results demonstrate successful use of DNA metabarcoding for understanding diet via a novel noninvasive sample type to identify common and uncommon species. More work is needed to understand how raptor diets vary locally, and the mechanisms that enable exploitation of similar dietary resources. This approach has wide ranging applicability to other birds of prey, and demonstrates the power of using DNA metabarcoding to study species noninvasively. Keywords: Avian, Eurasian Eagle Owl, Molecular diet analysis, Next-generation sequencing, Raptor, Saker Falcon, Steppe Eagle, Upland Buzzard Background Understanding predator–prey interactions  is an impor- tant component of community ecology and management (Estes et al. 2011). Sympatric species with similar ecologi- *Correspondence: yugzhang@sina.com; janeckaj@duq.edu cal demands must find ways to reduce competition. One Research Institute of Forest Ecology, Environment and Protection, Chinese Academy of Forestry, and Key Laboratory of Biodiversity way this is accomplished is through dietary niche parti- Protection of National Forestry and Grassland Administration, tioning (Schoener 1974). Understanding this overlap can Beijing 100091, China discern how species allocate resources. Such knowledge Department of Biological Sciences, Duquesne University, Pittsburgh, PA 15282, USA is important for conservation planning as communities Full list of author information is available at the end of the article © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Hacker et al. Avian Res (2021) 12:42 Page 2 of 11 with lower niche overlap can support greater biodiver- MT-RNR1 is an RNA gene previously used in studies sity (Pianka 1974). Unfortunately, food web dynamics examining the identity and genetic relationships of ani- are complex and require accurate information of items mals (Riaz et  al. 2011). However, MT-RNR1 lacks the consumed (Pompanon et  al. 2012). Understanding the genetic diversity to discern wild versus domestic goat trophic niche of birds is important as knowledge of prey (Capra hircus) and sheep species (Shehzad et  al. 2012; composition plays an important role in shaping conser- Hacker et  al. 2021). This is problematic for the discern - vation policies (Grier 1982). Dietary assessment methods ment of domestic sheep and argali (Ovis ammon), as for avian species include direct observation (Margalida well as domestic goat and Siberian Ibex (Capra sibirica), et  al. 2005, 2009), camera placement at nest sites (Mar- in areas where they are sympatric (Reading et  al. 2020). galida et al. 2005; Tremblay et al. 2005), stomach pump- uTh s, an additional marker is necessary. MT-COI is ing (Wilson 1984; Walter and O’Neill 1986), examination widely used for DNA barcoding as it has both conserved of stomach contents (Miller and McEwen 1995), diges- regions and segments with high divergence (Hebert et al. tive tract flushing (Moody 1970), forced vomiting (Valera 2003). Recent work identified a segment of MT-COI et  al. 1997), examination of fatty acid and isotope sig- capable of discerning wild and domestic goat and sheep natures (Iverson et  al. 2007), molecular fecal analysis taxa in Central Asia, making it an important addition (Treves et  al. 2016; Jedlicka et  al. 2017; Trevelline et  al. for diet research of predators and scavengers distributed 2018a, b) and morphological investigation of pellets (Li there (Hacker et al. 2021). et al. 2004). Numerous sympatric birds of prey are found on the Bird pellets are the accumulation of undigested prey Qinghai-Tibetan Plateau (QTP). They play vital roles in that are regurgitated through the mouth in compact maintaining ecosystem balance (Xia et al. 1991) under the units (Taberlet and Fumagalli 1996). Examination of context of vast modernization and environmental shifts diet via pellets has historically involved collection and (Liu and Chen 2000; Foggin 2008). They face threats morphological assessment of contents (Ewins et  al. associated with climate change (Liu and Chen 2000) as 1994; Symondson 2002; Sándor and Ionescu 2009). This well as anthropogenically-induced mortality, such as method has many shortcomings. Digestive processes may electrocution from power lines (Dixon et  al. 2013) and render samples unrecognizable (Symondson 2002; Galan pika poisoning (Badingqiuying et  al. 2016). Raptor spe- et  al. 2012), pellet appearance can vary widely based on cies include the Eurasian Eagle Owl (Bubo bubo) (Birdlife life stage and sex (Galan et  al. 2012), small to medium International 2017), Upland Buzzard (Buteo hemilasius), size prey are often overestimated and unusual prey items Northern Goshawk (Accipiter gentilis), Golden Eagle unrecorded (Marchesi et  al. 2002), and taxonomic spe- (Aquila chrysaetos), Saker Falcon (Falco cherrug) (Dixon cialists are needed for wide ranging genera (Galan et  al. et  al. 2013; Birdlife International 2018a), Steppe Eagle 2012). (Aquila nipalensis), Himalayan Vulture (Gyps himalayen- Molecular approaches involving the examination sis), and Bearded Vulture (Gypaetus barbatus) (Schaller of DNA can circumvent these issues and increase the 1998; Cui et al. 2008; Birdlife International 2019). Under- detectable prey spectrum relative to sampling and analy- standing their diets is necessary for effective conserva - sis effort (Oehm et al. 2017). DNA sequencing techniques tion action, and DNA metabarcoding of regurgitated have been called for since the early 2000s (Symondson pellets provides a method to do so noninvasively. 2002), but remain under used despite the stability of The aims of this study were to (1) determine applicabil - DNA in bird pellets (Taberlet and Fumagalli 1996), and ity of DNA metabarcoding to species and prey identifi - the advent of Next-Generation Sequencing (NGS) tech- cation of avian pellets; (2) examine metrics surrounding nology which can identify species from a high volume species presence and mechanisms of coexistence; and (3) of samples (Galan et  al. 2012). In DNA metabarcoding, make suggestions based on our results for conservation generalized primers target and amplify a segment, the action planning. DNA barcoding region, of a conserved gene (Pompanon et  al. 2012). This gene must have low intra-species vari - Methods ation (Galan et al. 2012) and high inter-species variation Study site for taxonomic classification (Simon et al. 1994). Degrada- Samples were collected in the Qilian Shan Mountains tion of longer DNA segments remains problematic. Thus, (hereafter “Qilian Shan”) of Qinghai and Gansu Prov- researchers rely on shorter segments known as “mini- inces, China, in the eastern Kunlan Mountains in Dulan barcodes” (Meusnier et  al. 2008). Researchers also use County, Qinghai Province, China and in Zhiduo County, mitochondrial genes, as they have higher copy numbers Yushu Prefecture, Qinghai Province, China (Fig.  1). Qil- compared to nuclear genes and greater PCR amplifica - ian Shan runs along the northeastern corner of the QTP tion success (Freeland 2017). (> 3000  m above sea level). It comprises three parallel Hacker  et al. Avian Res (2021) 12:42 Page 3 of 11 September 2017 to July 2019. Sampling methods are described in Janečka et al. (2008, 2011). DNA extraction Samples were stored in a − 20  °C freezer. Pellets were defrosted and individually placed in a Petri dish. Twee- zers were then used to remove material from  the out- side and inside the pellet to capture DNA of both the prey and host and placed in a 1.5  mL centrifuge tube. DNA was extracted using QIAamp DNA Stool Mini Kits (QIAGEN, Hilden, Germany) with two additional centrifugation steps to remove residual Buffer AW1 and AW2. Aliquots were quantified using a NanoDrop Lite Spectrophotometer. PCR for species and diet analysis A short segment (~ 100-bp) of MT-RNR1 (primers 12SV5F/12SV5R; Riaz et  al. 2011) was used to discern all predator and prey species except those belonging to Fig. 1 Bird pellet collection locations and their identified host goats and sheep. The primers for MT-CO1 (MT-CO1- species. Host species was determined by sequencing of a segment 379F and MT-CO1-604Rd; Hacker et  al. 2021) were of the MT-RNR1. Figure made in QGIS and modified using Microsoft designed to amplify a gene segment (330-bp) capable of PowerPoint for Mac version 16.42 differentiating closely related caprines. Each segment was amplified separately using a PCR reaction containing 1.5 µL of DNA template, 7.94 µL of KAPA HiFi HotStart ReadyMix (2 ×) (Kapa Biosystems, Wilmington, MA, subsidiary ranges—the Tuali Nanshan, Shule Nanshan, USA), 0.16 µL of 20 µM forward primer, 0.16 µL of 20 µM and Danghe Nanshan (Schaller et  al. 1988). Qilian Shan reverse primer, and 5.2 µL of PCR grade water. PCR con- is composed of deserts at lower elevations giving way ditions consisted of 95 °C for 3 min, 35 cycles of 95 °C for to shrubs, grasses, and alpine meadows (Schaller et  al. 30 s, 60 °C for 30 s, and 72 °C for 30 s, followed by a 5 min 1988). Yushu Prefecture is in the southwestern corner extension step at 72 °C and 4 °C hold. of Qinghai Province, and has alpine meadow vegetation with small rugged ranges surrounded by rolling grass- Next‑generation sequencing land, with juniper forests along mountainsides (Schaller Amplicons were mixed in equal ratios determined by et  al. 1988). The Kunlan Mountains are the longest measuring gel band brightness using GeneTools Analy- mountain system in Asia with its eastern end south west sis Software Version 4.03.05.0 (SynGene, Frederick, MD, of Qilian Shan (Miller and Bedunah 1994). The landscape USA). The E.Z.N.A. Gel Extraction Kit (Omega Bio-Tek, of Dulan County is primarily rugged grassland with rock Inc., Norcross, GA, USA) was used to isolate and purify slopes (Liu 1993). Species found on the QTP include ® ™ products. The NEBNext Ultra II DNA Library Prep Kit the Blue Sheep (Pseudois nayaur), Tibetan Gazelle (Pro- for Illumina (New England Biolabs, MA, USA) was used capra picticaudata), Argali (Ovis ammon), and White- to prepare sequencing libraries. Indexing oligonucleo- lipped Deer (Cervus albirostris), and carnivores such as tides (Nextera XT, Illumina, San Diego, CA, USA) were the Tibetan Wolf (Canis lupus), Red Fox (Vulpes vulpes), incorporated and the pooled ampliconic library quanti- Snow Leopard (Panthera uncia), Eurasian Lynx (Lynx fied using an Invitrogen Qubit 2.0 Fluorometer (Thermo lynx), and Pallas’s Cat (Otocolobus manul), among others. Fisher Scientific, Waltham, MA, USA). Paired-end 250- Small mammals include voles (Neodon sp.), pikas (Ocho- bp sequencing was completed on an Illumina NovaSeq tona sp.), zokors (Myospalax sp.), Woolly Hare (Lepus 6000 by Guangdong Magigene Biotechnology Co., Ltd. oiostolus), and Himalayan Marmots (Marmota himaly- (Guangzhou, China). ana) (Schaller et al. 1988; Jackson 2012). Diet analysis Sample collection FASTQ sequences were demultiplexed, sequencing Permits were obtained prior to sample collection. Pel- adapters removed, and reads imported into CLC Genom- lets were collected opportunistically as part of a sepa- ics Workbench v12.0 (CLC bio, QIAGEN, Aarhus, rate snow leopard study over seven sampling trips from Hacker et al. Avian Res (2021) 12:42 Page 4 of 11 Denmark). Raw sequencing reads were trimmed using a by the total number of pellets for each predator species. quality score limit of 0.1. A reference FASTA file was cre - Observed dietary taxonomic richness, as well as Shan- ated by downloading MT-CO1 and MT-RNR1 sequences non’s index (Shannon and Weaver 1949) and Simpson’s for potential prey and host species, which were deter- index (Simpson 1949; Marchesi et  al. 2002) were cal- mined via a literature search and by consulting with local culated using the “iNEXT” function (Hsieh et  al. 2016). experts (Additional file  1). Sequencing reads were then Effective taxonomic richness and diversity of predator mapped to the reference FASTA file using local align - diet was calculated using Hill numbers (Hill 1973) and ment with the following parameters—mismatch cost: the “iNEXT” function (Hsieh et  al. 2016). This enabled 2; insertion cost: 3; deletion cost: 3; length fraction: 0.9; comparison of dietary diversity between species with similarity: 0.94; non-specific matches mapped randomly varying sample sizes (Hurlbert 1971; Heck et  al. 1975) (Hacker et al. 2021). To identify the host species of each and allowed for evaluation of sampling completeness. pellet, identification was made when the sequenced reads For the Upland Buzzard and the Eurasian Eagle Owl, were mapped to a reference sequence with > 98% similar- Pianka’s metric of niche overlap (Pianka 1974) was cal- ity. Unique haplotypes of the MT-RNR1 segment for each culated using the ‘niche_null_model’ function in the Eco- host species were determined in MEGA 7.0 by aligning SimR package (Gotelli et  al. 2015) and compared to 999 sequences using MUSCLE (Edgar 2004). Prey identifi - null model simulations to determine if observed dietary cation was made by selecting the prey species from the niche overlap was higher or lower than expected. The reference database with the highest number of mapped occurrence of interspecific niche differentiation between reads and fewest number of mismatches. To ensure prey these two species was determined by performing a DNA sequences were correctly classified, the consensus permutational multivariate analysis of variance (PER- sequence for the host and prey items were extracted and MANOVA; Anderson 2017) using the “adonis” function a blastn search performed against the nr/nt nucleotide and Jaccard distance matrix within the vegan R package collection GenBank databases using megablast for highly (Oksanen et al. 2013). Additional information on statisti- similar sequences (Additional file  2). In addition, collec- cal analyses can be found in Additional file 3. tion sites were compared with known species distribu- tions (CITES Red List range maps, https:// www. iucnr edlist. org/ search/ map). Results Samples with a large proportion of unmapped reads Predator species detected (> 5% of total reads) were analyzed to rule out an incom- Using the MT-RNR1 genetic marker, four host bird spe- plete reference file. This was done by performing a de cies were identified among 45 pellets—Upland Buzzard novo assembly with the following parameters—minimum (n = 26, 23 from Qilian Shan and 3 from Yushu), Eura- contig length of 100; mismatch cost: 2; insertion cost: sian Eagle Owl (n = 15, 14 from Qilian Shan and 1 from 3; deletion cost: 3; length fraction: 0.9; similarity: 0.98. Yushu), Steppe Eagle (n = 3, from Dulan County), and Consensus sequences were extracted for contigs with the Saker Falcon (n = 1, from Qilian Shan) (Fig. 2). A total of highest number of mapped reads. At least 10,000 reads 8 unique MT-RNR1 haplotypes were found for Upland were required to generate a consensus sequence. Nucle- Buzzard, 5 for Eurasian Eagle Owl, 1 for Saker Falcon, otides in sites with conflicting reads were resolved via and 1 for Steppe Eagle (Additional file 4). majority rule and ambiguous sites were coded with an “N”. Species were identified using the same blastn search described above. Predator diet composition A total of 41 of the 45 collected pellets (91.1%) had prey DNA sequences discerned. A total of 11 unique prey taxa Statistical analysis spanning 7 orders and 4 classes were detected (Table  1). All statistical analyses were performed in R version The most frequently detected prey species, Plateau Pika 3.5.2 (R Core Team 2018) using base R functions, vegan (Ochotona curzoniae), was found in 77.8% (35/45) of all (Oksanen et  al. 2013), iNEXT (Hsieh et  al. 2016), and diets analyzed and was the only prey taxa detected in the EcoSimR packages (Gotelli et  al. 2015). The mean num - Saker Falcon samples (Table 1). While Mammalia was the ber of unique prey taxa per pellet was calculated by sum- most taxonomically-rich prey class (7 species; Table  1), ming the number of detected prey species among each predator species in this study also consumed Ray-finned pellet and dividing this sum by the number of pellets Fishes (Actinopterygii, 2 species), small birds (Aves), and analyzed. This calculation was also repeated for each bird one species of toad (Amphibia). species sampled for interspecific comparisons. Dietary frequency of occurrence was calculated by dividing the number of pellets in which a prey species was detected Hacker  et al. Avian Res (2021) 12:42 Page 5 of 11 Fig. 2 The host bird species identified via DNA-metabarcoding of bird pellets collected on the Qinghai-Tibetan Plateau. a Steppe Eagle (photo credit: Jia Li). b Upland Buzzard (photo credit: Charlotte Hacker). c Saker Falcon (photo credit: Charlotte Hacker). d Eurasian Eagle Owl (photo credit: Munib Khanyari). Figure made in Microsoft PowerPoint for Mac version 16.42 Dietary richness, diversity and overlap each for Steppe Eagle and Saker Falcon. Steppe Eagle The average number of unique prey taxa detected in (n = 3) and Saker Falcon (n = 1) were removed from fur- a single pellet across all species was 1.16 (range 0–3), ther analyses due to limited sample sizes. with Upland Buzzard and Eurasian Eagle Owl having Though these data are preliminary, they provide an an average of 1.16 and 1.13 unique prey taxa per pellet, important starting point for understanding dietary over- respectively. We observed the average number of unique lap between two large bird species. The expected dietary prey taxa per pellet for Saker Falcon and Steppe Eagle richness of Upland Buzzard and Eurasian Eagle Owl was in our study to be 1 and 0.333, respectively. Our analy- 14.692 ± 11.227 (mean ± SE) and 11.6 ± 6.594, respec- sis detected 7 prey species within pellets collected from tively (Fig.  3). DNA barcoding detected 87.3% of the Upland Buzzard, 6 from Eurasian Eagle Owl, and one Upland Buzzard’s and 79.9% of the Eurasian Eagle Owl’s Hacker et al. Avian Res (2021) 12:42 Page 6 of 11 Table 1 The number (n) and percent frequency (%) of prey species identified Prey item Upland Buzzard Eurasian Eagle Owl Steppe Eagle Saker Falcon (Buteo hemilasius) (Bubo bubo) (Aquila nipalensis) (Falco cherrug) 26 pellets 15 pellets 3 pellets 1 pellet n % n % n % n % Plateau Pika (Ochotona curzoniae) 22 84.60 12 80.00 0 0 1 100.00 Plateau Vole (Neodon fuscus) 3 11.50 0 0 0 0 0 0 Sikkim Mountain Vole (Neodon sikimensis) 0 0 1 6.67 0 0 0 0 Woolly Hare (Lepus oiostolus) 1 3.85 3 20.00 1 33.33 0 0 Tibetan Snow Finch (Montifringilla henrici) 2 7.69 0 0 0 0 0 0 Long-tailed Dwarf Hamster (Cricetulus longicaudatus) 0 0 1 6.67 0 0 0 0 Przewalksi’s Naked Carp (Gymnocypris przewalskii) 0 0 1 6.67 0 0 0 0 Chinese Perch (Siniperca sp.) 1 3.85 0 0 0 0 0 0 Lazy Toad (Scutinger sp.) 0 0 1 6.67 0 0 0 0 Domestic Goat (Capra hircus) 1 3.85 0 0 0 0 0 0 Blue Sheep (Pseudois nayaur) 1 3.85 0 0 0 0 0 0 Taxonomic richness 7 6 1 1 dietary niche overlap was calculated to be 0.511, within the expected 95% confidence interval derived from null model simulations (CI range: 0.477–0.522). Discussion Pika was the dominant consumed species in this study. Pikas are non-hibernating and diurnal, providing a year-long accessible food source (Badingqiuying et  al. 2016). Pikas have been labeled as pests, with a poison- ing program to control and eradicate them launched in 1958 (Smith et  al. 1990; Smith and Foggin 1999). How- ever, more recent research recognizes pikas as important environmental engineers that cause minimal damage to alpine grassland ecosystems (Smith and Foggin 1999; Wei Fig. 3 Estimation of prey taxonomic richness with rarefaction and et  al. 2020). Poisoning them is more likely to have nega- extrapolation. Using the iNEXT function, rarefaction curves were tive impacts (Lai and Smith 2003; Badingqiuying et  al. made to estimate sample completeness and expected taxonomic 2016). In a study by Badingqiuying et  al. (2016), record- richness with additional sampling for the Eurasian Eagle Owl ings of Steppe Eagles, Saker Falcons, and Upland Buz- (n = 15) and Upland Buzzard (n = 26). Interpolation (solid lines) was performed using the sample sizes for each predator species while zards were three times less frequent at sites with active extrapolation (dotted lines) of species richness was performed on poisoning programs compared to those without; though double the sample size. Figure made in R version 3.5.2 data were insufficient to make any conclusions about Eurasian Eagle Owls. Similarly, Lai and Smith (2003) reported that Upland Buzzards were found 11.2 times expected prey taxa. Observed Shannon’s and Simp- more frequently at non-poisoned versus poisoned pika son’s diversity for Upland Buzzard were 1.089 and 0.479, sites. respectively, while expected values were 1.312 ± 0.266 The one pellet collected from a Saker Falcon revealed and 0.485 ± 0.086. Observed Shannon’s and Simp- pika as the dietary item. In previous studies, pikas were son’s diversity for Eurasian Eagle Owl were 1.202 and found to comprise 90% of the food items a Saker Fal- 0.565, while expected values were 1.502 ± 0.340 and con pair fed to their young in the Chang Tang  region 0.580 ± 0.114, respectively. Our PERMANOVA results (Schaller 2012). In this study, pikas were detected in 80% did not detect interspecific dietary niche partition - of pellets from Eurasian Eagle Owl and 84.6% of Upland ing between Upland Buzzard and Eurasian Eagle Owl Buzzard pellets, aligning with previous studies. Our (Pseudo-F = 1.86, p = 0.129) and Pianka’s index of (1, 36) results further substantiate the important role pika play Hacker  et al. Avian Res (2021) 12:42 Page 7 of 11 as a primary food source for large bird species. However, occur at different parts of the day. Across a broader some species may be able to adapt to reductions in pika temporal scale, Upland Buzzards partially migrate, populations by exploiting other prey species (Cui et  al. spending the breeding season in China, while Eurasian 2008). Eagle Owls are not migratory (Birdlife International Dietary diversity assessments could only be done with 2017). However, with the exception of two samples Upland Buzzards and Eurasian Eagle Owls due to the collected in September, all remaining pellets were col- low sample size of Saker Falcon and Steppe Eagle pellets. lected during the corresponding breeding season for While the Upland Buzzard had one more unique spe- Upland Buzzards (April through July) (Rasmussen and cies represented in their diet compared to the Eurasian Anderson 2005), and both would be expected to be pre- Eagle Owl, the observed diversity indices suggested that sent. Eurasian Eagle Owls and Upland Buzzards also Eurasian Eagle Owls exhibit greater dietary taxonomic deploy different hunting strategies  which may separate diversity. The values resulting from the Shannon-Weiner them  behaviorally. Eurasian Eagle Owls are oppor- diversity index and Simpson’s index between the two spe- tunistic (Hiraldo et  al. 1975), swooping down on prey cies were relatively similar with a difference of 0.113 for and departing whether the kill is successful or not, Shannon-Weiner and 0.086 for Simpson’s. Previous work while Upland Buzzards deploy a sit-and-wait strategy by Cui et al. (2008), who compared the dietary diversity of (Cui et  al. 2008). In addition, we observed ungulates Upland Buzzards and Eurasian Eagle Owls via morpho- (Domestic Goat and Blue Sheep) in two Upland Buz- logical assessment of regurgitated pellets found that Eur- zard pellets, likely attributed to scavenging. asian Eagle Owls exhibited higher, but similar diversity to Unique and previously under-reported prey items that of the Upland Buzzard with a difference between the were identified. It may be that birds at present are rap - two species of 0.17. When calculated to expected degrees idly adapting to changes in prey base; however, it is also of dietary diversity, differences between Upland Buzzard possible that traditional methods reliant on the mor- and Eurasian Eagle Owl in this study were a bit larger, phological assessment of pellets may have missed rare with a difference of 0.19 expected for Shannon-Weiner or unexpected prey items. In this study, Przewalski’s and 0.095 for Simpson’s. Greater expected dietary diver- Naked Carp (Gymnocyris przewalksii) was found in one sity for Eurasian Eagle Owls is not unexpected as they are Eurasian Eagle Owl pellet. Przewalski’s Naked Carp is known to be generalists (Hiraldo et al. 1975). found in Qinghai Lake and is classified as Endangered Despite the relatively small number of pellets collected, on the China Species Red List (Chen et al. 2009; Zhang the sampling effort in this study captured 87.3% of the et  al. 2015). Loss of the species could cause severe expected prey taxa for the Upland Buzzard and 79.9% negative consequences within the lake (Chen et  al. of expected prey taxa for the Eurasian Eagle Owl. Spe- 2009). A previous study examining the distribution of cies that have been previously recorded as present in the Przewalski’s Naked Carp in Qinghai Lake showed that diets of birds of prey living on the QTP but not found the species was found ~ 2  m below the surface (Chen in this study include Zokors (Myospalax fontanierii; et al. 2009), where individuals moving upward could be M. baileyi), Gansu Pika (Ochotona cansus), Root Voles caught by birds of prey. (Micrtous oeconomus), Plateau Voles (Lasiopodomys fus- Although DNA metabarcoding is sensitive enough to cus), and Mountain Weasels (Mustela altaica) (Lai and detect the presence of rare and elusive species (Gran- Smith 2003; Li et al. 2004; Cui et al. 2008). The presence jon et  al. 2002; Thiam et  al. 2008), the degree to which and frequency of food items in the diets of birds of prey they are being predated would be unknown due to are typically reflective of prey presence and frequency in the inability to count the number of individuals con- their habitats (Bontzorlos et  al. 2005). While data could sumed within in a pellet (Symondson 2002; Emmrich reflect low numbers or absence of expected species that and Düttmann 2011). This may be problematic as raw may spark concern surrounding conservation status of occurrence counts of dietary items can artificially wildlife, species absence in this dataset is most likely due inflate the occurrence of rare food taxa (Deagle et  al. to low sample size or a result of variable methodology in 2019). Previous research has suggested that the num- comparison to previous studies. ber of mapped reads may be quasi-indicative of the We found no evidence of dietary niche partitioning quantity of a dietary item within a sample (Kartzinel between Upland Buzzard and Eurasian Eagle Owl, indi- et  al. 2015; Deagle et  al. 2019; Lamb et  al. 2019). We cating that temporal or spatial partitioning is responsi- greatly caution against this assumption, as read map- ble for coexistence. Previous work by Cui et  al. (2008) ping success can be influenced by other factors such as had similar findings. Temporally, Eurasian Eagle Owls PCR amplification bias, sample quality, and reference are nocturnal while Upland Buzzards are diurnal (Lei sequence availability (Pompanon et al. 2012; Lamb et al. 1995; Yang et  al. 2000), and resource exploitation may 2019). Other potential pitfalls of DNA metabarcoding Hacker et al. Avian Res (2021) 12:42 Page 8 of 11 which should be considered include the level of DNA caprids in the diet of the Upland Buzzard, one match- degradation and whether the genetic marker used pro- ing to Domestic Goat and one to Blue Sheep. Unfor- vides necessary taxonomic resolution. tunately, and as would be an issue with morphological In this study, four of the pellets collected were unable analyses, DNA metabarcoding is unable to distinguish to have prey DNA amplified. This may be because the between scavenging versus legitimate kills (Symondson primer pairs used did not amplify the prey’s target gene, 2002). Larger birds of prey, such as the Golden Eagle or could be a result of DNA degradation. The QTP is a have been observed taking larger prey such as domestic dry, cold environment. Regardless, DNA degradation calves and sheep in North America (Avery and Cum- may make the rendering of sequences from some samples mings 2004). Upland Buzzards are considerably smaller impossible (Guimaraes et  al. 2016). Shorter diagnostic in size, and would most likely not be able to kill caprids; fragments are preferred as they are more likely to remain with the exception of newborns. Scavenging on the intact (Meusnier et  al. 2008). However, use of shorter carcass of an already deceased individual or placenta DNA segments warrants caution as this may increase is more likely. Determining whether native raptors kill difficulty in discerning species (Symondson 2002), espe- newborn livestock would be of particular interest to cially for those belonging to taxonomically-rich groups. herders, and would present a further challenge for con- Rodents represent 40% of all mammalian species glob- servation as this could lead to conflict. ally (Musser and Carleton 2005), and were found in our The genetic identification of pellet hosts allowed for dataset. While unmapped reads were examined to iden- accurate information of species presence in a particu- tify any species potentially missing from the reference lar area. Of the 45 pellets identified in this study, only 3 file, it remains possible that the MT-RNR1 segment does belonged to Steppe Eagle and 1 to Saker Falcon. China is not amplify all possible rodent species within our study breeding habitat for both species, corresponding to when area. Rodents are a food source for many predators on samples were collected, but both birds are classified as the QTP, extending beyond birds (Hacker et  al. 2021), Endangered by the IUCN and lack of representation in and thus an accurate assessment of their consump- our dataset is consistent with these low population num- tion would be of use in large-scale conservation plan- bers (Birdlife International 2018a, 2019). In contrast, the ning. Sequences which cannot be discerned down to Upland Buzzard and Eurasian Eagle Owl are listed as the species-level may be grouped into a “prey OTU”, as Least Concern (Birdlife International 2017, 2018b). More commonly done in dietary and microbial studies for iden- detailed studies may be able to apply noninvasive surveys tification and analysis purposes (Marchesi et  al. 2002). using regurgitated pellets and DNA metabarcoding to Galan et  al. (2012) designed a mini-barcode marker map geographic distribution or shifts in temporal pres- based on a 136-bp DNA segment of MT-CYB capable of ence during breeding seasons and migration events, as discerning a vast array of Rodentia. While our study did well as habitat preference and overall population status. not face intensive challenges associated with rodent dis- This can similarly be applied to prey in an area, though cernment, our sample size was small and concentrated in birds may be moving long distances and thus the pellets two areas. Study questions tied to rich taxonomic groups they regurgitate may not be indicative of immediate local as a resource in an area with high biodiversity may want wildlife presence per se. to consider the addition of such a marker. Researchers Albeit a small sample size, our findings provide infor - should also consider the likelihood of false identifications mation important for conservation action on the QTP. which may result in the query sequence being assigned First, results demonstrate that pikas are an important to the incorrect taxon and subsequently choose appro- food source, thus large-scale policies to eradicate them priate sequence based species identification methods to should be reconsidered. Second, Eurasian Eagle Owls increase confidence, accuracy, and efficiency of data pro - and Upland Buzzards depend on similar dietary items. cessing (Soergel et  al. 2012; Boyer et  al. 2016; Shi et  al. Persistence of these prey will be necessary for survival 2018). of both species. Eurasian Eagle Owls may be better able The MT-COI marker was used as a diagnostic marker to exploit a wider range of prey as they demonstrate the to differentiate wild and domestic caprid species found ability to increase dietary plasticity, while Upland Buz- in the study area (Hacker et  al. 2021). Though ungu - zards may lack the adaptability to quickly changing and lates were not necessarily expected in the diet based resource poor landscapes. Third, birds of prey may con - on previously available literature, it was surmised tribute to conflict with humans in the form of livestock that past work using morphological analysis may have depredation or death. Reports of such from local herders, missed these larger species as bones and other hard or findings of larger than expected mammals in pellets, parts would likely not be ingested by birds due to should be taken seriously to build mitigation actions for their size. Indeed, our study found two occurrences of reducing livestock loss. Hacker  et al. Avian Res (2021) 12:42 Page 9 of 11 Sabin Family Foundation (G1900011, G2000017), and Welfare Project of the Conclusions National Scientific Research Institution (CAFYBB2019ZE003). Biodiversity tends to be richer and more quickly chang- ing than acknowledged, and assessment methods must Availability of data and materials The datasets generated during the current study are available via email to the be reliable and comprehensive (Pimm et al. 1995). DNA corresponding authors. metabarcoding is a conservation technology tool capa- ble of providing such information. We show the utility Declarations of this technique in determining dietary items from regurgitated bird pellets, and subsequently identified 11 Ethics approval and consent to participate Not applicable. unique prey species for 4 raptor species with pika con- firmed as an important food resource, with high dietary Consent for publication niche overlap between Eurasian Eagle Owl and Upland Not applicable. Buzzard. While we focus on bird of prey species on the Competing interests QTP, our methodology could be applied to pellet analy- The authors declare that they have no competing interests. sis for species worldwide. Our work also demonstrates Author details the power of DNA metabarcoding with noninvasive Research Institute of Forest Ecology, Environment and Protection, Chinese samples, encouraging other sample types to also be Academy of Forestry, and Key Laboratory of Biodiversity Protection of National explored. We also showed that sampling efforts may not Forestry and Grassland Administration, Beijing 100091, China. Depar tment of Biological Sciences, Duquesne University, Pittsburgh, PA 15282, USA. need to be as intensive as previously assumed, given the Yanchiwan National Nature Reserve of Gansu Province, Subei 736300, China. relatively high number of observed taxa captured com- Institute for Ecology and Environmental Resources, Chongqing Academy pared to the expected number of taxa despite limited of Social Sciences, Chongqing 400020, China. The High School Affiliated to Renmin, University of China, Beijing 100872, China. sample sizes. This could decrease effort, time, and costs associated with fieldwork. However, we also contend Received: 26 February 2021 Accepted: 1 August 2021 that the application of this technique has great poten- tial  to answer complex questions surrounding birds of prey that will require rigorous and large-scale study design. References Anderson MJ. Permutational multivariate analysis of variance (PERMANOVA). Wiley StatsRef Stat Refer Online. 2017. https:// doi. org/ 10. 1002/ 97811 Supplementary Information 18445 112. stat0 7841. The online version contains supplementary material available at https:// doi. Avery ML, Cummings JL. Livestock depredations by black vultures and golden org/ 10. 1186/ s40657- 021- 00276-3. eagles. Sheep Goat Res J. 2004;19:58–63. Badingqiuying, Smith AT, Senko J, Siladan MU. Plateau pika Ochotona curzoniae poisoning campaign reduces carnivore abundance in southern Qinghai, Additional file 1. The FASTA file of reference sequences used for read China. Mammal Study. 2016;41:1–8. mapping of possible host and prey species. Birdlife International. Bubo bubo (amended version of 2016 assessment). The Additional file 2. Results of subject sequences blasted to reference IUCN Red List of Threatened Species 2017: e.T22688927A113569670. sequences of determined host and prey items. 2017. https:// doi. org/ 10. 2305/ IUCN. UK. 2017-1. RLTS. T2268 8927A 11356 9670. en. Additional file 3. The r-script used in the statistical analyses of the Birdlife International. Falco cherrug (amended version of 2016 assessment). dataset. The IUCN Red List of Threatened Species 2017: e.T22696495A110525916. Additional file 4. Unique haplotypes for a segment of MT-RNR1 amplified 2018a. https:// doi. org/ 10. 2305/ IUCN. UK. 2017-1. RLTS. T2269 6495A 11052 by primers (12SV5F/12SV5R, Riaz et al. 2011) by species. 5916. en. Birdlife International. Buteo hemilasius. The IUCN Red List of Threatened Spe- cies 2018: e.T22695967A131937792. 2018b. https:// doi. org/ 10. 2305/ IUCN. Acknowledgements UK. 2018-2. RLTS. T2269 5967A 13193 7792. en. We would like to thank Mr. Yun Ma, Jiong He, and Rou Bao for their generous Birdlife International. Aquila nipalensis (amended version of 2017 assessment). assistance in the field surveys. The IUCN Red List of Threatened Species 2019: e.T22696038A155419092. 2019. https:// doi. org/ 10. 2305/ IUCN. UK. 2019-3. RLTS. T2269 6038A 15541 Authors’ contributions 9092. en. CEH formulated study concept. BDH performed statistical analyses. LW led Bontzorlos VA, Peris SJ, Vlachos CG, Bakaloudis DE. The diet of barn owl in the field teams. CEH, LW, WC, YD, YL, JL, YX, YZ, HC, and YZ collected study samples. agricultural landscapes of central Greece. Folia Zool. 2005;54:99–110. CEH, WC, JY, and YJ performed DNA extractions. CEH, WC, and JY performed Boyer F, Mercier C, Bonin A, Le Bras Y, Taberlet P, Coissac E. Obitools: a unix- sequence read mapping and determination of dietary items. CEH, YZ, and JEJ inspired software package for DNA metabarcoding. Mol Ecol Resour. acquired funding for the study. CEH wrote the study manuscript. CEH, BDH, 2016;16:176–82. YZ, and JEJ interpreted study results. BDH, YZ, and JEJ edited the manuscript Chen D, Zhang X, Tan X, Wang K, Qiao Y, Chang Y. Hydroacoustic study of spa- and provided additional expertise. All authors read and approved the final tial and temporal distribution of Gymnocypris przewalskii (Kessler, 1876) in manuscript. Qinghai Lake, China. Environ Biol Fish. 2009;84:231–9. Cui Q, Su J, Jiang Z. Summer diet of two sympatric species of raptors Upland Funding Buzzard (Buteo hemilasius) and Eurasian Eagle Owl (Bubo bubo) in alpine This project was funded by the National Key Technology R&D Program of meadow: problem of coexistence. Pol J Ecol. 2008;56:173–9. China (2019YFA0607103), Snow Leopard Conservancy (G1900014, G2000019), Cleveland Metroparks Zoo (G1800082), Panthera Corporation and The Andrew Hacker et al. Avian Res (2021) 12:42 Page 10 of 11 Deagle BE, Thomas AC, McInnes JC, Clarke LJ, Vesterinen EJ, Clare EL, et al. Lai CH, Smith AT. Keystone status of plateau pikas (Ochotona curzoniae): Counting with DNA in metabarcoding studies: how should we convert effect of control on biodiversity of native birds. Biodivers Conserv. sequence reads to dietary data? Mol Ecol. 2019;28:391–406. 2003;12:1901–12. Dixon A, Maming R, Gunga A, Purev-Ochir G, Batbayar N. The problem of Lamb PD, Hunter E, Pinnegar JK, Creer S, Davies RG, Taylor MI. How quantitative raptor electrocution in Asia: case studies from Mongolia and China. Bird is metabarcoding: a meta-analytical approach. Mol Ecol. 2019;28:420–30. Conserv Int. 2013;23:520–9. Lei F. A study on diet of the Little Owl (Athene noctua plumipes) in Qishan, Edgar RC. MUSCLE: multiple sequence alignment with high accuracy and high Shanxi Province, China. Wuyi Sci J. 1995;12:136–42 (in Chinese). throughput. Nucleic Acids Res. 2004;32:1792–7. Li L, Yi X, Li M, Zhang X. Diet composition of upland buzzard: analysis on stom- Emmrich M, Düttmann H. Seasonal shifts in diet composition of Great Cormo- ach content and food pellet. Zool Res. 2004;25:162–5 (in Chinese). rants Phalacrocorax carbo sinensis foraging at a shallow eutrophic inland Liu X, Chen B. Climatic warming in the Tibetan Plateau during recent decades. lake. Ardea. 2011;99:207–16. Int J Climatol. 2000;20:1729–42. Estes JA, Terborgh J, Brashares JS, Power ME, Berger J, Bond WJ, et al. Trophic Liu Y. International hunting and the involvement of local people, Dulan, downgrading of planet earth. Science. 2011;333:301–6. Qinghai, People’s Republic of China. Master’s Thesis. Missoula: University Ewins PJ, Weseloh DV, Groom JH, Dobos RZ, Mineau P. The diet of Herring of Montana; 1993. Gulls (Larus argentatus) during winter and early spring on the lower Great Marchesi L, Pedrini P, Sergio F. Biases associated with diet study methods in the Lakes. Hydrobiologia. 1994;279(280):39–55. Eurasian Eagle-Owl. J Raptor Res. 2002;36:11–6. Foggin JM. Depopulating the Tibetan grasslands: national policies and per- Margalida A, Bertran J, Boudet J. Assessing the diet of nestling Bearded spectives for the future of Tibetan herders in Qinghai Province, China. Mt Vultures: a comparison between direct observation methods. J Field Res Dev. 2008;28:26–31. Ornithol. 2005;76:40–5. Freeland JR. The importance of molecular markers and primer design Margalida A, Bertran J, Heredia R. Diet and food preferences of the endan- when characterizing biodiversity from environmental DNA. Genome. gered Bearded Vulture Gypaetus barbatus: a basis for their conservation. 2017;60:358–74. Ibis. 2009;151:235–43. Galan M, Pagès M, Cosson JF. Next-generation sequencing for rodent barcod- Meusnier I, Singer GAC, Landry JF, Hickey DA, Hebert PDN, Hajibabaei M. A ing: species identification from fresh, degraded and environmental universal DNA mini-barcode for biodiversity analysis. BMC Genomics. samples. PLoS ONE. 2012;7:e48374. 2008;9:214. Gotelli NJ, Hart EM, Ellison AM. EcoSimR: null model analysis for ecological Miller C, McEwen L. Diet of nesting Savannah Sparrows in interior Alaska. J data. R Package version 0.1.0. 2015. https:// doi. org/ 10. 5281/ zenodo. Field Ornithol. 1995;66:152–8. 16522. Miller DJ, Bedunah DJ. Rangelands of the Kunlun Mountains in Western China. Granjon L, Bruderer C, Cosson JF, Dia AT, Colas F. The small mammal commu- Rangelands. 1994;16:71–6. nity of a coastal site of south-west Mauritania. Afr J Ecol. 2002;40:10–7. Moody T. A method for obtaining food samples from insectivorous birds. Auk. Grier JW. Ban of DDT and subsequent recovery of reproduction in bald eagles. 1970;87:579. Science. 1982;218:1232–5. Musser GG, Carleton MD. Superfamily Muroidea. In: Wilson D, Reeder D, Guimaraes S, Fernandez-Jalvo Y, Stoetzel E, Gorgé O, Bennett EA, Denys C, editors. Mammal species of the world: a taxonomic and geographic refer- et al. Owl pellets: a wise DNA source for small mammal genetics. J Zool. ence. 3rd ed. Baltimore: John Hopkins University; 2005. p. 894–1531. 2016;298:64–74. Oehm J, Thalinger B, Eisenkölbl S, Traugott M. Diet analysis in piscivorous birds: Hacker CE, Jevit M, Hussain S, Muhammad G, Munkhtsog B, Munkhtsog B, et al. what can the addition of molecular tools offer? Ecol Evol. 2017;7:1984–95. Regional comparison of snow leopard (Panthera uncia) diet using DNA Oksanen J, Blanchet FG, Kindt R, Legendre P, Minchin PR, O’Hara RB, et al. Pack- metabarcoding. Biodivers Conserv. 2021;30:797–817. age vegan: community ecology package. R Package. version 2.3-1. 2013. Hebert PDN, Cywinska A, Ball SL, deWaard JR. Biological identifications Pianka ER. Niche overlap and diffuse competition. Proc Natl Acad Sci USA. through DNA barcodes. Proc R Soc B Biol Sci. 2003;270:313–21. 1974;71:2141–5. Heck KL, van Belle G, Simberloff D. Explicit calculation of the rarefaction Pimm SL, Russell GJ, Gittleman JL, Brooks TM. The future of biodiversity. Sci- diversity measurement and the determination of sufficient sample size. ence. 1995;269:347–50. Ecology. 1975;56:1459–61. Pompanon F, Deagle BE, Symondson WOC, Brown DS, Jarman SN, Taberlet P. Hill MO. Diversity and evenness: a unifying notation and its consequences. Who is eating what: diet assessment using next generation sequencing. Ecology. 1973;54:427–32. Mol Ecol. 2012;21:1931–50. Hiraldo F, Andrada J, Parreño F. Diet of the eagle owl (Bubo bubo) in Mediter- Rasmussen P, Anderson J. Birds of South Asia: the ripley guide. Washington, ranean Spain. Doñana Acta Vertebr. 1975;2:161–77. DC, and Barcelona: Smithsonian Institution and Lynx Edicions; 2005. Hsieh TC, Ma KH, Chao A. iNEXT: an R package for rarefaction and extrapolation Reading R, Michel S, Amgalanbaatar S. Ovis ammon. The IUCN Red List of of species diversity (Hill numbers). Method Ecol Evol. 2016;7:1451–6. Threatened Species 2020: e.T15733A22146397. 2020. https:// doi. org/ 10. Hurlbert SH. The nonconcept of species diversity: a critique and alternative 2305/ IUCN. UK. 2020-2. RLTS. T1573 3A221 46397. en. parameters. Ecology. 1971;52:577–86. Riaz T, Shehzad W, Viari A, Pompanon F, Taberlet P, Coissac E. ecoPrimers: Iverson SJ, Springer AM, Kitaysky AS. Seabirds as indicators of food web struc- inference of new DNA barcode markers from whole genome sequence ture and ecosystem variability: qualitative and quantitative diet analyses analysis. Nucleic Acids Res. 2011;39:e145. using fatty acids. Mar Ecol Prog Ser. 2007;352:235–44. Sándor AD, Ionescu DT. Diet of the eagle owl (Bubo bubo) in Braşov, Romania. Jackson R. Fostering community-based stewardship of wildlife in Central Asia: North West J Zool. 2009;5:170–8. transforming snow leopards from pests into valued assets. In: Squires VR, Schaller GB. Wildlife of the Tibetan Steppe. Chicago: University of Chicago editor. Rangeland stewardship in Central Asia. Dordrecht: Springer; 2012. Press; 1998. p. 357–80. Schaller GB. Tibet Wild. Washington: Island Press; 2012. Janečka JE, Jackson R, Yuquang Z, Diqiang L, Munkhtsog B, Buckley-Beason V, Schaller GB, Junrang R, Mingjiang Q. Status of the snow Leopard Pan- et al. Population monitoring of snow leopards using noninvasive collec- thera uncia in Qinghai and Gansu Provinces, China. Biol Conserv. tion of scat samples: a pilot study. Anim Conserv. 2008;11:401–11. 1988;45:179–94. Janečka JE, Munkhtsog B, Jackson RM, Naranbaatar G, Mallon DP, Murphy WJ. Schoener TW. Competition and the form of habitat shift. Theor Popul Biol. Comparison of noninvasive genetic and camera-trapping techniques for 1974;6:265–307. surveying snow leopards. J Mammal. 2011;92:771–83. Shannon C, Weaver W. The mathematical theory of communication. Urbana: Jedlicka JA, Vo ATE, Almeida RPP. Molecular scatology and high-throughput University of Illinois Press; 1949. sequencing reveal predominately herbivorous insects in the diets of Shehzad W, McCarthy TM, Pompanon F, Purevjav L, Coissac E, Riaz T, et al. Prey adult and nestling Western Bluebirds (Sialia mexicana) in California vine- preference of snow leopard (Panthera uncia) in South Gobi, Mongolia. yards. Auk. 2017;134:116–27. PLoS ONE. 2012;7:e32104. Kartzinel TR, Chen PA, Coverdale TC, Erickson DL, Kress WJ, Kuzmina ML, et al. Shi ZY, Yang CQ, Hao MD, Wang XY, Ward RD, Zhang AB. FuzzyID2: a software DNA metabarcoding illuminates dietary niche partitioning by African package for large data set species identification via barcoding and large herbivores. Proc Natl Acad Sci USA. 2015;112:8019–24. Hacker  et al. Avian Res (2021) 12:42 Page 11 of 11 metabarcoding using hidden Markov models and fuzzy set methods. Mol associated with dietary shifts in an obligate riparian Neotropical migra- Ecol Resour. 2018;18:666–75. tory songbird. PeerJ. 2018a;16:e5141. Simon C, Frati F, Beckenbach A, Crespi B, Liu H, Flook P. Evolution, weighting, Trevelline BK, Nuttle T, Hoenig BD, Brouwer NL, Porter BA, Latta SC. DNA and phylogenetic utility of mitochondrial gene sequences and a compi- metabarcoding of nestling feces reveals provisioning of aquatic prey and lation of conserved polymerase chain reaction primers. Ann Entomol Soc resource partitioning among Neotropical migratory songbirds in a ripar- Am. 1994;87:651–701. ian habitat. Oecologia. 2018b;187:85–98. Simpson EH. Measurement of diversity. Nature. 1949;163:688. Treves A, Krofel M, McManus J. Predator control should not be a shot in the Smith AT, Formozov NA, Hoffman RS, Zeng C, Erbajeva MA. The pikas. In: dark. Front Ecol Environ. 2016;14:380–8. Chapman J, Flux J, editors. Rabbits, hares and pikas. Status survery and Valera F, Gutiérrez JE, Barrios R. Eec ff tiveness, biases and mortality in the use of conservation action plan. Chapter 3. Gland, Switzerland: IUCN; 1990. pp. apomorphine for determining the diet of granivorous passerines. Condor. 14–60. 1997;99:765–72. Smith AT, Foggin JM. The plateau pika (Ochotona curzoniae) is a keystone spe- Walter CB, O’Neill E. Electrophoresis in the study of diets and digestive rates of cies for biodiversity on the Tibetan plateau. Anim Conserv. 1999;2:235–40. seabirds. Comp Biochem Physiol A Comp Physiol. 1986;84:763–5. Soergel DAW, Dey N, Knight R, Brenner SE. Selection of primers for optimal Wei WR, He JD, Zheng QY. Plateau pikas (Ochotona curzoniae) at low densities taxonomic classification of environmental 16S rRNA gene sequences. have no destructive effect on winter pasture in alpine meadows. Rangel ISME J. 2012;6:1440–4. J. 2020;42:55–61. Symondson WOC. Molecular identification of prey in predator diets. Mol Ecol. Wilson RP. An improved stomach pump for penguins and other seabirds. J 2002;11:627–41. Field Ornithol. 1984;55:109–12. Taberlet P, Fumagalli L. Owl pellets as a source of DNA for genetic studies of Xia W, Zhou X, Liu J, Zhang X. The bio-community in the region of alpine small mammals. Mol Ecol. 1996;5:301–5. meadow. In: Liu J, Wang Z, editors. Alpine meadow ecosystem. Beijing, Thiam M, Bâ K, Duplantier JM. Impacts of climatic changes on small mammal China: Science Press; 1991. p. 1–7. communities in the Sahel ( West Africa) as evidenced by owl pellet analy- Yang Z, Gong M, Huang X, Li C, Liu X. Biodiversity of birds of prey in Shiqu sis. African Zool. 2008;43:135–43. county. J Sichuan Teach Coll. 2000;21:137–40 (in Chinese). Tremblay I, Thomas D, Blondel J, Perret P, Lambrechts MM. The effect of habitat Zhang R, Ludwig A, Zhang C, Tong C, Li G, Tang Y, et al. Local adaptation of quality on foraging patterns, provisioning rate and nestling growth in Gymnocypris przewalskii (Cyprinidae) on the Tibetan Plateau. Sci Rep. Corsican Blue Tits Parus caeruleus. Ibis. 2005;147:17–24. 2015;5:9780. Trevelline BK, Nuttle T, Porter BA, Brouwer N, Hoenig BD, Steffensmeier ZD, Latta SC. Stream acidification and reduced migratory prey availablity are Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Avian Research Springer Journals

Use of DNA metabarcoding of bird pellets in understanding raptor diet on the Qinghai-Tibetan Plateau of China

Download PDF
11 pages

Loading next page...
 
/lp/springer-journals/use-of-dna-metabarcoding-of-bird-pellets-in-understanding-raptor-diet-ECkaakwikH

References (73)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2021
eISSN
2053-7166
DOI
10.1186/s40657-021-00276-3
Publisher site
See Article on Publisher Site

Abstract

Background: Diet analysis is essential to understanding the functional role of large bird species in food webs. Mor- phological analysis of regurgitated bird pellet contents is time intensive and may underestimate biodiversity. DNA metabarcoding has the ability to circumvent these issues, but has yet to be done. Methods: We present a pilot study using DNA metabarcoding of MT-RNR1 and MT-CO1 markers to determine the species of origin and prey of 45 pellets collected in Qinghai and Gansu Provinces, China. Results: We detected four raptor species [Eurasian Eagle Owl (Bubo bubo), Saker Falcon (Falco cherrug), Steppe Eagle (Aquila nipalensis), and Upland Buzzard (Buteo hemilasius)] and 11 unique prey species across 10 families and 4 classes. Mammals were the greatest detected prey class with Plateau Pika (Ochotona curzoniae) being the most frequent. Observed Shannon’s and Simpson’s diversity for Upland Buzzard were 1.089 and 0.479, respectively, while expected values were 1.312 ± 0.266 and 0.485 ± 0.086. For Eurasian Eagle Owl, observed values were 1.202 and 0.565, while expected values were 1.502 ± 0.340 and 0.580 ± 0.114. Interspecific dietary niche partitioning between the two spe - cies was not detected. Conclusions: Our results demonstrate successful use of DNA metabarcoding for understanding diet via a novel noninvasive sample type to identify common and uncommon species. More work is needed to understand how raptor diets vary locally, and the mechanisms that enable exploitation of similar dietary resources. This approach has wide ranging applicability to other birds of prey, and demonstrates the power of using DNA metabarcoding to study species noninvasively. Keywords: Avian, Eurasian Eagle Owl, Molecular diet analysis, Next-generation sequencing, Raptor, Saker Falcon, Steppe Eagle, Upland Buzzard Background Understanding predator–prey interactions  is an impor- tant component of community ecology and management (Estes et al. 2011). Sympatric species with similar ecologi- *Correspondence: yugzhang@sina.com; janeckaj@duq.edu cal demands must find ways to reduce competition. One Research Institute of Forest Ecology, Environment and Protection, Chinese Academy of Forestry, and Key Laboratory of Biodiversity way this is accomplished is through dietary niche parti- Protection of National Forestry and Grassland Administration, tioning (Schoener 1974). Understanding this overlap can Beijing 100091, China discern how species allocate resources. Such knowledge Department of Biological Sciences, Duquesne University, Pittsburgh, PA 15282, USA is important for conservation planning as communities Full list of author information is available at the end of the article © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Hacker et al. Avian Res (2021) 12:42 Page 2 of 11 with lower niche overlap can support greater biodiver- MT-RNR1 is an RNA gene previously used in studies sity (Pianka 1974). Unfortunately, food web dynamics examining the identity and genetic relationships of ani- are complex and require accurate information of items mals (Riaz et  al. 2011). However, MT-RNR1 lacks the consumed (Pompanon et  al. 2012). Understanding the genetic diversity to discern wild versus domestic goat trophic niche of birds is important as knowledge of prey (Capra hircus) and sheep species (Shehzad et  al. 2012; composition plays an important role in shaping conser- Hacker et  al. 2021). This is problematic for the discern - vation policies (Grier 1982). Dietary assessment methods ment of domestic sheep and argali (Ovis ammon), as for avian species include direct observation (Margalida well as domestic goat and Siberian Ibex (Capra sibirica), et  al. 2005, 2009), camera placement at nest sites (Mar- in areas where they are sympatric (Reading et  al. 2020). galida et al. 2005; Tremblay et al. 2005), stomach pump- uTh s, an additional marker is necessary. MT-COI is ing (Wilson 1984; Walter and O’Neill 1986), examination widely used for DNA barcoding as it has both conserved of stomach contents (Miller and McEwen 1995), diges- regions and segments with high divergence (Hebert et al. tive tract flushing (Moody 1970), forced vomiting (Valera 2003). Recent work identified a segment of MT-COI et  al. 1997), examination of fatty acid and isotope sig- capable of discerning wild and domestic goat and sheep natures (Iverson et  al. 2007), molecular fecal analysis taxa in Central Asia, making it an important addition (Treves et  al. 2016; Jedlicka et  al. 2017; Trevelline et  al. for diet research of predators and scavengers distributed 2018a, b) and morphological investigation of pellets (Li there (Hacker et al. 2021). et al. 2004). Numerous sympatric birds of prey are found on the Bird pellets are the accumulation of undigested prey Qinghai-Tibetan Plateau (QTP). They play vital roles in that are regurgitated through the mouth in compact maintaining ecosystem balance (Xia et al. 1991) under the units (Taberlet and Fumagalli 1996). Examination of context of vast modernization and environmental shifts diet via pellets has historically involved collection and (Liu and Chen 2000; Foggin 2008). They face threats morphological assessment of contents (Ewins et  al. associated with climate change (Liu and Chen 2000) as 1994; Symondson 2002; Sándor and Ionescu 2009). This well as anthropogenically-induced mortality, such as method has many shortcomings. Digestive processes may electrocution from power lines (Dixon et  al. 2013) and render samples unrecognizable (Symondson 2002; Galan pika poisoning (Badingqiuying et  al. 2016). Raptor spe- et  al. 2012), pellet appearance can vary widely based on cies include the Eurasian Eagle Owl (Bubo bubo) (Birdlife life stage and sex (Galan et  al. 2012), small to medium International 2017), Upland Buzzard (Buteo hemilasius), size prey are often overestimated and unusual prey items Northern Goshawk (Accipiter gentilis), Golden Eagle unrecorded (Marchesi et  al. 2002), and taxonomic spe- (Aquila chrysaetos), Saker Falcon (Falco cherrug) (Dixon cialists are needed for wide ranging genera (Galan et  al. et  al. 2013; Birdlife International 2018a), Steppe Eagle 2012). (Aquila nipalensis), Himalayan Vulture (Gyps himalayen- Molecular approaches involving the examination sis), and Bearded Vulture (Gypaetus barbatus) (Schaller of DNA can circumvent these issues and increase the 1998; Cui et al. 2008; Birdlife International 2019). Under- detectable prey spectrum relative to sampling and analy- standing their diets is necessary for effective conserva - sis effort (Oehm et al. 2017). DNA sequencing techniques tion action, and DNA metabarcoding of regurgitated have been called for since the early 2000s (Symondson pellets provides a method to do so noninvasively. 2002), but remain under used despite the stability of The aims of this study were to (1) determine applicabil - DNA in bird pellets (Taberlet and Fumagalli 1996), and ity of DNA metabarcoding to species and prey identifi - the advent of Next-Generation Sequencing (NGS) tech- cation of avian pellets; (2) examine metrics surrounding nology which can identify species from a high volume species presence and mechanisms of coexistence; and (3) of samples (Galan et  al. 2012). In DNA metabarcoding, make suggestions based on our results for conservation generalized primers target and amplify a segment, the action planning. DNA barcoding region, of a conserved gene (Pompanon et  al. 2012). This gene must have low intra-species vari - Methods ation (Galan et al. 2012) and high inter-species variation Study site for taxonomic classification (Simon et al. 1994). Degrada- Samples were collected in the Qilian Shan Mountains tion of longer DNA segments remains problematic. Thus, (hereafter “Qilian Shan”) of Qinghai and Gansu Prov- researchers rely on shorter segments known as “mini- inces, China, in the eastern Kunlan Mountains in Dulan barcodes” (Meusnier et  al. 2008). Researchers also use County, Qinghai Province, China and in Zhiduo County, mitochondrial genes, as they have higher copy numbers Yushu Prefecture, Qinghai Province, China (Fig.  1). Qil- compared to nuclear genes and greater PCR amplifica - ian Shan runs along the northeastern corner of the QTP tion success (Freeland 2017). (> 3000  m above sea level). It comprises three parallel Hacker  et al. Avian Res (2021) 12:42 Page 3 of 11 September 2017 to July 2019. Sampling methods are described in Janečka et al. (2008, 2011). DNA extraction Samples were stored in a − 20  °C freezer. Pellets were defrosted and individually placed in a Petri dish. Twee- zers were then used to remove material from  the out- side and inside the pellet to capture DNA of both the prey and host and placed in a 1.5  mL centrifuge tube. DNA was extracted using QIAamp DNA Stool Mini Kits (QIAGEN, Hilden, Germany) with two additional centrifugation steps to remove residual Buffer AW1 and AW2. Aliquots were quantified using a NanoDrop Lite Spectrophotometer. PCR for species and diet analysis A short segment (~ 100-bp) of MT-RNR1 (primers 12SV5F/12SV5R; Riaz et  al. 2011) was used to discern all predator and prey species except those belonging to Fig. 1 Bird pellet collection locations and their identified host goats and sheep. The primers for MT-CO1 (MT-CO1- species. Host species was determined by sequencing of a segment 379F and MT-CO1-604Rd; Hacker et  al. 2021) were of the MT-RNR1. Figure made in QGIS and modified using Microsoft designed to amplify a gene segment (330-bp) capable of PowerPoint for Mac version 16.42 differentiating closely related caprines. Each segment was amplified separately using a PCR reaction containing 1.5 µL of DNA template, 7.94 µL of KAPA HiFi HotStart ReadyMix (2 ×) (Kapa Biosystems, Wilmington, MA, subsidiary ranges—the Tuali Nanshan, Shule Nanshan, USA), 0.16 µL of 20 µM forward primer, 0.16 µL of 20 µM and Danghe Nanshan (Schaller et  al. 1988). Qilian Shan reverse primer, and 5.2 µL of PCR grade water. PCR con- is composed of deserts at lower elevations giving way ditions consisted of 95 °C for 3 min, 35 cycles of 95 °C for to shrubs, grasses, and alpine meadows (Schaller et  al. 30 s, 60 °C for 30 s, and 72 °C for 30 s, followed by a 5 min 1988). Yushu Prefecture is in the southwestern corner extension step at 72 °C and 4 °C hold. of Qinghai Province, and has alpine meadow vegetation with small rugged ranges surrounded by rolling grass- Next‑generation sequencing land, with juniper forests along mountainsides (Schaller Amplicons were mixed in equal ratios determined by et  al. 1988). The Kunlan Mountains are the longest measuring gel band brightness using GeneTools Analy- mountain system in Asia with its eastern end south west sis Software Version 4.03.05.0 (SynGene, Frederick, MD, of Qilian Shan (Miller and Bedunah 1994). The landscape USA). The E.Z.N.A. Gel Extraction Kit (Omega Bio-Tek, of Dulan County is primarily rugged grassland with rock Inc., Norcross, GA, USA) was used to isolate and purify slopes (Liu 1993). Species found on the QTP include ® ™ products. The NEBNext Ultra II DNA Library Prep Kit the Blue Sheep (Pseudois nayaur), Tibetan Gazelle (Pro- for Illumina (New England Biolabs, MA, USA) was used capra picticaudata), Argali (Ovis ammon), and White- to prepare sequencing libraries. Indexing oligonucleo- lipped Deer (Cervus albirostris), and carnivores such as tides (Nextera XT, Illumina, San Diego, CA, USA) were the Tibetan Wolf (Canis lupus), Red Fox (Vulpes vulpes), incorporated and the pooled ampliconic library quanti- Snow Leopard (Panthera uncia), Eurasian Lynx (Lynx fied using an Invitrogen Qubit 2.0 Fluorometer (Thermo lynx), and Pallas’s Cat (Otocolobus manul), among others. Fisher Scientific, Waltham, MA, USA). Paired-end 250- Small mammals include voles (Neodon sp.), pikas (Ocho- bp sequencing was completed on an Illumina NovaSeq tona sp.), zokors (Myospalax sp.), Woolly Hare (Lepus 6000 by Guangdong Magigene Biotechnology Co., Ltd. oiostolus), and Himalayan Marmots (Marmota himaly- (Guangzhou, China). ana) (Schaller et al. 1988; Jackson 2012). Diet analysis Sample collection FASTQ sequences were demultiplexed, sequencing Permits were obtained prior to sample collection. Pel- adapters removed, and reads imported into CLC Genom- lets were collected opportunistically as part of a sepa- ics Workbench v12.0 (CLC bio, QIAGEN, Aarhus, rate snow leopard study over seven sampling trips from Hacker et al. Avian Res (2021) 12:42 Page 4 of 11 Denmark). Raw sequencing reads were trimmed using a by the total number of pellets for each predator species. quality score limit of 0.1. A reference FASTA file was cre - Observed dietary taxonomic richness, as well as Shan- ated by downloading MT-CO1 and MT-RNR1 sequences non’s index (Shannon and Weaver 1949) and Simpson’s for potential prey and host species, which were deter- index (Simpson 1949; Marchesi et  al. 2002) were cal- mined via a literature search and by consulting with local culated using the “iNEXT” function (Hsieh et  al. 2016). experts (Additional file  1). Sequencing reads were then Effective taxonomic richness and diversity of predator mapped to the reference FASTA file using local align - diet was calculated using Hill numbers (Hill 1973) and ment with the following parameters—mismatch cost: the “iNEXT” function (Hsieh et  al. 2016). This enabled 2; insertion cost: 3; deletion cost: 3; length fraction: 0.9; comparison of dietary diversity between species with similarity: 0.94; non-specific matches mapped randomly varying sample sizes (Hurlbert 1971; Heck et  al. 1975) (Hacker et al. 2021). To identify the host species of each and allowed for evaluation of sampling completeness. pellet, identification was made when the sequenced reads For the Upland Buzzard and the Eurasian Eagle Owl, were mapped to a reference sequence with > 98% similar- Pianka’s metric of niche overlap (Pianka 1974) was cal- ity. Unique haplotypes of the MT-RNR1 segment for each culated using the ‘niche_null_model’ function in the Eco- host species were determined in MEGA 7.0 by aligning SimR package (Gotelli et  al. 2015) and compared to 999 sequences using MUSCLE (Edgar 2004). Prey identifi - null model simulations to determine if observed dietary cation was made by selecting the prey species from the niche overlap was higher or lower than expected. The reference database with the highest number of mapped occurrence of interspecific niche differentiation between reads and fewest number of mismatches. To ensure prey these two species was determined by performing a DNA sequences were correctly classified, the consensus permutational multivariate analysis of variance (PER- sequence for the host and prey items were extracted and MANOVA; Anderson 2017) using the “adonis” function a blastn search performed against the nr/nt nucleotide and Jaccard distance matrix within the vegan R package collection GenBank databases using megablast for highly (Oksanen et al. 2013). Additional information on statisti- similar sequences (Additional file  2). In addition, collec- cal analyses can be found in Additional file 3. tion sites were compared with known species distribu- tions (CITES Red List range maps, https:// www. iucnr edlist. org/ search/ map). Results Samples with a large proportion of unmapped reads Predator species detected (> 5% of total reads) were analyzed to rule out an incom- Using the MT-RNR1 genetic marker, four host bird spe- plete reference file. This was done by performing a de cies were identified among 45 pellets—Upland Buzzard novo assembly with the following parameters—minimum (n = 26, 23 from Qilian Shan and 3 from Yushu), Eura- contig length of 100; mismatch cost: 2; insertion cost: sian Eagle Owl (n = 15, 14 from Qilian Shan and 1 from 3; deletion cost: 3; length fraction: 0.9; similarity: 0.98. Yushu), Steppe Eagle (n = 3, from Dulan County), and Consensus sequences were extracted for contigs with the Saker Falcon (n = 1, from Qilian Shan) (Fig. 2). A total of highest number of mapped reads. At least 10,000 reads 8 unique MT-RNR1 haplotypes were found for Upland were required to generate a consensus sequence. Nucle- Buzzard, 5 for Eurasian Eagle Owl, 1 for Saker Falcon, otides in sites with conflicting reads were resolved via and 1 for Steppe Eagle (Additional file 4). majority rule and ambiguous sites were coded with an “N”. Species were identified using the same blastn search described above. Predator diet composition A total of 41 of the 45 collected pellets (91.1%) had prey DNA sequences discerned. A total of 11 unique prey taxa Statistical analysis spanning 7 orders and 4 classes were detected (Table  1). All statistical analyses were performed in R version The most frequently detected prey species, Plateau Pika 3.5.2 (R Core Team 2018) using base R functions, vegan (Ochotona curzoniae), was found in 77.8% (35/45) of all (Oksanen et  al. 2013), iNEXT (Hsieh et  al. 2016), and diets analyzed and was the only prey taxa detected in the EcoSimR packages (Gotelli et  al. 2015). The mean num - Saker Falcon samples (Table 1). While Mammalia was the ber of unique prey taxa per pellet was calculated by sum- most taxonomically-rich prey class (7 species; Table  1), ming the number of detected prey species among each predator species in this study also consumed Ray-finned pellet and dividing this sum by the number of pellets Fishes (Actinopterygii, 2 species), small birds (Aves), and analyzed. This calculation was also repeated for each bird one species of toad (Amphibia). species sampled for interspecific comparisons. Dietary frequency of occurrence was calculated by dividing the number of pellets in which a prey species was detected Hacker  et al. Avian Res (2021) 12:42 Page 5 of 11 Fig. 2 The host bird species identified via DNA-metabarcoding of bird pellets collected on the Qinghai-Tibetan Plateau. a Steppe Eagle (photo credit: Jia Li). b Upland Buzzard (photo credit: Charlotte Hacker). c Saker Falcon (photo credit: Charlotte Hacker). d Eurasian Eagle Owl (photo credit: Munib Khanyari). Figure made in Microsoft PowerPoint for Mac version 16.42 Dietary richness, diversity and overlap each for Steppe Eagle and Saker Falcon. Steppe Eagle The average number of unique prey taxa detected in (n = 3) and Saker Falcon (n = 1) were removed from fur- a single pellet across all species was 1.16 (range 0–3), ther analyses due to limited sample sizes. with Upland Buzzard and Eurasian Eagle Owl having Though these data are preliminary, they provide an an average of 1.16 and 1.13 unique prey taxa per pellet, important starting point for understanding dietary over- respectively. We observed the average number of unique lap between two large bird species. The expected dietary prey taxa per pellet for Saker Falcon and Steppe Eagle richness of Upland Buzzard and Eurasian Eagle Owl was in our study to be 1 and 0.333, respectively. Our analy- 14.692 ± 11.227 (mean ± SE) and 11.6 ± 6.594, respec- sis detected 7 prey species within pellets collected from tively (Fig.  3). DNA barcoding detected 87.3% of the Upland Buzzard, 6 from Eurasian Eagle Owl, and one Upland Buzzard’s and 79.9% of the Eurasian Eagle Owl’s Hacker et al. Avian Res (2021) 12:42 Page 6 of 11 Table 1 The number (n) and percent frequency (%) of prey species identified Prey item Upland Buzzard Eurasian Eagle Owl Steppe Eagle Saker Falcon (Buteo hemilasius) (Bubo bubo) (Aquila nipalensis) (Falco cherrug) 26 pellets 15 pellets 3 pellets 1 pellet n % n % n % n % Plateau Pika (Ochotona curzoniae) 22 84.60 12 80.00 0 0 1 100.00 Plateau Vole (Neodon fuscus) 3 11.50 0 0 0 0 0 0 Sikkim Mountain Vole (Neodon sikimensis) 0 0 1 6.67 0 0 0 0 Woolly Hare (Lepus oiostolus) 1 3.85 3 20.00 1 33.33 0 0 Tibetan Snow Finch (Montifringilla henrici) 2 7.69 0 0 0 0 0 0 Long-tailed Dwarf Hamster (Cricetulus longicaudatus) 0 0 1 6.67 0 0 0 0 Przewalksi’s Naked Carp (Gymnocypris przewalskii) 0 0 1 6.67 0 0 0 0 Chinese Perch (Siniperca sp.) 1 3.85 0 0 0 0 0 0 Lazy Toad (Scutinger sp.) 0 0 1 6.67 0 0 0 0 Domestic Goat (Capra hircus) 1 3.85 0 0 0 0 0 0 Blue Sheep (Pseudois nayaur) 1 3.85 0 0 0 0 0 0 Taxonomic richness 7 6 1 1 dietary niche overlap was calculated to be 0.511, within the expected 95% confidence interval derived from null model simulations (CI range: 0.477–0.522). Discussion Pika was the dominant consumed species in this study. Pikas are non-hibernating and diurnal, providing a year-long accessible food source (Badingqiuying et  al. 2016). Pikas have been labeled as pests, with a poison- ing program to control and eradicate them launched in 1958 (Smith et  al. 1990; Smith and Foggin 1999). How- ever, more recent research recognizes pikas as important environmental engineers that cause minimal damage to alpine grassland ecosystems (Smith and Foggin 1999; Wei Fig. 3 Estimation of prey taxonomic richness with rarefaction and et  al. 2020). Poisoning them is more likely to have nega- extrapolation. Using the iNEXT function, rarefaction curves were tive impacts (Lai and Smith 2003; Badingqiuying et  al. made to estimate sample completeness and expected taxonomic 2016). In a study by Badingqiuying et  al. (2016), record- richness with additional sampling for the Eurasian Eagle Owl ings of Steppe Eagles, Saker Falcons, and Upland Buz- (n = 15) and Upland Buzzard (n = 26). Interpolation (solid lines) was performed using the sample sizes for each predator species while zards were three times less frequent at sites with active extrapolation (dotted lines) of species richness was performed on poisoning programs compared to those without; though double the sample size. Figure made in R version 3.5.2 data were insufficient to make any conclusions about Eurasian Eagle Owls. Similarly, Lai and Smith (2003) reported that Upland Buzzards were found 11.2 times expected prey taxa. Observed Shannon’s and Simp- more frequently at non-poisoned versus poisoned pika son’s diversity for Upland Buzzard were 1.089 and 0.479, sites. respectively, while expected values were 1.312 ± 0.266 The one pellet collected from a Saker Falcon revealed and 0.485 ± 0.086. Observed Shannon’s and Simp- pika as the dietary item. In previous studies, pikas were son’s diversity for Eurasian Eagle Owl were 1.202 and found to comprise 90% of the food items a Saker Fal- 0.565, while expected values were 1.502 ± 0.340 and con pair fed to their young in the Chang Tang  region 0.580 ± 0.114, respectively. Our PERMANOVA results (Schaller 2012). In this study, pikas were detected in 80% did not detect interspecific dietary niche partition - of pellets from Eurasian Eagle Owl and 84.6% of Upland ing between Upland Buzzard and Eurasian Eagle Owl Buzzard pellets, aligning with previous studies. Our (Pseudo-F = 1.86, p = 0.129) and Pianka’s index of (1, 36) results further substantiate the important role pika play Hacker  et al. Avian Res (2021) 12:42 Page 7 of 11 as a primary food source for large bird species. However, occur at different parts of the day. Across a broader some species may be able to adapt to reductions in pika temporal scale, Upland Buzzards partially migrate, populations by exploiting other prey species (Cui et  al. spending the breeding season in China, while Eurasian 2008). Eagle Owls are not migratory (Birdlife International Dietary diversity assessments could only be done with 2017). However, with the exception of two samples Upland Buzzards and Eurasian Eagle Owls due to the collected in September, all remaining pellets were col- low sample size of Saker Falcon and Steppe Eagle pellets. lected during the corresponding breeding season for While the Upland Buzzard had one more unique spe- Upland Buzzards (April through July) (Rasmussen and cies represented in their diet compared to the Eurasian Anderson 2005), and both would be expected to be pre- Eagle Owl, the observed diversity indices suggested that sent. Eurasian Eagle Owls and Upland Buzzards also Eurasian Eagle Owls exhibit greater dietary taxonomic deploy different hunting strategies  which may separate diversity. The values resulting from the Shannon-Weiner them  behaviorally. Eurasian Eagle Owls are oppor- diversity index and Simpson’s index between the two spe- tunistic (Hiraldo et  al. 1975), swooping down on prey cies were relatively similar with a difference of 0.113 for and departing whether the kill is successful or not, Shannon-Weiner and 0.086 for Simpson’s. Previous work while Upland Buzzards deploy a sit-and-wait strategy by Cui et al. (2008), who compared the dietary diversity of (Cui et  al. 2008). In addition, we observed ungulates Upland Buzzards and Eurasian Eagle Owls via morpho- (Domestic Goat and Blue Sheep) in two Upland Buz- logical assessment of regurgitated pellets found that Eur- zard pellets, likely attributed to scavenging. asian Eagle Owls exhibited higher, but similar diversity to Unique and previously under-reported prey items that of the Upland Buzzard with a difference between the were identified. It may be that birds at present are rap - two species of 0.17. When calculated to expected degrees idly adapting to changes in prey base; however, it is also of dietary diversity, differences between Upland Buzzard possible that traditional methods reliant on the mor- and Eurasian Eagle Owl in this study were a bit larger, phological assessment of pellets may have missed rare with a difference of 0.19 expected for Shannon-Weiner or unexpected prey items. In this study, Przewalski’s and 0.095 for Simpson’s. Greater expected dietary diver- Naked Carp (Gymnocyris przewalksii) was found in one sity for Eurasian Eagle Owls is not unexpected as they are Eurasian Eagle Owl pellet. Przewalski’s Naked Carp is known to be generalists (Hiraldo et al. 1975). found in Qinghai Lake and is classified as Endangered Despite the relatively small number of pellets collected, on the China Species Red List (Chen et al. 2009; Zhang the sampling effort in this study captured 87.3% of the et  al. 2015). Loss of the species could cause severe expected prey taxa for the Upland Buzzard and 79.9% negative consequences within the lake (Chen et  al. of expected prey taxa for the Eurasian Eagle Owl. Spe- 2009). A previous study examining the distribution of cies that have been previously recorded as present in the Przewalski’s Naked Carp in Qinghai Lake showed that diets of birds of prey living on the QTP but not found the species was found ~ 2  m below the surface (Chen in this study include Zokors (Myospalax fontanierii; et al. 2009), where individuals moving upward could be M. baileyi), Gansu Pika (Ochotona cansus), Root Voles caught by birds of prey. (Micrtous oeconomus), Plateau Voles (Lasiopodomys fus- Although DNA metabarcoding is sensitive enough to cus), and Mountain Weasels (Mustela altaica) (Lai and detect the presence of rare and elusive species (Gran- Smith 2003; Li et al. 2004; Cui et al. 2008). The presence jon et  al. 2002; Thiam et  al. 2008), the degree to which and frequency of food items in the diets of birds of prey they are being predated would be unknown due to are typically reflective of prey presence and frequency in the inability to count the number of individuals con- their habitats (Bontzorlos et  al. 2005). While data could sumed within in a pellet (Symondson 2002; Emmrich reflect low numbers or absence of expected species that and Düttmann 2011). This may be problematic as raw may spark concern surrounding conservation status of occurrence counts of dietary items can artificially wildlife, species absence in this dataset is most likely due inflate the occurrence of rare food taxa (Deagle et  al. to low sample size or a result of variable methodology in 2019). Previous research has suggested that the num- comparison to previous studies. ber of mapped reads may be quasi-indicative of the We found no evidence of dietary niche partitioning quantity of a dietary item within a sample (Kartzinel between Upland Buzzard and Eurasian Eagle Owl, indi- et  al. 2015; Deagle et  al. 2019; Lamb et  al. 2019). We cating that temporal or spatial partitioning is responsi- greatly caution against this assumption, as read map- ble for coexistence. Previous work by Cui et  al. (2008) ping success can be influenced by other factors such as had similar findings. Temporally, Eurasian Eagle Owls PCR amplification bias, sample quality, and reference are nocturnal while Upland Buzzards are diurnal (Lei sequence availability (Pompanon et al. 2012; Lamb et al. 1995; Yang et  al. 2000), and resource exploitation may 2019). Other potential pitfalls of DNA metabarcoding Hacker et al. Avian Res (2021) 12:42 Page 8 of 11 which should be considered include the level of DNA caprids in the diet of the Upland Buzzard, one match- degradation and whether the genetic marker used pro- ing to Domestic Goat and one to Blue Sheep. Unfor- vides necessary taxonomic resolution. tunately, and as would be an issue with morphological In this study, four of the pellets collected were unable analyses, DNA metabarcoding is unable to distinguish to have prey DNA amplified. This may be because the between scavenging versus legitimate kills (Symondson primer pairs used did not amplify the prey’s target gene, 2002). Larger birds of prey, such as the Golden Eagle or could be a result of DNA degradation. The QTP is a have been observed taking larger prey such as domestic dry, cold environment. Regardless, DNA degradation calves and sheep in North America (Avery and Cum- may make the rendering of sequences from some samples mings 2004). Upland Buzzards are considerably smaller impossible (Guimaraes et  al. 2016). Shorter diagnostic in size, and would most likely not be able to kill caprids; fragments are preferred as they are more likely to remain with the exception of newborns. Scavenging on the intact (Meusnier et  al. 2008). However, use of shorter carcass of an already deceased individual or placenta DNA segments warrants caution as this may increase is more likely. Determining whether native raptors kill difficulty in discerning species (Symondson 2002), espe- newborn livestock would be of particular interest to cially for those belonging to taxonomically-rich groups. herders, and would present a further challenge for con- Rodents represent 40% of all mammalian species glob- servation as this could lead to conflict. ally (Musser and Carleton 2005), and were found in our The genetic identification of pellet hosts allowed for dataset. While unmapped reads were examined to iden- accurate information of species presence in a particu- tify any species potentially missing from the reference lar area. Of the 45 pellets identified in this study, only 3 file, it remains possible that the MT-RNR1 segment does belonged to Steppe Eagle and 1 to Saker Falcon. China is not amplify all possible rodent species within our study breeding habitat for both species, corresponding to when area. Rodents are a food source for many predators on samples were collected, but both birds are classified as the QTP, extending beyond birds (Hacker et  al. 2021), Endangered by the IUCN and lack of representation in and thus an accurate assessment of their consump- our dataset is consistent with these low population num- tion would be of use in large-scale conservation plan- bers (Birdlife International 2018a, 2019). In contrast, the ning. Sequences which cannot be discerned down to Upland Buzzard and Eurasian Eagle Owl are listed as the species-level may be grouped into a “prey OTU”, as Least Concern (Birdlife International 2017, 2018b). More commonly done in dietary and microbial studies for iden- detailed studies may be able to apply noninvasive surveys tification and analysis purposes (Marchesi et  al. 2002). using regurgitated pellets and DNA metabarcoding to Galan et  al. (2012) designed a mini-barcode marker map geographic distribution or shifts in temporal pres- based on a 136-bp DNA segment of MT-CYB capable of ence during breeding seasons and migration events, as discerning a vast array of Rodentia. While our study did well as habitat preference and overall population status. not face intensive challenges associated with rodent dis- This can similarly be applied to prey in an area, though cernment, our sample size was small and concentrated in birds may be moving long distances and thus the pellets two areas. Study questions tied to rich taxonomic groups they regurgitate may not be indicative of immediate local as a resource in an area with high biodiversity may want wildlife presence per se. to consider the addition of such a marker. Researchers Albeit a small sample size, our findings provide infor - should also consider the likelihood of false identifications mation important for conservation action on the QTP. which may result in the query sequence being assigned First, results demonstrate that pikas are an important to the incorrect taxon and subsequently choose appro- food source, thus large-scale policies to eradicate them priate sequence based species identification methods to should be reconsidered. Second, Eurasian Eagle Owls increase confidence, accuracy, and efficiency of data pro - and Upland Buzzards depend on similar dietary items. cessing (Soergel et  al. 2012; Boyer et  al. 2016; Shi et  al. Persistence of these prey will be necessary for survival 2018). of both species. Eurasian Eagle Owls may be better able The MT-COI marker was used as a diagnostic marker to exploit a wider range of prey as they demonstrate the to differentiate wild and domestic caprid species found ability to increase dietary plasticity, while Upland Buz- in the study area (Hacker et  al. 2021). Though ungu - zards may lack the adaptability to quickly changing and lates were not necessarily expected in the diet based resource poor landscapes. Third, birds of prey may con - on previously available literature, it was surmised tribute to conflict with humans in the form of livestock that past work using morphological analysis may have depredation or death. Reports of such from local herders, missed these larger species as bones and other hard or findings of larger than expected mammals in pellets, parts would likely not be ingested by birds due to should be taken seriously to build mitigation actions for their size. Indeed, our study found two occurrences of reducing livestock loss. Hacker  et al. Avian Res (2021) 12:42 Page 9 of 11 Sabin Family Foundation (G1900011, G2000017), and Welfare Project of the Conclusions National Scientific Research Institution (CAFYBB2019ZE003). Biodiversity tends to be richer and more quickly chang- ing than acknowledged, and assessment methods must Availability of data and materials The datasets generated during the current study are available via email to the be reliable and comprehensive (Pimm et al. 1995). DNA corresponding authors. metabarcoding is a conservation technology tool capa- ble of providing such information. We show the utility Declarations of this technique in determining dietary items from regurgitated bird pellets, and subsequently identified 11 Ethics approval and consent to participate Not applicable. unique prey species for 4 raptor species with pika con- firmed as an important food resource, with high dietary Consent for publication niche overlap between Eurasian Eagle Owl and Upland Not applicable. Buzzard. While we focus on bird of prey species on the Competing interests QTP, our methodology could be applied to pellet analy- The authors declare that they have no competing interests. sis for species worldwide. Our work also demonstrates Author details the power of DNA metabarcoding with noninvasive Research Institute of Forest Ecology, Environment and Protection, Chinese samples, encouraging other sample types to also be Academy of Forestry, and Key Laboratory of Biodiversity Protection of National explored. We also showed that sampling efforts may not Forestry and Grassland Administration, Beijing 100091, China. Depar tment of Biological Sciences, Duquesne University, Pittsburgh, PA 15282, USA. need to be as intensive as previously assumed, given the Yanchiwan National Nature Reserve of Gansu Province, Subei 736300, China. relatively high number of observed taxa captured com- Institute for Ecology and Environmental Resources, Chongqing Academy pared to the expected number of taxa despite limited of Social Sciences, Chongqing 400020, China. The High School Affiliated to Renmin, University of China, Beijing 100872, China. sample sizes. This could decrease effort, time, and costs associated with fieldwork. However, we also contend Received: 26 February 2021 Accepted: 1 August 2021 that the application of this technique has great poten- tial  to answer complex questions surrounding birds of prey that will require rigorous and large-scale study design. References Anderson MJ. Permutational multivariate analysis of variance (PERMANOVA). Wiley StatsRef Stat Refer Online. 2017. https:// doi. org/ 10. 1002/ 97811 Supplementary Information 18445 112. stat0 7841. The online version contains supplementary material available at https:// doi. Avery ML, Cummings JL. Livestock depredations by black vultures and golden org/ 10. 1186/ s40657- 021- 00276-3. eagles. Sheep Goat Res J. 2004;19:58–63. Badingqiuying, Smith AT, Senko J, Siladan MU. Plateau pika Ochotona curzoniae poisoning campaign reduces carnivore abundance in southern Qinghai, Additional file 1. The FASTA file of reference sequences used for read China. Mammal Study. 2016;41:1–8. mapping of possible host and prey species. Birdlife International. Bubo bubo (amended version of 2016 assessment). The Additional file 2. Results of subject sequences blasted to reference IUCN Red List of Threatened Species 2017: e.T22688927A113569670. sequences of determined host and prey items. 2017. https:// doi. org/ 10. 2305/ IUCN. UK. 2017-1. RLTS. T2268 8927A 11356 9670. en. Additional file 3. The r-script used in the statistical analyses of the Birdlife International. Falco cherrug (amended version of 2016 assessment). dataset. The IUCN Red List of Threatened Species 2017: e.T22696495A110525916. Additional file 4. Unique haplotypes for a segment of MT-RNR1 amplified 2018a. https:// doi. org/ 10. 2305/ IUCN. UK. 2017-1. RLTS. T2269 6495A 11052 by primers (12SV5F/12SV5R, Riaz et al. 2011) by species. 5916. en. Birdlife International. Buteo hemilasius. The IUCN Red List of Threatened Spe- cies 2018: e.T22695967A131937792. 2018b. https:// doi. org/ 10. 2305/ IUCN. Acknowledgements UK. 2018-2. RLTS. T2269 5967A 13193 7792. en. We would like to thank Mr. Yun Ma, Jiong He, and Rou Bao for their generous Birdlife International. Aquila nipalensis (amended version of 2017 assessment). assistance in the field surveys. The IUCN Red List of Threatened Species 2019: e.T22696038A155419092. 2019. https:// doi. org/ 10. 2305/ IUCN. UK. 2019-3. RLTS. T2269 6038A 15541 Authors’ contributions 9092. en. CEH formulated study concept. BDH performed statistical analyses. LW led Bontzorlos VA, Peris SJ, Vlachos CG, Bakaloudis DE. The diet of barn owl in the field teams. CEH, LW, WC, YD, YL, JL, YX, YZ, HC, and YZ collected study samples. agricultural landscapes of central Greece. Folia Zool. 2005;54:99–110. CEH, WC, JY, and YJ performed DNA extractions. CEH, WC, and JY performed Boyer F, Mercier C, Bonin A, Le Bras Y, Taberlet P, Coissac E. Obitools: a unix- sequence read mapping and determination of dietary items. CEH, YZ, and JEJ inspired software package for DNA metabarcoding. Mol Ecol Resour. acquired funding for the study. CEH wrote the study manuscript. CEH, BDH, 2016;16:176–82. YZ, and JEJ interpreted study results. BDH, YZ, and JEJ edited the manuscript Chen D, Zhang X, Tan X, Wang K, Qiao Y, Chang Y. Hydroacoustic study of spa- and provided additional expertise. All authors read and approved the final tial and temporal distribution of Gymnocypris przewalskii (Kessler, 1876) in manuscript. Qinghai Lake, China. Environ Biol Fish. 2009;84:231–9. Cui Q, Su J, Jiang Z. Summer diet of two sympatric species of raptors Upland Funding Buzzard (Buteo hemilasius) and Eurasian Eagle Owl (Bubo bubo) in alpine This project was funded by the National Key Technology R&D Program of meadow: problem of coexistence. Pol J Ecol. 2008;56:173–9. China (2019YFA0607103), Snow Leopard Conservancy (G1900014, G2000019), Cleveland Metroparks Zoo (G1800082), Panthera Corporation and The Andrew Hacker et al. Avian Res (2021) 12:42 Page 10 of 11 Deagle BE, Thomas AC, McInnes JC, Clarke LJ, Vesterinen EJ, Clare EL, et al. Lai CH, Smith AT. Keystone status of plateau pikas (Ochotona curzoniae): Counting with DNA in metabarcoding studies: how should we convert effect of control on biodiversity of native birds. Biodivers Conserv. sequence reads to dietary data? Mol Ecol. 2019;28:391–406. 2003;12:1901–12. Dixon A, Maming R, Gunga A, Purev-Ochir G, Batbayar N. The problem of Lamb PD, Hunter E, Pinnegar JK, Creer S, Davies RG, Taylor MI. How quantitative raptor electrocution in Asia: case studies from Mongolia and China. Bird is metabarcoding: a meta-analytical approach. Mol Ecol. 2019;28:420–30. Conserv Int. 2013;23:520–9. Lei F. A study on diet of the Little Owl (Athene noctua plumipes) in Qishan, Edgar RC. MUSCLE: multiple sequence alignment with high accuracy and high Shanxi Province, China. Wuyi Sci J. 1995;12:136–42 (in Chinese). throughput. Nucleic Acids Res. 2004;32:1792–7. Li L, Yi X, Li M, Zhang X. Diet composition of upland buzzard: analysis on stom- Emmrich M, Düttmann H. Seasonal shifts in diet composition of Great Cormo- ach content and food pellet. Zool Res. 2004;25:162–5 (in Chinese). rants Phalacrocorax carbo sinensis foraging at a shallow eutrophic inland Liu X, Chen B. Climatic warming in the Tibetan Plateau during recent decades. lake. Ardea. 2011;99:207–16. Int J Climatol. 2000;20:1729–42. Estes JA, Terborgh J, Brashares JS, Power ME, Berger J, Bond WJ, et al. Trophic Liu Y. International hunting and the involvement of local people, Dulan, downgrading of planet earth. Science. 2011;333:301–6. Qinghai, People’s Republic of China. Master’s Thesis. Missoula: University Ewins PJ, Weseloh DV, Groom JH, Dobos RZ, Mineau P. The diet of Herring of Montana; 1993. Gulls (Larus argentatus) during winter and early spring on the lower Great Marchesi L, Pedrini P, Sergio F. Biases associated with diet study methods in the Lakes. Hydrobiologia. 1994;279(280):39–55. Eurasian Eagle-Owl. J Raptor Res. 2002;36:11–6. Foggin JM. Depopulating the Tibetan grasslands: national policies and per- Margalida A, Bertran J, Boudet J. Assessing the diet of nestling Bearded spectives for the future of Tibetan herders in Qinghai Province, China. Mt Vultures: a comparison between direct observation methods. J Field Res Dev. 2008;28:26–31. Ornithol. 2005;76:40–5. Freeland JR. The importance of molecular markers and primer design Margalida A, Bertran J, Heredia R. Diet and food preferences of the endan- when characterizing biodiversity from environmental DNA. Genome. gered Bearded Vulture Gypaetus barbatus: a basis for their conservation. 2017;60:358–74. Ibis. 2009;151:235–43. Galan M, Pagès M, Cosson JF. Next-generation sequencing for rodent barcod- Meusnier I, Singer GAC, Landry JF, Hickey DA, Hebert PDN, Hajibabaei M. A ing: species identification from fresh, degraded and environmental universal DNA mini-barcode for biodiversity analysis. BMC Genomics. samples. PLoS ONE. 2012;7:e48374. 2008;9:214. Gotelli NJ, Hart EM, Ellison AM. EcoSimR: null model analysis for ecological Miller C, McEwen L. Diet of nesting Savannah Sparrows in interior Alaska. J data. R Package version 0.1.0. 2015. https:// doi. org/ 10. 5281/ zenodo. Field Ornithol. 1995;66:152–8. 16522. Miller DJ, Bedunah DJ. Rangelands of the Kunlun Mountains in Western China. Granjon L, Bruderer C, Cosson JF, Dia AT, Colas F. The small mammal commu- Rangelands. 1994;16:71–6. nity of a coastal site of south-west Mauritania. Afr J Ecol. 2002;40:10–7. Moody T. A method for obtaining food samples from insectivorous birds. Auk. Grier JW. Ban of DDT and subsequent recovery of reproduction in bald eagles. 1970;87:579. Science. 1982;218:1232–5. Musser GG, Carleton MD. Superfamily Muroidea. In: Wilson D, Reeder D, Guimaraes S, Fernandez-Jalvo Y, Stoetzel E, Gorgé O, Bennett EA, Denys C, editors. Mammal species of the world: a taxonomic and geographic refer- et al. Owl pellets: a wise DNA source for small mammal genetics. J Zool. ence. 3rd ed. Baltimore: John Hopkins University; 2005. p. 894–1531. 2016;298:64–74. Oehm J, Thalinger B, Eisenkölbl S, Traugott M. Diet analysis in piscivorous birds: Hacker CE, Jevit M, Hussain S, Muhammad G, Munkhtsog B, Munkhtsog B, et al. what can the addition of molecular tools offer? Ecol Evol. 2017;7:1984–95. Regional comparison of snow leopard (Panthera uncia) diet using DNA Oksanen J, Blanchet FG, Kindt R, Legendre P, Minchin PR, O’Hara RB, et al. Pack- metabarcoding. Biodivers Conserv. 2021;30:797–817. age vegan: community ecology package. R Package. version 2.3-1. 2013. Hebert PDN, Cywinska A, Ball SL, deWaard JR. Biological identifications Pianka ER. Niche overlap and diffuse competition. Proc Natl Acad Sci USA. through DNA barcodes. Proc R Soc B Biol Sci. 2003;270:313–21. 1974;71:2141–5. Heck KL, van Belle G, Simberloff D. Explicit calculation of the rarefaction Pimm SL, Russell GJ, Gittleman JL, Brooks TM. The future of biodiversity. Sci- diversity measurement and the determination of sufficient sample size. ence. 1995;269:347–50. Ecology. 1975;56:1459–61. Pompanon F, Deagle BE, Symondson WOC, Brown DS, Jarman SN, Taberlet P. Hill MO. Diversity and evenness: a unifying notation and its consequences. Who is eating what: diet assessment using next generation sequencing. Ecology. 1973;54:427–32. Mol Ecol. 2012;21:1931–50. Hiraldo F, Andrada J, Parreño F. Diet of the eagle owl (Bubo bubo) in Mediter- Rasmussen P, Anderson J. Birds of South Asia: the ripley guide. Washington, ranean Spain. Doñana Acta Vertebr. 1975;2:161–77. DC, and Barcelona: Smithsonian Institution and Lynx Edicions; 2005. Hsieh TC, Ma KH, Chao A. iNEXT: an R package for rarefaction and extrapolation Reading R, Michel S, Amgalanbaatar S. Ovis ammon. The IUCN Red List of of species diversity (Hill numbers). Method Ecol Evol. 2016;7:1451–6. Threatened Species 2020: e.T15733A22146397. 2020. https:// doi. org/ 10. Hurlbert SH. The nonconcept of species diversity: a critique and alternative 2305/ IUCN. UK. 2020-2. RLTS. T1573 3A221 46397. en. parameters. Ecology. 1971;52:577–86. Riaz T, Shehzad W, Viari A, Pompanon F, Taberlet P, Coissac E. ecoPrimers: Iverson SJ, Springer AM, Kitaysky AS. Seabirds as indicators of food web struc- inference of new DNA barcode markers from whole genome sequence ture and ecosystem variability: qualitative and quantitative diet analyses analysis. Nucleic Acids Res. 2011;39:e145. using fatty acids. Mar Ecol Prog Ser. 2007;352:235–44. Sándor AD, Ionescu DT. Diet of the eagle owl (Bubo bubo) in Braşov, Romania. Jackson R. Fostering community-based stewardship of wildlife in Central Asia: North West J Zool. 2009;5:170–8. transforming snow leopards from pests into valued assets. In: Squires VR, Schaller GB. Wildlife of the Tibetan Steppe. Chicago: University of Chicago editor. Rangeland stewardship in Central Asia. Dordrecht: Springer; 2012. Press; 1998. p. 357–80. Schaller GB. Tibet Wild. Washington: Island Press; 2012. Janečka JE, Jackson R, Yuquang Z, Diqiang L, Munkhtsog B, Buckley-Beason V, Schaller GB, Junrang R, Mingjiang Q. Status of the snow Leopard Pan- et al. Population monitoring of snow leopards using noninvasive collec- thera uncia in Qinghai and Gansu Provinces, China. Biol Conserv. tion of scat samples: a pilot study. Anim Conserv. 2008;11:401–11. 1988;45:179–94. Janečka JE, Munkhtsog B, Jackson RM, Naranbaatar G, Mallon DP, Murphy WJ. Schoener TW. Competition and the form of habitat shift. Theor Popul Biol. Comparison of noninvasive genetic and camera-trapping techniques for 1974;6:265–307. surveying snow leopards. J Mammal. 2011;92:771–83. Shannon C, Weaver W. The mathematical theory of communication. Urbana: Jedlicka JA, Vo ATE, Almeida RPP. Molecular scatology and high-throughput University of Illinois Press; 1949. sequencing reveal predominately herbivorous insects in the diets of Shehzad W, McCarthy TM, Pompanon F, Purevjav L, Coissac E, Riaz T, et al. Prey adult and nestling Western Bluebirds (Sialia mexicana) in California vine- preference of snow leopard (Panthera uncia) in South Gobi, Mongolia. yards. Auk. 2017;134:116–27. PLoS ONE. 2012;7:e32104. Kartzinel TR, Chen PA, Coverdale TC, Erickson DL, Kress WJ, Kuzmina ML, et al. Shi ZY, Yang CQ, Hao MD, Wang XY, Ward RD, Zhang AB. FuzzyID2: a software DNA metabarcoding illuminates dietary niche partitioning by African package for large data set species identification via barcoding and large herbivores. Proc Natl Acad Sci USA. 2015;112:8019–24. Hacker  et al. Avian Res (2021) 12:42 Page 11 of 11 metabarcoding using hidden Markov models and fuzzy set methods. Mol associated with dietary shifts in an obligate riparian Neotropical migra- Ecol Resour. 2018;18:666–75. tory songbird. PeerJ. 2018a;16:e5141. Simon C, Frati F, Beckenbach A, Crespi B, Liu H, Flook P. Evolution, weighting, Trevelline BK, Nuttle T, Hoenig BD, Brouwer NL, Porter BA, Latta SC. DNA and phylogenetic utility of mitochondrial gene sequences and a compi- metabarcoding of nestling feces reveals provisioning of aquatic prey and lation of conserved polymerase chain reaction primers. Ann Entomol Soc resource partitioning among Neotropical migratory songbirds in a ripar- Am. 1994;87:651–701. ian habitat. Oecologia. 2018b;187:85–98. Simpson EH. Measurement of diversity. Nature. 1949;163:688. Treves A, Krofel M, McManus J. Predator control should not be a shot in the Smith AT, Formozov NA, Hoffman RS, Zeng C, Erbajeva MA. The pikas. In: dark. Front Ecol Environ. 2016;14:380–8. Chapman J, Flux J, editors. Rabbits, hares and pikas. Status survery and Valera F, Gutiérrez JE, Barrios R. Eec ff tiveness, biases and mortality in the use of conservation action plan. Chapter 3. Gland, Switzerland: IUCN; 1990. pp. apomorphine for determining the diet of granivorous passerines. Condor. 14–60. 1997;99:765–72. Smith AT, Foggin JM. The plateau pika (Ochotona curzoniae) is a keystone spe- Walter CB, O’Neill E. Electrophoresis in the study of diets and digestive rates of cies for biodiversity on the Tibetan plateau. Anim Conserv. 1999;2:235–40. seabirds. Comp Biochem Physiol A Comp Physiol. 1986;84:763–5. Soergel DAW, Dey N, Knight R, Brenner SE. Selection of primers for optimal Wei WR, He JD, Zheng QY. Plateau pikas (Ochotona curzoniae) at low densities taxonomic classification of environmental 16S rRNA gene sequences. have no destructive effect on winter pasture in alpine meadows. Rangel ISME J. 2012;6:1440–4. J. 2020;42:55–61. Symondson WOC. Molecular identification of prey in predator diets. Mol Ecol. Wilson RP. An improved stomach pump for penguins and other seabirds. J 2002;11:627–41. Field Ornithol. 1984;55:109–12. Taberlet P, Fumagalli L. Owl pellets as a source of DNA for genetic studies of Xia W, Zhou X, Liu J, Zhang X. The bio-community in the region of alpine small mammals. Mol Ecol. 1996;5:301–5. meadow. In: Liu J, Wang Z, editors. Alpine meadow ecosystem. Beijing, Thiam M, Bâ K, Duplantier JM. Impacts of climatic changes on small mammal China: Science Press; 1991. p. 1–7. communities in the Sahel ( West Africa) as evidenced by owl pellet analy- Yang Z, Gong M, Huang X, Li C, Liu X. Biodiversity of birds of prey in Shiqu sis. African Zool. 2008;43:135–43. county. J Sichuan Teach Coll. 2000;21:137–40 (in Chinese). Tremblay I, Thomas D, Blondel J, Perret P, Lambrechts MM. The effect of habitat Zhang R, Ludwig A, Zhang C, Tong C, Li G, Tang Y, et al. Local adaptation of quality on foraging patterns, provisioning rate and nestling growth in Gymnocypris przewalskii (Cyprinidae) on the Tibetan Plateau. Sci Rep. Corsican Blue Tits Parus caeruleus. Ibis. 2005;147:17–24. 2015;5:9780. Trevelline BK, Nuttle T, Porter BA, Brouwer N, Hoenig BD, Steffensmeier ZD, Latta SC. Stream acidification and reduced migratory prey availablity are Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions

Journal

Avian Research – Springer Journals

Published: Aug 21, 2021

Keywords: Avian; Eurasian Eagle Owl; Molecular diet analysis; Next-generation sequencing; Raptor; Saker Falcon; Steppe Eagle; Upland Buzzard

There are no references for this article.