Community Ecology (2023) 24:147–157 https://doi.org/10.1007/s42974-023-00148-4 ORIGINAL ARTICLE DNA metabarcoding Passerine bird feces at tree‑line uncovers little intra‑ and inter‑species dietary overlap Jamie Bookwalter1,2 · Afaq M. Mohamed Niyas3 · Berta Caballero‑López4 · Caterina Villari3 · Jana Marco‑Tresserras2 · Albert Burgas2 · Mariona Ferrandiz‑Rovira1,2 · Bernat Claramunt‑López1,2 Received: 30 September 2022 / Accepted: 11 May 2023 / Published online: 4 July 2023 © The Author(s) 2023 Abstract High-elevation insectivorous birds are currently confronted with the reality of a changing climate, land use shifts, and the decline of many prey groups. The diet dynamics among many imperiled animals in this group are still unresolved. Exam- ining the diets of tree-line Passerine birds to the species level of the prey allows for stronger population predictions. This study uses DNA metabarcoding to identify prey arthropods from adult Passerine bird feces at and slightly below tree-line in a Pyrenean forest. Our objective was to quantify the intra-and inter-species richness and overlap of Passerine bird diet over time and space. The results showed that adult Passerine diets have high inter- and intra-species dietary variability and low inter- and intra-species dietary overlap. The lack of association between dietary richness and open space, season, and eleva- tion and lack of differences between dietary overlap and open space and elevation suggest high-elevation Passerine birds have very high dietary flexibility. The results also showed that aphids known to be pests to conifers, and other conifer pests, were prevalent in the birds’ diets. The Passerine diets and high rate of rare dietary items are mainly in line with other recent DNA metabarcoding studies. Implications for the long-term projections relative to tree-line Passerine populations are discussed. Keywords Insecta · Arachnida · Aphid Introduction non-insectivorous bird populations have faced a substantial decline over the past 30 years in Europe (Inger et al., 2015). Insect-eating birds across the globe consume 400–500 mil- Upper elevation birds are particularly at risk because lion tons of insects annually (Nyffeler et al., 2018), sur- mountains are expected to be more affected by climate passing the 350 million tons of meat that humans consume change than lowland areas due to faster and enhanced warm- each year (Hicks et al., 2018). However, insectivorous bird ing (Mountain Research Initiative EDW Working Group, populations are particularly at risk due to climate change. In 2015). For example, high-elevation populations of Canada North America, some terrestrial insectivorous bird popula- jays (Perisoreus canadensis (L., 1766)) declined 50% over a tions have decreased by 33% in the past five decades (Tal- 30-year period, and the decline was attributed to warmer and lamy & Shriver, 2021). Similarly, many insectivorous and more variable weather (Sutton et al., 2021). This change in weather pattern increased the number of freeze–thaw events, which caused an increase in the spoilage of cached food * Jamie Bookwalter jdink123@gmail.com items. The survival of another high-elevation Passerine, the white winged snowfinch (Montifringilla nivalis (L., 1766)) 1 Unitat d’Ecologia, BABVE, Edifici Ciències, Universitat is in doubt because its foraging behavior is closely tied to Autònoma de Barcelona, 08193 Bellaterra, Catalonia, Spain snow retreat conditions which are becoming increasingly 2 CREAF, Centre de Recerca Ecològica i Aplicacions less consistent (Resano-Mayor et al., 2019). Finally, Barras Forestals, E08193 Bellaterra (Cerdanyola de Vallès), et al. (2021) found that elevated ambient temperatures at the Catalonia, Spain tree-line in the Swiss alps negatively affected nestling prey 3 Warnell School of Forestry and Natural Resources, provisioning rates of the Alpine ring ouzel (Turdus torquatus University of Georgia, Athens, Georgia 30602, USA alpestris L., 1758). 4 Museu de Ciències Naturals de Barcelona, Passeig Picasso, Castell Tres Dragons 08003, Barcelona, Spain 13 Vol.:(0123456789) 148 Community Ecology (2023) 24:147–157 The flora and fauna of the Pyrenees mountains are espe- examined the adult diet of one species (McClenaghan et al., cially threatened due to climate change and land use shifts 2019; Moran et al., 2019; Shutt et al., 2020, and Snider et al., (OPCC-CTP, 2018). The snowpack is warmer than many 2021 among others) or that of nestlings (see Rytkönen et al., other mountain ranges across the world and thus particularly 2019, da Silva et al., 2020, and Jarrett et al., 2020). The sensitive to slight changes in ambient temperature (Lopez- goal of our study was to examine the diets of co-occurring Moreno et al., 2017), and the decline of agropastoral prac- high-elevation Pyrenean Passerines and to assess the effects tices in the Pyrenees has led to transitions of open grassland of open space, elevation, and season on them. We expected into forest (Roura et al., 2005). The Pyrenean tree-line is also higher niche differentiation (i.e., difference in diet compo- shifting upward (Ameztegui et al., 2016). Tree-line dynam- sition) in morphologically and behaviorally similar spe- ics globally are affected by a variety of factors, including cies and higher dietary richness and higher dietary overlap precipitation, tree community composition, and soil struc- among species as spring progressed to autumn. Passerines ture (Grace et al., 2002; Körner, 2012). The upward shift that have similar traits often have competition-driven niche and densification of the tree-line in the Pyrenees is generally separation (Alatalo et al., 1986; Cowie & Hinsley, 1988; linked to local agricultural abandonment (Batllori & Gutiér- Sottas et al., 2020), and higher abundance of prey is linked rez, 2008), but there can be locally important factors such to less dietary partitioning (Davies et al., 2022). In some spe- as slope morphometry and lithology (Feuillet et al., 2020). cies, we expected diet richness to positively correlate with It is within this context that we examined the diet of Pas- the percentage of open space because patchier habitats have serines at elevations located below and at tree-line. The diet been shown to benefit some species but not others (Suarez- of many European Passerine birds, e.g., Paridae, has been Seoane et al., 2002). Finally, we expected higher overlap in examined closely, even though most studies were limited to below tree-line plots because we expected the conditions to estimating the diet of nestlings using either methods that are be more favorable to Passerines. Higher overlap is common invasive (neck collars (Barba & Gil-Delgado, 1990; Pagani- in more favorable habitats (Hou et al., 2021). Núñez et al., 2011) and stomach flushing (Senécal et al., 2021)), noninvasive but less detailed (e.g., cameras and sta- ble isotope analysis (Currie et al., 1996; Şekercioğlu et al., Methods 2023, respectively)), or lethal (gizzard extraction (Sehhatisa- bet et al., 2008)). Recent advances in DNA metabarcoding Study area and feces collection technology (hereafter metabarcoding) have increased our ability to analyze the diet of adult Passerines in a noninva- Ten plots were selected in grid format within a black pine sive manner at a high level of taxonomic classification (see forest (Pinus mugo species complex, hereafter P. mugo) Crisol-Martínez et al., 2016; Ribeiro et al., 2019; da Silva in Vall d’Ordino, a valley located within three km of Vall et al., 2020; Shutt et al., 2020). de Sorteny Natural Park in the parish of Ordino, Andorra. Metabarcoding supports high-throughput (i.e., massively Plots were situated between 1729 and 2352 m above sea parallel) taxonomic classification within a sample (Bush level (hereafter MASL). Percent open space surrounding et al., 2019). A short portion of a gene (barcode) from an each plot (1000 m radius) was calculated using QGIS3.4 environmental or biological sample is amplified by a primer and the MCSA 2012 landcover map downloaded from the designed to provide taxonomic resolution of a target organ- Institute of Andorran Studies (Centre de Biodiversitat de ism or taxonomic group (Deagle et al., 2014; Hajibabaei l'Institut d'Estudis Andorrans, 2012). Testing various scales et al., 2007). However, there are many fundamental limita- of open space was out of the scope of this study; therefore, tions of fecal metabarcoding. For example, raw prey abun- a landscape level scale of 1000 m radius was chosen as a dances cannot be determined from the number of reads in 1000 m radius is a generally accepted radius for a landscape a similar DNA sequence, and relative prey abundance is level measure of “forest amount” (Shoffner et al., 2018). difficult to recover because of technological and biological Plots were characterized as “below tree-line” or “at tree- biases including primer mismatch and differences in PCR line” depending on positioning above or below the median amplification due to primer sequence length (Deagle et al., elevation of all plots (i.e., 2077 MASL). In most of Andorra, 2013; Krehenwinkel et al., 2017; Piñol et al., 2015). Sample tree-line occurs between 2200 and 2400 MASL and in a few contamination and differing rates of DNA preservation in areas between 2100 and 2500 MASL (Carreras et al., 1996). the gut can also present issues (Galan et al., 2018; Nielsen Birds were captured using Ecotone mist-nets (9 m and et al., 2018). 6 m long and 2.5 m high, with 5 shelves and a mesh size of A handful of studies have examined the diet of multiple 16 ­mm2) stretched between 4 m poles inserted perpendicu- adult Passerines species using metabarcoding (see Crisol- larly in the ground. Three nets were used in eight of the sites, Martínez et al., 2016, Sottas et al., 2020, and Garfinkel et al., and two nets were used in the remaining two sites due to dif- 2022). Most Passerine dietary metabarcoding studies have ficultly of access, and openness of site (see Supplementary 13 Community Ecology (2023) 24:147–157 149 material 1). Mist nets were deployed between May 15, 2018, amplicon was purified with bead cleanup. Final libraries and September 30, 2018, and number of net days for each underwent quality checks by SYBR fluorometry, Fragment site are also documented in Supplementary material 1. The Analyzer and qPCR (KAPA Library Quantification—Roche start date for the field component of this study coincided Sequencing & Life Sciences). Libraries were normalized and with the date when snow historically has retreated from the pooled equally. The pooled amplicons were sequenced on Andorran tree-line. Mist nets were not set at a plot when the MiSeq platform (PE250) (Illumina). there was precipitation or high winds to ensure good capture conditions and welfare of the bird. Once a bird was captured, Bioinformatic analysis it was placed in a single use individual paper bag. After defecation, feces were carefully removed from the paper bag Within the QIIME 2 2020.6 environment, tagged feces using a single-use toothpick and stored in plastic vials. Vials sequence reads generated from the Illumina MiSeq were placed on ice in the field and transferred to long-term sequencer were demultiplexed and primers trimmed to create deep-freeze as soon as possible. All birds caught were iden- fastq files (Bolyen et al., 2019). We then used the DADA2 tified at species level, ringed, aged, sexed, and measured pipeline for further downstream analysis, which created a following standard ringing procedure. Birds were handled table of amplicon sequence variants (ASVs) rather than by certified ringers and all the procedures approved by the traditional operational taxonomic units (OTUs), thereby Environment and Sustainability Department of Andorran improving reproducibility, comprehensiveness, and accuracy Government. (Callahan et al., 2016). Potential contaminants in the ASV A total of 132 fecal samples were collected. No bird was table using the “frequency” method were identified by the captured twice. Samples collected in May and June were package Decontam using the default threshold of 0.1 (Davis considered to be from the spring, July and August samples et al., 2018). The “frequency” method identifies contami- were considered to be from the summer, and September nants whose concentrations in blank samples are inversely samples were considered to be from the autumn. correlated with sample DNA concentrations, based on a user-specified threshold. As the metabarcoding workflow DNA extraction and amplification introduces quantitative bias into results, ASV raw counts were transformed into a presence/absence matrix (Martoni DNA in the collected fecal samples and in four DNA extrac- et al., 2022). tion blanks (i.e., vials with no fecal samples that served as ASVs were taxonomically classified by aligning contaminant control) was extracted using the QIAamp DNA sequences to those within the arthropod training database Stool Mini Kit (QIAGEN) following the manufacture’s pro- “tidybug” (O’Rourke et al., 2020) via the classy-sklearn tocol with modifications suggested by Davies et al. (2022). naïve Bayes method implemented in QIIME 2’s q2-feature- Only samples of similar size (approximately 3 mg) were classifier plugin. The full QIIME script can be found in processed. Three samples were smaller than 3 mg and thus Supplementary material 2. Each taxonomic assignment was discarded from analysis. DNA concentration was quantified further examined individually using the following protocol: using a Qubit 3.0 Fluorometer (Thermo Fisher Scientific (1) The geographic range of the assignment was assessed, Inc.). Amplicon library preparation and sequencing were and species that do not occur in Europe were removed. (2) carried out by the Georgia Genomics and Bioinformatics Species considered to be rare in Europe but are not known Core (University of Georgia, Athens GA, USA) as follows: to occur in the Pyrenees were flagged. (3) Flagged assign- the concentration and size distribution of the DNA were ments were verified by submitting query sequences to the determined using SYBR fluorometry and a fragment ana- NCBI BLAST tool and assignments that did not score at or lyzer, respectively. The DNA samples were normalized above 98% identity were removed. All single flagged species to the same concentration before library preparation. The composed of multiple ASV sequences were aligned to check amplicon library was prepared beginning with a 2X KAPA for sequencing error, and sequences above an 80% sequenc- HiFi HotStart ReadyMix (Roche Sequencing & Life Sci- ing error were kept in the analysis (Brandt et al., 2021; Ritter ences). The first PCR was assembled with target-specific et al., 2022). (4) If an ASV showed multiple hits with the primers. Initial plans called for the use of a primer amplify- same max score on the NCBI BLAST tool, the ASV was ing a longer region, but a preliminary experiment (data not removed from the analysis. shown) indicated higher efficacy of a shorter primer, the mini-barcode mitochondrial primer (ANML). ANML ampli- Statistical analyses fies a smaller 180 bp segment on the cytochrome oxidase C subunit 1 (COI) (Jusino et al., 2017). In the second round A Wilcoxon–Mann–Whitney test was performed to test for of PCR, unique Illumina Indexed primers (i5 and i7) were correlations between open space and elevation as a discrete added to each sample and at the end of each round of PCR, variable, using the percent of open space as a dependent 13 150 Community Ecology (2023) 24:147–157 variable. We calculated mean prey richness of all bird species were taxonomically classified as the MOTUs of 713 arthro- and of the top two commonly collected bird species (Peri- pod classifications (representing six classes, 30 orders, parus ater (L., 1758) and Lophophanes cristatus (L., 1758)). 163 families, 466 genera) and then only MOTUs classified Predictive roles of independent factors affecting of the com- to genus or species were kept for a total of 590 arthropod bined prey richness of these two bird species (season, eleva- MOTUs. We refer to these MOTUs as species. tion as a continuous gradient, and open space) were found by fitting data to a negative binomial or Poisson model (GLMM) Inter‑ and intra‑species dietary richness using the lme4 v.26 package (Bates et al., 2015) in R Version of Passerines 1.3.1056 (R Core Team, 2021). Models were chosen through a combination of (1) residual plotting with the DHARMa pack- Fecal samples were collected and analyzed from 14 bird age (Hartig, 2022); (2) model performance testing using Pear- species (Supplementary material 4). The bird species P. ater son, Kendall, and Spearman correlation coefficients; and (3) (coal tit), L. cristatus (crested tit), and Prunella modula- model accuracy evaluation by measuring the root mean square ris (L., 1758) (dunnock) accounted for 40%, 15%, and 14% error and the mean-absolute deviation of each model. Plot was (respectively) of the bird species from which samples were used as a random variable. We also tested “individual bird” collected. No significant seasonal differences were found as the only random factor and it yielded very similar results, when all bird species were combined or when just P. ater quantitatively and qualitatively. Nevertheless, the “individual and L. cristatus were combined (P values in Supplemen- bird” choice also caused some models to fail convergence and tary material 5). No captures of P. modularis were made in these problems were more severe when both “plot” and “indi- autumn. Seventy-three and 59 birds were caught in the high- vidual” were chosen together as a random factor. Therefore, elevation and low-elevation plots, respectively. our final choice for random factor was “plot.” Some models, The mean prey richness per bird capture was 13.5 ± 6.9 however, did not accept any random factor and were thus run species. Phoenicurus ochruros (Gmelin, 1774) (black red- without one after numerical testing showed little differences start) displayed the highest species richness per bird capture, between models with random factors and those without. After followed by Sylvia atricapilla (L., 1758) (blackcap) and P. the prey richness model of each bird group was fit, post hoc modularis, although standard deviation bars do not show Tukey tests were carried out to investigate error rates of the differences within these three species (Fig. 1). categorical factor of season. Comparison of beta diversity (i.e., According to Wilcoxon–Mann–Whitney tests, the percent dissimilarity of diet) among and within bird species was deter- of open space was significantly higher in the plots at tree- mined by a Jaccard dissimilarity matrix using the R vegan line (mean: 88 ± 13.7) than the plots below tree-line (mean: package, with a value of 1 indicating there were no shared 45.2 ± 17.5) (Z = 4.25, P < 0.001). However, GLMM results species and a value of 0 indicating complete sharing of species showed that the percentage of open space was not correlated (Oksanen et al., 2020). A Jaccard dissimilarity matrix was also with prey richness in the diets of P. ater and L. cristatus used to examine the dissimilarity of diet of all individual birds when combined, or when all bird species were combined combined and the individuals of the two most common bird (Beta estimates and P values in Supplementary material 5). species combined in regard to season and the discrete vari- There were no elevational differences in prey richness when able of elevation. Frequency of occurrence (number of times a all bird species were combined, or when P. ater and L. cris- prey item appeared in a fecal sample, divided by total number tatus were combined (Supplementary material 5). of samples) was calculated for the six most frequently occur- ring prey groups (Arachnida, Diptera, Coleoptera, Hemiptera, Hymenoptera, and Lepidoptera). Predictive roles affecting spe- cies richness within each prey group (season, elevation as a continuous gradient, and open space) were calculated using the same model selection process as bird groups. Sample cover- age was examined using the iNEXT package to create sample size-based rarefaction curves and extrapolation curves (Chao & Jost, 2012; Chao et al., 2014). Results A total of 8.95 million sequence reads were produced in Fig. 1  Prey species richness per bird capture with standard deviation the feces samples, with ASV counts per feces sample rang- bars. Values in parentheses represent the captures (i.e., number of ing from four to 268,999 (Supplementary material 3). ASVs feces collected) of each bird species 13 Community Ecology (2023) 24:147–157 151 Inter‑ and intra‑species dietary overlap Presence of prey species and prey species trends of Passerines Most prey species were rare; 58.64% of the prey species Jaccard dissimilarity index showed very little overlap in were collected only once (i.e., collected in one sample) the diet between and within bird species, and there was (Table 1). However, eleven species were present in over a mean dissimilarity of 0.91 ± 0.05 in the diet of the 14 15% of samples (Supplementary material 7). Of these bird species. Beta diversity was very high among species. eleven species, five were conifer pests. Aphid conifer pests Compared to each other, P. modularis /L. cristatus and P. (Hemiptera) were the two species most likely to be present modularis /P. ater had a higher rate of dissimilarity (0.86 (Supplementary material 7). Diptera and Lepidoptera repre- and 0.83, respectively) than the rate between P. ater/ L. sented 21.78% and 20.11%, respectively, of all prey species cristatus (0.73). The dietary variability within P. ater, (Fig. 3). GLMM results indicated that the richness of some L. cristatus, and P. modularis was high as well: P. ater groups of prey (Coleoptera, Hemiptera, and Hymenoptera) (0.82 ± 0.02), L. cristatus (0.93 ± 0.05), and P. modularis significantly varied between some seasons, and elevation (0.95 ± 0.04). Accordingly, sample size-based rarefaction and open space did not drive richness of prey groups (Beta curves indicated it would be necessary to capture over 100 estimates, P values, and seasonality Tukey tests in Supple- more P. ater individuals than P. modularis and L. crista- mentary material 8). tus to reach 99% sample coverage (Fig. 2a). At 99% cover- age, P. modularis is expected to have a higher diversity of diet than P. ater and P. cristatus, while P. ater is expected Discussion to have the lowest (Fig. 2b). The mean overlap among all individual birds captured was not different between the The difficulty in accurately and directly identifying species- three seasons (spring, 0.9 ± 0.05; summer, 0.93 ± 0.07; level dietary components of adult insectivorous Passerines in autumn, 0.93 ± 0.06) (Supplementary material 6a). Simi- a non-lethal manner is a quandary that has long confounded larly, when the mean overlap between the two most cap- ornithologists. A metabarcoding approach allows for the tured birds (P. ater and L. cristatus) was calculated by direct study of species composition of Passerine feces. With season, no difference in overlap was recorded (spring, this technique, we were able to determine that: (1) there 0.92 ± 0.07; summer, 0.91 ± 0.08; autumn, 0.90 ± 0.07) was extremely high inter- and intra-species variability and (Supplementary material 6a). The mean overlap between low inter- and intra-species overlap in the diet of captured all birds and between P. ater and P. cristatus in the plots Passerines, a result that contrasts with some traditional Pas- at tree-line and below tree-line was similarly low (see serine nestling studies and agrees with many metabarcoding Supplementary material 6b). studies; (2) dietary richness did not correlate with season nor with open space, and there was no difference in dietary overlap relative to open space or elevation, which suggests that high-elevation Passerine birds in our study have high Fig. 2  Sample size-based rarefaction curves (solid line) and extrapo- b Prey diversity of bird species per number of individual birds sam- lation curves (dotted line) with 95% confidence intervals. a Number pled. Numbers in parentheses indicate number of captures of birds of birds caught per sample coverage. Numbers in parentheses indicate per species necessary to reach 99% coverage, and prey diversity at bird species and number of individual birds caught at 99% sample 99% coverage with 95% confidence intervals. Prunella modularis is coverage. To reach 99% coverage, it would be necessary to capture expected to have a higher diversity of diet at 99% sample coverage, over 100 more P. ater individuals than P. modularis and L. cristatus. while P. ater is expected to have lowest diversity 13 152 Community Ecology (2023) 24:147–157 Table 1  Prey species distribution among bird feces collections Insecta: Insecta: Insecta: Coleoptera Orthoptera Diptera Number of feces recorded (Number of prey spe- 11.46% 2.23% 21.78% cies) % of prey species recorded 1 (346) 58.64 Arachnida: Araneae 2 (103) 17.46 13.07% 3 (43) 7.29 4 (22) 3.73 5 (7) 1.19 6 (8) 1.36 Insecta: 7 (10) 1.69 Lepidoptera Insecta: 8 (5) 0.85 20.11% Hemiptera 9 (7) 1.19 15.19% 10 (5) 0.85 Insecta: Hymenoptera 11 (6) 1.02 16.16% 12 (3) 0.51 13 (4) 0.68 Fig. 3  Pie graph of relative species richness by order, i.e., percentage 14 (2) 0.34 of total number of prey species belonging to each order. Only orders 16 (3) 0.51 with relative species richness above 0.00% are included in pie graph 18 (1) 0.17 19 (2) 0.34 1788)) (Moran et al., 2019), and barn swallows (Hirundo 21 (1) 0.17 rustica (L. 1758)) (McClenaghan et al., 2019) postulated 22 (1) 0.17 that the high intraspecific dietary variation found in the 23 (2) 0.34 birds’ diets was likely due to prey availability and dietary 25 (1) 0.17 flexibility. 26 (2) 0.34 Despite a mean richness per bird capture of only 30 (1) 0.17 13.5 ± 6.9 species, the dietary overlap among and within bird 33 (1) 0.17 species was very low. We expected lower overlap (higher 37 (1) 0.17 niche differentiation) between closely related bird species, 39 (1) 0.17 50 (1) 0.17 as it is well established that segregated foraging behavior occurs between closely related European insectivorous 70 (1) 0.17 Passerines. When a potential niche is left unoccupied by Number in parentheses is tally of an individual prey species recorded a Passerine bird species, the species that most resembles among the bird feces collection. For example, of the 590 prey species the absent species in body size is the species most likely classified in the study, eight (or 1.36% of all prey species recorded) were recorded in six feces. Most prey species were rare; 58.64% of to fill it (for a review of geographic niche changes in insec- the prey species (or 346) were recorded in only one feces collection tivorous Passerines, see Alatalo et al. (1986)). Segregated foraging behavior makes sense in light of a study showing negative impacts upon a less dominant but closely related dietary mobility; and (3) Passerine diets were dominated by species sharing geographic space; Parus major (L. 1758) conifer pests. (great tit) nestlings raised sympatrically with C. caeruleus There was high biological richness in our analysis of the weighed less than those raised allopatrically, suggesting that 132 fecal samples: over 594 taxonomically classified arthro- a large overlap of resource utilization exists between the two pod species were identified. Most prey species were rare; closely related species (Torok & Tóth, 1999). Most dietary 58.64% of the prey species (or 346) were recorded in only studies comparing insectivorous hole-nesting Passerines one feces collection. A recent metabarcoding study examin- (mainly some combination of P. ater, P. major, L. cristatus, ing diets of 25 North American grassland Passerines also and C. caeruleus) have historically examined nestling diets found very diverse diets and rare prey items among bird and reported a high overlap when dietary components are species and linked this inter-species dietary diversity to classified to a combination of class and family (Grzędzicka, opportunistic availability (Garfinkel et al., 2022). Similarly, 2018; Michalski et al., 2011; Nour et al., 1998). One study authors of other metabarcoding studies examining (respec- that classified Passerine prey of Lepidoptera and Arachnida tively) Eurasian blue tits (Cyanistes caeruleus (Shutt et al., to species also found high overlap (Atiénzar et al., 2013). 2020)), Rufous hummingbirds (Selasphorus rufus (Gmelin However, at least two studies have shown that P. major 13 Community Ecology (2023) 24:147–157 153 and C. caeruleus feed differing sizes of caterpillar prey to Passerines during times of limited insect availability, in con- nestlings (Ceia et al., 2016; Torok & Tóth, 1999), a result trast to many studies that show high dietary overlap during that would not be observable in a DNA-based study such times of high resource availability. The researchers postu- as ours. In Ceia et al. (2016) the authors showed that the lated this result indicated that during times of very low food composition of prey (classified to family) between the two availability, bird species were unable to avoid competition bird species were similar, and they postulated that difference (Kent et al., 2022). Our study found no seasonal differences in prey size resulted from either interspecific competition in dietary overlap between the two most commonly captured between P. major and C. caeruleus or the segregation of bird birds (P. ater and L. cristatus). foraging guilds; Cyanistes caeruleus is primarily a foliage- Our results also did not show seasonal dietary richness gleaner and more likely to come in contact with smaller differences of these two bird species combined or when die- instar caterpillars, while P. major are bark–foliage glean- tary richness of all individual birds captured were combined ers and therefore more likely to come in contact with later (Supplementary material 5). Our seasonal richness results instar caterpillars. Numerous dietary metabarcoding studies, are different from the findings of an adult metabarcoding however, have found lower inter- and intra-species overlap, study done in a deciduous forest (Shutt et al., 2020). In that including a dietary metabarcoding study of insectivorous study, Shutt et al. (2020) tested 793 fecal samples collected Malaysian Babbler species (genera within the families Pel- from March through early May of C. caeruleus and linked lorneidae and Timaliidae) which found lower than 35% over- rising seasonal dietary richness to rising herbivorous insect lap in three of the ten species (Mansor et al., 2022). Lower abundance and availability. It is possible that the lack of dif- inter- and intra-species overlap was also found in two other ferences in seasonal dietary richness found in our study is metabarcoding studies that examined dietary overlap within a result of our sampling time frame (mid-May through late European insectivorous Passerines species (see Rytkönen September) and/or the relatively lower number of samples et al., 2019; Shutt et al., 2020). taken in our study (132). Regardless, our results (lack of While inter- and intra-species overlap was very low in our seasonal prey overlap among bird species, no difference in study, the Jaccard dissimilarity indexes displayed slightly dietary richness from spring to autumn, and no difference higher index values between P. modularis/L. cristatus and in richness levels among all bird species) likely indicate that P. modularis/P. ater than between the more closely related P. the birds captured in our study have a high level of dietary ater and L. cristatus. These results are likely due to diverg- flexibility. ing foraging habits, a finding supported by a metabarcoding We expected dietary richness levels of ground and shrub study that found differing dietary compositions of ground- foraging species, such as P. modularis, to be correlated with vs arboreal-foraging bird species in macadamia orchards open space, as structure and composition of vegetation can (Crisol-Martínez et al., 2016). Periparus ater and L. cris- be very influential in nestling success of some European tatus both forage in trees (Alatalo, 1981; Hartley, 1987; Passerine species (Orłowski et al., 2017). We also expected Lens, 1996), while P. modularis are mainly ground feeders higher dietary overlap in plots below tree-line; while (Bishton, 1986). response of insects to elevation is species specific (Hod- Historical data report divergent timing of clutch-laying of kinson, 2005), many montane fauna either decrease with closely related insectivorous hole-nesting Passerines (Sanz elevation or have a humped shaped distribution along an et al., 2010), and a more recent study reported that many elevational gradient (McCain, 2009; Rahbek, 2005). Tem- Passerine nestlings are provisioned with differing prey types perature is known to be a major driver in insect community depending on the nestling’s development stage (Orłowski structure (Bale et al., 2002), and temperature swings are et al., 2017). Historical data also suggest that resident insec- wider at higher elevations in the Pyrenees (Navarro-Serrano tivorous Passerines in many temperate forests are less seg- et al., 2020). Therefore, insects are likely more abundant regated in both foraging sites and dietary components in in plots below tree-line, and thus, insectivorous Passerines summer, when insect prey is more abundant. Insect prey in might be less segregated at these lower elevations. However, autumn and winter is less abundant, leading to resource par- in our study percent of open space and/or elevation had no titioning and inter-species competition (Betts, 1955; Gibb, effect upon the richness of the bird diets, or when the most 1954; Lister, 1980). There has been disagreement over sea- common birds were examined separately. While more birds sonal segregation and diet. For example, Ulfstrand (1977) were caught in plots below tree-line than in plots at tree- found spatial segregation in summer compared to autumn, line, the diet composition within both these groups showed whereas Wagner (1981) and Almeida and Granadeiro (2000) low overlap, i.e., the diet among birds in plots at tree-line found no significant seasonal spatial differences. Obeso had as much overlap as the diet of birds in plots below tree- (1987) found no spatial difference but did find significantly line. More data would be needed to document and compare different dietary components. Finally, a recent study found the diet of each 14 bird species we studied, but this lack of high dietary overlap between communities of insectivorous link between open space and elevation may indicate high 13 154 Community Ecology (2023) 24:147–157 mobility of the more common generalists that occupy high- may therefore not be a significant factor affecting the diet of elevation Pyrenean landscapes. some adult generalist insectivorous Passerines. Our adult Passerine barcoding study was performed in Much remains to be discovered regarding the diets P. mugo forests, so it is unsurprising that five of the eleven of adult European insectivorous Passerines (Cholewa & species most likely to be present in the bird feces from this Wesołowski, 2011). In the future, metabarcoding will habitat were conifer pests. The two species most likely to be undoubtably continue to elucidate the relationship between present in the bird feces were conifer aphids; 53.03% of the birds, insects, and landscape and has the potential to reveal samples contained Cinara pini (L., 1758) and 37.88% con- vast quantities of dietary data. Our results showed very high tained Eulachnus rileyi (Williams, 1911) (Supplementary prey diversity and very little overlap within and among material 6). Cinara pini is a common and native conifer pest insectivorous Passerines. Spatial trends (open space and in Europe. Eulachnus riley, however, is considered rare to elevation) had little effect on prey diversity and overlap. uncommon in its native range in Europe and is considered While these data indicate that the dietary plasticity of the a pest outside its native range (Blackman & Eastop, 1994), more common birds is high, more studies are needed to so it is interesting to find this species to be common in our reveal dietary components of mountain species, such as A. study. Even though, as a group, the European Paridae are spinoletta. some of most intensely studied birds in the world (Betts, Supplementary Information The online version contains supplemen- 1955; Cowie & Hinsley, 1988; Gibb, 1954; Lack, 1964; Ulf- tary material available at https://d​ oi.o​ rg/1​ 0.1​ 007/s​ 42974-0​ 23-0​ 0148-4. strand, 1976), data are limited relative to their adult diets. However, the abundance of aphids in adult diets in our study Acknowledgements Many people were integral to the success of is similar to two other studies: Shutt et al. (2020) reported this project. Earthwatch Institute volunteers donated countless hours dedicated to capturing birds and collecting samples. Cristina Ame- aphids comprised three of the top ten prey taxa and had the tller Quero was extremely helpful with organizing and filtering DNA highest presence incidences, and Betts (1955) found aphids sequences and initiated much of the statistical work. Cesc Murria and comprised over 50% of the adult diet of three species of Josep Piñol were especially helpful in regard to the bioinformatic pro- hole-nesting European Paridae in June. Birds play important cesses. Roberto Molowny provided critical expertise to statistics sec- tions. Gerald Dinkins contributed thoughtful proofreading assistance. roles in top-down control of forest arthropod populations Amie Carlone’s GIS advice was very helpful. Funding was provided by (Fayt et al., 2005; Gunnarsson, 1995, 1996; Philpott et al., Earthwatch Institute, Daniel B. Warnell School of Forestry and Natural 2004; Schwenk et al., 2010). It is possible the Passerines in Resources within the University of Georgia, and the Collections Sec- our study are shaping arthropod communities and causing a tion of the Museu de Ciències Naturals de Barcelona trophic cascade by affecting tree growth. Research examin- Funding Open Access Funding provided by Universitat Autonoma de ing trophic cascade affects by bird predation, however, have Barcelona. revealed complex interactions or mixed results (Gruner, 2004; Schwenk et al., 2010). Data availability The raw data present at this link: https://​doi.​org/​10.​ 5281/​zenodo.​69687​70. Regardless, we now know many high-elevation species are under pressure (Öztürk et al., 2015), and alpine-breeding Declarations Passerines such as Anthus spinoletta (L., 1758) (water pipit) are precipitously declining in some areas (Flousek et al., Conflict of interest On behalf of all authors, the corresponding author 2015). We caught only one specimen of A. spinoletta; the states that there is no conflict of interest. remainder of our species are elevational generalists and not Open Access This article is licensed under a Creative Commons Attri- confined to high elevation. While our data does not shed bution 4.0 International License, which permits use, sharing, adapta- light on alpine specialists, the lack of differences in dietary tion, distribution and reproduction in any medium or format, as long overlap and diversity relative to open space and elevation, 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 not to mention the extremely low levels of intra-species die- were made. The images or other third party material in this article are tary overlap, suggest that adult diet may not be a constrain- included in the article's Creative Commons licence, unless indicated ing factor in populational growth of some generalist insec- otherwise in a credit line to the material. If material is not included in tivorous Passerines in and around the Pyrenean tree-line. At the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will least one European generalist insectivorous Passerine bird need to obtain permission directly from the copyright holder. To view a seems to display extreme plasticity in timing of egg-laying copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. (Wesołowski et al., 2016), and other ecological requirements such as suitable nesting sites and the provisioning needs for nestling could be more plausible population constraints. The upward migration of the tree-line in the Pyrenees, a phe- nomenon likely caused by both land-use shifts and climate change (Batllori & Gutiérrez, 2008; Batllori et al., 2010), 13 Community Ecology (2023) 24:147–157 155 References from biomonitoring of aquatic macroinvertebrates. Frontiers in Ecology and Evolution, 7, 434. Callahan, B. J., McMurdie, P. J., Rosen, M. J., Han, A. W., Johnson, A. Alatalo, R. V. (1981). Interspecific competition in tits: Parus spp. and J. A., & Holmes, S. P. (2016). DADA2: High-resolution sample the goldcrest Regulus regulus: Foraging shifts in multispecific inference from Illumina amplicon data. Nature Methods, 13(7), flocks. Oikos, 37(3), 335–344. 581–583. Alatalo, R. V., Gustafsson, L., & Lundberg, A. (1986). Interspecific Carreras, J., Carrillo, E., Massalles, R., & Ninot, J. M. (1996). Delimi- competition and niche changes in tits (Parus spp.): Evaluation tation of the supra-forest zone in the Catalan Pyrenees. Bulletin of nonexperimental data. The American Naturalist, 127(6), De La Societé Linnéenne De Provence, 47, 27–36. 819–834. Ceia, R., Machado, R., & Ramos, J. (2016). Nestling food of three Almeida, J., & Granadeiro, J. (2000). Seasonal variation of foraging hole-nesting passerine species and experimental increase in their niches in a guild of passerine birds in a cork-oak woodland. densities in Mediterranean oak woodlands. European Journal of Ardea, 88, 243–252. Forest Research, 135, 839–847. Ameztegui, A., Coll, L., Brotons, L., & Ninot, J. M. (2016). Land-use Centre de Biodiversitat de l'Institut d'Estudis Andorrans. (2012). legacies rather than climate change are driving the recent upward Sistema d’Informació Geogràfica i Mediambiental d’Andorra. shift of the mountain tree line in the Pyrenees. Global Ecology Retrieved from 12 May 2021 https://​www.​iea.​ad/​mapa-​de-​cober​ and Biogeography, 25(3), 263–273. tes-​del-​sol-d-​andor​ra-​2012. Atiénzar, F., Belda Eduardo, J., & Barba, E. (2013). Coexistence of Chao, A., Gotelli, N. J., Hsieh, T. C., Sander, E. L., Ma, K. H., Colwell, mediterranean tits: A multidimensional approach. Écoscience, R. K., & Ellison, A. M. (2014). Rarefaction and extrapolation 20(1), 40–47. with Hill numbers: A framework for sampling and estimation in Bale, J. S., Masters, G. J., Hodkinson, I. D., Awmack, C., Bezemer, species diversity studies. Ecological Monographs, 84(1), 45–67. T. M., Brown, V. K., Butterfield, J., Buse, A., Coulson, J. C., Chao, A., & Jost, L. (2012). Coverage-based rarefaction and extrapo- Farrar, J., & Good, J. E. (2002). Herbivory in global climate lation: Standardizing samples by completeness rather than size. change research: Direct effects of rising temperature on insect Ecology, 93(12), 2533–2547. herbivores. Global Change Biology, 8(1), 1–16. Cholewa, M., & Wesołowski, T. (2011). Nestling food of European Barba, E., & Gil-Delgado, J. (1990). Seasonal variation in nestling hole-nesting Passerines: Do we know enough to test the adap- diet of the great tit Parus major in orange groves in eastern tive hypotheses on breeding seasons? Acta Ornithologica, Spain. Ornis Scandinavica, 21, 296–298. 46(2), 105–116. Barras, A. G., Niffenegger, C. A., Candolfi, I., Hunziker, Y. A., & Cowie, R. J., & Hinsley, S. A. (1988). Feeding ecology of great Arlettaz, R. (2021). Nestling diet and parental food provision- tits (Parus major) and blue tits (Parus caeruleus), breeding in ing in a declining mountain passerine reveal high sensitivity suburban gardens. Journal of Animal Ecology, 57(2), 611–626. to climate change. Journal of Avian Biology, 52(2). https://​doi.​ Crisol-Martínez, E., Moreno-Moyano, L. T., Wormington, K. R., org/​10.​1111/​jav.​02649 Brown, P. H., & Stanley, D. (2016). Using next-generation Bates, D., Mächler, M., Bolker, B., & Walker, S. (2015). Fitting sequencing to contrast the diet and explore pest-reduction linear mixed-effects models using lme4. Journal of Statistical services of sympatric bird species in macadamia orchards in Software, 67(1), 1–48. Australia. PLoS ONE, 11(3), e0150159. Batllori, E., Camarero, J. J., & Gutiérrez, E. (2010). Current regen- Currie, D., Nour, N., & Adriaensen, F. (1996). A new technique for eration patterns at the tree line in the Pyrenees indicate simi- filming prey delivered to nestlings, making minimal alterations lar recruitment processes irrespective of the past disturbance to the nest box. Bird Study, 43, 380–382. regime. Journal of Biogeography, 37(10), 1938–1950. da Silva, L. P., Mata, V. A., Lopes, P. B., Lopes, R. J., & Beja, Batllori, E., & Gutiérrez, E. (2008). Regional tree line dynamics in P. (2020). High-resolution multi-marker DNA metabarcod- response to global change in the Pyrenees. Journal of Ecol- ing reveals sexual dietary differentiation in a bird with minor ogy, 96(6), 1275–1288. https://​doi.​org/​10.​1111/j.​1365-​2745.​ dimorphism. Ecology and Evolution, 10(19), 10364–10373. 2008.​01429.x Davies, S., Vaughan, I., Thomas, R., Marchbank, A., Drake, L., & Betts, M. M. (1955). The food of titmice in oak woodland. Journal Symondson, W. (2022). Seasonal and ontological variation in of Animal Ecology, 24(2), 282–323. diet and age-related differences in prey choice, by an insectivo- Bishton, G. (1986). The diet and foraging behaviour of the dun- rous songbird. Ecology and Evolution, 12(8), e9180. nock Prunella modularis in a hedgerow habitat. Ibis, 128(4), Davis, N. M., Proctor, D. M., Holmes, S. P., Relman, D. A., & Calla- 526–539. han, B. J. (2018). Simple statistical identification and removal Blackman, R. L., & Eastop, V. F. (1994). Aphids on the world’s trees: of contaminant sequences in marker-gene and metagenomics An identification and information guide. CAB International in data. Microbiome, 6(1), 226. association with The Natural History Museum. Deagle, B. E., Jarman, S. N., Coissac, E., Pompanon, F., & Taberlet, Bolyen, E., Rideout, J. R., Dillon, M. R., Bokulich, N. A., Abnet, C. P. (2014). DNA metabarcoding and the cytochrome c oxidase C., Al-Ghalith, G. A., Alexander, H., Alm, E. J., Arumugam, M., subunit I marker: not a perfect match. Biology Letters, 10(9), Asnicar, F., & Bai, Y. (2019). Reproducible, interactive, scalable 20140562. and extensible microbiome data science using QIIME 2. Nature Deagle, B. E., Thomas, A. C., Shaffer, A. K., Trites, A. W., & Jar- Biotechnology, 37(8), 852–857. man, S. N. (2013). Quantifying sequence proportions in a Brandt, M. I., Trouche, B., Quintric, L., Günther, B., Wincker, P., Pou- DNA-based diet study using Ion Torrent amplicon sequenc- lain, J., & Arnaud-Haond, S. (2021). Bioinformatic pipelines ing: Which counts count? Molecular Ecology Resources, 13(4), combining denoising and clustering tools allow for more com- 620–633. prehensive prokaryotic and eukaryotic metabarcoding. Molecular Fayt, P., Machmer, M. M., & Steeger, C. (2005). Regulation of Ecology Resources, 21(6), 1904–1921. spruce bark beetles by woodpeckers—a literature review. For- Bush, A., Compson, Z. G., Monk, W. A., Porter, T. M., Steeves, R., est Ecology and Management, 206(1), 1–14. Emilson, E., Gagne, N., Hajibabaei, M., Roy, M., & Baird, D. J. Feuillet, T., Birre, D., Milian, J., Godard, V., Clauzel, C., & Ser- (2019). Studying ecosystems with DNA metabarcoding: Lessons rano-Notivoli, R. (2020). Spatial dynamics of alpine tree lines under global warming: What explains the mismatch between 13 156 Community Ecology (2023) 24:147–157 tree densification and elevational upward shifts at the tree line differences are widespread in food-limited warbler (Parulidae) ecotone? Journal of Biogeography, 47(5), 1056–1068. communities. Ibis, 164(1), 44–59. Flousek, J., Telenský, T., Hanzelka, J., & Reif, J. (2015). Population Körner, C. (2012). Alpine treelines: Functional ecology of the global trends of central European montane birds provide evidence for high elevation tree limits. Springer. adverse impacts of climate change on high-altitude species. Krehenwinkel, H., Wolf, M., Lim, J. Y., Rominger, A. J., Simison, PLoS ONE, 10(10), e0139465. W. B., & Gillespie, R. G. (2017). Estimating and mitigating Galan, M., Pons, J. B., Tournayre, O., Pierre, É., Leuchtmann, M., amplification bias in qualitative and quantitative arthropod Pontier, D., & Charbonnel, N. (2018). Metabarcoding for the metabarcoding. Science and Reports, 7, 17668–17612. parallel identification of several hundred predators and their Lack, D. (1964). A long-term study of the great tit (Parus major). prey: Application to bat species diet analysis. Molecular Ecol- Journal of Animal Ecology, 33, 159–173. ogy Resources, 18(3), 474–489. Lens, L. (1996). Wind stress affects foraging site competition Garfinkel, M., Minor, E., & Whelan, C. J. (2022). Using faecal meta- between crested tits and willow tits. Journal of Avian Biology, barcoding to examine consumption of crop pests and benefi- 27(1), 41–46. cial arthropods in communities of generalist avian insectivores. Lister, B. C. (1980). Resource variation and the structure of British Ibis, 164(1), 27–43. bird communities. Proceedings of the National Academy of Gibb, J. (1954). Feeding ecology of tits, with notes on treecreeper Sciences—PNAS, 77(7), 4185–4187. and goldcrest. Ibis, 96(4), 513–543. López-Moreno, J. I., Gascoin, S., Herrero, J., Sproles, E. A., Pons, Grace, J., Berninger, F., & Nagy, L. (2002). Impacts of climate M., Alonso-González, E., Hanich, L., Boudhar, A., Mus- change on the tree line. Annals of Botany, 90(4), 537–544. selman, K. N., Molotch, N. P., & Sickman, J. (2017). Different Gruner, D. S. (2004). Attenuation of top-down and bottom-up sensitivities of snowpacks to warming in Mediterranean cli- forces in a complex terrestrial community. Ecology, 85(11), mate mountain areas. Environmental Research Letters, 12(7), 3010–3022. 074006. Grzędzicka, E. (2018). Habitat and diet variability of two coexisting Mansor, M. S., Rozali, F. Z., Davies, S., Nor, S. M., & Ramli, R. tit species in central European forests. Bird Study, 65(1), 52–61. (2022). High-throughput sequencing reveals dietary segregation Gunnarsson, B. (1995). Arthropods and passerine birds in coniferous in Malaysian babblers. Current Zoology, 68, 381–389. forest: The impact of acidification and needle-loss. Ecological Martoni, F., Piper, A. M., Rodoni, B. C., & Blacket, M. J. (2022). Bulletins, 44, 248–258. Disentangling bias for non-destructive insect metabarcoding. Gunnarsson, B. (1996). Bird predation and vegetation structure PeerJ, 10, e12981–e12981. affecting spruce-living arthropods in a temperate forest. Jour- McCain, C. M. (2009). Global analysis of bird elevational diversity. nal of Animal Ecology, 65(3), 389–397. Global Ecology and Biogeography, 18(3), 346–360. Hajibabaei, M., Singer, G. A. C., Hebert, P. D. N., & Hickey, D. McClenaghan, B., Nol, E., & Kerr, K. C. R. (2019). DNA metabar- A. (2007). DNA barcoding: How it complements taxonomy, coding reveals the broad and flexible diet of a declining aerial molecular phylogenetics and population genetics. Trends in insectivore. The Auk, 136(1), 1–11. Genetics, 23(4), 167–172. Michalski, M., Nadolski, J., Marciniak, B., Loga, B., & Bańbura, J. Hartig, F. (2022). DHARMa: Residual diagnostics for hierarchi- (2011). Faecal analysis as a method of nestling diet determina- cal (multi-level/mixed) regression models. R package version tion in insectivorous birds: A case study in blue tits Cyanistes 0.4.6. R Development Core Team. caeruleus and great tits Parus major. Acta Ornithologica, Hartley, P. H. T. (1987). Ecological aspects of the foraging behaviour 46(2), 164–172. of crested tits Parus cristatus. Bird Study, 34(2), 107–111. Moran, A. J., Prosser, S. W. J., & Moran, J. A. (2019). DNA meta- Hicks, T. M., Knowles, S. O., & Farouk, M. M. (2018). Global pro- barcoding allows non-invasive identification of arthropod prey visioning of red meat for flexitarian diets. Frontiers in Nutri- provisioned to nestling Rufous hummingbirds (Selasphorus tion, 5, 50–50. rufus). PeerJ, 7, e6596. Hodkinson, I. D. (2005). Terrestrial insects along elevation gradi- Mountain Research Initiative EDW Working Group. (2015). Eleva- ents: Species and community responses to altitude. Biological tion-dependent warming in mountain regions of the world. Nat Reviews, 80(3), 489–513. Clima Chang., 5(5), 424–430. Hou, J., Li, L., Wang, Y., Wang, W., Zhan, H., Dai, N., & Lu, P. Navarro-Serrano, F., López-Moreno, J., Azorin-Molina, C., Alonso- (2021). Influences of submerged plant collapse on diet compo- González, E., Aznarez-Balta, M., Buisan, S., & Revuelto, J. sition, breadth, and overlap among four crane species at Poy- (2020). Elevation effects on air temperature in a topographi- ang Lake, China. Frontiers in Zoology, 18(1), 24. cally complex mountain valley in the Spanish Pyrenees. Atmos- Inger, R., Gregory, R., Duffy, J. P., Stott, I., Vorisek, P., & Gaston, phere, 11, 656. K. J. (2015). Common European birds are declining rapidly Nielsen, J. M., Clare, E. L., Hayden, B., Brett, M. T., & Kratina, while less abundant species’ numbers are rising. Ecology Let- P. (2018). Diet tracing in ecology: Method comparison and ters, 18(1), 28–36. selection. Methods in Ecology and Evolution, 9(2), 278–291. Jarrett, C., Powell, L. L., McDevitt, H., Helm, B., & Welch, A. J. Nour, N., Currie, D., Matthysen, E., Van Damme, R., & Dhondt, A. (2020). Bitter fruits of hard labour: Diet metabarcoding and A. (1998). Effects of habitat fragmentation on provisioning telemetry reveal that urban songbirds travel further for lower- rates, diet and breeding success in two species of tit (great tit quality food. Oecologia, 193, 377–388. and blue tit). Oecologia, 114(4), 522–530. Jusino, M. A., Banik, M. T., Palmer, J. M., Wray, A. K., Xiao, L., Nyffeler, M., Şekercioğlu, Ç. H., & Whelan, C. J. (2018). Insectivo- Pelton, E., Barber, J. R., Kawahara, A. Y., Gratton, C., Peery, rous birds consume an estimated 400–500 million tons of prey M. Z., & Lindner, D. L. (2017). An improved method for uti- annually. The Science of Nature, 105(7), 47. lizing high-throughput amplicon sequencing to determine the O’Rourke, D., Bokulich, N., Jusino, M., MacManes, M., & Foster, J. diets of insectivorous animals. Molecular Ecology Resources, (2020). A total crapshoot? Evaluating bioinformatic decisions 19, 176–190. in animal diet metabarcoding analyses. Ecology and Evolution, Kent, C. M., Huh, K. M., Hunter, S. C., Judson, K., Powell, L. L., & 10(18), 9721–9739. Sherry, T. W. (2022). High resource overlap and small dietary 13 Community Ecology (2023) 24:147–157 157 Obeso, J. R. (1987). Uso del espacio y alimentation de los Parus spp Sehhatisabet, M. E., Kiabi, B., Pazuki, A., Alipanah, H., Khaleghi- en bosques mixtos de la Sierra de Cazorla. Ardeola, 34(1), zadeh, A., Barari, H., Basiri, R., & Aghabeigi, F. (2008). Food 61–77. diversity and niche-overlap of sympatric tits (great tit, Parus Oksanen, J., Blanchet, F.G., Kindt, R., Legendre, P., Minchin, P.R., major, blue tit, Cyanistes caeruleus and coal tit Periparus ater) O'Hara, R.B. and Wagner, H. (2020). vegan: Community Ecol- in the Hyrcanian Plain forests. Zoology in the Middle East, 44(1), ogy Package Version R package version 2.5-7 18–30. OPCC-CTP. (2018). Climate change in the Pyrenees: Impacts, vulner- Şekercioğlu, Ç. H., Fullwood, M. J., Cerling, T. E., Brenes, F. O., Daily, abilities and adaptation. Observatory PCC. G. C., Ehrlich, P. R., Chamberlain, P., & Newsome, S. D. (2023). Orłowski, G., Frankiewicz, J., & Karg, J. (2017). Nestling diet optimi- Using stable isotopes to measure the dietary responses of Costa zation and condition in relation to prey attributes and breeding Rican forest birds to agricultural countryside. Frontiers in Ecol- patch size in a patch-resident insectivorous passerine: An opti- ogy and Evolution, 11, 202. mal continuum and habitat constraints. Journal of Ornithology, Senécal, S., Riva, J.-C., O’Connor, R. S., Hallot, F., Nozais, C., & 158(1), 169–184. Vézina, F. (2021). Poor prey quality is compensated by higher Öztürk, M. A., Hakeem, K. R., Hanum, I. F., & Efe, R. (2015). Climate provisioning effort in passerine birds. Science and Reports, 11, change impacts on high-altitude ecosystems. Springer. 11182. Pagani-Núñez, E., Ruiz, I., Quesada, J., Negro, J., & Senar, J. C. Shoffner, A., Wilson, A. M., Tang, W., & Gagné, S. A. (2018). The (2011). The diet of great tit Parus major nestlings in a Medi- relative effects of forest amount, forest configuration, and urban terranean Iberian forest: The important role of spiders. Animal matrix quality on forest breeding birds. Science and Reports, Biodiversity and Conservation, 34, 355–361. 8, 17140. Philpott, S. M., Greenberg, R., Bichier, P., & Perfecto, I. (2004). Shutt, J. D., Nicholls, J. A., Trivedi, U. H., Burgess, M. D., Stone, G. Impacts of major predators on tropical agroforest arthro- N., Hadfield, J. D., & Phillimore, A. B. (2020). Gradients in rich- pods: comparisons within and across taxa. Oecologia, 140(1), ness and turnover of a forest passerine’s diet prior to breeding: 140–149. A mixed model approach applied to faecal metabarcoding data. Piñol, J., Mir, G., Gomez-Polo, P., & Agusti, N. (2015). Universal and Molecular Ecology, 29(6), 1199–1213. blocking primer mismatches limit the use of high-throughput Snider, A. M., Bonisoli-Alquati, A., Pérez-Umphrey, A. A., Stouffer, P. DNA sequencing for the quantitative metabarcoding of arthro- C., & Taylor, S. S. (2021). Metabarcoding of stomach contents pods. Molecular Ecology Resources, 15(4), 819–830. and fecal samples provide similar insights about Seaside Sparrow R Core Team. (2021). R: A language and environment for statisti- diet. Ornithological Applications, 124(1), 1–12. cal computing. R foundation for statistical computing version Sottas, C., Reif, J., Kreisinger, J., Schmiedová, L., Sam, K., Osiejuk, T. 1.3.1056. R Core Team. S., & Reifová, R. (2020). Tracing the early steps of competition- Rahbek, C. (2005). The role of spatial scale and the perception of large- driven eco-morphological divergence in two sister species of scale species-richness patterns. Ecology Letters, 8(2), 224–239. passerines. Evolutionary Ecology, 34(4), 501–524. Resano-Mayor, J., Korner-Nievergelt, F., Vignali, S., Horrenberger, Suarez-Seoane, S., Osborne, P. E., & Baudry, J. (2002). Responses of N., Barras, A. G., Braunisch, V., Pernollet, C. A., & Arlettaz, birds of different biogeographic origins and habitat requirements R. (2019). Snow cover phenology is the main driver of foraging to agricultural land abandonment in northern Spain. Biological habitat selection for a high-alpine passerine during breeding: Conservation, 105(3), 333–344. Implications for species persistence in the face of climate change. Sutton, A. O., Strickland, D., Freeman, N. E., & Norris, D. R. (2021). Biodiversity and Conservation, 28(10), 2669–2685. Climate-driven carry-over effects negatively influence population Ribeiro, Â., Smit, B., & Gilbert, M. (2019). 31° South: Dietary niche growth rate in a food-caching boreal passerine. Global Change of an arid-zone endemic passerine. Environmental DNA, 1, Biology, 27(5), 983–992. 109–118. Tallamy, D. W., & Shriver, W. G. (2021). Are declines in insects and Ritter, C. D., Dal Pont, G., Stica, P. V., Horodesky, A., Cozer, N., Netto, insectivorous birds related? Ornithological Applications, 123(1), O. S. M., Henn, C., Ostrensky, A., & Pie, M. R. (2022). Wanted 8. not, wasted not: Searching for non-target taxa in environmen- Torok, J., & Tóth, L. (1999). Asymmetric competition between two tal DNA metabarcoding by-catch. Environmental Advances, 7, tit species: A reciprocal removal experiment. Journal of Animal 100169. Ecology, 68(2), 338–345. Roura, N., Pons, P., Etienne, M., & Lambert, B. (2005). Transforma- Ulfstrand, S. (1976). Feeding niches of some passerine birds in a south tion of a rural landscape in the eastern Pyrenees between 1953 Swedish coniferous plantation in winter and summer. Ornis and 2000. Mountain Research and Development, 25, 252–261. Scandinavica, 7(1), 21–27. Rytkönen, S., Vesterinen, E. J., Westerduin, C., Leviäkangas, T., Vatka, Ulfstrand, S. (1977). Foraging niche dynamics and overlap in a guild of E., Mutanen, M., Välimäki, P., Hukkanen, M., Suokas, M., & Passerine birds in a south Swedish coniferous woodland. Oeco- Orell, M. (2019). From feces to data: A metabarcoding method logia, 27(1), 23–45. for analyzing consumed and available prey in a bird-insect food Wagner, J. L. (1981). Seasonal change in guild structure: Oak wood- web. Ecology and Evolution, 9(1), 631–639. land insectivorous birds. Ecology, 62(4), 973–981. Sanz, J. J., Garcia-Navas, V., & Ruiz-Peinado, J. V. (2010). Effect of Wesołowski, T., Cholewa, M., Hebda, G., Maziarz, M., & Rowiński, P. habitat type and nest-site characteristics on the breeding perfor- (2016). Immense plasticity of timing of breeding in a sedentary mance of great and blue tits (Parus major and P. caeruleus) in a forest passerine, Poecile palustris. Journal of Avian Biology, Mediterranean landscape. Ornis Fennica, 87(2), 41. 47(1), 129–213. Schwenk, W. S., Strong, A. M., & Sillett, T. S. (2010). Effects of bird predation on arthropod abundance and tree growth across an elevational gradient. Journal of Avian Biology, 41(4), 367–377. 13