Conserva

Francisco J. García-De León ^a, Gorgonio Ruiz-Campos ^b, *, Jesús Roberto Oyervides-Figueroa ^c, Gonzalo De León-Girón ^b, Carlos Alberto Flores-López ^b, Dante Magdaleno-Moncayo ^d, Alicia Abadía-Cardoso ^e

^a Centro de Investigaciones Biológicas del Noroeste, S.C., Laboratorio de Genética para la Conservación, Av. Instituto Politécnico Nacional 195, Playa Palo de Santa Rita Sur, 23096 La Paz, Baja California Sur, Mexico

^b Universidad Autónoma de Baja California, Facultad de Ciencias, Carretera Ensenada-Tijuana Km. 103 s/n, 22860 Ensenada, Baja California, Mexico

^c Centro de Investigación Científica y de Educación Superior de Ensenada, Departamento de Acuicultura, Carretera Tijuana -Ensenada 3918, Zona Playitas, 22860 Ensenada, B.C., Mexico

^d Universidad Autónoma de Baja California, Facultad de Ingeniería, Arquitectura y Diseño, Carretera Ensenada-Tijuana Km. 103 s/n, 22860 Ensenada, Baja California, Mexico 

^e Universidad Autónoma de Baja California, Facultad de Ciencias Marinas, Carretera Ensenada-Tijuana Km. 103 s/n, 22860 Ensenada, Baja California, Mexico

*Corresponding author: gruiz@uabc.edu.mx (G. Ruiz-Campos)

Received: 20 August 2024; accepted: 20 February 2025

Abstract

We assembled and annotated the mitochondrial genome of golden eagles from the northwest of Baja California, Mexico, using reference and de novo strategies to analyze the synteny of mitochondrial genes, the phylogenetic relationships, and genetic variation of mitochondrial DNA. The length of the Golden Eagle mitogenome was 17,472 bp (base pairs) with a base composition of A (29.8%), C (32.5%), G (14.0%), and T (23.6%). The mitogenome contains 13 genes coding for the protein complexes coxI II-III, Cytb, cATP 6 and 8, and Nicotinamide Adenine Dinucleotide (NADH 1-6); this arrangement is consistent with the general model of the mitogenome reported in other congeneric members. Mitogenomes of individuals from northwestern Baja California are unique and they differ from the mitogenome of southern California golden eagles in 3 traits: 1) molecule size is 140 bp larger than that previously reported, 2) the addition in the annotation of a region called pseudo control (ψRC), and 3) the annotation in 2 fractions of the coding region for the protein NADH dehydrogenase subunit 3 (ND3). The genetic diversity and phylogenetic analyses of the individual genes and mitogenome support a close genetic relatedness between the golden eagles from northwestern Baja California and southern California region. 

Keywords: Mediterranean region; Mitochondrial lineages; Mitochondrial genome; Next generation sequencing; Non-model species

Mitogenoma de águila real (Aquila chrysaetos) en el noroeste de Baja California, México: relaciones filogenéticas y variación genética

Resumen

Ensamblamos y anotamos el genoma mitocondrial del águila real del noroeste de Baja California, México, utilizando estrategias de referencia y de novo para analizar la sintenia de genes mitocondriales, las relaciones filogenéticas y la variación genética del DNA mitocondrial. La longitud del mitogenoma fue 17,472 pb (pares de bases) con una composición de bases de A (29.8%), C (32.5%), G (14.0%) y T (23.6%). El mitogenoma contiene 13 genes codificantes de los complejos proteínicos coxI II-III, Cytb, cATP 6 y 8, y nicotinamida adenina dinucleótido (NADH 1-6); esto fue consistente con el modelo general del mitogenoma reportado en otros congéneres. Los mitogenomas de individuos del noroeste de Baja California son únicos y se diferencian del mitogenoma de individuos del sur de California en 3 rasgos: 1) el tamaño de molécula es 140 pb más grande que el reportado, 2) adición de la región llamada pseudocontrol (ψRC) y 3) anotación en 2 fracciones de la región codificante de la proteína NADH deshidrogenasa subunidad 3 (ND3). La diversidad genética y los análisis filogenéticos de los genes individuales y el mitogenoma respaldan una estrecha relación genética entre las águilas reales del noroeste de Baja California y la región del sur de California.

Palabras clave: Región mediterránea; Linajes mitocondriales; Genoma mitocondrial; Secuenciación de próxima generación; Especies no modelo

Introduction

The Golden Eagle, Aquila chrysaetos, is an accipitrid of boreal distribution that inhabits a variety of open and semi-open biotopes from sea level to 3,630 m in altitude in biomes as tundra, chaparral, temperate grassland, temperate deciduous forest, and coniferous forest (De León-Girón et al., 2016; Flesch et al., 2020; Kochert et al., 2002; Watson et al., 2011). This emblematic species is known to occur in Mexico in the arid and semiarid environments of the northern and central regions, including some locations as south as Oaxaca (Bolger et al., 2014; De León-Girón et al., 2016; Howell & Webb, 1995; Rodríguez-Estrella, 2002; Rodríguez-Estrella et al., 1991, 2020). Although the demography of the Golden Eagle has been determined with different genetic methods that confirm the structuring of its populations (Craig et al., 2016; Doyle et al., 2016), little is known about the genetic identity of populations of this eagle in its distribution range. Northwestern Baja California is considered the greatest nesting and conservation potential region for the Golden Eagle in Mexico (De León-Girón et al., 2016; Rodríguez-Estrella, 2002; Rodríguez-Estrella et al., 1991; Tracey et al., 2017), with vast areas of habitats still pristine for the population conservation of this species shared with the United States of America (Craig et al., 2016; De León-Girón et al., 2016, 2024; Doyle et al., 2014; Katzner et al., 2023).

Studies based on mitochondrial DNA (mtDNA) allow determining the current state of conservation of species, as well as the identification of significant evolutionary and management units with their demographic aspects (Moritz, 1994). With the advent of next-generation sequencing (NGS) technologies (especially on complete mitochondrial genomes), new studies with non-model species became more common. As was the case for various species of the family Accipitridae, including the Golden Eagle. Doyle et al. (2014) were the first to describe the nuclear and mitochondrial genome of the Golden Eagle from the central California region, in the USA.

In this study we focus on the synteny of mitochondrial genes, the phylogenetic relationships, and the mitochondrial genetic diversity among Golden Eagle individuals in northwestern Baja California, Mexico. Given the high migration potential of this species, individuals collected in northwestern Baja California are predicted to be phylogenetically related to the westernmost Golden Eagle population in the USA. Therefore, studies focused on examining the genetic variation at the mitochondrial level will allow better decision-making for management and conservation programs for the species in a binational spectrum. In the same way, our study will be a valuable reference for analyzing other Golden Eagle populations in Mexico and other regions of its distribution range.

Materials and methods

Tissue samples of 6 Golden Eagle specimens from northwestern Baja California, Mexico were obtained. These specimens were found dead in different agriculture valleys of the Mediterranean region in northwestern Baja California, Mexico, between the municipalities of Rosarito and San Quintín, from 1995 to 2014, and deposited as vouchers in the Bird Collection of the Science Faculty, at the Autonomous University of Baja California (UABC), campus Ensenada (Fig. 1, Supplementary material: T1). 

For the DNA extraction, tissue samples of the pads of the feet of 5 Golden Eagle specimens referred to here as Ach 1 to Ach 5 were obtained. DNA using the nucleic acid purification method by differential saline precipitation (Aljanabi & Martinez, 1997) was extracted. Both feathers and blood for the Ach 6 sample (Supplementary material: T1) were used and extracted DNA with the Qiagen Blood & Tissue kit. The quality and quantity of the extracted DNA in both cases was evaluated using Nanodrop, Qubit and agarose gel electrophoresis. 

Extracted DNA (from 77 to 210 ng/µL per sample quantified in Nanodrop for Ach1-Ach5 individuals and 0.581 ng/µL for Ach6 quantified in Qubit) was purified with SpeedBeads magnetic beads (Thermo-Scientific, Waltham, MA, USA) at a 1.2:1 beads:DNA ratio and resuspended in 50 µl of TLE1X buffer. Libraries for shotgun sequencing were prepared, for which purified DNA from each individual was fragmented by sonication in a Bioruptor® (Diagenode, Liege, Belgium) using 2 rounds, each consisting of 5 cycles of 30 sec of sonication and 30 sec without sonication at the highest setting. DNA fragments were then prepared using the Kapa Biosystems® Hyper Prep Kit (KR0961–v4.15, Roche, Basel, Switzerland), with which end repair and A-tailing were performed, adapters were ligated, and PCR was performed with indexed primers (Glenn et al., 2020) for 14 cycles. Following amplification, fragment size selection was performed by double-dip with SpeedBeads magnetic beads (Thermo-Scientific, Waltham, MA, USA) that allow to preserve fragments between ~250 and ~700 base pairs (bp) for each of the samples, which are the appropriate insert size for the Illumina platform. Sequencing was performed paired end on 2 different platforms. We sequenced the Ach1-Ach5 samples on the Illumina MiSeq V.3 at the Georgia Genomics and Bioinformatics Core (GGBC) to generate 300-bp paired-end fragments, and we sequenced Ach6 on an Illumina HiSeq 4000 at the Oklahoma Medical Research Foundation Clinical Genomics Center to generate 150-bp paired-end fragments.

Bioinformatic analysis. The quality of the raw sequences was evaluated using the FastQC program (Andrews, 2010). Subsequently, the sequences using a standard treatment were filtered using the Trimmomatic software (Bolger et al., 2014) with 4 steps. In the first step, the sequencing adapters and over-represented sequences were removed by means of the ILLUMINACLIP function. The second step discarded sequences with a quality value below 30 QS with the AVGQUAL function. The third step removed fragments below 25 QS with the SLIDINGWINDOW function. The fourth step did an even stricter cleanup using MAXINFO. This step conserved sequences of at least 100 base pairs and was configured to preferentially conserve longer sequences with a value of 0.3. For the Ach 5 specimen, an extended and modified version of the standard treatment was used due to the low quality of sequences. This involved applying the SLIDINGWINDOW function, which modifies the quality value.

Assembly and annotation. Two strategies were followed to assemble the mitochondrial genomes: the reference based and the de novo strategy (Machado et al., 2015, 2018). Mapping against a reference mtDNA genome was performed using the Bowtie2 program version 2.3.4.2 with clean reads (Langmead & Salzberg, 2012). The reference mitochondrial genome was from the same species A. chrysaetos, with the GenBank accession number of KF905228.1. De novo assembly only for samples Ach 2 to Ach 6 was performed using the A5-miseq pipeline (Coil et al., 2015). Due to the large number of reads after the filter (Supplementary material: T2) sample Ach 1 was analyzed with the Velvet software version 1.1 (Zerbino & Birney, 2008). The mitogenome was annotated from the scaffolds obtained from the de novo assemblies. We determined scaffolds longer than 4 Kpb by applying a search with the BLAST tool (Basic Local Alignment Search Tool) of the NCBI (National Center for Biotechnology Information) website. The scaffolds were aligned to achieve greater coverage considering the length of the reference genome used previously. Afterwards, RNAweasel (http://megasun.bch.umontreal.ca/RNAweasel/) was implemented to the identification of tRNA’s (transfer RNA), rRNA’s (ribosomal RNA) and introns and, in turn, MFannot (http://megasun.bch.umontreal.ca/RNAweasel/) for the identification of proteins and open reading frames (Sieber et al., 2018). 

In addition to the previous annotation, MITOS web server (Bernt et al., 2013) was used to assist in the annotation of de novo mitochondrial genomes, allowing gene names, tRNA and rRNA secondary structures, and codon usage to be obtained. Finally, a manual curation of the annotations was used to review the 6 existing reading frames with the UGENE software (Okonechnikov et al., 2012). The control region was identified based on 99% similarity with an A. chrysaetos partial control region sequence (EF459579.1). Genome Vx software (Conant and Wolfe, 2008) was used to map the Golden Eagle’s mitogenome, using individual Ach 6 to achieve this analysis.

In the annotation of the de novo assembly, the ND3 protein appeared divided into 2 fractions (see Results), a trait that was not reported by Doyle et al. (2014). Therefore, an experimental verification was required through PCR amplification and its subsequent Sanger sequencing. We designed primers from de novo assembly using the NCBI Primer-BLAST program: ND3Ach-1 (5’GCCTGATACTGGCACTTCGT3’ and 5’CCCTATCAATCTGACCCACCG3’) which generates a 716 bp fragment and ND3Ach-2 (5’CTTCTTCGTCGCTACAGGCT3’ and 5’CCTTCCACCGAACCCACTTAA3’) which generates a 774 bp fragment. Eight Golden Eagle samples from the Ornithological Collection of the Faculty of Sciences of the Autonomous University of Baja California were used for PCR amplification. These samples correspond to the 6 individuals used for sequencing and 2 extra samples that are not part of the mitogenome assembly study. The conditions of the PCR reaction were as follows: 6 min at 94°C, 35 cycles of denaturation, annealing, and extension at 94°C for 30 sec, 51°C for 30 sec and at 72°C for 1.5 min, respectively. A final elongation stage at 70°C for 7 min, and a final conservation stage at 15°C. We sent the amplified fragments for sequencing to the SeqXcel.inc Company in San Diego, California, and analyzed with the ABI PRISM® 3130xl Genetic Analyzer DNA kit.

Synteny and search for polymorphisms. A posteriori synteny analysis was performed between the Golden Eagle mitogenome reported by Doyle et al. (2014) and the newly assembled mitochondrial genome of this study. This analysis was carried out with the MAUVE software (Darling et al., 2004). The same reference mitogenome was used to call SNP’s and INDEL’s for the 6 genomes assembled in this work using the SAMtools program (Li, 2011). Filtering the call quality of the variants involved discarding those with a value less than 20 Qs and keeping those with at least 5X depth (Li, 2011). Each variant was manually evaluated, retaining SNPs and INDELs where at least half of the aligned reads showed the change and verified its existence in at least one other individual (Fridjonsson et al., 2011).

Phylogenetic relationships. Phylogenetic relationships were inferred using the individual gene dataset and the mitogenome. The individual gene fragments analyzed consisted of the ND2, ND3, coxI and Cytb recovered from Aquila chrysaetos individuals from GenBank (Supplementary material: T3) and those produced in this study. The mitogenome analysis for phylogenetic relationships consisted of 10 complete mitogenomes: 6 obtained in this study, 2 reference samples of Aquila chrysaetos (LR822062 from the United Kingdom and NC_024087.1 from California, USA), 1 reference sample classified as Aquila heliaca (NC_035806.1), but has been reported to actually consist of a A. chrsyasetos individual (Sangster & Lukesenburg, 2021) and 1 outgroup (Aquila nipalensis, GenBank accession number NC_045042.1). Both datasets were aligned using the MAFFT 7 algorithm (Katoh & Standley, 2013). Removing the non-conserved regions between the sequences of the multiple alignments was done using the online program Gblocks with the default parameters (Castresana, 2000). The nucleotide substitution model and the mutation rate were determined using JModelTest (Posada, 2008), considering the Bayesian information criterion (BIC).

Two phylogenetic reconstruction methods were used: Maximum Likelihood (ML) and Bayesian analysis. In IQ-TREE (Nguyen et al., 2015) the ML method (Felsenstein, 1981) was run with the DNA substitution model selected by ModelFinder (Kalyaanamoorthy et al., 2017) and 1,000 bootstrap pseudo replicates were performed (Hoang et al., 2017). Bayesian inference was used with MrBayes program (Ronquist et al., 2012); for this, 4 Markov Monte Carlo chains (MCMC) were implemented with a total of 10 million generations, and a sampling every 1,000 generations. To construct the consensus phylogenetic tree, a 25% of burn-in was applied. The support of the nodes was evaluated via the posterior probability values. The results of both approaches were visualized with FigTree 1.4 (Rambaut, 2009). For the mitogenome phylogenetic tree, the dataset on genes that were annotated across the entire mitogenome were partitioned (Supplementary material: T4). To determine the most appropriate DNA substitution model for each gene, the Akaike Information Criterion test implemented in jModelTest2 (Darriba et al., 2012; Guindon & Gascuel, 2003) was used (see selected models for each gene in Supplementary material: T4). The partitioned mitogenome dataset was used to estimate a Bayesian phylogenetic tree in MrBayes (Ronquist et al., 2012). The Bayesian analysis was carried out by running 4 MCMC chains for 5 million generations and saving the trees every 1,000 generations. The consensus phylogenetic tree was constructed using a 10% burn-in.

Genetic diversity. Individual DNA segments from the coxI, Cytb, ND2, and -ND3 genes were used to estimate the genetic diversity parameters for the A. chrysaetos species complex since these were the genes that had a relative amount of A. chrysaetos DNA sequences available in GenBank. Using DNA sequences from the Baja California individuals produced in this study and reference sequences from the species complex (Supplementary material: T3) the following indices were estimated with DnaSP 6 (Rozas et al., 2017): the number of segregating sites (S), number of haplotypes (Nh), haplotype diversity (Hd), and nucleotide diversity (π). Genetic diversity indices were calculated among all sequences included in the respective datasets and within specific groups of sequences according to phylogenetic clades or geographic origin of samples. In addition, the genetic distance (uncorrected p-distance) between and within the phylogenetic clades observed within the mitochondrial genetic lineages in PAUP software was estimated (Swofford, 1993). The clade OTU1 included sequences from a diverse geographical background, including all the DNA sequences of the Baja California golden eagles. In contrast, the A. chrysaetos DNA sequences grouped in the other closely related clade were classified as OTU2. Additionally, the DNA sequence of an individual of A. chrysaetos from California (Southern Sierra Nevada) was analyzed separately to compare its genetic distance with those from Baja California, given the geographical proximity between both populations.

Results

Raw sequencing reads ranged from 3,324,579 (individual Ach 1) to 539,825 (individual Ach 4). The number of total sequences after the filter ranged between 3,251,746 (Ach 1) and 526,875 (Ach4) (Supplementary material: T2A). 

Assembly and annotation. There were large differences in assembly between individuals; for example, the average depth in the reference mapping for individual Ach 1 is 30.5X, while for Ach 6 is 168.3X, even though Ach 6 had fewer total reads than Ach 1. Assemblies for individuals Ach 1-5 showed fewer aligned sequences (Supplementary material: T2A). For de novo assembly, the average depth for individual Ach 6 is 8.0X. The longest scaffold generated (17,472 bp) allowed us to assemble the entire mitogenome of the Ach 6 individual, eliminating the need to align multiple scaffolds to recover the full length (Supplementary material: T2B). The data yield was inferior when doing de novo assembly for Ach 1-5 individuals, and none of these individuals allowed the recovery of the complete mitochondrial genome (Supplementary material: T3B). 

Based on the above results, we annotated and curated the mitochondrial genome using only the complete mitogenome obtained by de novo assembly of the Ach 6 individual. All 3 annotation tools identified tRNAs, and none of the tools identified introns within the sequence. The MITOS program (Fig. 2, Supplementary material: T4) was the only software that identified all the genetic elements, such as the 22 tRNAs, 2 rRNAs, and the 13 protein-coding genes that make up the mitogenome. This last annotation was the most complete, so the curation was made from it. We found a region with 7 nucleotide bases that is composed of an AGA stop codon of the ND3 protein, followed by an adenine (A) and finally an ATC start codon that codes for isoleucine. This region divides the ND3 fragment into the 2 regions proposed here (ND3a and ND3b, Fig. 2).

Synteny and search for polymorphisms. The linear order of the mitogenome of Ach 6 coincides with the reference genome (Accession number MT319112.1, Supplementary material: T5). We consider the SNPs and INDELs identified in the Ach 6 reliable, given that the average coverage is 168.3X, and their call quality was high (Supplementary material: T2A). Most SNPs and INDELs are found within genes that code for some protein, transfer RNA and ribosomal genes. The coxI gene presented the greatest number of SNPs with respect to the reference genome in the 6 individuals, while the Cytb and tRNA F genes presented the greatest length changes in the INDELs. In addition, in the ND3 gene, there is a consistent change in the sequences of the 6 individuals with respect to the reference genome (Table 1). 

Phylogenetic relationships. With the use of the mitogenome we constructed a phylogenetic tree that identified 2 clades with strong support (1): OTU1 that includes all the individuals analyzed in this study, plus 2 GenBank sequences (Aquila chryseatos, LR82062.1 from the United Kingdom and A. heliatica, NC_035806.1), and OTU2 that corresponds to a GenBank sequence of a specimen collected in California (Fig. 3). For their part, the phylogenetic trees obtained with the fragments of individual genes (ND2, ND3, Cytb and coxI) did not resolve the divergence between individuals from Baja California and California obtained with the mitogenome, but all except the ND3 gene showed a divergence (OTUs 1 and 2), mainly with respect to eagles from the European continent (when information on the collection location is reported, Supplementary material: F1). 

Genetic diversity. We observed low genetic diversity, for instance, ND3, ND2, and coxI each had a single segregating site (Table 2), resulting in low nucleotide diversity values (0.0005, 0.0008, and 0.0002, respectively). In contrast, Cytb was the most polymorphic, with a total of 10 segregating sites and a nucleotide diversity an order of magnitude larger (0.00297) (Table 2). The genetic diversity found within OTU1 was higher than that found in OTU2 (Table 2), although OTU1 was not a geographically homogeneous clade among the trees of the different mitochondrial genes, for example in the case of ND2 it was made up of North American eagles, including the individuals of this study, but coxI apart from the previous ones was constituted with eagles from Sweden, Norway and Japan (Table 2).

Discussion

Despite using 2 sequencing platforms for the mitogenome of Golden Eagle from northwestern Baja California, Mexico, we successfully assembled the mitochondrial genome of all 6 individuals by the reference method. This included some individuals (Ach 1-5) with non-consensus regions. Bolger et al. (2014) recommend a pre-processing step of the readings before any analysis, be it assembly by reference or de novo, and mention that if the library and identification adapters are not removed, they can be incorporated in the final assembly. In our case, the pre-processing of the sequences positively influenced the performance of both assemblies and the SNP’s call since the quality of the sequence of the 6 individuals was considerably improved after pre-processing in all aspects of quality reporting. 

Table 1

Location of SNP’s and INDEL’s in mitogenomes. Call of SNP’s and INDEL’s for each the Ach 1-6 specimens versus the reference genome of Golden Eagle from the southern Sierra Nevada in California, USA (GenBank accession: KF905228). Pos: Mitogenome position; Ref: nucleotide present in reference sequence. ψCR: Mitochondrial pseudogene control region. 

Table 2

Genetic diversity and haplotype composition from gene segments. Tax Set: Group of sequences included in taxonomic set; N: number of sequences; bp: base pairs of DNA sequence included in alignment; S: number of segregating sites; Nh: number of haplotypes; Hd: haplotype diversity; Nd: nucleotide diversity; North America: sequences from either Canada or USA. All sequences from ND3 formed a single clade, and thus were analyzed as a single group. OTUs 1 and 2 in each individual gene refer to the clades detected in each phylogenetic analysis, see Supplementary material F1. 

The most reliable mitogenome assembly was that of Ach 6 individual due to its higher alignment rate, its greater depth, and its 100% coverage of the reference genome. This ensemble can be considered true and not an artifact due to the procedures used, such as sampling every 500 generations and a burn-in value of 600,000 generations (Lerner & Mindell, 2005). In addition, we analyzed the data with Bowtie2, which implements an alignment strategy based on the FM-Index and Burrows-Wheeler. These programs have been shown to work well for applications such as INDEL discovery (Lindner & Friedel, 2012) and for aligning long sections of the reference genome (Thankaswamy-Kosalai et al., 2017). Hunt et al. (2014) pointed out that the best result of assembling a genome de novo is when it is contained in a single scaffold for the entire mitochondrial DNA molecule or for each chromosome in the case of the nuclear genome. The Golden Eagle mitochondrial genome assembly of individual Ach 6 was the best because it was found within a single scaffold (17,472 bp) and its depth was at least 8.0X (Supplementary material: T2B; Baker, 2012). 

The extension and position of the genes in the mitogenome of the Golden Eagle described in this study (Fig. 2, Supplementary material: T4) showed differences compared to those reported by Doyle et al. (2014) from an individual sampled in central California. For example, the size of the mitogenome described here is larger (17,472 bp) than the one reported by Doyle et al. (2014) (17,332 bp), with a difference of 140 bp. When experimentally analyzing the size of the ND3 gene amplified fragments, the size was between 700 and 800 bp, that is, the theoretically expected size. Another difference consisted in the fraction of the gene encoding the ND3 protein reported here as split into 2 segments (ND3a and ND3b). The distinction of the 2 subunits is, within the reading frame, a region made up of 7 nucleotides between the ATC start codon that starts the protein, and the TAA stop codon that marks the end of the protein. The first 3 of these 7, code for a new AGA stop codon, followed by an adenine (A) nucleotide and 3 nucleotides that code for a new ATC start codon. This causes 2 coding portions to coexist, annotated as ND3, separated by a single nucleotide that changes the reading frame, causing the second portion to appear with its start and stop codons. Mindell et al. (1998) report a nucleotide that is not translated within the ND3 protein sequence. Eberhard and Wright (2016) mentioned that such a trait is observed in the entire order Psittaciformes (parrots, parakeets, and allies). 

Specifically, a variety of distinct indels have been found within several of the mitochondrial protein-coding genes in Psittaciformes (Eberhard & Wright, 2016), with variations in terms of their evolutionary origin since some are present in all Psittaciformes and some are more recent in origin and only found within specific taxa. Slack et al. (2003) reported variation in the length of the mitochondrial ND6 gene in other avian taxa. Overall, the biological implications of these variations remain unclear until more studies on the proteins are performed to determine if the protein compositions become altered by these variations. However, the experimental verification carried out in the present study strengthens our annotation as 2 fractions for the ND3 region of the A. chrysaetos mitogenome. Also, we noted the regulatory non-coding region (the pseudo control region, ψCR) to be highly conserved among the A. chrysaetos individuals analyzed, which is a relatively conserved region within the order Accipitriformes (cf. Song et al., 2015) and is also found in the same position between the tRNA -E and tRNA-F, as shown by Liu et al. (2017) for the mitogenome of Accipiter gularis, another congeneric species. 

Synteny and search for polymorphisms. The assembled Golden Eagle mitogenome did not present changes in gene order with respect to the reference (KF905228.1, a male Golden Eagle from southern Sierra Nevada, California). As expected, neither did it present changes in the order of the genes with respect to other species of the same group of birds (Accipitridae), such as Buteo buteo (Haring et al., 2001), Accipiter virgatus (Song et al., 2015), Aquila fasciata (Jiang et al., 2015) and Accipiter gularis (Liu et al., 2017). In addition, it is consistent with one of the general models of the birds’ mitogenome that is characterized by a duplication of the control region and other adjacent genes, which were subsequently degraded, giving rise to the ψCR (Eberhard & Wright, 2016). SNP’s and INDEL’s observed in the mitogenome of the Ach 6 individual meet the needed quality (probability of not being an error) and sequence depth (Li, 2011), so they can be considered as potential markers that should be tested in future population studies, considering some technical aspects, and thus avoid the call of false positives (The 1000 Genomes Project Consortium, 2015).

Reconstruction of phylogenetic relationships. The phylogenetic tree classified all Golden Eagle individuals into 2 mitogenome clades (OTU1 and OTU2). All individuals from Baja California, an individual from the United Kingdom, and an A. heliaca individual clustered within OTU1 (Fig. 3), while OTU2 was composed of a single Golden Eagle individual from California. The genetic distance between the 2 OTUs was 0.002 nucleotide differences per site (Supplementary material: T6). This pattern is not surprising given the poor sampling of taxa for both the mitogenomes and individual genes of A. chrysaetos. Once additional mitogenomes are available for this species, it will become possible to increase the phylogenetic resolution that could confirm or reject the potential presence of multiple mitochondrial lineages circulating within Golden Eagle populations. Nevertheless, given the broad geographic area included in the mitogenome dataset (i.e., an A. chrysaetos individual from the United Kingdom, as well an A. heliaca individual), and the low genetic diversity observed within these mitochondrial sequences (see below), suggests that at the mitochondrial level, the genetic diversity found within A. chrysaetos populations is low. 

The use of genes or fragments of the mitochondrial genome allows for the analysis to include a larger data set in the number of locations and individuals. The topology of 3 of the 4 individual gene trees (ND2, Cytb and coxI) did support the presence of at least 2 mitochondrial clades (Supplementary material: F1). However, these gene trees were not consistent when grouping the North American Golden Eagle individuals on the same OTUs. 

Future studies targeting molecular markers with a higher mutational rate (i.e., microsatellites) or genomics approach could potentially uncover the presence of genetic differences that are not well defined at the mitochondrial level but might be biologically important. One such difference is seen in the species’ bivalent behavior and its migration pattern in the west (Mcintyre & Lewis, 2016) and northwest of the USA, and through Canada (Bedrosian et al., 2018). The mitogenomic lineages that potentially infer the difference between the eagles of Baja California and central California could correspond to the area where 2 ecological populations are coexisting (De León-Girón et al., 2016). The first population is migratory, originating in western North America (Oregon), while the second lineage is resident in Baja California. According to Craig et al. (2016), this lineage has 1 of the less frequent haplotypes of the mitochondrial control region and is restricted to California. The ability of the Golden Eagle to adapt to different environments (Judkins & Van Den Bussche, 2018), the large dispersal distances of reproductive or floating between Mexico and the USA (De León-Girón et al., 2016, 2024; Rodríguez-Estrella et al., 2020; Tracey et al., 2017), and its extensive home range of reproductive pairs (D’Addario et al., 2019), would support these hypotheses. Besides, the presence of these groups of individuals (reproductive and non-reproductive) in Baja California would be part of the reproductive behavior of the species (Watson et al., 2011), that is, the process of succession and substitution of reproductive pairs in the region (De León-Girón et al., 2016).

Genetic diversity. The values of mitochondrial genetic diversity calculated for most genes (except Cytb) were near zero (Table 2). We expected this result, considering that they are genes that code for proteins with highly conserved regions (Dawnay et al., 2007) and are subject to biochemical limitations that cause high levels of homoplasy (Faria et al., 2007). However, Bates et al. (2003) mentioned that ND2 is genetically more diverse than Cytb, while Faria et al. (2007) stated that ND2 is one of the most variable mitochondrial genes within birds and, therefore, regularly implemented in population genetics. A pattern we did not observe in A. chrysaetos since Cytb was the gene with the highest genetic diversity observed, with a haplotype diversity of 0.154 and 0.667 for OTU1 and OTU2, respectively, while both OTUs had a haplotype diversity of zero for the ND2 gene (Table 2).

Implications for conservation. Golden eagles are recognized for presenting large expanses of territory, and their resident, migratory and floating populations (reproductive adults without territories), promote the population gene flow (Craig et al., 2016; Poessel et al., 2022). The mitochondrial DNA data produced in this study confirm a very close genetic relationship between the northwestern Baja California individuals and those from the USA. Therefore, the execution of bi-national conservation programs by the agencies of both countries (SEMARNAT and USFWS) are priorities for the “Californian-Baja Californian” metapopulation of Golden Eagle. It is necessary to increase the number of young individuals with satellite tracking in the southern Baja California peninsula, to continue the genetic characterization (with different types of markers, mitochondrial DNA, microsatellites and SNPs obtained by Next Generation Sequencing) and population monitoring to evaluate the structure and connectivity of the species in both countries.

Our findings warrant developing joint conservation efforts between the governments of Mexico and the USA to monitor and preserve the North American Golden Eagle.

Acknowledgements

Funding for the sequencing was covered by the SAGARPA-INAPESCA PIDETEC project 2017/0647. The authors thank Travis Glen and Natalia Juliana Bayona Vásquez for their Illumina sequencing services. We thank 3 anonymous reviewers who helped improve the manuscript.

References

Aljanabi, S. M., & Martinez, I. (1997). Universal and rapid salt-extraction of high-quality genomic DNA for PCR-based techniques. Nucleic Acids Research, 25, 4692–4693. https://doi.org/10.1093/nar/25.22.4692 

Auton, A., Brooks, L. D., Durbin, R. M., Garrison, E. P., Kang, H. M., Korbel, J. O. et al. (2015). A global reference for human genetic variation. The 1000 Genomes Project Consortium. Nature, 526, 68–74. https://doi.org/10.1038/nature15393

Baker, M. (2012). De novo genome assembly: What every biologist should know. Nature Methods, 9, 333–337. https://doi.org/10.1038/nmeth.1935

Bates, J. M., Tello, J. G., & Da Silva, J. M. C. (2003). Initial assessment of genetic diversity in ten bird species of South American Cerrado. Studies on Neotropical Fauna and Environment, 38, 87–94. https://doi.org/10.1076/snfe.38.2.87.15924

Bernt, M., Donath, A., Jühling, F., Externbrink, C., Florentz, G., & Fritzsch, J. (2013). MITOS: Improved de novo metazoan mitochondrial genome annotation. Molecular Phylogenetics and Evolution, 69, 313–319. https://doi.org/10.1016/j.ympev.2012.08.023

Bolger, A. M., Lohse, M., & Usadel, B. (2014). Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics, 30, 2114–2120. https://doi.org/10.1093/bioinformatics/btu170

Castresana, J. (2000). Selection of conserved blocks from multiple alignments for their use in Phylogenetic Analysis. Molecular Biology and Evolution, 17, 540–552. https://doi.org/10.1093/oxfordjournals.molbev.a026334

Coil, D., Jospin, G., & Darling, A. E. (2015). A5-miseq: an updated pipeline to assemble microbial genomes from Illumina MiSeq data. Bioinformatics, 31, 587–589. https://doi.org/10.1093/bioinformatics/btu661

Conant, G. C., & Wolfe, K. H. (2008). GenomeVx: Simple web based creation of editable circular chromosome maps. Bioinformatics, 24, 861–862. https://doi.org/10.1093/bioinformatics/btm598

Craig, E. H., Adams, J. R., Waits, L. P., Fuller, M. R., & Whittington, D. M. (2016). Nuclear and mitochondrial DNA analyses of golden eagles (Aquila chrysaetos canadensis) from three areas in western North America; initial results and conservation implications. Plos One, 11, e0164248. https://doi.org/10.1371/journal.pone.0164248

D’Addario, M., Monroy-Vilchis, O., Zarco-González, M. M., & Santos-Fita, D. (2019). Potential distribution of Aquila chrysaetos in Mexico: Implications for conservation. Avian Biology Research, 12, 33–41. https://doi.org/10.1177/1758155918823424

Darling, A. C. E., Mau, B., Blattner, F. R., & Perna, N. T. (2004). Mauve: Multiple alignment of conserved genomic sequence with rearrangements. Genome Research, 14, 1394–1403. https://doi.org/10.1101/gr.2289704 

Darriba, D., Taboada, G., Doallo, R., & Posada, D. (2012). jModelTest 2: more models, new heuristics and parallel computing. Nature Methods, 9, 772. https://doi.org/10.1038/nmeth.2109 

Dawnay, N., Ogden, R., McEwing, R., Carvalho, G. R., & Thorpe, R. S. (2007). Validation of the barcoding gene COI for use in forensic genetic species identification. Forensic Science International, 173, 1–6. https://doi.org/10.1016/j.forsciint.2006.09.013

De León-Girón, G., Rodríguez-Estrella, R., & Ruiz-Campos, G. (2016). Estatus de distribución actual del águila real (Aquila chrysaetos) en el noroeste de Baja California, México. Revista Mexicana de Biodiversidad, 87, 1328–1335. https://doi.org/10.1016/j.rmb.2016.10.003

De León-Girón, G., Toscano, D., Ruiz-Campos, G., Porras-Peña, C., Escoto-Rodríguez, F., Sánchez-Sotomayor, V. G. et al. (2024). Aerial identification and quantification of nesting sites of Golden Eagle (Aquila chrysaetos) in the mountain ranges of central Baja California Peninsula, Mexico. Revista Mexicana de Biodiversidad, 95, e955247.https://doi.org/10.22201/ib.20078706e.2024.95.5247

Doyle, J. M., Katzner, T. E., Bloom, P. H., Ji, Y., Wijayawardena, B. K., & DeWoody, J. A. (2014). The genome sequence of a widespread apex predator, the golden eagle (Aquila chrysaetos). Plos One, 9, e95599. https://doi.org/10.1371/journal.pone.0095599

Doyle, J. M., Katzner, T. E., Roemer, G. W., Cain, J. W., Millsap, B. A., McIntyre, C. L. et al. (2016). Genetic structure and viability selection in the golden eagle (Aquila chrysaetos), a vagile raptor with a Holarctic distribution. Conservation Genetics, 17, 1307–1322. https://doi.org/10.1007/s10592-016-0863-0

Eberhard, J. R., & Wright, T. F. (2016). Rearrangement and evolution of mitochondrial genomes in parrots. Molecular Phylogenetics and Evolution, 94, 34–46. https://doi.org/10.1016/j.ympev.2015.08.011

Faria, P. J., Baus, E., Morgante, J. S., & Bruford, M. W. (2007). Challenges and prospects of population genetic studies in terns (Charadriiformes, Aves). Genetics and Molecular Biology, 30, 681–689. www.sbg.org.br

Felsenstein, J. (1981). Evolutionary trees from DNA sequences: a maximum likelihood approach. Journal of Molecular Evolution, 17, 368–376. https://doi.org/10.1007/BF01734359

Flesch, A. D., Rodríguez-Estrella, R., Gallo-Reynoso, J. P., Armenta-Méndez, L., & Montiel-Herrera, M. (2020). Distribution and habitat of the Golden Eagle (Aquila chrysaetos) in Sonora, Mexico. Revista Mexicana de Biodiversidad, 91, 1892–2019. https://doi.org/10.22201/ib.20078706e.2020.91.3056

Fridjonsson, O., Olafsson, K., Tompsett, S., Bjornsdottir, S., Consuegra, S., Knox, D. et al. (2011). Detection and mapping of mtDNA SNPs in Atlantic salmon using high throughput DNA sequencing. BMC Genomics, 12, 179. https://doi.org/10.1186/1471-2164-12-179.

Glenn, T. C., Nilsen, R. A., Kieran, T. J., Sanders, J. G., Bayona-Vásquez, N. J., Finger, J. W. et al. (2020). Adapterama I: universal stubs and primers for 384 unique dual-indexed or 147,456 combinatorially-indexed Illumina libraries (iTru & iNext). PeerJ, 7, e7755. https://doi.org/10.7717/peerj.7755

Guindon, S., & Gascuel, O. (2003). A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Systematic Biology, 52, 696–704. https://doi.org/10.1080/10635150390235520

Haring, E., Kruckenhauser, L., Gamauf, A., Riesing, M. J., & Pinsker W. (2001). the complete sequence of the mitochondrial genome of Buteo buteo (Aves, Accipitridae) indicates an early split in the phylogeny of raptors. Molecular Biology and Evolution, 18, 1892–1904. https://doi.org/10.1093/oxfordjournals.molbev.a003730

Hoang, D. T., Chernomor, O., Von Haeseler, A., Quang Minh, B., Sy Vinh, L., & Rosenberg, M. S. (2017). UFBoot2: improving the ultrafast bootstrap approximation. Molecular Biology and Evolution, 35, 518–522. https://doi.org/10.5281/zenodo.854445

Howell, S., & Webb, S. (1995). A guide of the birds of Mexico and northern Central America. Oxford, UK: Oxford University Press.

Hunt, M., Newbold, C., Berriman, M., & Otto, T. D. (2014). A comprehensive evaluation of assembly scaffolding tools. Genome Biology, 15, R42. https://doi.org/10.1186/gb-2014-15-3-r42

Jiang, L., Chen, J., Wang, P., Ren, Q., Yuan, J., Qian, C. et al. (2015). The mitochondrial genomes of Aquila fasciata and Buteo lagopus (Aves, Accipitriformes): sequence, structure and phylogenetic analyses. Plos One, 10, e0141037. https://doi.org/10.1371/journal.pone.0136297

Judkins, M. E., & Van Den Bussche, R. A. (2018). Holarctic phylogeography of golden eagles (Aquila chrysaetos) and evaluation of alternative North American management approaches. In Biological Journal of the Linnean Society, 123, 471–482. https://doi.org/10.1093/biolinnean/blx138

Kalyaanamoorthy, S., Minh, B. Q., Wong, T. K. F., Von Haeseler, A., & Jermiin L. S. (2017). ModelFinder: fast model selection for accurate phylogenetic estimates. Nature Methods, 14, 587–589. https://doi.org/10.1038/nmeth.4285

Katoh, K., & Standley, D. M. (2013). MAFFT multiple sequence alignment software Version 7: improvements in performance and usability article fast track. Molecular Biology and Evolution, 30, 772–780. https://doi.org/10.1093/molbev/mst010

Katzner, T., Kochert, M., Steenhof, K., McIntyre, C., Craig, E., & Miller, T. (2023). Golden Eagle (Aquila chrysaetos), version 2.0. InP. G. Rodewald, & B. K. Keeney (Eds.), Birds of the World. Ithaca, NY: Cornell Lab of Ornithology, USA. https://doi.org/10.2173/bow.goleag.02

Kochert, M. N., Steenhof, K., Mcintyre, C. L., & Craig, E. H. (2002). Golden Eagle (Aquila chrysaetos). In A. Poole, & F. Gill (Eds.), The birds of North America, No. 684 (pp. 1–44). Philadelphia, PA: The Birds of North America, Inc. 

Langmead, B., & Salzberg, S. L. (2012). Fast gapped-read alignment with Bowtie 2. Nature Methods, 9, 357–359. https://doi.org/10.1038/nmeth.1923

Lerner, H. R. L., & Mindell, D. P. (2005). Phylogeny of eagles, Old World vultures, and other Accipitridae based on nuclear and mitochondrial DNA. Molecular Phylogenetics and Evolution, 37, 327–346. https://doi.org/10.1016/j.ympev.2005.04.010

Li, H. (2011). A statistical framework for SNP calling, mutation discovery, association mapping and population genetical parameter estimation from sequencing data. Bioinformatics, 27, 2987–2993. https://doi.org/10.1093/bioinformatics/btr509

Lindner, R., & Friedel, C. C. (2012). A comprehensive evaluation of alignment algorithms in the context of RNA-Seq. Plos One, 7,e52403. https://doi.org/10.1371/journal.pone.0052403

Liu, G., Li, C., Du, Y., & Liu X. (2017). The complete mitochondrial genome of Japanese sparrowhawk (Accipiter gularis) and the phylogenetic relationships among some predatory birds. Biochemical Systematics and Ecology, 70,116–125. https://doi.org/10.1016/j.bse.2016.11.007

Machado, D. J., Janies, D., Brouwer, C., & Grant T. (2018). A new strategy to infer circularity applied to four new complete frog mitogenomes. Ecology and Evolution, 8, 4011–4018. https://doi.org/10.1002/ece3.3918

Machado, D. J., Lyra, M. L., & Grant, T. (2016). Mitogenome assembly from genomic multiplex libraries: comparison of strategies and novel mitogenomes for five species of frogs. Molecular Ecology Resources, 16, 686–693. https://doi.org/10.1111/1755-0998.12492

McIntyre, C. L., & Lewis, S. B. (2016). Observations of migrating golden eagles (Aquila chrysaetos) in eastern interior Alaska offer insights on population size and migration monitoring. The Journal of Raptor Research, 50, 254–264. https://doi.org/10.3356/jrr-15-13.1 

Mindell, D. P., Sorenson, M. D., & Dimcheff, D. E. (1998). An extra nucleotide is not translated in mitochondrial ND3 of some birds and turtles. Molecular Biology and Evolution, 15, 1568–1571. https://doi.org/10.1093/oxfordjournals.molbev.a025884

Moritz, C. (1994). Applications of mitochondrial DNA analysis in conservation: a critical review. Molecular Ecology, 3,
401–411. https://doi.org/10.1111/j.1365-294X.1994.tb00080.x

Nguyen, L. T., Schmidt, H. A., Von Haeseler, A., & Minh, B. Q. (2015). IQ-TREE: a fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies. Molecular Biology and Evolution, 32, 268–274. https://doi.org/10.1093/molbev/msu300

Okonechnikov, K., Golosova, O., Fursov, M., Varlamov, A., Vaskin, Y., Efremov, I. et al. (2012). Unipro UGENE: a unified bioinformatics toolkit. Bioinformatics, 28, 1166–1167. https://doi.org/10.1093/bioinformatics/bts091

Poessel, S. A., Woodbridge, B., Smith, B. W., Murphy, R. K., Bedrosian, B. E., Bell, D. A. et al. (2022). Interpreting long-distance movements of non-migratory golden eagles: Prospecting and nomadism? Ecosphere, 13,e4072. https://doi.org/10.1002/ecs2.4072 

Posada, D. (2008). jModelTest: phylogenetic model averaging. Molecular Biology and Evolution, 25,1253–1256. https://doi.org/10.1093/molbev/msn083

Rambaut, A. (2010). FigTree v1.3.1. Institute of Evolutionary Biology, University of Edinburgh, Edinburgh. http://tree.bio.ed.ac.uk/software/figtree/

Rodríguez-Estrella, R. (2002). A survey of golden eagles in Northern Mexico in 1984 and recent records in Central and Southern Baja California Peninsula. Journal of Raptor Research, 36, 3–9. https://digitalcommons.usf.edu/jrr/vol36/iss5/3

Rodríguez-Estrella, R., De León-Girón, G., Nocedal, J., Chapa, L., Scott, L., Eccardi, F. et al. (2020). Informe del Programa de monitoreo del águila real en México. Ciudad de México: Comisión Nacional para el Conocimiento y Uso de la Biodiversidad/Centro de Investigaciones Biológicas del Noroeste (CIBNOR). 

Rodríguez-Estrella, R., Llinas-Gutiérrez, J., & Cancino, J. (1991). New Golden Eagle records from Baja California. Journal Raptor Research, 25, 68–71. https://digitalcommons.usf.edu/jrr/vol25/iss3/3

Ronquist, F., Teslenko, M., Van Der Mark, P., Ayres, D. L., Darling, A., Höhna, S. et al. (2012). Mrbayes 3.2: efficient bayesian phylogenetic inference and model choice across a large model space. Systematic Biology, 61, 539–542. https://doi.org/10.1093/sysbio/sys029

Rozas, J., Ferrer-Mata, A., Sánchez-Del Barrio, J. C., Guirao-Rico, S., Librado, P., Ramos-Onsins, S. E. et al. (2017). DnaSP 6: DNA sequence polymorphism analysis of large data sets. Molecular Biology and Evolution, 34, 3299–3302. https://doi.org/10.1093/molbev/msx248

Sangster, G., & Luksenburg, J. A. (2021). Sharp increase of problematic mitogenomes of birds: causes, consequences, and remedies. Genome Biology and Evolution, 13, 1–14. https://doi.org/10.1093/gbe/evab210

Sieber, P., Platzer, M., & Schuster, S. (2018). The definition of open reading frame revisited. Trends in Genetics, 34, 167–170. https://doi.org/10.1016/j.tig.2017.12.015

Slack, K. E., Janke, A., Penny, D., & Arnason, U. (2003). Two new avian mitochondrial genomes (penguin and goose) and a summary of bird and reptile mitogenomic features. Gene, 302, 43–52. https://doi.org/10.1016/S0378111902010533 

Song, X., Huang, J., Yan, C., Xu, G., Zhang, X., & Yue, B. (2015). The complete mitochondrial genome of Accipiter virgatus and evolutionary history of the pseudo-control regions in Falconiformes. Biochemical Systematics and Ecology, 58, 75–84. https://doi.org/10.1016/j.bse.2014.10.013

Thankaswamy-Kosalai, S., Sen, P., & Nookaew, I. (2017). Evaluation and assessment of read-mapping by multiple next-generation sequencing aligners based on genome-wide characteristics. Genomics, 109, 186–191. https://doi.org/10.1016/j.ygeno.2017.03.001

Tracey, J. A., Madden, M. C., Sebes, J. B., Bloom, P. H., Katzner, T. E., & Fisher, R. N. (2017). Biotelemetery data for Golden Eagles (Aquila chrysaetos) captured in coastal southern California, February 2016-February 2017. U.S. Geological Survey Data Series 1051. https://doi.org/10.3133/ds1051 

Watson, J., Riley, H., & Thomson, D. (2011). The Golden Eagle (2nd Ed.). New Haven, Connecticut: Yale University Press.

Zerbino, D. R., & Birney, E. (2008). Velvet: algorithms for de novo short read assembly using de Bruijn graphs. Genome Research, 18, 821–829. https://doi.org/10.1101/gr.074492.107

Omar A. Hernández-Dávila ^a, *, Vinicio Sosa ^b, Javier Laborde ^b

^a Instituto de Biotecnología y Ecología Aplicada, Av. de las Culturas Veracruzanas No. 101, Col. Emiliano Zapata, 91090 Xalapa, Veracruz, Mexico 

^b Instituto de Ecología, A.C., Red de Ecología Funcional, Carretera Antigua a Coatepec, No. 351, El Haya, 91073 Xalapa, Veracruz, Mexico

*Corresponding author: omar.hernandez.davila@outlook.com (O.A. Hernández-Dávila)

Abstract

Cloud forests are known for their remarkable biodiversity and provide many ecosystem services. However, this biodiversity is in jeopardy due to the conversion of forests to other land uses. At its northernmost range in the Neotropics, cloud forest persists in remnant fragments immersed in an agricultural matrix that still has arboreal elements, such as riparian corridors. In this study we characterize the avifauna present in cloud forest riparian corridors in a highly degraded landscape of Mexico. We classified the avifauna in terms of migratory and conservation status, trophic guild, body mass, forest stratum and habitat preference. In 14 riparian corridors we recorded 86 bird species (75% were resident). Insectivorous and frugivorous species represented 79% of total richness. Almost 65% of species prefer the mid-story or canopy forest strata, while 46% were habitat generalists. Despite crossing open agricultural areas, cloud forest riparian corridors still harbor a diverse assemblage of bird species, which includes not only those tolerant to disturbance, but also species that are typical of old-growth cloud forest. We suggest that these remnants may be crucial for forest birds that move across the fragmented landscape.

Keywords: Riparian strips; Bird community; Fragmented landscape; Reservoirs

Avifauna de corredores ribereños de bosque de niebla en un paisaje degradado del este de México

Resumen 

Los bosques de niebla son conocidos por su notable biodiversidad y sus servicios ecosistémicos. Sin embargo, esta biodiversidad está en peligro debido a su conversión a otros usos de suelo. En su distribución más septentrional en el Neotrópico, el bosque de niebla persiste en remanentes inmersos en una matriz agrícola que aún conserva elementos arbóreos, como los corredores riparios. En este estudio caracterizamos la avifauna presente en dichos corredores en un paisaje altamente degradado de México. Clasificamos a la avifauna en términos de su estatus migratorio y de conservación, gremio trófico, masa corporal y preferencia de hábitat y estrato forestal. En 14 corredores riparios registramos 86 especies de aves. Las especies insectívoras y frugívoras representaron 79% de la riqueza total. Casi 65% de las especies prefieren los estratos forestales medios o de dosel, mientras que 46% fueron generalistas de hábitat. Aunque los corredores riparios atraviesan zonas agrícolas, siguen albergando un conjunto diverso de aves, que incluye no solo aquellas tolerantes a las perturbaciones, sino también especies características del bosque de niebla conservado. Sugerimos que estos remanentes pueden ser cruciales para las aves forestales que se desplazan por el paisaje fragmentado.

Palabras clave: Franjas riparias; Comunidad de aves; Paisaje fragmentado; Reservorios

Introduction

Tropical montane cloud forest (hereafter, cloud forest) is one of the most important terrestrial ecosystems worldwide. It provides environmental services such as carbon sequestration and water capture, and also mitigates flooding and drought (Bruijnzeel et al., 2011). Cloud forest is among the most biodiverse ecosystems in the world and hosts a remarkable diversity of flora and fauna in which spatial variation is prominent (i.e., high beta diversity), as well as a high proportion of endemic species (Aldrich et al., 2000; Karger et al., 2021). Mexican cloud forests are particularly rich in species of trees, shrubs and epiphytes, along with amphibians, reptiles, birds and mammals (Gual-Díaz & Rendón-Correa, 2014). It is estimated that Mexican cloud forests are home to 551 bird species (i.e., 50% of the total richness of the avifauna in the country), and the cloud forests in the state of Veracruz are home to 346 bird species (Navarro-Sigüenza et al., 2014). The richness of cloud forest avifauna in other parts of the country ranges from 196 to 335 species per region (Hernández-Baños et al., 1995; Navarro-Sigüenza et al., 2014). Despite its relevance, Mexican cloud forest is currently in jeopardy, not only because of agricultural expansion and uncontrolled urban growth, but also due to the illegal extraction of its species and unsustainable use of its resources (Toledo-Aceves et al., 2011). In the central part of Veracruz in the 1990s, it was estimated that there were 426 km² of cloud forest, an area that was reduced to 279 km² by 2003 (Muñoz-Villers & López-Blanco, 2008). Even though the rate of Mexican cloud forest deforestation has slowed over the last decade, today it is estimated to cover only 175 km² in Veracruz (Bonilla-Moheno & Aide, 2020). Cloud forest remnants are currently found as numerous fragments of different sizes (1 to 30 ha, usually), which are immersed in an extensive agricultural matrix (Williams-Linera et al., 2002) that still contains distinct arboreal elements such as isolated trees, living fences and forested riparian corridors. The presence of these arboreal elements within the agricultural matrix, which also include small patches (< 5 ha) of secondary forest, could be relevant to the maintenance of different groups of native flora and fauna in anthropic landscapes (Toledo-Aceves et al., 2014).

Cloud forest riparian corridors are single rows of trees growing along each side of permanent streams or rivers that have been left uncut by farmers when converting the forest to crop fields or pastures. These narrow, elongated belts or strips of tall trees have a dense woody undergrowth (riparian corridors, hereafter) and extend along rivers for kilometers. They are usually the most conspicuous arboreal element in agricultural landscapes. Forested riparian corridors that cross agricultural plots provide several environmental services, such as riverbank stabilization, nutrient recycling and enrichment, and filtering and retention of agrochemical pollutants in surface runoff, among other services (Cole et al., 2020). Additionally, riparian corridors contribute to the conservation of several taxonomic groups in anthropic landscapes, including native woody plant species (Hernández-Dávila et al., 2020), amphibians (Rodríguez-Mendoza & Pineda, 2010), bats and non-flying mammals (Griscom et al., 2007; Zarazúa-Carbajal et al., 2017). For birds, forested riparian corridors have been shown to function as refuges, foraging and nesting sites, as well as habitat corridors or stepping stones for moving across fragmented landscapes (Domínguez-López & Ortega-Álvarez, 2014; Kontsiotis et al., 2019; Lees & Peres, 2008). In lowland tropical regions that have been converted to agriculture, relatively wide riparian corridors connected to large forest remnants have been found to harbor a higher richness and abundance of forest birds than narrower and unconnected riparian corridors (Arizmendi et al., 2008; Domínguez-López & Ortega-Álvarez, 2014; Pliscoff et al., 2020). The presence of linear arboreal elements within the agricultural matrix might help to maintain bird diversity in anthropic landscapes (de Zwaan et al., 2022). To date, the majority of studies on the avifauna that uses riparian corridors in anthropic or/and fragmented landscapes has been done in lowland areas originally covered by tropical rainforest or seasonally dry tropical forest (Graham et al., 2002; Latta et al., 2012; Villaseñor-Gómez, 2008). In particular, for riparian strips of tropical montane cloud forest, Hernández-Dávila et al. (2021) found that landscape composition (i.e., urban and forest area) and vegetation structure (mean height of vegetation) positively influence the richness and abundance of generalist and specialist birds using riparian strips. However, the bird community of cloud forest riparian habitats that could potentially be using these remnants in anthropic landscapes has not been characterized to date. In this sense, taking into account that the original area covered by cloud forest has been drastically reduced by human activities and that the remaining forest fragments are surrounded by an extensive agricultural matrix, in which still there are arboreal riparian strips, we wanted to determine and characterize the composition and structure of the bird community present in these remnants. The latter, in order to find out if these riparian strips can serve as biodiversity reservoirs or landscape connectors, that facilitate the movement of birds across the landscape and which bird species use them. The objectives of this paper were: 1) to determine the richness, diversity and composition of birds that use riparian corridors of cloud forest in a highly modified anthropic landscape, and 2) to characterize the avifauna that uses these corridors in terms of their conservation and migratory status, trophic guild, and habitat preference. To date, no characterization of the bird community present in riparian corridors of cloud forest has been carried out. Given that the cloud forest is fragmented and natural or semi-natural remnants, such as arboreal riparian corridors, can support different groups of flora and fauna, it is necessary to determine and analyze the avifauna present in these corridors. This knowledge is needed to design and implement management strategies of riparian corridors and improve the odds of native cloud forest species conservation in current landscapes.

Materials and methods

The study area is located in the central part of the state of Veracruz, Mexico in the upper basin of the La Antigua River within 19°31’59”-19°22’42” N, 97°05’36”-96°57’43” W (Fig. 1). The original vegetation was cloud forest (i.e., Tropical Montane Cloud Forest) with an average annual temperature of 18 °C and total annual precipitation from 1,500 to 2,000 mm/year. Fourteen riparian corridor sites (with elevations from 1,190 up to 1,780 m asl) were selected for bird sampling. The sites selected are near to the cities of Xalapa, Coatepec, and Xico and are part of a highly fragmented landscape and are representative of the riparian corridors in the region (i.e., narrow linear bands of remnant cloud forest 2 to 5 m wide growing on both sides of permanent rivers). Since riparian corridors can be several kilometers long, we delimited each of our 14 sampling sites as a riparian tract or segment approximately 400 m long (± 16 m, s.e.) with a continuous tree canopy (i.e., uninterrupted arboreal cover). The 14 riparian segments selected were all separated by more than 1 km. Most of the segments of riparian corridors that we selected for bird sampling cross open cattle pastures, with a few of them adjoining small patches (< 4 ha) of secondary old-growth forest or different types of crop fields (mostly maize or shaded coffee plantations). Dominant tree species in the sampling sites include Platanus mexicana, Liquidambar styraciflua, Palicourea padifolia, Styrax glabrescens, and Perrottetia longistylis. The canopy of these riparian corridors is formed by tall trees usually 15 to 20 m in height, with some surpassing 30 m, however, the average tree height in the sampling sites was 6.3 (± 7.0) m. See Hernández-Dávila et al. (2020) for more details on the vegetation structure and composition of the sites sampled. 

The richness and number of birds visiting each sampling site were recorded at 2 fixed point counts set along each segment at least 250 m apart (Gregory et al., 2004). Field observations were conducted in October 2017, and in January, April and July 2018, for a total of 4 visits per point over the course of 1 year, with a 75-day interval between visits to each point. For each point count, all bird sightings were recorded with binoculars over the course of 15 minutes and within 35 m distance by one of us between sunrise and 10:30 am, except on rainy days. It took 3 days to complete the 28 points of a given period. The starting point during each period was alternated to cover the whole morning schedule of observation at each point. Only birds that were perching on the woody vegetation or on the ground or riverbank were recorded. Those flying overhead or perching in open areas nearby outside the riparian corridor were not counted. Thus, a total of 112 counts (14 sites × 2 points/site × 4 visits/site) were done at 28 points, totaling 28 h of observation. In addition to the visual records, we also set mist nets to capture birds to record understory birds visiting some of the sampled corridors, but owing to time and monetary constraints, we were only able to place the nets in 6 of the 14 riparian corridor sites. These 6 sites were chosen randomly and in each of them a total of 15 nets (10 × 2.5 m, each) were set parallel to the river flow on both riversides and at least 50 m apart along the sampled segment. From August 2016 to July 2017, one of the 6 riparian corridors was selected each month for mist-netting. Nets were left in place for 3 consecutive days avoiding rainy days and opened twice a day from sunrise to 11:00 and from 16:00 to sunset. This was done at each site twice over the sampled year. Sampling effort was 1,743 h and 2,250 m² of nets (25 m²/net × 15 nets × 6 sites) in 2 sampling periods at each site. Nets were inspected by 2 people every 30 minutes or less, depending on capture intensity. Each captured bird was marked by trimming a notch at the tip of one of the tail feathers to recognize recaptured individuals. Birds were released in situ immediately after sexing, weighing, and measuring (tarsus, tail, and total length; wing chord length; beak length, width, and depth).

Bird species were identified using the Sibley (2000) and Howell and Webb (1995) field guides. Nomenclature follows the IOC World Bird List checklist (Gill et al., 2024). Recorded birds were classified as either migratory from North America or resident species, and we also noted whether they were endemic, following Navarro-Sigüenza et al. (2007). The conservation status of each species was based on Mexican federal law NOM-059 (Semarnat 2010), and The IUCN Red List (IUCN, 2024). For trophic guild, each species was classified as: carnivore, insectivore, frugivore, granivore, nectarivore and omnivore (González-Salazar et al., 2014). Additionally, species were categorized into 3 size categories (body mass): small (< 40 g), medium (40 – 100 g) and large birds (> 100 g). Forest stratum preferences were based on Martínez-Morales (2001) and classified as: canopy species (those associated with the upper forest stratum, > 10 m above the ground); midstory species (those that preferentially use forest stratum between 5 and 10 m above the ground); midstory and canopy species (those using both midstory and canopy strata); understory species (those that frequent the forest floor up to 5 m above the ground); understory and midstory species (those using both understory and midstory strata); and all strata species (those that use all forest strata). Finally, main habitat preference was also based on Martínez-Morales (2001) and noted as: forest interior species (those which prefer forest sites away from the forest edge); forest edge species (those that preferentially use the fragment edge less than 100 m away from open areas); forest generalist species (those that are common both at the edge and the interior of forest fragments); and vegetation matrix species (those associated with the agricultural matrix or open areas). For species not reported by Martínez-Morales (2001) we did not assign categories for forest stratum preference and habitat preference. For these cases, species were classified as unknown.

The total richness recorded in riparian corridors was determined by the sum of records obtained from point records and those from mist-nets. Each bird species was characterized in relation to its migratory and conservation status, size, trophic guild, and habitat and forest stratum preferences. Due to the differences in the number of riparian corridors sampled by each sampling method (i.e., 14 for count points and 6 for mist nets) and their differences in data obtained from each method, for all the remaining analyses, only data from count points were taken into account: species accumulation curve and sample coverage were estimated to assess the sample completeness. Hill numbers were calculated to analyze the diversity of bird species in terms of effective number of species (q0), effective number of abundant species or Shannon diversity (q1) and effective number of dominant species or Simpson diversity (q2) (Hsieh et al., 2016; Jost, 2006). The rank-abundance curve of overall recorded avifauna was drawn to show graphically the structure of the bird community (Kindt & Coe, 2005). Analyses of accumulation curves, Hill numbers and range-abundance curves were performed for all riparian strips combined and separately. Finally, similarity in species composition among the 14 sampling sites was estimated using the Jaccard index distance, which varies from 0 to 1. Zero indicates that no species are shared between compared sites, and 1 indicates that species composition is identical between the sites. All analyses were run in R software using the vegan (Oksanen et al., 2020) and BiodiversityR (Kindt, 2021) packages.

Results

We recorded a total of 86 bird species in the 14 riparian corridors. These 86 species belonged to 9 orders, 27 families and 67 genera (Table 1). The richest families were Parulidae (16 species); Tyrannidae (13 spp.); Trochilidae (9 spp.) and Turdidae (6 spp.). Of the 86 bird species, there were 67 recorded at the point-counts and 48 trapped in the nets, with 29 recorded with both field methods (Table 1). 

Of the 86 recorded species, 65 were resident species and 21 were migratory. Only 2 of the recorded species are endemic to Mexico: Melanotis caerulescens (Blue Mockingbird) and Cardellina rubra (Red Warbler). Four species are under special protection status by Mexican federal law (Semarnat, 2010): Accipiter striatus (Sharp-shinned Hawk), Cinclus mexicanus (American Dipper), Psarocolius montezuma (Montezuma Oropendola) and Catharus mexicanus (Black-headed Nightingale-thrush), and 1 species is threatened: Catharus frantzii (Ruddy-capped Nightingale-thrush). One species, Selasphorus rufus (Rufous Hummingbird), is regarded as Near Threatened in the IUCN list. The richest trophic guilds were insectivorous (63% of species) and frugivorous birds (15%) (Fig. 2A). Regarding forest stratum preference, 18% of the species recorded prefer the midstory to canopy strata, and 17% are midstory specialists (Fig. 2B). For habitat preference, habitat generalists were the most strongly represented accounting for 30% of richness, followed by forest-interior species with 18% (Fig. 2C). Most recorded species (66%) were relatively small with a body mass < 40 g, and only 12% of species were heavier than 100 g (Fig. 2D).

During the 28 point counts we recorded a total of 816 detections of 67 bird species using the corridors. In the mist nets placed in 6 of the corridors, we captured 273 birds belonging to 48 species. The species accumulation curve for all corridors reached a sample completeness of 97%. According to Hill numbers the bird diversity was of 67 species observed (q0), 18 common species (q1) and 8 dominant species (q2) (Fig. 3). Individually, the sample completeness for each of the 14 riparian segments varied between 79% and 93% per segment; and regarding diversity, q0 ranged from 9 to 24 bird species per segment, while q1 ranged from 5 to 13 species and q2 from 2 to 13 species (Fig. 4).

Table 1

List of the 86 bird species recorded in 14 cloud forest riparian corridors in the anthropic landscape of central Veracruz, Mexico. Shown are the number of visual detections (# detections) of each species estimated by point counts and the number of birds
captured in mist nets. Seed dispersing birds are indicated with an asterisk (*). Species are ordered by rank (only for data obtained by point counts). We also show for each species: migratory status, resident (R), North American migrant (M). Protection status: threatened (A), special protection (Pr), least concern (LC), near threatened (NT). Trophic guild: insectivore (In), frugivore (Fr), nectarivore (Ne), omnivore (Om), granivore (Gr), carnivore (Ca). Size (body mass in g): small < 40 g (1), medium-sized 40 – 100 g (2), large > 100 g (3). Habitat preference: forest interior (FI), forest edge (FE), forest generalist (FG), vegetation matrix (VM), and no information available (-). Forest stratum preference: understory (U), understory and midstory (UM), midstory (M), midstory and canopy (MC), canopy (C), and no information available (-). See below for bibliographic sources.

*Seed dispersing birds: based on field data from Hernández-Dávila et al. (2022) and Hernández-Ladrón De Guevara et al. (2012). Migratory status: from Howell and Webb (1995) and Sibley (2000). Size: from Martínez-Morales (2001), Sibley (2000), and birds captured in mist nets (this study). Habitat preference and stratum preference: from Martínez-Morales (2001).

The most common species recorded visually was Chlorospingus flavopectus (Common Chlorospingus) with 234 detections (Fig. 5A), followed by Cardellina pusilla (Wilson’s Warbler) with 102, Psilorhinus morio (Brown Jay) with 79 and Myadestes occidentalis (Brown-backed Solitaire) with 51 detections (see species detection data in Table 1). These 4 dominant species accounted for 45% of total detections recorded in the point counts. There were 10 species with only 2 detections (i.e., doubletons) and 18 species with only 1 (singletons), and these 28 extremely rare species accounted for less than 5% of total detections and 42% of the 67 species recorded in the point counts. In general, the pattern of dominance by the 4 species mentioned occurred in each riparian corridor sampled, concentrating most of the bird detections at each site, with several species having much fewer detections per site (Fig. 5B).

Bird species richness per site estimated in point counts varied from 9 (TM site) to 33 species (MA site), with 10 of the 14 segments sampled having fewer than 20 species each. Similarity between sites was low (Jaccard index, J < 0.4) for most of the paired comparisons (Table 2), with the highest level of similarity between the TL and AB sites (0.53) and the lowest between MA and AF (0.10).

Discussion

In general, and pooling all sampled riparian corridors, the sample completeness was high (97%), recording a total richness of 86 bird species in the 14 segments sampled in the fragmented cloud forest landscape of central Veracruz, Mexico. This richness is comparable to that reported in similar studies carried out in relatively large (> 3 ha) remnant fragments of cloud forest in the same region, where up to 75-100 bird species have been detected (Rueda-Hernández et al., 2015; Serna-Lagunes et al., 2023). These 14 riparian corridors harbor 24% of the bird richness reported for cloud forest throughout the entire state of Veracruz (Navarro-Sigüenza et al., 2014). The richness recorded in our study represents between 25 and 43% of the avifauna reported in other regions of Mexico with cloud forest, where 196 to 335 bird species have been found (Martínez-Morales, 2007; Navarro-Sigüenza et al., 2014). The richness detected suggests that riparian corridors are important elements in deforested landscapes of cloud forest for numerous birds. Other studies in different sites have shown that these elements of the landscape can serve as refuges, foraging areas and even as reproductive (nesting) sites (Hawes et al., 2008; de Zwaan et al., 2022), as well as making it possible for birds to move across large open areas in agricultural landscapes (Gillies & St. Clair, 2010; Pliscoff et al., 2020). Both resident species and migratory birds, visit and use forested riparian corridors during their autumn-winter stay in the tropics (Skagen et al., 1998; Villaseñor-Gómez, 2008). As shown in our results: 24% of the recorded species were North American migratory species. The migratory bird Cardellina pusilla, abundant in cloud forest as well as in shaded coffee plantations in Veracruz (Navarro-Sigüenza et al., 2014), was the second most common species in our study. By far, the most dominant species in our study was Chlorospingus flavopectus, a resident bird regarded as a forest generalist that is common to old-growth forest, forest edge habitats and patches of secondary forest (Cruz-Angón et al., 2008; Martínez-Morales, 2007; Renner et al., 2006). Myadestes occidentalis was the fourth most dominant species in riparian corridors, which is considered a typical cloud forest species (Caballero-Cruz et al., 2020). This pattern of dominance is similar to that recorded by Martínez-Morales (2001) who reported C. flavopectus, M. occidentalis, Henicorhina leucophrys, Catharus mexicanus, and Trogon mexicanus as the most dominant species in conserved fragments of cloud forest. Except for T. mexicanus, the mentioned species were recorded in this study. Another very common bird in our study was Psilorhinus morio, a habitat generalist associated with disturbed areas, and common in small forest fragments of cloud forest (Serna-Lagunes et al., 2023), in rainforest riparian corridors that cross pastures (Graham et al., 2002), as well as in open agricultural areas with scant arboreal cover (Cerezo et al., 2009). It is important to mention that P. morio is a large species (> 100 gr), only 12% of the species recorded belong to this size category, while 65% are small species (< 40 gr). The conversion of forest for agricultural purposes mainly affects the presence of large bird species due to the reduction in food availability as well as fewer nesting and roosting sites (Gomes et al., 2008; Martínez-Morales, 2001). Thus, large bird species are the most strongly affected by the reduction of forest cover, while small birds are more vagile and tolerant to deforestation, explaining the predominance of species smaller than 40 g in the riparian corridors studied.

Table 2

Similarity distance in bird species composition (Jaccard index) among the 14 riparian corridors sampled (upper-right side of Table) with point counts (see Materials and methods), showing the number of species shared between riparian corridors (lower-left side), and the total number of species in each corridor (diagonal black cells). Gray-shaded cells highlight the highest and lowest values of similarity distance. The names of riparian corridors sampled are abbreviated as in Figure 1.

Insectivorous birds were the richest and most abundant trophic guild in the riparian corridors sampled and are also the most common guild in intact cloud forest (Martínez-Morales, 2007). Frugivorous birds were the second richest guild and were relatively abundant in our riparian corridors. These species, together with omnivorous species, are particularly important in forest regeneration due to their role as seed dispersers of forest plants. Some of the most important frugivores that are efficient dispersers of cloud forest trees, shrubs and other zoochorous plants include M. occidentalis, C. flavopectus, P. morio, and several species of the genera Turdus, Catharus, and Euphonia (Hernández-Dávila et al., 2022; Hernández-Ladrón De Guevara et al., 2012), all of which were recorded in our study. Seed dispersal by birds is crucial for cloud forest restoration since most plant species native to this forest depend on vertebrate frugivores for dispersal (Jordano et al., 2011). Forest frugivores usually avoid open areas that are devoid of perching sites, and this is one of the strongest limitations to forest restoration due to the limited or absent immigration of woody plant seeds into agricultural areas (Holl et al., 2000). However, it has been recorded that the density of linear forest or wooded patches such as riparian strips and live fences can increase bird diversity in agricultural landscapes (Wilson et al., 2017). In this sense cloud forest riparian corridors could contribute to species movement across the landscape. Thus, forested riparian corridors within the agricultural matrix facilitate that frugivorous birds will visit these disturbed sites and disperse seeds across and into the site. Another group of birds that is crucial for plant reproduction are the nectarivorous species, with several species of hummingbirds being particularly important. Of the 26 hummingbird species reported for Mexican cloud forest (Navarro-Sigüenza et al., 2014), 9 were recorded in the riparian corridors that we studied. The presence of birds that are seed or pollen vectors in riparian corridors contributes greatly to connectivity in anthropic landscapes and are essential to biodiversity conservation and forest regeneration and restoration in these landscapes.

As expected by the high degree of anthropic disturbance in the landscape we studied, the richest groups of birds in the cloud forest riparian corridors were forest generalist species and those associated with the vegetation of the agricultural matrix (sensu Martínez-Morales, 2001), which together represented 53% of the avifauna we recorded. The presence and wide distribution of different arboreal elements within the current anthropic landscape could explain the presence of forest interior bird species in riparian corridors. Changes in the structure and floristic composition of the original vegetation of a given site, resulting from human activities also lead to changes in the community attributes of the avifauna (Martínez-Morales, 2005). Among the most salient changes in the forested riparian corridors that cross agricultural areas is the high abundance of plants that are common in large canopy gaps or associated with disturbed sites, including several species of the families Piperaceae, Melastomataceae, and Rubiaceae, which were common in our study (Hernández-Dávila et al., 2020). These vegetation changes are more favorable to habitat generalists than to bird species associated with the interior of large forest fragments, explaining the lower proportion of species whose preferred habitat is the forest interior in our results. Studies in different regions report an increase in the richness and abundance of habitat generalist birds in forest edge habitats, where forest interior birds decrease (de Zwaan et al., 2022; Watson et al., 2004; Wilson et al., 2017).

For cloud forest, Martínez-Morales (2005) found that fragment size positively affects the richness and abundance of both generalist and forest-interior birds. In addition, for riparian corridors of cloud forest, Hernández-Davila et al. (2021) found that the richness and abundance of generalist and specialist birds showed a differential response to the amount of forest and urban cover in the vicinity of riparian strips. The percentage of urban cover near the riparian strip negatively affected the abundance of forest interior species and positively affected generalist species, whereas surprisingly the amount of forest area nearby the strip does not seem to influence the richness and abundance of birds using the riparian strip. These results could explain the differences between the number of generalist and interior species found in our study, as well as the differences of the Hill diversity values among the sampled corridors. Although this study did not analyze explicitly aspects of landscape configuration, it is relevant to mention that, although the riparian corridors are narrow remnants of just a few meters wide (< 10 m), both generalist and forest interior species were recorded in them. This suggests that these remnants may harbor a wealth of bird species regardless of their habitat preference. In fact, 2 of the bird species recorded are endemic to Mexico, another 4 are protected by law and 1 is threatened. This highlights the importance of riparian corridors as reservoirs of birds within anthropic landscapes, particularly species native to the cloud forest including resident and migratory birds. It is important to note that 3 of the species recorded in riparian corridors; Quiscalus mexicanus, Rupornis magnirostris, and Pyrocephalus rubinus could be regarded as urban birds (Maya-Elizarrarás, 2011; Ruelas & Aguilar, 2010).

As far as we know, this study is the first to describe and characterize the avifauna present in riparian corridors of the threatened cloud forest, however, it is important to take into account that, despite the fact that our point counts had a sufficient separation and elapsed time between observations to warrant independent detections and no overflying individuals or auditory recordings were included, we might have overestimated species abundances, in particular because birds move frequently along the forest strips (personal observation, OHD). Also, because mist nets capture understory species more frequently, species richness of mid- to high- strata birds are usually underestimated with nets. This study focused on riparian corridors and did not include other types of natural remnants or conserved forest fragments present in the region and part of the current mosaic of the anthropic fragmented landscape. Therefore, more studies are needed to determine the importance of riparian corridors for conserving bird diversity in comparison with other natural remnants of cloud forest. 

In conclusion, our results show that the cloud forest riparian corridors that cross open agricultural areas harbor a diverse assemblage of bird species, which includes not only those tolerant of or associated with disturbance, but also species that are typical of intact patches of old-growth cloud forest. These would be absent in areas that are completely devoid of trees. Additionally, several of the birds recorded in our study are effective seed dispersers of cloud forest plants. For current deforested landscapes, this strongly suggests that cloud forest riparian corridors could be key elements in forest restoration efforts and for the conservation of bird biodiversity in agricultural landscapes. Landscape management plans designed to encourage the permanence and sustainable management of these forested corridors within the agricultural matrix will not only help in the conservation of the avifauna, but also in the restoration of degraded landscapes, thanks to the ecosystem services provided by birds, including the pollination and seed dispersal of plants native to the cloud forest.

Acknowledgement

We are grateful to María de los Ángeles García and Diana Vázquez for their valuable help in the field. The Instituto de Ecología, A.C. and Idea Wild provided the space and equipment that made this study possible. Bianca Delfosse translated the text from the original in Spanish and edited subsequent versions of the manuscript. We also thank two anonymous reviewers for their suggestions to improve the manuscript. This work was supported by The Rufford Foundation (grant number 20471-1 to OHD) and the Consejo Nacional de Ciencia y Tecnología (grant numbers CONACYT-CB-2016-01 to VJS, and graduate scholarship CONACYT-285962 to OHD).

References

Aldrich, M., Bubb, P., Hostettler, S., & van de Wiel, H. (2000). Tropical montane cloud forests: time for action. WWF International/IUCN The World Conservation Union.

Arizmendi, M. D. C., Dávila, P., Estrada, A., Figueroa, E., Márquez-Valdelamar, L., Lira, R. et al. (2008). Riparian Mesquite bushes are important for bird conservation in tropical arid Mexico. Journal of Arid Environments, 72, 1146–1163. https://doi.org/https://doi.org/10.1016/j.jaridenv.2007.12.017 

Bonilla-Moheno, M., & Aide, T. M. (2020). Beyond deforestation: Land cover transitions in Mexico. Agricultural Systems, 178, 102734. https://doi.org/https://doi.org/10.1016/j.agsy.2019.102734 

Bruijnzeel, L. A., Scatena, F. N., & Hamilton, L. S. (2011). Tropical montane cloud forests. Science for conservation and management (International Hydrology Series). Cambridge University Press.

Caballero-Cruz, P., Vargas-Noguez, G., & Ortiz-Pulido, R. (2020). Especies de aves en riesgo en el bosque mesófilo de montaña en cinco AICA de la sierra Madre Oriental, México. Huitzil, Revista Mexicana de Ornitología, 21, e-490. https://doi.org/https://doi.org/10.28947/hrmo.2020.21.1.472 

Cerezo, A., Robbins, C. S., & Dowell, B. (2009). Uso de hábitats modificados por aves dependientes de bosque tropical en la región caribeña de Guatemala. Revista de Biología Tropical, 57, 401–419. https://doi.org/https://doi.org/10.15517/rbt.v57i1-2.11355 

Cole, L. J., Stockan, J., & Helliwell, R. (2020). Managing riparian buffer strips to optimise ecosystem services: A review. Agriculture, Ecosystems and Environment, 296, 106891. https://doi.org/https://doi.org/10.1016/j.agee.2020.106891 

Cruz-Angón, A., Scott, S. T., & Greenberg, R. (2008). An experimental study of habitat selection by birds in a coffee plantation. Ecology, 89, 921–927. https://doi.org/https://doi.org/10.1890/07-0164.1 

de Zwaan, D. R., Alavi, N., Mitchell, G. W., Lapen, D. R., Duffe, J., & Wilson, S. (2022). Balancing conservation priorities for grassland and forest specialist bird communities in agriculturally dominated landscapes. Biological Conser-
vation, 265, 109402. https://doi.org/10.1016/j.biocon.2021.109402 

Domínguez-López, M. E., & Ortega-Álvarez, R. (2014). The importance of riparian habitats for avian communities in a highly human-modified Neotropical landscape. Revista Mexicana de Biodiversidad, 85, 1217–1227. https://doi.org/10.7550/rmb.43849 

Gill, F., Donsker, D., & Rasmussen, P. (2024). IOC World Bird List (v14.1). https://www.worldbirdnames.org/new/ 

Gillies, C. S., & St. Clair, C. C. (2010). Functional responses in habitat selection by tropical birds moving through fragmented forest. Journal of Applied Ecology, 47, 182–190. https://doi.org/10.1111/j.1365-2664.2009.01756.x 

Gomes, G. L. L., Oostra, V., Nijman, V., Cleef, A. M., & Kappelle, M. (2008). Tolerance of frugivorous birds to habitat disturbance in a tropical cloud forest. Biological Conservation, 141, 860–871. https://doi.org/10.1016/j.biocon.2008.01.007 

González-Salazar, C., Martínez-Meyer, E., & López-Santiago, G. (2014). A hierarchical classification of trophic guilds for North American birds and mammals. Revista Mexicana de Biodiversidad, 85, 931–941. https://doi.org/10.7550/rmb.38023 

Graham, C., Martínez-Leyva, J. E., & Cruz-Paredes, L. (2002). Use of fruiting trees by birds in continuous forest and riparian forest remnants in Los Tuxtlas, Veracruz, Mexico. Biotropica, 34, 589–597. https://doi.org/10.1111/j.1744-7429.2002.tb00578.x 

Gregory, R. D., Gibbons, D. W., & Donald, P. F. (2004). Bird census and survey techniques. In W. J. Sutherland, I. Newton, and R. Green (Eds.), Bird ecology and conservation: a handbook of techniques (pp. 17–56). Oxford: Oxford University Press.

Griscom, H. P., Kalko, E. K., & Ashton, M. S. (2007). Frugivory by small vertebrates within a deforested, dry tropical region of Central America. Biotropica, 39, 278–282. https://doi.org/10.1111/j.1744-7429.2006.00242.x 

Gual-Díaz, M., & Rendón-Correa, A. (2014). Bosques mesófilos de montaña de México: diversidad, ecología y manejo. México D.F.: Comisión Nacional para el Conocimiento y Uso de la Biodiversidad.

Hawes, J., Barlow, J., Gardner, T. A., & Peres, C. A. (2008). The value of forest strips for understorey birds in an Amazonian plantation landscape. Biological Conservation, 141, 2262–2278. https://doi.org/10.1016/j.biocon.2008.06.017 

Hernández-Baños, B. E., Peterson, A. T., Navarro-Sigüenza, A. G., & Escalante-Pliego, P. B. (1995). Bird faunas of the humid montane forests of Mesoamerica: biogeographic patterns and priorities for conservation. Bird Conservation International, 5, 251–277. https://doi.org/10.1017/S0959270900001039 

Hernández-Dávila, O. A., Laborde, J., Sosa, V. J., & Díaz-Castelazo, C. (2022). Interaction network between frugivorous birds and zoochorous plants in cloud forest riparian strips immersed in anthropic landscapes. Avian Research, 13, 100046. https://doi.org/10.1016/j.avrs.2022.100046 

Hernández-Dávila, O. A., Laborde, J., Sosa, V. J., Gallardo-Hernández, C., & Díaz-Castelazo, C. (2020). Forested riparian belts as reservoirs of plant species in fragmented landscapes of tropical mountain cloud forest. Botanical Sciences, 98, 288–304. https://doi.org/https://doi.org/10.17129/botsci.2497 

Hernández-Dávila, O. A., Sosa, V. J., & Laborde, J. (2021). Effects of landscape context and vegetation attributes on understorey bird communities of cloud forest riparian belts. Ecological Engineering, 167, 106269. https://doi.org/10.1016/j.ecoleng.2021.106269 

Hernández-Ladrón De Guevara, I., Rojas-Soto, O. R., López-Barrera, F., Puebla-Olivares, F., & Díaz-Castelazo, C. (2012). Dispersión de semillas por aves en un paisaje de bosque mesófilo en el centro de Veracruz, México: Su papel en la restauración pasiva. Revista Chilena de Historia Natural, 85, 89–100. https://doi.org/10.4067/S0716-078X2012000100007 

Holl, K. D., Loik, M. E., Lin, E. H. V., & Samuels, I. A. (2000). Tropical montane forest restoration in Costa Rica: overcoming barriers to dispersal and establishment. Restoration Ecology, 8, 339–349. https://doi.org/https://doi.org/10.1046/j.1526-100x.2000.80049.x 

Howell, N. G., & Webb, S. (1995). A guide to the birds of Mexico and Northern Central America. Oxford: Oxford University Press.

Hsieh, T. C., Ma, K. H., & Chao, A. (2016). iNEXT: an R package for rarefaction and extrapolation of species diversity (Hill numbers). Methods in Ecology and Evolution, 7, 1451–1456. https://doi.org/10.1111/2041-210X.12613 

IUCN (08 de junio de 2024). The IUCN Red List of Threatened Species. Version 2024-1. https://www.iucnredlist.org 

Jordano, P., Pierre-Michel, F., Lambert, J. E., Böhning-Gaese, K., Traveset, A., & Wright, S. J. (2011). Frugivores and seed dispersal: mechanisms and consequences for biodiversity of a key ecological interaction. Biology Letters, 7, 321–323. https://doi.org/https://doi.org/10.1098/rsbl.2010.0986 

Jost, L. (2006). Entropy and diversity. Oikos, 113, 363–375. https://doi.org/https://doi.org/10.1111/j.2006.0030-1299.14714.x 

Karger, D. N., Kessler, M., Lehnert, M., & Jetz, W. (2021). Limited protection and ongoing loss of tropical cloud forest biodiversity and ecosystems worldwide. Nature Ecology and Evolution, 5, 854–862. https://doi.org/10.1038/s41559-021-01450-y 

Kindt, R. (2021). BiodiversityR. Package for Community Ecology and Suitability Analysis (2.16-1). http://www.worldagroforestry.org/output/tree-diversity-analysis 

Kindt, R., & Coe, R. (2005). Tree diversity analysis. A manual and software for common statistical methods for ecological and biodivserity studies. Nairobi: World Agroforestry Centre (ICRAF).

Kontsiotis, V., Zaimes, G. N., Tsiftsis, S., Kiourtziadis, P., & Bakaloudis, D. (2019). Assessing the influence of riparian vegetation structure on bird communities in agricultural Mediterranean landscapes. Agroforestry Systems, 93, 675–687. https://doi.org/10.1007/s10457-017-0162-x 

Latta, S. C., De La Cueva, H., & Harper, A. B. (2012). Abundance and site fidelity of migratory birds wintering in riparian habitat of Baja California. Western Birds, 43, 90–101.

Lees, A. C., & Peres, C. A. (2008). Conservation value of remnant riparian forest corridors of varying quality for amazonian birds and mammals. Conservation Biology, 22, 439–449. https://doi.org/10.1111/j.1523-1739.2007.00870.x 

Martínez-Morales, M. A. (2001). Forest fragmentation effects on bird communities of tropical montane cloud forests in eastern Mexico (Ph.D. Thesis). Cambridge UK: University of Cambridge.

Martínez-Morales, M. A. (2005). Landscape patterns influencing bird assemblages in a fragmented Neotropical cloud forest. Biological Conservation, 121, 117–126. https://doi.org/10.1016/j.biocon.2004.04.015 

Martínez-Morales, M. A. (2007). Avifauna del bosque mesófilo de montaña del noreste de Hidalgo, México. Revista Mexicana de Biodiversidad, 78, 149–162. https://doi.org/https://revista.ib.unam.mx/index.php/bio/article/view/395 

Maya-Elizarrarás, E. (2011). Aves explotadoras de áreas verdes urbanas: un ejemplo de la zona metropolitana de Guadalajara, Jalisco. El Canto Del Cenzontle, 2, 104–109.

Muñoz-Villers, L. E., & López-Blanco, J. (2008). Land use/cover changes using Landsat TM/ETM images in a tropical and biodiverse mountainous area of central-eastern Mexico. International Journal of Remote Sensing, 29, 71–93. https://doi.org/10.1080/01431160701280967 

Navarro-Sigüenza, A. G., Lira-Noriega, A., Peterson, A. T., Oliveras de Ita, A., & Gordillo-Martínez, A. (2007). Diversidad, endemismo y conservación de las aves. In I. Luna, J. J. Morrone, & D. Espinosa (Eds.), Biodiversidad de la Faja Volcánica Transmexicana (pp. 461–483). México D.F.: Universidad Nacional Autónoma de México.

Navarro-Sigüenza, A. G., Gómez de Silva, H., Gual-Díaz, M., Sánchez-Gonzales, L. A., & Pérez-Villafaña, M. (2014). La importancia de las aves del bosque mesófilo de montaña de México In M. Gual-Díaz, & A. Rendón-Correa (Eds.), Bosques mesófilos de montaña de México: diversidad, ecología y manejo (pp. 279–299). Conabio.

Oksanen, J., Guillaume, B. F., Kindt, R., Legendre, P., McGlinn, D., Minchin, P. R. et al. (2020). Community Ecology Package (2.5-7). http://mirror.bjtu.edu.cn/cran/web/packages/vegan/vegan.pdf 

Pliscoff, P., Simonetti, J. A., Grez, A. A., Vergara, P. M., & Barahona-Segovia, R. M. (2020). Defining corridors for movement of multiple species in a forest-plantation landscape. Global Ecology and Conservation, 23, e011108. https://doi.org/10.1016/j.gecco.2020.e01108 

Renner, S. C., Walterrt, M., & Mühllenberg, M. (2006). Comparison of bird communities in primary vs. young secondary tropical montane cloud forest in Guatemala. Biodiversity and Conservation, 15, 1545–1575. https://doi.org/10.1007/s10531-005-2930-6 

Rodríguez-Mendoza, C., & Pineda, E. (2010). Importance of riparian remnants for frog species diversity in a highly fragmented rainforest. Conservation Biology, 6, 781–784. https://doi.org/10.1098/rsbl.2010.0334 

Rueda-Hernández, R., MacGregor-Fors, I., & Renton, K. (2015). Shifts in resident bird communities associated with cloud forset patch size in Central Veracruz, Mexico. Avian Conservation and Ecology, 10. https://doi.org/http://dx.doi.org/10.5751/ACE-00751-100202 

Ruelas, I. E., & Aguilar, R. S. H. (2010). La avifauna urbana del parque ecológico Macuiltépetl en Xalapa. Ornitología Neotropical, 21, 87–103.

Semarnat (Secretaría del Medio Ambiente y Recursos Naturales). (2010). Norma Oficial Mexicana NOM-059-SEMARNAT-2010, Protección ambiental – Especies nativas de México de flora y fauna silvestres – Categorías de riesgo y especificaciones para su inclusión, exclusión o cambio – Lista de especies en riesgo. Diario Oficial de la Federación. 30 de diciembre de 2010, Segunda Sección, México.

Serna-Lagunes, R., Torres-Cantú, G. B., & García-Martínez, M. Á. (2023). Avifauna en fragmentos de bosque mesófilo de montaña y vegetación secundaria en el municipio de Huatusco, Veracruz. CienciaUAT, 1, 06–24. https://doi.org/https://doi.org/10.29059/cienciauat.v18i1.1669 

Sibley, D. A. (2000). The Sibley guide to birds. New York: Alfred A. Knopf.

Skagen, S. K., Melcher, C. P., Howe, W. H., & Knopf, F. L. (1998). Comparative use of riparian corridors and oases by migrating birds in southeast Arizona. Conservation Biology, 12, 896–909. https://doi.org/https://doi.org/10.1111/j.1523-1739.1998.96384.x 

Toledo-Aceves, T., García-Franco, J., Williams-Linera, G., Mac-Millan, K., & Gallardo-Hernández, C. (2014). Significance of remnant cloud forest fragments as reservoirs of tree and epiphytic bromeliad diversity. Tropical Conservation Science, 7, 230–243. https://doi.org/10.1177/194008291400700205 

Toledo-Aceves, T., Meave, J., González-Espinosa, M., & Ramírez-Marcial, N. (2011). Tropical montane cloud forests: Current threats and opportunities for their conservation and sustainable management in Mexico. Journal of Environmental Management, 92, 974–981. https://doi.org/10.1016/j.jenvman.2010.11.007 

Villaseñor-Gómez, J. F. (2008). Habitat use of wintering bird communities in Sonora Mexico: the importance of riparian habitats. Studies in Avian Biology, 37, 53–68.

Watson, J. E. M., Whittaker, R. J., & Dawson, T. P. (2004). Habitat structure and proximity to forest edge affect the abundance and distribution of forest-dependent birds in tropical coastal forests of southeastern Madagascar. Biological Conservation, 120, 311–327. https://doi.org/10.1016/j.biocon.2004.03.004 

Williams-Linera, G., Manson, R. H., & Isunza, V. E. (2002). La fragmentación del bosque mesófilo de montaña y patrones de uso del suelo en la región oeste de Xalapa, Veracruz, México. Madera y Bosques, 8, 69–85. https://doi.org/https://doi.org/10.21829/myb.2002.811307 

Wilson, S., Mitchell, G. W., Pasher, J., McGovern, M., Hudson, M. A. R., & Fahrig, L. (2017). Influence of crop type, heterogeneity and woody structure on avian biodiversity in agricultural landscapes. Ecological Indicators, 83, 218–226. https://doi.org/10.1016/j.ecolind.2017.07.059 

Zarazúa-Carbajal, M., Avila-Cabadilla, L. D., Álvarez-Añorve, M. Y., Benítez-Malvido, J., & Stoner, K. E. (2017). Importance of riparian habitat for frugivorous bats in a tropical dry forest in western Mexico. Journal of Tropical Ecology, 33, 74–82. https://doi.org/10.1017/S0266467416000572