Identification and Diversity Analysis of Avian Remains in Tianjin Urban Area Inferring from DNA Barcoding

Abstract

(1) Background: The acceleration of urbanization poses an increasingly serious threat to avian diversity. Consequently, accurate species identification of avian remains is essential for biodiversity monitoring, bird rescue operations, and conservation management. (2) Methods: This study employed DNA barcoding technology based on the cytochrome c oxidase subunit I (COI) gene to analyze 112 avian remains samples collected from urban and peri-urban areas of Tianjin city. (3) Results: A total of 47 bird species were identified, belonging to 11 orders, 24 families, and 30 genera, achieving an overall identification success rate of 95.54%. Passeriformes were dominant, accounting for 70.21% of the identified species. The species list includes 3 species listed as Class I nationally protected birds and 7 species as Class II nationally protected birds. Discrepancies between preliminary morphological identification and molecular identification results highlighted the complementary roles of the two approaches. (4) Conclusions: This study demonstrates that DNA barcoding is an effective tool for efficiently identifying degraded avian remains in urban environments. It provides reliable data for biodiversity assessments, wildlife rescue, and conservation management, while also supporting improved identification accuracy through the integration of molecular and morphological methods.

1. Introduction

With the continuous advancement of urbanization, the structure and function of urban ecosystems are undergoing profound transformations. In urban ecosystems, wildlife constitutes a vital resource for human survival and development, playing a key role in maintaining ecosystem security, stability, and ecological services [1,2]. Birds are among the most common wild animals in urban environments, characterized by their wide distribution and accessibility [3]. They play indispensable roles in maintaining ecological balance, facilitating seed dispersal, and managing pest. The community composition and population dynamics directly of avians reflect the ecological quality and biological carrying capacity of urban environments, substantially enhancing both the physical and mental health, as well as the overall welfare of urban dwellers [4].

Avian diversity, serving as a crucial indicator of ecosystem health and stability, faces mounting pressures from habitat fragmentation, environmental pollution, and human disturbances. Particularly in urban settings, incidents leading to bird mortality, such as collisions with buildings, illegal trade, and poisoning, occur frequently [5]. Without effective identification and documentation of these remains, gaps in biodiversity monitoring data will arise, subsequently compromising the scientific basis for conservation decisions and forensic investigations [6]. Against the backdrop of increasing integration between ecological conservation and judicial practice, there is a growing demand for the scientific and accurate identification of wildlife remains. This need extends to critical areas such as species protection, determination of legal liability, ecological and environmental damage compensation assessments, and compliance with international conventions [7,8].

Tianjin (38°34′–40°15′ N, 116°43′–118°04′ E), located in the northeastern part of the North China Plain and bordering the Bohai Sea to the east, is one of the four municipalities directly under the central government of China. The city covers an area of approximately 11,917 km², with plains accounting for about 94% of its territory and low mountains in the north reaching an elevation of 1078 m at Jiushanding. These diverse habitats, ranging from mountains to coastal wetlands, support abundant bird populations, particularly along the East Asian–Australasian Flyway.

Tianjin encompasses diverse habitats ranging from mountains to coastal wetlands that support abundant bird populations. Over the past two decades, avian diversity in this region has been extensively documented. Within Tianjin alone, 251 species have been documented in urban areas, with species richness being higher in rural and wild landscapes, underscoring the suppressive effect of urbanization on local avian diversity [9].

Focusing specifically on Tianjin, recent studies have identified 228 protected bird species as conservation targets, revealing a synergistic relationship between avian biodiversity and ecosystem services [10]. However, protected areas cover only 37.7% of the overlapping biodiversity and ecosystem service hotspots, highlighting critical conservation gaps. To address these gaps, an ecological network comprising 20 ecological sources and 31 corridors has been proposed [10]. Research on wetland waterbirds further emphasizes the ecological importance of Tianjin’s wetlands, which cover approximately 21.05% of the city’s total area, with key habitats spanning 2695 km² [11].

Despite these contributions, existing studies have predominantly focused on natural reserves and coastal wetlands, with limited attention to avian diversity within urban built-up areas such as parks, green spaces, and building perimeters. Methodologically, traditional morphological identification and field observation have been the primary approaches, whereas the application of DNA barcoding for identifying degraded avian remains has not yet been systematically explored in Tianjin. Traditional methods for monitoring avian diversity primarily rely on direct field observations, vocalization recognition, or feather morphology identification [12]. Nevertheless, these approaches often prove inadequate for deceased, fragmented, or morphologically indistinct individuals. In recent years, DNA barcoding technology has emerged as a powerful tool that employs sufficiently variable, easily amplified, and relatively short standardized DNA fragments, which specificity within species and diversity between species, for species identification [13,14]. Consequently, this method maintains high reliability even when samples are incomplete, degraded, or consist of minimal tissue, allowing for the simultaneous processing of multiple specimens [15]. Previous studies, which encompassed an extensive analysis of 13,320 species across 11 animal phyla has demonstrated that a segment of the mitochondrial cytochrome c oxidase subunit I (COI) gene near the 5′ end enables effective species identification within the animal kingdom, while also revealing significant sequence variation in the COI gene across the majority of animal groups [16]. This gene exhibits characteristics such as maternal inheritance, a relatively fast evolutionary rate, and a high nucleotide substitution rate, leading to its widespread application in wildlife forensics, biodiversity monitoring, and conservation biology research [17].

With the continuous improvement of the COI gene database, DNA barcoding has been applied to the identification of remains from a variety of organisms. In the fields of conservation biology and forensic science, accurate species identification forms a critical foundation for implementing effective conservation management and law enforcement supervision. However, when confronted with materials lacking morphological features or highly degraded specimens—such as wildlife products, trade residues, or environmental samples—traditional morphological identification methods often face significant limitations. In such cases, mitochondrial DNA barcoding provides a key tool for the precise identification of wildlife and endangered plants [18,19,20]. By amplifying short and conserved gene fragments, this technique can effectively overcome the challenges posed by incomplete morphological information or severe degradation of samples, making it particularly suitable for identifying materials such as hair, scales, bone fragments, wood remnants, and powdered products.

In law enforcement and regulatory practice, this technology has been widely applied in tracing the origin of illegally traded products, identifying components in mixed herbal medicines, and determining the species of decomposed biological samples. For instance, it has supplied reliable molecular evidence in cases involving illegal trade of animal skins [21,22] and identification of ivory remains [23]. Moreover, a study based on the COI gene involving over 23,000 avian remains [24] further demonstrates its practical value in forensic identification. Markedly, research indicates that building collisions rank as the second leading direct human-caused source of bird mortality in the United States, following predation by both wild and free-ranging domestic cats [25], with bird–window collisions being a major contributing factor among such human-related causes of death [26]. DNA barcoding has played a significant role in the rapid and accurate identification of avian remains in such contexts, providing standardized molecular evidence that supports efforts against poaching and smuggling, and enhances the scientific rigor and effectiveness of species conservation. For instance, a study successfully identified three species of Ascidiacea from northeastern China utilizing COI sequences, while another research effort determined that bird remains from an aircraft belonged to Hirundo rustica using COI and cytochrome b (Cytb) genes [5,27]. Thus, the COI sequence is the earliest proposed and most extensively used barcode marker, widely employed for precise species identification [28].

This study, focusing on the Tianjin urban area and its surrounding regions, collected avian remains from locations such as parks, green spaces, and building collision sites, and employed DNA barcoding technology for species identification, to elucidate their taxonomic status and analyze the species composition, faunal characteristics, and threat status of the urban avian community. This study employed DNA barcoding technology based on the COI gene to conduct molecular identification of 112 avian remains collected from urban Tianjin and its surrounding areas. A total of 47 bird species were successfully identified, belonging to 30 genera, 24 families, and 11 orders, achieving an overall identification success rate of 95.54%. Among these, three species are classified as national first-class protected species and seven as second-class protected species. The study also revealed discrepancies and complementary relationships between morphological and molecular identification results.

By unveiling the current state of avian diversity and primary risk factors within the urban environment, this research provides fundamental data for local avian resource inventorying, endangered species protection, and facilitating ecological restoration efforts, while simultaneously contributing valuable genetic information to advance studies in species biology and systematics. On a practical level, the identification results can aid in tracing illegal hunting and trade activities, offering technical support for wildlife conservation law enforcement and the refinement of relevant regulations. Moreover, this study provides scientific reference and practical support for enhancing public ecological awareness, reconciling urban development with biodiversity conservation, and promoting harmonious coexistence between humanity and the natural world.

2. Materials and Methods

2.1. Sample Collection

This study conducted avian sample collection within Tianjin urban area and its surrounding regions in 2025, accumulating a total of 112 avian remains samples. These samples encompassed various tissue types, including blood, muscle, and feathers. The collection methods and sources of the samples are summarized in

Table 1. Summary of sample collection sites, causes of death, and sources.

Collection Method/Source	Number of Samples	Description
Collected by research team	87	Samples collected from parks, building perimeters, and relevant management institutions; causes of death included glass collisions, natural mortality, and wildlife rescue cases.
Provided by Public Security Bureau	25	Samples obtained from bird rescue operations conducted by the Public Security Bureau.
Total	112

.

All collected samples, accompanied by clear background information, were obtained and utilized in strict compliance with ethical review approval and applicable regulations, providing a solid foundation for this study.

Among the collected samples, 87 were severely damaged in their external morphological features, rendering species identification based solely on morphology impossible. The remaining 25 samples underwent preliminary identification based on morphological characteristics, corresponding to 25 different bird species. Among the 112 samples collected, 48 were blood samples, 35 were muscle samples, and the remaining 29 were feather samples. All samples were stored at −20 °C for subsequent experimental analysis.

2.2. DNA Extraction and PCR Amplification

An appropriate amount of tissue sample was used to extract genomic DNA with the QIAGEN DNeasy Blood & Tissue Kit (Tiangen Biochemical Technology Co., Ltd., Beijing, China), strictly following the manufacturer’s instructions, for subsequent PCR use. PCR primers for amplifying the COI sequence (Fa1COFa, BirdR2) were referenced from Hebert et al. [29]:

Fa1COFa: 5′-TCAACAAACCACAAAGACATCGGCAC-3′;

BirdR2: 5′-ACTACATGTGAGATGATTCCGAATCCAG-3′.

Primers were synthesized by Sangon Biotech (Shanghai ) Co., Ltd, Shanghai, China. The PCR reaction volume was 50 μL, containing 25 μL of TaqTM HS Perfect Mix, 2 μL each of Fa1COFa and BirdR2 primers, 2 μL of DNA template, and 19 μL of ultrapure sterile water.

The PCR program commenced with an initial denaturation step at 94 °C for 1 min to ensure complete denaturation of the template DNA into single strands. This was followed by the first-stage amplification, consisting of 5 cycles. Each cycle comprised denaturation at 94 °C for 1 min, annealing at 45 °C for 1.5 min, and extension at 72 °C for 1.5 min. This stage employed a lower annealing temperature to facilitate effective primer binding during the initial cycles. Subsequently, the reaction proceeded to the second-stage amplification for 35 cycles. In this stage, the annealing temperature was increased to 52 °C, with each cycle including 94 °C for 1 min, 52 °C for 1.5 min, and 72 °C for 1 min. The higher annealing temperature promoted stringent and specific primer-template pairing, significantly reducing non-specific amplification. After the cycling steps, a final extension was performed at 72 °C for 5 min to ensure the integrity of the newly synthesized DNA strands. Upon completion, PCR products were stored at 4 °C for subsequent use.

PCR amplification products were detected via 1% TAE agarose gel electrophoresis, using Marker 2000 (Silca, Treviso, Italy) as the reference marker. The PCR products were then sent to Beijing Genomics Institute Co., Ltd., Beijing, China for bidirectional sequencing using the Sanger method [30].

2.3. Data Comparison and Analysis

Sequences in FASTA format were subjected to local alignment analysis using the Basic Local Alignment Search Tool (BLAST, version 4.0) provided by the National Center for Biotechnology Information (NCBI) database and BOLD Identification Engine provided by BOLD Systems v4. Upon sequence submission, the system automatically returned alignment results. For further screening, sequences with a length and sequence identity greater than 99% were selected for subsequent phylogenetic analysis. Concurrently, COI sequences of Dromaius novaehollandiae (accession number: HQ910428.1) and Struthio camelus (accession number: MG930899.1) were downloaded from NCBI to serve as outgroups for sequence comparison [31,32]. Utilizing MEGA 11 software [33], a phylogenetic tree was constructed using the Neighbor-Joining (NJ) method, with the reliability of the phylogenetic relationships for each branch was assessed through Bootstrap analysis with 1000 replicates, ensuring the scientific rigor and robustness of the findings [34].

3. Results

3.1. Taxonomic Identification Results

Among the total of 112 samples obtained in this study, 25 samples were preliminarily identified as corresponding to 25 valid bird species. This initial identification was based on a comprehensive assessment integrating collection location, time, and specific morphological characteristics. The preliminary identification results are presented in

Table 2. Summary of Preliminary Identification Results for the 25 Avian Remains.

	Order	Family	Species	Common Names (English)
G1	Anseriformes	Anatidae	Anas formosa	Baikal Teal
G2	Anseriformes	Anatidae	Anas zonorhyncha	Chinese Spot-billed Duck
G3	Anseriformes	Anatidae	Anas platyrhynchos	Mallard
G4	Gruiformes	Gruidae	Grus vipio	White-naped Crane
G5	Anseriformes	Anatidae	Anser cygnoides	Swan Goose
G6	Otidiformes	Otididae	Otis tarda	Great Bustard
G7	Ciconiiformes	Ciconiidae	Ciconia boyciana	Oriental Stork
G8	Passeriformes	Muscicapidae	Luscinia megarhynchos	Common Nightingale
G9	Charadriiformes	Scolopacidae	Scolopax rusticola	Eurasian Woodcock
G10	Ciconiiformes	Ardeidae	Botaurus stellaris	Eurasian Bittern
G11	Charadriiformes	Recurvirostridae	Recurvirostra avosetta	Pied Avocet
G12	Passeriformes	Passeridae	Carpodacus erythrinus	Common Rosefinch
G13	Galliformes	Phasianidae	Pavo cristatus	Indian Peafowl
G14	Accipitriformes	Accipitridae	Accipiter gentiis	Northern Goshawk
G15	Coraciiformes	Alcedinidae	Halcyon pileata	Black-capped Kingfisher
G16	Passeriformes	Passeridae	Passer montanus	Eurasian Tree Sparrow
G17	Passeriformes	Emberizidae	Emberiza pusilla	Little Bunting
G18	Passeriformes	Emberizidae	Emberiza spodocephala	Black-faced Bunting
G19	Passeriformes	Sylviinae	Acrocephalus orientalis	Reed Warbler
G20	Passeriformes	Emberizidae	Emberiza rutila	Chestnut Bunting
G21	Passeriformes	Sylviinae	Phylloscopus fuscatus	Dusky Warbler
G22	Passeriformes	Muscicapidae	Rhyacornis fuliginosus	Plumbeous Water-redstart
G23	Passeriformes	Emberizidae	Emberiza yessoensis	Japanese Reed Bunting
G24	Passeriformes	Zosteropidae	Zosterops erythropleurus	Chestnut-flanked White-eye
G25	Passeriformes	Muscicapidae	Luscinia calliope	Siberian Rubythroat

.

3.2. Molecular Sequencing and Identification Results

DNA was successfully extracted from all 112 avian remains samples. Among the resulting DNA samples, the COI gene was successfully amplified and sequenced from 107 samples, while amplification failed in two samples and three samples exhibited double peaks in their sequencing profiles. Using the online NCBI BLAST tool and the BOLD Identification Engine, sequences with similarity greater than 99% were selected for species assignment, resulting in the identification of 47 bird species. The overall identification success rate was 95.54%. Success rates varied by tissue type, with blood samples showing 97.92% success, muscle samples 88.57%, and feather samples 100%. A total of 47 bird species were successfully identified using DNA barcoding. The molecular identification results, together with the local residency status of each species, are summarized in

Table 3. Summary of Molecular Identification Results for the 47 Avian Remains.

Order	Family	Species	Common Names (English)	Residency and Status in Tianjin
Anseriformes	Anatidae	Anas formosa	Baikal Teal	Migratory (uncommon passage migrant)
Anseriformes	Anatidae	Anas platyrhynchos	Mallard	Breeding resident (common)
Apodiformes	Hirundinidae	Apus apus	Common Swift	Breeding resident (common)
Charadriiformes	Recurvirostridae	Recurvirostra avosetta	Pied Avocet	Breeding resident (locally common)
Charadriiformes	Scolopacidae	Gallinago stenura	Pin-tailed Snipe	Migratory (common passage migrant)
Ciconiiformes	Ardeidae	Botaurus stellaris	Eurasian Bittern	Breeding resident (rare)
Ciconiiformes	Ciconiidae	Ciconia boyciana	Oriental Stork	Migratory (rare passage migrant)
Columbiformes	Columbidae	Streptopelia orientalis	Oriental Turtle Dove	Breeding resident (common)
Coraciiformes	Alcedinidae	Halcyon pileata	Black-capped Kingfisher	Breeding resident (uncommon)
Cuculiformes	Cuculidae	Eudynamys scolopaceus	Asian Koel	Vagrant (extremely rare)
Galliformes	Phasianidae	Pavo cristatus	Indian Peafowl	Introduced (locally established)
Gruiformes	Gruidae	Grus vipio	White-naped Crane	Migratory (rare passage migrant)
Gruiformes	Rallidae	Gallinula chloropus	Common Moorhen	Breeding resident (common)
Otidiformes	Otididae	Otis tarda	Great Bustard	Winter visitor (rare)
Passeriformes	Acrocephalidae	Acrocephalus aedon	Thick-billed Warbler	Migratory (rare passage migrant)
Passeriformes	Alaudidae	Alauda arvensis	Eurasian Skylark	Breeding resident (uncommon)
Passeriformes	Emberizidae	Emberiza pusilla	Little Bunting	Winter visitor (common)
Passeriformes	Emberizidae	Emberiza rutila	Chestnut Bunting	Migratory (uncommon passage migrant)
Passeriformes	Emberizidae	Emberiza spodocephala	Black-faced Bunting	Migratory (common passage migrant)
Passeriformes	Emberizidae	Emberiza yessoensis	Japanese Reed Bunting	Winter visitor (rare)
Passeriformes	Hirundinidae	Cecropis daurica	Red-rumped Swallow	Breeding resident (common)
Passeriformes	Hirundinidae	Hirundo rustica	Barn Swallow	Breeding resident (abundant)
Passeriformes	Laniidae	Lanius cristatus	Brown Shrike	Breeding resident (common)
Passeriformes	Locustellidae	Locustella lanceolata	Lanceolated Warbler	Migratory (uncommon passage migrant)
Passeriformes	Motacillidae	Anthus richardi	Richard’s Pipit	Winter visitor (uncommon)
Passeriformes	Motacillidae	Motacilla flava	Western Yellow Wagtail	Migratory (common passage migrant)
Passeriformes	Motacillidae	Motacilla tschutschensis	Eastern Yellow Wagtail	Migratory (common passage migrant)
Passeriformes	Muscicapidae	Ficedula albicilla	Taiga Flycatcher	Migratory (uncommon passage migrant)
Passeriformes	Muscicapidae	Ficedula parva	Red-breasted Flycatcher	Migratory (rare passage migrant)
Passeriformes	Muscicapidae	Luscinia calliope	Siberian Rubythroat	Migratory (uncommon passage migrant)
Passeriformes	Muscicapidae	Luscinia cyane	Siberian Blue Robin	Migratory (rare passage migrant)
Passeriformes	Muscicapidae	Luscinia megarhynchos	Common Nightingale	Migratory (uncommon passage migrant)
Passeriformes	Muscicapidae	Luscinia sibilans	Rufous-tailed Robin	Migratory (rare passage migrant)
Passeriformes	Muscicapidae	Luscinia svecica	Bluethroat	Winter visitor (uncommon)
Passeriformes	Muscicapidae	Phylloscopus tenellipes	Pale-legged Leaf Warbler	Migratory (uncommon passage migrant)
Passeriformes	Muscicapidae	Tarsiger cyanurus	Red-flanked Bluetail	Winter visitor (uncommon)
Passeriformes	Passeridae	Carpodacus erythrinus	Common Rosefinch	Migratory (uncommon passage migrant)
Passeriformes	Passeridae	Passer montanus	Eurasian Tree Sparrow	Breeding resident (abundant)
Passeriformes	Sylviinae	Acrocephalus bistrigiceps	Black-browed Reed Warbler	Migratory (common passage migrant)
Passeriformes	Sylviinae	Acrocephalus orientalis	Reed Warbler	Breeding resident (common)
Passeriformes	Sylviinae	Locustella certhiola	Pallas’s Grasshopper Warbler	Migratory (uncommon passage migrant)
Passeriformes	Sylviinae	Phylloscopus borealis	Arctic Warbler	Migratory (common passage migrant)
Passeriformes	Sylviinae	Phylloscopus fuscatus	Dusky Warbler	Winter visitor (uncommon)
Passeriformes	Sylviinae	Phylloscopus inornatus	Yellow-browed Warbler	Winter visitor (common)
Passeriformes	Sylviinae	Phylloscopus schwarzi	Radde’s Warbler	Migratory (uncommon passage migrant)
Passeriformes	Turdidae	Turdus cardis	Japanese Thrush	Migratory (rare passage migrant)
Passeriformes	Zosteropidae	Zosterops erythropleurus	Chestnut-flanked White-eye	Migratory (common passage migrant)

.

Table 3 lists the species identified through DNA barcoding, along with their English common names, scientific names, and residency status (including breeding residents, migrants, winter visitors, and vagrants) and relative abundance (rare, uncommon, common, abundant) in Tianjin. This information provides essential ecological context for interpreting the species composition derived from molecular identification.

The 107 COI sequences obtained from the 47 bird species were compiled and consolidated. Additionally, COI sequences of D. novaehollandiae and S. camelus were downloaded from NCBI to serve as the outgroup. An NJ phylogenetic tree was constructed using the total set of 109 sequences, with the results presented in Figure 1.

The topological analysis demonstrated that sequences corresponding to the same species consistently formed distinct, well-supported monophyletic clades, confirming the high resolution of the COI gene at the species level. Genetic distances were markedly lower among congeners than between different genera, and even subtle differentiations among closely related species, such as within the Hirundo rustica complex, were clearly resolved. These findings further validate the precision of COI-based species identification and the reliability of the phylogenetic reconstruction.

It is noteworthy that several discrepancies were observed between preliminary morphological identifications and molecular phylogenetic placements. Sample G2, initially identified as Anas zonorhyncha, clustered within the clade of A. platyrhynchos with high sequence similarity and a divergence of less than 2%. Similarly, sample G20, preliminarily assigned to Emberiza rutila, grouped with E. spodocephala, exhibiting high similarity and less than 1% sequence divergence. Likewise, sample G22, tentatively identified as Rhyacornis fuliginosus, clustered with Ficedula parva, also showing high sequence similarity and a divergence below 2%.

3.3. Analysis of Species Composition Based on Molecular Sequencing Results

The 47 bird species identified via molecular methods belong to 11 orders, 24 families, and 30 genera. Among them, Passeriformes comprised 33 species, accounting for 70.21% of the total species count. Furthermore, this order represented 12 families, constituting 50% of all families identified. Thus, Passeriformes dominated both in terms of family-level diversity and species richness, which aligns with general patterns of avian diversity distribution [35,36]. At the genus level, most genera were represented by only a single species, reflecting a high degree of genus-level differentiation. Of note, the genus Luscinia contained five species, the highest among all genera, followed by the genus Emberiza with four species. The detailed species composition is presented in Figure 2.

The conservation status of the 47 bird species identified in this survey exhibits a hierarchical distribution. According to China’s national wildlife protection classification system, the list includes three Class I nationally protected species, seven Class II nationally protected species, and 33 terrestrial wildlife species identified as having significant ecological, scientific, and social value (Three-Value Protected Species). Four species were not listed under any of these statutory protection categories. When assessed against the IUCN Red List threat categories, most species are classified as Least Concern (LC). Nonetheless, the assemblage also includes threatened species, with 4 classified as Vulnerable (VU); 3 as Near Threatened (NT); and 1 as Endangered (EN).

A striking divergence between national and global assessments is observed for five species—Luscinia calliope, Luscinia megarhynchos, Luscinia svecica, Turdus cardis, and Zosterops erythropleurus—which are listed as Class II protected species in China but are globally assessed as LC on the IUCN Red List. This discrepancy highlights the differences between regional conservation policies and global species risk assessments. The detailed conservation profiles of all species are summarized in

Table 4. Conservation Status of the Identified Bird Species.

Species	Endangered Status	Protection Level
Acrocephalus aedon	Three-Value Protected Species	LC
Acrocephalus bistrigiceps	Three-Value Protected Species	LC
Acrocephalus orientalis	Three-Value Protected Species	LC
Alauda arvensis	II	NT
Anas formosa	II	VU
Anas platyrhynchos	Three-Value Protected Species	LC
Anthus richardi	Three-Value Protected Species	LC
Apus apus	——	LC
Botaurus stellaris	Three-Value Protected Species	NT
Carpodacus erythrinus	Three-Value Protected Species	LC
Cecropis daurica	——	LC
Ciconia boyciana	I	EN
Emberiza pusilla	Three-Value Protected Species	LC
Emberiza rutila	Three-Value Protected Species	LC
Emberiza spodocephala	Three-Value Protected Species	LC
Emberiza yessoensis	Three-Value Protected Species	NT
Eudynamys scolopaceus	Three-Value Protected Species	LC
Ficedula albicilla	Three-Value Protected Species	LC
Ficedula parva	Three-Value Protected Species	LC
Gallinago stenura	Three-Value Protected Species	LC
Gallinula chloropus	Three-Value Protected Species	LC
Grus vipio	I	VU
Halcyon pileata	Three-Value Protected Species	VU
Hirundo rustica	Three-Value Protected Species	LC
Lanius cristatus	Three-Value Protected Species	LC
Locustella certhiola	Three-Value Protected Species	LC
Locustella lanceolata	Three-Value Protected Species	LC
Luscinia calliope	II	LC
Luscinia cyane	Three-Value Protected Species	LC
Luscinia megarhynchos	II	LC
Luscinia sibilans	Three-Value Protected Species	LC
Luscinia svecica	II	LC
Motacilla flava	Three-Value Protected Species	LC
Motacilla tschutschensis	Three-Value Protected Species	LC
Otis tarda	I	VU
Passer montanus	Three-Value Protected Species	LC
Pavo cristatus	——	LC
Phylloscopus borealis	Three-Value Protected Species	LC
Phylloscopus fuscatus	Three-Value Protected Species	LC
Phylloscopus inornatus	Three-Value Protected Species	LC
Phylloscopus schwarzi	Three-Value Protected Species	LC
Phylloscopus tenellipes	——	LC
Recurvirostra avosetta	Three-Value Protected Species	LC
Streptopelia orientalis	Three-Value Protected Species	LC
Tarsiger cyanurus	Three-Value Protected Species	LC
Turdus cardis	II	LC
Zosterops erythropleurus	II	LC

Note: “I” refers to the national first-class protected wildlife species in China. “II” refers to the national second-class protected wildlife species in China.

.

4. Discussion

4.1. Applicability and Success Rate of DNA Barcoding for Identifying Avian Remains in Urban Environments

Utilizing DNA barcoding technology based on the COI gene, this study conducted molecular identification on 112 avian remains collected from Tianjin’s urban area and its surroundings, achieving an overall identification success rate of 95.54%. Remarkably, feather samples achieved a 100% success rate, strongly underscoring the substantial advantage of DNA barcoding in accurately identifying avian remains that are incomplete, degraded, or lacking distinct morphological features [37]. Compared to traditional methods reliant on intact morphology, DNA barcoding requires only trace amounts of tissue for accurate identification, thereby greatly expanding the range of identifiable samples, which is particularly valuable for the highly damaged, decomposed, or feather-only remains commonly encountered in urban settings [38,39]. The variation in success rates across different tissue types observed in this study, 97.92% for blood, 88.57% for muscle, also reflects the impact of sample preservation status and DNA extraction quality on identification outcomes [40,41,42].

In addition, the successful identification of 47 bird species, spanning 11 orders and 24 families, demonstrates the high resolution of the COI gene at the avian species level. Intraspecific genetic variation rarely exceeds 2% and is generally less than 1% [19,43], enabling effective discrimination of closely related species and accurately reflecting phylogenetic relationships among taxa. In total, DNA barcoding not only serves as a powerful tool for rapid species identification in forensic investigations and biodiversity monitoring, but it also provides reliable technical support for assessing the structure, composition, and conservation status of urban avian communities [8,44].

4.2. Complementarity and Conflict Resolution Between Morphological and Molecular Identification

In this research, 25 samples initially underwent morphological identification, followed by molecular identification for all samples. The findings revealed a high degree of concordance between most morphological identifications and their molecular results, confirming that traditional morphology remains valuable when samples are intact and features are distinct. Conversely, discrepancies were observed between preliminary morphological and final molecular identifications for several samples. For instance, sample G2 was initially identified as Anas zonorhyncha but matched A. platyrhynchos molecularly; sample G20 was initially identified as Emberiza rutila but matched E. spodocephala, among others. These discrepancies may stem from misjudgment due to damaged morphological features, morphological similarities caused by subspecies or geographic variation, or lagging updates in taxonomic systems.

Morphological and molecular identification are not mutually exclusive but are distinctly complementary. Morphological identification is rapid and intuitive, suitable for initial field screening and whole-specimen examination. In contrast, molecular identification offers objectivity, repeatability, and minimal influence from sample condition, proving particularly effective for identifying damaged remains, juveniles, females, or morphologically conservative groups [45]. In practical application, it is advisable to adopt a combined strategy that involves initial morphological screening followed by molecular verification. Furthermore, fostering the development of local integrated databases containing both morphological and molecular reference data for birds is crucial to reduce identification ambiguity and enhance the overall reliability of the identification system.

4.3. The Value of Avian Remains for Rescue and Conservation in Urban Environments

Of the 112 samples analyzed in this study, 25 avian specimens were provided by public security authorities during rescue and law enforcement operations, highlighting the critical role of species identification in wildlife rescue and conservation practices. DNA barcoding can provide species-level scientific evidence for avian conservation and rescue efforts, supporting the characterization and sentencing for illegal activities such as hunting, trade, and poisoning, and also furnishes a basis for species identification in ecological and environmental damage assessments [46]. In the context of animal rescue, precise species identification not only serves as a prerequisite for appropriate intervention measures but also provides essential baseline data for subsequent ecological monitoring, population assessment, and conservation management. Characterized by standardization, traceability, and verifiability, DNA barcoding constitutes a valuable technical tool for wildlife protection, demonstrating considerable potential for improving the efficiency of species identification and strengthening the scientific foundation of conservation strategies.

From a conservation management standpoint, the identification of remains in this study reveals the composition and conservation status of avian diversity within and around Tianjin’s urban area. Notably, among the 47 identified species, 3 are nationally Class I protected, 7 are Class II protected, and 8 are assessed as Vulnerable or higher on the IUCN Red List, highlighting the critical conservation needs in the area. These data offer a crucial addition to the existing inventories and dynamic monitoring efforts of local avian resources. They also indicate the presence of a notable number of threatened species within urban habitats, underscoring the need to strengthen the protection and restoration of key habitats such as urban green spaces and wetlands, as well as to reduce anthropogenic threats such as glass collisions and illegal hunting activities.

5. Conclusions

This study employed DNA barcoding based on the COI gene to conduct molecular identification and diversity analysis of 112 avian remains collected from Tianjin urban area and its surrounding regions. A total of 47 bird species belonging to 11 orders, 24 families, and 30 genera were successfully identified, yielding an overall identification success rate of 95.54%. Notably, feather samples achieved a 100% success rate, underscoring the robustness of DNA barcoding in identifying highly degraded or morphologically unidentifiable avian remains across different tissue types.

While the avian diversity of this region is already relatively well documented, the principal value of this study lies not in such descriptive accounts, but in demonstrating the effectiveness and reliability of DNA barcoding for species identification of avian remains under challenging urban conditions. The high success rate and the ability to resolve taxonomically challenging cases highlight the advantages of this molecular approach over traditional morphological methods, especially when samples are fragmented or lack diagnostic features. The discrepancies observed between morphological and molecular identification further emphasize the necessity of integrating both approaches to enhance identification accuracy in practical applications.

Overall, this study provides robust technical validation for the application of DNA barcoding in wildlife forensic investigations, biodiversity monitoring, and conservation management. It supplies essential molecular data for regional avian resource inventories and establishes a scientific foundation for rescue operations, species conservation, and urban ecological planning. Future efforts should focus on developing integrated local databases combining morphological and molecular reference data to further strengthen the comprehensiveness and utility of avian identification systems.