Effects of Fruiting Plants on Frugivorous Bird Diversity Across Different Disturbed Habitats

Abstract

Bird–plant interactions are critical for maintaining biodiversity and ecosystem function, and represent a key research focus in modern ecology. Using the line transect method, we surveyed bird diversity and collected plant trait data in four habitat types in the southern zone of Fujian’s Meihuashan National Nature Reserve during October–December 2021 and July–August 2022. This study investigated how plant traits (tree height, diameter at breast height (DBH), canopy density fruit amount) influence the diversity of frugivorous birds (species richness, abundance, Shannon–Wiener, Pielou, Simpson) across four disturbed habitats—villages (residential areas), bamboo forests (economic plantations), unguarded broad-leafed forests (wild forests), and nurtured broad-leafed forests (managed forests)—during both summer (breeding season) and autumn–winter (fruiting season). The key findings revealed that (1) significant correlations between plant traits and bird diversity were exclusive to the fruiting season, with no associations found in summer; (2) during autumn–winter, the key plant traits driving bird diversity varied distinctively by habitat: tree height and canopy density were paramount in villages; both habitat structure (canopy density) and fruit amount were important in bamboo forests, whereas in both broad-leafed forests, a combination of tree structure (height, DBH, canopy density) and fruit amount determined bird abundance; (3) a significant interaction between season and habitat was detected for community evenness, indicating that habitat type modulates the seasonal effects on community composition. This study underscores that in human-modified landscapes, conserving habitat structural complexity and key resource plants is crucial for sustaining frugivorous bird diversity and its ecological functions. Conservation strategies must account for seasonal dynamics to be effective.

1. Introduction

Patterns of species diversity and their underlying mechanisms remain fundamental research priorities and challenges in modern ecosystem studies [1,2,3]. Understanding the spatial distribution of species diversity and its driving forces continues to be a central focus in contemporary ecosystem research [4]. Avian diversity functions as a crucial indicator of ecosystem health, with its richness and rarity being influenced by multiple interacting environmental factors [5].

Current investigations primarily examine habitat variables (climate, precipitation) and vegetation types as influential factors [6]. Notable differences exist in avian diversity across various habitats [7], with variations in habitat structure resulting in distinct avian community compositions [8]. In forest ecosystems, fruiting plants directly influence the distribution patterns and diversity of fruit-eating birds through food resource provision (fruits) [7]. Research has established that forest habitats sustain higher avian diversity compared to more heavily disturbed village habitats, with distinct differences in species composition between these habitat types [9]. In highly disturbed habitats, most bird species prefer to forage and perch in tall trees [10,11]. In subtropical forest ecosystems, variations in plant traits explain a considerable proportion of avian diversity distribution patterns.

The morphological characteristics of zoochorous plants, including tree height, canopy density, and mean number of fruits, substantially influence avian diversity [12]. Recent investigations have demonstrated that plant morphological features (e.g., tree height, canopy structure) and resource traits (e.g., fruit amount) influence avian community assembly by modifying habitat heterogeneity and resource availability [13,14,15]. Zhang et al. [12] determined that canopy density and mean number of fruits of Taxus chinensis, a zoochorous plant, correlated significantly with frugivorous bird presence. Similarly, Jin [16] documented that tall zoochorous plants with dense canopies attracted a wider range of bird species compared to other plants.

Although numerous studies have been conducted in recent years, several significant limitations remain. First, habitat coverage is incomplete, with most research focusing on comparisons between forests and farmland or agroforestry systems [17,18], and monoculture bamboo stands, thus failing to fully capture how different types of disturbance affect plant–bird systems [19]. Second, research on seasonal dynamics is limited. Many studies concentrate on a single season, overlooking variations in fruit resource availability and bird foraging demand between summer and autumn. Autumn is a critical period for energy accumulation in preparation for winter, during which birds rely more heavily on fruit resources than in summer. However, comparative studies across seasons—particularly those quantifying the importance of autumn plant resources—remain scarce [20,21]. Third, analyses of plant traits tend to be narrow in scope. Many studies focus on isolated traits such as fruit size or tree height, without integrating key features including tree height, DBH, canopy density (indicating habitat structure), and fruit amount (reflecting food availability). This makes it difficult to clarify the combined effects and relative importance of these traits across disturbed habitats [22,23].

This study addresses these gaps through three key innovations, simultaneously incorporating four distinct habitat types—villages, unguarded broad-leafed forests, nurtured broad-leafed forests, and bamboo forests—with contrasting disturbance and vegetation characteristics. Drawing on multi-habitat comparison approaches [19], we dissect differences in plant–bird relationships across disturbance contexts, covering both summer and autumn seasons to investigate how seasonal dynamics influence the association between plant traits and bird diversity, informed by fruit phenology studies [20,24], while integrating structural traits (tree height, DBH, canopy density) and resource traits (fruit amount) to quantify the importance of plant characteristics in autumn—a key seasonal bottleneck—using multi-trait analysis methods [22], thereby identifying core influencing factors during critical periods.

This study was conducted in the southern region of Fujian Meihuashan National Nature Reserve, located at the junction between the southern slopes of Wuyi Mountains and Boping Ridge. The study addresses the following key questions: (1) Do relationships between bird diversity and plant traits differ between seasons? (2) Which plant traits best explain bird diversity in each habitat during the fruiting season? (3) Do season and habitat type interact to influence bird diversity metrics?

We predicted that bird–plant correlations would be stronger in the fruiting season due to resource tracking; habitat type would modulate the strength of plant–bird relationships, with more structured forests (e.g., broad-leaved forests) supporting stronger associations; and that nurtured habitats would show simplified interactions due to reduced structural and resource diversity.

2. Study Area

Fujian Meihuashan National Nature Reserve is located in southwestern Fujian Province (25°15′–25°35′ N, 116°45′–116°57′ E) (Figure 1), spanning the border of Shanghang County, Liancheng County, and Xinluo District. This subtropical climate zone exhibits distinct seasonal patterns: July–August represent the bird breeding-dominated summer, while October–December constitute the fruit resource-abundant autumn–winter period. The reserve extends approximately 20 km east to west and 19 km north to south, encompassing a total area of 22,168.5 hm². This investigation was conducted in the southern portion of the reserve, which features diverse habitat types, including evergreen broad-leaved forests, bamboo and broad-leaved mixed forests, bamboo forests, villages, and farmlands. The study focused on the following four representative habitat types: nurtured broad-leaved forest habitat, unguarded broad-leaved forest habitat, village habitat, and bamboo forest habitat. The nurtured broad-leaved forest represents a managed and cultivated forest, while the unguarded broad-leaved forest develops naturally without human intervention (

Table A1. Study site information table.

Habitat	Autumn and Winter	Summer	Number of Sampling Days	Transect Number	Starting Geographic Coordinates of the Transect	Altitude/m
Village	2021.10.08–2021.11.14	2022.07.15–2022.08.09	23	VL01	25°17′13.46″ N, 116°52′39.21″ E–25°17′16.15″ N, 116°52′51.56″ E	879
				VL02	25°16′31.38″ N, 116°53′50.5″ E–25°16′40.77″ N, 116°54′1.35″ E	864
				VL03	25°17′42.77″ N, 116°53′29.61″ E–25°17′46.7″ N, 116°53′23.47″ E	778
Unguarded broad-leaved forest	2021.10.06–2021.12.05	2022.07.14–2022.08.11	23	UL01	25°16′38.52″ N, 116°53′25.85″ E–25°16′47″ N, 116°52′24.3″ E	780
				UL02	25°16′57.22″ N, 116°52′37.64″ E–25°17′7.36″ N, 116°52′35.62″ E	780
				UL03	25°17′11.08″ N, 116°52′22.47″ E–25°17′16.48″ N, 116°52′18.58″ E	780
Nurtured broad-leaved forest	2021.10.07–2021.12.05	2022.07.20–2022.08.06	22	NL01	25°16′55.06″ N, 116°53′27.08″ E–25°16′50.77″ N, 116°53′39.35″ E	780
				NL02	25°16′56.58″ N, 116°53′29.79″ E–25°16′51.97″ N, 116°53′41.45″ E	780
				NL03	25°16′40.97″ N, 116°53′41.75″ E–25°16′52.54″ N, 116°53′53.49″ E	780
Bamboo forest	2021.10.09–2021.12.06	2022.07.22–2022.08.11	21	BL01	25°16′39.85″ N, 116°53′31.65″ E–25°16′36.65″ N, 116°53′44.86″ E	780
				BL02	25°17′31.74″ N, 116°53′22.62″ E–25°17′36.69″ N, 116°53′28.17″ E	779
				BL03	25°16′49.22″ N, 116°53′1.57″ E–25°16′44.28″ N, 116°53′5.36″ E	780

).

The nurtured broad-leaved forest habitat primarily comprises the Taxus chinensis Ecological Garden, situated in the southern section of Meihuashan National Nature Reserve. The area contains not only a high-density population of Taxus chinensis, but also other contemporary fruiting plants, such as Quercus glauca and Sloanea sinensis. While local conservation authorities have enhanced protection measures in recent years, tourism activities continue to present a source of human disturbance.

The unguarded broad-leaved forest habitat is located around the bamboo forest habitat, with a relatively scattered distribution, belonging to a distinct patchy habitat. The main fruit-bearing plants in this habitat include Taxus chinensis, Rhus chinensis, etc. Compared with the nurtured broad-leaved forest habitat, it has more plant species. This habitat has no artificial management and is subject to certain noise and road disturbances.

The village habitat encompasses the residential areas of Fudi Village, Jiazikeng Village, and Mafang Village, along with their surrounding farmland. This habitat features a complex landscape with numerous crops. Taxus chinensis predominates near houses and farmland edges, while other fruit-bearing plants consist mainly of vines or herbs, such as Callicarpa bodinieri and Phytolacca acinosa. This habitat exhibits lower vegetation canopy density and tree height compared to both broad-leaved forest habitats. The area demonstrates high land-use diversity due to crop cultivation and human activities.

The bamboo forest habitat is extensively distributed in the southern part of the nature reserve. Phyllostachys heterocycla dominates the vegetation in this habitat, accompanied by fruit-bearing plants such as Fissistigma oldhamii and Photinia serratifolia. Local residents manage and operate the Phyllostachys heterocycla stands, with bamboo harvesting activities resulting in substantial disturbances.

3. Methods

3.1. Survey Methods

Avian diversity data were collected as follows: The research site comprised 12 permanent line transects with individual lengths averaging 1 km (Table A1). Observers walking along transects recorded frugivorous bird species richness and abundance within 50 m on either side of the transect line. Surveys were conducted daily from 06:00 to 10:00 and 16:00 to 18:00. Surveys were conducted under clear, windless conditions, with each 8–10 day period constituting one survey cycle (the time required to complete all 12 transects sequentially). Bird diversity data for each season were derived from the sum of four survey cycles. Bird classification follows “The Checklist of Birds of China (3rd Edition)” [25].

Field investigations were conducted to survey various species of fruiting plants (woody fruiting plants) and measure their growth performance parameters. Surveys of fruiting plants were also conducted using the line transect method, with transect layout and survey timing identical to those used in bird surveys to ensure data comparability. During each transect walk, we thoroughly recorded the species, abundance, and geographical locations of all fruiting plants encountered. Additionally, we specifically observed and documented bird foraging behaviors on these plants. During each transect walk, we systematically recorded abundance and spatial distribution of fruiting plants, with particular emphasis on documenting avian frugivory events. When frugivorous birds were observed visiting and consuming fruits from fruiting plants, researchers waited until feeding activities concluded before establishing the focal plant as a sampling point. For each bird-visited fruiting plant, precise location data were recorded, followed by measurements of key plant traits including height, diameter at breast height (DBH), canopy density, fruit crop size, fruit color, and fruit dimensions. Canopy closure was calculated as the ratio of crown projection area to forest floor area, following Li Yongning’s method [26]. Tree height measurements followed Fang Jingyun’s methodology [27]. In this study, the fruit production of tree species within the transects was estimated using the visual estimation method, a widely adopted semi-quantitative approach that balances efficiency and reliability [12]. Within each study plot, typical sampling was conducted by observing all fruiting tree species during their fruit-bearing period. For each individual tree, the same trained observer systematically estimated the fruit crop using binoculars from multiple angles to minimize perspective-related errors. Fruits were assessed across different canopy layers based on standardized branch units (with uniform branch length and fruit load), enabling an extrapolation to the total number of fruiting branchlets across the whole plant.

To investigate potential differences in how plant traits influence bird diversity between seasons (resource-rich autumn–winter versus breeding-dominated summer), this study employed line-transect surveys during autumn–winter (October–December 2021) and summer (July–August 2022) to examine bird species richness and abundance, along with characteristics of bird-consumed plants, across four habitat types (nurtured broad-leafed forest, unguarded broad-leafed forest, village areas, and bamboo forest), with three parallel 1 km transects established in each habitat.

3.2. Analytical Methods

Shannon–Wiener diversity index (H′), Pielou evenness index (J), and Simpson dominance index (C) were used to evaluate avian community diversity between seasons, while Sorensen similarity index (S′) was employed to compare bird community similarities between seasons. The formulae for H′, J, C, and S′ was calculated following the method described by Ma Keping et al. [28].

$$H\prime =-{\displaystyle \sum _{i=1}^S{P}_i}\times \ln P_i$$

$$J=H\prime /\ln S$$

$$C=1-{{\displaystyle \sum _{i=1}^S\left(P_i\right)}}^2$$

S′ = 2c/(a + b)

where S represents the total species richness; P_i is the proportion of individuals belonging to species i; c is the number of species common to both communities; a and b are the number of species in communities A and B, respectively.

Pearson correlation analysis was performed using R version 4.1.2 to examine the relationship between resultant plants and frugivorous birds. The correlation heatmap was generated using OriginPro 2024.

To avoid multicollinearity among variables, predictors with a variance inflation factor (VIF) greater than 5 were considered to exhibit significant collinearity and were removed from the model. Due to strong collinearity between DBH and other variables in the village habitat (VIF = 11.42), it was excluded prior to model fitting. Generalized linear models (GLMs) were constructed using plant traits that showed significant correlations with avian species diversity. Poisson distributions were applied for species richness and abundance, while Gamma distributions were used for diversity indices. The best-fitting model was selected based on the lowest Akaike Information Criterion (AIC) value and a ΔAIC < 4. Model averaging was performed using the model avg function to generate averaged parameter estimates from the top model set. The averaged results reflect the relative importance of each explanatory variable. All procedures were conducted in R Studio. Model construction utilized the lme4 package [29], and model averaging was implemented with the MuMIn package [30].

A multifactorial ANOVA was conducted in IBM SPSS Statistics 27 to assess the interactive effects of season and habitat type on the diversity indices of frugivorous birds.

4. Results and Analysis

4.1. Avian and Fruiting Plants Composition

A total of 27 bird species were recorded consuming fruits, including insectivorous birds that also feed on fruits during the autumn and winter seasons (Table A1), with Hemixos castanonotus and Hypsipetes leucocephalus being the primary frugivorous species, followed by Chloropsis hardwickii, Urocissa erythroryncha, Ixos mcclellandii, Pycnonotus sinensis, Spizixos semitorques, and Pericrocotus solaris. Hypsipetes leucocephalus was identified as the dominant species.

During the survey period, a total of 23 fruiting plant species were measured (see

Table A2. Frugivorous bird species composition at southern region of Meihuashan National Nature Reserve, Fujian.

Order/Family	Species	IUCN	Nurtured Broad-Leaved Forest	Unguarded Broad-Leaved Forest	Village	Bamboo Forest	Trophic Level	Residency Status
GALLIFORMES
Phasianidae	Lophura nycthemera	II	√	√	√	√	R	O
PICIFORMES
Capitonidae	Psilopogon virens		√	√	√	√	R	O
PASSERIFORMES
Vireonidae	Erpornis zantholeuca		√	√	√	√	R	O
Campephagidae	Pericrocotus flammeus		√	√		√	R	I
	Pericrocotus solaris		√	√	√	√	R	I
Corvidae	Urocissa erythroryncha		√	√	√	√	R	O
	Dendrocitta formosae		√	√	√	√	R	O
Paridae	Parus major		√	√	√	√	R	I
	Machlolophus spilonotus		√	√	√	√	R	O
Pycnonotidae	Pycnonotus sinensis		√	√	√	√	RWP	O
	Hypsipetes leucocephalus		√	√	√	√	RS	O
	Pycnonotus jocosus		√	√	√		R	O
	Hemixos castanonotus		√	√	√	√	R	O
	Spizixos semitorques		√	√	√	√	R	O
	Ixos mcclellandii		√	√	√	√	R	O
Aegithalidae	Aegithalos concinnus		√	√	√	√	R	I
Zosteropidae	Zosterops japonicus		√	√	√		RSWP	O
	Yuhina castaniceps			√			R	O
Pellorneidae	Alcippe morrisonia		√	√	√	√	W	O
Leiothrichidae	Garrulax pectoralis		√	√	√	√	R	O
	Garrulax monileger		√	√			R	O
Muscicapidae	Rhyacornis fuliginosa			√	√	√	R	I
	Tarsiger cyanurus		√			√	W	O
	Cyanoptila cyanomelana		√	√	√	√	P	I
Chloropseidae	Chloropsis hardwickii		√	√	√	√	R	O
Nectariniidae	Aethopyga christinae			√	√	√	R	P
Passeridae	Passer montanus			√	√		R	O

Note: “II” means national second-class protected bird; “√” means that this species of bird has been recorded. O: omnivore; I: insectivore; P: palynivore; R: residents; W: winter visitors; S: summer visitors; P: passage migrant.

). Among these, 18 species (88 individuals) were recorded during the autumn–winter period of 2021, with 18 species (85 individuals) showing evidence of bird foraging; in the summer of 2022, 5 species (20 individuals) were measured, all of which exhibited signs of bird consumption. The species comprised large trees such as Taxus Chinensis and Photinia serratifolia, along with shrubs including Viburnum dilatatum, Phytolacca acinosa, and Callicarpa bodinieri.

4.2. Correlation Between Frugivorous Bird Diversity and Plant Traits

During autumn and winter, 24 bird species feeding on 18 fruit-bearing plant species were analyzed for correlations between bird diversity and plant characteristics (

Table 1. Species diversity of frugivorous bird communities in different habitats and seasons at southern region of Meihua Mountain, Fujian.

Season	Habitat	Species Richness	Abundance	Shannon–Wiener	Pielou
Autumn and Winter	Nurtured broad-leaved forest	16	366	2.004	0.723
	Unguarded broad-leaved forest	20	258	2.730	0.911
	Bamboo forest	15	214	1.958	0.723
	Village	13	224	2.366	0.922
Summer	Nurtured broad-leaved forest	5	46	1.196	0.743
	Unguarded broad-leaved forest	7	26	1.778	0.914
	Bamboo forest	4	8	1.321	0.953
	Village	4	20	1.258	0.907

). The results (Figure 2) show significant correlations between tree height and species richness (p = 0.001), abundance (p = 0.026), Shannon–Wiener diversity index (p = 0.001), and Simpson dominance index (p = 0.001). However, tree height showed no significant correlation with Pielou evenness index (p > 0.05). Canopy density significantly correlated with species richness (p = 0.001), abundance (p = 0.003), Shannon–Wiener diversity index (p = 0.000), Pielou evenness index (p = 0.009), and Simpson dominance index (p = 0.000). Mean number of fruits showed significant correlations with species richness (p = 0.027), abundance (p = 0.009), Shannon–Wiener diversity index (p = 0.042), and Simpson dominance index (p = 0.062), but not with Pielou evenness index (p > 0.05).

In summer, 13 bird species feeding on five fruit-bearing plant species were analyzed. The results (Figure 1) indicate no significant correlations (p > 0.05) between tree height, canopy density, or fruit mean number of fruits and any of the following parameters: species richness, abundance, abundance, Shannon–Wiener diversity index, Pielou evenness index, and Simpson dominance index. Neither canopy density nor fruit mean number of fruits showed significant correlations (p > 0.05) with these bird diversity indices.

4.3. Variations in Avian Diversity Across Habitats and Seasons

Based on generalized linear models of plant traits, the relative importance of each factor on frugivorous bird species diversity was obtained. In the village habitat, DBH was excluded due to a VIF > 5. Both bird abundance (σ² = 135.333) and Shannon–Wiener index (σ² = 0.124) showed an importance value of 1 for tree height (σ² = 18.969) and canopy density (σ² = 50.705), while importance for fruit amount (σ² = 3,351,560.302) was 0; all other importance values were below 0.5. In nurtured broad-leaved forests, bird abundance (σ² = 306.969) had an importance of 1 for tree height (σ² = 71.342), DBH (σ² = 1187.497), canopy density (σ² = 0.021), and fruit amount (σ² = 15,265,130.316), while all other values were below 0.5. In unguarded broad-leaved forests, bird abundance (σ² = 521.21) also had an importance of 1 for tree height (σ² = 24.212), DBH (σ² = 24.054), canopy density (σ² = 0.027), and fruit amount (σ² = 2,864,545.988), while the Shannon–Wiener index (σ² = 0.158) showed an importance greater than 0.5 (0.69) for canopy density, and 0 importance for tree height and fruit amount; all other values were below 0.5. In bamboo forests, bird abundance (σ² = 128.886) had an importance of 1 for canopy density (σ² = 0.027), and importance above 0.5 (0.69) for tree height (σ² = 69.911), DBH (σ² = 127.334), and fruit amount (σ² = 33,584,570.242), while all other importance values remained below 0.5 (

Table 2. The relative importance of plant traits for avian species diversity during autumn and winter.

		Tree Height	DBH	Canopy Density	Fruit Amount
Village	Species richness	0.28	-	0.19	0.3
	Abundance	1	-	1	0
	Shannon–Wiener	1	-	1	0
	Pielou	0.13	-	0.13	0.13
	Simpson	0.22	-	0.1	0.17
Nurtured broad-leaved forest	Species richness	0.17	0.13	0.13	0.14
	Abundance	1	1	1	1
	Shannon–Wiener	0.37	0.18	0.11	0.1
	Pielou	0.13	0.13	0.13	0.13
	Simpson	0.13	0.14	0.13	0.13
Unguarded broad-leaved forest	Species richness	0.07	0.19	0.34	0.07
	Abundance	1	1	1	1
	Shannon–Wiener	0	0.16	0.69	0
	Pielou	0.11	0.11	0.11	0.11
	Simpson	0.12	0.11	0.18	0.11
Bamboo forest	Species richness	0.13	0.11	0.17	0.14
	Abundance	0.69	0.69	1	0.69
	Shannon–Wiener	0.14	0.28	0.09	0.19
	Pielou	0.12	0.11	0.11	0.12
	Simpson	0.1	0.21	0.08	0.19

).

The multifactorial ANOVA reveals that the interaction effect between habitat and season had no significant impact on bird species richness (F = 0.316, p = 0.813), abundance (F = 0.110, p = 0.953), Shannon–Wiener diversity index (F = 0.789, p = 0.519), or Simpson’s dominance index (F = 0.720, p = 0.555). However, a significant effect was observed on Pielou’s evenness index (F = 3.852, p = 0.032).

5. Discussion

Our study demonstrates that the relationships between fruiting plants and frugivorous bird species diversity are strongly influenced by both seasonal dynamics and habitat type. Below, we interpret our key findings in the context of the existing literature and highlight their implications for conservation.

Consistent with our predictions, we found significant correlations between bird species diversity metrics and plant traits only during the autumn–winter fruiting season. During this period, birds are more likely to track fruit resources [20,31], explaining the significant associations between bird abundance, Shannon–Wiener, Pielou, and Simpson with tree height, DBH, and canopy density. By contrast, the lack of significant correlations in summer suggests that birds may rely more on arthropods or other resources during the breeding season, reducing their dependence on fruiting plants [32].

Our habitat-specific GLMs reveal that the importance of plant traits varied considerably across disturbance contexts.

Villages: Tree height and canopy density were the most important predictors of bird abundance and Shannon–Wiener, while fruit amount had no effect. This suggests that in highly disturbed habitats, structural attributes may provide refuge or perching sites more than food resources [18]. The exclusion of DBH due to high VIF (11.42) further indicates multicollinearity among structural variables in simplified habitats.

Nurtured broad-leaved forest and unguarded broad-leaved forest: All plant traits (height, DBH, canopy density, fruit amount) were important for bird abundance, highlighting the role of multi-dimensional habitat complexity in supporting frugivore bird communities [22]. However, only canopy density consistently influenced Shannon–Wiener, underscoring the importance of vertical stratification for niche partitioning [33].

Bamboo Forests: Canopy density was the most important trait for bird abundance, but tree height, DBH, and fruit amount also contributed significantly. This suggests that even in monocultural plantations, vegetation structure and resource availability can support bird communities, though likely with lower diversity than in natural forests [34].

Our study reveals an important implicit finding: the significant reliance of bird diversity on fruit resources during autumn and winter is likely contributed to by a much broader array of species than just typical frugivores. As our results show, plant traits were not significantly correlated with bird diversity in summer (the breeding season), whereas these correlations became strongly significant in autumn and winter. This pronounced seasonal contrast may be explained by a fundamental ecological mechanism: during autumn and winter, when insect availability declines sharply, many insectivorous or omnivorous birds shift their diet and increasingly depend on fruit resources [35,36].

Our results emphasize that habitat management should prioritize structural complexity (e.g., canopy density, tree height) alongside fruit availability, especially in human-dominated landscapes; nurtured forests—whether broad-leaved or bamboo—can support bird communities if structural and resource diversity are maintained; and that even anthropized habitats like villages and bamboo forests can contribute to conservation by providing resources, especially during critical seasons. Their value lies not in replicating natural forests but in providing complementary resources that support different aspects of biodiversity.

This study did not account for fruit nutritional quality, spatial arrangement of trees, or bird movement patterns, which could further clarify plant–bird interactions. Future studies should integrate remote sensing, molecular dietary analysis, and tracking technologies to better understand how frugivores navigate and utilize disturbed landscapes across seasons.