Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 Jun 29;94(9):1866–1878. doi: 10.1111/1365-2656.70097

Asian elephants are associated with a more robust mammalian community in tropical forests

Li‐Li Li 1,2, Ru‐Chuan He 1,2, Cheng Chen 3,4,5, Rui‐Chang Quan 1,2,✉
PMCID: PMC12424276  PMID: 40583232

Abstract

  1. Megaherbivores are experiencing a global extinction crisis before we fully understand their ecological functions. While the role of megaherbivores as ecosystem engineers—enhancing environmental structure complexity and facilitating seed dispersal—is well‐documented, their influence on animal community assemblies remains less explored, especially in tropical forests. This knowledge gap is crucial for effective, functional‐oriented conservation planning.

  2. Therefore, we investigated the association between Asian elephants (Elephas maximus) and mammalian community assemblages—from community to species level—in tropical forests of Southwest China, using long‐term monitoring data from camera traps.

  3. Our results revealed that the presence of Asian elephants was associated with a more robust co‐occurrence network within mammalian communities. Additionally, elephants were positively correlated with the abundance of mammal species, especially ungulates and primates. At the species level, while some mammals temporarily avoided Asian elephants, most retained their diel activity patterns, presumably because they were neither in a predator–prey relationship nor intense competitors.

  4. Our results show that Asian elephants not only affect vegetation but also are associated with a more robust mammalian community. Consequently, protecting elephants is a pivotal conservation action towards maintaining robust animal communities in Asian tropical forests.

Keywords: abundance, community, Elephas maximus, mammal, megaherbivore, network analysis

It's the first evidence that Asian elephants are positively associated with robustness of mammalian networks, increases ungulate/primate abundances and minimally disrupts activity patterns. Highlights elephants' overlooked role as keystone architects beyond vegetation engineering, urging conservation prioritization to safeguard ecosystem resilience.

graphic file with name JANE-94-1866-g006.jpg

1. INTRODUCTION

Many vertebrates are experiencing population declines and range restrictions due to rapid environmental changes driven by human activities, leading to a global extinction crisis (Ceballos et al., 2015; Dirzo et al., 2014; Ripple et al., 2019). Specifically, megafauna (defined as animals >44 kg; Martin & Klein, 1995) are particularly vulnerable to extinction (Cardillo et al., 2005) due to climatic changes (Wroe et al., 2013) and, predominantly, human disturbances (Bergman et al., 2023; Saltré et al., 2019; Svenning et al., 2024). Many are often locally extirpated or extinct before their ecological functions can be fully understood (Hansen & Galetti, 2009). Due to their large body size, their diet and behavior can strongly affect sympatric species (Wright & Jones, 2004). Understanding how these species affect ecosystems is essential for conservationists and policymakers to develop comprehensive, function‐oriented conservation schemes.

Increasing evidence shows that megafauna play unique and irreplaceable roles in ecosystems (Malhi et al., 2015; Trepel et al., 2024). Through activities such as foraging and movement (Campos‐Arceiz & Blake, 2011; Sekar et al., 2015), they modify resource availability and influence the abundance and diversity of other species (Ogada et al., 2008; Pringle, 2008; Wright & Jones, 2004). Particularly, megaherbivores (defined as species weighing over 1000 kg; Moleón et al., 2020; Owen‐Smith, 1988) usually function as ecosystem engineers by dispersing seeds and increasing environmental structure complexity (Sekar et al., 2015; Terborgh et al., 2025). However, while most of the research on the ecological function of megaherbivores focuses on their impacts on vegetation (Ong et al., 2023), their influence on the assembly of animal communities is rarely studied and often overlooked. Especially in tropical forests, how the animal communities respond to the presence of megaherbivores is to be explored (Pringle et al., 2023; Trepel et al., 2024).

For instance, megaherbivores may play important roles in structuring animal communities. Coexisting species enable this by adopting diverse strategies, including modifications of their timing, space use and resource partitioning, to coexist with megaherbivores (He et al., 2023, 2024). Some smaller animals avoid megaherbivores spatially to reduce competition associated with their presence, which can lead to a decrease in their abundance (Keesing, 1998). Such behavioral adaptations contribute significantly to the spatial and diel activity patterns of animals (He et al., 2024), mediating interspecific interactions and potentially altering competitive hierarchies or predator–prey dynamics for other species. Given the complexity of these interactions, the responses of sympatric mammals are unlikely to be uniform across the entire community. Instead, the presence of megaherbivores may foster a spectrum of outcomes—facilitating some species or functional groups while potentially disadvantaging or having neutral effects on others, depending on their specific ecological niches and adaptations to megaherbivore‐modified environments. This may lead to the formation of different community properties, such as network robustness—defined as the system's resilience and ability to maintain structural integrity following potential species loss (Landi et al., 2018; Peng & Wu, 2016). However, while much research has focused on the association between megaherbivores and the abundance and diversity of other species (Keesing & Young, 2014; Kerlay & Landman, 2006), there remains a gap in our understanding of how these megaherbivores associate with the community structure by influencing the behavioural and ecological strategies of coexisting species, thereby altering species composition and interactions in the community. This potential for varied, species‐specific responses underscores the need for a multi‐level analytical approach. Addressing this gap could provide key insights into the role of megaherbivores in ecosystem functioning and the processes that govern biodiversity in complex animal communities.

The Asian elephant (Elephas maximus), one of the largest terrestrial mammals and a megaherbivore, is known to function as an ecosystem engineer in tropical forests (Campos‐Arceiz & Blake, 2011; Ong et al., 2021). For instance, their diet, and more importantly, their movement patterns shape the fates of seed dispersal in tropical areas (Campos‐Arceiz et al., 2008) and can serve as ecological filters by suppressing their preferred food plants (Ong et al., 2023). Although Asian elephants are herbivores, which attracts more research attention to their impacts on the plant communities, their non‐trophic interactions with the animal community are less well understood. Studies on African savanna elephants (Loxodonta africana) and Asian elephants have shown that these megaherbivores can create refuges for other animals by increasing environmental heterogeneity (Platt et al., 2018; Pringle, 2008). Accordingly, we predicted that Asian elephants could have important non‐trophic interactions with animal communities in tropical forest ecosystems because, for instance, the open habitat Asian elephants create may affect the movements and foraging of other animals (Ong et al., 2023). The loss of elephants in tropical forests could result in significant changes to animal communities. However, no evidence has yet clarified how animal communities might respond to the extirpation of Asian elephants. Given that Asian elephants are classified as an endangered species by the IUCN Red List (Williams et al., 2020), research on their ecological function should be prioritized before their potential extirpation to further support comprehensive conservation strategies and maintain an intact ecosystem.

Using camera trap data collected over 5 years, this study aims (1) to explore whether the presence of Asian elephants associates with the mammal assembly from the community to species level; and (2) if such associations are confirmed, investigate the underlying mechanisms by analysing how elephant presence correlates with shifts in the diel activity and temporal avoidance patterns of other mammal species. We predicted that mammal abundance could increase with the presence of Asian elephants and present a more robust co‐occurrence network. Although other mammals might temporarily avoid the large‐bodied Asian elephants, they are unlikely to alter their diel activity patterns, as they are neither in a predator–prey relationship nor experiencing intense competition.

2. MATERIALS AND METHODS

2.1. Study area

We selected the eastern part of Xishuangbanna Dai Autonomous Prefecture (eastern XSBN), Yunnan province, southwest China, as our study area. Eastern XSBN is located to the east of Lancang River, home to the majority of Asian elephants in China (more than 80% of China's ~300 elephants reside in eastern XSBN; Figure 1; Liu et al., 2018; Tang et al., 2023). In this region, Xishuangbanna National Nature Reserve (including four sub‐reserves in eastern XSBN: Mengyang, Menglun, Mengla and Shangyong) and Yiwu Prefectural Nature Reserve were established to conserve Asian elephants and the tropical ecosystem. This study area's vegetation includes tropical rainforest, tropical seasonal moist forest, tropical montane evergreen broad‐leaved forest and tropical monsoon forest (Zhu et al., 2006). Within such landscapes, Asian elephants typically utilize habitats composed of broad‐leaved, coniferous or mixed coniferous‐broad‐leaved forests (Li et al., 2023). This region experiences a dry‐cool season from November to February, a dry‐hot season from March to April and a rainy/wet season from May to October (Zhang & Cao, 1995), with an average annual temperature of 21°C and a mean annual precipitation of 1500 mm (Zhu et al., 2006).

FIGURE 1.

FIGURE 1

Open in a new tab

Locations of selected camera traps in the study area (Eastern part of Xishuangbanna, Yunnan Province, China).

2.2. Camera trap survey and data selection

We used camera traps (Loreda L710, Shenzhen, China) to monitor the occurrence and activity of animals and to investigate the mammal species community with and without the presence of Asian elephants. To ensure sampling independence, we set 431 camera traps at a minimum distance of 500 m along 65 transect lines, mainly in Nature Reserves (Mengyang, Menglun, Mengla, Shangyong and Yiwu) and partly in state‐owned forests (outside Nature Reserves) from 2017 to 2021. We attached the camera traps to trees at a height of 0.5 m with no bait. The camera traps operated 24 h a day with high sensitivity, and three photographs were taken at each trigger without delay. We replaced the batteries and SD cards every 6 months.

From the photographs captured by the camera traps, we manually filtered false triggers and extracted data on mammal species, capture date, capture time and GPS location. In total, 38 camera traps detected the presence of Asian elephants (Figure 1). To compare the mammal community in areas with and without the presence of Asian elephants, we made a buffer zone with a 10 km radius around camera locations where Asian elephants were detected and excluded the data from camera traps that detected no elephants within the buffer zones, to ensure that the camera traps we selected as samples with no elephants were not influenced by the presence of elephants. We selected 10 km as the threshold based on the home range of elephants in this region (Zhang et al., 2006). These procedures were conducted using the ArcMap tool (version 10.6) of the ArcGIS (ESRI, Redlands, CA). Since 38 camera traps detected the presence of Asian elephants, we balanced the sampling efforts by selecting 38 camera traps that did not detect elephants, thereby ensuring a direct comparison between areas with and without elephants. We selected these camera traps based on similar environmental factors, such as elevation, slope, annual average precipitation, nearest distance to water, normalized difference vegetation index (hereafter NDVI), tree density, human relative abundance index (RAI), livestock RAI, nearest distance to the road and nearest distance to the cropland (see Table S1 for the source and resolution of the data). To confirm their similarity, we applied Wilcoxon tests and found no significant difference between all the environmental variables between sites with and without elephants (elevation: W = 825.5, p = 0.2846; slope: W = 720, p = 0.9876; annual average precipitation: W = 808, p = 0.3744; nearest distance to water: W = 782, p = 0.5386; NDVI: W = 835, p = 0.2441; tree density: W = 839, p = 0.2261; human RAI: W = 695, p = 0.7475; livestock RAI: W = 751.5, p = 0.6796; nearest distance to the road: W = 665, p = 0.5593, and nearest distance to the cropland: W = 664, p = 0.5524). All the camera trap surveys in this study were approved by the State Forestry Administration of China (No. 439). We used all data across 5 years as these areas were part of the elephants' regular distribution range and the lack of detection during a particular time period does not necessarily imply their absence from the area.

2.3. Asian elephants' association with the overall mammal community

To assess whether mammal assemblages in habitats with and without Asian elephants differed, we conducted non‐metric multidimensional scaling (NMDS; McClune & Mefford, 1999) ordination analysis based on mammals' RAI data and Bray–Curtis dissimilarity distance, using R package vegan 2.6‐4 (Oksanen et al., 2022). The RAI was defined as the integral number of independent detections per 100 trapping days. Consecutive photographs of a given species at the same camera station were considered independent if they were taken at least 30 min apart. An NMDS stress value smaller than 0.2 indicates adequate representation of the data across three axes (Clarke, 1993). Based on Bray–Curtis dissimilarity distance, we quantified the differences in species composition with and without the presence of Asian elephants by performing an analysis of similarities (ANOSIM). The R values range from −1 to 1, where a positive value represents a higher variation between groups than that within a group and vice versa. A larger R value means a larger difference between groups.

We further applied network analysis to explore the overall co‐occurrence patterns in mammalian community with and without the presence of Asian elephants. To mitigate the impact of false co‐occurrence resulting from species sharing similar environmental needs, we employed Hierarchical Modelling of Species Communities (HMSC), a type of joint species distribution model that allows the simultaneous assessment of species‐environment and species‐species relationships (Ovaskainen et al., 2017), to calculate species‐to‐species associations for network inference. This model was fitted using R's package Hmsc 3.0‐13 (Tikhonov et al., 2020), with the presence‐absence (probit distribution) and relative abundance (poisson distribution) species matrices as the response variables, and the selected environmental variables (elevation, slope, annual average precipitation, nearest distance to water, NDVI, tree density, human RAI, livestock RAI, nearest distance to the road and nearest distance to the cropland; see Table S1 for the data source and resolution) as the independent variables. The model setting followed Aung et al. (2022). We defined a co‐occurrence event as valid if the statistical support of the association value was larger than 0.95, indicating that 95% of the 1000 posteriors were either below or above zero (Tikhonov et al., 2017). The nodes in the network represent the species, and the edges represent the correlation between nodes. We then calculated the average degree, average path length, network diameter, network density, transitivity and clustering coefficient to describe the network topology (Newman, 2003). The average degree measures the average number of interactions a species has with others in the network, which ecologically means that a higher average degree can make the ecosystem more resilient to the removal of one or more species by providing compensatory interactions. Average path length and network diameter provide complementary information about the global connectivity and efficiency of the network, where shorter average path lengths and smaller diameters typically indicate more cohesive ecosystems where species are closely interconnected. Network density refers to the proportion of actual interactions (edges) relative to the total possible interactions in the network, with higher densities indicating a highly interdependent ecosystem. Moreover, transitivity measures the likelihood of indirect interactions, reflecting the overall integration of the ecosystem. Finally, the clustering coefficient indicates how likely species are to form cohesive groups or clusters of interacting species within the ecosystem (Newman, 2010). These statistical analyses were performed using vegan (Oksanen et al., 2022) and igraph (Csárdi & Nepusz, 2006) R packages. We quantified network robustness against random species loss using simulations. Nodes from the network were removed randomly, and the impact on network structure was measured using natural connectivity, a sensitive discrimination of network robustness (Peng & Wu, 2016). The simulation tracked the decline in natural connectivity as the proportion of removed nodes increased, to estimate how quickly the network structural robustness degraded. From an ecological standpoint, this quantitative measure of robustness reflects the structural resilience of a community. Specifically, a higher robustness score indicates a greater capacity for the network to absorb random species loss without triggering widespread secondary extinctions (extinction cascades, Landi et al., 2018), thereby preserving overall connectivity and function (Peng & Wu, 2016).

To examine whether the strength of associations among sympatric mammal species differs in areas with and without elephants, we compared the pairwise correlations, which represent the extent to which the presence or RAI of one species could associate with the presence or RAI of another species. The Spearman correlation test was deployed using the cor function based on the presence‐absence and RAI matrix from areas with and without elephant presence. After calculating the correlations, a one‐tailed t‐test was used to compare the Spearman correlations.

2.4. Asian elephants' association with the presence, richness and RAI of mammals at different scales

Using generalized linear models (GLMs), we explored how the presence of Asian elephants was correlated with the richness and RAI of mammals at community (overall) and order levels. We set the richness or RAI as response variable, respectively, and the environmental variables and the presence of Asian elephants as independent variables. To improve model convergence, we increased the iteration number to the maximum (10,000). The GLMs were first fitted using the Poisson distribution, and in the case of overdispersion, we fitted them with a ‘quasipoisson’ distribution. We used the binomial distribution for tree shrews, since there was only one species in the order Scandentia. To determine the percentage of dependent variables that explained the response variable, we use rsq function in R package rsq to calculate the coefficient of determination (Zhang, 2023).

Using the HMSC modelling approach described above, we explored how the presence of Asian elephants was correlated with the presence and RAI at the species level. Small rodents were grouped together as a single category (referred to as ‘small rodents’) because it was difficult to identify them from the photos of camera trapping. Unlike the HMSC modelling used in network analysis above, this approach not only incorporates environmental variables but also includes the presence of elephants as an independent variable. The coefficients (posterior means) were considered significant if the statistical support exceeded 0.95 (Ovaskainen et al., 2017). The goodness of model fit was evaluated by species‐specific AUC and R 2 values.

2.5. The change of mammal activities when elephants are presented in the habitat

To understand how the presence of Asian elephants influences mammal activities, we first used non‐parametric kernel density estimation to calculate the diel activity patterns of each mammal species both with and without the presence of elephants. We considered consecutive photographs taken within 30 min of the same species at the same camera station as non‐independent detections, to avoid overestimating species occurrences (O'Brien et al., 2003). Only species with more than 10 independent detections were included in this analysis. The capture time of photographs was transformed into circular solar time using the R package activity 1.3.2 (Nouvellet et al., 2012), and we measured the overlap of temporal segregation in the package overlap 0.3.4 (Ridout & Linkie, 2009) by the coefficient of overlap D‐hat (Δ). Δ1 was used when the detection number of a species (under circumstances with or without the presence of elephants) was smaller than 50; otherwise, Δ4 was used. A D‐hat value higher than 0.75 was considered strong overlap, 0.5–0.75 as medium overlap and lower than 0.5 as weak overlap. We then used a permutation approach to test the significance of activity differences (Niedballa et al., 2019).

In addition to changes in diel activity patterns, some species may avoid other species by not using the same site within a short period of time, which is defined as temporary avoidance. We calculated the avoidance ratio, which is the median of the time intervals when species A is detected after the detection of Asian elephants divided by that of the successive detection of species A, to explore if mammal species avoid temporally the sites utilized by Asian elephants. The avoidance ratios greater than 1 indicate a stronger interspecific avoidance than intra‐specific avoidance. We tested the statistical difference using Wilcoxon signed‐rank exact tests.

3. RESULTS

From January 2017 to April 2021, we amassed a total of 38,550 camera trap days (mean camera trapping days per site = 507.24 ± 247.16 SD; for information on the number of camera traps, deployment dates, camera days and mean trapping days per camera, see Table S2). We obtained a total of 9822 independent events, of which 78 were independent events of Asian elephants. In areas where Asian elephants were present, we recorded 6001 independent events involving 32 mammal species. In contrast, areas without Asian elephants yielded 3821 independent events involving 29 mammal species.

3.1. Asian elephants' association with the mammal community and co‐occurrence networks

The NMDS ordination analysis revealed a stress value of 0.158, with R 2 of non‐metric fit and linear fit at 0.98 and 0.85, respectively, suggesting that the data were adequately represented by the three NMDS axes (Figure 2). The ANOSIM results showed a significant dissimilarity between the mammalian communities with and without Asian elephants (ANOSIM R = 0.105, p = 0.001; Figure 2).

FIGURE 2.

FIGURE 2

Open in a new tab

The non‐metric multidimensional scaling (NMDS) plot showing the dissimilarities in the mammalian community with and without the presence of Asian elephants (ANOSIM R = 0.105, Stress = 0.158, p = 0.001).

Both co‐occurrence networks based on presence/absence data (with and without elephants) comprised a total of 18 species (nodes; Table 1). The RAI‐based co‐occurrence networks showed similar values for the number of the edges, average degree, average path length, network diameter, network density, transitivity and clustering coefficient, indicating that the complexity of the mammalian community remained similar with or without elephants (Figure 3). However, the correlation between mammal species—indicating how the presence of one species is associated with the presence of another—became significantly weaker when Asian elephants were absent based on the presence/absence data: t = −6.51, df = 152, p < 0.001, but was not significantly different based on the RAI data: t = 0.3, df = 152, p = 0.62 (Figure 3). In addition, RAI‐based network robustness was significantly higher with elephants than without (t = −2.56, df = 24.96, p = 0.01704).

TABLE 1.

Parameters of the association of Asian elephants with the mammal species co‐occurrence networks.

Network data Asian elephants Nodes Edges Average degree Average path length Network diameter Network density Transitivity Clustering coefficient
Presence/absence Presence 18 0 0 NA 0 0 NA NA
Absence 18 0 0 NA 0 0 NA NA
RAI Presence 18 34 3.78 1.00 1.60 0.22 0.77 0.77
Absence 18 32 3.56 1.20 2.76 0.21 0.77 0.77
Open in a new tab

Abbreviation: RAI, relative abundance index.

FIGURE 3.

FIGURE 3

Open in a new tab

The association of Asian elephants with the mammal species co‐occurrence networks and the correlations. Panel (a) presents the differences of the correlation of other mammals between the scenarios with and without Asian elephants based on presence‐absence data. Panel (b) illustrates the difference of the correlation of other mammals and the co‐occurrence networks between the scenarios with and without Asian elephants based on RAI data. Panel (c) compares the robustness of the co‐occurrence networks based on presence‐absence data between the scenarios with and without Asian elephants, and panel (d) compares the robustness of the co‐occurrence networks based on RAI data between the scenarios with and without Asian elephants. RAI, relative abundance index.

3.2. Asian elephants' association with the mammal richness and RAI at different scales

At the community (overall) level, elephant presence had no significant correlation with mammal species richness (estimate = 0.002, t = 0.026, p = 0.979; Figure 4; Table S3). However, elephant presence was significantly positively associated with the mammals' RAI (estimate = 0.44, t = 3.16, p < 0.01; Figure 4; Table S3). At the order level, elephant presence had no significant correlation with the richness of Rodentia (estimate = 0.094, z = 0.352, p = 0.725), Scandentia (estimate = −0.641, z = −0.886, p = 0.376), Artiodactyla (estimate = 0.083, z = 0.577, p = 0.564), Carnivora (estimate = −0.184, t = −1.342, p = 0.184), nor with the RAI of Rodentia (estimate = −0.355, t = −0.731, p = 0.467), Scandentia (estimate = −0.069, z = −0.256, p = 0.798; Figure 4; Table S3) and Carnivora (estimate = −0.239, t = −0.99, p = 0.3261). In contrast, elephant presence had significant positive correlations with the richness of Primates (estimate = 1.055, z = 3.053, p < 0.01), as well as with the RAI of Primates (estimate = 1.663, t = 4.262, p < 0.001) and Artiodactyla (estimate = 1.052, t = 5.535, p < 0.001; Figure 4; Table S3).

FIGURE 4.

FIGURE 4

Open in a new tab

Association of Asian elephants with the richness/presence and RAI of mammals at different scales (community scale, order scale and species scale). RAI, relative abundance index.

At the species level, Asian elephants had a significant positive correlation with the presence of Cervus equinus (coefficient = 0.51, statistical support = 0.96), Macaca mulatta (coefficient = 1.11, statistical support = 1) and Hystrix brachyura (coefficient = 0.73, statistical support = 0.96; Figure 4; Table S3), while they had a significant negative correlation with the presence of small rodents (coefficient = −0.6, statistical support = 0.95) and Melogale moschata (coefficient = −0.61, statistical support = 0.96; Figure 4; Table S3). Additionally, Asian elephants had a significant positive correlation with the RAI of Muntiacus vaginalis (coefficient = 1.301, statistical support = 1), Sus scrofa (coefficient = 0.59, statistical support = 1), M. leonina (coefficient = 2.24, statistical support = 1), C. equinus (coefficient = 0.84, statistical support = 0.95), Paradoxurus hermaphroditus (coefficient = 0.54, statistical support = 0.97), H. brachyura (coefficient = 1.39, statistical support = 1), Ursus thibetanus (coefficient = 1.31, statistical support = 1), Arctonyx albogularis (coefficient = 0.76, statistical support = 0.95), M. mulatta (coefficient = 2.38, statistical support = 1) and Tragulus williamsoni (coefficient = 1.68, statistical support = 1; Figure 4; Table S3), while they had a significant negative correlation with the RAI of M. moschata (coefficient = −1.11, statistical support = 0.99) and Capricornis milneedwardsii (coefficient = −1.2, statistical support = 1).

3.3. Mammal activity patterns in the presence of Asian elephants

The presence of Asian elephants generally had no significant association with the diel activity patterns of most mammal species detected in our camera traps (Figure S1). However, six carnivore species, Chrotogale owstoni (Dhat1 = 0.84, p < 0.001), Arctonyx albogularis (Dhat1 = 0.58, p < 0.001), Paradoxurus hermaphroditus (Dhat4 = 0.76, p < 0.05), Herpestes urva (Dhat1 = 0.79, p < 0.05), Prionailurus bengalensis (Dhat1 = 0.83, p < 0.01) and Martes flavigula (Dhat4 = 0.52, p < 0.05), exhibited statistically significant changes in their activity patterns where Asian elephants were present. Among them, Arctonyx albogularis showed a decreased activity density between 6:00 and 18:00 (when the elephants were most active), while Martes flavigula showed an increased activity density during this period in the presence of elephants (Figure S1).

Instead of changing diel activity patterns, mammal species avoided Asian elephants at a fine time scale, with all species showing a temporary avoidance index greater than 1 (Figure 5). At the order level, rodents and carnivores showed the strongest temporary avoidance of elephants, followed by ungulates (Artiodactyla) and primates.

FIGURE 5.

FIGURE 5

Open in a new tab

The temporary avoidance relationship between Asian elephants and other mammal species in the tropical areas. The temporary avoidance ratio larger than 1 (where the dashed line marks) represents significant avoidance of the presence of Asian elephants.

4. DISCUSSION

Focusing on Asian elephants, an endangered megaherbivore, our study highlights their significant association with the mammalian community in tropical forests, particularly their positive association with network robustness and animal abundance, an aspect that has long been neglected in research on the ecological functions of megaherbivores. Under rapid global economic development, many animals, particularly megaherbivores, have become endangered or even extinct before their ecological functions are adequately understood (Ripple et al., 2019). Our findings underscore the adverse consequence on the local animal community that could result from the extirpation of a megaherbivore such as Asian elephant.

4.1. Asian elephants are associated with the mammalian community in various ways

Our results show that, although the association between elephants and species richness was low, elephant presence was significantly associated with higher mammal abundance at the community level, with more taxa increasing their abundances than decreasing. This suggests that elephants may act as ecological facilitators by enhancing resource availability or modifying habitats (Ong et al., 2023) in ways that support higher densities of coexisting species without necessarily expanding the taxonomic breadth of the community. For instance, elephants' role as ecosystem engineers, through seed dispersal (Campos‐Arceiz et al., 2008) or creating microhabitats via foraging and trampling (Ong et al., 2023), could promote localized resource hotspots that attract or sustain larger populations of other mammals. However, the lack of a corresponding increase in species richness implies that these modifications do not substantially alter niche diversity or allow for the colonization of new species, potentially reflecting saturation of ecological niches in these tropical forests or limitations in the spatial scale of elephant‐mediated effects.

Further supporting their critical ecological role, our finding that elephant presence is associated with a more robust mammalian community network offers an additional layer of insight. This heightened robustness, defined in our study as a greater ability to maintain natural connectivity despite simulated random species loss (Peng & Wu, 2016), suggests that communities with the presence of elephants are structurally more resilient to perturbations. This increased resilience could be linked to the aforementioned positive effects of elephants: by associating with higher overall mammal abundance and potentially increasing the diversity or strength of interactions, elephant‐modified habitats may foster networks with greater redundancy and a stronger capacity to buffer against the impacts of species declines. Ecologically, this implies that communities with elephants are less susceptible to cascading secondary extinctions, a critical attribute for maintaining biodiversity and ecosystem function in the face of environmental change or other stressors.

At the community level, contrary to previous studies that found elephant presence in African savannas suppressed the abundance of small mammals (Keesing & Young, 2014), our study showed a positive association between the presence of Asian elephants and the abundance of other mammal species in tropical forests. Two factors may explain this discrepancy: firstly, the density of Asian elephants in our study area was relatively low (Tang et al., 2023), which might not lead to significant reductions in local food availability that would cause intense inter‐species competition. Secondly, African savannas and Asian tropical forests differ in the food resources they provide, which could also be crucial in influencing interspecific relationships (Ong et al., 2023; Trepel et al., 2024). However, at the order level, we also found a negative association between Asian elephants and the RAI of Rodentia and Scandentia, which partially aligns with the previous study. These results highlight the need for further studies at multiple levels in tropical areas to comprehensively understand the ecological roles of megaherbivores.

Notably, the differential responses of carnivores and ungulates to elephant presence highlight taxon‐specific mechanisms of coexistence. While ungulate abundance correlated positively with elephant presence—given their dietary similarities, this could indicate indirect benefits from elephant‐driven habitat heterogeneity—their diel activity patterns remained unaltered, indicating a lack of temporal avoidance (Figure 4; Figure S1; Table S3). In contrast, certain carnivores exhibited shifts in activity timing despite unchanged abundance or richness, suggesting behavioural adjustments to mitigate competition or predation risk in elephant‐occupied zones. This dichotomy underscores elephants' role as a key species affecting community structure: their physical presence may impose a ‘landscape of fear’ on the small carnivores detected in our study (Laundré et al., 2010), prompting temporal niche partitioning, while simultaneously fostering conditions that support higher ungulate densities through bottom‐up processes (Figure 4; Figure S1; Table S3). Interestingly, our study also documents a significant increase in both the richness and abundance of primates in areas with presence of Asian elephants (Figure 4). This aligns with a report that M. leonina are known to follow Asian elephants and feed on their leftovers, such as sugarcane (Choudhury, 2010). This result suggests that elephants may facilitate access to certain food resources for primates. These findings align with growing evidence that megaherbivores shape ecosystems not only through direct interactions but also by modulating spatiotemporal dynamics among sympatric species (Keesing & Young, 2014; Ripple et al., 2015).

4.2. Essential ecological functions of megaherbivores

Megaherbivores can change the environmental structure and vegetation type through movement and foraging (Trepel et al., 2024). From the perspectives of trophic cascades, more attention has been paid to megaherbivores' effects on vegetation (Sekar et al., 2017; Tan et al., 2021). However, through direct or indirect influences, megaherbivores are also indispensable in composing an integral animal community (Ripple et al., 2015). For instance, megaherbivores can, although rarely, be prey for some large top predators (Loveridge et al., 2006); their carcasses provide nutrients for scavengers, plants and microbes (Arnberg et al., 2024; Pereira et al., 2013); they influence sympatric herbivores by altering vegetation (Rutina et al., 2005; Tambling et al., 2013); and animals may rely on megaherbivores' defecation (Awasthi et al., 2024; Campos‐Arceiz, 2009). These studies align with our findings, which reveal that Asian elephants can interact with various mammalian species, and these interactions create complex patterns of community assembly at both the order and community levels, contributing to a more robust mammalian community in tropical forests. Therefore, the extirpation of megaherbivores could lead to unintended consequences, such as population reductions of ungulates or to animal communities more vulnerable to environmental changes.

Despite their positive effect on the animal community, an overloaded megaherbivore density may become detrimental to an ecosystem. For instance, overconsumption of plants by a dense population of megaherbivores could suppress the survival of other sympatric herbivores (Landman et al., 2013), especially when food resources are limited. Yet in tropical forests, the ecological consequences of overpopulated megaherbivores remain unclear. Furthermore, while we considered a comprehensive set of environmental variables in this study, it is possible that some relevant factors were overlooked. Further studies are needed to provide insight for conservation management of megaherbivores.

4.3. Megaherbivores serve as umbrella species

Megaherbivores are mostly threatened by hunting, habitat loss, deforestation and human disturbance (Ripple et al., 2015). Their population recovery could be challenging once they become endangered, as the reproduction of megaherbivores is generally slow (Lueders et al., 2012). As megaherbivores like Asian elephants usually require a large home range (Bahar et al., 2018), the conservation strategies protecting these animals could also preserve the natural habitats shared by other species. As the case in our study, protecting Asian elephants will protect the habitat they inhabit, and the sympatric animals will also benefit from such conservation efforts. Therefore, consistent with Ripple et al. (2017), we argue that megaherbivore conservation could serve as an ‘umbrella’ strategy to protect many species and ecosystems, and that protecting megaherbivores before their extirpation is a crucial step to prevent ecosystems from becoming more vulnerable.

5. CONCLUSIONS

In tropical forest ecosystems, Asian elephants were strongly associated with a more robust mammalian community. Network robustness increased potentially due to the variety of interactions with elephants, involving mammals from different taxonomic groups. The presence, abundance and various diel activity shifts at the species level associated with Asian elephants contribute to the heterogeneity of the overall mammal community. While the ecological role of elephants in altering vegetation is well recognized (Ong et al., 2021, 2023; Sekar et al., 2015, 2017), their functional role in maintaining animal communities may be equally important. The extirpation of Asian elephants from their habitats could severely impact the animal communities and ecological processes. Given the uniqueness of megaherbivores, we advocate for further studies to explore their ecological roles and the underlying mechanisms by which they influence other animals, to better protect the entire animal community in a function‐oriented manner.

AUTHOR CONTRIBUTIONS

Li‐Li Li and Rui‐Chang Quan designed the study. Ru‐Chuan He and Rui‐Chang Quan gathered the data. Ru‐Chuan He and Li‐Li Li performed the analysis and graphing. Li‐Li Li, Cheng Chen and Rui‐Chang Quan wrote the manuscript, with all authors contributing critically to the drafts and giving final approval for publication.

FUNDING INFORMATION

This work was funded by the National Natural Science Foundation of China (32301294, 32371745), the Transboundary Cooperation on Biodiversity Research and Conservation in Gaoligong Mountains (E1ZK251), Yunnan Province Science and Technology Department (No. 202203AP140007), the 14th Five‐Year Plan of the Xishuangbanna Tropical Botanical Garden, Chinese Academy of Sciences (E3ZKFF7B, E3ZKFF9B) and the Yunnan Revitalization Talent Support Program ‘Innovation Team’ Project (202405AS350019).

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest.

Supporting information

Table S1. The source and resolution of the environment covariates used in statistic analysis in this study.

Table S2. Information on the final samplings of camera trap survey used in the study.

Table S3. The association of Asian elephants with the richness, presence and RAI of mammal species on different levels.

Figure S1. The impact of Asian elephants (where the solid line represents the presence of Asian elephants, and the dashed line represents the absence of Asian elephants) on the activity patterns of different mammal species in the tropical areas.

JANE-94-1866-s001.docx (337.9KB, docx)

ACKNOWLEDGEMENTS

We are grateful to Jia‐Bin Li, Ying Geng, Hui Cao, Ying Liu for their help in extracting data from photographs of camera traps. We also thank the offices of Xishuangbanna National Nature Reserve and Yiwu Prefectural Nature Reserve for supporting this research. We thank Xiao‐Bao Deng, Guang‐Hong Cao and Guo‐Gang Li for setting the cameras, and Nan Sun, Wei Ao, Jia‐Zhou Yang and Xiao‐Ying Mi for parts of the species identification. We are also grateful to Ahimsa Campos‐Arceiz for polishing the language and refining the main text. This study is supported by biodiversity conservation initiatives from Southeast Asia Biodiversity Research Institute (SEABRI) and Xishuangbanna Tropical Botanical Garden (XTBG), Chinese Academy of Sciences, advancing the implementation of Goal A of the Kunming‐Montreal Global Biodiversity Framework (GBF).

Li, L.‐L. , He, R.‐C. , Chen, C. , & Quan, R.‐C. (2025). Asian elephants are associated with a more robust mammalian community in tropical forests. Journal of Animal Ecology, 94, 1866–1878. 10.1111/1365-2656.70097

Li‐Li Li and Ru‐Chuan He contributed equally to the work.

Handling Editor: Saskya van Nouhuys

DATA AVAILABILITY STATEMENT

Data and R code available from Zenodo https://doi.org/10.5281/zenodo.14994541 (Li et al., 2025).

REFERENCES

  1. Arnberg, M. P. , Eycott, A. E. , & Steyaert, S. M. J. G. (2024). From death comes life: Large vertebrate carrion enhances seedling establishment in clonal ericaceous shrubs. Functional Ecology, 38, 1284–1295. [Google Scholar]
  2. Aung, K. M. , Chen, H.‐H. , Segar, S. T. , Miao, B.‐G. , Peng, Y.‐Q. , & Liu, C. (2022). Changes in temperature alter competitive interactions and overall structure of fig wasp communities. Journal of Animal Ecology, 91, 1303–1315. [DOI] [PubMed] [Google Scholar]
  3. Awasthi, B. , McConkey, K. R. , Aluthwattha, S. T. , Chen, C. , & Chen, J. (2024). Restoring ecological function: Interactions between vertebrates and latrines in a reintroduced population of Rhinoceros unicornis . Biological Conservation, 294, 110611. [Google Scholar]
  4. Bahar, A. , Kasim, N. H. A. , & Hambali, K. (2018). Home range and movement patterns of Asian elephant (Elephas maximus) in Gua Musang, Kelantan, Malaysia. Malayan Nature Jounal, 70, 221–232. [Google Scholar]
  5. Bergman, J. , Pedersen, R. Ø. , Lundgren, E. J. , Lemoine, R. T. , Monsarrat, S. , Pearce, E. , Pearce, E. A. , Schierup, M. H. , & Svenning, J. C. (2023). Worldwide Late Pleistocene and Early Holocene population declines in extant megafauna are associated with Homo sapiens expansion rather than climate change. Nature Communications, 14, 7679. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Campos‐Arceiz, A. (2009). Shit happens (to be useful)! Use of elephant dung as habitat by amphibians. Biotropica, 41, 406–407. [Google Scholar]
  7. Campos‐Arceiz, A. , & Blake, S. (2011). Megagardeners of the forest: The role of elephants in seed dispersal. Acta Oecologica, 37, 542–553. [Google Scholar]
  8. Campos‐Arceiz, A. , Larrinaga, A. R. , Weerasinghe, U. R. , Takatsuki, S. , Pastorini, J. , Leimgruber, P. , Fernando, P. , & Santamaría, L. (2008). Behavior rather than diet mediates seasonal differences in seed dispersal by Asian elephants. Ecology, 89, 2684–2691. [DOI] [PubMed] [Google Scholar]
  9. Cardillo, M. , Mace, G. M. , Jones, K. E. , Bielby, J. , Bininda‐Emonds, O. R. P. , Sechrest, W. , Orme, C. D. L. , & Purvis, A. (2005). Multiple causes of high extinction risk in large mammal species. Science, 309, 1239–1241. [DOI] [PubMed] [Google Scholar]
  10. Ceballos, G. , Ehrlich, P. R. , Barnosky, A. D. , García, A. , Pringle, R. M. , & Palmer, T. M. (2015). Accelerated modern human‐induced species losses: Entering the sixth mass extinction. Science Advances, 1, e1400253. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Choudhury, A. (2010). Notes on behaviour of pig‐tailed Macaque Macaca nemestrina leonine towards sparrow‐halk, crows, drongo and elephants. Zoos' Print (Web Version), XXV(7), 23–24. [Google Scholar]
  12. Clarke, K. R. (1993). Non‐parametric multivariate analyses of changes in community structure. Australian Journal of Ecology, 18, 117–143. [Google Scholar]
  13. Csárdi, G. , & Nepusz, T. (2006). The igraph software package for complex network research. Inter Journal, Complex Systems, 1695, 1–9. [Google Scholar]
  14. Dirzo, R. , Young, H. S. , Galetti, M. , Ceballos, G. , Isaac, N. J. , & Collen, B. (2014). Defaunation in the Anthropocene. Science, 345, 401–406. [DOI] [PubMed] [Google Scholar]
  15. Hansen, D. M. , & Galetti, M. (2009). The forgotten megafauna. Science, 324, 42–43. [DOI] [PubMed] [Google Scholar]
  16. He, R.‐C. , Wang, L. , Goodale, E. , Li, K. , & Quan, R.‐C. (2023). The extremely small body size of Williamson's mouse deer (Tragulus williamsoni) allows coexistence with sympatric larger ungulates through temporal avoidance. Frontiers in Ecology and Evolution, 11, 1125840. [Google Scholar]
  17. He, R.‐C. , Wang, L. , Hughes, A. C. , Geng, Y. , Cao, H. , Liu, Y. , He, R.‐. C. , Li, J.‐. B. , Wang, Q.‐. Y. , Yang, Z.‐. B. , & Quan, R.‐. C. (2024). Anthropogenic disturbance and competitive avoidance shape the coexistence pattern of macaque species in tropical forests. Ecosphere, 15, e70001. [Google Scholar]
  18. Keesing, F. (1998). Impacts of ungulates on the demography and diversity of small mammals in central Kenya. Oecologia, 116, 381–389. [DOI] [PubMed] [Google Scholar]
  19. Keesing, F. , & Young, T. P. (2014). Cascading consequences of the loss of large mammals in an African savanna. BioScience, 64, 487–495. [Google Scholar]
  20. Kerlay, G. I. H. , & Landman, M. (2006). The impacts of elephants on biodiversity in the Eastern Cape subtropical thickets: Elephant conservation. South African Journal of Science, 102, 395–402. [Google Scholar]
  21. Landi, P. , Minoarivelo, H. O. , Brännström, Å. , Hui, C. , & Dieckmann, U. (2018). Complexity and stability of ecological networks: A review of the theory. Population Ecology, 60, 319–345. [Google Scholar]
  22. Landman, M. , Schoeman, D. S. , & Kerley, G. I. H. (2013). Shift in black rhinoceros diet in the presence of elephant: Evidence for competition? PLoS One, 8, e69771. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Laundré, J. W. , Hernández, L. , & Ripple, W. J. (2010). The lanscape of fear: Ecological implications of being afraid. The Open Ecology Journal, 3, 1–7. [Google Scholar]
  24. Li, L.‐L. , He, R.‐C. , Chen, C. , & Quan, R.‐C. (2025). Data from: Asian elephants are associated with a more robust mammalian community in tropical forests. Zenodo. 10.5281/zenodo.14994541 [DOI] [PMC free article] [PubMed]
  25. Li, W. W. , Liu, P. , Yang, N. , Chen, S. , Guo, X. , Wang, B. , Li, W. , & Zhang, L. (2023). Improving landscape connectivity through habitat restoration: Application for Asian elephant conservation in Xishuangbanna prefecture, China. Integrative Zoology, 19, 319–335. [DOI] [PubMed] [Google Scholar]
  26. Liu, S. , Yin, Y. , Li, J. , Cheng, F. , Dong, S. , & Zhang, Y. (2018). Using cross‐scale landscape connectivity indices to identify key habitat resource patches for Asian elephants in Xishuangbanna, China. Landscape and Urban Planning, 171, 80–87. [Google Scholar]
  27. Loveridge, A. J. , Hunt, J. E. , Murindagomo, F. , & Macdonald, D. W. (2006). Influence of drought on predation of elephant (Loxodonta africana) calves by lions (Panthera leo) in an African wooded savannah. Journal of Zoology, 270, 523–530. [Google Scholar]
  28. Lueders, I. , Niemuller, C. , Rich, P. , Gray, C. , Hermes, R. , Goeritz, F. , & Hildebrandt, T. B. (2012). Gestating for 22 months: Luteal development and pregnancy maintenance in elephants. Proceedings of the Royal Society B: Biological Sciences, 279, 3687–3696. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Malhi, Y. , Doughty, C. E. , Galetti, M. , & Terborgh, J. W. (2015). Megafauna and ecosystem function from the Pleistocene to the Anthropocene. Proceedings of the National Academy of Sciences of the United States of America, 113, 838–846. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Martin, P. S. , & Klein, R. G. (1995). Quaternary extinctions: A prehistoric revolution. University of Arizona Press. [Google Scholar]
  31. McClune, B. , & Mefford, M. J. (1999). Multivariate analysis of ecological data. MjM Software Design. [Google Scholar]
  32. Moleón, M. , Sánchez‐Zapata, J. A. , Donázar, J. A. , Revilla, E. , Martín‐López, B. , Gutiérrez‐Cánovas, C. , Getz, W. M. , Morales‐Reyes, Z. , Campos‐Arceiz, A. , Crowder, L. B. , Galetti, M. , González‐Suárez, M. , He, F. , Jordano, P. , Lewison, R. , Naidoo, R. , Owen‐Smith, N. , Selva, N. , Svenning, J. C. , … Tockner, K. (2020). Rethinking megafauna. Proceedings of the Royal Society B: Biological Sciences, 287, 20192643. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Newman, M. E. J. (2003). The structure and function of complex networks. SIAM Review, 45, 167–256. [Google Scholar]
  34. Newman, M. E. J. (2010). Networks: An introduction (1st ed.). Oxford Academic. [Google Scholar]
  35. Niedballa, J. , Wilting, A. , Sollmann, R. , Hofer, H. , & Courtiol, A. (2019). Assessing analytical methods for detecting spatiotemporal interactions between species from camera trapping data. Remote Sensing in Ecology and Conservation, 5, 272–285. [Google Scholar]
  36. Nouvellet, P. , Rasmussen, G. S. A. , MacDonald, D. W. , & Courchamp, F. (2012). Noisy clocks and silent sunrises: Measurement methods of daily activity pattern. Journal of Zoology, 286, 179–184. [Google Scholar]
  37. O'Brien, T. G. , Kinnaird, M. F. , & Wibisono, H. T. (2003). Crouching tigers, hidden prey: Sumatran tiger and prey populations in a tropical forest landscape. Animal Conservation, 6, 131–139. [Google Scholar]
  38. Ogada, D. L. , Gadd, M. E. , Ostfeld, R. S. , Young, T. P. , & Keesing, F. (2008). Impacts of large herbivorous mammals on bird diversity and abundance in an African savanna. Oecologia, 156, 387–397. [DOI] [PubMed] [Google Scholar]
  39. Oksanen, J. , Simpson, G. , Blanchet, F. , Kindt, R. , Legendre, P. , & Minchin, P. (2022). vegan: Community ecology package . R package version 2.6‐4. https://CRAN.R‐project.org/package=vegan
  40. Ong, L. , McConkey, K. R. , & Campos‐Arceiz, A. (2021). The ability to disperse large seeds, rather than body mass alone, defines the importance of animals in a hyper‐diverse seed dispersal network. Journal of Ecology, 110, 313–326. [Google Scholar]
  41. Ong, L. , Tan, W. H. , Daveport, L. C. , McConkey, K. R. , Mat Amin, M. K. A. B. , Campos‐Arceiz, A. , & Terborgh, J. W. (2023). Asian elephants as ecological filters in Sundaic forests. Frontiers in Forests and Global Change, 6, 1143633. [Google Scholar]
  42. Ovaskainen, O. , Tikhonov, G. , Norberg, A. , Guillaume Blanchet, F. , Duan, L. , Dunson, D. , Roslin, T. , & Abrego, N. (2017). How to make more out of community data? A conceptual framework and its implementation as models and software. Ecology Letters, 20, 561–576. [DOI] [PubMed] [Google Scholar]
  43. Owen‐Smith, R. N. (1988). Megaherbivores: The influence of very large body size on ecology. Cambridge University Press. [Google Scholar]
  44. Peng, G.‐S. , & Wu, J. (2016). Optimal network topology for structural robustness based on natural connectivity. Physica A: Statistical Mechanics and it Applications, 443, 212–220. [Google Scholar]
  45. Pereira, L. M. , Owen‐Smith, N. , & Moleón, M. (2013). Facultative predation and scavenging by mammalian carnivores: Seasonal, regional and intra‐guild comparisons. Mammal Review, 44, 44–45. [Google Scholar]
  46. Platt, S. G. , Bickford, D. P. , Win, M. M. , & Rainwater, T. R. (2018). Water‐filled Asian elephant tracks serve as breeding sites for anurans in Myanmar. Mammalia, 83, 287–289. [Google Scholar]
  47. Pringle, R. M. (2008). Elephants as agents of habitat creation for small vertebrates at the patch scale. Ecology, 89, 26–33. [DOI] [PubMed] [Google Scholar]
  48. Pringle, R. M. , Abraham, J. O. , Anderson, T. M. , Coverdale, T. C. , Davies, A. B. , Dutton, C. L. , Gaylard, A. , Goheen, J. R. , Holdo, R. M. , Hutchinson, M. C. , Kimuyu, D. M. , Long, R. A. , Subalusky, A. L. , & Veldhuis, M. P. (2023). Impacts of large herbivores on terrestrial ecosystems. Current Biology, 33, R584–R610. [DOI] [PubMed] [Google Scholar]
  49. Ridout, M. S. , & Linkie, M. (2009). Estimating overlap of daily activity patterns from camera trap data. Journal of Agricultural, Biological, and Environmental Statistics, 14, 322–337. [Google Scholar]
  50. Ripple, W. J. , Chapron, G. , López‐Bao, J. V. , Durant, S. M. , Macdonald, D. W. , Lindsey, P. A. , Bennett, E. L. , Beschta, R. L. , Bruskotter, J. T. , Campos‐Arceiz, A. , Corlett, R. T. , Darimont, C. T. , Dickman, A. J. , Dirzo, R. , Dublin, H. T. , Estes, J. A. , Everatt, K. T. , Galetti, M. , Goswami, V. R. , … Zhang, L. (2017). Conserving the world's megafauna and biodiversity: The fierce urgency of now. BioScience, 67, 197–200. [Google Scholar]
  51. Ripple, W. J. , Newsome, T. M. , Wolf, C. , Dirzo, R. , Everatt, K. T. , Galetti, M. , Hayward, M. W. , Kerley, G. I. , Levi, T. , Lindsey, P. A. , Macdonald, D. W. , Malhi, Y. , Painter, L. E. , Sandom, C. J. , Terborgh, J. , & Van Valkenburgh, B. (2015). Collapse of the world's largest herbivores. Science Advances, 1, e1400103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Ripple, W. J. , Wolf, C. , Newsome, T. M. , Betts, M. , Ceballos, G. , Courchamp, R. , Hayward, M. W. , Van Valkenburgh, B. , Wallach, A. D. , & Worm, B. (2019). Are we eating the world's megafauna to extinction? Conservation Letters, 12, e12627. [Google Scholar]
  53. Rutina, L. R. , Moe, S. R. , & Swenson, J. E. (2005). Elephant Loxodonta africana driven woodland conversion to shrubland improves dry‐season browse availability for impalas Aepyeros melampus . Wildlife Biology, 11, 207–213. [Google Scholar]
  54. Saltré, F. , Chadoeuf, J. , Peters, K. J. , McDowell, M. C. , Friedrich, T. , Timmermann, A. , Ulm, S. , & Bradshaw, C. J. A. (2019). Climate‐human interaction associated with southeast Australian megafauna extinction patterns. Nature Communications, 10, 5311. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Sekar, N. , Lee, C.‐L. , & Sukumar, R. (2015). In the elephants seed shadow: The prospects of domestic bovids as replacement dispersers of three tropical Asian trees. Ecology, 9, 2093–2105. [DOI] [PubMed] [Google Scholar]
  56. Sekar, N. , Lee, C.‐L. , & Sukumar, R. (2017). Functional nonredundancy of elephants in a disturbed tropical forest. Conservation Biology, 31, 1152–1162. [DOI] [PubMed] [Google Scholar]
  57. Svenning, J.‐C. , Lemoine, R. T. , Bergman, J. , Buitenwerf, R. , Le Roux, E. , Lundgren, E. , Mungi, N. , & Pedersen, R. Ø. (2024). The late‐Quaternary megafauna extinctions: Patterns, causes, ecological consequences and implications for ecosystem management in the Anthropocene. Cambridge Prisms: Extinction, 2, e5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Tambling, C. J. , Minnie, L. , Adendorff, J. , & Kerley, G. I. H. (2013). Elephants facilitate impact of large predators on small ungulate prey species. Basic and Applied Ecology, 14, 694–701. [Google Scholar]
  59. Tan, W. H. , Hii, A. , Solana‐Mena, A. , Wong, E. P. , Loke, V. P. W. , Tan, A. S. L. , Kromann‐Clausen, A. , Hii, N. , Pura, P. B. , Tunil, M. T. B. , Al Din, S. , Chin, C. F. , & Campos‐Arceiz, A. (2021). Long‐term monitoring of seed dispersal by Asian elephants in a Sundaland rainforest. Biotropica, 00, 1–13. [Google Scholar]
  60. Tang, Y. , Yang, C. , Yang, Z. , & Chen, F. (2023). Population dynamics of wild Asian elephants over the past fifty years. Acta Theriologica Sinica, 43, 593–607. (In Chinese with English abstract). [Google Scholar]
  61. Terborgh, J. , Ong, L. , Davenport, L. C. , Tan, W. H. , Mena, A. S. , McConkey, K. , & Campos‐Arceiz, A. (2025). Release of tree species diversity follows loss of elephants from evergreen tropical forests. Proceedings of the Royal Society B: Biological Sciences, 292, 20242026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Tikhonov, G. , Abrego, N. , Dunson, D. , & Ovaskainen, O. (2017). Using joint species distribution models for evaluating how species‐to‐species associations depend on the environmental context. Methods in Ecology and Evolution, 8, 443–452. [Google Scholar]
  63. Tikhonov, G. , Opedal, Ø. H. , Abrego, N. , Lehikoinen, A. , de Jonge, M. M. J. , Oksanen, J. , & Ovaskainen, O. (2020). Joint species distribution modelling with the r‐package Hmsc. Methods in Ecology and Evolution, 11, 442–447. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Trepel, J. , le Roux, E. , Abraham, A. J. , Buitenwerf, R. , Kamp, J. , Kristensen, J. A. , Tietje, M. , Lundgren, E. J. , & Svenning, J. C. (2024). Meta‐analysis shows that wild large herbivores shape ecosystem properties and promote spatial heterogeneity. Nature Ecology & Evolution, 8, 705–716. [DOI] [PubMed] [Google Scholar]
  65. Williams, C. , Tiwari, S. K. , Goswami, V. R. , de Silva, S. , Kumar, A. , Baskaran, N. , Yoganand, K. , & Menon, V. (2020). Elephas maximus. The IUCN Red List of threatened species 2020 . e.T7140A45818198. 10.2305/IUCN.UK.2020-3.RLTS.T7140A45818198.en [DOI]
  66. Wright, J. P. , & Jones, C. G. (2004). Predicting effects of ecosystem engineers on patch‐scale species richness from primary productivity. Ecology, 85, 2071–2081. [Google Scholar]
  67. Wroe, S. , Field, J. H. , Archer, M. , Grayson, D. K. , Price, G. J. , Louys, J. , Faith, J. T. , Webb, G. E. , Davidson, I. , & Mooney, S. D. (2013). Climate change frames debate over the extinction of megafauna in Sahul (Pleistocene Australia‐New Guinea). Proceedings of the National Academy of Sciences of the United States of America, 110, 8777–8781. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Zhang, D. (2023). rsq: R‐squared and related measures. R package version 2.6. https://CRAN.R‐project.org/package=rsq
  69. Zhang, J. , & Cao, M. (1995). Tropical forest vegetation of Xishuangbanna, SW China and its secondary changes, with special reference to some problems in local nature conservation. Biological Conservation, 73, 229–238. [Google Scholar]
  70. Zhang, L. , Wang, N. , Wang, Y. N. , & Ma, L. C. (2006). A preliminary study on the habitat and behaviors of Asian elephant (Elephas maximus) in Simao, Yunnan, China. Acta Theriologica Sinica, 23, 185–192. (In Chinese with English abstract). [Google Scholar]
  71. Zhu, H. , Cao, M. , & Hu, H. B. (2006). Geological history, flora, and vegetation of Xishuangbanna, southern Yunnan, China. Biotropica, 38, 310–317. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Table S1. The source and resolution of the environment covariates used in statistic analysis in this study.

Table S2. Information on the final samplings of camera trap survey used in the study.

Table S3. The association of Asian elephants with the richness, presence and RAI of mammal species on different levels.

Figure S1. The impact of Asian elephants (where the solid line represents the presence of Asian elephants, and the dashed line represents the absence of Asian elephants) on the activity patterns of different mammal species in the tropical areas.

JANE-94-1866-s001.docx (337.9KB, docx)

Data Availability Statement

Data and R code available from Zenodo https://doi.org/10.5281/zenodo.14994541 (Li et al., 2025).

Articles from The Journal of Animal Ecology are provided here courtesy of Wiley

RESOURCES