Habitat prioritization for bat conservation: A case study in Vietnam

Le Quang Tuan; Vuong Tan Tu; Tran Anh Tuan; Vu Dinh Thong; Nguyen Truong Son; Nguyen Thanh Luong; Nguyen Tran Vy; Hoang Trung Thanh; Gábor Csorba; Tamás Görföl; Mao-Ning Tuanmu

doi:10.1371/journal.pone.0331094

Abstract

Bats play a vital role in ecosystems through providing essential functions and services, such as pollination and pest control, but they face many threats from anthropogenic activities, climate change, and diseases. While establishing protected areas can be a conservation measure to mitigate bats’ population declines and habitats loss, the effectiveness of current protected areas in conserving bat species remains poorly understood. Using Vietnam as a case, this study aims to evaluate the representation of the existing protected area network for bats in terms of the proportion of their distribution ranges covered and the number of well-covered species. We also identified areas that would improve the representation using the spatial prioritization approach. Our results showed that while Vietnam’s protected area network provides better representation than random allocation, it currently only covers 6.12% of bat distribution ranges on average, leaving the distribution areas of more than 78 (96%) of the 81 species examined in this study insufficiently represented. Through the spatial prioritization, the forests in Central Highlands, western Central Coast, and Northwest were identified as the key regions for bats. Expanding the current network to cover 9% and 30% of the land under the national and global targets could improve the range coverage to 11.89% and 41.99%, respectively, reducing the number of poorly represented bats to 10–23 (12.3–28.4%) species. These findings offer critical insights for conservation practitioners and policymakers into prioritizing resources and efforts for improving conservation of not only bats in Vietnam but also biodiversity in general in other places.

Citation: Tuan LQ, Tu VT, Tuan TA, Thong VD, Son NT, Luong NT, et al. (2025) Habitat prioritization for bat conservation: A case study in Vietnam. PLoS One 20(9): e0331094. https://doi.org/10.1371/journal.pone.0331094

Editor: Beza Ramasindrazana, Institut Pasteur de Madagascar, MADAGASCAR

Received: November 28, 2024; Accepted: August 12, 2025; Published: September 11, 2025

Copyright: © 2025 Tuan et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All the data used in the study are available from the Zenodo (DOI: 10.5281/zenodo.15771217).

Funding: This study was supported by the fellowship from the Taiwan International Graduate Program to LQT, grant from Vietnam Academy of Science and Technology on behalf of the Program of Developing Basic Science in Chemistry, Life Sciences, Earth Sciences and Marine Science between 2017 and 2025 (KHCBTĐ.02/22–24) to TAT, LQT and VTT, the grants CSCL.09.02/25-26 and QTBG.01.04/26-27 to LQT, TAT and VTT, the grants KHCBSS.01/24-25, UQĐTCB.07/23-25 and NVCC09.04.04/24-25 to VDT, the National Research, Development and Innovation Fund of Hungary (NKFIH FK137778) to GC and TG, and the grant from Biodiversity Research Center, Academia Sinica to MNT. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: No authors have competing interests.

Introduction

Bats (Order Chiroptera) play important roles in various ecosystems and provide key services to humans, including pest suppression, seed dispersal, and pollination [1]. Additionally, bats serve as valuable indicators for biodiversity assessments due to their quick reactions to environmental changes [2]. However, bats face numerous pressures that threaten their survival. Specifically, the destruction and alteration of roosts and foraging habitats due to urbanization, agriculture, and deforestation are primary threats [3]. Other human activities, such as cave tourism, wind energy development, and bushmeat hunting, also disturb bat colonies, reducing their survival rates, and contributing to population declines [4]. Moreover, changes in climate patterns, such as rising temperatures, further threaten bat survival, particularly by disrupting their hibernation patterns, reducing prey availability, and altering habitats [5].

Establishing protected areas (PAs) is a primary tool for biodiversity conservation, but its lack of effectiveness is not uncommon around the world, especially in developing countries [6,7]. This phenomenon is often a consequence of inadequate management and lack of financial support due to poor governance quality and bureaucratic inefficiency [8]. However, it can also stem from the flaws in the design of PAs [9], especially the poor representativeness of PAs for targeted species or ecosystems [10]. For instance, a study on African bats found that only 5% of their suitable habitats are covered by protected areas [11]. Similarly, 67.6% of Brazil’s bat species have their range poorly protected by PAs [12]. This lack of representativeness is also common in other taxa [13]. Globally, only 4–9% of terrestrial amphibians, birds and mammals are sufficiently represented within the existing protected areas [14], and many PAs are biased toward protecting charismatic and well-known species, leaving less attention for inconspicuous taxa (e.g., small bodied vertebrates and invertebrates) [15]. The inadequate representativeness is particularly profound in tropical and lower income regions, although many of them are biodiversity hotspots [7].

The lack of representativeness in protected areas has been found to undermine biodiversity conservation efforts on reducing species extinction risk [16], preventing ecosystem services loss [17], and enhancing climate resilience [18], and thus result in the failure of global conservation strategies [7]. For example, bird populations in areas with limited PA coverage experience greater exposure to habitat destruction, hunting and other anthropogenic threats, compared to those with greater PA coverage, leading to increased extinction risk [16]. Additionally, underrepresented species and ecosystems may be more vulnerable to climate change if PAs do not include diverse and resilient habitats that allow species to move or adapt to changing conditions [18]. Accordingly, as a global target has been set to increase the protected lands to 30% by 2030 [19], it is crucial to ensure the increased coverage of PAs can effectively improve their biodiversity representativeness in order to safeguard biodiversity and maintain the health and functionality of ecosystems worldwide.

An integration of species distribution models (SDMs) and spatial conservation prioritization (SCP) is a powerful tool that can significantly improve the representativeness of protected areas [11]. SDMs can offer critical insights into where species currently exist, where species could potentially occur, and how species distributions might shift in response to environmental changes [20]. Based on this information, SCP uses data-driven approaches to identify areas that are most important for biodiversity conservation, ensuring that conservation efforts are placed in the areas which can maximize ecological benefits [21]. SCP has been applied in many cases to support and enhance conservation design [22]. For example, Sinclair et al. (2018) found that 74% of global conservation prioritizations have led to on-the-ground actions, demonstrating collaboration between the academic sector and practitioners [23]. Additionally, Lagabrielle et al. (2018) used a SCP method to identify priority conservation areas in Brazil’s Atlantic forests, significantly improving land allocation efficiency and reducing conflicts between development and conservation [24]. However, despite the availability of the powerful tools, they have seldom used for guiding bat conservation, especially in the tropics, because spatial information on bat distributions is often unavailable, incomplete or inaccurate [3,25].

Vietnam, home to approximately 30% of Asia’s bat fauna and 10% of the global bat fauna, is a crucial region for bat diversity [26]. It harbours three endangered and five vulnerable species, as listed in the IUCN Red List of Threatened Species [27]. Bats in Vietnam are regarded as important indicators of overall biodiversity, both for their significant representation within the mammal fauna and their role as keystone species providing essential ecosystem services [28,29]. However, several bats in Vietnam are at risk of extinction due to habitat loss, disturbance, and overhunting [29,30]. While PAs protect critical habitats for many wildlife species in Vietnam [28,31], the extent to which these areas represent suitable habitats for bats remain poorly understood. To fill this gap, this study aims to (i) assess the representativeness of the current protected area network in Vietnam for bats and (ii) identify priority areas for expanding the network to align with national and global PA coverage target.

Materials and methods

Study area

Vietnam, located in Southeast Asia, spans a considerable latitudinal range (Fig 1). It exhibits a significant temperature gradient, with the annual mean temperature increasing from 19^oC in the northern regions to 27^oC in the south. Annual precipitation also varies widely, ranging from under 800 mm in the coastal southeast to nearly 5,000 mm in the mountainous northeast, characterized by distinct wet and dry seasons. Generally, the degree of seasonality in both temperature and precipitation diminishes from north to south [32].

Download:

Fig 1. Land cover and the locations of current protected areas in Vietnam.

The information on the land cover and the protected areas were obtained from the ESA CCI land cover 2018 product [33], and the World Database on Protected Areas [34]. Copyright notice: © ESA Climate Change Initiative – Land Cover project 2017; © UNEP-WCWC and IUCN – WDPA 2023.

https://doi.org/10.1371/journal.pone.0331094.g001

Bat occurrence data

In total, 8431 unique occurrence records of 116 bat species in Vietnam were compiled from various sources (S1 Table). Of these, 65.4% were obtained from the Global Biodiversity Information Facility (GBIF; downloaded on June 3rd, 2023, DOI: 10.15468/dl.28jn5y), 26.6% from published literature, 5.1% from unpublished records of the Institute of Ecology and Biological Resources (IEBR), 1.0% from unpublished records in the Hungarian Natural History Museum, and 1.7% from our field notes. The number of records per species ranges from 21 for Murina annamitica to 838 for Pteropus vampyrus. Detailed information on the data sources and the data searching and cleaning approaches is provided in Tuan et al. (2023) [35].

Environmental parameters

Climate data.

We obtained 19 bioclimatic variables from CHELSA (Version 2.1). These variables were derived from the mean monthly maximum and minimum temperatures and mean monthly precipitations between 1979 and 2013 [36,37]. Compared to other products (e.g., WorldClim), CHELSA provides more accurate predictions of precipitation patterns and yields better species range predictions [36]. To address the multicollinearity among the variables, we selected one from highly correlated variables (Spearman ) based on their biological or ecological relevance to bat distributions. Six variables which have been found to influence the reproductive success, mortality rates, activity patterns, roost selection, and/or torpor-arousal cycles of bats [30,38,39], including the mean temperatures in the warmest and coldest quarters of the year, mean diurnal temperature range, annual precipitation, and precipitations in the warmest and coldest quarters, were selected for building species distribution models (S1 Fig).

Land cover data.

We obtained the ESA CCI Land Cover Product [33] for the year of 2018 from ArcGIS REST Services (https://env1.arcgis.com/arcgis/rest/services/esa_cci_landcover_2018/ImageServer, accessed on Apr 2024). This product classified the global land cover into 10 types at a 300 m resolution with the World Cylindrical Equal Area projection. We calculated the percent cover of each land cover type within every 1-by-1 km grid cell with the same projection across the study area. We only used the percent cover of cropland, deciduous forest, evergreen forest, flooded vegetation, and grassland/scrub to build species distribution models (see below) because other land cover types cover a small proportion of the study area.

Karst data.

Given that karst areas provide caves as roosts for many bat species in Vietnam [29], we also included the distribution of karst as a variable in our species distribution models. We obtained a global data layer with polygons delineating areas with carbonate rocks from the World Map of Carbonate Rock Outcrops V3.0 (https://digital.lib.usf.edu/SFS0055342/00001). Since most carbonate rocks are prone to karstification, these areas can be considered as potential karst regions [40]. However, it is important to note that not all areas with carbonate rocks provide suitable caves for bats, introducing some uncertainty in this data layer. We converted this data into presence or absence of karst at 1-by-1-km resolution by rasterizing it with the World Cylindrical Equal Area projection.

Species distribution modelling.

We built species distribution models with the above-mentioned bat occurrence records and the climate, land-cover and karst data for the 81 bat species that had more than 20 occurrence records (S1 Table). All environmental data were re-projected to the World Cylindrical Equal Area projection at the 1-by-1 km resolution. We built models for individual species using the Maxent (ver. 3.4.1; [41,42]) with the target-background approach [43], which used all of the presence locations of all the 116 species as the background. We used Maxent because it does not require absence data and consistently outperforms many other modelling algorithms [44,45]. We assessed model performance using a 5-fold spatial block cross-validation [46] with the values of the area under the receiver operating characteristic curve (AUC) and the True Skill Statistic (TSS). The model configuration that achieved the highest AUC value across all 81 species—characterized by a regularization multiplier of 2.5 and a combination of linear, quadratic, and hinge feature types (S2 Fig) —was used to build the final models. To maximize the information used for predicting species distribution, we included all presence records of a species to build the final model. The continuous model outputs (i.e., predictions of species presence probabilities) were converted into a binary format (i.e., presence/absence of the species) with the threshold determined by the 10^th percentile of the training presences [47] because this approach does not require absence data. The model generation and validation were conducted in R (version 4.2.2) with the dismo (version 1.1.4; [48] and blockCV ([49]) packages. Detailed modelling approaches are available in Tuan et al. (2023) [35].

Spatial prioritization analyses.

We quantified the importance of every1-by-1-km cell in Vietnam for bat conservation and identified the priority areas with different PA coverage targets using Zonation [50]. Zonation ranks all cells in a study region by their conservation values, quantified by the complementarity and irreplaceability of the species occurring in each cell, and then removes the cells with the lowest ranks [50]. It repeats this process iteratively until the remaining cells achieve the conservation target set by the user. In this study we employed the Additive Benefit Function (ABF) cell-removal rule, which balances multiple species in systematic conservation planning (Moilanen, 2007). The conservation value for the cell i (δi) was calculated as:

where qj indicates the representation of species j in remaining cells, qj-i indicates the representation of species j of the remaining cells excluding cell i, and thus ΔVj indicates the loss of the representation of species j when cell i is removed; wj is the weight of the species j with a higher value indicating higher conservation importance; ci is the cost for including cell i into the remaining cell patches; is the change in the ratio of the boundary length to the area of the remaining cell patches following the removal of cell i; and β is a constant that defines the strength of the boundary length penalty (BLP) with a higher value leading to a lower edge-to-area ratio of remaining areas thus increasing the compactness and ecological connectivity among prioritized conservation areas. Therefore, a higher δi value indicates that cell i is more worthy of protection relative to the associated cost.

We weighted the species based on their IUCN Red List category with a value of 1 for least-concern species, 2 for near threatened species, 3 for vulnerable species, 4 for endangered species, and 5 for critically endangered species, following Kullberg et al. [51]. We assigned a cost value, as proxy for land acquisition cost, for a cell based on its land cover with a value of 10 for nature land cover types (e.g., evergreen forest) and a value of 1 for human-associated land cover types (e.g., urban), following Whitehead et al. [52]. In each iteration, we used the “edge removal” function to remove lowest-ranked cells from the edges of remaining cell patches to maintain the connectivity of remaining cells as high as possible [53]. In the prioritization process, we also used the approach of boundary length penalties (BLP), which imposed a higher penalty to a higher perimeter-to-area ratio of the remaining cell patches, to increase the cohesion of the patches [54]. We set the BLP as 0.1, which means that the conservation value of cell i was add a value equal to the change of ratio of boundary length to area of the reserve network following removal of cell i, multiple by 0.1. To reduce the computation time, we removed 50 lowest-ranked cells in each iteration. We assigned the highest conservation value to the cells within current PAs to ensure their inclusion in the prioritized areas. The boundaries of the current PAs were obtained from the World Database on Protected Areas [34]. We included the PAs of the IUCN categories from I (strict nature reserve) to VI (protected areas with sustainable use of natural resources) (S3 Table).

We considered two conservation targets in the prioritization process. First, according to the Vietnam National Strategy on Biodiversity to 2030, vision 2050 [55], protected areas are expected to cover 9% of the land area by 2030. Second, aligning with the Target 3 of the Kunming-Montreal Global Biodiversity Framework of the Convention on Biological Diversity, which aims to increase the coverage of protected areas and other effective area-based measures to 30% by 2030 [56], we also set a target of 30% land coverage. Since the current protected areas (covering ~5% of land in Vietnam) were always included in the prioritized areas, we identified additional 4% and 25% of areas to achieve the national and global conservation targets, respectively.

Area representativeness for bats.

We quantified the representativeness of current protected areas and the prioritized areas from Zonation for bats in Vietnam using two approaches. First, we calculated the proportion of the distribution range of each bat species covered by the protected areas or prioritized areas [57], and then calculated the average across all 81 species. Second, we measured the representativeness using the number of bat species whose distribution ranges were well-represented by the current protected areas or the prioritized areas. The term “well-represented” was defined by following the approach outlined by González-Maya et al. (2015) [58] but adapting it to bats in Vietnam. Specifically, species with restricted distribution (range size <1,000 km²), were considered as well-represented if 100% of their range was covered by protected areas or prioritized areas; those with wide distribution (range size > 25,000 km²), were considered as well-represented if at least 10% of their range was covered; and for those in between, the coverage criterium was determined by a linear interpolation between 10% and 100% based on the range size.

Results

Predictions of bat distributions

The mean AUC value of the distribution models was 0.770 ± 0.102 (mean ± SD), with a value greater than 0.7 for 61 (70.3%) of the 81 bat species modelled (S2 Fig, S1 Table). The mean TSS value of the distribution models was 0.495 ± 0.184 (mean ± SD), with 54 models (66.7%) showing values greater than 0.4 (S1 Table). These results indicated that most models performed well. We present the results with all 81 species included in the main text and the results including only the 61 species with an AUC value ≥0.7 in the Supplementary Materials, because the results showed similar patterns (see below). Although the contributions of predictor variables in the model varied among species, the mean temperature in the coldest quarter was the most important factor determining bat species distributions, followed by the evergreen forest coverage and precipitation in the coldest quarter (S3 Fig, S2 Table).

The predicted distribution ranges of bat species within Vietnam varied significantly in size, ranging from 33 to 309,825 km² (Fig 2). Most taxa have a wide distribution (i.e., range size > 25,000 km²), and only two species, Harpiola isodon and Barbastella darjelingensis had a restricted distribution (i.e., range size < 1,000 km²) (Fig 2).

Download:

Fig 2. Predicted distribution range size for the 81 bat species modelled.

Species ordered by their range size within Vietnam.

https://doi.org/10.1371/journal.pone.0331094.g002

The representativeness of current protected areas for bats

The current protected areas (PAs) covered, on average, 6.12% (range: 0.42%−37.50%) of the distribution range of the bat species within Vietnam (Fig 3). The mean coverage was higher than the land coverage of current PAs in Vietnam (i.e., 5%), indicating that the current PAs had greater representativeness for bats compared to randomly placed PAs. However, 78 of the 81 bat species (~96.3%), including all threatened species (i.e., endangered and vulnerable species), had their range poorly represented by the current PAs (Fig 3). The representativeness for threatened species was particular low with only 3.05%, 7.34%, and 0.42% of the range being protected for Pteropus vampyrus, Myotis pilosus and Pteropus lylei, respectively (Fig 3). The results for the 61 species with models’ AUC higher than 0.7 are similar with those for all 81 species, with 6.64% (range: 0.01%−13.95%) of the distribution range covering by the PAs and 48 of the 61 bat species (~96.3%) having their range poorly represented (S5 Fig).

Download:

Fig 3. Representativeness of current protected areas, and the prioritized areas with the conservation targets of 9% and 30% land coverage.

The representativeness was measured as the proportion of a species’ distribution range covered by the protected areas or prioritized areas. The dots indicate the criterium for a species being considered as a well-represented species. The species are ordered by their current range size within Vietnam.

https://doi.org/10.1371/journal.pone.0331094.g003

Representativeness of prioritized areas for bats

With the target of 9% land coverage (i.e., national target), the spatial prioritization analysis indicated that some evergreen forest areas in Central Highlands and western Central Coast should be added to the PA network to improve the representativeness for bats (Fig 4a). The priority areas covered 11.89% (range: 0.95%−62.50%) of the distribution range of the bat species (Fig 3). Additionally, 58 species (71.6%), including one of the three threatened species, had their range well-represented by the priority areas (Fig 3). These results indicated that a 79.28% increase (from 5% to 9%) in the size of the PA network can lead to a 94.28% (from 6.12% to 11.89%) to 17.6-fold (from 3 to 58 species) increase in the representativeness for bats.

Download:

Fig 4. Distribution of the prioritized conservation areas for bat species in Vietnam identified by the spatial prioritization analysis.

Nine percent (a) and 30% (b) of the land area with the highest priority were identified corresponding to the national and global targets, respectively. The boundaries of current protected areas in Vietnam are also shown. The protected area boundaries were obtained from the World Database on Protected Areas [34]. Copyright notice: © UNEP-WCWC and IUCN – WDPA 2023.

https://doi.org/10.1371/journal.pone.0331094.g004

With the target of 30% land coverage (i.e., global target), the spatial prioritization analysis indicated that in addition to the Central Highlands and western Central Coast, the forests in Northwest are also important for bat representation (Fig 4b). The identified priority areas covered 41.99% (range: 1.86%−97.50%) of the distribution range of the bat species (Fig 3). Additionally, 71 species (87.6%), including one of the three threatened species, had their range well-represented by the prioritized areas (Fig 3). These results indicated that a 6-fold increase (from 5% to 30%) in the size of the PA network can lead to a 7-fold (from 6.12% to 41.99%) to 22.6-fold (from 3 to 71 species) increase in the representativeness for bats. The prioritization results for only the 61 species with models’ AUC higher than 0.7 are similar (S5 and S6 Figs).

Discussion

Bats provide important ecosystem services and are threatened by various human-induced environmental changes [5,11], but they are rarely considered as a conservation target in the process of establishing protected areas (PAs) [59]. The effectiveness of current PAs for bats has been poorly studied, especially in tropical countries like Vietnam [12]. This study presents the first assessment of the representativeness of current PAs for protecting bats in Vietnam and identifies the priority areas for improving the representativeness under both national and global conservation targets. Our results show that while the current PAs are more representative than randomly placed PAs, they are far from sufficient to protect bat habitats, suggesting a need of establishing more protected areas. By integrating the spatial information of bats’ distributions and the spatial conservation prioritization approach, we also show that the country’s current plan to expand the protected area network to 9% of land coverage by 2030s has the potential to significantly increase the representativeness. However, to ensure that the network can well cover distribution ranges of a majority of bat species in Vietnam, achieving the global target (i.e., protecting 30% of the land by 2030s) should be the goal. Our identification of the priority areas provides essential information for conservation planning. The findings also highlight the need to adjust current conservation strategies to ensure the effective protection of bat populations and their habitats.

In this study, we found that the protected areas in Vietnam exhibit low representativeness for bat distribution ranges. The representativeness was similar to that for bats in Brazil (6.8%) [12] and lower than that for many other taxa, including global terrestrial mammals (20%) [57] and mammals in Costa Rica (28.3%) [58]. The low representativeness of protected areas could be attributed to the good flying ability of bats, similar to birds [60], leading to a large distribution range [12]. The low representativeness may also result from the fact that bats are usually not the conservation target during the protected area design in Vietnam [61], like in other countries [62].

Despite being a hotspot for bat diversity, our findings reveal that Vietnam’s current protected areas network is insufficient in safeguarding bats. Many previous studies have highlighted the necessity of expanding protected areas to prevent long-term major losses of species diversity in Vietnam [61,63]. In response, Vietnam is working towards the National Biodiversity Strategy to 2030 and the vision for 2050, aiming to protect 9% of the country’s territory [55]. Considering both the national and global targets on land protection, spatial prioritization provides crucial information on where to expand current protected areas. Protecting these prioritized conservation areas can significantly enhance the representation of bats within the protected area network.

According to the spatial prioritization, the forest in Central Highlands, Northwest and western Central Coastal are the key regions to improve bat representation. Locations of the priority conservation areas are consistent with those designated for protecting Vietnamese mammals in general [64]. Most of these areas have been recognized as biodiversity hotspots in Vietnam. In particular, the forests in the Central Highlands are one of the most biodiversity rich area in Vietnam, hosting animals and plants from diverse biogeographical regions in Asia [65,66]. Northwest and Western Central Coast are situated in the transition zone of the Greater Annamites Ecoregion, one of the World Wildlife Fund’s Global 200 ecoregions and known for its exceptional biodiversity values [67]. However, rapid deforestation is threatening biodiversity in these regions [68], indicating a urgent need of protecting these areas.

Threatened and narrowly distributed bat species, such as Coelops frithii, Ia io, Kerivoula picta, Myotis formosus, Myotis pilosus, Pteropus hypomelanus, P. lylei, P. vampyrus and Rousettus leschenaultii, particularly need greater conservation efforts. These species face numerous threats, including habitat loss, climate change, and human disturbance [29,35]. Moreover, our findings suggest that current protected areas provide insufficient protection for these species according to their particular low representation in PAs (Figs 2 and 3). This suggests that their extinction risk might be higher than previously estimated, indicating the need for greater conservation investments, particularly outside current PAs, to effectively safeguard these bats.

Although this study considers all bats in Vietnam, the same approach can be applied to a subset of the species (e.g., cave-dwelling species or forest specialists) to guide specific measures for various conservation goals. Furthermore, it can be applied to other aspect of biodiversity, such as phylogenetic and functional diversity [9], as well as to other taxa and other conservation targets. By integrating multiple taxa and conservation targets into the prioritization process, we can create a more comprehensive conservation strategy that reflects the complex interdependencies within ecosystems. Expanding the approach to include a wider range of species and ecological functions would ensure that conservation efforts are more holistic, improving outcomes for both species and overall biodiversity [69].

The conservation value of different species is inherently context-dependent and can vary based on the specific conservation goals being pursued [70]. For instance, if the objective is to preserve evolutionary distinctiveness or functional traits that support ecosystem resilience, species prioritization may differ from approaches focusing solely on extinction risk [71]. In this study, species were weighted according to their IUCN Red List categories, which emphasizes the urgency of protecting threatened species. However, alternative weighting strategies could be employed in future applications to better reflect goals such as safeguarding ecosystem services, maintaining ecological interactions, or preserving phylogenetic diversity. Therefore, it is important for conservation planners to adapt the weighting scheme according to the intended conservation outcomes.

Our study relied on Maxent, a widely adopted algorithm known for its strong performance with presence-only data [41,42], for modelling species distributions. However, like all other algorithms, Maxent operates under specific assumptions and modelling decisions, such as background point selection, regularization settings, and feature types, which can introduce uncertainty into the resulting species distributions [41,43]. While we carefully evaluated various modelling settings and used the best one in this study, future studies could incorporate multiple SDM algorithms to better capture model uncertainties.

In conclusion, this study presents the first attempt to evaluate and improve the representativeness of protected areas for bat conservation in Vietnam. Our results show that current protected areas fall short in adequately safeguarding bats. However, the spatially explicit information on priority conservation areas obtained from an integration of species distribution modelling and spatial conservation prioritization offers a solution by optimizing the design of a protected area network. These findings provide great contributions to the conservation of not only bats but also broader biodiversity.

Supporting information

S1 Table. Bat species included in the study.

The Red List category of the species, the number of occurrence records used in modeling, the model performance measures (i.e., AUC and TSS values), the predicted range size, and estimated range representations are also shown. The taxonomy and nomenclature used follow the Mammal Diversity Database (Mammal Diversity Database. (2024). Mammal Diversity Database (Version 2.0) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.15007505), with updated names according to recent taxonomic revisions in parentheses.

https://doi.org/10.1371/journal.pone.0331094.s001

(PDF)

S2 Table. The contributions of environmental variables in the model of individual species.

The value measures an environmental variable’s share in the model’s overall gain (a measure of fit). It’s calculated by summing each variable’s incremental gain during the model’s training process and converting these sums to percentages.

https://doi.org/10.1371/journal.pone.0331094.s002

(PDF)

S3 Table. The number and total area of protected areas in Vietnam by the IUCN category.

The values were calculated from the data in the World Database on Protected Areas. The corresponding national categories are also provided.

https://doi.org/10.1371/journal.pone.0331094.s003

(PDF)

S1 Fig. Pairwise correlations of the six bioclimatic variables used for modeling the distributions of Vietnamese bat species.

The values shown are Spearman’s rank correlation coefficients.

https://doi.org/10.1371/journal.pone.0331094.s004

(PDF)

S2 Fig. The AUC values of the distribution models built for individual bat species with different combinations of feature types and regularization values.

Each boxplot shows the minimum, 25^th percentile, median, 75^th percentile and maximum values across all species modeled in the study. The boxplots and associated combinations are ordered by the AUC values, with those on the top having the highest value. The feature types are linear (L), quadratic (Q) and hinge (H).

https://doi.org/10.1371/journal.pone.0331094.s005

(PDF)

S3 Fig. AUC values of the distribution models built for the bat species in Vietnam.

The red line shows the value of 0.7.

https://doi.org/10.1371/journal.pone.0331094.s006

(PDF)

S4 Fig. The percent contributions of the environmental variables across the bat species distribution models.

Each boxplot shows the minimum, 25^th percentile, median, 75^th percentile and maximum values across all species modelled in the study. The variables are ordered by the medians of the percent contributions, with the one on the top having the highest median.

https://doi.org/10.1371/journal.pone.0331094.s007

(PDF)

S5 Fig. Representativeness of current protected areas, and the prioritized areas with the conservation targets of 9% and 30% land coverage.

Same as Fig 3, but for bat species with an AUC value greater than 0.7.

https://doi.org/10.1371/journal.pone.0331094.s008

(PDF)

S6 Fig. Distribution of prioritized conservation areas for bat species in Vietnam identified by the spatial prioritization analysis.

Same as Fig 4, but for the bat species with an AUC value greater than 0.7. The protected area boundaries were obtained from the World Database on Protected Areas [34]. Copyright notice: © UNEP-WCWC and IUCN – WDPA 2023.

https://doi.org/10.1371/journal.pone.0331094.s009

(PDF)

Acknowledgments

We extend special thanks to Dr. Joe Chun Chia Huang for his contributions in compiling and cleaning the bat occurrence data. We also thank Dr. I-Ching Chen, Dr. Jai-Chyi Pei, Dr. Jian-Nan Liu, and Dr. Pei-Jen Lee Shaner for their valuable insights and guidance on this study.

References

1. Ghanem SJ, Voigt CC. Increasing awareness of ecosystem services provided by bats. Adv Study Behav. 2012;44:279–302.
- View Article
- Google Scholar
2. Russo D, Salinas-Ramos VB, Cistrone L, Smeraldo S, Bosso L, Ancillotto L. Do we need to use bats as bioindicators? Biology (Basel). 2021;10(8):693. pmid:34439926
- View Article
- PubMed/NCBI
- Google Scholar
3. Frick WF, Kingston T, Flanders J. A review of the major threats and challenges to global bat conservation. Ann N Y Acad Sci. 2020;1469(1):5–25. pmid:30937915
- View Article
- PubMed/NCBI
- Google Scholar
4. Mickleburgh S, Waylen K, Racey P. Bats as bushmeat: a global review. Oryx. 2009;43(2):217–34.
- View Article
- Google Scholar
5. Festa F, Ancillotto L, Santini L, Pacifici M, Rocha R, Toshkova N, et al. Bat responses to climate change: a systematic review. Biol Rev Camb Philos Soc. 2023;98(1):19–33. pmid:36054527
- View Article
- PubMed/NCBI
- Google Scholar
6. Leverington F, Costa KL, Pavese H, Lisle A, Hockings M. A global analysis of protected area management effectiveness. Environ Manage. 2010;46(5):685–98. pmid:20859627
- View Article
- PubMed/NCBI
- Google Scholar
7. Watson JEM, Dudley N, Segan DB, Hockings M. The performance and potential of protected areas. Nature. 2014;515(7525):67–73. pmid:25373676
- View Article
- PubMed/NCBI
- Google Scholar
8. Borrini G, Dudley N, Jaeger T, Lassen B, Pathak N, Phillips A. Governance of protected areas: from understanding to action. Best Pract Prot Area Guidel Ser. 2013.
9. Brum FT, Graham CH, Costa GC, Hedges SB, Penone C, Radeloff VC, et al. Global priorities for conservation across multiple dimensions of mammalian diversity. Proc Natl Acad Sci U S A. 2017;114(29):7641–6. pmid:28674013
- View Article
- PubMed/NCBI
- Google Scholar
10. Chauvenet ALM, Kuempel CD, McGowan J, Beger M, Possingham HP. Methods for calculating protection equality for conservation planning. PLoS One. 2017;12(2):e0171591. pmid:28199341
- View Article
- PubMed/NCBI
- Google Scholar
11. Smith A, Schoeman MC, Keith M, Erasmus BFN, Monadjem A, Moilanen A, et al. Synergistic effects of climate and land-use change on representation of African bats in priority conservation areas. Ecol Indic. 2016;69:276–83.
- View Article
- Google Scholar
12. Pessoa da Silva F, De Marco Júnior P. The role of protected areas in safeguarding bat diversity in Brazil. Biodivers Conserv. 2023;32(8–9):2909–24.
- View Article
- Google Scholar
13. Rodrigues ASL, Andelman SJ, Bakarr MI, Boitani L, Brooks TM, Cowling RM, et al. Effectiveness of the global protected area network in representing species diversity. Nature. 2004;428(6983):640–3. pmid:15071592
- View Article
- PubMed/NCBI
- Google Scholar
14. Coad L, Watson JE, Geldmann J, Burgess ND, Leverington F, Hockings M, et al. Widespread shortfalls in protected area resourcing undermine efforts to conserve biodiversity. Front Ecol Environ. 2019;17(5):259–64.
- View Article
- Google Scholar
15. Delso Á, Fajardo J, Muñoz J. Protected area networks do not represent unseen biodiversity. Sci Rep. 2021;11(1):12275. pmid:34112867
- View Article
- PubMed/NCBI
- Google Scholar
16. Butchart SHM, Scharlemann JPW, Evans MI, Quader S, Aricò S, Arinaitwe J, et al. Protecting important sites for biodiversity contributes to meeting global conservation targets. PLoS One. 2012;7(3):e32529. pmid:22457717
- View Article
- PubMed/NCBI
- Google Scholar
17. Laurance WF, Useche DC, Rendeiro J, Kalka M, Bradshaw CJA, Sloan SP, et al. Averting biodiversity collapse in tropical forest protected areas. Nature. 2012;489(7415):290–4. pmid:22832582
- View Article
- PubMed/NCBI
- Google Scholar
18. Eigenbrod F, Gonzalez P, Dash J, Steyl I. Vulnerability of ecosystems to climate change moderated by habitat intactness. Glob Chang Biol. 2015;21(1):275–86. pmid:25059822
- View Article
- PubMed/NCBI
- Google Scholar
19. Secretariat CBD. Global biodiversity framework. 2023. Available from: https://www.cbd.int/gbf/targets
20. Zamora‐Gutierrez V, Pearson RG, Green RE, Jones KE. Forecasting the combined effects of climate and land use change on Mexican bats. Divers Distrib. 2017;24(3):363–74.
- View Article
- Google Scholar
21. Moilanen A, Leathwick JR, Quinn JM. Spatial prioritization of conservation management. Conserv Lett. 2011;4(5):383–93.
- View Article
- Google Scholar
22. Giakoumi S, Richardson AJ, Doxa A, Moro S, Andrello M, Hanson JO, et al. Advances in systematic conservation planning to meet global biodiversity goals. Trends Ecol Evol. 2025;40(4):395–410. pmid:39880725
- View Article
- PubMed/NCBI
- Google Scholar
23. Sinclair SP, Milner‐Gulland EJ, Smith RJ, McIntosh EJ, Possingham HP, Vercammen A, et al. The use, and usefulness, of spatial conservation prioritizations. Conserv Lett. 2018;11(6).
- View Article
- Google Scholar
24. Lagabrielle E, Lombard AT, Harris JM, Livingstone T-C. Multi-scale multi-level marine spatial planning: a novel methodological approach applied in South Africa. PLoS One. 2018;13(7):e0192582. pmid:29969441
- View Article
- PubMed/NCBI
- Google Scholar
25. Cayuela L, Golicher DJ, Newton AC, Kolb M, De Alburquerque FS, Arets EJMM. Species distribution modeling in the tropics: problems, potentialities, and the role of biological data for effective species conservation. Tropical Conserv Sci. 2009.
- View Article
- Google Scholar
26. Simmons N., Cirranello AL. Bat species of the world: A taxonomic and geographic database. Version 1.6. 2024. Available from: https://www.batnames.org/
27. IUCN. The IUCN Red List of Threatened Species. Version 2017-3. 2023 [cited 2 Jun 2023]. Available from: https://www.iucnredlist.org
28. Thong VD, Denzinger A, Sang NV, Huyen NTT, Thanh HT, Loi DN, et al. Bat Diversity in Cat Ba Biosphere Reserve, Northeastern Vietnam: A Review with New Records from Mangrove Ecosystem. Diversity. 2021;13(8):376.
- View Article
- Google Scholar
29. Furey NM, Mackie IJ, Racey PA. Bat diversity in Vietnamese limestone karst areas and the implications of forest degradation. Biodivers Conserv. 2010;19(7):1821–38.
- View Article
- Google Scholar
30. Furey NM, Mackie IJ, Racey PA. Reproductive phenology of bat assemblages in Vietnamese Karst and its conservation implications. Acta Chiropterologica. 2011;13(2):341–54.
- View Article
- Google Scholar
31. Tu VT, Estók P, Csorba G, Son NT, Thanh HT, Tuan LQ, et al. Recent remarkable records reveal that Phia Oac-Phia Den Nature Reserve is a priority area for bat conservation in Northern Vietnam. J Asia-Pac Biodivers. 2016;9(3):312–22.
- View Article
- Google Scholar
32. Van NK, Hien NT, Loc PK, Hiep NT. Bioclimatic diagrams of Vietnam. Van NK, editor. Hanoi National University Publisher; 2000.
33. ESA. Land Cover CCI Product User Guide Version 2. In: Tech. Rep [Internet]. 2017. Available from: maps.elie.ucl.ac.be/CCI/viewer/download/ESACCI-LC-Ph2-PUGv2_2.0.pdf
34. UNEP-WCMC and IUCN. Protected Planet: The World Database on Protected Areas (WDPA), downloaded on November, 2023, Cambridge, UK: UNEP-WCMC and IUCN. 2023. Available from: www.protectedplanet.net
35. Tuan LQ, Thong VD, Son NT, Tu VT, Tuan TA, Luong NT, et al. Potential individual and interactive effects of climate and land-cover changes on bats and implications for conservation planning: a case study in Vietnam. Biodivers Conserv. 2023;32(13):4481–508.
- View Article
- Google Scholar
36. Karger D, Conrad O, Böhner J, Kawohl T, Kreft H, Soria-Auza R, et al. Data from: Climatologies at high resolution for the earth’s land surface areas. EnviDat. 2018.
- View Article
- Google Scholar
37. Booth TH, Nix HA, Busby JR, Hutchinson MF. Bioclim: the first species distribution modelling package, its early applications and relevance to most current MaxEnt studies. Divers Distribut. 2013;20(1):1–9.
- View Article
- Google Scholar
38. Burles DW, Brigham RM, Ring RA, Reimchen TE. Influence of weather on two insectivorous bats in a temperate Pacific Northwest rainforest. Can J Zool. 2009;87(2):132–8.
- View Article
- Google Scholar
39. Turbill C. Winter activity of Australian tree‐roosting bats: influence of temperature and climatic patterns. J Zool. 2008;276(3):285–90.
- View Article
- Google Scholar
40. Williams Ford YTF. World map of carbonate rock outcrops v3.0. 2016 [cited 5 Mar 2020]. Available from: https://digital.lib.usf.edu/SFS0055342/00001
41. Phillips SJ, Anderson RP, Schapire RE. Maximum entropy modeling of species geographic distributions. Ecol Modell. 2006;190(3–4):231–59.
- View Article
- Google Scholar
42. Phillips SJ, Anderson RP, Dudík M, Schapire RE, Blair ME. Opening the black box: an open‐source release of Maxent. Ecography. 2017;40(7):887–93.
- View Article
- Google Scholar
43. Phillips SJ, Dudík M, Elith J, Graham CH, Lehmann A, Leathwick J, et al. Sample selection bias and presence-only distribution models: implications for background and pseudo-absence data. Ecol Appl. 2009;19(1):181–97. pmid:19323182
- View Article
- PubMed/NCBI
- Google Scholar
44. Duan R-Y, Kong X-Q, Huang M-Y, Fan W-Y, Wang Z-G. The predictive performance and stability of six species distribution models. PLoS One. 2014;9(11):e112764. pmid:25383906
- View Article
- PubMed/NCBI
- Google Scholar
45. Hernandez PA, Graham CH, Master LL, Albert DL. The effect of sample size and species characteristics on performance of different species distribution modeling methods. Ecography. 2006;29(5):773–85.
- View Article
- Google Scholar
46. Roberts DR, Bahn V, Ciuti S, Boyce MS, Elith J, Guillera‐Arroita G, et al. Cross‐validation strategies for data with temporal, spatial, hierarchical, or phylogenetic structure. Ecography. 2017;40(8):913–29.
- View Article
- Google Scholar
47. Cao Y, DeWalt RE, Robinson JL, Tweddale T, Hinz L, Pessino M. Using Maxent to model the historic distributions of stonefly species in Illinois streams: The effects of regularization and threshold selections. Ecol Modell. 2013;259:30–9.
- View Article
- Google Scholar
48. Hijmans RJ, Phillips S, Leathwick J, Maintainer JE. Package “dismo” type package title species distribution modeling. 2017. Available from: https://cran.r-project.org/web/packages/dismo/dismo.pdf
49. Valavi R, Elith J, Lahoz‐Monfort JJ, Guillera‐Arroita G. blockCV: An r package for generating spatially or environmentally separated folds for k‐fold cross‐validation of species distribution models. Methods Ecol Evol. 2018;10(2):225–32.
- View Article
- Google Scholar
50. Moilanen A, Franco AMA, Early RI, Fox R, Wintle B, Thomas CD. Prioritizing multiple-use landscapes for conservation: methods for large multi-species planning problems. Proc Biol Sci. 2005;272(1575):1885–91. pmid:16191593
- View Article
- PubMed/NCBI
- Google Scholar
51. Kullberg P, Toivonen T, Montesino Pouzols F, Lehtomäki J, Di Minin E, Moilanen A. Complementarity and area-efficiency in the prioritization of the global protected area network. PLoS One. 2015;10(12):e0145231. pmid:26678497
- View Article
- PubMed/NCBI
- Google Scholar
52. Whitehead AL, Kujala H, Ives CD, Gordon A, Lentini PE, Wintle BA, et al. Integrating biological and social values when prioritizing places for biodiversity conservation. Conserv Biol. 2014;28(4):992–1003. pmid:24617898
- View Article
- PubMed/NCBI
- Google Scholar
53. Moilanen A, Pouzols FM, Meller L, Veach V, Arponen A, Leppanen J. Zonation - User Manual. 2014. Available from: http://cbig.it.helsink
54. Moilanen A, Wintle BA. The boundary-quality penalty: a quantitative method for approximating species responses to fragmentation in reserve selection. Conserv Biol. 2007;21(2):355–64. pmid:17391186
- View Article
- PubMed/NCBI
- Google Scholar
55. Prime Minister. 149/QD-TTg. 2022. Available from: www.aging-us.com
56. Secretariat CBD. Global biodiversity framework. 2023. https://www.cbd.int/gbf
57. Cantú-Salazar L, Orme CDL, Rasmussen PC, Blackburn TM, Gaston KJ. The performance of the global protected area system in capturing vertebrate geographic ranges. Biodivers Conserv. 2013;22(4):1033–47.
- View Article
- Google Scholar
58. González-Maya JF, Víquez-R LR, Belant JL, Ceballos G. Effectiveness of protected areas for representing species and populations of terrestrial mammals in Costa Rica. PLoS One. 2015;10(5):e0124480. pmid:25970293
- View Article
- PubMed/NCBI
- Google Scholar
59. Tena E, Tellería JL. Modelling the distribution of bat activity areas for conservation in a Mediterranean mountain range. Anim Conserv. 2021;25(1):65–76.
- View Article
- Google Scholar
60. Prieto-Torres DA, Nori J, Rojas-Soto OR. Identifying priority conservation areas for birds associated to endangered Neotropical dry forests. Biol Conserv. 2018;228:205–14.
- View Article
- Google Scholar
61. Vu XD, Csaplovics E, Marrs C, Nguyen TT. Criteria and indicators to define priority areas for biodiversity conservation in Vietnam. Forests. 2022;13(9):1341.
- View Article
- Google Scholar
62. Asaad I, Lundquist CJ, Erdmann MV, Costello MJ. Ecological criteria to identify areas for biodiversity conservation. Biol Conserv. 2017;213:309–16.
- View Article
- Google Scholar
63. Nhan HTT, Tinh TK, Hung PV. Current situation of nature reserved area management in Vietnam. Hanoi: Vietnam National University; 2015.
64. Höffner H, Nguyen ST, Dang PH, Motokawa M, Oshida T. Conservation priorities for threatened mammals of Vietnam: Implementation of the IUCN’s One Plan Approach. 2024. pp. 1–20.
- View Article
- Google Scholar
65. Do HTT, Grant JC, Trinh BN, Zimmer HC, Nichols JD. Diversity depends on scale in the forests of the Central Highlands of Vietnam. J Asia-Pac Biodivers. 2017;10:472–88.
- View Article
- Google Scholar
66. Pham NT, To VQ, Lohman DJ, Monastyrskii AL. High Species Richness and Endemism Characterize the Butterfly Fauna of Vietnam’s Central Highlands (Lepidoptera, Papilionoidea). J Lepid Soc. 2022;76(1).
- View Article
- Google Scholar
67. Tordoff AW, R M, Bezuijen J, Duckworth W, Fellowes JR, Koenig K. Ecosystem profile: Indo-Burma biodiversity hotspot 2011 update. 2012.
68. Tran DX, Tran TV, Pearson D, Myint SW, Lowry J, Nguyen TT. Spatiotemporal analysis of forest cover change and associated environmental challenges: a case study in the Central Highlands of Vietnam. Geocarto Int. 2021;37(25):9277–97.
- View Article
- Google Scholar
69. Cadotte MW, Tucker CM. Difficult decisions: Strategies for conservation prioritization when taxonomic, phylogenetic and functional diversity are not spatially congruent. Biol Conserv. 2018;225:128–33.
- View Article
- Google Scholar
70. Arponen A. Prioritizing species for conservation planning. Biodivers Conserv. 2012;21(4):875–93.
- View Article
- Google Scholar
71. Redding DW, Mooers AØ. Incorporating evolutionary measures into conservation prioritization. Conserv Biol. 2006;20(6):1670–8. pmid:17181802
- View Article
- PubMed/NCBI
- Google Scholar

[ref1] 1. Ghanem SJ, Voigt CC. Increasing awareness of ecosystem services provided by bats. Adv Study Behav. 2012;44:279–302.
View Article
Google Scholar

[2] View Article

[3] Google Scholar

[ref2] 2. Russo D, Salinas-Ramos VB, Cistrone L, Smeraldo S, Bosso L, Ancillotto L. Do we need to use bats as bioindicators? Biology (Basel). 2021;10(8):693. pmid:34439926
View Article
PubMed/NCBI
Google Scholar

[5] View Article

[6] PubMed/NCBI

[7] Google Scholar

[ref3] 3. Frick WF, Kingston T, Flanders J. A review of the major threats and challenges to global bat conservation. Ann N Y Acad Sci. 2020;1469(1):5–25. pmid:30937915
View Article
PubMed/NCBI
Google Scholar

[9] View Article

[10] PubMed/NCBI

[11] Google Scholar

[ref4] 4. Mickleburgh S, Waylen K, Racey P. Bats as bushmeat: a global review. Oryx. 2009;43(2):217–34.
View Article
Google Scholar

[13] View Article

[14] Google Scholar

[ref5] 5. Festa F, Ancillotto L, Santini L, Pacifici M, Rocha R, Toshkova N, et al. Bat responses to climate change: a systematic review. Biol Rev Camb Philos Soc. 2023;98(1):19–33. pmid:36054527
View Article
PubMed/NCBI
Google Scholar

[16] View Article

[17] PubMed/NCBI

[18] Google Scholar

[ref6] 6. Leverington F, Costa KL, Pavese H, Lisle A, Hockings M. A global analysis of protected area management effectiveness. Environ Manage. 2010;46(5):685–98. pmid:20859627
View Article
PubMed/NCBI
Google Scholar

[20] View Article

[21] PubMed/NCBI

[22] Google Scholar

[ref7] 7. Watson JEM, Dudley N, Segan DB, Hockings M. The performance and potential of protected areas. Nature. 2014;515(7525):67–73. pmid:25373676
View Article
PubMed/NCBI
Google Scholar

[24] View Article

[25] PubMed/NCBI

[26] Google Scholar

[ref8] 8. Borrini G, Dudley N, Jaeger T, Lassen B, Pathak N, Phillips A. Governance of protected areas: from understanding to action. Best Pract Prot Area Guidel Ser. 2013.

[ref9] 9. Brum FT, Graham CH, Costa GC, Hedges SB, Penone C, Radeloff VC, et al. Global priorities for conservation across multiple dimensions of mammalian diversity. Proc Natl Acad Sci U S A. 2017;114(29):7641–6. pmid:28674013
View Article
PubMed/NCBI
Google Scholar

[29] View Article

[30] PubMed/NCBI

[31] Google Scholar

[ref10] 10. Chauvenet ALM, Kuempel CD, McGowan J, Beger M, Possingham HP. Methods for calculating protection equality for conservation planning. PLoS One. 2017;12(2):e0171591. pmid:28199341
View Article
PubMed/NCBI
Google Scholar

[33] View Article

[34] PubMed/NCBI

[35] Google Scholar

[ref11] 11. Smith A, Schoeman MC, Keith M, Erasmus BFN, Monadjem A, Moilanen A, et al. Synergistic effects of climate and land-use change on representation of African bats in priority conservation areas. Ecol Indic. 2016;69:276–83.
View Article
Google Scholar

[37] View Article

[38] Google Scholar

[ref12] 12. Pessoa da Silva F, De Marco Júnior P. The role of protected areas in safeguarding bat diversity in Brazil. Biodivers Conserv. 2023;32(8–9):2909–24.
View Article
Google Scholar

[40] View Article

[41] Google Scholar

[ref13] 13. Rodrigues ASL, Andelman SJ, Bakarr MI, Boitani L, Brooks TM, Cowling RM, et al. Effectiveness of the global protected area network in representing species diversity. Nature. 2004;428(6983):640–3. pmid:15071592
View Article
PubMed/NCBI
Google Scholar

[43] View Article

[44] PubMed/NCBI

[45] Google Scholar

[ref14] 14. Coad L, Watson JE, Geldmann J, Burgess ND, Leverington F, Hockings M, et al. Widespread shortfalls in protected area resourcing undermine efforts to conserve biodiversity. Front Ecol Environ. 2019;17(5):259–64.
View Article
Google Scholar

[47] View Article

[48] Google Scholar

[ref15] 15. Delso Á, Fajardo J, Muñoz J. Protected area networks do not represent unseen biodiversity. Sci Rep. 2021;11(1):12275. pmid:34112867
View Article
PubMed/NCBI
Google Scholar

[50] View Article

[51] PubMed/NCBI

[52] Google Scholar

[ref16] 16. Butchart SHM, Scharlemann JPW, Evans MI, Quader S, Aricò S, Arinaitwe J, et al. Protecting important sites for biodiversity contributes to meeting global conservation targets. PLoS One. 2012;7(3):e32529. pmid:22457717
View Article
PubMed/NCBI
Google Scholar

[54] View Article

[55] PubMed/NCBI

[56] Google Scholar

[ref17] 17. Laurance WF, Useche DC, Rendeiro J, Kalka M, Bradshaw CJA, Sloan SP, et al. Averting biodiversity collapse in tropical forest protected areas. Nature. 2012;489(7415):290–4. pmid:22832582
View Article
PubMed/NCBI
Google Scholar

[58] View Article

[59] PubMed/NCBI

[60] Google Scholar

[ref18] 18. Eigenbrod F, Gonzalez P, Dash J, Steyl I. Vulnerability of ecosystems to climate change moderated by habitat intactness. Glob Chang Biol. 2015;21(1):275–86. pmid:25059822
View Article
PubMed/NCBI
Google Scholar

[62] View Article

[63] PubMed/NCBI

[64] Google Scholar

[ref19] 19. Secretariat CBD. Global biodiversity framework. 2023. Available from: https://www.cbd.int/gbf/targets

[ref20] 20. Zamora‐Gutierrez V, Pearson RG, Green RE, Jones KE. Forecasting the combined effects of climate and land use change on Mexican bats. Divers Distrib. 2017;24(3):363–74.
View Article
Google Scholar

[67] View Article

[68] Google Scholar

[ref21] 21. Moilanen A, Leathwick JR, Quinn JM. Spatial prioritization of conservation management. Conserv Lett. 2011;4(5):383–93.
View Article
Google Scholar

[70] View Article

[71] Google Scholar

[ref22] 22. Giakoumi S, Richardson AJ, Doxa A, Moro S, Andrello M, Hanson JO, et al. Advances in systematic conservation planning to meet global biodiversity goals. Trends Ecol Evol. 2025;40(4):395–410. pmid:39880725
View Article
PubMed/NCBI
Google Scholar

[73] View Article

[74] PubMed/NCBI

[75] Google Scholar

[ref23] 23. Sinclair SP, Milner‐Gulland EJ, Smith RJ, McIntosh EJ, Possingham HP, Vercammen A, et al. The use, and usefulness, of spatial conservation prioritizations. Conserv Lett. 2018;11(6).
View Article
Google Scholar

[77] View Article

[78] Google Scholar

[ref24] 24. Lagabrielle E, Lombard AT, Harris JM, Livingstone T-C. Multi-scale multi-level marine spatial planning: a novel methodological approach applied in South Africa. PLoS One. 2018;13(7):e0192582. pmid:29969441
View Article
PubMed/NCBI
Google Scholar

[80] View Article

[81] PubMed/NCBI

[82] Google Scholar

[ref25] 25. Cayuela L, Golicher DJ, Newton AC, Kolb M, De Alburquerque FS, Arets EJMM. Species distribution modeling in the tropics: problems, potentialities, and the role of biological data for effective species conservation. Tropical Conserv Sci. 2009.
View Article
Google Scholar

[84] View Article

[85] Google Scholar

[ref26] 26. Simmons N., Cirranello AL. Bat species of the world: A taxonomic and geographic database. Version 1.6. 2024. Available from: https://www.batnames.org/

[ref27] 27. IUCN. The IUCN Red List of Threatened Species. Version 2017-3. 2023 [cited 2 Jun 2023]. Available from: https://www.iucnredlist.org

[ref28] 28. Thong VD, Denzinger A, Sang NV, Huyen NTT, Thanh HT, Loi DN, et al. Bat Diversity in Cat Ba Biosphere Reserve, Northeastern Vietnam: A Review with New Records from Mangrove Ecosystem. Diversity. 2021;13(8):376.
View Article
Google Scholar

[89] View Article

[90] Google Scholar

[ref29] 29. Furey NM, Mackie IJ, Racey PA. Bat diversity in Vietnamese limestone karst areas and the implications of forest degradation. Biodivers Conserv. 2010;19(7):1821–38.
View Article
Google Scholar

[92] View Article

[93] Google Scholar

[ref30] 30. Furey NM, Mackie IJ, Racey PA. Reproductive phenology of bat assemblages in Vietnamese Karst and its conservation implications. Acta Chiropterologica. 2011;13(2):341–54.
View Article
Google Scholar

[95] View Article

[96] Google Scholar

[ref31] 31. Tu VT, Estók P, Csorba G, Son NT, Thanh HT, Tuan LQ, et al. Recent remarkable records reveal that Phia Oac-Phia Den Nature Reserve is a priority area for bat conservation in Northern Vietnam. J Asia-Pac Biodivers. 2016;9(3):312–22.
View Article
Google Scholar

[98] View Article

[99] Google Scholar

[ref32] 32. Van NK, Hien NT, Loc PK, Hiep NT. Bioclimatic diagrams of Vietnam. Van NK, editor. Hanoi National University Publisher; 2000.

[ref33] 33. ESA. Land Cover CCI Product User Guide Version 2. In: Tech. Rep [Internet]. 2017. Available from: maps.elie.ucl.ac.be/CCI/viewer/download/ESACCI-LC-Ph2-PUGv2_2.0.pdf

[ref34] 34. UNEP-WCMC and IUCN. Protected Planet: The World Database on Protected Areas (WDPA), downloaded on November, 2023, Cambridge, UK: UNEP-WCMC and IUCN. 2023. Available from: www.protectedplanet.net

[ref35] 35. Tuan LQ, Thong VD, Son NT, Tu VT, Tuan TA, Luong NT, et al. Potential individual and interactive effects of climate and land-cover changes on bats and implications for conservation planning: a case study in Vietnam. Biodivers Conserv. 2023;32(13):4481–508.
View Article
Google Scholar

[104] View Article

[105] Google Scholar

[ref36] 36. Karger D, Conrad O, Böhner J, Kawohl T, Kreft H, Soria-Auza R, et al. Data from: Climatologies at high resolution for the earth’s land surface areas. EnviDat. 2018.
View Article
Google Scholar

[107] View Article

[108] Google Scholar

[ref37] 37. Booth TH, Nix HA, Busby JR, Hutchinson MF. Bioclim: the first species distribution modelling package, its early applications and relevance to most current MaxEnt studies. Divers Distribut. 2013;20(1):1–9.
View Article
Google Scholar

[110] View Article

[111] Google Scholar

[ref38] 38. Burles DW, Brigham RM, Ring RA, Reimchen TE. Influence of weather on two insectivorous bats in a temperate Pacific Northwest rainforest. Can J Zool. 2009;87(2):132–8.
View Article
Google Scholar

[113] View Article

[114] Google Scholar

[ref39] 39. Turbill C. Winter activity of Australian tree‐roosting bats: influence of temperature and climatic patterns. J Zool. 2008;276(3):285–90.
View Article
Google Scholar

[116] View Article

[117] Google Scholar

[ref40] 40. Williams Ford YTF. World map of carbonate rock outcrops v3.0. 2016 [cited 5 Mar 2020]. Available from: https://digital.lib.usf.edu/SFS0055342/00001

[ref41] 41. Phillips SJ, Anderson RP, Schapire RE. Maximum entropy modeling of species geographic distributions. Ecol Modell. 2006;190(3–4):231–59.
View Article
Google Scholar

[120] View Article

[121] Google Scholar

[ref42] 42. Phillips SJ, Anderson RP, Dudík M, Schapire RE, Blair ME. Opening the black box: an open‐source release of Maxent. Ecography. 2017;40(7):887–93.
View Article
Google Scholar

[123] View Article

[124] Google Scholar

[ref43] 43. Phillips SJ, Dudík M, Elith J, Graham CH, Lehmann A, Leathwick J, et al. Sample selection bias and presence-only distribution models: implications for background and pseudo-absence data. Ecol Appl. 2009;19(1):181–97. pmid:19323182
View Article
PubMed/NCBI
Google Scholar

[126] View Article

[127] PubMed/NCBI

[128] Google Scholar

[ref44] 44. Duan R-Y, Kong X-Q, Huang M-Y, Fan W-Y, Wang Z-G. The predictive performance and stability of six species distribution models. PLoS One. 2014;9(11):e112764. pmid:25383906
View Article
PubMed/NCBI
Google Scholar

[130] View Article

[131] PubMed/NCBI

[132] Google Scholar

[ref45] 45. Hernandez PA, Graham CH, Master LL, Albert DL. The effect of sample size and species characteristics on performance of different species distribution modeling methods. Ecography. 2006;29(5):773–85.
View Article
Google Scholar

[134] View Article

[135] Google Scholar

[ref46] 46. Roberts DR, Bahn V, Ciuti S, Boyce MS, Elith J, Guillera‐Arroita G, et al. Cross‐validation strategies for data with temporal, spatial, hierarchical, or phylogenetic structure. Ecography. 2017;40(8):913–29.
View Article
Google Scholar

[137] View Article

[138] Google Scholar

[ref47] 47. Cao Y, DeWalt RE, Robinson JL, Tweddale T, Hinz L, Pessino M. Using Maxent to model the historic distributions of stonefly species in Illinois streams: The effects of regularization and threshold selections. Ecol Modell. 2013;259:30–9.
View Article
Google Scholar

[140] View Article

[141] Google Scholar

[ref48] 48. Hijmans RJ, Phillips S, Leathwick J, Maintainer JE. Package “dismo” type package title species distribution modeling. 2017. Available from: https://cran.r-project.org/web/packages/dismo/dismo.pdf

[ref49] 49. Valavi R, Elith J, Lahoz‐Monfort JJ, Guillera‐Arroita G. blockCV: An r package for generating spatially or environmentally separated folds for k‐fold cross‐validation of species distribution models. Methods Ecol Evol. 2018;10(2):225–32.
View Article
Google Scholar

[144] View Article

[145] Google Scholar

[ref50] 50. Moilanen A, Franco AMA, Early RI, Fox R, Wintle B, Thomas CD. Prioritizing multiple-use landscapes for conservation: methods for large multi-species planning problems. Proc Biol Sci. 2005;272(1575):1885–91. pmid:16191593
View Article
PubMed/NCBI
Google Scholar

[147] View Article

[148] PubMed/NCBI

[149] Google Scholar

[ref51] 51. Kullberg P, Toivonen T, Montesino Pouzols F, Lehtomäki J, Di Minin E, Moilanen A. Complementarity and area-efficiency in the prioritization of the global protected area network. PLoS One. 2015;10(12):e0145231. pmid:26678497
View Article
PubMed/NCBI
Google Scholar

[151] View Article

[152] PubMed/NCBI

[153] Google Scholar

[ref52] 52. Whitehead AL, Kujala H, Ives CD, Gordon A, Lentini PE, Wintle BA, et al. Integrating biological and social values when prioritizing places for biodiversity conservation. Conserv Biol. 2014;28(4):992–1003. pmid:24617898
View Article
PubMed/NCBI
Google Scholar

[155] View Article

[156] PubMed/NCBI

[157] Google Scholar

[ref53] 53. Moilanen A, Pouzols FM, Meller L, Veach V, Arponen A, Leppanen J. Zonation - User Manual. 2014. Available from: http://cbig.it.helsink

[ref54] 54. Moilanen A, Wintle BA. The boundary-quality penalty: a quantitative method for approximating species responses to fragmentation in reserve selection. Conserv Biol. 2007;21(2):355–64. pmid:17391186
View Article
PubMed/NCBI
Google Scholar

[160] View Article

[161] PubMed/NCBI

[162] Google Scholar

[ref55] 55. Prime Minister. 149/QD-TTg. 2022. Available from: www.aging-us.com

[ref56] 56. Secretariat CBD. Global biodiversity framework. 2023. https://www.cbd.int/gbf

[ref57] 57. Cantú-Salazar L, Orme CDL, Rasmussen PC, Blackburn TM, Gaston KJ. The performance of the global protected area system in capturing vertebrate geographic ranges. Biodivers Conserv. 2013;22(4):1033–47.
View Article
Google Scholar

[166] View Article

[167] Google Scholar

[ref58] 58. González-Maya JF, Víquez-R LR, Belant JL, Ceballos G. Effectiveness of protected areas for representing species and populations of terrestrial mammals in Costa Rica. PLoS One. 2015;10(5):e0124480. pmid:25970293
View Article
PubMed/NCBI
Google Scholar

[169] View Article

[170] PubMed/NCBI

[171] Google Scholar

[ref59] 59. Tena E, Tellería JL. Modelling the distribution of bat activity areas for conservation in a Mediterranean mountain range. Anim Conserv. 2021;25(1):65–76.
View Article
Google Scholar

[173] View Article

[174] Google Scholar

[ref60] 60. Prieto-Torres DA, Nori J, Rojas-Soto OR. Identifying priority conservation areas for birds associated to endangered Neotropical dry forests. Biol Conserv. 2018;228:205–14.
View Article
Google Scholar

[176] View Article

[177] Google Scholar

[ref61] 61. Vu XD, Csaplovics E, Marrs C, Nguyen TT. Criteria and indicators to define priority areas for biodiversity conservation in Vietnam. Forests. 2022;13(9):1341.
View Article
Google Scholar

[179] View Article

[180] Google Scholar

[ref62] 62. Asaad I, Lundquist CJ, Erdmann MV, Costello MJ. Ecological criteria to identify areas for biodiversity conservation. Biol Conserv. 2017;213:309–16.
View Article
Google Scholar

[182] View Article

[183] Google Scholar

[ref63] 63. Nhan HTT, Tinh TK, Hung PV. Current situation of nature reserved area management in Vietnam. Hanoi: Vietnam National University; 2015.

[ref64] 64. Höffner H, Nguyen ST, Dang PH, Motokawa M, Oshida T. Conservation priorities for threatened mammals of Vietnam: Implementation of the IUCN’s One Plan Approach. 2024. pp. 1–20.
View Article
Google Scholar

[186] View Article

[187] Google Scholar

[ref65] 65. Do HTT, Grant JC, Trinh BN, Zimmer HC, Nichols JD. Diversity depends on scale in the forests of the Central Highlands of Vietnam. J Asia-Pac Biodivers. 2017;10:472–88.
View Article
Google Scholar

[189] View Article

[190] Google Scholar

[ref66] 66. Pham NT, To VQ, Lohman DJ, Monastyrskii AL. High Species Richness and Endemism Characterize the Butterfly Fauna of Vietnam’s Central Highlands (Lepidoptera, Papilionoidea). J Lepid Soc. 2022;76(1).
View Article
Google Scholar

[192] View Article

[193] Google Scholar

[ref67] 67. Tordoff AW, R M, Bezuijen J, Duckworth W, Fellowes JR, Koenig K. Ecosystem profile: Indo-Burma biodiversity hotspot 2011 update. 2012.

[ref68] 68. Tran DX, Tran TV, Pearson D, Myint SW, Lowry J, Nguyen TT. Spatiotemporal analysis of forest cover change and associated environmental challenges: a case study in the Central Highlands of Vietnam. Geocarto Int. 2021;37(25):9277–97.
View Article
Google Scholar

[196] View Article

[197] Google Scholar

[ref69] 69. Cadotte MW, Tucker CM. Difficult decisions: Strategies for conservation prioritization when taxonomic, phylogenetic and functional diversity are not spatially congruent. Biol Conserv. 2018;225:128–33.
View Article
Google Scholar

[199] View Article

[200] Google Scholar

[ref70] 70. Arponen A. Prioritizing species for conservation planning. Biodivers Conserv. 2012;21(4):875–93.
View Article
Google Scholar

[202] View Article

[203] Google Scholar

[ref71] 71. Redding DW, Mooers AØ. Incorporating evolutionary measures into conservation prioritization. Conserv Biol. 2006;20(6):1670–8. pmid:17181802
View Article
PubMed/NCBI
Google Scholar

[205] View Article

[206] PubMed/NCBI

[207] Google Scholar

Figures

Abstract

Introduction

Materials and methods

Study area

Bat occurrence data

Environmental parameters

Climate data.

Land cover data.

Karst data.

Species distribution modelling.

Spatial prioritization analyses.

Area representativeness for bats.

Results

Predictions of bat distributions

The representativeness of current protected areas for bats

Representativeness of prioritized areas for bats

Discussion

Supporting information

S1 Table. Bat species included in the study.

S2 Table. The contributions of environmental variables in the model of individual species.

S3 Table. The number and total area of protected areas in Vietnam by the IUCN category.

S1 Fig. Pairwise correlations of the six bioclimatic variables used for modeling the distributions of Vietnamese bat species.

S2 Fig. The AUC values of the distribution models built for individual bat species with different combinations of feature types and regularization values.

S3 Fig. AUC values of the distribution models built for the bat species in Vietnam.

S4 Fig. The percent contributions of the environmental variables across the bat species distribution models.

S5 Fig. Representativeness of current protected areas, and the prioritized areas with the conservation targets of 9% and 30% land coverage.

S6 Fig. Distribution of prioritized conservation areas for bat species in Vietnam identified by the spatial prioritization analysis.

Acknowledgments

References