https://orcid.org/0000-0001-8078-4172
Figures
Abstract
Acoustic monitoring is emerging as a key tool in wildlife conservation, especially for species in inaccessible habitats like the rock cavy (Kerodon rupestris), an endangered species native to Brazil’s threatened Caatinga biome. Emotional stress from threatening situations affects breathing, heart rate, and vocal muscle tension, altering vocal acoustic parameters. This allows researchers to gauge the animal’s environmental perception through its vocalizations. We aimed to evaluate emotional disturbance indicators in free-range rock cavies’ vocalizations to suggest an acoustic index during threats. We compared calls from rock cavies in two areas with similar habitats but that differ in terms of anthropic impacts. Area 1 (A1) is near urban areas and disturbed by livestock and dogs, and Area 2 (A2) is farther from urban areas and free from human disturbance. Data on calls and behaviors were collected ad libitum in both areas. The alarm whistle call, making up 73.5% of total calls, was most common. Across 108 observation hours per area, 392 alarm whistle calls were recorded, with more calls in A1 than A2 (223 vs. 169; Chi-square = 29.44, DF = 1, P < 0.001). This resulted in a 32% higher hourly call rate in A1 (2.6 calls/h vs. 1.6 calls/h). Both male and female cavies in A1 had higher high-frequency (F1, 388 = 7.80, P = 0.005) and peak-frequency calls (F1, 388 = 21.32, P < 0.001). Given the similar landscape and resource availability in both areas, the differences in call emission rate and parameters are likely linked to emotional responses to human disturbances in A1. Thus, alarm whistle calls at an hourly rate of 2.6 calls/h or higher, with high-frequency and peak-frequency at or above 7222 Hz and 2603 Hz, can indicate anthropogenic disturbance in the Caatinga biome, aiding remote monitoring efforts.
Citation: Almeida WN, Barros KS, Nogueira-Filho SLG, Nogueira SSC (2025) Assessing anthropogenic impact on the habitat of threatened rock cavy (Kerodon rupestris) through its alarm calls. PLoS One 20(5): e0323711. https://doi.org/10.1371/journal.pone.0323711
Editor: Vitor Hugo Rodrigues Paiva, MARE – Marine and Environmental Sciences Centre, PORTUGAL
Received: October 25, 2024; Accepted: April 13, 2025; Published: May 30, 2025
Copyright: © 2025 Almeida 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 raw data were included in this submission as Supplementary Information.
Funding: Coordination for the Improvement of Higher Education Personnel (CAPES - Finance Code 001, WNA Grant, Process #88882.451299/2019-01 #8888485144/2020-00). and National Institute of Science and Technology in Interdisciplinary and Transdisciplinary Studies in Ecology and Evolution (IN-TREE – Process CNPq #465767/ 2014-1 and CAPES #23038.000776/2017-54), Bahia, Brazil; SSCN and SLGNF received a grant from the Council for Scientific and Technological Development (CNPq) (Processes # 303448/2019-9 and # 04226/ 2019-0, respectively). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors of the ms ‘Assessing anthropogenic impact on the habitat of threatened rock cavy (Kerodon rupestris) through its alarm calls’ have no actual or potential conflict of interest including any financial, personal or other relationships with other people or organizations within three years of beginning the submitted work that could inappropriately influence, or be perceived to influence, this work.
1. Introduction
The rapid loss of habitats due to anthropogenic causes [1–4] has hastened the decline of Neotropical diversity [5]. To make improvements in the monitoring of free-living animals, studies have explored new technologies such as bioacoustic methods [6–8]. Recent advances in technological and analytical approaches to sounds [9] have allowed researchers to investigate several issues, such as the presence of species [10–13], species richness [14], abundance and density [15].
Acoustic signals have also been used to assess the affective states of wild animals [16–18], both in captivity [19] as well as in a natural wild environment [20,21]. This is possible because when emitting a call an animal uses specific organs such as the larynx, syrinx and vocal fold [16,22], which carry physiological [23], emotional [24,25], and individual information [26,27]. By analyzing call parameters such as frequency, rate, duration, number of notes, and other acoustic features [28], it is possible to assess the perception of the environment as either positive or negative [19], group member recognition [29,30], predator detection [31,32], and the impact of anthropogenic disturbances [33,34].
Alarm calls, for instance, which are reported in several species during threatening situations [35,36], can be widespread in the environment, alerting conspecifics to potential danger [37]. This has suggested acoustic monitoring as a new tool for wildlife conservation [34]. This tool is particularly useful for monitoring species that live in places that are difficult to access. Some studies with the Australasian heron (Botaurus poiciloptilus) [12] and African forest elephant (Loxodonta cyclotis) [8] are examples of how bioacoustic techniques are effective in monitoring species living in environments where it is extremely difficult to move around and access and collect regular ecological data. These new techniques increase research on data-poor species by providing ecological indicators that can be used for their management and conservation purposes [7,33], such as the rock cavy (Kerodon rupestris).
The rock cavy, a rodent from the Caviidae family [38], plays a crucial ecological role in maintaining the diversity of the Caatinga, where it is endemic. The Caatinga, an exclusively Brazilian biome, is currently under threat [39], and Kerodon rupestris is at the base of the food chain of the predators that occur in this region, such as the jaguar (Panthera onca), cougar (Puma concolor), ocelot (Leopardus pardalis), wild fox (Cerdocyon thirty), oncilla (Leopardus tigrinus) and laughing falcon (Herpetotheres cachinnans) [40–42]. Additionally, the species is also used as an alternative source of animal protein by the human populations that inhabit the Caatinga and live below the poverty line [43–45]. Nevertheless, the rock cavy is threatened due to the accelerated process of environmental change and deterioration caused by the unsustainable use of natural resources. This has led to the rapid loss of this endemic species [46]. Due to a 30% population decline over the last 10 years, the species has been included in the vulnerable category [47]. However, monitoring activities to protect the species and its habitat in the area are challenging due to the conditions of the Caatinga. The Caatinga region has a complex topography, consisting of a rocky landscape, which makes it challenging to move around. Additionally, only 7% of the original Caatinga area is protected by conservation units [39]. Environmental stressors caused by human activities, such as habitat degradation, can threaten biodiversity and can also cause suffering to individuals, impacting their welfare [see 48]. Thus, stressor agents affect animals’ acoustic behavior [16,21,48]. It is therefore urgent to conduct studies that identify ecological indicators using non-invasive techniques such as bioacoustics for assessing the rock cavy’s response to environmental disturbances using remote acoustic monitoring.
The species lives in a harem, usually composed of one male and three to four adult females and their offspring [49]. Adapted to a semi-arid environment, it lives in rocky terrain, using rock crevices as shelter [50]. The acoustic repertoire of K. rupestris comprises 11 different types of calls and two mechanical signals [49,51]. Among them, the whistle is an alarm call emitted when the animal is facing threatening situations, such as the presence of a potential predator or during intraspecific agonistic interactions [49,51,52]. The alarm whistle call is thus related to a negative affective state caused by negative events such as predator attack or aggression between conspecifics, both associated with high arousal (e.g., to run). Therefore, we aimed to compare the type and the number of calls, in addition to the acoustic parameters of alarm whistle calls emitted by K. rupestris living in two areas with different degrees of anthropic impact. We expected a higher emission of alarm calls in the area with greater anthropogenic disturbance than in the less disturbed one. Additionally, if high-pitched calls are emitted in hostile contexts [16,53] and are related to negative motivational states in animals [54,55], we also predicted that the calls emitted by rock cavies in the more anthropogenically disturbed area would exhibit higher frequencies compared to those from the less disturbed area.
2. Materials and methods
2.1. Ethics statement
This study was approved by the Chico Mendes Institute for Biodiversity and Conservation (ICMBio/SISBIO) (#59323), the Environment and Water Resources Institute of Bahia (INEMA) (#20065), and the Commission for Ethics in the Use of Animals (CEUA) of the State University of Bahia (UNEB) (Protocol #07/2019).
2.2. Study area and animals
The study was carried out with free-ranging animals in two interconnected conservation units (14°16’ 14°39’S 42°46’ 43°46’W): Serra dos Montes Altos wildlife refuge (REVIS) and Serra dos Montes Altos state park (PESMA), both covering approximately 46,000 Ha. These conservation units are located in the southwestern part of Bahia state, Brazil, and are monitored by the regional environmental agency. Both REVIS and PESMA are characterized by a semi-arid climate, an ecotone transition area between herbaceous and arboreal caatinga vegetation, rupestrian fields and gallery forests [56]. Topographically, both conservation units are characterized by the São Francisco River basin depression and a sedimentary structural plateau – the Espinhaço Plateau [57]. There are several endemic and rare species living in these areas, highlighting the importance of maintaining this habitat [58]. These conservation units contain two distinct sampling areas, named Serra dos Montes Altos wildlife refuge (A1) and Serra dos Montes Altos state park (A2), located within 15 km of each other.
Both areas (A1 and A2) share similar natural landscapes. They are characterized by large rock formations with dystrophic litholic neosol and dystrophic red-yellow latosol pedology [59], proximity to springs and bodies of water, and a predominance of shrubby and herbaceous vegetation strata typical of the Caatinga, composed of dry branches and thorny plants, with overall vegetation cover classified as savanna formation [60]. They present similar hypsometric data, ranging from 750 m to 1,150 m [61], and show no signs of recent wildfires or other severe anthropogenic alterations in the landscape. The availability of water and shelter is also similar. However, A1 (86,675.03 m², 14°19’07” S 43°03’42” W) is located inside REVIS and was characterized as the most disturbed area due to its hunting history. Additionally, this area is 630 m away from the urban zone, and shows anthropogenic changes, such as drinking troughs arranged along the trails for domestic animals that usually pass by. There are fences delimiting the area for animals, the presence of feral and domestic dogs, small livestock production, farmers, and a pipe network for water distribution. In contrast, A2 (401,212.48 m²; 14°25’45” S 42°57’51” W) was characterized as the less disturbed area, mainly because it receives few visits due to difficult access. This area is located inside PESMA and 6.1 km away from the nearest urban area. In this area there are no drinking troughs, fences, livestock, domestic or feral dogs, characterized as the opposite of A1.
The sample areas were chosen during the 10-day pilot period that occurred prior to data collection. During this period, we conducted indirect searches for animal traces, such as footprints and feces, and registered general activities and behaviors of animals (see Table 1). The rock cavy has latrine behavior, which makes its detection easier in the wild [50]. Moreover, the park rangers, who monitor the security of these sites and usually sight the rock cavy groups, helped to determine the sample areas. During data collection we recorded quite a similar number of rock cavies in both areas (A1 = 91 and A2 = 97).
2.3. Data collection
The study was carried out from November 2019 to March 2020, with a total of 24 expeditions (A1: 12 A2: 12). We interrupted our data collection in March 2020 due to the COVID-19 pandemic, when all conservation units and national reserves in Brazil were closed to visitors. All acoustic calls and associated behaviors were collected simultaneously (Table 1) by ad libitum method [62]. One observer (W. N. A.) and a field assistant monitored and registered the emissions of animals’ calls and described the associated behaviors. During the field period, the expeditions were interspersed between the two areas, two days in A1 and two days in A2 and so on. Given that there are conflicting findings regarding the species’ peak activity period [48,63–65], the observation sessions took place between 6h00 and 17h00 (encompassing the crepuscular hours of the early morning), totaling 216 hours of data collection (A1: 108 h and A2: 108 h). The observation session was chosen following reports on the species’ general activity [49,66] and confirmed here during the pilot period.
We used a portable digital Tascam DR-44WL recorder (Tokyo, Japan), coupled to a Sennheiser ME-66 unidirectional microphone (Wedemark, Germany) to record the emitted calls. The recorder settings were mono mode, WAV format, 48 kHz sampling rate and 16-bit resolution. To capture the beginning of the sound emission of the animals, we used the “pre-REC” function programmed in 5s. The observer manually activated the recorder as soon as the individuals emitted a call, without cuts until the call ended.
The observation sites had been previously selected during the pilot period, when the observer chose the locations that provided the best view of the animals. From these sites, we first habituated the animals to the presence of the observer to minimize the observation impact that occurred during the pilot period. During data collection, the observer used trees as observation sites, keeping a distance of approximately 2.0 m to 2.5 m from the animals. There were eight observation sites in A1 and nine observation sites in A2. The observer occupied these observation sites and set up the equipment at least 40 min before the animals returned to their general activities outside their burrows. The identification of the sex of the animal was made from direct visualization of the animal’s genital tract or, if in doubt, the observer recorded the animal as unidentified sex category (Table 2).
2.4. Acoustic analysis
The call emissions of K. rupestris were classified by aural and visual inspection of spectrograms using Raven Pro Software version 1.6 (Cornell Lab of Ornithology, Ithaca, NY). The following settings were used: Hann type window, window size 1351 samples, 90% overlap in time domain, and DFT size 4096. The smallest vocal unit of calls was called a pulse, but it can also be called an element, note or syllable, defined as a continuous sound without internal interruptions, following Barros et al. [29]. For each pulse (element), the following parameters were measured: high-frequency (Hz) (higher frequency threshold), low-frequency (Hz) (lower frequency threshold), peak-frequency (Hz) (frequency that corresponds to the maximum occurrence of power in the selection of elements or phrases), and duration (total duration in seconds) [67]. The emission rate per hour of each call or hourly rate (HR) was calculated by dividing the total number of each call type by the total observation time for each sampled area.
2.5. Data and statistical analysis
First, we described the calls and their associated behaviors to better understand their possible functions (see Table 1). We then quantified the number of call types emitted by rock cavies according to age, caller’s sex and area of occurrence (Table 2). The age classification was based on the observation of individual size, as adults and juveniles are distinguishable by body size [68]. We compared the emission of the different calls according to the sampled area (A1 and A2) using the chi-square test. Following that, we used discriminant function analysis (DFA) [69] of the acoustic parameters (low-frequency, high-frequency, peak-frequency, and duration) to test the classification of the different calls initially categorized by aural and visual inspection. Prior to conducting the DFA analysis, we normalized and standardized the variables by subtracting the mean of each variable from its corresponding data points and dividing the result by the variable’s standard deviation, a process known as Z-score transformation. This standardization was necessary to prevent the spurious attribution of weights to acoustic parameters measured in different units. Thereafter, we performed the DFA cross-validation test to ascertain whether the prediction for each vocal type based on the measured acoustic parameters was correct.
Snort-like and snort calls were emitted exclusively by adult females, while scream and whine calls did not occur very often and were only emitted by juvenile animals in the most disturbed area (A1) (Table 2). In turn, the alarm whistle was the prevalent call, emitted exclusively by adult rock cavies within both sampled areas (Table 2). Therefore, we applied general linear mixed models (GLMs) to verify whether the high-frequency as well as other acoustic parameters (low-frequency, peak-frequency, and duration) varied according to the sampled areas (A1 and A2) just for alarm whistle calls. We applied one model per acoustic parameter (high-frequency, low-frequency, peak-frequency, and duration). In these models we included the sex (male and female), the sampled areas (A1 and A2), and the interaction between them (sex x area) as fixed factors. When the interaction was significant, we used Tukey HSD post hoc tests. We graphically checked the residuals of every model for normal distribution and homoscedasticity, and all data were Box-Cox transformed to meet these assumptions. For all analysis, we used the MINITAB 19.0 program (Minitab Inc.) considering α < 0.05.
3. Results
3.1. Recordings and discrimination of calls
We recorded a total of 616 calls (A1: 387; A2: 229), which were initially aural and visual, inspected the generated spectrograms and classified them into five types of calls (Fig 1, see also S1–S5 Files). K. rupestris emitted these five types of calls while performing four behavioral patterns: alarm, agonistic, social isolation, and pain (Table 1).
The alarm whistle calls accounted for 73.5% and were emitted by adult animals of both sexes in both areas (Table 2). There were, however, a greater number of alarm whistle calls emitted in the more disturbed area (A1) than in the less disturbed area (A2) (Chi-square = 29.44, DF = 1, P < 0.001, Fig 2). However, this disparity between the two areas was entirely attributable to males (Chi-square = 35.57, DF = 1, P < 0.001 and not to females (Chi-square = 0.75, DF = 1, P = 0.387). The total observation time was the same (108 hours) for both sampled areas, which led to an hourly rate (HR) for the alarm whistle call that was 32% higher in A1 than A2 (2.6 calls/h vs 1.6 call/h, respectively).
The second most prevalent call was the snort-like call (20.2%) (A1: HR = 0.7 call/h A2: HR = 0.4 call/h). This call was more frequently emitted in the more disturbed area (A1) compared to the less disturbed area (A2) (Chi-square = 6.73, DF = 1, P = 0.009). The third most prevalent call was the snort call (2.4%) (A1: HR = 0.09 call/h A2: HR = 0.04 call/h) and was similarly emitted in both areas (Chi-square = 1.67, DF = 1, P = 0.197). The snort-like call and the snort call were emitted only by females across the two areas (Table 2). We recorded screams (1.9%) and whines (1.8%) only in the most disturbed area (scream HR = 0.1 call/h and whine HR = 0.1 call/h). Additionally, these last calls were emitted only by juveniles (Table 2).
Feral dogs were observed exclusively in the most disturbed area (A1). During the 12 expeditions conducted in A1, feral dogs were seen near or around the data collection points and roaming the access trails. These dogs posed a threat to K. rupestris, as we witnessed a Canis lupus familiaris catching and killing a rock cavy. In response to the presence of feral dogs, six alarm whistle calls were emitted. Five more alarm whistle calls were emitted in response to humans walking through A1. Additionally, one scream call was recorded due to the presence of feral dogs nearby. During the collection period in A1, two to three dogs were present in each of the alarm whistle call recordings where a dog threat was detected. No natural predators were observed, and there was no evidence of their presence in A1. No dogs or humans were seen in in the less disturbed area (A2). However, during the observation, we detected a direct threat from the natural predator Eira barbara, which triggered the emission of two alarm whistle calls in A2.
The DFA of all calls showed differences among the five call types (Wilks λ = 0.03, F16, 1858 = 245.91, P < 0.0001, DFA = 0.86, Table 3), confirming and validating the initial inspection, with a cross-validation value of 0.86 (see raw data in S6 File). All acoustic parameters used for DFA contributed to differentiating the types of calls signaled (Fig 3).
3.2. Acoustic characteristics of alarm whistle calls according to the sampled areas
The interaction between the sex and area affected the high-frequency type of alarm whistle call (F1, 388 = 7.80, P = 0.005). The post hoc tests showed that both female and male rock cavies living in the most disturbed area (A1) emitted alarm whistles with a higher high-frequency than those living in the less disturbed area (A2). This difference was found to be 1.3 times more pronounced in females than in males (Fig 4a). The interaction between the sex and area also affected the low-frequency (F1, 388 = 148.52, P < 0.001) and the duration of the alarm whistle calls (F1, 388 = 27.14, P < 0.001). The post hoc tests showed that low-frequency calls (Fig 4b) and duration (Fig 4c) only differed in males between both areas (males in A1 emitted longer whistle calls with higher low frequency), than females living in the same area and both females and males living in the less disturbed area (A2). The peak-frequency (F1, 388 = 21.32, P < 0.001) of the rock cavy’s alarm whistle calls was only affected by the sample areas. Rock cavies living in the most disturbed area (A1) emitted alarm whistles with a higher peak-frequency than those living in the less disturbed area (A2) (Fig 4d) (see raw data and complete statistical analysis in – S7 File).
Columns that do not share the same letter differed according to Tukey’s post hoc tests (P < 0.05).
4. Discussion
Our results showed five types of calls emitted by free-ranging K. rupestris, which was a smaller acoustic repertoire than that reported for captive animals (Lacher [49]: six types of calls; Monticelli and Alencar-Junior [51]: 11 types of calls and two mechanical signals). Different stimuli in the two environmental conditions (wild x captive) and the proximity to the caller may explain the different number of types of calls reported in these studies.
The snort-like and snort calls recorded in both sampled areas were emitted exclusively by adult females and are not easy to record in the wild. They are barely perceptible to the human, and only at a close distance (within 3 m from the caller). The snort call described by Monticelli and Alencar-Junior [51] seems to be the same call recorded in the present study, as they occurred in a similar agonistic context. Alencar-Junior [52] described an association between alarm whistle and snort calls occurring in a territorial dispute, while the caller was under attack. This description is compatible with the behavioral context described in the present study for both the snort and snort-like calls, when only the individual who is attacked and expelled from the rock site emits these calls. Thus, the emission of snort and snort-like is probably related to a submissive behavior. This is compatible with a hierarchically organized species, such as K. rupestris [49]. This kind of vocal subordination signal has also been described for other mammalian species (grunt calls in Papio cynocephalus [70]; giggle calls in Crocuta crocuta, [71]; grunt calls in Tayassu pecari [72]). However, further experimental study must be done to test this hypothesis for K. rupestris. Our study draws heavily on previous foundational work about the repertoire of this species [51]. However, it also diverged by introducing new methodologies and findings that enrich our understanding of the rock cavy’s acoustic behavior, diversity, and plasticity. For instance, our study reveals that males and females adjust their vocalizations differently, and males in disturbed areas exhibit multidimensional vocal modulation (details below). Moreover, the snort-like call is not comparable to any call described before [49,51]. Therefore, this would be a new call to be included in the species’ acoustic repertoire. This underscores the importance of continued investigation to fully capture the intricate vocalizations of this species.
The scream and whine calls did not occur very often and were only emitted by juvenile animals in the most disturbed area (A1). The scream was categorized herein as an emission of pain, emitted when an individual was captured and bitten by a feral dog (Canis lupus familiaris). It is reasonable to suppose that the scream is the squeal call, previously described by Lacher [49] for both the rock cavy and the guinea pig (Cavia aperea). The author describes the squeal call in the context of fear and pain, during handling of captive animals and after being bitten by a conspecific [49]. In our study, the whine was emitted by an individual isolated from the group; a rare context for free-ranging rock cavy due to their cohesive nature. The spectral features of whine and scream described in the present study were like the whine and yelp calls described before for the rock cavy [51], which indicates the similarity of the calls emitted both in wild and captivity conditions. However, screams and whines are underrepresented, with only 12 and 11 samples, respectively. This limited data collection may affect the robustness of conclusions drawn about these less frequent vocalizations. Therefore, further studies should aim to obtain larger datasets to better understand the occurrence and functions of these calls, especially in different environmental contexts and age groups.
The alarm whistle was the most prevalent call emitted by the rock cavies in both areas. The emission of this call occurs in contexts of alarm and aggression and only by adult individuals, as previously reported [49,51]. Therefore, it is reasonable to suppose that the alarm whistle can exert both functions, acting as an alarm call after predator detection and for disputes between conspecifics, as described for other mammals [73]. Further studies must be done to clarify the function of this call, as already done for calls of other caviomorphs (e.g., Hydrochoerus hydrochaeris [30,74]).
The overall higher emission of high-pitched alarm whistle calls is associated with the presence of humans and negative anthropic changes in the environments [21,75–77]. Humans and exotic predators (domestic and feral dogs) that usually transit in A1 may be the main stimuli for the high rate of emission of alarm whistle calls and changes in their acoustic parameters. During the data collection period, no exotic predators were observed, and no humans other than the observer and the field assistant were present in the less disturbed area (A2). Anthropogenic actions usually enforce new challenges for free-ranging animals [78–80]. In these disturbed contexts, animals can respond differently to ensure survival and reproductive success, which can generate new selective pressures [21,54,81]. Alarm calls are usually emitted by threatened animals [35], which need to balance the cost and benefit for their emission [82,83]. These calls are louder and can easily expose the caller to a nearby audience, turning it into easy prey for opportunistic predators [37], as described for the European pied flycatcher (Ficedula hypoleuca) [84] and vervet monkey (Chlorocebus pygerythrus) [85].
Our results suggest that the interaction between sex and area has influenced the acoustic characteristics of alarm calls in K. rupestris, implying that males and females adjust their vocalizations differently in response to environmental disturbance. Both males and females in the more disturbed area (A1) emitted alarm calls with higher high-frequencies compared to individuals in the less disturbed area (A2). However, only males from A1 produced calls with higher low-frequencies and longer durations, highlighting a multidimensional vocal modulation in males. This pattern may be related to differences in social and ecological roles between the sexes. Males of K. rupestris, being more territorial and frequently involved in agonistic interactions [49,52], may exhibit greater vocal flexibility to ensure efficient detection of their calls by both group members and potential competitors. The emission of longer calls with greater frequency variation may enhance signal propagation in a fragmented environment, thereby maximizing their alert function [86,87]. In contrast, females, by adjusting only the high frequency, may be reducing predator exposure and minimizing the energy costs associated with vocalization [28]. These findings seem to show vocal plasticity in K. rupestris.
It is important to highlight the limitation of our study due to the absence of replicates for each level of disturbance (A1 disturbed area; A2 undisturbed area); due to this, other factors could also play a role, besides the anthropogenic factors such as diseases, parasites, and social events, for instance. However, given the similarity in habitat structure and other variables, such as temperature and vegetation cover, as well as food, water, and shelter resources for rock cavies in both areas, it is reasonable to infer that those variations in the emission rate and acoustic parameters of rock cavy whistle calls between areas may be linked to the animals’ emotional state in response to human disturbance in the most disturbed area. Therefore, the emission of alarm whistle calls with an hourly rate of 2.6 calls/h or higher, along with high-frequency and peak-frequency of 7222 Hz and 2603 Hz or higher, respectively, may serve as ecological indicators of anthropogenic disturbance in the Caatinga biome, enabling remote monitoring. Furthermore, future studies focusing on monitoring animals’ distress in response to negative events—such as by measuring cortisol levels and correlating these with acoustic variations—could provide valuable insights into the relationship between whistle call emissions, negative emotional states, and high arousal [53,54]. This approach could enhance our understanding of the stressors influencing vocalizations and offer a more comprehensive ecological indicator of anthropogenic impact.
In addition, the ad libitum method of vocalization collection used herein played a crucial role in documenting vocalizations within our study. This method provided the flexibility needed to capture a wide range of vocal behaviors in a natural setting, reflecting the genuine conditions of the animals’ environment. However, while ad libitum sampling allows for comprehensive data collection without strict constraints, it is important to acknowledge its limitations. Systematic recording methods, on the other hand, offer the advantage of yielding more precise insights into specific phenomena, such as predator presence or isolation frequency. These methods involve structured and consistent data collection protocols, which can enhance the reliability and comparability of the findings. Thus, future research directions could explore a comparison of these approaches. Therefore, conducting parallel studies using both ad libitum and systematic recording methods could provide a deeper understanding of their respective strengths and limitations. This comparative analysis would help identify the most effective strategies for documenting vocalizations and interpreting their ecological significance.
As previously stated, K. rupestris and its habitat are under threat. However, the conservation units in the Caatinga are not adequately protected due to insufficient financial and human resources. For instance, only six park rangers are responsible for enforcing the law across the entire 46,000 hectares of the two conservation units studied, namely the Serra dos Montes Altos wildlife refuge (REVIS) and the Serra dos Montes Altos state park (PESMA). Utilizing the results of this study will enable remote bioacoustics monitoring, allowing authorities to better allocate scarce resources to areas with higher levels of human interference. Nonetheless, due to the study’s limitations, such as the lack of behavioral quantification data in the observed areas, further research is required to validate this hypothesis. Moreover, additional study may be required to determine whether K. rupestris adjusts its acoustic responses based on the type of threat, which would improve remote monitoring. Furthermore, long-term monitoring could be suggested to assess the stability of observed patterns and their potential changes over time in response to varying anthropic pressures.
5. Conclusions
The acoustic characteristics of alarm whistle calls emitted by K. rupestris in the most disturbed area can be attributed to negative events and high arousal. Although our study had limitations that prevented more systematic tests on the emotions associated with negative events experienced by K. rupestris, our findings allow us to infer the quality of their environment and, consequently, the rock cavy’s vocal responses, suggesting impaired animal welfare in the most disturbed area. These results validate our predictions and highlight the potential of using acoustic monitoring of alarm whistle calls by free-ranging K. rupestris as an ecological indicator of environmental disturbance in the Caatinga biome.
Supporting information
S1 File. Example of rock cavies’ alarm whistle recorded in the Caatinga.
https://doi.org/10.1371/journal.pone.0323711.s001
(WAV)
S2 File. Example of rock cavies’ scream recorded in the Caatinga.
https://doi.org/10.1371/journal.pone.0323711.s002
(WAV)
S3 File. Example of rock cavies’ snort recorded in the Caatinga.
https://doi.org/10.1371/journal.pone.0323711.s003
(WAV)
S4 File. Example of rock cavies’ snot-like recorded in the Caatinga.
https://doi.org/10.1371/journal.pone.0323711.s004
(WAV)
S5 File. Example of rock cavies’ whine recorded in the Caatinga.
https://doi.org/10.1371/journal.pone.0323711.s005
(WAV)
S6 File. Raw data and DFA analysis of rock cavies’ calls recorded in the Caatinga.
https://doi.org/10.1371/journal.pone.0323711.s006
(DOCX)
S7 File. Raw data and generalized linear models (GLMs) analyses of rock cavies’ alarm whistle calls recorded in the Caatinga.
https://doi.org/10.1371/journal.pone.0323711.s007
(DOCX)
Acknowledgments
We thank park rangers João C. Santos, Edvan F. Pereira and Antônio Marcos S. Brito for all their support in the field.
References
- 1. Benítez-López A, Santini L, Schipper AM, Busana M, Huijbregts MAJ. Intact but empty forests? Patterns of hunting-induced mammal defaunation in the tropics. PLoS Biol. 2019;17(5):e3000247. pmid:31086365
- 2. Betts MG, Wolf C, Pfeifer M, Banks-Leite C, Arroyo-Rodríguez V, Ribeiro DB, et al. Extinction filters mediate the global effects of habitat fragmentation on animals. Science. 2019;366(6470):1236–9. pmid:31806811
- 3. Krauss J, Bommarco R, Guardiola M, Heikkinen RK, Helm A, Kuussaari M, et al. Habitat fragmentation causes immediate and time-delayed biodiversity loss at different trophic levels. Ecol Lett. 2010;13(5):597–605. pmid:20337698
- 4.
Radić B, Gavrilović S. Natural Habitat Loss: Causes and Implications of Structural and Functional Changes. In: Leal Filho W, Azul A, Brandli L, Lange Salvia A, Wall T. editors. Life on Land. Encyclopedia of the UN Sustainable Development Goals. Springer; Cham: 2020. https://doi.org/10.1007/978-3-319-71065-5_6-1
- 5. Theng M, Jusoh WF, Jain A, Huertas B, Tan DJ, Tan HZ. A comprehensive assessment of diversity loss in a well-documented tropical insect fauna: Almost half of Singapore’s butterfly species extirpated in 160 years. Biol Conserv. 2020:242.
- 6. Jorge FC, Machado CG, da Cunha Nogueira SS, Nogueira-Filho SLG. The effectiveness of acoustic indices for forest monitoring in Atlantic rainforest fragments. Ecol Indic. 2018;91:71–6.
- 7. Penar W, Magiera A, Klocek C. Applications of bioacoustics in animal ecology. Ecol Complex. 2020;43:100847.
- 8. Wrege PH, Rowland ED, Keen S, Shiu Y. Acoustic monitoring for conservation in tropical forests: examples from forest elephants. Methods Ecol Evol. 2017;8(10):1292–301.
- 9. Deichmann JL, Acevedo‐Charry O, Barclay L, Burivalova Z, Campos‐Cerqueira M, d’Horta F, et al. It’s time to listen: there is much to be learned from the sounds of tropical ecosystems. Biotropica. 2018;50(5):713–8.
- 10. Bardeli R, Wolff D, Kurth F, Koch M, Tauchert K-H, Frommolt K-H. Detecting bird sounds in a complex acoustic environment and application to bioacoustic monitoring. Pattern Recognit Lett. 2010;31(12):1524–34.
- 11.
Spence HR. Bioacoustic monitoring: Urgent challenges and opportunities on the MesoAmerican Reef System. Ocean. MTS/IEEE Charleston; 2018. https://doi.org/10.1109/OCEANS.2018.8604744
- 12. Williams EM, O’Donnell CFJ, Armstrong DP. Cost-benefit analysis of acoustic recorders as a solution to sampling challenges experienced monitoring cryptic species. Ecol Evol. 2018;8(13):6839–48. pmid:30038779
- 13. Zwart MC, Baker A, McGowan PJK, Whittingham MJ. The use of automated bioacoustic recorders to replace human wildlife surveys: an example using nightjars. PLoS One. 2014;9(7):e102770. pmid:25029035
- 14. Wimmer J, Towsey M, Roe P, Williamson I. Sampling environmental acoustic recordings to determine bird species richness. Ecol Appl. 2013;23(6):1419–28. pmid:24147413
- 15. Borker AL, McKown MW, Ackerman JT, Eagles-Smith CA, Tershy BR, Croll DA. Vocal activity as a low cost and scalable index of seabird colony size. Conserv Biol. 2014;28(4):1100–8. pmid:24628442
- 16. Briefer EF. Vocal expression of emotions in mammals: mechanisms of production and evidence. J Zool. 2012;288(1):1–20.
- 17.
Briefer EF. Coding for ‘Dynamic’ Information: Vocal Expression of Emotional Arousal and Valence in Non-human Animals. In: Aubin T, Mathevon N, editos. Coding Strategies in Vertebrate Acoustic Communication. Animal Signals and Communication, vol 7. Springer; Cham: 2020. https://doi.org/10.1007/978-3-030-39200-0_6
- 18. Mendl M, Burman OHP, Paul ES. An integrative and functional framework for the study of animal emotion and mood. Proc Biol Sci. 2010;277(1696):2895–904. pmid:20685706
- 19. Lima AF, Lima SGC, Nogueira-Filho SLG, Held S, Paul E, Mendl M, et al. Vocal expression of emotions in farmed spotted paca (Cuniculus paca). Appl Anim Behav Sci. 2022;256:105753.
- 20. Blumstein DT, Mennill DJ, Clemins P, Girod L, Yao K, Patricelli G, et al. Acoustic monitoring in terrestrial environments using microphone arrays: applications, technological considerations and prospectus. J Appl Ecol. 2011;48(3):758–67.
- 21. Gogoleva SS, Volodina EV, Volodin IA, Kharlamova AV, Trut LN. The gradual vocal responses to human-provoked discomfort in farmed silver foxes. Acta Ethol. 2010;13(2):75–85. pmid:22865950
- 22. Baciadonna L, Briefer EF, McElligott AG. Investigation of reward quality-related behaviour as a tool to assess emotions. Appl Anim Behav Sci. 2020;225:104968.
- 23. Mateo JM. Alarm calls elicit predator-specific physiological responses. Biol Lett. 2010;6(5):623–5. pmid:20236965
- 24. Grandin T. The feasibility of using vocalization scoring as an indicator of poor welfare during cattle slaughter. Appl Anim Behav Sci. 1998;56(2–4):121–8.
- 25. Linhart P, Ratcliffe VF, Reby D, Špinka M. Expression of emotional arousal in two different piglet call types. PLoS One. 2015;10(8):e0135414. pmid:26274816
- 26. Keenan S, Mathevon N, Stevens JMG, Nicolè F, Zuberbühler K, Guéry J-P, et al. The reliability of individual vocal signature varies across the bonobo’s graded repertoire. Anim Behav. 2020;169:9–21.
- 27. Monteiro R de A, Ferreira CD, Perbiche-Neves G. Vocal repertoire and group-specific signature in the Smooth-billed Ani, Crotophaga ani Linnaeus, 1758 (Cuculiformes, Aves). Pap Avulsos Zool. 2021;61:e20216159.
- 28.
Bradbury JW, Vehrencamp SL. Principles of animal communcation. 2nd ed. Sinauer Associates; Sunderland, Massachusetts: 1998.
- 29. Barros KS, Tokumaru RS, Pedroza JP, Nogueira SSC. Vocal repertoire of captive capybara (Hydrochoerus hydrochaeris): structure, context and function. Ethology. 2010;117(1):83–94.
- 30. Nogueira SSC, Pedroza JP, Nogueira‐Filho SLG, Tokumaru RS. The function of click call emission in Capybaras (Hydrochoerus hydrochaeris). Ethology. 2012;118(10):1001–9.
- 31. Morales MA, Barone JL, Henry CS. Acoustic alarm signalling facilitates predator protection of treehoppers by mutualist ant bodyguards. Proc Biol Sci. 2008;275(1645):1935–41. pmid:18480015
- 32. Platzen D, Magrath RD. Parental alarm calls suppress nestling vocalization. Proc Biol Sci. 2004;271(1545):1271–6. pmid:15306351
- 33. Mcloughlin MP, Stewart R, McElligott AG. Automated bioacoustics: methods in ecology and conservation and their potential for animal welfare monitoring. J R Soc Interface. 2019;16(155):20190225. pmid:31213168
- 34. Teixeira D, Maron M, van Rensburg BJ. Bioacoustic monitoring of animal vocal behavior for conservation. Conservat Sci and Prac. 2019;1(8).
- 35. Hollén LI, Radford AN. The development of alarm call behaviour in mammals and birds. Anim Behav. 2009;78(4):791–800.
- 36. Templeton CN. Animal communication: learning by listening about danger. Curr Biol. 2018;28(16):R892–4. pmid:30130514
- 37.
Berthet M, Zuberbühler K. Alarm calling. Encyclopedia of Evolutionary Psychological Science. 2020. pp. 1–5. https://doi.org/10.1007/978-3-319-16999-6_1235-1
- 38.
Bonvicino C, Oliveira J, D’Andrea O. Guia dos roedores do Brasil com chaves para gêneros baseadas em caracteres externos. Organização Pan-Americana da Saúde. 2008.
- 39. Oliveira CDL, Silva APA, Moura PAG. Distribution and importance of conservation units in the caatinga domain. Anu do Inst Geociencias. 2019;42(1):425–9.
- 40. Amora TD, Ferrari SF. Predation of small-bodied mammals (Callithrix and Kerodon) by Laughing Falcons (Herpetotheres cachinnans) in the semi-arid Caatinga Scrub Forest of the Brazilian Northeast. Wilson J Ornithol. 2014;126(4):771–5.
- 41. Marinho PH, Bezerra D, Antongiovanni M, Fonseca CR, Venticinque EM. Activity patterns of the threatened northern tiger cat Leopardus tigrinus and its potential prey in a Brazilian dry tropical forest. Mamm Biol. 2018;89:30–6.
- 42. Penido G, Astete S, Jácomo ATA, Sollmann R, Tôrres N, Silveira L, et al. Mesocarnivore activity patterns in the semiarid Caatinga: limited by the harsh environment or affected by interspecific interactions? J Mammal. 2017;98(6):1732–40.
- 43. Alves RRN, Gonçalves MBR, Vieira WLS. Caça, uso e conservação de vertebrados no semiárido Brasileiro. Trop Conserv Sci. 2012;5(3):394–416.
- 44. Costa LP, Leite YLR, Mendes SL, Ditchfield AD. Conservação de mamíferos no Brasil. Megadiversidade. 2005;1:103–12.
- 45. Zogno MA, Miglino MA, Oliveira MF. Análise bioquímica dos líquidos fetais e citologia do fluido amniótico da fêmea de Mocó (Kerodon rupestris). Braz J Vet Res Anim Sci. 2004;41(4):228–35.
- 46. ICMBio. Livro vermelho da fauna brasileira ameaçada de extinção. 2018a;1:66–70. [cited 20 Jun 2021. ]. Available from: https://www.icmbio.gov.br/portal/images/stories/comunicacao/publicacoes/publicacoes-diversas/livro_vermelho_2018_vol1.pdf
- 47. ICMBio. Livro vermelho da fauna brasileira ameaçada de extinção. 2018b;2:401–3. [cited 20 Jun 2021. ]. Available from: https://www.icmbio.gov.br/portal/images/stories/comunicacao/publicacoes/publicacoes-diversas/livro_vermelho_2018_vol2.pdf
- 48. Paquet P, Darimont C. Wildlife conservation and animal welfare: two sides of the same coin? Anim Welf. 2010;19(2):177–90.
- 49. Lacher TE Jr. The comparative social behavior of Kerodon rupestris and Galea spixii and the evolution of behavior in the Caviidae. Bull Carnegie Mus Nat History. 1981;17:1–71.
- 50.
Oliveira JA, Bonvicino CR. Ordem rodentia. In: Reis NR, Peracchi AL, Pedro WA, Lima IP, editors. Mamíferos do Brasil. Londrina; 2006. pp. 347–400.
- 51. Monticelli PF, Alencar-Jr RN. The acoustic behavior of the Brazilian caatinga big rodent is incongruent to its actual position in Hydrochoerinae. Behav Processes. 2021;193:104523. pmid:34592346
- 52.
Alencar-Junior RN. O repertório acústico de um especialista de rochedos da caatinga, o Mocó [The acoustic repertory of a specialist of the rockery Caatinga: the rock cavy]. Dissertation (Master’s in Psychology). São Paulo: São Paulo University. Psychology Institute. 2011. pp. 6–102.
- 53. Briefer EF, Tettamanti F, McElligott AG. Emotions in goats: mapping physiological, behavioural and vocal profiles. Anim Behav. 2015;99:131–43.
- 54. Manteuffel G, Puppe B, Schön PC. Vocalization of farm animals as a measure of welfare. Appl Anim Behav Sci. 2004;88(1–2):163–82.
- 55. Paul ES, Harding EJ, Mendl M. Measuring emotional processes in animals: the utility of a cognitive approach. Neurosci Biobehav Rev. 2005;29(3):469–91. pmid:15820551
- 56. Tasso MDL, Chang MRC. Análise erosional da Serra de Monte Alto entre Palmas de Monte Alto e Sebastião Laranjeiras, BA. Actageo. 2013;7:107–21.
- 57. Freire Júnior JMS, Silva JS. Clado Mimosoide (Leguminosae e Caesalpinioideae) no Parque Estadual da Serra dos Montes Altos, Bahia, Brasil. Rodriguésia. 2019;70.
- 58. Barros KS, Almeida WN. Mamíferos terrestres do parque estadual da serra dos montes altos, bahia: primeiros foto-registros. Bol Soc Bras Mastozool. 2018;83:162–7.
- 59. Embrapa, Empresa Brasileira de Pesquisa Agropecuária. Sistema de classificação de solos do Brasil. 2022. [cited 13 Nov 2023. ]. Available from: https://www.embrapa.br/solos/sibcs
- 60. Mapbiomas. Coleção 8 da Série Anual de Mapas de Cobertura e Uso de Solo do Brasil. 2022. [cited 15 Nov 2023. ]. Available from: https://mapbiomas.org/
- 61. USGS. United States Geological Survey. 2022. [cited 15 Nov 2023. ]. Available from: https://www.usgs.gov/
- 62. Altmann J. Observational study of behavior: sampling methods. Behav. 1974;49:227–67.
- 63. Delciellos AC. Mammals of four Caatinga areas in northeastern Brazil: inventory, species biology, and community structure. check list. 2016;12(3):1916.
- 64. Dias DM, Campos CB, Rodrigues FHG. Behavioural ecology in a predator-prey system. Mamm Biol. 2018;92:30–36.
- 65. Mares MA, Willig MR, Streilein KE, Lacher TE Jr. The mammals of northeastern Brazil: a preliminary assessment. Ann Carnegie Mus. 1981;50:81–137.
- 66. Oliveira F, Andrade-Da-Costa B, Cavalcante J, Silva S, Soares J, Lima R, et al. The eye of the crepuscular rodent rock cavy (Kerodon rupestris) (Wied, 1820). J Morphol Sci. 2014;31(02):089–97.
- 67.
Charif RA, Waack AM, Strickman LM. Raven pro 1.4 user`s manual. Ithaca, NY: Cornell Lab of Ornithology. 2010.
- 68. Conceição AM, Bocchiglieri A. Population density and use of space by Kerodon rupestris: an endemic and threatened rodent in the semiarid areas of Brazil. J Arid Environ. 2021;186:104425.
- 69.
Tabachnick BG, Fidell LS. Multivariate statistics. Pearson and Bacon. 2007.
- 70. Cheney DL, Seyfarth RM, Silk JB. The role of grunts in reconciling opponents and facilitating interactions among adult female baboons. Anim Behav. 1995;50(1):249–57.
- 71. Mathevon N, Koralek A, Weldele M, Glickman SE, Theunissen FE. What the hyena’s laugh tells: sex, age, dominance and individual signature in the giggling call of Crocuta crocuta. BMC Ecol. 2010;10:9. pmid:20353550
- 72. Nogueira SSC, Caselli CB, Costa TSO, Moura LN, Nogueira-Filho SLG. the role of grunt calls in the social dominance hierarchy of the white-lipped Peccary (Mammalia, Tayassuidae). PLoS One. 2016;11(7):e0158665. pmid:27409797
- 73. Macedonia JM, Evans CS. Essay on contemporary issues in ethology: variation among mammalian alarm call systems and the problem of meaning in animal signals. Ethology. 1993;93(3):177–97.
- 74. Dos Santos E, Tokumaru RS, Nogueira-Filho SLG, Nogueira SSC. The effects of unrelated offspring whistle calls on capybaras (Hydrochoerus hydrochaeris). Braz J Biol. 2014;74(3 Suppl 1):S171–6. pmid:25627382
- 75. Fischer J, Hammerschmidt K, Todt D. Factors affecting acoustic variation in barbary‐macaque (Macaca sylvanus) disturbance calls. Ethology. 1995;101(1):51–66.
- 76. Schrader L, Todt D. Contact call parameters covary with social context in common marmosets, Callithrix J. jacchus. Anim Behav. 1993;46(5):1026–8.
- 77. Schehka S, Esser K-H, Zimmermann E. Acoustical expression of arousal in conflict situations in tree shrews (Tupaia belangeri). J Comp Physiol A Neuroethol Sens Neural Behav Physiol. 2007;193(8):845–52. pmid:17520262
- 78. Costa TSO, Nogueira-Filho SLG, De Vleeschouwer KM, Oliveira LC, de Sousa MBC, Mendl M, et al. Individual behavioral differences and health of golden-headed lion tamarins (Leontopithecus chrysomelas). Am J Primatol. 2020;82(5):e23118. pmid:32128855
- 79. Pamela Delarue EM, Kerr SE, Lee Rymer T. Habitat complexity, environmental change and personality: a tropical perspective. Behav Processes. 2015;120:101–10. pmid:26386151
- 80. Nogueira SSC, Bernardi LG, Nogueira-Filho SLG. A note on comparative enclosure facility usage by wild and captive-born capybaras (Hydrochoerus hydrochaeris). Appl Anim Behav Sci. 2004;89(1–2):139–43.
- 81. Laiolo P. The emerging significance of bioacoustics in animal species conservation. Biol Conserv. 2010;143(7):1635–45.
- 82. Tamachi N. The evolution of alarm calls: an altruism with nonlinear effect. J Theor Biol. 1987;127(2):141–53.
- 83. Taylor RJ, Balph DF, Balph MH. The evolution of alarm calling: a cost-benefit analysis. Anim Behav. 1990;39(5):860–8.
- 84. Krama T, Krams I. Cost of mobbing call to breeding pied flycatcher, Ficedula hypoleuca. Behav Ecol. 2004;16(1):37–40.
- 85. Cheney DL, Seyfarth RM. Vervet monkey alarm calls: manipulation through shared information? Behav. 1985;94(1–2):150–66.
- 86. Blumstein DT, Armitage KB. Does sociality drive the evolution of communicative complexity? A comparative test with ground-dwelling sciurid alarm calls. Am Nat. 1997;150(2):179–200. pmid:18811281
- 87. Manser MB. The acoustic structure of suricates’ alarm calls varies with predator type and the level of response urgency. Proc Biol Sci. 2001;268(1483):2315–24. pmid:11703871