Miriam Selwyn1, Pedro J Garrote1, Antonio R Castilla1, Jose M Fedriani2,3. 1. Centro de Ecologia Aplicada "Prof. Baeta Neves" CEABN/InBIO, Instituto Superior de Agronomia, Universidade de Lisboa, Lisboa, Portugal. 2. Centro de Investigaciones sobre Desertificación (CIDE-CSIC) Carretera Moncada - Náquera, Moncada, Valencia, Spain. 3. Estación Biológica de Doñana (EBD - C.S.I.C.), Seville, Spain.
Abstract
The study of plant-frugivore interactions is essential to understand the ecology and evolution of many plant communities. However, very little is known about how interactions among frugivores indirectly affect plant reproductive success. In this study, we examined direct interactions among vertebrate frugivores sharing the same fruit resources. Then, we inferred how the revealed direct interspecific interactions could lead to indirect (positive or negative) effects on reproductive success of fleshy fruited plants. To do so, we developed a new analytical approach that combines camera trap data (spatial location, visitor species, date and time, activity) and tailored null models that allowed us to infer spatial-temporal interactions (attraction, avoidance or indifference) between pairs of frugivore species. To illustrate our approach, we chose to study the system composed by the Mediterranean dwarf palm, Chamaerops humilis, the Iberian pear tree, Pyrus bourgaeana, and their shared functionally diverse assemblages of vertebrate frugivores in a Mediterranean area of SW Spain. We first assessed the extent to which different pairs of frugivore species tend to visit the same or different fruiting individual plants. Then, for pairs of species that used the same individual plants, we evaluated their spatial-temporal relationship. Our first step showed, for instance, that some prey frugivore species (e.g. lagomorphs) tend to avoid those C. humilis individuals that were most visited by their predators (e.g. red foxes). Also, the second step revealed temporal attraction between large wild and domestic frugivore ungulates (e.g. red deer, cows) and medium-sized frugivores (e.g. red foxes) suggesting that large mammals could facilitate the C. humilis and P. bourgaeana exploitation to other smaller frugivores by making fruits more easily accessible. Finally, our results allowed us to identify direct interaction pathways, that revealed how the mutualistic and antagonistic relations between animal associates derived into indirect effects on both plants seed dispersal success. For instance, we found that large-sized seed predators (e.g. ungulates) had a direct positive effect on the likelihood of visits by legitimate seed dispersers (e.g. red foxes) to both fleshy fruited plants. Then, seed predators showed an indirect positive effect on the plants' reproductive success. Our new analytical approach provides a widely applicable framework for further studies on multispecies interactions in different systems beyond plant-frugivore interactions, including plant-pollinator interactions, the exploitation of plants by herbivores, and the use of carcasses by vertebrate scavengers.
The study of plant-frugivore interactions is essential to understand the ecology and evolution of many plant communities. However, very little is known about how interactions among frugivores indirectly affect plant reproductive success. In this study, we examined direct interactions among vertebrate frugivores sharing the same fruit resources. Then, we inferred how the revealed direct interspecific interactions could lead to indirect (positive or negative) effects on reproductive success of fleshy fruited plants. To do so, we developed a new analytical approach that combines camera trap data (spatial location, visitor species, date and time, activity) and tailored null models that allowed us to infer spatial-temporal interactions (attraction, avoidance or indifference) between pairs of frugivore species. To illustrate our approach, we chose to study the system composed by the Mediterranean dwarf palm, Chamaerops humilis, the Iberian pear tree, Pyrus bourgaeana, and their shared functionally diverse assemblages of vertebrate frugivores in a Mediterranean area of SW Spain. We first assessed the extent to which different pairs of frugivore species tend to visit the same or different fruiting individual plants. Then, for pairs of species that used the same individual plants, we evaluated their spatial-temporal relationship. Our first step showed, for instance, that some prey frugivore species (e.g. lagomorphs) tend to avoid those C. humilis individuals that were most visited by their predators (e.g. red foxes). Also, the second step revealed temporal attraction between large wild and domestic frugivore ungulates (e.g. red deer, cows) and medium-sized frugivores (e.g. red foxes) suggesting that large mammals could facilitate the C. humilis and P. bourgaeana exploitation to other smaller frugivores by making fruits more easily accessible. Finally, our results allowed us to identify direct interaction pathways, that revealed how the mutualistic and antagonistic relations between animal associates derived into indirect effects on both plants seed dispersal success. For instance, we found that large-sized seed predators (e.g. ungulates) had a direct positive effect on the likelihood of visits by legitimate seed dispersers (e.g. red foxes) to both fleshy fruited plants. Then, seed predators showed an indirect positive effect on the plants' reproductive success. Our new analytical approach provides a widely applicable framework for further studies on multispecies interactions in different systems beyond plant-frugivore interactions, including plant-pollinator interactions, the exploitation of plants by herbivores, and the use of carcasses by vertebrate scavengers.
A major long-established goal in community ecology and evolutionary biology is to understand how interspecific interactions influence population density, distribution, phenotypes, and genotypes [1], crucial to the selection and evolution of life-history traits. During the few last decades, the study of interspecific interactions has experienced an outstanding progress [2] such as, for example, moving from a traditional pair-wise perspective [3-6] to a more realistic and complex multispecific approach, where multiple species interact with each other [2, 7, 8]. However, most of these studies have focused on a single type of interaction at a time, usually studying either mutualistic or antagonistic interactions among species of particular taxonomic groups [9]. Examples of such interspecific interactions that have been most often investigated separately are competitive interactions among vertebrates [e.g., 10, 11] facilitative interactions among plants [e.g., 12, 13] and mutualistic/antagonistic plant-animal interactions [e.g., 8, 14]. Many of these interaction types take place within the same habitats [15-17] and thus, species involved in one interaction type (e.g. predation, competition) can also participate in other interactions (e.g. plant-animal interactions). However, to our knowledge, studies integrating these interaction types are still scarce [but see 18].New analytical approaches have emerged allowing coping with the methodological challenge that represents studying systems comprising different species that interact among them in variable ways. For instance, multilayer networks [19], spatially explicit agent-based simulation modeling [20] and tailored null models based on resampling techniques [21] represent powerful tools to investigate such complex systems. In many situations, however, a basic challenge remains as to how to monitor subtle and mixed interactions under suboptimal field conditions, especially when target species are nocturnal, secretive, or otherwise elusive animals (e.g. many vertebrates). Importantly, during the last few decades camera traps have revolutionized wildlife research, enabling the collection of precise photographic evidence of rarely seen species, with relatively little cost [22-24]. Camera traps record very accurate data while barely disturbing the photographed animal, operate continually and silently, providing proof of individuals at a spatially restricted spot (e.g. a fruiting plant, a carcass, a water point), their precise time of visit, and their activity (foraging, perching, fighting, etc.) among other relevant data. Hundreds of thousands, if not millions, of photographs of large numbers of vertebrate species are indeed being systematically collected in diverse habitats worldwide [see review 25]. Surprisingly, until very recently camera trap surveys have been seldom used as a quantitative tool to thoroughly measure direct interactions among individuals of different vertebrate species [but see 26–28].Here we propose a combination of camera-trap survey data and tailored null models as a unifying methodological framework to investigate whether and how interactions between foraging animals (predation, facilitation) alter subsequent plant-animal interactions (pollination, frugivory, herbivory). To illustrate the usefulness of our approach, we evaluate whether interspecific interactions among vertebrate frugivores alter the likelihood of their subsequent interactions with fleshy-fruited plants. Our analytical approach comprises two basic steps. First, we evaluated if different frugivore species within a study area visited the same fruiting individuals. Then, we used date-time data of photographs and resampling techniques to tailor null models that allowed us to evaluate whether different species of frugivores showed attraction, aversion, or indifference in their timing of visiting (and interacting) with such fruiting plants. Finally, we inferred the indirect consequences of these direct frugivore responses (attractive or aversive) on the plant’s reproductive success [e.g., 18].To illustrate the value of our approach, we chose as study systems the Mediterranean dwarf palm Chamaerops humilis and the Iberian pear tree Pyrus bourgaeana as well as their shared community of vertebrate frugivores (mostly mammals) at Doñana National Park (SW Spain) [e.g., 18, 29]. This diverse assemblage of frugivores that consume C. humilis and P. bourgaeana’s fruits can be grouped into five functional frugivore groups: carnivores, wild ungulates, domestic ungulates, pulp feeders (i.e. lagomorphs and rodents) and birds, which may act as seed disperses or seed predators [18, 29–33]; thus, providing a wide spectrum of antagonistic and mutualistic interactions among multiple species across trophic levels [32, 34, 35].In plant communities, their functional traits vary within environmental gradients and among species occupying similar conditions, raising a challenge for the synthesis of functional and community ecology [36]. The coexistence of consumer species is fostered by resource-use and niche differences, leading to greater resource use in communities with higher number of species [37]. The ecological niche of species is multi-dimensional, including three axes of particular importance that explicitly impact diets and spatial-temporal patterns of abundance: trophic interactions, habitat use and their temporal variability [38]. As a result of interspecific facilitative and competitive interactions, resource partitioning, and niche differentiation, our general hypothesis is that visitation of a fruiting plant by an individual of a given species of vertebrate frugivore could alter (either increasing or lessening) the likelihood of subsequent visits by individuals of other vertebrate frugivore species. Specifically, we expected that (i) different frugivores species would tend to visit different fruiting individual plants lessening thus interspecific competition or predation risk, (ii) when using the same fruiting plant, prey frugivore species (e.g. small mammals) would tend to temporally avoid plants being foraged by their predators (e.g. carnivores) to reduce predation risk, and (iii) large-bodied frugivores (e.g. ungulates) would facilitate the plant exploitation to smaller frugivores by making the fruit more easily accessible (e.g. by giving up ripe fruits in the plants’ immediacy).
Materials and methods
Study area and sites
The study was carried out during the fruiting seasons (September to December) of 2018 and 2019 in Doñana National Park, SW of the Iberian Peninsula (37° 9′ N, 6° 26′ W, Fig 1a). The region is characterized by sub-humid Mediterranean climate, with hot and dry summers (June-September) and mild, wet winters (November-February [39]). The average annual temperature ranges between 15.4 and 18.7°C (mean ± SE = 16.9 ± 1°C; n = 35; period 1978–2017 and annual precipitation is highly variable, ranging between 170 and 1028 mm (mean ± SE = 542.6 ± 32.8 mm; n = 35; period 1978–2017), with most rainfall during winter (271.4 ± 27.1 mm) and extreme drought during summer (33.3 ± 5.3 mm) (data from Monitoring Team of Natural Process of Doñana Biological Station; http://icts.ebd.csic.es/en/web/icts-ebd/monitoring-program-physical-environment).
Fig 1
Study area.
a) Location of Doñana National Park in Europe and the Iberian Peninsula. b) Location of the study sites: Matasgordas (area = 274 ha) and Martinazo (area = 16.7 ha).
Study area.
a) Location of Doñana National Park in Europe and the Iberian Peninsula. b) Location of the study sites: Matasgordas (area = 274 ha) and Martinazo (area = 16.7 ha).We selected two C. humilis populations (10 km apart) (Matasgordas and Martinazo sites, Fig 1b) and one population of P. bourgaeana located at Matasgordas’ site (as density of P. bourgaeana is extremely low in Martinazo). Both sites have suffered consequences derived from human activities which have made them differ in vegetation and physiographic characteristics. In Matasgordas, Mediterranean shrub species (C. humilis, Pistacia lentiscus, Halimium halimifolium, Cistus spp.) as well as some scattered Pyrus bourgaeana, Quercus suber, Fraxinus angustifolia and Olea europaea var. sylvestris trees are slowly recolonizing the old-field area [40-42]. In Martinazo site, the area has been recolonized mainly by early-successional species (H. halimifolium, Ulex spp., and Stauracanthus genistoides) and animal-dispersed native plants (C. humilis, Rubus ulmifolius, Phillyrea angustifolia or Asparagus aphyllus [43]).
Study species
Chamaerops humilis (Mediterranean dwarf palm) and Pyrus bourgaeana (Iberian pear tree) are endemic species to the Mediterranean region, specifically distributed in southern Europe and northern Africa [30, 44]. Chamaerops humilis is a dioecious palm that blooms from March to May and is mostly pollinated by insects [45-48]. Its fruits are ‘polydrupes’ (usually 1–3 drupes, which vary from 1–3 cm in diameter [49]) attached to infrutescences of up to 30 cm long (7–120 fruits per infructescence) and usually located at ~10–30 cm above the ground level. The ripening occurs in autumn (September-November). Seedlings emerge during spring and early summer, experiencing high mortality due to summer droughts and herbivory [29]. Pyrus bourgaeana is a small (3–6 m height) deciduous monoecious tree which flowers during February and March and is pollinated by numerous Hymenoptera, Diptera, and Coleoptera species [50, 51]. Each tree typically produces 200–450 fruits [29] which are non-dehiscent globose pomes (2–3 cm diameter), with a sugary water-rich pulp, that ripe and drop to the ground from September to December [18]. Seedlings emerge during early spring and experience high mortality due to summer droughts and herbivory [52].In Doñana National Park, C. humilis and P. bourgaeana’s seed dispersal is mainly accomplished by medium-size carnivores such as Eurasian badgers (Meles meles) and red foxes (Vulpes vulpes) [28, 29]. These two carnivores intensively prey on lagomorphs (Oryctolagus cuniculus) and up to five rodent species (mostly Apodemus sylvaticus and Mus spretus) [32] that feed on C. humilis and P. bourgaeana’s pulp and seeds [29, 30, 33]. Lagomorphs and rodents act mostly as pulp feeders, although may occasionally act as short-distance seed dispersers for C. humilis [29]. Whereas for P. bourgaeana, lagomorphs act mostly as pulp feeders and may occasionally disperse some seeds at short distances, rodents damage all seeds eaten [18, 30]. Fruit predators such as red (Cervus elaphus) and fallow deer (Dama dama) grind all ingested seeds [29, 30], although can disperse viable seeds of C. humilis by regurgitation [31]. Wild boars (Sus scrofa) act mostly as seed predators though may occasionally disperse some viable seeds of both species [30, P.J. Garrote unpublished data]. Free-ranging cows and horses are abundant in some areas of Doñana and are known to consume and usually predate both C. humilis and P. bourgaeana seeds. Large mammals (e.g. deer, cows) seem to be better adapted in their foraging, either by breaking many of the defensive needle-like spines in fruiting C. humilis or by plucking P. bourgaeana’s fruits still attached to the branches (Authors personal observations, see S1A File). Hence, ungulates remove many fruits of both plant species and often leave some of them whole or partially consumed around mother plants (Authors unpublished data). These left fruits are more easily accessible to medium-sized frugivores which hardly cope with the palm’s spines and cannot reach the fruits attached to P. bourgaeana branches. Therefore, we predict that the foraging of smaller frugivores could be facilitated by larger ones [e.g., 18, 53–55]. Also, birds such as azure-winged magpieCyanopica cyanus and blackcap Silvia atricapilla consume P. bourgaeana’s fruits mostly acting as pulp-predators rather than seed dispersers [30]. As a result, this diverse assemblage of frugivores can be grouped into the above-mentioned five functional groups (carnivores, wild ungulates, domestic ungulates, pulp feeders and birds).
Data collection
During 2018 fruiting season, we collected data for 29 fruiting C. humilis individuals in both study sites (N = 9 for Matasgordas and N = 20 for Martinazo). As for 2019 fruiting season, data was collected for 15 C. humilis and 21 P. bourgaeana fruiting individuals, all located at Matasgordas site. Data on crop size was recorded for all selected individuals during both study seasons, using visual survey methods (VSM) [56-58]. Camera traps (LTL ACORN 5310A, detection range = 18 m) were installed to collect data regarding to frugivore species visits and use of fruits. Cameras were placed from three to five meters distance from the focal plants and were automatically activated any time a movement occurred, taking a three-photo sequence every second. Camera sampling effort for each individual plant (i.e. camera active days) was carried out during the time period since the fruits were ripe until there were not any fruits left or these were musty. For a given camera and vertebrate species, we considered successive visits separated by more than 5 minutes between them. Although 5 minutes might lead to recording the same individuals several times, we were interested in quantifying successive visits by the same individuals since such accumulated number of visits is likely to alter the behavior of other frugivores. Also, to test our results, we used a stricter criterion, considering independent species visits when the time window was higher than 30 minutes, and therefore individuals were less likely to be resampled [59-61]. For all considered visits, the following data was recorded: date, time, day number since camera was active, frugivore species and their use of the fruiting individual plant. We classified the use of individual plants by each frugivore visitor into two types: (i) recorded mammals and birds that were not physically interacting with the target plant (value = 0) and (ii) mammals or birds clearly interacting with the plant (most likely eating fruit; value = 1) (see S1B and S1C File). Accordingly, we described as total number of visit data all types of plant use (i.e. value = 0 and 1), whilst total number of interaction data refers only to those in which visitors were clearly interacting with the plant (i.e. value = 1). Data on sampling effort days and total number of photographs used in the analysis can be found in S1 Table.
Analytical approach
To explore our data, we estimated the total number of visits and interactions (as defined above) for each functional group of frugivores (carnivores, wild ungulates, domestic ungulates, pulp feeders and birds) for every focal individual plant. We considered, for each study site, the same pool of vertebrate species due to their high mobility and therefore we did not predict changes in frugivore community structure within each site. Environmental conditions, related to each plant’s micro or meso habitat (e.g. crop size), may affect the visitation patterns of different frugivore species [62]. Thus, we assessed the relation between fruit availability (crop size) and total number of visits and interactions by each functional group using Pearson’s correlation coefficient applying the function corrplot implemented in the R package corrplot [63].Because the same frugivore individual could have been photographed several times within a short time period, the analyses described below (i.e. null models) were carried out considering photographs of a given frugivore species at a given individual plant separated by at least 5 minutes (or 30 minutes in a second set of analyses). Our approach was first applied to the whole set of visit data and then to the subset of data corresponding to physical interactions between frugivores and C. humilis and P. bourgaeana individuals separately.We assessed the potential interactions between frugivore species by means of three spatially explicit null models that allow coping with the effect of each individual plant’s spatial location:
Null model 1
We compared the average number of visits by a frugivore species (sp1) to individual plants visited (PV) and not visited (PNV) by a second frugivore species (sp2). Under the null hypothesis of no interaction between sp1 and sp2, and all else being equal, the average number of visits by sp1 to PV and PNV by sp2 should not differ. If for a given pair of frugivore species the average number of visits by sp1 to PV was significantly larger or smaller than to PNV by sp2 we assumed spatial attraction or avoidance, respectively, between the pair of frugivore species (i.e. sp1 and sp2). For each pair of frugivore species, we examined potential significant differences between the number of visits by one particular species to individual plants visited and not visited by the paired species by fitting Generalized Linear Models with Poisson distribution and log link function [64]. In a second set of analyses, we added “crop size” as a co-variable in the models. Since the new results did not vary regarding the models without the covariate, for major simplicity, we omitted these results.
Null models 2 and 3
To examine temporal interactions among different frugivore species, we compared the observed mean values of time differences between successive visits by different frugivore species (OMTD) with the 95% percentile of the expected time difference value (95%ETD) based on two different null models [21]. We considered aversive or attractive responses between pairs of frugivore species when OMTD was above or below the 95%ETD limits, respectively. When OMTD was within 95%ETD limits we concluded no significant response between pairs of frugivore species [21].The metric used to determine observed time differences was the time elapsed (in hours) between the visit of one frugivore species (sp1) until the first visit of a second frugivore species (sp2; i.e. minimum time elapsed between successive interspecific visits). Because we considered the possibility that, for example, sp1 altered the temporal pattern of the plants visits and interactions by sp2 but, for example, sp2 did not alter the temporal pattern of the plants visits and interactions by sp1, the order of species occurrence was taken into account. Thus, for each pair of species, we calculated the time elapsed between a specific species to another (e.g. sp1-sp2) and vice versa (e.g. sp2-sp1). This metric was determined considering the all camera traps with valid data (two camera traps at Martinazo site were discarded) and for all sampling effort days. For all analyses, we only considered visits and interactions of frugivore pairs separated by < 36 hours, as we assume there would not be any species interactions in a greater time window. Interactions between pair of species were only contemplated when the number of observations for OMTD and 95%ETD was ≥4.For null model 2, the expected minimum time elapsed between successive interspecific visits was calculated for each target pair of frugivore species as for OMTD but considering their occurrences at different and spatially independent (i.e. >100m or >200m apart) plants (S1 Fig). By considering frugivore occurrences at two plants substantially separated we ensured the condition of no potential direct interactions between such frugivore visitors.For null model 3, calculation of expected time differences between pairs of interspecific frugivore visits was based on the randomization of frugivore occurrence (and timing of occurrence) for each individual plant (S2 Fig). To do so, we run one thousand iterations that assigned occurrence (i.e. visits or interactions) and their timing to a randomly chosen different individual plant. Our procedure preserved not only the observed number of occurrences of each frugivore species, but also their observed circadian rhythms as the observed times linked to each frugivore occurrence (i.e. recorded picture) were not altered. After each iteration, we calculated the time difference between such simulated co-occurrences.Because results for both C. humilis fruiting seasons were consistent, so as not to be redundant, analyses were carried out uniting data of both seasons. Our analytical approach was performed using free software R 3.5.0 [65]. The implemented packages used to tailor the null models were plyr [66], dplyr [67] and lubridate [68]. See S2 File for the main functions used in the analysis.
Results
Chamaerops humilis
Do pairs of frugivore species visit the same fruiting C. humilis? (null model 1)
For the complete set of visits in a time window higher than 5 minutes, we found that, in decreasing order, the functional groups with higher number of visits were wild ungulates, domestic ungulates, carnivores, birds and pulp feeders. However, for interaction data the most frequent frugivores were wild ungulates, carnivores, pulp feeders, birds and domestic ungulates (S2 Table).For the complete visit data, only total number of carnivores’ visits was positively correlated with crop size (r5’ = 0.75, n5’ = 44, P5’ <0.001). Regarding to data corresponding to total number of interactions with C. humilis we found that crop size was positively correlated with total number of interactions (r5’ = 0.57, n5’ = 44, P5’ <0.001), as well as with carnivores (r5’ = 0.78, n5’ = 44, P5’ <0.001) and wild ungulates (r5’ = 0.44, n5’ = 44, P5’ <0.01).In relation to all visits, we found evidence of spatial attraction and segregation in the use of C. humilis by some predator-prey species (Fig 2). For instance, on average, badgers visited 2.28 times more often those palms visited by lagomorphs (Fig 2a). Also, lagomorph’s visits to palms not visited by red foxes were 8.6 times higher as compared to the mean number of visits to palms visited by red foxes (Fig 2b). Our results showed that palms visited by large ungulates, were more often visited by medium-sized frugivores. For example, red foxes visited palms which were visited by cows 3.34 times more often than those that were not (Fig 2c) and also, visited palms which were visited by horses 7.90 times more often than those that were not (Fig 2d).
Fig 2
Difference in Chamaerops humilis visit patterns for frugivore pairs of species (null model 1).
The observed values (black bars) represent the mean number of visits by a frugivore species (sp1) to individual plants visited (PV) by a second frugivore species (sp2). The expected values (grey bars) represent the mean number of visits by sp1 to plants not visited (PNV) by sp2. (* P<0.05, ** P<0.01, *** P<0.001).
Difference in Chamaerops humilis visit patterns for frugivore pairs of species (null model 1).
The observed values (black bars) represent the mean number of visits by a frugivore species (sp1) to individual plants visited (PV) by a second frugivore species (sp2). The expected values (grey bars) represent the mean number of visits by sp1 to plants not visited (PNV) by sp2. (* P<0.05, ** P<0.01, *** P<0.001).
Are there differences between observed and expected minimum time elapsed between successive interspecific visits to C. humilis? (null model 2)
When considering the whole set of visits in a time window higher than 5 minutes, null model 2 showed significant interactions between some frugivore pairs. Specifically, the observed mean difference between the time of cow and red fox visits was lower (0.16 times for 100m and 0.14 times for 200 m) than the expected mean difference, therefore indicating strong temporal attraction between both species (Fig 3A and 3B). Also, the observed mean difference between the time of cow and bird was between 0.93 and 0.90 times lower than expected (depending on the criteria used), indicating a moderate attraction between them (Fig 3A and 3B). Conversely, the observed mean differences between the visitation times of horse and badger was 1.94 times greater than the expected mean differences for this frugivore pair, indicating temporal aversion (Fig 3B). As to the subset of data corresponding to interactions between frugivores and C. humilis, and separated at least 100 or 200 meters, we found a significant and strong attractive response for the boar-badger pair, which was 0.35 times lower than the expected value for both criterions (Fig 3C and 3D). Also, we found that for a separation of at least 200 meters, that the fox-horse pair showed a significant and attractive response, with the observed mean time difference between the visitations being only 0.62 times the expected value (Fig 3D).
Fig 3
Significant interactions between frugivores of Chamaerops humilis resulting from applying null model 2.
Black dots indicate the observed mean time differences (OMTD). Black lines and grey bars represent the expected 95% and 90% intervals of time elapsed between visits, respectively. (A) Interactions derived from the whole set of visit data, using C. humilis individuals at a minimum distance of 100 m. (B) Interactions derived from the whole set of visit data, using C. humilis individuals at a minimum distance of 200 m. (C) Interactions derived from data corresponding to frugivores physical interactions with C. humilis at a minimum distance of 100 meters. (D) Interactions derived from data corresponding to frugivores physical interactions with C. humilis at a minimum distance of 200 meters.
Significant interactions between frugivores of Chamaerops humilis resulting from applying null model 2.
Black dots indicate the observed mean time differences (OMTD). Black lines and grey bars represent the expected 95% and 90% intervals of time elapsed between visits, respectively. (A) Interactions derived from the whole set of visit data, using C. humilis individuals at a minimum distance of 100 m. (B) Interactions derived from the whole set of visit data, using C. humilis individuals at a minimum distance of 200 m. (C) Interactions derived from data corresponding to frugivores physical interactions with C. humilis at a minimum distance of 100 meters. (D) Interactions derived from data corresponding to frugivores physical interactions with C. humilis at a minimum distance of 200 meters.
Are there differences between the observed and the expected time difference between interspecific frugivore visits to C. humilis? (null model 3)
When considering all visits within a time window higher than 5 minutes, the difference between the observed and the expected time differences between interspecific visits by frugivore pairs was significant in a few cases. Specifically, the observed mean difference between the pairs cow-red fox and horse-lagomorph visits was much lower (0.17 and 0.30 times, respectively) than the expected mean difference, thus indicating temporal attraction (Fig 4A). Besides, the observed mean differences between the visit times by the lagomorph-badger and lagomorph-rodent pairs was 1.40 and 1.15 times greater, respectively, than the expected mean differences, indicating temporal aversion (Fig 4A). Furthermore, when considering physical interaction data, we found an attractive response for the bird-boar pair, which observed mean difference was 0.33 times lower than the expected one (Fig 4B). Finally, we found a significant aversive response for the lagomorph-rodent pair, which was 1.33 times greater than the expected mean time difference, thus indicating aversion (Fig 4B).
Fig 4
Significant interactions between frugivores of Chamaerops humilis resulting from applying null model 3.
Black dots indicate the observed mean time differences (OMTD). Black lines and grey bars represent the expected 95% and 90% intervals of time elapsed between visits, respectively. (A) Interactions derived from the whole set of visit data. (B) Interactions derived from data corresponding to frugivore physical interactions with C. humilis.
Significant interactions between frugivores of Chamaerops humilis resulting from applying null model 3.
Black dots indicate the observed mean time differences (OMTD). Black lines and grey bars represent the expected 95% and 90% intervals of time elapsed between visits, respectively. (A) Interactions derived from the whole set of visit data. (B) Interactions derived from data corresponding to frugivore physical interactions with C. humilis.
Pyrus bourgaeana
Do pairs of frugivore species visit the same fruiting P. bourgaeana? (null model 1)
For the complete set of visits in a time window higher than 5 minutes, we found that, in decreasing order, the functional frugivore groups with higher number of visits were wild ungulates, pulp feeders, birds and carnivores. However, for interaction data the most frequent frugivores were wild ungulates, pulp feeders, carnivores and birds (S2 Table).For the complete visit data, total number of carnivores’ and wild ungulate visits were positively correlated with crop size (r5’ = 0.54, n5’ = 21, P5’ <0.01 and r5’ = 0.52, n5’ = 21, P5’ <0.001, respectively). Regarding to data corresponding to total number of interactions with P. bourgaeana, we found that crop size was positively correlated with total number of interactions (r5’ = 0.52, n5’ = 21, P5’ <0.001), as well as with carnivores (r5’ = 0.55, n5’ = 21, P5’ <0.01) and wild ungulates (r5’ = 0.52, n5’ = 21, P5’ <0.001).In relation to all visits (Fig 5), we found significant evidence of spatial avoidance by lagomorphs which visited 0.98 times less often those P. bourgaeana trees visited by badgers (Fig 5a). Also, we observed that trees visited by large ungulates, were more often visited by medium or small-sized frugivores. For example, lagomorphs only visited trees that were visited by boars (Fig 5b) or red deer (Fig 5c).
Fig 5
Difference in Pyrus bourgaeana visit patterns for frugivore pairs of species (null model 1).
The observed values (black bars) represent the mean number of visits by a frugivore species (sp1) to individual plants visited (PV) by a second frugivore species (sp2). The expected values (grey bars) represent the mean number of visits by sp1 to plants not visited (PNV) by sp2. (* P<0.05, ** P<0.01, *** P<0.001).
Difference in Pyrus bourgaeana visit patterns for frugivore pairs of species (null model 1).
The observed values (black bars) represent the mean number of visits by a frugivore species (sp1) to individual plants visited (PV) by a second frugivore species (sp2). The expected values (grey bars) represent the mean number of visits by sp1 to plants not visited (PNV) by sp2. (* P<0.05, ** P<0.01, *** P<0.001).
Are there differences between observed and expected minimum time elapsed between successive interspecific visits to P. bourgaeana? (null model 2)
When taking into account the whole set of visit data, within a time window higher than 5 minutes, null model 2 showed only one significant response regardless of the spatial independence distance criterion. Specifically, the observed mean difference between the time of fallow deer and red deer visits was lower (0.13 times for 100 and 200 meters) than the expected mean difference for this frugivore pair, therefore indicating temporal attraction.
Are there differences between the observed and the expected time difference between interspecific frugivore visits to P. bourgaeana? (null model 3)
When considering all visits within a time window higher than 5 minutes, the difference between the observed and the expected time differences between interspecific visits by frugivore pairs was significant in only a few cases. Specifically, the observed mean time difference between the pairs red deer-red fox, lagomorph-fallow deer and fallow deer-red deer visits was always lower (0.73, 0.66 and 0.09 times, respectively) than the expected mean difference, thus indicating temporal attraction (Fig 6A).
Fig 6
Significant interactions between frugivores of Pyrus bourgaeana resulting from applying null model 3.
Black dots indicate the observed mean time differences (OMTD). Black lines and grey bars represent the expected 95% and 90% intervals of time elapsed between visits, respectively. (A) Interactions derived from the whole set of visit data.
Significant interactions between frugivores of Pyrus bourgaeana resulting from applying null model 3.
Black dots indicate the observed mean time differences (OMTD). Black lines and grey bars represent the expected 95% and 90% intervals of time elapsed between visits, respectively. (A) Interactions derived from the whole set of visit data.Similar patterns were found when applying the 3 null models for visit and interaction data within a time window higher than 30 minutes. Thus, regardless of the criteria used to select the dataset, for most cases, frugivore pairs used the same C. humilis and P. bourgaeana fruiting individuals. Most of the resulting interactions were consistent when restricting the analysis to a narrower time window (30 minutes). Further details on the complete outcome, figures and statistical summaries (for both 5 and 30 minutes) can be found in S3 File and S3 Table, respectively.
Indirect effects on plant reproductive success
As a result of summing up the outcomes of null models 2 and 3, we assessed both direct and indirect interaction pathways among C. humilis and P. bourgaeana frugivores’ in relation to these plants reproductive success. For instance, we found that large ungulates such cows or red deer, had a direct (mostly) negative effect on seed dispersal as they grind most ingested seeds (Fig 7A). However, cows had a direct positive effect for legitimate seed dispersers, such as red foxes which were attracted to palms visited by this large ungulate and thus, lead to an indirect positive effect on C. humilis’ reproductive success (Fig 7A). Our results also showed a similar situation for the interactions between red deer, red foxes, and P. bourgaeana (Fig 7A). Finally, our study also disclosed how, for instance, lagomorphs which mainly had a direct negative effect on C. humilis reproductive success acting as pulp feeders, also had a direct negative effect on seed predators such as red deer. Thus, these direct interspecific interactions lead to an indirect positive effect on C. humilis reproductive success (Fig 7B).
Fig 7
Effects on C. humilis and P. bourgaeana’s reproductive success.
Direct and indirect interaction pathways between pairs of frugivore species in relation to Chamaerops humilis and Pyrus bourgaeana’s reproductive success based on interaction patterns obtained from null models 2 and 3.
Effects on C. humilis and P. bourgaeana’s reproductive success.
Direct and indirect interaction pathways between pairs of frugivore species in relation to Chamaerops humilis and Pyrus bourgaeana’s reproductive success based on interaction patterns obtained from null models 2 and 3.
Discussion
This study represents one of the first assessments of whether and how functionally diverse vertebrate frugivores alter fruiting plant interactions with subsequent frugivore associates [but see 18]). Specifically, our results revealed, for example, how some frugivore prey (e.g. lagomorphs) avoid those C. humilis individuals most often visited by one of its major predators (e.g. red foxes [32]). Additionally, we revealed temporal attraction between large-sized frugivores (e.g. cows, red deer) and medium-sized frugivores (e.g. red foxes) suggesting that ungulates facilitate C. humilis and P. bourgaeana exploitation to other frugivores by making fruits more accessible. Our analytical approach, based on camera trap data and tailored null models, has also allowed us to identify a number of interactions among frugivores with potential indirect effects on C. humilis and P. bourgaeana’s reproductive success. Our approach can be widely applied by taking advantage of the large amount of data generated by numerous (on-going and completed) camera-traps surveys done all over the world [25]. This will certainly foster a most comprehensive understanding of the direct and indirect effects of interspecific interactions on ecosystem functioning within and across spatial scales.
Plant crop size variation and frugivore visitation
Our results revealed that frugivores respond to crop sizes of fruiting plants, by visiting more frequently those plants bearing more fruits. These results are in line with previous studies that have also found a positive relationship between frugivore visitation and plant resource availability [see 69 and references therein]. Interestingly, the positive relationship between crop size and frugivore visitation was mostly driven by two functional groups of frugivores (i.e. carnivores and wild ungulates). Carnivores (i.e. badgers and foxes) are the most effective seed dispersers of C. humilis and P. bourgaeana [29, 30], whereas wild ungulates can behave as seed predators or seed dispersers [29-31]. Therefore, a greater dispersal success of individuals bearing more fruits due to higher visitation of the most effective dispersers can be limited depending on whether ungulates behave mostly as dispersers or as seed predators [70]. Other environmental conditions derived from the micro (e.g. plant’s location, herbaceous cover, distance to shrub and light availability) and meso habitat (e.g. elevation and type of forest cover) related to each individual plant as well as each animal’s niche differences, that allow the coexistence of consumer species, could also exert an effect on the spatial pattern of frugivore visitation [62, 71]. In our system, most of the species (carnivores and ungulates) show high mobility, being able to travel several kilometers within the same day. Therefore, and given the relatively small spatial scale of our study sites, it can be assumed that different species of frugivores had access to all fruiting plants. This scenario however differs for rodents and lagomorphs since they are less mobile; thus, a few cases of spatial avoidance by small mammals identified (null model 1) should be considered with caution.
Competitive, predatory and facilitative interactions among functionally diverse frugivores
Decrease in fruit display due to exploitation by frugivores is expected to lessen subsequent visitation rates by other fruit consumers [e.g., 72]. In this line, null model 1 results showed that lagomorphs visited less often palms visited by boars, and that red deer visited less often palms visited by rodents, likely because they lessened available ripe fruits. Predator-prey interactions may also affect the spatial and temporal foraging patterns by frugivores [73-75]. Results from null model 1 revealed that lagomorphs tended to avoid C. humilis individuals visited by red foxes, likely to reduce predation risk. Also, null models 2 and 3 suggest that badgers seemed to temporally avoid palms visited by lagomorphs. This unexpected result turns less surprising if we consider the fact that badgers do not generally prey upon adult lagomorphs (i.e. the ones that feed on C. humilis fruit) but upon small juveniles by digging them out from their burrows [32].Facilitative interactions in the acquisition of food are highly prevalent among many terrestrial vertebrates and have critical importance in structuring and function of many communities [53–55, 76, 77]. In contrast with Carreira et al.’s [27] findings, we detected that medium-sized frugivores tended to be temporally attracted by larger ungulates. For instance, null models 2 and 3 indicate that red foxes visited those palms previously visited by cows more often than would be expected under the null hypothesis of no interaction between both mammals. This pattern could be a result of ungulates giving up ripe fruits in the palm’s immediacy or reducing the number of defensive needle-like spines in fruiting C. humilis (Authors personal observations) making its fruits more easily accessible to such medium-sized frugivores. Also, null model 3 showed that red foxes seem to be attracted to P. bourgaeana individuals previously visited by red deer. This could relate to red deer plucking the fruits from the trees leaving fallen ripe fruits within the tree’s immediacy (Authors personal observations; see S1C File) which might attract smaller frugivores. Our study shows how this functionally diverse assemblage of frugivores leads to interspecific facilitation in foraging, a widespread pattern in other temperate and tropical habitats [55, 76, 77]. Positive interactions may be an important mechanism linking high diversity to high productivity under stressful environmental conditions [54, 78]. Identifying the specific mechanisms by which frugivore species sense each other is out of the scope of the present investigation, although, signals such odors and trail and territory marks, amongst others, are frequently perceived by numerous species of terrestrial vertebrates [79-83] and may be involved in our system.
Community level and indirect effects on C. humilis and P. bourgaeana reproductive success
Our approach allowed us to identify direct interactions among frugivores of C. humilis and P. bourgaeana but also potential indirect effects on the reproductive success of plant populations. These indirect effects can either be positive or negative, depending on whether and how they change subsequent visits by antagonistic or mutualistic animal associates. This is an important consideration since the fruits of most plant species are consumed by functionally diverse assemblages of frugivores such as legitimate seed dispersers, seed predators, and pulp feeders [52, 84–86]. As predicted, seed predators and pulp feeders can alter the interaction frequency between fleshy-fruited plants and their mutualistic and antagonistic animal associates [18]. In the first case, ungulate seed predators increased the interaction frequency with legitimate seed dispersers, for both C. humilis and P. bourgaeana yielding thus an indirect positive effect on seed survival (Fig 7A). Secondly, pulp feeders, that often lessen long-distance dispersal [52], decreased C. humilis’ foraging by ungulate seed predators, enhancing thus seed survival and fruit removal by mutualistic animal associates (Fig 7B). Therefore, both, mutualistic and antagonistic partners that shape the interaction outcomes among individual plants lead to contrasting seed dispersal success [18, 70], appearing to shape a complex web of direct and indirect effects which’s net effect is most likely dependent on the community context [52]. This outcome may be strongly dependent on initial population densities that would most likely alter the dynamics of the system, thus, the net effect of dispersal success, among other factors, will probably be a result of the abundance of different functional guild of frugivores (legitimate seed dispersers, pulp feeders and seed predators) [87-89].
Conclusions
The combined use of camera trap data and tailored null models has proved to be an effective tool to assess positive and negative interactions between functionally diverse frugivore in relation to a shared resource and infer direct and indirect interaction pathways that influence the reproductive success of fleshy-fruited plants. This new approach may provide advantages and complementarities to ongoing research and methodologies for multispecies interactions such as multilayer networks or spatially explicit agent-based simulation modeling [19–20, 89]. For instance, by relaying on data from camera-trapping, our approach can be applied to uncover interspecific interactions among understudied nocturnal or otherwise elusive animals such as many mammals and other secretive vertebrates. Our analytical approach enabled us to evaluate the specific hypothesis and predictions that we had projected for our particular study system and would hardly have been able to achieve using other similar approaches [see 26–28] that impose a number of requirements and assumptions that did not always fit our specific system.This study may also foster a greater understanding of the implications of multispecies plant-animal interactions on ecosystem function and restoration by means of direct and indirect interaction pathways that influence seed dispersal and recruitment in disturbed areas [90]. It can also enrich a great potential of interaction network researches predicting indirect novel interactions or changes in functional network structures under global changing conditions (e.g. land use changes, human disturbance and climate change) [35]. Indeed, although we have focused here on interactions among frugivores and two fruiting plants, our approach can be easily expanded at a community level by increasing the sampling effort (i.e. number of camera traps). Finally, it can also be applied to other varied systems and species such as, for example, plant-herbivore systems [54], plant-pollinator systems [91], or carcass-scavenger systems [92], taking advantage of the large amount of camera-trap surveys that are collecting data all over the world [25].
Classified frugivore plant use.
A) Red deer plucking the fruits from Pyrus bourgaeana. B) Frugivores not interacting with the plant (value = 0). C) Frugivores physically interacting with the plant, most likely eating its fruits (interaction value = 1).(PDF)Click here for additional data file.
R functions and packages.
Main functions for tailored models 1, 2 and 3 used in the analysis.(PDF)Click here for additional data file.
Complete outcome, figures and statistical summaries (for both 5 and 30 minutes).
A) Difference in Chamaerops humilis visit patterns for frugivore pairs of species within a time window higher than 5 minutes (null model 1). B) Difference in Chamaerops humilis visit patterns for frugivore pairs of species within a time window higher than 30 minutes (null model 1). C) Zoom of difference in Chamaerops humilis visit patterns for frugivore pairs of species within a time window higher than 5 minutes (null model 1). D) Difference in P. bourgaeana visit patterns for frugivore pairs of species within a time window higher than 5 minutes (null model 1). E) Difference in P. bourgaeana visit patterns for frugivore pairs of species within a time window higher than 30 minutes (null model 1). F) Zoom of difference in P. bourgaeana visit patterns for frugivore pairs of species within a time window higher than 5 minutes (null model 1). G) Significant interactions between frugivores of Chamaerops humilis resulting from applying null model 2 within a time window higher than 30 minutes. H) Significant interactions between frugivores of Chamaerops humilis resulting from applying null model 3 within a time window higher than 30 minutes. I) Significant interactions between frugivores of Pyrus bourgaeana resulting from applying null model 3 within a time window higher than 30 minutes.(PDF)Click here for additional data file.
Null model 2.
Dwarf palms (Chamaerops humilis) A and B are spatially independent individuals, in this case, separated by 100 meters from each other (the set of analysis was also run for palm trees separated 200 meters for a stricter spatial independence criterion). For this example, a badger has been recorded at dwarf palm 1 at the time bt0. For dwarf palm 2, two species have been recorded twice each: as for the first species, a fox has been recorded at time ft1 and time after, another fox has been recorded at time ft2. Whilst the second species was a horse recorded at time ht3 and after a period of time, another horse has been recorded at time ht4. To calculate expected time differences between pairs of interspecific species we have only considered the first successive visit of a different species. Therefore, for this example we have calculated time elapsed between b t0—ht3 and bt0- ft1. This scheme also applies to the Iberian pear (Pyrus bourgaeana) tree system.(TIF)Click here for additional data file.
Null model 3.
A, B and C represent Chamaerops humilis individuals at which one or both target frugivore groups have been recorded. For this example, a badger has been recorded at dwarf palms A (at time bt1) and B (at time bt2), and a red fox has been recorded at dwarf palms A (at time ft1), B (at time ft2) and C (at time ft3). To calculate expected time differences between the pair of interspecific species, we randomly assigned occurrence of frugivore groups (and the timing of the occurrences) by shuffling them 1000 times. When, by chance, both frugivore species (badger and red fox in this case) meet at the same C. humilis individual, we calculated time differences between such co-occurrence. In this example we have calculated time elapsed between bt1 and ft2 for C. humilis A and between bt2 and ft1 for C. humilis B. This scheme also applies to the Iberian pear tree (Pyrus bourgaeana) system.(TIF)Click here for additional data file.
Sampling effort days and total number of photographs used in the analysis for both C. humilis and P. bourgaeana.
(PDF)Click here for additional data file.
Overall patterns.
Total number of visits and interactions for each frugivore functional group and plant species for both 5 and 30 minutes between successive species records.(PDF)Click here for additional data file.
Statistical summary for null models 2 and 3.
Table 1) Chamaerops humilis 5 minutes statistical summary for null model 2. Table 2) Chamaerops humilis 30 minutes statistical summary for null model 2. Table 3) Chamaerops humilis 5 minutes statistical summary for null model 3. Table 4) Chamaerops humilis 30 minutes statistical summary for null model 3. Table 5) Pyrus bourgaeana 5- and 30-minutes statistical summary for null model 2. Table 6) Pyrus bourgaeana 5- and 30-minutes statistical summary for null model 3.(PDF)Click here for additional data file.14 Jul 2020PONE-D-20-17034Unravelling interspecific interactions among frugivores: a new approach based on camera-trap data and tailored null modelsPLOS ONEDear Miriam SelwynThank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.Two reviewers have evaluated your manuscript and made constructive contributions. While they found valuable information and were positive, also showed concern on several major issues, so it requires MAJOR CHANGES for your manuscript does meet our criteria for publication. In particular, Reviewer #2 had several major concerns with data analysis.I mostly agree with the reviewers, and added some comments. You need to solve these shortcomings of the manuscript before it can be considered suitable for publication in this journal. If you can work on the comments, we are looking forward to a new version.Please submit your revised manuscript by August 25th 2020. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.Please include the following items when submitting your revised manuscript:A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocolsWe look forward to receiving your revised manuscript.Kind regards,Pedro G. Blendinger, PhDAcademic EditorPLOS ONEJournal Requirements:When submitting your revision, we need you to address these additional requirements.1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found athttps://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf andhttps://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf2. Please note that in order to use the direct billing option the corresponding author must be affiliated with the chosen institute. Please either amend your manuscript or remove this option (via Edit Submission).Additional Editor Comments:Consider modifying the title, so that it is not so general and fully reflects the main objectives of the study; see also the comment of Reviewer # 1.Avoid making lack of knowledge a strong point of the study (e.g. lines 53-54, 59-62, 70-71). Instead, highlight why this new knowledge is important, what its contribution is, and how it enables progress on ecological theory.Lines 88-96. Significantly reduce the species description in the Introduction. For the reader unfamiliar with the species on your study site, it is more important to know the functions of the species than their identity. Reviewer's # 1 recommendation on citing evidence of competitive interactions in the system should be included in Methods.Line 132. Please inform on the polydrupe size.Lines 178-191 should be condensed but be careful not to omit relevant information.Line 217. Use italics for “expected time difference”.I do not understand why you show plant crop size variation and frugivore visitation as a separate result. This is confusing and contributes to unnecessarily complexing a manuscript that is already quite complex. Although it is an interesting topic, it is not one of your objectives. In this study, crop size should be more a source of variation to be controlled than a main effect to be evaluated.The potential influence on visit frequency of the fruiting neighborhood does not appear to have been considered. This may be an important limitation of the study; I would expect to see some consideration in this regard in the Discussion. Along the same line, see the comments of both reviewers about the limitations of a non-spatially explicit design in interpreting the results.Results are hard to follow, and even more difficult to interpret without falling into successive ad-hoc explanations. I recommend reducing them, keeping only those most closely related to the main objective and the proposed conceptual framework. In the context of plant-animal interactions in which you framed the study (e.g. lines 76-78), interactions are the most important, not visits. On the other hand, it appears that a 5 min interval between visits may better reflect an effect of responses to previous visitors than a 30 min interval. Therefore, I suggest that you only focus the results in number of interaction using the 5 min as a reference. The other results can be omitted entirely, or can be reduced to the minimum possible in the main text, referring to the supplementary material to see the complete complementary analyzes. This includes the corresponding figures and tables.The Discussion is somewhat poor and does not take advantage of the potential of the results. For example, it could be greatly enriched by focusing more on functional interpretation and the implications for community organization.Figure 8 should move to results.Finally, there is a marked tendency to self-refer, and not always because they are the only references available. It is important that you smooth this trend significantly.==============================[Note: HTML markup is below. Please do not edit.]Reviewers' comments:Reviewer's Responses to QuestionsComments to the Author1. Is the manuscript technically sound, and do the data support the conclusions?The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.Reviewer #1: YesReviewer #2: Partly**********2. Has the statistical analysis been performed appropriately and rigorously?Reviewer #1: YesReviewer #2: No**********3. Have the authors made all data underlying the findings in their manuscript fully available?The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.Reviewer #1: YesReviewer #2: Yes**********4. Is the manuscript presented in an intelligible fashion and written in standard English?PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.Reviewer #1: YesReviewer #2: No**********5. Review Comments to the AuthorPlease use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)Reviewer #1: This manuscript addresses in detail how animal functional groups interact with two fruiting plants, and how some functional groups influence the exploitation of those fruiting plants on a temporal and spatial scale in the Doñana National Park, SW of the Iberian Peninsula. More importantly, the authors provide new a new approach by combining camera trapping and null models to figure out those interactions. It appears that some functional groups of frugivores have negative or positive effects on the exploitation of fruit plants by others and this may have consequences on the reproductive success of fruiting plants. The study opens a new door for studying plant-animal interaction and seed dispersal at the community level to understand the functioning of ecosystems. I was particularly interested in the level of details given in the result section. However, some aspects need clarification in the abstract, introduction, methods, and discussion.AbstractLines 16-20: This sentence refers to lines 73-75. However here the word “combination” that makes the approach new according to me is lacking.Lines 30-32: This objective seems to be part of the research. So I would suggest considering revising the third sentence of the abstract (lines 15-16) to include this aspect and give here the outcome of this analysis, instead of stating the objective.Line 35 (carcasses): Maybe "fruit remains", it is generally called "feeding remains"IntroductionLines 51-53: Please can the author provide some examples of studies of different types of interactions taking place in the same habitat?Lines 78-82: a) the authors only treated this in part in the abstract. Please provide information on the outcome of the first step in the abstract. b) Please consider revising the tense of the verbs in both sentences. The first is in the present tense and the second in the past tense.Lines 82-84: a) The way this research question is stated (using "how") directly means that we already acknowledge that there would be an effect. I would suggest saying, "we inferred the consequence of these direct responses on the plants", instead of "how they would affect".b) This question seems to form an integral part of the present research, as we can see in the second sentence of the abstract. Although it is mention in the abstract, it is not clear from the title of the manuscript that the interaction between the frugivores and plants will be analyzed.Line 85: Please consider revising the tense. Some other mistakes may be found elsewhere.Lines 96-99: The authors have presented the different species or groups of species forming their study system (Lines 85-96). However, there is no clear evidence for the reader why the authors think that the visitation by some species may influence the visitation by other species remain unclear. It would good to set the scene by first talking about "niche partitioning" and "ecological plasticity" that allow ecologically similar species to coexist.These articles may help:Salinas‐Ramos, V.B., Ancillotto, L., Bosso, L., Sánchez‐Cordero, V. and Russo, D. (2020), Interspecific competition in bats: state of knowledge and research challenges. Mam Rev, 50: 68-81. doi:10.1111/mam.12180Ruczyńki L, Zahorowicz P, Borowik T, Hałat Z (2017) Activity patterns of two syntopic and closely related aerial‐hawking bat species during the breeding season in Bialowieza Primaeval Forest. Mammal Research 62: 65– 73.Schimpp SA, Li H, Kalcounis‐Rueppell MC (2018) Determining species specific nightly bat activity in sites with varying urban intensity. Urban Ecosystems 21: 541– 550.Line 101: Please clarify in which context the differential visitation of different fruiting plants by different frugivores will contribute to lessening predation risk. To my knowledge, no species was presented in the system as a predator of another in the same system.Lines 101-103: As mentioned in my previous comment of line 101, it is only here that it appears that some species are the prey of others in this study system. Please make it clear in the description of the study system (lines 85-96). It would also be good to mention some evidence from the literature showing which species in the system are known to compete for food.Lines 103-105: It is difficult to understand in which ecological context the authors are testing this hypothesis. The authors may provide some evidence or state a theory that will help understand the relationship between the larger and smaller frugivores.Material and methodsLine 109: Can the author please present a map of the study area so that the reader can quickly have an idea on the scale of the study?Line 150: In my opinion, this prediction should be in the introduction, and may certainly help understand hypothesis 3, as in the previous comment. Giving evidence from the literature will be important.Lines 160-163: I would like to know if there any reason why data were not collected for C. humilis at Martinazo in 2019, and why data where not collected for P. bourgaeana in 2018?Line 163: a) Was there any reference size for each species to be selected in the study sample?b) Until the end of the document, I have not seen what was the importance of measuring the plant size. Please can you clarify?Line 205: A part of the aim of this research seems to be an example of an analytical approach for the interactions between animal communities and plants. However, the reason for the choice for the main statistical approach (Null model) has not been made clear to the reader. It would be good to know if there exists another method and why the choice for this particular method.Lines 206-2012: Please consider revising. The relationship between the sp1 and the sp2 is not very clear.Line 250: For reproducibility of the method, I encourage the authors to indicate the packages used for the analyses, as well as the principal functions.ResultsLine 254: When talking about "the most abundant visitors", are you referring to the species with the highest number of visits? If so, please consider revising.Discussion1) The results showed the influence of some animal species on the visitation rates of some individual plants by other animal species. However, the authors did not discuss the fact that the spatial locations of individual plants coupled with the spatial distribution of the different animal functional groups may influence such interaction; Although all animal species visited all plant individuals. I suggest discussing this aspect and including it as a limit of the study.I suggest reading these articles:García, D., Zamora, R. and Amico, G.C. (2011), The spatial scale of plant–animal interactions: effects of resource availability and habitat structure. Ecological Monographs, 81: 103-121. https://doi.org/10.1890/10-0470.1Morales, J.M. and Vázquez, D.P. (2008), The effect of space in plant–animal mutualistic networks: insights from a simulation study. Oikos, 117: 1362-1370. https://doi.org/10.1111/j.0030-1299.2008.16737.x2) Secondly, the authors presented their research using a new approach (combining camera traps and null models). However, they did not discuss their approach with other approaches in the literature and present the advantages of their approach to those in the literature.Line 542: Change "such" into "such as"Reviewer #2: This study presents a framework to evaluate interspecific interactions (attraction, avoidance or indifference) among homeoterms vertebrates across a meta-region harboring population of a palm (Chamaerops humilis) and a flesh-fruited tree species (Pyrus bourgaeana). This assay was based on camera-trapping during two seasons of plant fruiting contrasting the empirical results with three approaches of null models. I have only a few, but rather fundamental, concerns about this study, as below. Moreover, some parts of the paper are hard to understand or unnecessarily long, for this, I suppose that the paper audience will be better attracted to a leaner version. I hope that my suggestions can contribute to the improvement of the manuscript.Major comments:1) Once your approach is spatially explicit at individual-plant-scale, some environmental characteristics can affect your results? For example, micro- and meso-habitat features at each plant buffer may bias the results in terms of species interactions. If these factors were not considered or controlled in your analysis, I recommend acknowledging this issue in a paragraph in the discussion section. If you have some environmental or spatial co-variables, I suggest to include that in the regression models to understand the role of environmental features in the influence on vertebrates co-occurrence patterns.2) Further, you assumed that all sites have the same pool of vertebrate species? According to your introduction and details in methods, both fruits used by the widest spectrum of consumers, thereby underpinning the trophic structure according to consumer ecological traits (e.g., your functional groups). You can at least discuss this issue because the difference in the local pool of species (real or potential in the future) can cause a rearrangement or bias in the dynamics that you are evaluating.3) If you had data of camera-trapping during two successive resources seasons, what is the reason to only “infer” about the seed dispersal/frugivory? I suggest using the values on fruit removal as a function of species co-occurrence in each plant individual into a regression model. This can reveal an interesting contrast between trees that were patrolled more frequently by predators. I think that this quantitative approach can provide results to your implicit concern (i.e., how much the vertebrate (in)direct interactions influence the frugivory, and potentially the quantitative impacts on seed dispersal and plant fitness).4) I recognized that you evaluated the visitation according to crop size via correlation, but I suppose that the crop size should be addressed as a co-variable in your regression models to predict the (co)visits. I’m afraid this invalidates the mechanics or the interpretation of much of this very nuanced analysis because the vertebrates' attraction and indifference can be strongly linked to crop-size (as your correlation reveals).4) Once both sites have suffered human-induced alterations, do you have a recommendation derived from your findings for the conservation viewpoint? The novelty of your study is showing an effect of the co-occurrence between frugivores attracted in the space by flesh-fruited plants — an issue little addressed in the literature — but you are assuming a mesocosm equal for all site and plant-individual, for example: (i) the local pool of species; (iii) some environmental features; (iv) past human-induced effects (e.g. disturbance); and (iv) individuals crop-size in each temporal replica.5) I identified a disequilibrium between the sections of the manuscript. The manuscript is long but the discussion section is condensed to a few general paragraphs (6 considering the modest conclusion). Whereas the result section needs a rethink about the size and fashion. I regret that some specific comments about them are lacking, but I recommend rethink the results section because it is too long. Some specific results could be synthesized in the result section but better developed their ecological implications in the discussion.5) Figures: Although I recognize that there were several paired-interaction to show, I suggest that the figures of the manuscript deserve a reformulation to attract the paper audience. I encourage moving the current version of figures (e.g. Fig 1, Fig 2, Fig 5, and Fig 6) to the supporting information files. In doing so, you can select only the most important interactions to show in a new simplified figure avoiding different scales in the y-axis and providing the same color to each species.Minor comments1) L103-105: “and (iii) large-bodied frugivores (e.g. ungulates) would facilitate the plant exploitation to smaller frugivores by making the fruit more easily accessible (e.g. by giving up ripe fruits in the plants’ immediacy).”I try to understand this presumed facilitation relationship. It seems counter-intuitive because the tendency of large-bodied frugivores ingests the largest portion of fruits. Thus, the interaction expected for the small-bodied frugivores viewpoint is the competition. Do you need to rewrite this premise with more details (as I see in the S1 File).2) L163-165: “Data on plant size, as well as crop size was recorded for all selected individuals during both study seasons, using sector estimate techniques and if possible total count of fruits.”I strongly suggest referencing methodological sentences such as this example. Several seminal studies adopted this approach (and other approaches, such as lines 193-195). Further, this sentence apparently is repeated in lines 195-197.3) L212-214: “We examined potential significant differences between the average number of visits to PV and PNV in each pair of frugivore species, by fitting Generalized Linear Models with Poisson distribution and log link function.”You need to depict in more detail how these regression models were performed. Further, I do not understand the reason to use Poisson distribution to continuous data (average number of visits). Is the “average” a count data (sum) to justify the use of Poisson distribution?4) L249-250: “Our analytical approach was performed using free software R 3.5.0 [48].”Are the functions to perform the null models self-created by you or the analysis was based on a specific R package. Provide more details (but be cohesive) about the data analysis, to this, open space in the manuscript by synthesizing your results.**********6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.If you choose “no”, your identity will remain anonymous but your review may still be made public.Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.Reviewer #1: Yes: Luc Roscelin Dongmo TédonzongReviewer #2: No[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.23 Aug 2020Dear Editor,On behalf of my coauthors, I would like to thank you for the opportunity to revise and resubmit ourmanuscript entitled “Interspecific interactions among functionally diverse frugivores and their outcomes for plant reproduction: a new approach based on camera-trap data and tailored null models” (PONE-D-20-17034).We are also thankful to the Associate Editor Pedro G. Blendinger and the two reviewers for their valuable feedback on an early draft of this manuscript. We found the reviewers’ and the Editor's comments to be very helpful in revising the manuscript and have carefully considered and responded to each suggestion. In the majority of cases we were successful in incorporating the feedback into our revised manuscript. Please, find attached our Responses to Reviewers.We hope that this revised version now meets PLOS ONE's requirements and criteria for publication.Thank you again for your consideration of our revised manuscript.Sincerely,Miriam SelwynSubmitted filename: Selwyn et al. Response to Reviewers.pdfClick here for additional data file.30 Sep 2020Interspecific interactions among functionally diverse frugivores and their outcomes for plant reproduction: a new approach based on camera-trap data and tailored null modelsPONE-D-20-17034R1Dear Dr. José M. FedrianiWe are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.Within one week, you will receive an e-mail detailing the required amendments. When these have been addressed, you will receive a formal acceptance letter and your manuscript will be scheduled for publication.An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they will be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.Kind regards,Pedro G. Blendinger, PhDAcademic EditorPLOS ONEReviewers' comments:Reviewer's Responses to QuestionsComments to the Author1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.Reviewer #1: All comments have been addressedReviewer #2: All comments have been addressed**********2. Is the manuscript technically sound, and do the data support the conclusions?The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.Reviewer #1: YesReviewer #2: Yes**********3. Has the statistical analysis been performed appropriately and rigorously?Reviewer #1: YesReviewer #2: Yes**********4. Have the authors made all data underlying the findings in their manuscript fully available?The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.Reviewer #1: YesReviewer #2: Yes**********5. Is the manuscript presented in an intelligible fashion and written in standard English?PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.Reviewer #1: YesReviewer #2: Yes**********6. Review Comments to the AuthorPlease use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)Reviewer #1: (No Response)Reviewer #2: Selwyn and colleagues presented a much-improved version of the paper, embodying the major concerns raised in the first revision. The manuscript now is very better understandable and fashion. Thus, I do not have new concerns or recommendations for this version. I would like to congratulate the authors for the work.**********7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.If you choose “no”, your identity will remain anonymous but your review may still be made public.Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.Reviewer #1: Yes: Luc Roscelin Dongmo TedonzongReviewer #2: Yes: Juliano André Bogoni8 Oct 2020PONE-D-20-17034R1Interspecific interactions among functionally diverse frugivores and their outcomes for plant reproduction: a new approach based on camera-trap data and tailored null modelsDear Dr. Fedriani:I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.If we can help with anything else, please email us at plosone@plos.org.Thank you for submitting your work to PLOS ONE and supporting open access.Kind regards,PLOS ONE Editorial Office Staffon behalf ofDr. Pedro G. BlendingerAcademic EditorPLOS ONE
Authors: Jose A Barasona; M Cecilia Latham; Pelayo Acevedo; Jose A Armenteros; A David M Latham; Christian Gortazar; Francisco Carro; Ramon C Soriguer; Joaquin Vicente Journal: Vet Res Date: 2014-12-12 Impact factor: 3.683
Fig 1
Fig 2
Fig 3
Fig 4
Fig 5
Fig 6
Fig 7