Next Article in Journal
Diversity Patterns Associated with Varying Dispersal Capabilities as a Function of Spatial and Local Environmental Variables in Small Wetlands in Forested Ecosystems
Next Article in Special Issue
Seed Shadows of Northern Pigtailed Macaques within a Degraded Forest Fragment, Thailand
Previous Article in Journal
Phylogenetic Position of Geosmithia spp. (Hypocreales) Living in Juniperus spp. Forests (Cupressaceae) with Bark Beetles of Phloeosinus spp. (Scolytinae) from the Northeast of Mexico
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Forests
Volume 11
Issue 11
10.3390/f11111144
forests-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Seed Removal Rates in Forest Remnants Respond to Forest Loss at the Landscape Scale

by
Alain Hambuckers
1,*,
Franck Trolliet
1,
Astrid Simon
1,
Eliana Cazetta
2 and
Larissa Rocha-Santos
2
1
UR SPHERES, Biologie du Comportement, Université de Liège, 22 Quai Van Beneden, 4020 Liège, Belgium
2
Programa de Pós Graduação em Ecologia e Conservação da Biodiversidade, Laboratório de Ecologia Aplicada a Conservação, Universidade Estadual de Santa Cruz-UESC, Rodovia Jorge Amado km 16, Ilhéus, BA 45662-900, Brazil
*
Author to whom correspondence should be addressed.
Forests 2020, 11(11), 1144; https://doi.org/10.3390/f11111144
Submission received: 16 September 2020 / Revised: 26 October 2020 / Accepted: 27 October 2020 / Published: 29 October 2020
(This article belongs to the Special Issue Plant-Animal Interactions in Forests)
Browse Figures
Versions Notes

Abstract

:
Seed removal is a key component of seed dispersal and may be influenced by both landscape-scale and local attributes, and it has been used as an indicator of the intensity of interactions between ecosystem components. We examined how the seed removal rates, which integrate the activity of seed dispersers and seed predators, vary with landscape-scale forest cover. We collected data under 34 trees belonging to two zoochoric species (Helicostylis tomentosa (Poepp. and Endl.) J. F. Macbr. and Inga vera Willd.) in 17 remnants in the Brazilian Atlantic forest, with different percentages of forest cover. The seed removal rate was estimated using a fast method based on the abundance of intact fruits and fruit scraps on the ground. The amount of forest cover affected the rate of seed removal in a humpbacked shape, with a maximum seed removal rate at intermediate forest cover. Seed removal rates must be related to the amount of food resources offered and diversity of dispersers and predators in the region. In landscapes with intermediate forest amount, there is a better balance between supply and demand for fruits, leading to a higher seed removal rate than more deforested or forested landscape. Our results also show that local factors, such as crop size and canopy surface, together with forest cover amount, are also important to the removal rate, depending on the species. In addition, our results showed that plant–animal interactions are occurring in all fragments, but the health status of these forests is similar to disturbed forests, even in sites immersed in forested landscapes.

1. Introduction

The Anthropocene is characterized by the pervasive influence of human activity in every biome on Earth. Human impact has drastically increased and it is now comparable in magnitude to natural processes affecting many aspects of the global biosphere [1]. Anthropogenic activities are responsible for drastic land use changes, in which forests are replaced by agricultural and pasture areas [2]. This longstanding process of habitat conversion and consequently habitat loss is the main threat to biodiversity worldwide [3,4]. Therefore, there is an increasing demand for scientific information to mitigate habitat loss effects, but first it is necessary to quantify the status of these new ecosystems [5].
Assessing community diversity through conventional methods is extremely difficult and time consuming [6,7,8], especially in mega diverse environments such as tropical forests. Therefore, methods that quickly access species diversity have been increasingly used since large programs such as the Conservation International’s Rapid Assessment Program have been proposed [9]. Since then, rapid methods have been developed to access the diversity of different taxonomic groups, such as tape recordings for birds [10], acoustic inventories for bats [11], monitoring species by camera traps [12], and assessing forest structure by drone and lidar [13,14]. Recently, Boissier et al. [5] proposed a method to estimate ecological processes through the rate of interaction between plants and animals. This is of paramount importance because species interaction is an often-missed component of biodiversity loss that can even precede species loss [15]. This rapid seed removal method allows the easier implementation of large-scale studies and provides a snapshot of the proportion of removed seeds by frugivores and seed predators [5,16,17,18]. While not integrative of animal activity over the year, since seed removal of a given species can be estimated only during the fruiting period, it provides a realistic picture of it. Indeed, animals adapt the use of their home-range according to the seasonal availability of food resources, even in degraded or provisioned habitats, as demonstrated for primates [19,20,21]. The rapid seed removal method is able to estimate a functional aspect of ecosystem health, as it is an indicator of ecological sustainability [22].
The seed removal rate integrates the activity of seed predators and also of primary and secondary seed dispersers. Zoochoric seed dispersal is a primordial regeneration process for most tropical tree species because it prevents local extinction and allows seedlings and saplings to escape density-dependent mortality processes [23,24,25]. Seed dispersal by animals is performed by primary or secondary frugivores. The former consume fruits in the canopy and regurgitate or defecate seeds elsewhere, and the latter eat fallen fruits from the ground or directly interact with lose seeds, such as dung beetles [26] and ants [27]. Seed predators may also play a role in seed dispersal processes through scatter hoarding [28]. Seed predation occurs mainly on ripe fruits and mature seeds in the canopy or on the ground by different taxonomic groups, such as birds (e.g., parrots), mammals (e.g., squirrels, monkeys, rodents), or insects (e.g., Hemiptera, Coleoptera), and can be as high as 95% of the crop (by peccary [29]).
In tropical forests, human disturbances, such as habitat loss and hunting, mainly reduce the components of the zoochoric seed dispersal, i.e., tree visitation, dispersal distance, and seed removal [30]. On the contrary, the direction of the effects of human perturbations on seed predation depends on the preferences of the predators and on their responses to the perturbation [31]. For instance, seed predation by mammals decreased with disturbance [29,32,33], and consequently, seed removal rate also decreases [18].
In this context, we examined how forest loss affects the seed removal rates of two zoochoric tree species, in 17 remnants of Atlantic forest in Brazil. We tested for the relationship with forest cover percentage at different scales, taking into account cofactors known to influence seed removal rates (canopy surface and fruit crop). A decrease in seed dispersers and predators in abundance and diversity were previously recorded in the same landscapes due to habitat loss and decrease in fruit availability [34]. Thus, we predicted that seed removal rates are positively correlated to landscape forest cover.

2. Material and Methods

2.1. Study Area

The study was performed in the municipality of Una (15.2937° S, 39.0743° W), in the southern region of the Bahia State, Brazil. The vegetation in this region is classified as Tropical Lowland Rainforest, and the climate according Köppen system is Af type, hot and moist without a distinct dry season. The mean annual rainfall is 2082 mm, while the average temperature is 24.4 °C, and the area shows few seasonal climatic variations (Instituto National de Meterologia, Brazil). The region harbors high levels of plant and animal species richness and represents one of the three pockets of endemism in the Atlantic Forest [35,36]. However, hunting pressure has significantly reduced the population of seed dispersers, particularly large mammals, and species such as Tapirus terrestris (Linnaeus, 1758) and Tayassu pecari (Link, 1795) are locally extinct [37,38,39].

2.2. Sites and Forest Cover

The amount of landscape-scale forest cover is considered a meaningful proxy for habitat loss [40,41,42,43], and it is an easy to access metric due to the advance in remote sensing and image analysis methods. In addition, other human pressures, logging, wildfires, or hunting in general occur in synergy and concomitantly with forest loss [1], and thus the amount of landscape forest cover may integrate them. We conducted samplings in pre-selected landscapes of a large network project (SISBIOTA-Functionality), which uses satellites images to identify landscapes with different forest amount, considering only native forest. Each sample site was based around a central point allocated within a forest area, and the proportion of forest cover surrounding that central point was recorded. In this study, we considered each focal tree as the central point to calculate forest cover amount, and we adopted the patch-landscape approach [44] (for details see [45,46]). We considered the percentage of forest cover in different buffers, with a radius of 200 m to 2000 m. The shortest distance between sites was 1 km, while the average distance was 13.7 km (Figure 1).

2.3. Study Species

Helicostylis tomentosa (Poepp. and Endl.) J. F. Macbr. is an evergreen species belonging to the Moraceae family. It is a shade-tolerant species, in which the young stages need shade, but trees grow in full sun. The species is present in greater abundance in more conserved Atlantic Forest areas, however, it is also present in areas that have suffered medium to high anthropic disturbances such as slash-and-burn or logging [36,47]. It generally reaches 15 and 30 m in height, with a diameter at breast height of 40 to 70 cm, and crown up to 12 m in diameter [48]. The infructescence have around 7 g and 3 cm in diameter, and it is composed by four to ten fruits, with a single seed of about 0.73 cm diameter in a light-yellow pulp [48,49]. The fruits are consumed by primates [50,51,52], and according to [53], five or six primate species interact with H. tomentosa. However, only Leontopithecus chrysomelas (Kuhl, 1820) and Sapajus xanthosternos ((Wied-Neuwied, 1826) inhabit the study area. Infructescence with ripe fruits that are not consumed entirely in the treetops are easily found on the ground, and may be intact, partially consumed, or sometimes broken due to the impact upon reaching the ground. It seems there are more terrestrial animal species interacting with fruits and seeds: Tortoises (Chelonoidis denticulatus (Linnaeus, 1766); seed disperser [54]), tapirs (Tapirus terrestris seed disperser [55]), peccaries (Pecari tajacu (Linnaeus, 1758) and Tayassu pecari; seed predator, [56]), grey brockets (Mazama gouazoubira (G. Fisher [von Waldheim], 1814); seed disperser and predator [57]), and agoutis (Dasyprocta (Illiger, 1811); seed disperser and predator [48]). To the best of our knowledge, H. tomentosa was rarely consumed by large frugivorous birds (only sometimes by curassows [52]).
Inga vera Willd. is an evergreen pioneer species belonging to the Fabaceae family. It grows up to 18 m in height, with a diameter at breast height of approximately 60 cm, and 10 to 15 m crown diameter bearing long branches. The fruits are indehiscent legumes with an approximate size of 5.6 cm in length and 2 cm in diameter, and when ripe they are brownish to yellowish and contain approximately 14 seeds of 1.5 cm diameter enveloped in a white, sweetish aril [46,58,59,60]. The fruits and the seeds are mainly consumed by primate seed dispersers [53], but are also unequivocally a source of food for many different species of birds, e.g., five species of parrots (seed predators [61]) and several species of Passeriformes (seed dispersers [62]) and bats (seed dispersers [63]). Ripe fruits are usually not consumed entirely in the treetops, and fallen fruit usually has marks showing that they were partially consumed and are rarely found intact on the ground. Secondary dispersal or predation in Inga genera is accomplished by a variety of animals: Small and large rodents (seed predators), tapirs [64], and peccaries (seed predator [65]).
Considering the fruit characteristics of the tree species used and the fauna of the study site, we consider that it is very unlikely that predation by invertebrates will result in seed removal. Thus, we believe that the predation with removal is carried out only by vertebrates such as lizards, but mainly by mammals. Therefore, the removed seeds may have been dispersed by arboreal frugivores (mainly primates), or predated by arboreal animals (e.g., parrots), and on the ground may have been dispersed or predated by terrestrial animals (such as Pecari tajacu; Mazama gouazoubira; Dasyprocta leporina (Linnaeus, 1758); Cuniculus paca (Linnaeus, 1766); Didelphis albiventris (Lund, 1840); and Didelphis. aurita (Wied-Neuwied, 1826))

2.4. Seed Removal Rates

We visited 32 of the 40 available sites between February and March 2018 to search for fruiting specimens of four initially selected focal tree species. Indeed, the fruiting periods of the species could vary according to climate fluctuations, but have to match with the availability of investigators for field work, and it is also necessary to find enough fruiting specimens to adequately conduct the investigations. This resulted in 34 reproductive individuals marked (19 Helicostylis tomentosa and 15 Inga vera), which were sampled to estimate seed removal rates. To obtain the seed removal rate, we estimated seed production and seed removal from fallen fruits and fruit remains on the ground [5,17]. We estimated seed production (hereafter crop size) and seed removal by counting in quadrats, the number of intact ripe fruits, ripe fruit scraps (pedicels and peduncles, parts of fruits) and loose seeds. Crop size was obtained by adding the number of intact and ripe fruit scraps on the ground and multiplied by the mean seed number per fruit while removed seeds were the missing seeds in the scraps. The removal rate was computed as the ratio of removed seeds over crop size. The efficiency of the seed removal rate testing depends on the researchers’ ability to easily and quickly count seeds. Since fruit production was particularly low for H. tomentosa, we considered the entire surface under the crown of each studied tree. For I. vera, we first tested if our method was accurate by sampling with three 1 m2 random quadrats under each focal tree. For this, we sampled three trees, each with 20 random quadrats and obtained the mean squared errors (MSE) by bootstrap. The MSE (0.0156, 0.0148 and 0.0269) were in the range of the values obtained by [17], which guarantees the accuracy of the method. We also determined the canopy surface area of each sample tree using the standard forester method by estimating four random crown radii [66].

2.5. Statistical Analysis

All the analyses were done in R3.6.3. We first tested whether tree seed removal rate was spatially auto-correlated with the Moran’s I test while varying the number of neighbors (function ‘moransI.v’ of the lctools R package; [67]). We computed the mean distances between k neighbors with the ‘knearneigh’ function of the spdep R package [68]. We used auto-covariate logistic regression models following [69] to test for the relationships between the seed removal rate (the dependent variable) and a combination of landscape and local variables—forest cover, tree canopy surface (m2 per tree), seed crop (seed number per m2), and pairwise interactions of these two variables. The auto-covariate was computed using the ‘autocov dist’ function of spdep following [68]. We included in the analysis the linear and the quadratic effects of the landscape forest cover. We evaluated the scale of forest cover effect by varying the buffer radius (200 m to 2000 m). The analyses were performed separately for each species. We tested all the possible combinations of effects using the function ‘bestglm’ of the bestglm R package [70], comparing the fits of variable combinations with the Bayesian information criterion (BIC) and applying the marginality rule (i.e., when an interaction term is included in a model, the corresponding main effects should be also included). We tested the selected models against a null model with the likelihood ratio test and the spatial auto-correlation in the Pearson residuals as above (dataset and R code available in the Supplementary Materials: Supplementary information.zip file). To plot the model prediction of one factor over the observations, we set the other factors to their mean. As predictive capacity power of landscape forest cover, we calculated the partial Efron’s pseudo R-squared of the factor with the function ‘rsq.partial of the rsq R package [71] and compared the coefficients of the models using their 95% confidence intervals.

3. Results

Landscape-scale forest cover ranged between 36.5% and 100% in radius of 200 m to 2000 m. The seed removal rate was on average 0.56, but varied considerably among trees and between species (Figure 2); for H. tomentosa, the mean was 50.1%, and for I. vera, 62.4%. The canopy surface of the sampled H. tomentosa was 11.1 to 61.4 m2 (mean = 33.0 m2) and for I. vera, 12.6 to 139.3 m2 (mean = 64.9 m2). The crop size of H. tomentosa varied between 1 and 13 seeds/m2 (mean = 6.3 seeds/m2), and that of I. vera between 27 and 227 seeds/m2 (mean = 94.1 seeds/m2).
The Moran’s I test indicated positive and significant spatial auto-correlation decreasing with distance (number of conspecifics taken into account), which means that seed removal rates were more similar between closer trees. The trend was weaker for H. tomentosa than for I. vera (Figure 3). It also suggested that, for I. vera, spatial auto-correlation extended beyond landscape limits, since the mean number of trees per remnant was only two, while the Moran’s I remained significant for distances up to 4 km.
Model selection allowed to identify highly significant relationships between seed removal rate and forest cover having a humpbacked shape (Figure 4). The best model of H. tomentosa (Table 1) was obtained for landscape forest cover estimated in a radius of 1600 m while for I. vera, the best model was obtained for landscape forest cover estimated in a radius of 1400 m (Table 2). For both species, the model also included the canopy surface, the crop, their pairwise interaction, and the auto-covariate. The Pearson residuals were not spatially auto-correlated (Figure 3), which means that the models correctly captured the generating processes. The effects of landscape forest cover were supported by a higher partial pseudo-R squared for H. tomentosa than for I. vera. Nevertheless, the response of I. vera to forest cover was stronger than that of H. tomentosa (Figure 4), as also shown by the forest cover coefficients, which were significantly larger. The models selected with the estimations of landscape forest cover in other radii showed the same kind of relationship for forest cover (excepting two of them), i.e., a humpbacked curve, but with lower fit. The responses of the seed removal rate to crop and canopy surface were contrasted and clearly resulted from the interaction between both factors.

4. Discussion

The amount of forest cover affects seed removal rates of two plant species in the Atlantic Forest from southern Bahia. However, the effect of forest loss was humpbacked shaped and not linear as expected. Local factors were also important determinants of seed removal rates, depending on the tree species, such as fruit crop and the canopy surface. Thus, our study showed that the response of seed removal rate to forest cover is species-specific. More importantly, the seed removal rate results showed that plant–animal interaction is occurring in all fragments, but it was low in forested landscapes, an indication that they are also defaunated since it is established that the seed removal rate responds to defaunation [18].
The variation of spatial autocorrelation of the seed removal rate with number of neighbors indicated that the H. tomentosa seed removal was limited to each forest remnant and landscape, while for I. vera, it extended beyond landscape limits. Both species are known to be dispersed by primates, however, I. vera is also dispersed by several species of birds and bats, while H. tomentosa was only rarely observed being dispersed by curassows [52], which also has a low population in the region. Thus, I. vera probably has a more mobile animal community that interacts with seeds than those of H. tomentosa—this fact probably favors a higher rate of seed removal for this species, as observed in this work. Seed removal is probably occurring even in the most deforested fragments by some primates that may be relatively little affected or even favored by forest loss [72]. The primate Leontopithecus chrysomelas, an important seed disperser in our study area, as well as other tamarins such as Callithrix penicillata (E. Geoffroy Saint-Hilaire, 1812) and Callithrix jacchus (Linnaeus, 1758), are able to use degraded areas, such as cocoa-shaded plantations, degraded forests, or rubber plantations [73,74,75]. The seed dispersal by birds and bats may be also important, but only for I. vera.
Seed removal rates for both species were higher in fragments immersed in landscapes with intermediate amounts of forest. Seed removal rates must be related to the amount of food resources offered and diversity of seed dispersers and predators in the region. In forests immersed in more deforested landscapes there is a smaller amount of fruits being offered [46], however, the rate of defaunation is higher, so there are few animals (dispersers and predators) to remove the seeds [76,77,78] resulting in low seed removal rates. On the other hand, in more forested landscapes, there is a greater supply of fruits and more animals due a lower rate of defaunation, also resulting in low seed removal rates. However, in landscapes with an intermediate forest cover amount, a better balance between supply and demand of fruits is expected, leading to higher seed removal rates and explaining the humpback shape. The forests of this region have a high defaunation rate, in which the population of medium and large mammals has suffered a great decrease or even local extinction, such as the tapir [37,38]. In the same landscapes, it has been shown that deforestation leads to reduced seed predation by small mammals [77]. In this way, small and some species of medium mammals (such as agoutis, pacas, and peccaries) should play the main role in removing fruits on the ground (predation or secondary dispersal). However, it is noteworthy that the forest cover gradient evaluated comprises only 36 to 80% of forest cover in the selected buffers, and it is not possible to infer about fragments in more deforested landscapes due to the fact that fruiting individuals of these species were not found.
The mean seed removal rate was 0.56, but varied considerably among individuals and species. It was lower to Helicostylis tomentosa, a shade-tolerant species with more nutritive fruits than Inga vera, a shade-intolerant species with poor fruits [46]. However, both species showed individuals with high removal rates (maximum 0.95 for H. tomentosa and 0.97 for I. vera). In undisturbed fragments, with limited hunting pressure, previous studies showed that seed removal rates could be close to 1 [5,16,17], but this also depends on plant species [18]. The seed removal rate in a well-preserved area was 77%, and 47% in an area with a history of logging and high hunting pressure [18]. The fact that in our study region, there is high hunting pressure, with a decrease in or local extinction of medium and large mammals, [37,38,79], could explain the low rates of seed removal even in the most forested landscapes, even if precise information of these threats on each landscape is lacking.
Seed removal rates can be determined not only by forest amount or diversity of frugivores, but also by individual tree features, such as crown surface area and crop size, but these features interact with the animal communities. Plant size generally correlates with crop size (this is also the case in this study), and larger crops may attract more frugivores. Indeed, crop size is suggested as a key trait for conspicuousness [80], thus, increasing fruit removal. A recent review on the role of fruit crops provided evidence to support the hypothesis that crop size represents a conspicuous signal for the amount of reward for birds [81]. However, faster satiation of disturbed animal communities could occur with larger crops [82] or when important frugivorous species have been extirpated [83]. The larger trees may attract more animals but those trees may produce crops so large that only a smaller proportion of the seeds may finally be removed by dispersers or predators.

5. Conclusions

Seed removal rates in forest remnants were sensitive to forest loss in the landscape, but the responses were species specific. In the limited scope of our study, we did not register the animal communities and their responses to forest loss while it would have helped to strengthen our conclusions. Nevertheless, our analysis highlighted a humpbacked response peaking at intermediate values of landscape forest cover. Below those values, the conditions for the seed dispersers and seed predators probably unduly worsen and the seed removal rate strictly decline. Methodological difficulties are constituted by selecting appropriate species (i.e., species characterized by ripe fruits falling on the ground and fruits leaving traces after consumption) and by finding many individuals with enough ripe fruits of the same species in different landscapes. However, the method is not restricted to threatened tree species or to tropical climate ecosystems. A greater ability to generalize may allow us to deduce a pertinent indicator of forest functionality, and thus of forest health. Thus, the seed removal rate showed that animal–plant interaction occurs in all fragments, however, their health is already compromised since even fragments immersed in highly forested landscapes showed low seed removal rates.

Supplementary Materials

The following are available online at https://www.mdpi.com/1999-4907/11/11/1144/s1, data and R code for model selection.

Author Contributions

Conceptualization, A.H. and F.T.; methodology, A.H., F.T. and L.R.-S.; formal analysis, A.H.; investigation, A.S. and L.R.-S.; writing—original draft preparation, A.H. and E.C.; writing—review and editing, A.H., L.R.-S. and E.C. All authors have read and agreed to the published version of the manuscript.

Funding

We are grateful to ARES for providing a travel grant for flights between Belgium and Brazil.

Acknowledgments

We are grateful to the landowners for allowing us to work on their properties. We are grateful to Deborah Faria for coordinating the Rede SISBIOTA project (CNPq 563216/2010-7), in which this work is inserted, being the publication number 37 of this project.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Malhi, Y.; Gardner, T.A.; Goldsmith, G.R.; Silman, M.R.; Zelazowski, P. Tropical forests in the anthropocene. Annu. Rev. Environ. Resour. 2014, 39, 125–159. [Google Scholar] [CrossRef] [Green Version]
  2. Curtis, P.G.; Slay, C.M.; Harris, N.L.; Tyukavina, A.; Hansen, M.C. Classifying drivers of global forest loss. Science 2018, 361, 1108–1111. [Google Scholar] [CrossRef] [PubMed]
  3. Fahrig, L. Effects of habitat fragmentation on biodiversity. Annu. Rev. Ecol. Evol. Syst. 2003, 34, 487–515. [Google Scholar] [CrossRef] [Green Version]
  4. Betts, M.G.; Wolf, C.; Ripple, W.J.; Phalan, B.; Millers, K.A.; Duarte, A.; Butchart, S.H.M.; Levi, T. Global forest loss disproportionately erodes biodiversity in intact landscapes. Nature 2017, 547, 441–444. [Google Scholar] [CrossRef] [PubMed]
  5. Boissier, O.; Bouiges, A.; Mendoza, I.; Feer, F.; Forget, P.-M. Rapid assessment of seed removal and frugivore activity as a tool for monitoring the health status of tropical forests. Biotropica 2014, 46, 633–641. [Google Scholar] [CrossRef]
  6. Lawton, J.H.; Naeem, S.; Thompson, L.J.; Hector, A.; Crawley, M.J.; Thompson, K.; Hodgson, J.G. Biodiversity and ecosystem function: Getting the Ecotron experiment in its correct context. Funct. Ecol. 1998, 12, 848–853. [Google Scholar]
  7. Bibby, C.J.; Burgess, N.D.; Hill, D.A.; Mustoe, S. Bird Census Techniques, 2nd ed.; Academic Press: Amsterdam, The Netherlands, 2000. [Google Scholar]
  8. Gardner, T.A.; Barlow, J.; Araujo, I.S.; Ávila-Pires, T.C.; Bonaldo, A.B.; Costa, J.E.; Esposito, M.C.; Ferreira, L.V.; Hawes, J.; Hernandez, M.I.M.; et al. The cost-effectiveness of biodiversity surveys in tropical forests. Ecol. Lett. 2008, 11, 139–150. [Google Scholar] [CrossRef] [PubMed]
  9. Abate, T. Environmental rapid-assessment programs have appeal and critics. Are they the domain of the conservation elite? BioScience 1992, 42, 486–489. [Google Scholar] [CrossRef]
  10. Parker, T.A., III. On the use of tape recorders in avifaunal surveys. Auk 1991, 108, 443–444. [Google Scholar]
  11. O’Farrell, M.J.; Gannon, W.L. A comparison of acoustic versus capture techniques for the inventory of bats. J. Mamm. 1999, 80, 24–30. [Google Scholar] [CrossRef]
  12. Majumder, A.; Qureshi, Q.; Sankar, K.; Kumar, A. Long-term monitoring of a Bengal tiger (Panthera tigris tigris) population in a human-dominated landscape of Central India. Eur. J. Wildl. Res. 2017, 63, 17. [Google Scholar] [CrossRef]
  13. Messinger, M.; Asner, G.P.; Silman, M. Rapid assessments of amazon forest structure and biomass using small unmanned aerial systems. Remote Sens. 2016, 8, 615. [Google Scholar] [CrossRef] [Green Version]
  14. Kellner, J.R.; Armston, J.; Birrer, M.; Cushman, K.C.; Duncanson, L.; Eck, C.; Falleger, C.; Imbach, B.; Král, K.; Krůček, M.; et al. New opportunities for forest remote sensing through ultra-high-density drone lidar. Surv. Geophys. 2019, 40, 959–977. [Google Scholar] [CrossRef] [Green Version]
  15. Valiente-Banuet, A.; Aizen, M.A.; Alcántara, J.M.; Arroyo, J.; Cocucci, A.; Galetti, M.; García, M.B.; García, D.; Gómez, J.M.; Jordano, P.; et al. Beyond species loss: The extinction of ecological interactions in a changing world. Funct. Ecol. 2015, 29, 299–307. [Google Scholar] [CrossRef]
  16. Lermyte, C.; Forget, P.-M. Rapid assessment of dispersal failure and seedling recruitment of large-seeded non- timber forest products trees in a tropical rainforest. Trop. Conserv. Sci. 2009, 2, 404–424. [Google Scholar] [CrossRef] [Green Version]
  17. Hambuckers, J.; Dauvrin, A.; Trolliet, F.; Evrard, Q.; Forget, P.M.; Hambuckers, A. How can seed removal rates of zoochoric tree species be assessed quickly and accurately? For. Ecol. Manag. 2017, 403, 152–160. [Google Scholar] [CrossRef]
  18. Boissier, O.; Feer, F.; Henry, P.-Y.; Forget, P.-M. Modifications of the rainforest frugivore community are associated with reduced seed removal at the community level. Ecol. Appl. 2020, 30, e02086. [Google Scholar] [CrossRef] [PubMed]
  19. Albert, A.; Huynen, M.-C.; Savini, T.; Hambuckers, A. Influence of food resources on the ranging pattern of Northern pigtailed macaques (Macaca leonina). Int. J. Primatol. 2013, 34, 693–713. [Google Scholar] [CrossRef]
  20. José-Domínguez, J.M.; Huynen, M.C.; García, C.J.; Albert-Daviaud, A.; Savini, T.; Asensio, N. Non-territorial Macaques Can Range Like Territorial Gibbons When Partially Provisioned With Food. Biotropica 2015, 47, 733–744. [Google Scholar] [CrossRef]
  21. Gazagne, E.; José-Domínguez, J.M.; Huynen, M.C.; Hambuckers, A.; Poncin, P.; Savini, T.; Brotcorne, F. Northern pigtailed macaques rely on old growth plantations to offset low fruit availability in a degraded forest fragment. Am. J. Primatol. 2020, 82, e23117. [Google Scholar] [CrossRef]
  22. Patil, G.P.; Brooks, R.P.; Myers, W.L.; Rapport, D.J.; Taillie, C. Ecosystem health and its measurement at landscape scale: Toward the next generation of quantitative assessments. Ecosyst. Health 2001, 7, 307–316. [Google Scholar] [CrossRef]
  23. Mangan, S.A.; Schnitzer, S.A.; Herre, E.A.; Mack, K.M.L.; Valencia, M.C.; Sanchez, E.I.; Bever, J.D. Negative plant-soil feedback predicts tree-species relative abundance in a tropical forest. Nature 2010, 466, 752–755. [Google Scholar] [CrossRef] [PubMed]
  24. Swamy, V.; Terborgh, J.W. Distance-responsive natural enemies strongly influence seedling establishment patterns of multiple species in an Amazonian rain forest. J. Ecol. 2010, 98, 1096–1107. [Google Scholar] [CrossRef]
  25. Bagchi, R.; Gallery, R.E.; Gripenberg, S.; Gurr, S.J.; Narayan, L.; Addis, C.E.; Freckleton, R.P.; Lewis, O.T. Pathogens and insect herbivores drive rainforest plant diversity and composition. Nature 2014, 506, 85–88. [Google Scholar] [CrossRef] [PubMed]
  26. Culot, L.; Lazo, F.J.J.M.; Huynen, M.C.; Poncin, P.; Heymann, E.W. Seasonal variation in seed dispersal by tamarins alters seed rain in a secondary rain forest. Int. J. Primatol. 2010, 31, 553–569. [Google Scholar] [CrossRef] [Green Version]
  27. Lengyel, S.; Gove, A.D.; Latimer, A.M.; Majer, J.D.; Dunn, R.R. Convergent evolution of seed dispersal by ants, and phylogeny and biogeography in flowering plants: A global survey. Perspect. Plant Ecol. Evol. Syst. 2010, 12, 43–55. [Google Scholar] [CrossRef]
  28. Aliyu, B.; Adamu, H.; Moltchanova, E.; Forget, P.-M.; Chapman, H. The interplay of habitat and seed type on scatterhoarding behavior in a fragmented afromontane forest landscape. Biotropica 2014, 46, 264–267. [Google Scholar] [CrossRef]
  29. Keuroghlian, A.; Eaton, D.P. Removal of palm fruits and ecosystem engineering in palm stands by white-lipped peccaries (Tayassu pecari) and other frugivores in an isolated Atlantic Forest fragment. Biodivers. Conserv. 2009, 18, 1733–1750. [Google Scholar] [CrossRef]
  30. Markl, J.S.; Schleuning, M.; Forget, P.M.; Jordano, P.; Lambert, J.E.; Traveset, A.; Wright, S.J.; Böhning-Gaese, K. Meta-analysis of the effects of human disturbance on seed dispersal by animals. Conserv. Biol. 2012, 26, 1072–1081. [Google Scholar] [CrossRef] [Green Version]
  31. Chen, Q.; Tomlinson, K.W.; Cao, L.; Wang, B. Effects of fragmentation on the seed predation and dispersal by rodents differ among species with different seed size. Integr. Zool. 2017, 12, 468–476. [Google Scholar] [CrossRef]
  32. Wright, S.J.; Duber, H.C. Poachers and forest fragmentation alter seed dispersal, seed survival, and seedling recruitment in the palm Attalea butyraceae, with implications for tropical tree diversity. Biotropica 2001, 33, 583–595. [Google Scholar] [CrossRef]
  33. Herrerías-Diego, Y.; Quesada, M.; Stoner, K.E.; Lobo, J.A.; Hernández-Flores, Y.; Sanchez Montoya, G. Effect of forest fragmentation on fruit and seed predation of the tropical dry forest tree Ceiba aesculifolia. Biol. Conserv. 2008, 141, 241–248. [Google Scholar] [CrossRef]
  34. Pessoa, M.; Hambuckers, A.; Benchimol, M.; Rocha-Santos, L.; Bomfim, J.A.; Faria, D.; Cazetta, E. Deforestation drives functional diversity and fruit quality changes in a tropical tree assemblage. Perspect. Plant Ecol. Evol. Syst. 2017, 28, 78–86. [Google Scholar] [CrossRef]
  35. Prado, P.I.; Pinto, L.P.; Moura, R.T.; Carvalho, A.M.; Thomas, W.W.; Amorin, A.M.; Aguiar, C. Caracterizção dos registros de ocorrência de espécies de plantas e mamíferos no sul da Bahia, Brasil. In Corredor de biodiversidade da Mata Atlântica do sul da Bahia; Instituto de Estudos Sócio-Ambientais do Sul da Bahia e Conservation International do Brasil: Ilhéus, Bahia, Brazil, 2003; pp. 1–16. [Google Scholar]
  36. Martini, A.M.Z.; Fiaschi, P.; Amorim, A.M.; Paixão, J.L.D. A hot-point within a hot-spot: A high diversity site in Brazil’s Atlantic Forest. Biodivers. Conserv. 2007, 16, 3111–3128. [Google Scholar] [CrossRef]
  37. Canale, G.R.; Peres, C.A.; Guidorizzi, C.E.; Gatto, C.A.F.; Kierulff, M.C.M. Pervasive defaunation of forest remnants in a tropical biodiversity hotspot. PLoS ONE 2012, 7, e41671. [Google Scholar] [CrossRef] [Green Version]
  38. Cassano, C.R.; Barlow, J.; Pardini, R. Large mammals in an agroforestry mosaic in the Brazilian Atlantic Forest. Biotropica 2012, 44, 818–825. [Google Scholar] [CrossRef]
  39. Flesher, K.M.; Laufer, J. Protecting wildlife in a heavily hunted biodiversity hotspot: A case study from the Atlantic Forest of Bahia, Brazil. Trop. Conserv. Sci. 2013, 6, 181–200. [Google Scholar] [CrossRef]
  40. Beca, G.; Vancine, M.H.; Carvalho, C.S.; Pedrosa, F.; Alves, R.S.C.; Buscariol, D.; Peres, C.A.; Ribeiro, M.C.; Galetti, M. High mammal species turnover in forest patches immersed in biofuel plantations. Biol. Conserv. 2017, 210, 352–359. [Google Scholar] [CrossRef] [Green Version]
  41. Rocha-Santos, L.; Benchimol, M.; Mayfield, M.M.; Faria, D.; Pessoa, M.S.; Talora, D.C.; Mariano-Neto, E.; Cazetta, E. Functional decay in tree community within tropical fragmented landscapes: Effects of landscape-scale forest cover. PLoS ONE 2017, 12, e0175545. [Google Scholar] [CrossRef] [Green Version]
  42. Decaëns, T.; Martins, M.B.; Feijoo, A.; Oszwald, J.; Dolédec, S.; Mathieu, J.; Arnaud de Sartre, X.; Bonilla, D.; Brown, G.G.; Cuellar Criollo, Y.A.; et al. Biodiversity loss along a gradient of deforestation in Amazonian agricultural landscapes. Conserv. Biol. 2018, 32, 1380–1391. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  43. Macedo-Reis, L.E.; Quesada, M.; de Siqueira Neves, F. Forest cover drives insect guild diversity at different landscape scales in tropical dry forests. For. Ecol. Manag. 2019, 443, 36–42. [Google Scholar] [CrossRef]
  44. McGarigal, K.; Cushman, S.A. Comparative evaluation of experimental approaches to the study of habitat fragmentation effects. Ecol. Appl. 2002, 12, 335–345. [Google Scholar] [CrossRef]
  45. Rocha-Santos, L.; Pessoa, M.S.; Cassano, C.R.; Talora, D.C.; Orihuela, R.L.L.; Mariano-Neto, E.; Morante-Filho, J.C.; Faria, D.; Cazetta, E. The shrinkage of a forest: Landscape-scale deforestation leading to overall changes in local forest structure. Biol. Conserv. 2016, 196, 1–9. [Google Scholar] [CrossRef]
  46. Pessoa, M.; Rocca, L.; Talora, D.; Faria, D.; Mariano-Neto, E.; Hambuckers, A.; Cazetta, E. Fruit biomass availability along a forest cover gradiant. Biotropica 2016, 49, 45–55. [Google Scholar] [CrossRef]
  47. Rocha-Santos, L.; Talora, D.C. Recovery of atlantic rainforest areas altered by distinct land-use histories in northeastern Brazil. Trop. Conserv. Sci. 2012, 5, 475–494. [Google Scholar] [CrossRef] [Green Version]
  48. Arruda, Y.M.B.C.; Ferraz, I.D.K. Inharé-da-folha-peluda Helicostylis tomenosa (Poep. & Endl.) Rusby. In Informativo Técnico. Rede de Sementes da @mazônia; Ferraz, I.D.K., Sidney, A.N., Ferreira, S.A.N., Camargo, J.L.C., Eds.; INPA: Manaus, Brazil, 2008; pp. 1–2. [Google Scholar]
  49. de Souza Pessoa, M.; de Vleeschouwer, K.; Amorim, A.M.; Custódio Talora, D. Calendário Fenológico. Uma Ferramenta Para Auxiliar No Cultivo de Espécies Arbóreas Nativitas da Floresta Atläntica no Sul da Bahia; Edus-Editoria da UESC: Ilheus, Brazil, 2011. [Google Scholar]
  50. Izawa, K. Foods and feeding behavior of wild black-capped capuchin (Cebus apella). Primates 1979, 20, 57–76. [Google Scholar] [CrossRef]
  51. Catenacci, L.S.; De Vleeschouwer, K.M.; Nogueira-Filho, S.L.G. Seed dispersal by Golden-headed Lion Tamarins Leontopithecus chrysomelas in southern Bahian Atlantic Forest, Brazil. Biotropica 2009, 41, 744–750. [Google Scholar] [CrossRef]
  52. Stevenson, P.R.; Link, A.; González-Caro, S.; Torres-Jiménez, M.F. Frugivory in canopy plants in a western Amazonian forest: Dispersal systems, phylogenetic ensembles and keystone plants. PLoS ONE 2015, 10, e0140751. [Google Scholar] [CrossRef] [Green Version]
  53. Bello, C.; Galetti, M.; Montan, D.; Pizo, M.A.; Mariguela, T.C.; Culot, L.; Bufalo, F.; Labecca, F.; Pedrosa, F.; Constantini, R.; et al. Atlantic frugivory: A plant–frugivore interaction data set for the Atlantic Forest. Ecology 2017, 98, 1729. [Google Scholar] [CrossRef] [Green Version]
  54. Jerozolimski, A.; Ribeiro, M.B.N.; Martins, M. Are tortoises important seed dispersers in Amazonian forests? Oecologia 2009, 161, 517–528. [Google Scholar] [CrossRef]
  55. Salas, L.A.; Fuller, T.K. Diet of the lowland tapir (Tapirus terrestris L.) in the Tabaro River valley, southern Venezuela. Can. J. Zool. 1996, 74, 1444–1451. [Google Scholar] [CrossRef]
  56. Prado, H.M. Feeding ecology of five Neotropical ungulates: A critical review. Oecologia Aust. 2013, 17, 459–473. [Google Scholar] [CrossRef] [Green Version]
  57. Gayot, M.; Henry, O.; Dubost, G.; Sabatier, D. Comparative diet of the two forest cervids of the genus Mazama in French Guiana. J. Trop. Ecol. 2004, 20, 31–43. [Google Scholar] [CrossRef]
  58. Wiesbauer, M.B.; Giehl, E.L.H.; Jarenkow, J.A. Padrões morfológicos de diásporos de árvores e arvoretas zoocóricas no Parque Estadual de Itapuã, RS, Brasil. Acta Bot. Brasilica 2008, 22, 425–435. [Google Scholar] [CrossRef]
  59. Orwa, C.; Mutua, A.; Kindt, R.; Jamnadass, R.; Simons, A. Agroforestree Database: A Tree Reference and Selection Guide Version 4.0; World Agroforestry Centre: Nairobi, Kenya, 2009. [Google Scholar]
  60. Colmanetti, M.A.A.; Barbosa, L.M.; Shirasuna, R.T.; Couto, H.T.Z.D. Phytosociology and structural characterization of woody regeneration from a reforestation with native species of Southeastern Brazil. Rev. Árvore 2016, 40, 209–218. [Google Scholar] [CrossRef] [Green Version]
  61. Ragusa-Netto, J.; Fecchio, A. Plant food resources and the diet of a parrot community in a gallery forest of the Southern Pantanal (Brazil). Braz. J. Biol. 2006, 66, 1021–1032. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  62. Carlo, T.A.; Collazo, J.A.; Groom, M.J. Influences of fruit diversity and abundance on bird use of two shaded coffee plantations. Biotropica 2004, 36, 602–614. [Google Scholar] [CrossRef]
  63. Bevilacqua Marcuzzo, S.; Viera, M. Ecological restoration in Conservation units. In Biodiversity in Ecosystems-Linking Structure and Function; Lo, Y.-H., Blanco, J.A., Roy, S., Eds.; IntechOpen Limited: London, UK, 2015; pp. 493–509. [Google Scholar]
  64. Galetti, M.; Keuroghlian, A.; Hanada, L.; Morato, M.I. Frugivory and seed dispersal by the lowland tapir (Tapirus terrestris) in Southeast Brazil. Biotropica 2001, 33, 723–726. [Google Scholar] [CrossRef]
  65. Beck, H. Seed predation and dispersal by peccaries throughout the Neotropics and its consequences: A review and synthesis. In Seed Fate: Predation, Dispersal and Seedling Establishment; Forget, P.M., Hulme, P., Lambert, J.E., Vander Wall, S.B., Eds.; CABI Publishing: Wallinford, UK, 2004; pp. 77–115. [Google Scholar]
  66. Rondeux, J. La Mesure des Arbres et des Peuplements Forestiers; Presses Agronomiques de Gembloux: Gembloux, Belgium, 1993. [Google Scholar]
  67. Kalogirou, S. Package ‘lctools’. Local Correlation, Spatial Inequalities and Other Tools. Available online: https://CRAN.R-project.org/package=lctools (accessed on 15 September 2020).
  68. Bivand, R.; Altman, M.; Anselin, L.; Assunção, R.; Berke, O.; Bernat, A.; Blanchet, G. Package ‘spdep’. Available online: https://github.com/r-spatial/spdep/ (accessed on 15 September 2020).
  69. Dormann, C.F.; McPherson, J.M.; Araújo, M.B.; Bivand, R.; Bolliger, J.; Carl, G.; Davies, R.G.; Hirzel, A.; Jetz, W.; Kissling, W.D.; et al. Methods to account for spatial autocorrelation in the analysis of species distributional data: A review. Ecography 2007, 30, 609–628. [Google Scholar] [CrossRef] [Green Version]
  70. McLeod, A.I.; Xu, C. Best subset GLM and regression utilities. Available online: https://CRAN.R-project.org/package=bestglm (accessed on 16 June 2020).
  71. Zhang, D. R-squared and related measures. Available online: https://CRAN.R-project.org/package=rsq (accessed on 19 October 2020).
  72. Boyle, S.A.; Smith, A.T. Can landscape and species characteristics predict primate presence in forest fragments in the Brazilian Amazon? Biol. Conserv. 2010, 143, 1134–1143. [Google Scholar] [CrossRef]
  73. Raboy, B.E.; Christman, M.C.; Dietz, J.M. The use of degraded and shade cocoa forests by Endangered golden-headed lion tamarins Leontopithecus chrysomelas. Oryx 2004, 38, 75–83. [Google Scholar] [CrossRef] [Green Version]
  74. Tisovec, K.C.; Cassano, C.R.; Boubli, J.P.; Pardini, R. Mixed-species groups of marmosets and tamarins across a gradient of agroforestry intensification. Biotropica 2014, 46, 248–255. [Google Scholar] [CrossRef]
  75. De Vleeschouwer, K.M.; Oliveira, L.C. Report on the presence of a group of golden-headed lion tamarins (Leontopithecus chrysomelas), an endangered primate species in a rubber plantation in southern Bahia, Brazil. Primate Biol. 2017, 4, 61–67. [Google Scholar] [CrossRef] [Green Version]
  76. Galetti, M.; Donatti, C.I.; Pires, A.S.; Guimarães Jr, P.R.; Jordano, P. Seed survival and dispersal of an endemic Atlantic forest palm: The combined effects of defaunation and forest fragmentation. Bot. J. Linn. Soc. 2006, 151, 141–149. [Google Scholar] [CrossRef] [Green Version]
  77. Soares, L.A.S.S.; Faria, D.; Vélez-Garcia, F.; Vieira, E.M.; Talora, D.C.; Cazetta, E. Implications of habitat loss on seed predation and early recruitment of a keystone palm in anthropogenic landscapes in the Brazilian Atlantic rainforest. PLoS ONE 2015, 10, e0133540. [Google Scholar] [CrossRef] [Green Version]
  78. Pérez-Méndez, N.; Jordano, P.; García, C.; Valido, A. The signatures of Anthropocene defaunation: Cascading effects of the seed dispersal collapse. Sci. Rep. 2016, 6, 24820. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  79. Barcellos de Souza, J.B.; Nóbrega Alves, R.R. Hunting and wildlife use in an Atlantic Forest remnant of northeastern Brazil. Trop. Conserv. Sci. 2014, 7, 145–160. [Google Scholar] [CrossRef] [Green Version]
  80. Ordano, M.; Blendinger, P.G.; Lomáscolo, S.B.; Chacoff, N.P.; Sánchez, M.S.; Núñez Montellano, M.G.; Jiménez, J.; Ruggera, R.A.; Valoy, M. The role of trait combination in the conspicuousness of fruit display among bird-dispersed plants. Funct. Ecol. 2017, 31, 1718–1727. [Google Scholar] [CrossRef]
  81. Palacio, F.X.; Ordano, M. The strength and drivers of bird-mediated selection on fruit crop size: A meta-analysis. Front. Ecol. Evol. 2018, 6, 18. [Google Scholar] [CrossRef] [Green Version]
  82. Briani, D.C.; Guimarães, P.R., Jr. Seed predation and fruit damage of Solanum lycocarpum (Solanaceae) by rodents in the cerrado of central Brazil. Acta. Oecol. 2007, 31, 8–12. [Google Scholar] [CrossRef]
  83. Trolliet, F.; Forget, P.-M.; Huynen, M.-C.; Hambuckers, A. Forest cover, hunting pressure, and fruit availability influence seed dispersal in a forest-savanna mosaic in the Congo Basin. Biotropica 2017, 49, 337–345. [Google Scholar] [CrossRef]
Forests 11 01144 g001 550
Figure 1. Map depicting all individuals sampled (19 Helicostylis tomentosa and 15 Inga vera; some points are overlapping on the map, due to its large scale) which were used for the seed removal rate estimates and for calculating the amount of Brazilian Atlantic forest (Bahia, Brazil).
Figure 1. Map depicting all individuals sampled (19 Helicostylis tomentosa and 15 Inga vera; some points are overlapping on the map, due to its large scale) which were used for the seed removal rate estimates and for calculating the amount of Brazilian Atlantic forest (Bahia, Brazil).
Forests 11 01144 g001
Forests 11 01144 g002 550
Figure 2. Boxplots of seed removal rates for each species with means (+).
Figure 2. Boxplots of seed removal rates for each species with means (+).
Forests 11 01144 g002
Forests 11 01144 g003 550
Figure 3. Moran’s I for seed removal rates and Pearson residuals of the auto-covariate logistic regression models computed as a function of the number k of neighbors (accounted for k = 1 to 7, from left to right) and expressed as a function of corresponding mean distance. Significant values for p ≤ 0.05 are in solid symbols and non-significant values in hollow symbols.
Figure 3. Moran’s I for seed removal rates and Pearson residuals of the auto-covariate logistic regression models computed as a function of the number k of neighbors (accounted for k = 1 to 7, from left to right) and expressed as a function of corresponding mean distance. Significant values for p ≤ 0.05 are in solid symbols and non-significant values in hollow symbols.
Forests 11 01144 g003
Forests 11 01144 g004 550
Figure 4. Results of the logistic regression models. (a) Seed removal rates as a function of forest cover around forest remnants (H. tomentosa: Observed values in black squares and black prediction line, forest cover estimated in a landscapes of 1600 m radius, I. vera: Observed values in red circles and red dashed prediction line, forest cover estimated in landscapes of 1400 m radius); (b,c) seed removal rate as a function of crop size, and canopy surface (symbol size is a log function of total number of seeds taken into account for each tree).
Figure 4. Results of the logistic regression models. (a) Seed removal rates as a function of forest cover around forest remnants (H. tomentosa: Observed values in black squares and black prediction line, forest cover estimated in a landscapes of 1600 m radius, I. vera: Observed values in red circles and red dashed prediction line, forest cover estimated in landscapes of 1400 m radius); (b,c) seed removal rate as a function of crop size, and canopy surface (symbol size is a log function of total number of seeds taken into account for each tree).
Forests 11 01144 g004
Table
Table 1. Selected auto-covariate logistic regression models of seed removal rate with Efron’s pseudo R-squared and likelihood ratio tests. Helicostylis tomentosa, sample size = 19.
Table 1. Selected auto-covariate logistic regression models of seed removal rate with Efron’s pseudo R-squared and likelihood ratio tests. Helicostylis tomentosa, sample size = 19.
CoefficientsEstimatesStandard ErrorsZ-Valuesp-Values
Intercept−9.761.041−9.374<0.0001
Forest cover (1600 m radius)0.3330.0418.207<0.0001
Forest cover2−0.00290.0003−8.453<0.0001
Canopy surface0.0350.00605.904<0.0001
Crop size0.1300.03074.227<0.0001
Crop x Canopy surface−0.0060.0006−9.824<0.0001
Scaled auto-covariate−0.2100.0569−3.7600.0002
95% confidence interval of coefficients; Forest cover: 0.2633–0.4122, Forest cover2: −0.00363–−0.00226; Efron’s pseudo R-squared = 0.5878; Partial Efron’s pseudo R-squared for forest cover effects = 0.5204; Likelihood ratio test deviance = 645.4, ddl = 6, p-value < 0.0001.
Table
Table 2. Selected auto-covariate logistic regression models of seed removal rate with Efron’s pseudo R-squared and likelihood ratio tests. Inga vera, sample size = 15.
Table 2. Selected auto-covariate logistic regression models of seed removal rate with Efron’s pseudo R-squared and likelihood ratio tests. Inga vera, sample size = 15.
CoefficientsEstimatesStandard ErrorsZ-Valuesp-Values
Intercept−79.385.799−13.69<0.0001
Forest cover (1400 m radius)2.4930.182013.69<0.0001
Forest cover2−0.01930.0014−13.64<0.0001
Canopy surface0.01950.00375.329<0.0001
Crop size−0.00090.0021−0.4660.6415
Crop x Canopy surface−0.000070.00002−3.4830.0005
Scaled auto-covariate−0.18730.0635−2.8230.0048
95% confidence interval of coefficients; Forest cover: 2.136–2.850, Forest cover2: −0.0221–−0.0165; Efron’s pseudo R-squared 0.3867; Partial Efron’s pseudo R-squared for forest cover effects 0.1247; Likelihood ratio test deviance = 464.63, ddl = 6, p-value < 0.0001.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Hambuckers, A.; Trolliet, F.; Simon, A.; Cazetta, E.; Rocha-Santos, L. Seed Removal Rates in Forest Remnants Respond to Forest Loss at the Landscape Scale. Forests 2020, 11, 1144. https://doi.org/10.3390/f11111144

AMA Style

Hambuckers A, Trolliet F, Simon A, Cazetta E, Rocha-Santos L. Seed Removal Rates in Forest Remnants Respond to Forest Loss at the Landscape Scale. Forests. 2020; 11(11):1144. https://doi.org/10.3390/f11111144

Chicago/Turabian Style

Hambuckers, Alain, Franck Trolliet, Astrid Simon, Eliana Cazetta, and Larissa Rocha-Santos. 2020. "Seed Removal Rates in Forest Remnants Respond to Forest Loss at the Landscape Scale" Forests 11, no. 11: 1144. https://doi.org/10.3390/f11111144

APA Style

Hambuckers, A., Trolliet, F., Simon, A., Cazetta, E., & Rocha-Santos, L. (2020). Seed Removal Rates in Forest Remnants Respond to Forest Loss at the Landscape Scale. Forests, 11(11), 1144. https://doi.org/10.3390/f11111144

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop