We have made substantial changes to incorporate your suggestions. We hope the revisions are satisfactory. Here are my detailed suggestions: i lines , which summarize messages of other papers: why should increased light etc favor a smaller set of species? There might be new species that exist only at those edges And what is meant by "natural enemies are less able to promote plant diversity at forest edges"? Whose enemies are meant? Those of plants? Then it might be better to write "herbivores" Additionally, the effect of insects and fungi, natural enemies of plants, in maintaining seedling diversity can weaken near forest edges Krishnadas et al.
Which features of seed dispersal can modify the effect of stabilizing mechanisms? You had a census in April and one in November: did you add up the two seedling numbers or average them? Did you chacke whether the April seedling were still present on November, or did you consider as seedlings only plants younger than 6 months? We conducted two censuses but only used the data from the final census to quantify annual recruitment and seed-to-seedling transition.
Yes, we tagged seedlings in April to monitor whether they survived until the next census in November. In fact, all seedlings present before the experiment began were also tagged so as to distinguish them from new recruits. Additionally, all new recruits were tagged for future monitoring. We have now added these details to the methods. Please say in more detail how you determined the a and b coefficients and show also a sample of raw data. If I understand it correctly, you had at each site and for each distance to the edge 15 sampling areas, so each a and each be is determined from a fit of the J s data through 15 data points.
I find the descriptions of the two coefficients confusing. You call b the "effect of CNDD" conspecific negative density dependence. The functional form of the equation used to model seed-to-seedling transition, incorporating baseline mean transition rate and density-dependence, is well established and used in numerous studies Harms et al.
Furthermore, we have explained in detail the functional form in our previous work and hence did not repeat it here Krishnadas et al. To reiterate briefly though, the reviewer has correctly understood the equation and its relevance to CNDD. We used a generalized linear mixed-effects model with a negative binomial link function in a hierarchical Bayesian framework with separate intercepts and slopes estimated per species.
I learned this only by looking at the figures, this was not said sufficiently clearly here. We do use the a and b values estimated empirically in the study except for question 4, where they are altered to check how robust our results are to measurement error. We now include this information immediately after describing the empirical model Lines Also, the 'microhabitat effect' is introduced without motivation.
I guess you need it in order to obtain a better agreement of hte model with data. Please include an explanation of why you include it. You already have a and b to characterize differences between sites, so m should be an effect within sites? The motivation for the microhabitat effect was based on the random effects from the empirical models where sites explained some variance in baseline survival.
And one more question concerning your model: how large were the lattices used in your simulations? Do I understand correctly that you simulated entire habitats that contain sites that have different distances to the edges? Or did each simulation only consider a small area that has is meant to have fixed distance from the edge and fixed values of a and b?
The reviewer has correctly understood the simulation. To reiterate, we simulated fragments that contained a large number of sites or grids think of these as a spot for one adult tree.
Lines You find less species near edges. Where are the reasons for this discussed in more depth? My first intuition would be that edge sites have less neighbors and receive a smaller diversity of seeds from their surroundings. But this effect cannot be disentangled from the effect of a larger b. Did you ever check what happens when you use the same a and b on all lattice sites? How does the diversity decrease near edges in this case? Please see our detailed response to point 5 from Reviewer 1.
To reiterate briefly, edges had higher mean abundance and diversity of seeds per site Krishnadas et al. Regarding a and b being the same for all sites, we ran simulations and found that the model was a very poor match to observations. Line What are "edge specialists" please no hyphen , and how do you identify them? Do you want to say here that there are tree species that exist mainly at edges?
Do they have a larger b? Or did you now use a new parameter, Ej x , mentioned in lines ? All species had the same Ej x. The subsequent text makes clear that this was done in the simulation. Please state this clearly to avoid confusion. I guess this follows from the general reduction in lambda for all species close to edges.
The reviewer is correct that lambda or invader growth is the parameter which represents stabilization, i. We have now restructured this paragraph and the one following lines , to show which specific results support our stated conclusion.
Our results for lambda suggest that even if a species were to disperse into a patch after local extirpation, the loss of stabilizing mechanisms make it highly unlikely for that species to persist and contribute to diversity in the long run.
However, increased dispersal did have a small positive effect on diversity. We note these points in the discussion of our results. Alone the observation that b increases from 0 to 1 when one goes from the interior towards the edge allow already the conclusion that it is harder for invaders to become established at edges.
Maybe the lambda study can be moved to the appendix. Overall, I found the number of figures somewhat overwhelming as the main message can be based upon less figures. However, we have retained the analysis with lambda because it is a core feature of our analysis and interpretation of changes to community stability. Long-standing theory and simulations based on empirical data have established lambda as a key parameter to assess stability. Additionally, our lambda analysis allowed us to specifically tease apart the impact of the b-parameter from other factors affecting coexistence such as habitat partitioning.
For example, we found that habitat partitioning did have a stabilizing effect in fragments, it was just very small, and smaller than most of the fitness differences it generated. But interaction with pollinators, seed dispersers, herbivores, etc will also change near edges and affect diversity.
This is a fair point. Based on our analysis, we can say that processes that contribute to stabilization may be more important in general than processes that lead to fitness differences. However, the other species would still persist, albeit at lower numbers, if stabilizing mechanisms operate to prevent full competitive exclusion.
We derive support for this from the results of our sensitivity analysis, which showed that changes in the relative importance of stabilizing mechanisms was key to maintaining diversity. We see a need for more studies that quantify how stabilization is mediated by different processes at different life-stages of long-lived organisms. Your manuscript has now been assessed by reviewers and by Iain Stott, the Associate Editor handling your manuscript.
We require you to make some revisions, before we can make a final decision for your paper. I agree with the AE and reviewers that your study is interesting and worth publishing. However, as noted there are several areas of concern, many related to your modelling and analysis, that need revision. You will need to provide additional explanation of your analyses, justify your approach, and likely do some additional analyses to adequately deal with these concerns. The deadline to submit your revision is: Oct When reading the manuscript, I had at several places problems to understand what the authors mean or wether the explanations given by the authors are the best ones, so I suggest major revisions to improve clarity of presentation and arguments.
Log in. Web of Science. Identifiers publons. In this model, both distance to the edge and sites differed in seed predation with an interaction between these two variables. Seeds had a tendency of larger predation in the fragment interior Figure 2 , however there is an abruptly fall in predation at 50 m in the control site, F0, increasing afterwards.
This fall is likely to be responsible for the site effect on the analysis, differentiating between F0 and F1. The results suggest a possible edge effect for seed predation in this area. In accordance with previous studies Burkey, Burkey, TV. However, the increasing in predation rates were only for distances larger than 50 m, and differing from those previous studies, predation rates were higher in the edge than 50 m distance.
The differing matrix surrounding the fragment between these studies might be the reason for this variation near the edge, since in our study, the soybean matrix could attract seed predators. As previously discussed, the majority of studies on seed predation does not take in consideration ant species. Some ant species have a larger density in forest edges — up to 50 m — Wirth et al. The difference between predation in the two blocks F0 and F1 , was probably due to the larger predation rate at 50 m in the control site than in the fire site.
The vegetation structure of the sites differed at this distance from the edge, with the control site having a denser and more complete cover than the relatively open environment of the fire site. Some seed predators might avoid open areas due to fear of predation. Even if this difference between sites is a real effect for the area, the edge effect remains, with more seeds preyed in the interior than in the edges in both areas suggesting that, even with the fire disturbance and different vegetation structures, the edge effect is a factor affecting seed predation rates at this site.
It should be noted that natural and artificial predation rates might differ Wong et al. The utilization of exotic seed species in this artificial experiment does not allow a precise estimate of the natural predation rates in this site Wong et al. It is possible, however, to deduce the seed predation pressure by predators, which in turn, may be used to estimate seeds vulnerability at community level Sieving, Sieving, K.
Nest predation and differential insular extinction among selected forest birds of Central Panama. Nest predation in insular and Mainland Lowland Rainforest in Panama. The Condor, vol. Henceforth, this seed predation should be taken in consideration in reforestation projects, where the main source of plants species is from seed distribution Diaz et al. Abrir menu Brasil. Brazilian Journal of Biology.
Abrir menu. Amazon Forest; edge effect; seed predation. Figure 2 Mean preyed seeds for each distance for both areas independently: a control site F0, and b annual fire F1. Figure 3 Mean preyed seeds for each distance for both areas jointly. References Aide, TM. Balch, JK. Baldissera, R. Burkey, TV. Carvalho, KS. Atta Acta Amazonica, vol. Cochrane, MA. Diaz, I.
Diaz, M. Ferreira, AV. Holl, K. Hoover, JP. Laurance, WF. For forest specialist species, edges maintained by natural processes were penetrable, allowing these species to move right through the edges, while edges still under anthropogenic interventions were impenetrable, preventing the dispersal of forest specialists out of the forest.
However, natural forest edges constituted a barrier and prevented the invasion of matrix species into the forest interior. Preserving and protecting all edges maintained by natural processes, and preventing anthropogenic changes to their structure, composition, and characteristics are key factors to sustain biodiversity in forests.
Moreover, the increasing presence of anthropogenic edges in a landscape is to be avoided, as they contribute to the loss of biodiversity. Simultaneously, edges under continued anthropogenic disturbance should be restored by increasing habitat heterogeneity.
Because of the importance and ubiquity of edges, ecological responses to their presence have been extensively researched Ries et al. Ries and Sisk and Ries et al. This unified model identifies ecological flows, access to spatially separated resources, resource mapping, and species interactions as fundamental mechanisms that change species abundance patterns across habitat edges.
Edge orientation Ries et al. The history of a habitat edge may determine its structural and functional properties and its ecological conditions Strayer et al. This can be generalized into an edge history hypothesis: Edges created by forces no longer in operation, and maintained only by natural processes mainly by succession and edges repeatedly disturbed by anthropogenic activities forestry, agriculture, urbanization have different structural and functional characteristics and have different influence on species richness and assemblage composition Strayer et al.
Although the history and maintenance of habitat edges have relevance, their impact on the edge effect has not yet been tested. We focused on forest edges, which are one of the most common habitat edges within terrestrial landscapes Murcia, We distinguished forest edges maintained by natural processes succession, irregular extensive grazing, and irregular mowing from edges repeatedly disturbed and maintained by anthropogenic influence.
Studies of edge effect on ground beetles reported findings that are contradictory or inconsistent. Some papers reported higher species richness in edges than in the forest interiors e. We hypothesized that the reported inconsistency is caused by differences between forest edges maintained by natural processes vs. More specifically, we hypothesized that forest edges maintained by natural processes have significantly higher carabid diversity than their interiors, while edges with continued anthropogenic influence do not.
We also hypothesized that species with different habitat affinities show idiosyncratic responses to forest edges, and edges still under anthropogenic influences are more easily penetrable for matrix species.
Our analysis shows that carabid diversity is higher only at edges not under continued human influence. Additionally, we scanned the reference section of the publications found in this search for additional, undetected, relevant publications.
Data were extracted from text, tables, and graphs. From papers that studied carabids along transects, only data from the interiormost forest locations were used. Samples in the forest edges were collected in the 0—15 m edge zone, where 0 m represents the line of outermost trees. Forest edges were classified according to their maintaining processes. Edges whose neighboring habitats have been unmanaged forest interiors without fire, cutting or thinning; adjacent grasslands or meadows without burning, intensive grazing, or mowing for at least 50 years were classified as edges maintained by natural processes.
These edges are maintained by natural processes such as succession or irregular interventions irregular mowing and irregular extensive grazing , with succession starting between such disturbance events.
At the assemblage level, the mean overall abundance and species richness of ground beetles Coleoptera: Carabidae were analyzed. Species with different habitat affinities may show different responses to edge effect; therefore, ground beetles were classified by habitat preference, distinguishing 1 forest specialists species associated with forest habitats ; 2 habitat generalists species occurring both in forests and other habitats ; and 3 species associated with open habitats.
Species whose habitat affinity could not be unequivocally categorized were not included into the analyses. A negative g value means higher abundance or species richness in forest edges than interiors. The two main groups were forest edges maintained by natural or anthropogenic processes. Edges with anthropogenic disturbances were further divided into subgroups based on the type of human influence forestry, urbanization, or agriculture.
We assessed whether effect sizes were homogenous or varied across studies i. To describe the heterogeneity, complementary measures, Q , T 2 , and I 2 , were estimated Borenstein et al. Significant variance between groups Q between means that edge effect on species richness or abundance significantly differed according to the history or the continued anthropogenic influence. To evaluate the proportion of true variance explained by the covariates subgroup classification , the R 2 was calculated Borenstein et al.
During the calculations, subgroups with less than five cases were excluded from subgroup categorical analyses. Therefore, we tested the publication bias using funnel plots and the Egger test Borenstein et al. In case of significant asymmetry, the trim and fill method was used as suggested by Duval and Tweedie This method calculates the number of missing studies and estimates their effect sizes as well as standard errors; then, these missing studies are added to the data set and the summary effect size is recomputed.
This method yields an unbiased estimate of the summary effect size Borenstein et al. The literature search yielded publications. After applying the selection criteria, 53 papers were retained.
Twelve papers studied forest edges maintained by natural processes, 26 papers investigated edges maintained by continued anthropogenic interventions, and a single study examined both. Edges maintained by human influence were further grouped according to the activity type: forestry 10 papers , urbanization three papers , or agriculture 13 papers.
Studies were carried out on all continents except Antarctica , with a majority from Europe 21 papers ; the number of experiments from Asia 6 and North America 7 were almost equal.
In anthropogenically maintained edges, the edge effect on abundance was not significantly related to the disturbance type agriculture vs. The abundance of ground beetles in edges and their interiors was not significantly different when edges were under agricultural or forestry disturbance Figure 1 a.
Values in brackets refer to the number of comparisons from which the mean effect size was calculated. Forest edges maintained by natural processes had significantly higher species richness than their interiors, while edges under continued anthropogenic influence showed no such difference Figure 1 b.
In either type of anthropogenically disturbed edges forestry or agriculture , the species richness in edges vs. There was no publication bias regarding species richness see Appendix S3. Population viability and harvest sustainability for Madagascar lemurs. Buckland, S. Line transect sampling of primates: can animal-to-observer distance methods work?
Camara-Cabrales, L. Carlo, T. Generalist birds promote tropical forest regeneration and increase plant diversity via rare-biased seed dispersal. Ecology 97, — Chapman, C. Primate seed dispersal and forest restoration: an African perspective for a brighter future. Cordeiro, N. Low recruitment of trees dispersed by animals in African forest fragments. Cosyns, E. Endozoochorous seed dispersal by cattle and horse in a spatially heterogeneous landscape. Plant Ecol. Couvreur, M. Large herbivores as mobile links between isolated nature reserves through adhesive seed dispersal.
Dausmann, K. Improved recruitment of a lemur-dispersed tree in Malagasy dry forests after the demise of vertebrates in forest fragments. Biased seed rain in forest edges: evidence from the Brazilian Atlantic forest. Dew, J. Biotropica 30, — Dunham, A. Fruiting phenology is linked to rainfall variability in a tropical rain forest. Biotropica 50, — Emer, C. Seed-dispersal interactions in fragmented landscapes—a metanetwork approach.
Seedling recruitment in semi-natural pastures: the effects of disturbance, seed size, phenology and seed bank. Eriksson, O. Seedling recruitment in deciduous forest herbs: the effects of litter, soil chemistry and seed bank.
Flora , 65— Farwig, N. Trait-associated loss of frugivores in fragmented forest does not affect seed removal rates. Garbutt, N. Mammals of Madagascar: A Complete Guide. Gascon, C. Receding forest edges and vanishing reserves. Science , — George, L.
The fern understory as an ecological filter: emergence and establishment of canopy-tree seedlings. Ecology 80, — Glessner, K. Population density and home range size of Indri indri in a protected low altitude rain forest. Gomes, L. Tolerance of frugivorous birds to habitat disturbance in a tropical cloud forest. Fruit production along roads and footpaths in an Atlantic rain forest area. How does avian seed dispersal shape the structure of early successional tropical forests? Gray, M. The response of avian feeding guilds to tropical forest disturbance.
Haddad, N. Harper, G. Fifty years of deforestation and forest fragmentation in Madagascar. IUCN Version Irwin, M. Patterns of species change in anthropogenically disturbed forests of Madagascar. Johns, A.
Responses of rain-forest primates to habitat disturbance: a review. Kembel, S. Picante: R tools for integrating phylogenies and ecology. Bioinformatics 26, — Klanderud, K. Recovery of plant species richness and composition after slash-and-burn agriculture in a tropical rainforest in Madagascar. Knoop, S. Lemur News 21, 20— Kraft, N.
Community assembly, coexistence and the environmental filtering metaphor. Kurten, E. Hunting alters seedling functional trait composition in a Neotropical forest. Ecology 96, — Lawton, R. Natural disturbance and gap-phase regeneration in a wind-exposed tropical cloud forest.
Ecology 69, — Lehman, S. Edge effects and their influence on lemur density and distribution in southeast Madagascar. Edge effects on the density of Cheirogaleus major. Lenz, B. Edge effects in the primate community of the biological dynamics of forest fragments project, Amazonas, Brazil. Lindenmayer, D. Forest fragmentation drives Atlantic forest of northeastern Brazil to biotic homogenization.
Magrach, A. Meta-analysis of the effects of forest fragmentation on interspecific interactions. Mantia, T. Vertebrate-mediated seed rain and artificial perches contribute to overcome seed dispersal limitation in a Mediterranean old field. Martinez, B. Frugivory and seed dispersal patterns of the red-ruffed lemur, Varecia rubra , at a forest restoration site in Masoala National Park, Madagascar.
Folia Primatol. McArdle, B. Fitting multivariate models to community data: a comment on distance-based redundancy analysis. Ecology 82, — McConkey, K. Seed dispersal in changing landscapes. Melo, F. Forest fragmentation reduces recruitment of large-seeded tree species in a semi-deciduous tropical forest of southern Mexico. Messina, S. Selective logging reduces body size in omnivorous and frugivorous tropical forest birds. Michalski, F. Disturbance-mediated drift in tree functional groups in Amazonian forest fragments.
Biotropica 39, — Morelli, T. Murcia, C. Edge effects in fragmented forests: implications for conservation. Trends Ecol. Murphy, A. Estimating encounter rates and densities of three lemur species in northeastern Madagascar.
Myers, J. Seed arrival, ecological filters, and plant species richness: a meta-analysis.
0コメント