Ecology, 87(6), 2006, pp. 1489–1496 Ó 2006 by the Ecological Society of America




1 Ecology and Evolutionary Biology, University of Connecticut, Storrs, Connecticut 06269 USA Ecology and Evolutionary Biology, University of Colorado, Boulder, Colorado 80309-0334 USA

Abstract. For many plant species, seed dispersal is one of the most important spatial demographic processes. We used a diffusion approximation and a spatially explicit simulation model to explore the mechanisms generating seed dispersal kernels for plants dispersed by frugivores. The simulation model combined simple movement and foraging rules with seed gut passage time, plant distribution, and fruit production. A simulation experiment using plant spatial aggregation and frugivore density as factors showed that seed dispersal scale was largely determined by the degree of plant aggregation, whereas kernel shape was mostly dominated by frugivore density. Kernel shapes ranged from fat tailed to thin tailed, but most shapes were between an exponential and that of the solution of a diffusion equation. The proportion of dispersal kernels with fat tails was highest for landscapes with clumped plant distributions and increased with increasing number of dispersers. The diffusion model provides a basis for models including more behavioral details but can also be used to approximate dispersal kernels once a diffusion rate is estimated from animal movement data. Our results suggest that important characteristics of dispersal kernels will depend on the spatial pattern of plant distribution and on disperser density when frugivores mediate seed dispersal. Key words: animal movement; diffusion model; frugivory; kernel shape; landscape; plant and frugivore distribution; seed dispersal; spatial ecology; spatially explicit simulation.

INTRODUCTION For many plant species, seed dispersal is one of the most important spatial demographic processes, directly influencing the colonization of new habitats, population dynamics, genetic differentiation, and species interactions, as well as community structure and diversity (Nathan and Muller-Landau 2000, Levin et al. 2003, Levine and Murrell 2003). The probability of a seed being deposited at a particular distance from the parent plant can be described by functions called dispersal kernels (Nathan and Muller-Landau 2000). The characteristics of these kernels, in particular their scale and shape, can have significant ecological consequences. Mean dispersal distance and its variance set the spatial scale of dispersal, which, depending on the scale of individual interactions, can alter population dynamics, carrying capacity, and the coexistence of competitors (Bolker and Pacala 1999, Law et al. 2003, Snyder and Chesson 2003). Kernel shape can be summarized by its kurtosis, which indicates how the probability density is distributed among the peak and tails of the whole distribution. In nature, dispersal kernels often are Manuscript received 17 June 2005; revised 28 October 2005; accepted 6 December 2005. Corresponding Editor: E. Siemann. 3 Present address: The Statistical Laboratory, Centre for Mathematical Sciences, Wilberforce Road, Cambridge CB3 0WB UK. E-mail: [email protected] 4 Present address: University of Washington, 528 Kincaid Hall, Box 351800, Seattle, Washington 98115 USA.

leptokurtic, with a sharp peak near the point of origin and a long tail (Kot et al. 1996). Leptokurtosis can greatly increase the rate of spread of an invading organism or allele, and has been hypothesized to explain otherwise surprisingly fast rates of spread (Kot et al. 1996, Cain et al. 1998, Clark et al. 2001). Moreover, if the dispersal kernel is ‘‘fat’’ (i.e., its tail decays with distance at a slower rate than an exponential), invasion can progress with jumps, and with increasing rather than constant speed (Kot et al. 1996). What ecological mechanisms are behind the main attributes of dispersal kernels? Frugivorous (i.e., fruiteating) animals are the dominant seed dispersers for woody plant species in many temperate and tropical communities (Herrera 2002; see Plate 1). For these plants, seed dispersal kernels are a function of frugivore movement and gut passage (or regurgitation) times for seeds (Murray 1988, Schupp 1993). Assuming that, after consuming fruit, animals perform a random walk, it is possible to approximate movement with a diffusion equation (Turchin 1998). The solution of the diffusion equation can then be combined with a probability density function for gut passage time for seeds in order to solve for the distances at which they would be deposited. However, diffusion may be a poor approximation for frugivore movements (Holbrook and Smith 2000, Westcott and Graham 2000, Wenny 2001), especially at the small temporal scales defined by gut passage times. Furthermore, animals may detect and




Ecology, Vol. 87, No. 6

PLATE 1. Dendroica tigrina feeding on the fruits of Trema lamarkiana in Maricao, Puerto Rico. This Neotropical migrant feeds commonly on nectar and fruit in its Neotropical wintering grounds. Photo credit: T. A. Carlo.

track fruit abundance at several spatial and/or temporal scales (Levey 1988, Aukema and Martı´ nez and del Rio 2002, Kwit et al. 2004, Saracco et al. 2004). Thus, we expect that the spatial pattern of plant distribution will feed back into the characteristics of seed dispersal kernels via its effects on frugivore movements. In addition, frugivore density may also affect seed dispersal because fruit distribution and abundance would change as fruit ripens and animals track and consume them. Despite these expected connections among plant distribution, frugivore density, and dispersal, most theoretical studies set the characteristics of dispersal kernels first and then ask about ecological or evolutionary consequences. Interestingly, the consequences of dispersal, whether ecological or evolutionary, almost inevitably would be mediated by its effects on the spatial distribution of individuals. However, the arguments in the preceding paragraph suggest that, at least for animal-dispersed plants, changes in spatial aggregation of plants could change dispersal kernels. What kind of

changes should we expect? Would there be changes in both scale and shape? How would frugivore density affect these changes? To answer these questions, and in order to draw links between seed dispersal and plant population dynamics, we need a mechanistic understanding of frugivore-generated dispersal kernels. Toward this goal, we formulated a diffusion model as a baseline and then developed a spatially explicit stochastic model that simulated frugivores feeding on fruit and dispersing seeds in a plant population. We found that seed dispersal scale was largely determined by the degree of plant aggregation, whereas kernel shape was mostly dominated by frugivore density. MATERIALS



Diffusion model We first present a diffusion model as outlined in the Introduction. This model can be considered as a ‘‘null model’’ to be compared with the simulation models,

June 2006


which relax some of the assumptions that make the diffusion model mathematically tractable. According to this model, a bird consumes fruit and then starts a random walk, approximated by a diffusion equation:  2  ]nðx; y; tÞ ] n ]2 n ¼D ð1Þ þ ]t ]x2 ]y2 where n(x, y, t) is the probability of finding the individual near coordinates (x, y), assuming initial coordinates (0, 0) and given that it has been moving during time t with diffusion rate Dq (Turchin 1998). Defining displacement ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ffi

distance as r ¼ Eq. 1 is

ðx  x0 Þ2 þ ðy  y0 Þ2 , the solution for

nðr; tÞ ¼

  r r2 exp  : 2Dt 4Dt


The seeds will travel with the animal for a period of time given by gut passage time (Murray 1988, Wahaj et al. 1998). Gut passage time for seeds can be described by a Gamma distribution: 1 gðtÞ ¼ ðt=bÞa1 expðt=bÞ ð3Þ bCðaÞ where a and b are the shape and scale parameters, respectively. We solved for the seed dispersal kernel f(r) by integrating over time the product of the displacement probability density, Eq. 2, and the probability density for gut passage time, Eq. 3: Z ‘ f ðrÞ ¼ nðr; tÞgðtÞ dt ð4Þ 0

 a   2 1 kr 2aKðgÞ ðkrÞ  krKðgþ1Þ ðkrÞ ð5Þ rCðaÞ 2 pffiffiffiffiffiffiffiffiffiffiffi where k ¼ 1=bD and K(g)() are modified Bessel functions of the second kind of order (g). Eq. 5 can be evaluated numerically to show how dispersal kernels change with changes in gut passage time and with diffusion rate (see Results). Other studies have used similar methods to model dispersal when individuals settle after moving for a variable amount of time (Turchin and Thoeny 1993, Yamamura 2002). f ðrÞ ¼ 

Simulation model We developed a spatially explicit, event-driven, stochastic simulation of bird foraging and fruit production. Here we present a summary of the model, but a full description, together with default parameter values, can be found in Appendix A. Our choice of functional forms and parameter values was guided by behavioral observations and data reported in the literature, although some aspects were chosen for simplicity and flexibility. The importance of most assumptions was assessed with a sensitivity analysis (Appendix B). Simulated birds spent a variable amount of time (sampled from a probability distribution) perching and eating at fruiting plants. We kept perching time independent of fruit abundance because many factors other than resource


availability can influence how much time a bird spends at a perch. For example, birds might be chased by other birds, be ‘‘on the move’’ to avoid predators, or perch for a while to rest. Fruit consumption followed a hyperbolic functional response, but was kept within the limits of gut size. For simplicity, animals moved from plant to plant in straight lines and at constant speed. When choosing where to perch next, individuals sampled from an attraction distribution that was compiled by weighting fruit abundance and distance from current location (i.e., attraction increased with number of fruits and decreased with distance). Each fruit was assumed to contain a single seed, and after ingestion, gut passage time was sampled from a Gamma distribution with mean and SD of 27 6 15 minutes. The scale and shape of this Gamma were chosen to match gut passage rates of seeds reported from several frugivorous bird species (Murray 1988, Wahaj et al. 1998). To ease computations, all seeds from a frugivory event had identical gut passage time. Birds defecated seeds at the time dictated by gut passage time, irrespective of whether the animals were perching or flying. The program recorded the spatial coordinates of each dispersed seed, as well as the identity of the mother plant. Simulated birds kept moving, eating, and dispersing seeds until they accumulated 6 hours of daily activity. At the end of each simulated day, every plant produced new ripe fruits according to a regrowth model. We varied the degree of aggregation in plant distributions on simulated landscapes using a Neyman-Scott cluster point process (Appendix A). Simulated landscapes were composed of 200 clusters and a total of 1000 plants distributed over a 5000 3 5000 m area. Simulation experiment We performed simulation experiments following a factorial design. We varied number of birds in the landscape (1, 10, and 100) and the degree of aggregation of plants in the landscape. For each combination of factors, we ran 30 replicates with a new landscape for each replicate. Each bird started the simulations at a randomly chosen plant. During 30 simulated days, the program kept track of the number of fruits removed per plant and the dispersal distances of seeds. As stated in the Introduction, both scale and shape of dispersal kernels are of particular interest, given their potential influences on ecological and evolutionary processes. We quantified these properties of dispersal by both sample statistics (mean and kurtosis) and by fitting Weibull distributions to dispersal distances. Mean dispersal distance was calculated for all plants that dispersed seeds, but we only considered plants with at least 10 fruits removed for kurtosis and the fit of Weibull distributions. Maximum likelihood was used to fit the Weibull distributions using the density function: WðrÞ ¼ jmr m1 expbjr m c


where j is a scale parameter and m is a shape parameter. Note that when m ¼ 2, this is the same as Eq. 2, with j ¼



Ecology, Vol. 87, No. 6

FIG. 1. (a, c) Gut passage time densities and (b, d) the corresponding seed dispersal kernels, assuming different diffusion rates by dispersers. Gut passage time is given by a Gamma distribution (Eq. 3). Panel (a) has parameter values 10.0 for scale and 1.0 for shape. Panel (c) has 9.0 for scale and 3.0 for shape. Seed dispersal kernels are given by Eq. 5. Diffusion rate is 100 m2/min for thin lines, 500 for thick lines, and 1000 for thicker lines. Inserts in (b) and (d) show probability densities, in log scale, at large distances.

1/4Dt. For m ¼ 1, the distribution has an exponential tail, and when m , 1, the distribution has a fat tail. Thus, the Weibull distribution has a very flexible shape that may approximate a variety of dispersal kernels. To quantify how the movement of simulated animals differed from a simple diffusion, we calculated effective diffusion rates of birds at different time intervals from their mean squared displacement (Turchin 1998, Morales 2002). Effective diffusion rate indicated how fast individuals would have had to diffuse in order to match observed mean squared displacement. Effective diffusion rate at time t was defined as Dt ¼ R2t =4t, where R2t is the squared displacement after moving during time t, averaged over all individuals. RESULTS Diffusion model Gut passage time and diffusion rate can combine to generate a variety of seed dispersal kernels (Fig. 1), with scale determined mostly by diffusion rate and shape governed by gut passage time. The solution for the diffusion model (Eq. 5, Fig. 1), shows that increasing either or both the scale and shape parameters (a and b) in the gut passage time distribution and/or increasing the diffusion rate of animals (D) resulted in increasing

dispersal distance. The tails of the seed dispersal kernels produced by the diffusion model are always exponential (e.g., inserts in Fig. 1), and of course, there was no relationship between landscape characteristics and dispersal kernels. Simulation model Relaxing the random-walk assumption using the simulation model resulted in both the spatial distribution of plants and the density of dispersers having important effects on the characteristics of seed dispersal kernels. Mean dispersal distance increased with the number of dispersers and decreased as plants were more aggregated. However, there was little difference in mean dispersal distances when one or 10 frugivores were available. Also, there was little difference among the least clustered landscapes (Fig. 2a). A fixed-effects twoway ANOVA with number of dispersers and degree of aggregation in plants as factors showed that 89.01% of the variation in mean dispersal distance was related to the spatial aggregation of plants (SS ¼ 403130.4, F3, 348 ¼ 2708.99). The number of frugivores explained 6.84% of the variation (SS ¼ 30988.71, F2, 348 ¼ 312.36), and the interaction explained 0.33% (SS ¼ 1515.97, F6, 348 ¼ 252.66). Kurtosis in dispersal distance increased with increasing number of dispersers and with increasing

June 2006



138 328 plants with more than 10 fruits removed (Appendix C). Inspection of shape parameters indicated that kernels ranged from fat tailed (m , 1) to very thin tailed (m . 3), but most values were between 1 and 2, which correspond, respectively, to an exponential distribution (m ¼ 1), and the solution of simple diffusion (m ¼ 2; Eq. 2). The percentage of dispersal kernels with fat tails was highest for landscapes with the most clumped plant distributions, and increased with increasing number of dispersers (3.38% for one disperser; 5.07% for 10 dispersers; and 16.21% for 100). All other landscape and disperser density combinations had ,2% of fat-tailed dispersal kernels (Appendix C: Fig. C1). Frugivore movements and landscape structure The effective diffusion rate of simulated animals changed with time. In general, effective diffusion decreased sharply during the first few minutes of simulation and then increased to a maximum rate at about 10–20 minutes, followed by a smooth decline (Fig. 3a, b). Effective diffusion rates increased with the number of dispersers and decreased with increasing the degree of aggregation of plants (Fig. 3a, b). These changes in effective diffusion rate according to varying landscape properties and number of dispersers can explain some of the differences in dispersal kernels. Seed dispersal kernels that were calculated by plugging in averaged effective diffusion rate as D into Eq. 5 were similar to averaged dispersal kernels fitted to simulated data (Fig. 3c, d), but there were considerable differences at the tails of the distributions (inserts in Fig. 3c, d). Sensitivity analysis

FIG. 2. Changes in dispersal distance and kurtosis. (a) Dispersal (mean 6 SD) was shortest for highly clumped landscapes (circles), and increased with increasing number of frugivores. (b) Kurtosis (mean 6 SD) increased with increasing spatial aggregation of plants and with frugivore density. Data are from 30 replicates for each combination of landscape and number of frugivores (1, 10, or 100 dispersers).

spatial aggregation in plants (Fig. 2b). The number of dispersers accounted for 62.43% of the variability in kurtosis (SS ¼ 765.37, F2, 348 ¼ 727.11). The spatial aggregation of plants explained 18.23% (SS ¼ 223.57, F3, 348 ¼ 141.59), and the interaction explained 0.04% (SS ¼ 53.97, F6, 348 ¼ 17.09). The variability in dispersal distances was well described by Weibull distributions for 96.58% of the

The sensitivity analysis identified scale of perching time and scale and shape of gut passage time as the parameters with larger main effects on average dispersal distance (main effects are expected variance reduction in model output due to fixing a parameter while varying all other parameters). Total sensitivity for average dispersal distance (i.e., the sum of all the sensitivity indices, including all of the interaction effects) was largest for the parameter governing the scale of the distribution of perching time, followed by plant spatial aggregation and scale of gut passage time. However, shape of perching and gut passage, together with maximum gut capacity, were also important. Main effects on kurtosis of dispersal distances were largest for shape and scale of gut passage time, followed by number of frugivores. Total sensitivity for kurtosis was greater for gut passage time parameters, followed by number of frugivores. Perching time parameters and maximum gut capacity also had important total sensitivity values (Appendix B: Table B1). DISCUSSION Theoretical studies have shown that both the scale and shape of dispersal kernels can affect many



Ecology, Vol. 87, No. 6

FIG. 3. (a, b) Changes in ‘‘effective diffusion’’ with time for animals in different landscapes, and (c, d) comparison between averaged dispersal kernels from simulations with kernels from Eq. 5, using mean effective diffusion as parameter D. Results are for 10 individuals (a, c) and 100 individuals (b, d). Solid circles in (a) and (b) are for highly clustered landscapes (e.g., Appendix A: Fig. A1a), and open circles are for the least aggregated landscapes (e.g., Appendix A: Fig. A1d). Averaged kernels are in thin lines for the highly clustered landscapes and in thick lines for the least aggregated ones. Kernels obtained from Eq. 5 with averaged effective diffusion as D, are in dash-dot lines for clustered landscapes and in dotted lines for the least aggregated ones. Inserts in (c) and (d) show probability densities, in log scale, for large distances.

ecological processes (Kot et al. 1996, Bolker and Pacala 1999, Law et al. 2003, Levin et al. 2003, Levine and Murrell 2003, Snyder and Chesson 2003). Here we focus on bird-dispersed plants, a prevalent species group in many temperate and tropical woody plant communities (Herrera 2002). We started with a mathematically tractable diffusion model to explore how different seed dispersal kernels were produced by combining bird movement rates and gut passage times of seeds (Eq. 5 and Fig. 1). We followed with a spatially explicit simulation model, in which the unrealistic assumption of random-walk movement by frugivorous birds was relaxed. The simulation model also differed from the diffusion model in that it included more elements of the complex plant–frugivore seed dispersal interaction such as plant fruit production rates, crop size effects on bird visitation, and limits to the intake of fruits per plant visit. Both models were used to explore the mechanisms

generating seed dispersal kernels for plants dispersed by frugivores. Equipped with simple, stochastic movement rules, simulated birds adjusted their movements to the dynamics of different landscapes, which translated into differences in the scale and shape of dispersal kernels. The scale of dispersal kernels was largely influenced by the degree of plant aggregation, whereas kernel shape was mostly dominated by frugivore density (Fig. 2). Overall, mean dispersal distances were reduced as plant spatial aggregation increased, because simulated birds were retained in areas of high plant (and fruit) density. However, the changes in mean dispersal distances were not linearly related to the degree of plant aggregation (Fig. 2a); probably due to an interaction between movements and gut passage time. Changes in kernel shape are revealed by the average kurtosis of kernels, which increased with plant aggrega-

June 2006


tion but were strongly affected by frugivore density (Fig. 2b). Similarly, Weibull distributions fitted to seed dispersal distance data showed increased percentages of kernels with fat tails at high levels of frugivore density and plant aggregation (Appendix C: Fig. C1). High frugivore density resulted in resources being depleted locally, which, combined with a highly clustered landscape, caused simulated birds to make occasional long moves among clusters. The sensitivity analysis showed that both average kurtosis and mean dispersal distance were mainly influenced by the parameters of the distributions of perching time and gut passage time. Clearly, if birds stay long at perches, they will not be far from where they consumed fruits by the time seeds are deposited. Also, the longer seeds take to go through the gut, the greater the dispersal distance (Fig. 1), and if plants are clustered, birds can move short distances while foraging. Finally, when the distribution of gut passage time has a long tail, some seeds can travel within birds for long time, potentially dispersing far away from the mother plant. Maximum gut capacity was also important for mean dispersal distance and average kurtosis. Large gut capacity eases the constraint in fruit consumption and can be analogous to increasing frugivore density because, in both cases, local resources could be quickly depleted, promoting movement and thus increasing mean dispersal distances. Conveniently, all of these variables can be measured with relative ease in the field when compared to quantifying plant attractivity based on distances and fruit crop sizes (which turned out to be less important, although not negligible; Appendix B: Table B1). The diffusion model showed how gut passage time can interact with frugivore movement rates to produce different kernels (Eq. 5 and Fig. 1). This diffusion model not only can serve as a yardstick for models including more behavioral details, but also can be useful for approximating seed dispersal kernels once a rate of ‘‘effective diffusion’’ is estimated from empirical data or more complex models. This can be done by replacing D in Eq. 5 by the diffusion rate needed to match the squared displacement of simulated (or observed) animals (Fig. 3). Thus, different landscapes and frugivore densities would mean different effective diffusion rates, which would translate into different seed dispersal kernels. However, the tail of the dispersal kernels calculated from Eq. 5 will always be exponential (e.g., inserts in Fig. 1), which contrasts with the range of tail shapes, from fat to thin, that were observed in the simulation model (Appendix C: Fig. C1, and inserts in Fig. 3c, d). Nevertheless, kernel tails usually can be safely ignored for the study of plant population dynamics and species coexistence (Snyder and Chesson 2003). Details of kernel kurtosis and tail ‘‘fatness’’ are usually more relevant for the study of plant invasions and for modeling responses of plant communities to climate change (Levin et al. 2003).


To our knowledge, this is the first modeling effort linking plant distribution and frugivore movement behavior (modulated also by fruit production and animal density) to properties of seed dispersal kernels. Seeking generality and tractability, we made some simplifying assumptions and there are reasons to suspect that other behavioral details could further affect kernels. Still, we believe that adding more behavioral detail would increase, rather than decrease, the effects of plant distribution and frugivore density on seed dispersal kernels. For example, Wenny and Levey (1998) showed how male Three-wattled Bellbirds (Procnias tricarunculata) dispersed seeds of Ocotea endresiana to suitable microsites because they preferred perches in forest gaps for singing. Also, Carlo (2005) found that territorial behavior at some types of plant neighborhoods increased the spatial diffusion of seeds. The presence of other co-fruiting plant species can affect frugivore movements, causing interspecific seeds to gravitate toward one another or stay segregated, depending on dietary preferences of frugivores (Clark et al. 2004, Carlo and Aukema 2005). Finally, birds have spatial memory and show regularities in their movements, even within home ranges (Westcott and Graham 2000). Although different behaviors can alter seed dispersal, the validity of our findings should be quite general because as long as frugivores are able track resources, their movement will be related to the spatial distribution of fruiting plants. Also, if plants do not offer an unlimited supply of fruit, local depletion related to frugivore density would force animals to alternate between localized movement within a fruiting patch and longer moves between patches. Most theoretical studies set the characteristics of kernels first and then examine ecological and/or evolutionary consequences. Studies have shown that the behavior of individuals and the properties of landscapes can interact to produce different animal movement patterns (Morales 2002, Morales et al. 2004, Levey et al. 2005). Our models suggest that plant distribution patterns would feed back into the characteristics of dispersal kernels of frugivore-dispersed plants. Furthermore, the nature of this feedback most likely would be modulated by frugivore density. How would the linkage between spatial pattern and seed dispersal affect plant population dynamics? Clumped spatial distribution of plants resulted in shorter dispersal distances, which in theory should promote aggregation and increase competition. However, we found that a fraction of seed dispersal kernels in highly aggregated landscapes had fat tails, especially when frugivore density was also high. This potentially could take seeds far from crowded sites, creating new clusters of plants. Thus, the link between the spatial distribution of plants and properties of the seed dispersal kernels has the potential to create interesting plant population dynamics and deserves rigorous study. Understanding the relationships among frugivore movement, seed dispersal, and spatial hetero-



geneity should contribute to the general goal of linking the behavior of organisms to population and community dynamics. ACKNOWLEDGMENTS We thank Ine´s Armenda´riz, Ben Bolker, Robert Dunn, Nick Haddad, Carlos Martı´ nez del Rı´ o, and two anonymous reviewers for useful comments on the manuscript. This work was supported in part by NSF grant DEB 0407826 to T. A. Carlo, the University of Colorado, the University of Puerto Rico, and by E. Santiago-Valentı´ n and the Botanical Garden of the University of Puerto Rico, and by NSF grant 0078130 to J. M. Morales. LITERATURE CITED Aukema, J. E., and C. Martı´ nez del Rio. 2002. Where does a fruit-eating bird deposit mistletoe seeds? Seed deposition patterns and an experiment. Ecology 83:3489–3496. Bolker, B. M., and S. W. Pacala. 1999. Spatial moment equations for plant competition: understanding spatial strategies and the advantages of short dispersal. American Naturalist 153:575–602. Cain, M. L., H. Damman, and A. Muir. 1998. Seed dispersal and the Holocene migration of woodland herbs. Ecological Monographs 68:325–347. Carlo, T. A. 2005. Interspecific neighbors change seed dispersal pattern of an avian-dispersed plant. Ecology 86:2440–2449. Carlo, T. A., and J. E. Aukema. 2005. Female-directed dispersal and facilitation between a tropical mistletoe and its dioecious host. Ecology 86:3245–3251. Clark, C. J., J. R. Poulsen, E. F. Connor, and V. T. Parker. 2004. Fruiting trees as dispersal foci in a semi-deciduous tropical forest. Oecologia 139:66–75. Clark, J. S., M. Lewis, and L. Horvath. 2001. Invasion by extremes: population spread with variation in dispersal and reproduction. American Naturalist 157:537–554. Herrera, C. M. 2002. Seed dispersal by vertebrates. Pages 185– 208 in C. M. Herrera and O. Pellmyr, editors. Plant–animal interactions: an evolutionary approach. Blackwell, Oxford, UK. Holbrook, K. M., and T. B. Smith. 2000. Seed dispersal and movement patterns in two species of Ceratogymna hornbills in a West African tropical lowland forest. Oecologia 125:249– 257. Kot, M., M. A. Lewis, and P. van den Driessche. 1996. Dispersal data and the spread of invading organisms. Ecology 77:2027–2042. Kwit, C., D. J. Levey, C. H. Greenberg, S. F. Pearson, J. P. McCarty, and S. Sargent. 2004. Cold temperature increases winter fruit removal rate of a bird-dispersed shrub. Oecologia 139:30–34. Law, R., D. J. Murrell, and U. Dieckmann. 2003. Population growth in space and time: spatial logistic equations. Ecology 84:252–262.

Ecology, Vol. 87, No. 6

Levey, D. J. 1988. Topical wet forest treefall gaps and distributions of understory birds and plants. Ecology 69: 1076–1089. Levey, D. J., B. M. Bolker, J. J. Tewksbury, S. Sargent, and N. M. Haddad. 2005. Effects of landscape corridors on seed dispersal by birds. Science 309:146–148. Levin, S. A., H. C. Muller-Landau, R. Nathan, and J. Chave. 2003. The ecology and evolution of seed dispersal: a theoretical perspective. Annual Review of Ecology, Evolution and Systematics 34:575–604. Levine, J. M., and D. J. Murrell. 2003. The community-level consequences of seed dispersal patterns. Annual Review of Ecology, Evolution and Systematics 34:549–574. Morales, J. M. 2002. Behavior at habitat boundaries can produce leptokurtic movement distributions. American Naturalist 160:531–538. Morales, J. M., D. T. Haydon, J. Frair, K. E. Holsinger, and J. M. Fryxell. 2004. Extracting more out of relocation data: building movement models as mixtures of random walks. Ecology 89:2436–2445. Murray, K. G. 1988. Avian seed dispersal of three neotropical gap-dependent plants. Ecological Monographs 58:271–298. Nathan, R., and H. C. Muller-Landau. 2000. Spatial patterns of seed dispersal, their determinants and consequences for recruitment. Trends in Ecology and Evolution 15:278–285. Saracco, J. F., J. A. Collazo, and M. J. Groom. 2004. How do frugivores track resources? Insights from spatial analyses of bird foraging in a tropical forest. Oecologia 139:235–245. Schupp, E. W. 1993. Quantity, quality and the effectiveness of seed dispersal by animals. Vegetatio 107/108:15–29. Snyder, R. E., and P. Chesson. 2003. Local dispersal can facilitate coexistence in the presence of permanent spatial heterogeneity. Ecology Letters 6:301–309. Turchin, P. 1998. Quantitative analysis of movement: measuring and modeling population redistribution in animals and plants. Sinauer Associates, Sunderland, Massachusetts, USA. Turchin, P., and W. T. Thoeny. 1993. Quantifying dispersal of southern pine beetles with mark–recapture experiments and a diffusion model. Ecological Applications 3:187–198. Wahaj, S. A., D. J. Levey, A. K. Sanders, and M. L. Cipollini. 1998. Control of gut retention time by secondary metabolites in ripe Solanum fruits. Ecology 79:2309–2319. Wenny, D. G. 2001. Advantages of seed dispersal: a re-evaluation of directed dispersal. Evolutionary Ecology Research 3:51–74. Wenny, D. G., and D. J. Levey. 1998. Directed seed dispersal by bellbirds in a tropical cloud forest. Proceedings of the National Academy of Sciences (USA) 95:6204–6207. Westcott, D. A., and D. L. Graham. 2000. Patterns of movement and seed dispersal of a tropical frugivore. Oecologia 122:249–257. Yamamura, K. 2002. Dispersal distance of heterogeneous populations. Population Ecology 44:93–101.

APPENDIX A Model description (Ecological Archives E087-088-A1).

APPENDIX B Sensitivity analysis (Ecological Archives E087-088-A2).

APPENDIX C Scales and shapes of seed dispersal kernels (Ecological Archives E087-088-A3).

the effects of plant distribution and frugivore ... - Semantic Scholar

The simulation model combined simple movement and foraging rules with seed gut ... dispersal kernels once a diffusion rate is estimated from animal movement data. ..... A1a), and open circles are for the least aggregated landscapes (e.g., ...

258KB Sizes 0 Downloads 150 Views

Recommend Documents

The Smoothed Dirichlet distribution - Semantic Scholar
for online IR tasks. We use the .... distribution requires iterative gradient descent tech- .... ous degrees of smoothing: dots are smoothed-document models. )10.

on the probability distribution of condition numbers ... - Semantic Scholar
Feb 5, 2007 - distribution of the condition numbers of smooth complete ..... of the polynomial systems of m homogeneous polynomials h := [h1,...,hm] of.

The effects of a levels-of-processing manipulation ... - Semantic Scholar
ognize or recall an associate because it had been acti- ... major premise of SAC is that both the degree to which a concept is ..... Psychological Science, 9, 20-26.

Effects of Decision Variables and Intraparietal ... - Semantic Scholar
Oct 3, 2012 - V.W., and M.F.S.R. analyzed data; I.C.G., A.C.N., and M.F.S.R. wrote the paper. This work was .... EEG recording, preprocessing, and spectral analysis. ...... Oostenveld R, Fries P, Maris E, Schoffelen JM (2011) Fieldtrip: open source s

on the probability distribution of condition numbers ... - Semantic Scholar
Feb 5, 2007 - of the polynomial systems of m homogeneous polynomials h := [h1,...,hm] of ... We will concentrate our efforts in the study of homogeneous.

Effects of Hidden Unit Sizes and Autoregressive ... - Semantic Scholar
mental tasks from 4 subjects were used in the experimental study and combinations ... R.Palaniappan is with the Dept. of Computer Science, University of Essex,.

Effects of an Omnivorous Katydid, Salinity, and ... - Semantic Scholar
Dec 21, 2011 - they both support higher trophic levels and regulate plant community structure and biomass; however, understanding what regulates ... C.-K. Ho . S. C. Pennings (*). Department of Biology and Biochemistry, University of Houston,. Housto

Polarization effects and charge transfer in the KcsA ... - Semantic Scholar
b International School for Advanced Studies, SISSA-ISAS and INFM-Democritos Center, via Beirut 4, 34014 Trieste, ... The electronic structure of the selectivity filter of KcsA K+ channel is ... which features the conserved TVGYG signature [6,7].

The Redistributive Effects of Political Reservation ... - Semantic Scholar
The data source is the Planning Commission, Government of India.26. 25Specifically, for a ..... differential isolation story, we do not find direct evidence of it here.