Seed dispersal by animals is considered a pivotal ecosystem function that drives plant-community dynamics in natural habitats and vegetation recovery in human-altered landscapes. Nevertheless, there is a lack of suitable ecological knowledge to develop basic conservation and management guidelines for this ecosystem service. Essential questions, such as how well the abundance of frugivorous animals predicts seeding function in different ecosystems and how anthropogenic landscape heterogeneity conditions the role of dispersers, remain poorly answered. In three temperate ecosystems, we studied seed dispersal by frugivorous birds in landscape mosaics shaped by human disturbance. By applying a standardized design across systems, we related the frequency of occurrence of bird-dispersed seeds throughout the landscape to the abundance of birds, the habitat features, and the abundance of fleshy fruits. Abundance of frugivorous birds in itself predicted the occurrence of dispersed seeds throughout the landscape in all ecosystems studied. Even those landscape patches impoverished due to anthropogenic disturbance received some dispersed seeds when visited intensively by birds. Nonetheless, human-caused landscape degradation largely affected seed-deposition patterns by decreasing cover of woody vegetation or availability of fruit resources that attracted birds and promoted seed dispersal. The relative role of woody cover and fruit availability in seed dispersal by birds differed among ecosystems. Our results suggest that to manage seed dispersal for temperate ecosystem preservation or restoration one should consider abundance of frugivorous birds as a surrogate of landscape-scale seed dispersal and an indicator of patch quality for the dispersal function; woody cover and fruit resource availability as key landscape features that drive seedfall patterns; and birds as mobile links that connect landscape patches of different degrees of degradation and habitat quality via seed deposition.

No abstract

Abstract. Plant-animal interactions are crucial nodes in the structure of communities and pivotal drivers of ecosystem functioning. Much of this relevance may depend on how animals cope with plant resources at different spatial scales. However, little is known about how and why different interactions perform at different scales in the same environmental setting. In this study we assess the spatial scales at which two plant-animal interactions operate and disentangle the environmental factors (plant resource availability vs. habitat structure) underpinning these operational scales. We studied two interactions with opposite (mutualistic vs. antagonistic) ecological effects on fleshy-fruited trees, frugivory and seed dispersal by birds, and the later predation by rodents on bird-dispersed seeds. Employing a standardized sampling, we covered three temperate ecosystems hosting structurally similar plant-frugivoreseed predator systems: Cantabrian forest, Mediterranean shrubland, and Patagonian forest. We sampled habitat structure (tree and understory covers), fleshy-fruit abundance, birddispersed seed occurrence, frugivorous bird abundance, and seed predation rate, along 1500-2500 m transects. Using a spatially explicit approach, we broke down the predictable spatial patterns of bird abundance and seed predation rate into patchiness at three consecutive spatial scales (broad, intermediate, and fine). The degree of patchiness and the allocation of spatial variability at different scales suggested a hierarchically nested structure in frugivory and seed predation, but a larger operational scale in seed predation than in frugivory. Scale-specific spatial distributions were explained by the response of animals to plant resource availability and habitat structure. Birds tracked fruits at large spatial scales in all systems and, within some systems, even across consecutive scales. Seed predation distribution was more responsive to habitat features than to resource availability. The reinforcement of resource tracking patterns across scales sometimes occurred simultaneously with the dilution of habitat effects, suggesting that scale dependence may emerge from trade-offs between resource acquisition and the effects of other factors, such as predation risk, on interacting animals. Our findings suggest that scale dependence in frugivory and seed predation may affect the balance of demographic effects of these interactions in plant populations. Moreover, the consistency of frugivory patterns within and across spatial scales may condition the redundancy of seed dispersal as an ecosystem function.

Summary 1.Seed dispersal is considered critical for shaping the spatial structure of plant populations, though little empirical effort has been made to interpret this effect in terms of the scale at which plant species are distributed and cope with environmental heterogeneity. We assessed the spatial role of seed dispersal in Tristerix corymbosus , a mistletoe dispersed exclusively in the temperate forests of Patagonia by the endemic marsupial Dromiciops gliroides . 2. We examined how fruit resource tracking and seed dispersal by the marsupial affects mistletoe recruitment, employing a spatially explicit approach aimed at breaking down the spatial structure of the mistletoe and marsupial populations at different scales. 3. In a single fruiting season, we evaluated the abundance of mistletoe fruits, adult plants, dispersed seeds and recruits (seedlings and saplings), as well as the abundance of marsupials, along a 1500-m linear transect. 4. Both mistletoe and marsupial abundances were distributed hierarchically in space, with large patches containing smaller ones. Marsupial patchiness matched that of mistletoe fruits, at least at a broad scale within the transect. Marsupial abundance also varied at a large-scale, being conditioned by habitat features and decreasing progressively along the transect. Mistletoe seed rain accounted for the patchiness of adult plants and fruits, and for the large-scale pattern of marsupial activity. The spatial pattern of mistletoe recruitment closely matched seed rain. 5. Synthesis. Seed dispersal by marsupials shaped the scale of mistletoe recruitment in two ways. First, marsupials created a spatial match between mistletoe adults and recruits as a result of fruit resource tracking. Second, they generated patchiness in mistletoe offspring at a larger scale than in adults. Dispersal process performed as a strong demographic filter capable of changing the mistletoe spatial structure from adults to recruits, despite a low frequency of far-from-adult dispersal events. Similar effects of scale shaping by seed dispersers may be generalized among plants in which there is a sharp spatial match between fruits and frugivores, and whose dispersed seeds have a higher probability of recruiting than undispersed ones.

Fruit colour influences fruit choice by seed dispersers. The mistletoe Tristerix corymbosus (Loranthaceae) produces mature fruits of two different colours in two different biomes: yellow in the Chilean matorral and green in the temperate forest of southern South America. We conducted field surveys to establish the association between fruit colour and disperser identity throughout the entire geographical range of T. corymbosus. We selected 22 populations, eight of which were located in the Chilean matorral and 14 in the temperate forest south of the matorral. To identify the seed dispersers of the mistletoe we used direct observation, camera traps, and live‐trapping of small mammals. We also report experiments to assess fruit selection by seed dispersers based on differences in colour. The assemblages of dispersers of T. corymbosus differ between the two biomes: yellow fruits in Chilean matorral are exclusively dispersed by three bird species while green fruits in the temperate forest are exclusively dispersed by a marsupial. The differences in the assemblages of seed dispersers can be explained by differences in food‐finding strategies between the two assemblages. Green fruits in temperate forest are not easily detected by birds, while colour might not be an important cue for the marsupial because it is nocturnal and uses other senses to locate food. We propose that the association between the marsupial and the green‐fruited mistletoe constitutes an ecological fitting rather than the outcome of a co‐evolutionary process. The marsupial might have allowed the mistletoe T. corymbosus to retain green coloration in mature fruit, a condition to which it is preadapted by a slower ripening process in temperate forest populations.