Our environmental correlation results provide evidence for a habitat filtering process in which agricultural adapters are promoted by vineyard expansion, while co-occurrence of and negative interactions between agricultural adapters and oak woodland birds provide evidence for indirect detrimental effects of land conversion. These findings support a more pluralistic hypothesis that community structure is not influenced solely by habitat filtering or by species interactions, but rather is influenced by both mechanisms and interactions between them. It is not surprising that agricultural adapters and woodland dependent birds co-occur in oak woodlands where they can take advantage of woodland resources, such as food or nesting sites that may not be abundant in the vineyard matrix. Habitat filtering can act as a driver of change in communities ; considering that vineyards favor agricultural adapters , the relatively high abundance of these species within the agricultural matrix may increase the probability of percolation of these species into oak woodlands. Accounting for the influence of the vineyard matrix and other environmental site and landscape variables, our results reveal negative interactions between several agricultural adapters and a host of native woodland species – some of which are species of conservation concern. To the best of my knowledge this is the first report of these types of indirect interactions, in addition to predation and parasitism, related to birds near agroecosystems. Vineyards appear to influence bird communities at different scales, with increased vineyard influence in small fragments of oak woodland, which are more likely to be used by agricultural adapters and perhaps more vulnerable to indirect spillover effects that could devalue the remaining wild lands. The species identified by our study as agricultural adapters are among the most commonly recorded species in Californian vineyards suggesting that vineyards likely promote these species more broadly. My results suggest that vineyard expansion threatens bird diversity not only directly via habitat conversion, but also by promoting changes in community composition and species abundance that then have indirect, negative consequences for species in remaining natural habitat. Over time,plastic plant pot competitive interactions could change the community composition of oak woodlands, undermining their conservation value.

The combined impact of habitat conversion and reduced habitat quality is concerning for bird conservation. For instance, body condition and fitness are important for successful migration for Neotropical and altitudinal migratory species and are related to habitat quality. Abundance of some Neotropical migrant species is associated with large forest patches and negatively associated with agricultural/urban landscapes . Equally important are impacts to species with declining populations, in our context Spotted Towhee, Orange-crowned Warbler, and Bewick’s Wren or oak woodland specialist species as Nuttall’s Woodpeckers. The method I used does not explicitly model species unidirectional interactions, instead it directly estimates reciprocal interactions . I assume negative interactions to be negative effects of agricultural adapters on oak woodland birds. Several mechanisms documented in other studies may help interpret our results. Firstly, direct biotic interactions may be at play. For instance, starlings are extremely aggressive birds that displace other species from nest sites they want to use, which are mainly cavities , and are considered among the most spread invasive species worldwide that threaten species with extinction . Aggressive behavior by native birds towards European Starlings has also been observed during nesting . Similarly, nest usurpation and/or depredation by starlings has been recorded . Studies in fragmented forest reported a higher predation by birds or parasitism by Brown headed cowbirds on Neotropical migrants birds in small than in larger fragments . I suggest that European starlings can have similar effects on oak woodland birds, but with a different mechanism. There is at least three mechanism of species loss, high predation, low dispersal abilities, and detrimental conditions of the environment .Other evidence of these direct interactions include American crows, which aggressively chase away other birds and eating the eggs and nestlings of other birds , and Brown headed cowbirds, which are a brood parasite of several bird species . Secondly, indirect biotic interactions also be a role but are more difficult to observe and understand than direct interactions and could make community dynamics difficult to predict . In some cases, the mere presence of one bird can affect another’s feeding behavior or the inter specific interactions of the community .

This gap in behavioral knowledge for some of these species could be elucidated with future research. Thirdly, changes in bird community composition could affect the ecosystems functions and services through promote or attenuate complex interactions, that can cause cascade effects that change the ecosystem properties and affect biodiversity . Changes in composition of the landscape matrix, in our study vineyard extension, influence the bird community composition. Other studies previously documented the importance of landscape context and habitat quality for wildlife communities. Species benefitting from exotic plantations were promoted by mechanisms such as habitat filtering , temporal changes in the surrounding matrix causing shifts in species interactions , and new species introductions that change co-occurrence patterns in a community . Some of the relationships in our analysis appear to have a complex pattern of interactions with an unclear mechanism, for instance, the negative relationship of Orange-crowned Warbler with shrubs at the plot scale. Previous studies report varying bird-habitat relationships depending on the landscape context or interaction between species, such as in the case of a negative relationship between some Neotropical migrants and forest patch size . It is probable that for some species the proportion of oak woodland forest will be a better predictor than the proportion of vineyards, but these variables have a highly negative correlation at the landscape scale so are hard to address independently. Different landscape scales may more accurately predict the relationship between landscape variables and different species according to biological and ecological traits . Fragmentation and agricultural intensification have led to reductions in beta diversity and biological community homogenization in many temperate and tropical systems . These changes in native communities can have cascading effects on ecosystem function and resilience . Here, I show that vineyards can lead to regional biotic homogenization of neighboring natural areas through the expansion of the vineyard matrix, by increasing the abundance of birds adapted to vineyards, and by increasing competition between agricultural adapter and oak woodland birds. It is unlikely that all species will be equally affected , as habitat specialists are more likely to be threatened than generalists .

The underappreciated, indirect effects need to be considered as the agricultural footprint continues to expand across California’s coastal ranges and vineyards regions globally. Agriculture land conversion is recognized as one of the main threats to biodiversity due to habitat loss and fragmentation. In this study I found that the effects of agricultural land can spill over into natural areas. I see a clear relationship between the extent of agricultural land and bird species detection rates in adjacent natural areas. More surprisingly, I detected strong negative interactions through co-occurrence patterns between agriculturally adapted bird species and oak woodland associated birds. This research provides evidence that competition from species adapted to agricultural land use could be another driver of biotic homogenization in addition to habitat loss and fragmentation associated with habitat conversion. To the best of our knowledge, this is the first report of these types of indirect interactions beyond predation or parasitism between birds related to agroecosystems. Conserving biodiversity in agroecosystems is a global challenge . Agricultural impacts on biodiversity are widely recognized, and land use change and fragmentation are major drivers of species extinction and ecosystems degradation . Wine grapes are one of the most important crops in terms of land cover surface and economic relevance in Mediterranean type ecosystems worldwide,plastic planter pot which also overlapping with a designated biodiversity hotspot . Future projections for vineyards predict advancement to new areas, where climatic conditions will be favorable for wine grape production, in addition to demand from new markets, can increase vineyard expansion . Different strategies are proposed to increase biodiversity conservation in agricultural landscapes . Some studies indicate that biodiversity can best be conserved while increasing yield by dividing surface area into conservation zones and intensively managed agriculture . Others propose that increasing the quality of agroecosystem as habitat via agroecological management, enhances biodiversity. Still others propose an integration of both aforementioned strategies according the specific traits of the area of study/species . Finding new alternative approaches that enhance biodiversity within the agricultural matrix is part of the current policy efforts of “protect wildlife beyond the protected areas” of Chilean President M. Bachelet . The spatial configuration of the natural vegetation surrounding agriculture can significantly influence in the occurrence and abundance of bird species . It is recognized that increasing landscape complexity within agroecosystem can increase niche availability and thus host higher levels of biodiversity . An increasing number of studies indicate that remnant native vegetation can contribute to the persistence of birds. Birds can be favored by the presence of forest fragments , forest edges, and riparian vegetation in agroecosystems . Assessing the impacts of the native habitats on biodiversity in a dominant agroecosystem such as vineyards in central Chile is vital for conservation management within this biodiversity hotspot, and could provide management insight for other Mediterranean type ecosystems.

I used a natural experiment based on a gradient of landscape complexity within vineyards to investigate: whether bird trophic guilds change between continuous forests, fragments and vineyards, if changes in the presence and/or proportion of forest fragments influence birds communities, and if the proportion of continuous native vegetation influence birds communities. I hypothesized that vineyards promote some species, and that these functional guilds are different between vineyards and fragments within vineyards. My aim is to disentangle the changes in bird communities due to vineyards at different landscape scales. Conserving biodiversity in agricultural landscapes is an unresolved challenge, in particular in Mediterranean type ecosystems in where agricultural expansion and unique wildlife habitat overlap. I found that fragments of native vegetation within vineyards significantly retained bird communities in vineyard landscapes. Fragments significantly increased abundance and richness compared to the vineyards in which they are located, and also affected assemblages of endemics, insectivores, granivores and omnivores . An equally important finding is that the proportion of fragment native vegetation within a given landscape area affected similarly species than the mere presence of fragments. This highlights the important role of remnants of native vegetation within agroecosystem, particularly remnant size. Our evidence supports a land sharing approach, where more structurally diverse habitat within agroecosystems and habitat heterogeneity at landscape scale can enhance biodiversity in working landscapes . The results coincide with other studies finding that fragments, as well as hedge proportion and small woods within vineyards, support higher bird richness and abundance . In particular, the observed increase in insectivore birds due to large fragments is in agreement with other studies in vineyards . Insectivore species as Chilean swallow and House Wren, were more abundant in vineyards with fragments. Higher insectivore abundance could be associated to ecosystem services such as insect predation relevant to vineyard pest management . However, further research is required to better assess the impacts of biological control via bird predation of insects in Chilean agroecosystems. Fragments were particularly relevant for endemic species. Although fragments could be expected to favor mainly more mobile species such as Chilean Mockingbird and Dusky-tailed Canastero, this was observed also for the Dusky tapaculo, a rhinocryptid species which is less mobile . Other endemic rhinocryptids were excluded from the analysis due to their low rates of detectability within fragments and vineyards , indicating that these species are less likely to be found in anthropogenic environments. More focused research on these less mobile species would be needed to study their behavior. Forest patches were also relevant for Neotropical migratory species as White-crested Elaenia, showing that agroecosystems with large forest patches were indispensable for birds, and in particular for species that need connectivity at landscape scales . The results are consistent with existing literature demonstrating that vineyards support bird communities across the globe . Vineyards monocultures, without fragments of native vegetation favored a subset of species, all of them associated with open habitat and higher tolerance to anthropogenic environments.