| dc.description.abstracteng | Land-use change due to agricultural expansion is one of the greatest global threats to biodiversity and associated ecosystem functions and services. In the tropics in particular, conversion of forest and agroforests into monoculture agriculture such as oil palm is happening at an extremely rapid rate, leading to highly simplified landscapes. In Indonesia, currently the world’s leading palm oil producer, oil palm plantation crop cover was nine million hectares in 2010 with a projected increase to 18 million hectares by 2020. The predicted further large-scale expansion of oil palm plantations and other high-intensity agricultural systems has extensive implications for biodiversity loss. In agricultural production systems, loss of economically important functional groups can lead to decreases in essential ecosystem services such as pollination, biocontrol and soil turnover. Ants for example, dominate terrestrial biomass and are important for number of ecosystem services that are crucial in agriculture and have been shown to influence yield. In this thesis I aim to understand 1) how ant communities respond to land-use change, 2) if certain factors can alter their response, and 3) if changes in the ant community can alter their influence on biodiversity, ecosystem functions and yield.
In chapter 2 and 3 of this thesis I investigated land-use change effects on the taxonomic and functional diversity of ant communities in lowland forest, jungle rubber, monoculture rubber and oil palm plantation sites. In chapter 2 I focus on changes in species richness and composition and in chapter 3 changes in functional diversity and composition, including also birds and leaf-litter invertebrates in the analysis to broaden the study. The results show that ant species richness responded contrary to expected to land-use change, either increasing or not changing from forest to the agricultural systems studied dependent on the sampling methods used. However, species composition changed considerably and few ant species were shared among different land-use types. These results suggest land-use change would result in a net loss of ant species, even though ant species richness in plantations and forested habitats are similar. Furthermore, there was a linear relationship between species richness and functional diversity for ants, birds and leaf-litter invertebrates, indicating low redundancy in these systems. Finally, the functional composition of all three animal groups also changed with land-change. In particular, species from higher trophic guilds decreased from forest to oil palm. The observed decrease changes in species composition from rainforest to monoculture plantations, along with the tightly coupled decreases in functional diversity and low functional redundancy, could threaten long-term ecosystem stability through potential consequences for ecosystem processes
The results from chapter 2 and 3 provide strong evidence that overall impacts of conversion from natural ecosystems to land-use systems on ant communities are negative. However, there are factors which can influence the severity of these impacts. Therefore, in chapter 4 and 5 I investigate the response of ant communities and associated functions to landscape context and local management in oil palm plantations. In chapter 4 I examined the effects of location within the plantation (edge vs centre) and local characteristics (epiphyte cover, herbicide use, and local microclimate) on ant and other arthropod communities and litter decomposition in oil palm leaf axils. Arthropod abundance and taxa richness were higher at the edge of plantations than in the centre. Moreover, organic matter mass and height of the leaf axil were more important for arthropod communities and decomposition rate than epiphyte or ground vegetation cover. In chapter 5 I investigated the effect of surrounding landscape and distance from the edge on predation rates and predator occurrences in oil palm plantations. Predation rates were ~70% higher in non-oil palm habitat. This effect spilled over into the oil palm plantations, where predation increased by 55-100% 20 m from the edge of the plantation and by 40-55% 50 m from the edge when surrounded by a land-use other than oil palm. Taken together, chapters 4 and 5 show that oil palm plantations adjacent to different land-use systems have enhanced biodiversity and associated functioning, though this effect quickly declines with distance from the edge. Furthermore, local management of ground cover has mixed effects on arthropods and epiphyte management is unlikely to influence ant communities.
In chapter 6 and 7 I look further into the changes in ant communities shown in Chapter 1 and 2 by investigating the role of ant communities in shaping arthropod communities and associated ecosystem functions and services. In chapter 6 I studied the influence of ants on both above- and belowground invertebrate communities, soil and litter variables and decomposition across different land-use systems using ant suppression plots at forest, jungle rubber, rubber and oil palm sites. Belowground invertebrates, i.e. collembola, had 30% lower biomass in ant suppression plots. However, soil and litter variables and above-ground invertebrates were mostly unaffected. Rather, differences in local conditions between the four studies land-use systems were generally the most important predictors. Most notably, effects on decomposition rates were context-dependent, whereby ant suppression reduced decomposition in the forest sites only. Therefore, differences in ant communities between land-use systems alter their relationship to decomposition processes, however, the underlying drivers of these effects require further investigation. In chapter 7 I manipulated ant as well as flying vertebrate (birds and bats) access in only oil palm sites and measured effects on arthropod communities, related ecosystem functions (herbivory, predation, decomposition and pollination) and crop yield. Non-ant arthropod predator populations increased in response to reductions in ant and bird access, and the net effect of experimental manipulations on ecosystem functioning was minimal. Similarly, effects on yield were not significant. Chapter 7 shows that ecosystem functions and productivity in oil palm are, under current levels of pest pressure and pollinator populations, robust to large changes in the communities of major predator groups.
In conclusion, although forest conversion to oil palm and other agricultural systems in Indonesia has wide-ranging negative influences on biodiversity and function, there is the opportunity to enrich biodiversity in these systems. In oil palm plantations in particular this should be encouraged as changes in biodiversity do not compromise production. | de |