C3 - Insect diversity and trophic interactions in relation to agroforestry management and ENSO droughts

Abstract
Agricultural intensification and ENSO droughts are major drivers of global change. Here, we focus on little known effects of local agroforestry management (weeding, fertilization), the surrounding landscape (distance to nearest forest) and reduced rainfall (in agroforestry vs. forests) on insect diversity and ecosystem services such as pollination, biological control (herbivory and predation) as well as litter decomposition. A series of experiments address multitrophic interactions including cascades from plants (C2), mycorrhiza (C5), endophytes (C6) on community dynamics of functionally important below- and aboveground arthropods.

Summary
Agricultural intensification and ENSO droughts are major drivers of global environmental change and biodiversity losses. Little is known how these changes affect biodiversity and ecosystem functioning as well as biodiversity-ecosystem service management.

Within the second STORMA focus, we address the effects of local agroforestry management (weeding, fertilization) and the composition of the surrounding landscape (distance to nearest forest) on insect diversity and ecosystem services such as pollination, herbivory, predation (biological control) and litter decomposition. In a full factorial design with these three treatments (n= 64 plots), we analyze the responses of the insect communities in the cacao canopy (in close collaboration with ant and beetle taxonomists). Further we compare trees with ants and with experimentally excluded ant access.The ambiguous role of ants as vectors of cacao pathogens and biocontrol agents of insect pests will be studied using small-scale bioassays. In collaboration with C5 (mycorrhiza)
and C6 (endophytes), we focus on interactions of belowground (mycorrhiza, decomposition) and aboveground processes (endophyte infection, herbivory), including changes in leaf palatability, leaf chemicals and the role of N fertilization (with the help of 15N-labelled substrates).

Pollinator limitation appears to be a key factor in cacao productivity, so the relative importance of pollinator availability, diversification of deposited pollen, nutrient stress, and consequences for pod pests will be quantified. Video surveillance may allow us to directly detect the performance of cacao flower-midge pollinator and herbivore-predator interactions. The exposure of standardized traps for reproducing bees (as pollinators) and wasps (as predators) and their natural enemies will provide further data on alpha and beta diversity as well as multitrophic interactions in agroforestry. Earthworms, which are important ecosystem engineers, will be surveyed on all agroforestry and forest plots to allow an upscaling model for the study region.

Within the third STORMA focus, drought effects on tree growth should have cascading effects on leaf herbivores and their potential enemies. In both forest (natural trees) and agroforest (cacao trees), herbivory on drought-stressed and non-stressed trees will be compared and related to the results of the other plant and soil ecology groups. Soil moisture is known to affect decomposition processes, and litter-bag studies will reveal effects on decomposition processes contingent on partial exclusion of soil fauna.

Our Indonesian counterparts will address dung beetle-dung and parasitoid-butterfly interactions. In both STORMA foci, the role of beetle diversity for dung decomposition will be quantified, combining field with laboratory experiments. Further, parasitism of a lepidopteran cacao pest will be related to agroforestry management.

This project is closely linked to the socio-economic and ecological subprojects contributing to the second focus (agroforestry management) and third focus (ENSO drought effects) and will provide data for an upscaling process to include the landscape context (D6) and large-scale land-use scenarios (D5) as planned within the first STORMA focus (integrated modelling of land use).