|dc.description.abstracteng||Tropical low-land rainforests are one of the most diverse ecosystems in the world and provide valuable ecosystem services such as climate change mitigation. They are immensely threatened by expanding human land-use. Especially in South-East Asia, deforestation and replacement with cash crop monoculture plantations such as rubber (Hevea brasiliensis) and oil palm (Elaeis guineensis) have led to drastic losses in biodiversity and to ecosystem degradation. Recently, the research focus has increasingly extended to belowground demonstrating strong structuring effects of human land-use on soil microbial communities. Fungi fulfill various ecological functions and their interaction with plants include efficient degradation of dead plant material (saprotrophs), mutualistic mycorrhizal interactions with roots, essential for the nutrient uptake in a majority of land plants, and structuring effects on plant communities (pathogens). Thereby, fungi are often tightly associated with the plant community as a key group of organisms facilitating the flow of nutrients between the below- and aboveground biome. Conversion of tropical lowland rainforests plantations leads to drastic changes in fungal community structures. The magnitude of structuring effects by changes in root or soil properties on the composition of trophic groups (mycorrhiza, saprotrophs and pathogens) remains unknown. The present thesis, conducted on Sumatra (Indonesia), analyses the structuring effects of human land-use in tropical ecosystems on this important group of microorganisms using next generation sequencing methods and root and soil properties. The work is structured into three major research chapters.
In the first research chapter, I analyzed the effect of land-use intensity on root associated arbuscular mycorrhizal fungi (AM). Anthropogenic land-use severely affects the AM communities in grasslands but tropical forest transformation systems have rarely been studied. I hypothesized that increased land-use intensities negatively affect AM abundance and diversity because of impaired plant fungus interactions at the roots. I further hypothesized that increases in land-use intensity drive the composition of the AM community, causing decline in naturally occurring AM fungi. A land-use intensity index (LUI) based on yield, chemical input and plant richness across four major land-use systems (forest, jungle rubber, rubber and oil palm plantations) was developed and the effect of LUI on AM molecular richness and abundance as well as AM spore abundance and root colonization was tested. Indicator species analysis was used to investigate significant associations between AM species and land-use types. LUI structured the root associated AM community and negatively affected AM diversity and abundance but positively affected AM spore abundance in soil. Distinct land-use types harbored distinct AM communities; however, forest harbored a higher number of indicator species. In conclusion, land-use intensity strongly altered AM communities across land-use systems reducing specialized OTUs. Extensive management practices may help sustain a diverse and abundant AM community.
Local soil and root associated fungal communities often differ considerably. Likely, this is caused by varying magnitude of structuring effects by the plant root community (biotic environmental filter) and edaphic conditions (abiotic environmental filter). However, few studies analyzed the effects of those environmental drivers on root versus soil associated fungal communities across different land-use types. In chapter 2, I tested the hypothesis that root associated fungi respond to changes in root properties more strongly than to changes in soil properties, due to their strong dependence on the root community. In turn, the soil fungal community provides a species pool from which the root community is recruited and this pool is structured mainly by changes in soil properties and stochastic fluctuations. Shifts of different ecological groups of soil and root inhabiting fungi in response to spatial distance as well as changes in soil and root chemistry across different land-use systems (as above, including riparian sites) were investigated. Overall, environmental filters had a stronger effect on the fungal community composition than geographic distance. Unexpectedly, high turnover and low nestedness between local root and soil communities was found. Additionally to a strong structuring effect of soil pH, root chemistry, especially root C/N strongly affected the composition of the root-associated fungal assemblages, while root vitality also affected shifts in soil-residing fungal communities. Root and soil chemistry changes drove divergent turnover of different functional groups (saprotroph, mycorrhiza and plant pathoges) in soil and roots. An important novel result was that assemblages root associated fungal communities were promoted by changes of root chemistry largely independent of the surrounding soil community. Therefore, recovering chemical root traits in intensively managed systems may stabilize the fungal communities against human land-use.
The results of the previous chapters raised the question, whether enrichment of oil palm plantations with other tree species can help to reverse the strong structuring effects of human land-use and partly recover the mycorrhizal community. To address this question, I analyzed the effect of tree species enrichment islands in an intensively managed oil palm plantation on the soil fungal community composition. Islands of native tree species (Parkia speciosa, Archidendron pauciflorum, Durio zibethinus, Peronema canescens, Shorea leprosula, Dyera polyphylla) were planted in an oil palm monoculture and further management was stopped within the islands to allow for natural undergrowth succession. After three years of enrichment cultivation, I tested the hypothesis that tree enrichment alters the taxonomic and functional soil fungal community composition in comparison with that in the soil of intensely managed oil palm plantations. However, no significant effects of tree species richness, or presence of individual tree species on the fungal community composition were found. A small proportion of community variation (< 10 %) was explained by soil abiotic conditions (N, C/N and P) and the majority of variation remained unexplained. These results suggest that abiotic filters as the result of intensively managed land-use constitute a legacy to fungal communities, overruling structuring effects of the vegetation on soil fungal communities within the first years after stopping management.
This thesis demonstrated a severe structuring impact of anthropogenic land-use on the fungal community structures. Soil abiotic properties were a main driver of fungal community composition in roots and soil. For the first time, changes in root chemical traits were linked to changes in the root and soil fungal communities. The results of this thesis underpin that the observed community shifts may result in loss of ecosystem services such as tree nutrient provision and tree health because of impaired AM root colonization. Links between shifts in the fungal community and plant root vitality suggest negative plant soil feedbacks driven by fungal community shifts. Strong bottom-up regulatory effects by root chemical traits especially on the root associated fungal community was demonstrated. However, no structuring effects of three years of plant succession on soil fungal communities in a biodiversity enriched oil palm plantations was found. Time series are required to investigate long term structuring effects of plant top-down regulation of soil fungal communities and the spatial scale at which root traits can affect local soil fungal communities. In summary, this thesis provides valuable new insights in the fungal community assembly processes under human land-use and highlights important areas of future research.||de