The responses of soil microbes to climatic and anthropological factors in the Tibetan grasslands

Abstract

It is widely known that soil microbes play an important role in biogeochemical cycling, affecting plant growth, and creating soil structure. Faced with the widespread global changes, unravelling the response of soil microbes is essential for understanding soil feedbacks to climate changes. Therefore, there is an increasing interest in studying soil microbial diversity patterns from local to global spatial scales and in investigating changes in soil microbial community diversity and community structure under global changes. However, how soil microbial communities would be altered by natural environmental changes and/or by human induced environmental changes are still not well understood at the regional scale. The Tibetan plateau is known as ‘the third pole’ because of its high elevation of over 4000 m above sea level. Ecosystems in the Tibetan plateau, dominated by alpine grasslands, are fragile and vulnerable to global changes. The increase rate of air temperature is up to three times the global average level. However, we still lack knowledge on responses of soil microbial community to climate gradient and human-induced environmental changes, e.g. climate warming and grazing. This thesis investigated changes in bacterial and fungal community diversity, composition, and co-occurrence along a hydrothermal gradient in the Tibetan plateau grasslands, and the interactive effects of climate changes (warming) and human activities (grazing) on soil microbial functional communities. Specifically, this thesis includes four experiments:\nExperiment 1 studied the changes in soil bacterial community along an environmental gradient in the Tibetan plateau. The hypothesis was that alpha diversity of soil bacterial community would increase with higher precipitation and air temperature. We therefore selected sampling sites distributed in main grassland types in the Tibetan plateau. Microbial DNA extracted from soil samples were subjected to the next-generation sequencing to characterize soil bacterial diversity and community. Meanwhile, information of environmental factors 44 was also collected to clarify the driving forces of changes in soil bacterial communities. These factors included mean manual precipitation (MAP), mean manual temperature (MAT), soil moisture (SM), elevation, soil total organic carbon (SOC), total nitrogen (TN), soil carbon: nitrogen ratio (C/N), available phosphorus (AP), soil pH, plant richness, plant aboveground biomass (plant BiomassA), plant belowground biomass (plant BiomassB), NH4+-N, NO3--N, dissolved organic carbon (DOC), soil dissolved organic nitrogen (DON), soil microbial biomass carbon (MBC) and nitrogen (MBN). We used statistical methods including simple correlation, multivariable analysis and structural equation modelling (SEM) to elucidate spatial pattern and the relative importance of biotic and abiotic factors in structuring soil bacterial communities. In the SEM, climate factors were expected to affect soil microbial communities through their influences on plant and soil properties.\nMain findings included: in terms of overall bacterial diversity, the swamp meadow with better water conditions had significantly higher diversity than other habitat types, while the alpine desert had the lowest diversity. The overall alpha diversity was significantly correlated with factors that can indicate soil nutrient status including total soil organic carbon, total nitrogen, plant aboveground biomass, nitrate nitrogen (N) and other factors. The effect of environmental factors on the overall bacterial community structure variation was greater than that of spatial factors. The effects of mean annual precipitation (MAP) on soil bacterial alpha diversity were mostly indirect through affecting soil dissolved organic carbon (DOC) and plant richness based on the SEM. Among the environmental factors, the most influential factor for the differences of bacterial community structure was MAP. Unlike the effects on bacterial alpha diversity, MAP had a strong direct effect on soil bacterial community structure. Other factors such as soil DOC and soil pH affected soil bacterial community structure directly without through mediating intermediate variables. By contrast, mean annual 68 temperature (MAT) was not significantly related to soil bacterial diversity or community composition.\nExperiment 2 studied soil fungal community of the same soil samples as described in Experiment 1. We hypothesized that there would be a close association between soil fungal community diversity/composition and plant community diversity or composition. The main findings were: Ascomycetes were the most abundant phylum in all samples (84.56%), followed by Basidiomycetes and the Zygomycota. Significantly positive correlation between the relative abundance of Ascomycota and precipitation was found, while the relative abundance of Glomeromycota was not significantly correlated with precipitation. Based on the best-fitting regression model, the most important predictors of fungal species richness were pH and plant species richness. MAP also had substantial effects on soil fungal richness, mostly through its effects on soil pH and plant richness. Environmental factors and geographic distance can independently explain partial changes of fungal community structure with a higher effect by environmental factors. Compared to strong direct effects of soil pH, MAP affected soil fungal community composition by altering soil pH and plant community structure based on structural equation modeling (SEM).\nExperiment 3 studied the microbial co-occurrence network along the transect as described in Experiment 1 and 2 by integrating soil bacterial and fungal community data. In this study, Spearman correlation-based network was constructed, and a set of network topological properties were calculated. In addition, the impact factors of network properties were also investigated.\nThe main findings were: The microbial network size of alpine meadows was greater than that of the alpine steppes. However, the modularity of alpine steppe was higher compared to alpine meadow. In addition, networks of alpine steppe had a larger average path length. Based on these differences in network topological features, soil microbial communities of alpine steppe were considered more stable under environmental interferences. 93 Fungal networks were found to have larger modularity, but smaller inter-nodes connectivity compared with bacterial networks. Although betweenness centrality of bacterial nodes was higher in the alpine meadows, degree centrality was higher in the alpine steppes. The regional meta-network (integrating bacterial and fungal taxa) structure was mostly related to MAP. However, the network topological features of alpine meadows and alpine steppe were driven by different factors. MAP and soil moisture were the significant impact factors for alpine steppe network but not for alpine meadow. The network of alpine meadows was strongly associated with plant factors including biomass and diversity. Besides, fungal networks were not associated with plant community factors, but related to climate factors and soil properties, suggesting that drivers of soil fungal network were different from that of fungal community diversity and structure.\nIn Experiment 4, we investigated the effects of warming, grazing and their interaction in a factorial warming (+1.2-1.7 oC) and grazing (moderate intensity with ca. 50% vegetation consumption) experiment in a Tibetan alpine meadow on soil microbial communities by studying functional genes involved in soil carbon and nitrogen cycles. We hypothesize that warming would interact antagonistically with grazing to affect soil microbial functional communities. In this study, soil microbial communities were analysed by Geochip and environmental parameters including temperature, soil properties and plant communities were also collected.\nThe main findings include: microbial functional gene structure and abundances were largely affected by the interactive effect of grazing and warming, rather than the main effect of warming or grazing. Compared to the control, grazing alone significantly increased the functional gene alpha diversity, changed the overall functional community structure, and increased the abundances of C fixation, C degradation, N mineralization and denitrification genes, likely due to the stimulating impact of urine and 118 dung deposition. Warming alone did not change these microbial properties, possibly related to the unchanged soil nutrient status. Despite an increase in soil NO3- concentrations and the deposition of urine and dung, the combined warming and grazing treatment did not change functional gene alpha diversity, community structure, or C/N cycling gene abundances. Our study revealed antagonistic interactions between warming and grazing on microbial functional gene structure and abundances, which remained stable under moderate intensity of grazing in a future warming scenario in the Tibetan alpine meadow.\nIn sum, this thesis indicated that diversity, composition, and inter-taxa association of soil microbial communities are sensitive to environmental changes in the Tibetan plateau grassland. Climate changes and human activities affected soil microbial communities through direct and indirect pathways. More importantly, the evaluation of microbial-mediated processes in the Tibetan Plateau grasslands should take interactions between climate changes and anthropogenic activities into account.