| dc.description.abstract | Strong scientific evidence supports that anthropogenic activities since industrialization have caused instability in earth’s climate, featured by increasing global surface temperature, increasing greenhouse gas concentration in the atmosphere. Climate change has then caused a series of changes in the earth’s ecosystems, which can have significant impacts on the biosphere and our human being. Although huge efforts have been put into the research in climate science since the past century, due to the complexity of the climate system and its broad and long-lasting influence, there are still countless question marks and uncertainties in our understanding of climate change and its influence on earth and human society. Microorganisms are among the tiniest groups of life, but play important roles in the cycling of carbon and other nutrient elements in the biosphere. However, their response and feedback to climate warming in different ecosystems is still difficult to predict, limited by the lack of mechanistic understanding of the complex microbial community, their functions, their interactions among themselves and under warming perturbation. With the fast advance of high-throughput metagenomic technologies and the development of environmental microbiology, deep and detailed characterization of microbial diversity and functions became available, which provided great chances in promoting our insights into the mechanisms by which microbial communities mediate the carbon balance in a warmer world. This dissertation applied several metagenomic technologies to probe the soil microbial community responses to warming and permafrost thaw based on field observations and experiments in two ecosystems, a permafrost underlain Alaska tundra, and a temperate tall grass prairie in Oklahoma. Microbial decomposition of soil carbon in high latitude tundra underlain with permafrost is one of the most important, but poorly understood, potential positive feedbacks of greenhouse gas emissions from terrestrial ecosystems into the atmosphere in a warmer world. On the other hand, temperate grassland provided a contrast to the cold weather and huge soil carbon storage in the tundra, allowing the comparison of different ecosystems in terms of their sensitivity and vulnerability to warming.
In the beginning of this work, we sought answers to the question that how microbial functional diversity was affected by regional warming induced long-term permafrost thaw. Soil columns were collected from a tundra site where three locations with different lengths of permafrost degradation history were on record. A functional gene array (i.e. GeoChip 4.2) was used to analyze the functional capacities of soil microbial communities in these samples. Compared with the minimally thawed site, the number of detected functional gene probes across the 15-65 cm depth profile at the moderately and extensively thawed sites decreased by 25 % and 5 %, while the community functional gene β-diversity increased by 34% and 45%, respectively, revealing decreased functional gene richness but increased community heterogeneity along the thaw progression. Particularly, the moderately thawed site contained microbial communities with the highest abundances of many genes involved in prokaryotic C degradation, ammonification, and nitrification processes, but lower abundances of fungal C decomposition and anaerobic-related genes. Significant correlations were observed between functional gene abundance and vascular plant primary productivity, suggesting that plant growth and species composition could be co-evolving traits together with microbial community composition. This study reveals the complex responses of microbial functional potentials to thaw related soil and plant changes, and provides information on potential microbially mediated biogeochemical cycles in tundra ecosystems.
Next, a field warming experiment was set up to increase the winter soil temperature in tundra by snow cover coupled with spring snow removal. Using integrated metagenomic technologies, we showed that the microbial functional community structure in the active layer of tundra soil was significantly altered after only 1.5 years of warming, a rapid response demonstrating the high sensitivity of this ecosystem to warming. The abundance of microbial functional genes involved in both aerobic and anaerobic C decomposition was also markedly increased by this short-term warming. Consistent with this, ecosystem respiration (Reco) increased up to 38%. In addition, warming enhanced genes involved in nutrient cycling, which likely contributed to an observed increase (30%) in gross primary productivity (GPP). However, the GPP increase did not offset the extra Reco, resulting in significantly more net C loss in warmed plots compared to control plots. Altogether, our results demonstrate the vulnerability of active layer soil C in this permafrost-based tundra ecosystem to climate warming and the importance of microbial communities in mediating such vulnerability.
Then, we conducted quantitative comparisons of the responses of soil microbial communities to warming at tundra and the prairie ecosystems. Climate warming has been differentially increasing the global surface temperature, with the greatest temperature elevation observed in the northern high-latitude regions. Although tundra and underlain permafrost in those areas were predicted vulnerable to climate warming, few quantitative comparisons were reported between tundra and other grassland ecosystems, especially of the composition and structure of soil microbial communities and their functional diversity. We compared the early responses of soil microbial composition and functional gene abundance to experimental warming between a tundra site and a temperate tall grass prairie using several metagenomic technologies, including functional gene microarray, amplicon sequencing, and metagenomic shotgun sequencing. Despite distinct species and functional gene pools in soils from the two ecosystems, genes involved in carbon and nitrogen cycling showed positive responses to warming at both sites, but with 36% more significantly responding genes and a greater magnitude of response for 10 genes at the tundra site. The functional gene compositions were correlated with temperature, moisture, ecosystem respiration and gross primary production at the tundra sites, but mostly with substrate related variables, plant biomass and nitrate concentration, at the prairie, implying different limiting factors in microbial growth and functions. These results revealed the higher sensitivity of tundra soil microbial communities to warming, compared with those from temperature prairie, and provided field evidence in supporting that northern high-latitude regions might be more vulnerable to climate warming.
At last, we extended our exploration of the warming influence on the microbial community to the interaction of microorganism in the community by constructing co-occurrence networks for a time series sample set from the temperate prairie. Although intensive reports have shown that warming can influence the soil microbial community composition and structure, little is clear about how the microbial interactions among themselves would be influenced. Here, soil microbial co-occurrence networks were constructed using 16S rRNA gene amplicon sequences extracted from monthly samples collected in a long-term field warming experiment on a Central Oklahoma grassland. We observed substantially larger and more connected networks for warmed communities compared with control, despite huge variation in network structures along season. The increase in network complexity under warming was concurrent to decreased phylogenetic diversity, reflecting environmental filtering and increased functional association in altered soil and vegetation conditions. A portion of identified keystone taxa, which play important roles in network topology, reoccur in different networks, representing a preserved prominent group in grassland soils across the season and under warming. The structure of microbial networks introduced a dimension beyond species abundance, which revealed more complicated responses of microbial communities to climate warming.
Overall, this work provided valuable field evidence on microbial community’s response to climate warming, revealed different sensitivities of these responses in tundra and prairie, and captured the seasonal dynamics of soil microbial interactions under warming influence. Many of these findings represent novel insights into our understanding of the microbial-mediated carbon cycle in a warmer world, from which new hypothesis could be formulated, tested, and generate knowledge essential for including the microbial contributions to earth system models, eventually a better prediction of the future climate. | en_US |
