TY - THES T1 - Trophic relationships in soil communities how abiotic stress affects biotic interactions in the McMurdo Dry Valleys, Antarctica T2 - Ecology Y1 - 2018 A1 - Shaw, E. Ashley A1 - Diana H. Wall AB -

Understanding of the distribution and complexity of soil food webs and their role in ecosystem processes is limited. This is partially due to the difficulty studying the enormous diversity of species in belowground ecosystems and identifying the many roles of this diversity in ecosystem processes. Despite this, there is strong interest in understanding how the soil food web contributes to ecosystem processes such as decomposition, nutrient cycling, and carbon cycling. Yet, before we can fully understand how soil food webs are linked to ecosystem processes, more information is needed on their complex trophic interactions and how soil food webs respond to changing environmental variables. The McMurdo Dry Valleys in Antarctica provide an excellent opportunity to study soil communities and their trophic interactions because of soil food web simplicity and limited ecological interactions that are not easily distinguished in more diverse systems. However, it is unknown whether trophic interactions actually play a role in structuring soil communities in this ecosystem and whether these interactions are affected by environmental factors. The aim of this dissertation is to disentangle those questions.

In the first chapter of this dissertation, I give the background for my research. I introduce the challenges for studying soil biodiversity and its food web structure. Next, I discuss the usefulness of the McMurdo Dry Valleys as a simple, model system for researching trophic interactions in soil. The details of the current understanding of the McMurdo Dry Valley soil food web are demonstrated and I have highlighted gaps in this knowledge. In the second chapter of this dissertation, I address the question: What trophic interactions are present in the McMurdo Dry Valley soils? Here, I sought to elucidate the soil food web structure using stable isotopes (particularly 15N) and I present isotopic signatures for soil fauna taxa for one location in Taylor Valley, Antarctica. The natural abundance of 13C and 15N were measured for soil fauna and microbial mats sampled in both wet and dry soils near Von Guerard stream. This study revealed that three trophic levels were present in wet soils at this location and two trophic levels were present in dry soil. This is the first isotopic confirmation of Eudorylaimus antarcticus (Nematoda) as an omnivore-predator (in wet soil habitat), and challenges long-held assumptions of trophic simplicity of the McMurdo Dry Valley region.

Building on the findings of Chapter 2, Chapter 3 seeks to expand the understanding of dry valley food webs and the role of trophic interactions in structuring communities under environmental change. Specifically, I address the question: How do environmental variables (soil salinity and moisture) affect dry valley soil taxa and their trophic interactions? I show the results of a laboratory microcosm experiment on how elevated salinity and moisture affect four soil communities. Using soil collected from Taylor Valley, Antarctica, bacteria, bacteria with Scottnema lindsayae, bacteria with E. antarcticus, and bacteria with both S. lindsayae and E. antarcticus were established in microcosms under control or high salinity treatments and control or high moisture treatments (full factorial design). The results of this experiment showed that S. lindsayae has top down effects on bacterial abundance under control salinity but these top down effects were alleviated under high salinity. This study is the first to empirically show that biological interactions structure dry valley soil communities.

The fourth chapter follows the conclusions of Chapters 2 and 3, and seeks to determine food web structure and trophic interactions at the landscape scale in the McMurdo Dry Valleys. I sampled soil from 160 sites across 8 valleys ranging from the coast to high elevation near the polar plateau to address the question: How does the soil food web and its organic carbon sources vary across the McMurdo Dry Valley landscape with distance from coast and elevation? These valleys represent a temperature and moisture gradient, which affects ecosystem primary productivity. This study revealed that food web structure varies by habitat – the most diverse and complex trophic interactions exist in wet habitat near the coast where resources are more abundant. However, in dry habitat, where organic carbon resources are scarce, up to two trophic levels exist. These results build off of Chapter 2, and show that E. antarcticus can occupy either a predator trophic position when resources are high (wet soil) or a primary consumer position when resources are low (dry soil). Since climate-driven increases in hydrological connectivity are expected to alter soil moisture and resources, the distribution and abundance of soil biodiversity and their biotic interactions in formerly dry soil habitats may ultimately shift.

In Chapter 5, I asked if the lessons learned about soil food webs in the McMurdo Dry Valleys apply to a more complex ecosystem? In this study, I used soil nematode communities from the Loch Vale Watershed (Rocky Mountain National Park, Colorado) to test whether long-term nitrogen addition affected soil food web structure and function. Results from this study indicated that a faster-cycling, bacterial food web was prevalent in N-addition plots, as evidenced by abundance of r-selected bacterivore nematodes. Previously, lower bacterial abundance and soil carbon were found in the N-addition plots (compared to control) and the results presented in this dissertation suggest that these changes are likely trophic. Along with Chapter 3, the evidence that I present here support the hypothesis for top-down effects of microbivore nematodes on bacteria, which is consistent in subalpine and Antarctic soils.

In summary, through both field and laboratory experiments, my PhD project has: 1) defined the soil food web structure of the McMurdo Dry Valleys using stable isotopes; 2) revealed how top down interactions affect bacteria populations and how elevated stress (e.g. soil salinity) relieves the top down pressure; 3) showed how the soil food web structure varies across the landscape of the McMurdo Dry Valleys, Antarctica as related to soil C sources; and 4) shown how nitrogen addition affects soil food web dynamics in Colorado sub-alpine soil nematode community (Loch Vale Watershed, LVWS, Rocky Mountain National Park). These results have informed our understanding of soil communities and their trophic relationships in polar and subalpine ecosystems.

JF - Ecology PB - Colorado State University VL - Ph.D. UR - http://search.proquest.com/openview/0d494a3f115b75da1c7a2464e341808f/1?pq-origsite=gscholar&cbl=18750&diss=y ER -