Watershed Controls on Biogeochemical Processes in Aquatic Ecosystems

Thesis (Ph.D.)--University of Washington, 2014Freshwater ecosystems provide critical services to sustain human livelihoods, such as fisheries, agriculture and clean water, yet face increasing demands from rapidly expanding populations and globalizing societies. In addition, the threats to ecosystem services, such as climate and land use change, are increasingly large-scale and diffuse. The productivity and sustainability of freshwater ecosystem services rely on fundamental biogeochemical processes such as photosynthesis and nutrient cycling. Yet, biogeochemistry has traditionally been done at scales that do not match the spatial or temporal scale of modern issues like climate change. Therefore, it is increasingly clear that we need techniques that enable us to understand how the biogeochemical processes that support critical ecosystem services play out at scales that match these threats. In this dissertation, I address this gap by using novel approaches to evaluate aquatic carbon (C) and nitrogen (N) cycling at watershed scales. In chapter 2, I evaluate changes in N fixation and N cycling in small urban lakes in response to a gradient of human nutrient loading. I paired an isotope mass balance model with nutrient stoichiometry to estimate loads of nutrients from background, human and atmospheric sources; estimate N fixation at the ecosystem scale in response to eutrophication; and track the importance of N from various sources in supporting lake food webs. In chapter 3, I tie these changes in lake chemistry to shifts in bacterial community composition. Using a molecular approach, I examined how bacterial communities changed in response to eutrophication and whether the response to eutrophication differed among lake habitats. I found that surface and deep bacterial communities responded differently to eutrophication and were increasingly distinct as eutrophication progressed because of heterogeneity in resource distribution. My last two chapters focus on understanding how C cycling in boreal stream ecosystems will respond to warming temperatures associated with climate change. Using a mesocosm approach (chapter 4) and an ecosystem-scale Bayesian model (chapter 5), I examine how variation in watershed geomorphology influenced the temperature dependence of C metabolism in streams of the Wood River watershed in southwest Alaska. I found the response of stream respiration to temperature was not universal, but instead varied substantially among streams within a single watershed in southwest Alaska. Temperature sensitivity was hierarchically controlled by geomorphology; specifically, the sensitivity of stream respiration to increasing temperature was highest in streams draining flat watersheds. I found that this was linked to the quality and quantity of available carbon substrates. Together these results show that by using approaches that integrate over larger spatial and temporal scales, we can better understand how large-scale stressors like climate and land use change will influence aquatic ecosystems