Understanding effects of anthropogenic activities on element cycling in temperate forest watersheds.

摘要

Following an introductory Chapter 1, in Chapter 2 we present results from repeated synoptic surveys of streamwater chemistry for ∼30 watersheds spanning one of the largest nitrogen (N) deposition gradients in North America, located within the Great Smoky Mountains National Park. We primarily focus on patterns in dissolved organic matter (DOM) concentrations and composition across the N gradient, with particular attention given to dissolved organic nitrogen (DON). DON dominates the global flux of N between terrestrial and aquatic systems, yet we have little understanding of how this prevailing N form responds to human N pollution. We found that DON concentrations often declined significantly with increasing catchment N loading and, through laboratory bioavailability assays, found that when N limitation is alleviated increased microbial demand for labile carbon (C) may drive this pattern. We use these findings to propose a new hypothesis for the potential responses of DON to anthropogenic N pollution, which states that DON concentrations in forested watersheds may systematically decline with increasing N loading due to increased heterotrophic demand for labile reduced C.;Chapter 3 is an extension of Chapter 2, in which we attempt to assess the role of DON as either a C or N source within an entire stream reach through a series of independent manipulations of labile C and inorganic N availabilities. In the second order reach of Walker Branch, a well-studied stream in eastern Tennessee, we performed a series of progressive (i.e., sequentially increasing concentrations), kinetic (i.e., very short duration), enrichments of acetate and nitrate on two successive days during April of 2009 before the tree canopy emerged and when in-stream algal production was high. In this system and on these short timescales, we were unable to elicit the same responses observed at sites across the chronic N deposition gradient in Chapter 2. We did, however, observe that DOM processing and composition was significantly altered. Using fluorescence characterization of DOM, we found that adding acetate displaced heterotrophic demand for terrestrially derived DOM. Conversely, nitrate additions stimulated production of highly bioavailable autochthonous DOM within the stream channel, which resulted in an indirect displacement of demand for terrestrially derived DOM. Understanding DOM dynamics in streams has long been a priority for stream ecologists because it represents an important energy and nutrient source fueling stream metabolism. Our results not only provide new insight into the processes controlling DOM concentrations and composition in Walker Branch, but also demonstrate the potential of this method for future investigations of DOM in stream ecosystems.;In Chapter 4 we assess the role of climate change on long-term (1989--2009) streamwater concentrations and fluxes from the West Fork of the Walker Branch watershed. At this site, mean annual air temperatures have increased by ∼2°C, while mean annual precipitation and runoff have declined by ∼20% and >40%, respectively, since 1989. We use weekly streamwater samples to assess trends in concentrations and fluxes for 9 different solutes over this period and, using wet deposition data, also evaluate changes in approximate watershed input-output budgets. The observed change in runoff was accompanied by a change in the proportional contributions of different soil flowpaths to streamflow generation through time, with deep groundwater playing an increasingly important role in recent years. Solutes that increase in concentration deeper in the soil-saprolite profile exhibited significant increases in streamwater concentrations through time, while solutes with higher concentrations in soil solution in the upper profile decreased in concentration. Solutes that exhibit much less variation across soil flowpaths typically display large seasonal patterns in streamwater concentrations that are driven by in-stream biological uptake. However, most nutrient solutes exhibited little or no trend in concentrations through time, indicating that the biological controls on these solutes have remained relatively unaltered by the observed changes in climate over the 20-year period. On shorter timescales, changes in the frequency or severity of multi-year droughts, as well as changes in the frequency or intensity of storms that disrupt in-stream uptake, can have large impacts on watershed input-output budgets of nutrient solutes even if the effects are not linear through time. Our results demonstrate the important role that changing climates can have on watershed element cycles, illustrating that climate effects can manifest through either changes in hydrologic regime or through changing biogeochemical process rates. (Abstract shortened by UMI.)