9615563 Nadelhoffer Changes in terrestrial carbon cycles are often tightly constrained by interactions between C and other elements, especially N and P. The aim of this research is to improve the understanding of how carbon/nutrient interactions in soils might affect the responses of arctic tundra ecosystems to global environmental change. The research centers on the question, "What controls the amount of C lost from tundra soils per unit N made available for plant uptake?" This question is important because plant C gain in tundra ecosystems is often strongly N-limited, and virtually all of the N made available to vascular plants in tundra ecosystems comes from microbial mineralization of soil organic matter. Thus the overall C balance of the ecosystem is largely determined by the balance of plant C gains associated with N uptake versus C losses due to soil respiration. The design of the research is guided by a simple conceptual model, in which the large amount of soil organic matter in tundra ecosystems is viewed as being composed of three interacting organic matter pools with different turnover times and characteristic C:N ratios (microbes are a fourth pool that mediates turnover of the other three, but which is also much smaller than the other three). The overall hypothesis is that the balance of C lost: N made available in tundra soils is controlled by interactions between the chemical quality of soil organic matter (including fresh litter) and environmental factors that determine the relative rates of decomposition and interconversion of the various organic matter pools. Additional hypotheses deal with the individual effects of (1) soil organic matter quality, (2) drainage and associated aerobic/anaerobic conditions, and (3) temperature. These hypotheses will be tested by measuring C losses and N made available in soils from contrasting tundra ecosystem types that are already known to differ in initial organic matter quality. A field experiment will compare wet, moist, and dry heath tundras, and two laboratory experiments will compare wet and moist tundras. One laboratory experiment will be on whole plant-soil monoliths transported from the field, and the other laboratory experiment will be on soils only; treatments will include manipulations of temperature, drainage, and N and P availability. Isotopic tracers and natural abundances of C & N isotopes in various soil organic matter fractions will be used to help estimate interactions and turnover rates of the major organic matter pools. Because the experiments are long-term (the laboratory experiment will simulate 4 full growing seasons over the 3 years of the research, and the field measurements will be made in the 10th through the 12th years of an experiment begun in 1988), it may be possible to document changes in characteristics of organic matter pools themselves, as well as the fluxes into and out of them. In addition to direct tests of the hypotheses with data, a simulation model (MBL-GEM) will be used as a synthesis tool and for longer-term prediction and integration with past research on vegetation C-N interactions.
