Soil CO2 and N2O flux dynamics in a nitrogen-fertilized Pacific Northwest Douglas-fir stand

This study investigated how nitrogen (N) fertilization with 200 kg urea N ha− 1 of an intermediate-aged Pacific Northwest Douglas-fir (Pseudotsuga menziesii) stand influenced forest-floor CO2 and N2O fluxes over a period of two years. Forest-floor CO2 efflux (Rs) measurements were made continuously with an automated non-steady-state chamber system as well as at 2–4 weekly intervals with manual non-steady-state chambers. Soil N2O flux was measured using the static chamber technique. Soil CO2 efflux measurements were made (a) in a completely-randomized design field experiment with four replications of fertilization and trenching treatments, (b) in soil columns in the laboratory, and (c) as a part of long-term flux monitoring in the field, using an automated chamber, for four years before fertilization and two years post-fertilization. Another field experiment compared the effect of regular urea and a slow-release-urea (Environmentally Smart N (ESN)) on forest-floor CO2 and N2O fluxes. We also studied the effects of N fertilization on decomposition of a standard substrate in the field. Our field measurements showed that N fertilization resulted in a significant short-term (over 3–4 months) increase in Rs due to an increase in autotrophic (or rhizospheric) soil respiration after which a small decrease in heterotrophic soil respiration (Rh) was observed. However, our laboratory measurements showed that N application had no effect on Rh in mineral soil, but a small but consistent increase for 10 weeks when the mineral soil had an LFH (litter-fibric-humus) layer on the surface. Fertilization resulted in significant N2O emissions in the first year with its peak rate almost coinciding with peak CO2 emissions. In the second year, however, we found no N2O emissions, but rather a small uptake in all treatments. The increase in Rs with N addition generally continued until N2O emissions began to decline, suggesting that most of the applied urea-N was rendered unavailable within the first 4 months. The results further suggest the lack of bioavailable C and hence insignificant microbial immobilization of applied N, because otherwise the latter would have resulted in an appreciable increase in Rh.