Seasonal patterns of nitrogen fixation by Alnus tenuifolia within successional floodplain forests of the Bonaza Creek Experimantal Forest: 1992

Methods Nitrogenase activity was measured by acetylene reduction (Stewart et al. 1967) and denitrification activity was measured by the acetylene inhibition technique (Tiedje 1982). Both processes were measured simultaneously on the same alder root/nodule sample and on bulk soil samples that served as controls. We sampled all sites in the 1992 growing season; during 7-9 July, 10-14 August, and 6-11 September. Segments of alder roots and adhering rhizosphere soil, 3-7 cm in length, with obvious nodules were cut from live alder plants and immediately placed in 237 ml mason jars. Jars were sealed with a metal screw band and a gas tight lid fitted with a rubber septa. Acetylene, generated by the addition of water to CaC2, was added to sampling vessels at approximately 10% of the head space volume using a 30 ml syringe with subsequent pumping to thoroughly mix the total head space volume of the jar. Headspace volume was determined using a pressure transducer (Parkin et al. 1984). We used similar short-term incubation periods as thosed used by Silvester and Winship (1990), who tested these methods on individual physiological lab studies; ours take these methods to in situ (as much as is possible) ecosystem testing. This was done in an effort to minimize the numerous errors associated with the acetylene reduction assay (Minchin et al. 1983; Silvester and Winship 1990). At 30, 90, and 150 seconds, 10 ml of atmosphere was exchanged with a 10 ml gas sample for measurement of nitrogenase activity. Gas samples were stored in 2.5 ml glass vials which had been previously sealed with a butyl septa and flushed with 7 ml of sample. Three jars per replicate site were incubated at each sampling. Bulk soil, with no apparent roots included, approximately 40 g wet weight, was also sampled for both nitrogen fixation and denitrification and used as a control sample. Once nitrogenase activity samples were taken, the jars were returned to the laboratory, maintained at at room temperature, and sampled every 6 hr until N2O could be detected. This was done for up to 24 hr at which time if no N2O was detectable, incubations were terminated. Because the sampling jars were at ambient atmospheric pressure, a 60 ml air sample was added and mixed thoroughly with the vessel gas with a subsequent sampling of 60 ml. Each sampling syringe was fitted with a stopcock valve and gases were stored in the syringe until N2O analysis took place. All gas analyses were completed within 72 hr of sampling. After the final gas sampling, roots and nodules were washed free of soil, air dried at 40o C and weighed for total biomass. Nitrogenase activity was estimated by regressing the change in ethylene concentration against time, and expressed per unit nodule dry weight. Ethylene was measured with a Shimadzu H2 flame ionization detector with He as the carrier gas and a Poropak N column at 40o C. Denitrification activity was calculated as the accumulated N2O over a 24 hr incubation, and expressed per total unit root (root + nodule) dry weight. N2O was measured with a Shimadzu 63Ni electron capture detector at 345 C with a Poropak Q column (1.8 m, 80/100 mesh) at 65o C with argon/methane (95/5) as the carrier gas. Distribution and abundance of A. tenuifolia at each of the sites were measured on 4, 0.36 ha plots at each of the sites as described by Viereck et al. (1993a). Hourly readings of air and soil temperature, photosynthetically active radiation, and bimonthly measurements of soil moisture by time domain reflectometry were measured at each site as part of the Bonanza Creek LTER climate monitoring program (Viereck et al. 1993b), and used as independent variables in equations predicting seasonal and landscape patterns of acetylene reduction and denitrification activity.