Summary
Increasing anthropogenic pollution from urban centers and fossil fuel combustion can impact the carbon and nitrogen cycles in forests. To assess the impact of twentieth century anthropogenic pollution on forested system carbon and nitrogen cycles, variations in the carbon and nitrogen isotopic compositions of tree-rings were measured. Individual annual growth rings in trees from six sites across Ontario and one in New Brunswick, Canada were used to develop site chronologies of tree-ring δ15N and δ13C values. Tree-ring δ15N values were ∼0.5‰ higher and correlated with contemporaneous foliar samples from the same tree, but not with δ15N values of soil samples. Temporal trends in carbon and nitrogen isotopic compositions of these tree-rings are consistent with increasing anthropogenic influence on both the carbon and nitrogen cycles since 1945. Tree-ring δ13C values and δ15N values are correlated at both remote and urban-proximal sites, with δ15N values decreasing since 1945 and converging on 1‰ at urban-proximal sites and decreasing but not converging on a single δ15N value in remote sites. These results indicate that temporal trends in tree-ring nitrogen and carbon isotopic compositions record the regional extent of pollution.
Methodology
Coring. At each site, cores were extracted from dominant mature trees using a 5 mm diameter increment corer. At breast height, cores were taken at 90° intervals around each tree. For analysis, the four samples were grouped into a single suite for each tree. The increment corer was rinsed with deionized (DI) water between samples from the same tree and ethanol followed by copious amounts of DI water between trees. Even-numbered years from each tree sample suite were analyzed for carbon and nitrogen isotopic composition.
Tree-Ring Dating. Ten to twenty separate trees were sampled at each site. Cores were mounted, sanded, and polished with increasing grit sandpaper until individual rings were visible. Cores from the same site were cross-dated according to the principles of Stokes and Smiley (30). Distinctive ring-width patterns were identified in the crossdated core and used to dissect the core for analysis first into decades, then into individual annual rings using a stainless steel blade. Samples for each year were placed into individual vials, with each vial containing the annual ring from each of the four orthogonal cores taken from the tree. When possible, the distinctive ring patterns were used to cross-date trees from multiple sites.
Soil Sampling. At each site except New Brunswick, 3-10 soil samples were taken. The unconsolidated, recognizably organic detritus composing the O-horizon was removed with a trowel, and a sample of the top 5 cm of the A-horizon (the mineral horizon containing accumulated decomposed organic matter) was removed, placed in a plastic bag, and returned to the lab. Detailed soil horizon stratigraphies were not made. Rather, the top 5 cm of the A-horizon was taken as it represented the top of the rooting depth and potential source of bioavailable soil-derived nitrogen at each site. Samples were stored in a 4 °C refrigerator before and after processing. The samples were air-dried overnight and then sieved through a 2 mm mesh. Soil pH was determined on a 1:1 soil/water solution. A sub-sample was set aside for isotope measurement. Bulk soil was analyzed without carbonate extraction as none of the soil samples reacted with 20% HCl. This indicated that carbonate minerals were not present in the soil and carbon and nitrogen isotope analysis on a small selection (n ) 7) of soil samples indicated no significant difference between the isotopic composition of carbonate extracted and non-carbonate extracted soils.