Data Package Metadata   View Summary

We collected the top 10 cm of mineral soil using a tulip bulb corer (5 cm diameter) after first removing and discarding organic horizon material. Some sites across the gradient have thin or no organic horizon, thus we focused on mineral horizons to make cross-site comparisons. At sites with simulated N deposition treatments, we collected and homogenized into a single composite sample three soil cores within each of the four replicate control and treatment plots. At sites with ambient-only N deposition (NY2 and PA) there are no long-term plots. We therefore established four replicate sampling plots separated by at least 20 m. We collected and homogenized into a single composite sample three soil cores collected along a 1.5 m radius from the center of the plot. We chose this distance because fungal communities tend to be autocorrelated at that scale (Lilleskov et al. 2004). Soils were kept cool on ice in the field. Within two hours of sample collection, we subsampled and flash froze in liquid N 2 ml of soil for genomic analysis. We then sieved (<2 mm) the remaining sample, removing visible roots, rocks, and organic debris. Sieved soils were stored at 4°C prior to analysis, and flash-frozen soil subsamples for genomic analyses were stored at -20°C.

We measured biogeochemical and edaphic parameters important to fungal physiology and community structure, including soil moisture content, pH, texture, total soil C and N, C and N mineralization, and fungal biomass. Soil moisture content was measured gravimetrically by drying samples for at least 48 h at 105°C. Soil pH was determined in distilled water at a 1:2 (w/v) ratio using potentiometry. Soil texture was assessed on 30 g of field-moist soil using the hydrometer method (Robertson et al. 1999). Soil C and N content was measured on air-dried and finely ground subsamples (10 g) using an elemental analyzer (Perkin Elmer 2400 Series II CHN, Waltham, MA). To estimate C and N mineralization, we incubated 20 g field-moist soil for 28 days in the dark at 25°C in sealed Mason jars. We collected 2 ml headspace gas samples each day for the first seven days and on days 14 and 28. Anaerobic activity was minimized by flushing the jars with CO2-free air after each sampling. Carbon efflux was measured on an infrared gas analyzer (LiCor 7500). We extracted inorganic N using 2M KCl (40 mL) from soil sub-samples (10 g) taken prior to and after the incubation. Inorganic N was determined as the sum of NH4+ and NO3- measured colorimetrically (Braman and Hendrix, 1989), and N mineralization rate was calculated as the difference between final and initial total soil inorganic N concentrations.

Fungal biomass was estimated using phospholipid fatty acid (PLFA) analysis on freeze-dried soil samples (Freezone 6; Labconco, Kansas City, Missouri, USA). Lipids were extracted from freeze-dried soil (1 g) using phosphate buffer, chloroform, and methanol (0.8:1:2 v:v:v). The polar lipids were isolated and purified using silicic acid chromatography and methanol. Polar lipids were then methylated in 0.2 mol/L methanolic potassium hydroxide (1 mL) at 60°C for 30 min to form fatty acid methyl esters (FAMES). FAMES were dried and reconstituted in hexane for quantification on a Varian CP-3800 gas chromatograph equipped with a flame ionization detector. We compared FAME peaks against a standard library of FAMES specific to fungi (18:2w6,9c, 18:1w9c; Matreya, Pleasant Gap, Pennsylvania, USA). Standard biomarkers were used to convert peak area concentrations into nmol PLFA per g dry soil.