Correction Factors for Dissolved Organic Carbon Extracted from Soil in New England 2012-2013

Organic horizon (OH) and mineral soil (MS) samples were collected at the Hubbard Brook Experimental Forest, New Hampshire (43°56’N, 71°45’W) and Harvard Forest, Massachusetts (42°32’N, 72°11’W) in April, May, August, and October of 2012 and 2013. At Hubbard Brook, samples were collected in a mature sugar maple–yellow birch stand. Harvard Forest samples were collected in a mature red oak–red maple stand. Samples were 10×10 cm OH monoliths and 5 cm-diameter MS cores from the top 15 cm of the mineral soil. At Harvard Forest, OH samples were also collected in a primary-growth hemlock stand in August 2013. Peat samples were removed from the Central Unit of Caribou Bog, a 2200-ha peatland complex near Bangor, Maine (44°56’N, 68°46’W). We collected 5 cm-diameter peat cores from the top 20 cm of the dead moss layer in May, August, and October 2013 in sites dominated by (1) Sphagnum lawn, bryophytes, and sedges, (2) Sphagnum lawn and low ericaceous shrubs (Kalmia polifolia Wangenh., Rhododendron groenlandicum (Oeder) Kron and Judd) and (3) ericaceous shrubs and black spruce (Picea mariana (Mill.) B.S.P.). Samples were immediately brought back to the laboratory and stored at 4°C. OH and MS samples were homogenized by sieving through 2-mm mesh (4-mm mesh for hemlock OH) and removing rocks, roots, and woody debris. Roots were removed from peat samples without sieving. Five grams each of OH and peat sample or 10 g of MS sample were placed in 50-mL centrifuge tubes. 40 mL of 0.5 M K2SO4 were added to each tube. Slurries were shaken for 1 h on an oscillating table and filtered through a Whatman #1 paper filter. The extracts were frozen at −20°C until dissolved organic carbon (DOC) concentration was measured. Time between sample collection and freezing of the extracts was kept under 48 h. To measure DOC, we adapted Bartlett and Ross's (1988) method to use with a 96-well microplate reader. On each microplate, we pipetted duplicate 100-mL aliquots of eight oxalic acid standards with concentrations ranging from 0- to 4-mM C and triplicate 100-mL aliquots of 24 soil extracts. We added 50 mL of 10 mM Mn(III)-pyrophosphate solution and 50 mL of concentrated sulfuric acid to each well. The 96-well microplates were incubated in the dark at room temperature for 18 h before measuring absorbance at 495 nm on a microplate reader (VersaMax, Molecular Devices, Sunnyvale, California, USA). For each microplate, soil extract absorbance was converted to C concentration using a linear calibration curve based on the oxalic acid standards. We also measured total organic carbon content of the same soil extracts using an Apollo 9000 TOC Analyzer with autosampler (Teledyne Tekmar, Mason, Ohio, USA). Given the high variability in the recovery rate of the various organic compounds and the great difficulty in measuring the relative abundance of these and other substances in environmental samples, it is highly probable that reported DOC concentrations using the Bartlett and Ross (1988) colorimetric method are biased high or low depending on the particular mix of DOC compounds present. To determine the magnitude of bias in soil samples collected in New England ecosystems, we compared the DOC content of OH, MS, and peat samples measured using the colorimetric method and the TOC combustion analyzer.