Data Package Metadata   View Summary

We manipulated root access to soil microbes by constructing mesocosms made from a 15 cm long, 5 cm diameter PVC pipe. Two 10.5 cm × 8 cm openings were cut along the length of the pipe, one on each side. We covered these openings, as well as the bottom of the pipe, with stainless steel mesh attached with rivets and nutrient-free glue (Household Goop, Eclectic Products, Eugene, USA). We covered 135 mesocosms with one of three mesh sizes: 1.45 mm (n = 45), 38 microm (n = 45), or 5 micro m (n = 45). We intended to exclude roots from fine-mesh mesocosms and allow root access to soil with large-mesh mesocosms (Johnson, Leake, and Read, 2001; Langley, Chapman, and Hungate, 2006; Phillips et al., 2012). However, our previous work showed the mesh design does not generate absolute root exclusion in all forest types (Moore et al., 2015). Mesocosms in this study varied in root density, thus we analyzed our data across a gradient of root density.

On May 12, 2014, we installed mesocosms randomly throughout the study site. We removed any organic horizon material and excavated the top 15 cm of mineral soil using a 5 × 15 cm hammer corer (AMS, Inc., American Falls, USA). The top 15 cm included only the A horizon. We removed visible roots to avoid a litter fertilization effect, and then filled each mesocosm with this root-free native soil. We placed each mesocosm in the hole from which it was collected and ensured the mesh was completely below the soil surface. Mesocosms were placed at least 0.5 m away from each other.

In-situ 13C-starch and 13C-leaf material incubation

Adding stable isotopes to soil enabled us to track microbial activity within specific C pools. We applied a 99 atom-% 13C labeled algal starch (Cambridge Isotope Laboratories, Tewksbury, MA, USA) and a greater than 97 atom-% 13C labeled ground tulip-poplar (Liriodendron tulipiferae) leaf material (IsoLife, Wageningen, Netherlands). We suspended 5 mg of powdered starch (0.58 mg C) or ground leaf material (2.3 mg C) in 30 mL of deionized water and injected into mesocosms that contained approximately 350 g soil. The amount of C added to soil from starch (1.6 microg C g-1) or leaf material (6.6 microg C g-1) was large enough to have a traceable label but small enough to not fertilize the soil, which contained on average 20 mg C g-1 soil at our site and is a similar amount to other C tracer field studies (Zak and Kling, 2006). The injections were conducted on June 24, 2014, six weeks after installing the mesocosms. We injected the starch suspension into 45 mesocosms (15 of each mesh size) and the leaf solution into 45 mesocosms (15 of each mesh size). To control for moisture addition and disturbance, we injected deionized water into 21 starch-control mesocosms (7 of each mesh size) collected on the same day as starch-addition mesocosms, and into 24 leaf-control mesocosms (8 of each mesh size) collected on the same day as leaf-addition mesocosms. Mesocosms injected with the starch suspension were sampled for 13CO2 on days 1, 2, 3, 4 and 5 after injection, and those with the leaf suspension were sampled on days 2, 4, 6, 10, and 20 after injection. We sampled gasses across several days because we were unsure which day CO2 flux would peak and this timeframe ensured we would capture peak microbial respiration of the 13C-labeled substrate (Zak and Kling, 2006). To collect gas samples for 13CO2 analysis, we capped the cores with a tightly fitting 5 cm diameter PVC cap fitted with a rubber septum. After 20 min, we used a syringe to draw a 15 mL sample of gas from the cap and injected the sample into a 12 mL Exetainer vacuum vial (Labco Limited, Lampeter, UK). One gas sample per sampling day was taken from the cores. At the beginning and end of each sampling day two ambient samples were taken to establish background levels of 13CO2. All 13CO2 samples were analyzed at the UC Davis Stable Isotope Facility (Davis, USA) using a ThermoScientific PreCon-GasBench system interfaced to a ThermoScientific Delta V Plus isotope ratio mass spectrometer (ThermoScientific, Bremen, USA). We removed all starch and starch-control mesocosms on June 29, 2014 and all leaf and leaf-control mesocosms on August 4, 2014. We placed the contents of mesocosms in a plastic bag, transported them in a cooler on ice, and stored at 4degC until they were analyzed.

To quantify root density, we removed unsieved soil from the mesocosms and visually inspected soils for roots. We used forceps to collect fine (less than 2 mm) roots and placed field-moist root mass into a clear-bottomed reservoir filled with water to a depth of approximately 2 cm. We scanned the roots in the reservoir on a photo scanner at 300 dpi resolution. We cropped the images to remove the border created by the reservoir, and then calculated root length using the Morphology plug-in and IJ Rhizo script for ImageJ software (Lobet and Draye, 2013). Root density is equal to root length per volume soil.

We analyzed microbial biomass C (MBC) within 48 hours of soil collection using the chloroform fumigation-extraction method (Vance, Brooks, and Jenkinson, 1987), allowing fumigated samples to incubate at room temperature for 5 days. All samples were stored at 4degC until analysis. We measured C of the samples on a total organic carbon analyzer (TOC-V CPH Total Organic Carbon Analyzer, Shimadzu Scientific Instruments, Columbia, USA). Microbial biomass C was calculated using a correction factor of 0.38 (Voroney, Brooks, and Beyaert, 2007).

We analyzed the potential enzyme activity of our soils using methods described by Bell et al. (2013) within 48 h of collection. Briefly, we mixed 2.75 g of field moist soil (sieved to 2 mm) with 91 mL of 50 mM sodium acetate buffer at pH 5 using an immersion blender. We pipetted 800 microL of soil slurry into a column on a deep (2 mL) 96-well plate that contained 0 -100 microM of methylumbelliferyl (MUB) to establish a standardized MUB reaction for each soil sample. We then pipetted 800 microL of the soil slurry into a separate plate and added 200 microL of 4-MUB-beta-D-glucoside (beta-gluc), 4-MUB-cellobioside (CBH), 4-MUB-N-acetyl-beta-D-glucosaminide (NAG), or 4-MUB-phosphate (PHOS) to each soil sample. beta-gluc and CBH are hydrolytic enzymes that work in concert to break down cellulose into glucose, and NAG and PHOS are used by microbes to acquire nitrogen and phosphorus, respectively. We sealed each plate with a plate mat, agitated vigorously by hand, then incubated the MUB standard and sample plates in the dark at room temperature for 3 h. Using a fluorometer/spectrophotometer (Synergy HT, Biotek Inc, Winooski, USA) we measured fluorescence at an excitation wavelength of 365 nm and an emission wavelength of 450 nm.