Fine root biomass was measured in each plot in late August 2010 (pre-treatment) and in subsequent years by soil coring and manual dry sorting of live roots from soil. Multiple soil cores were collected in each plot from locations near our permanent soil respiration collars, adjusted to avoid large roots and rocks. After removing the litter layer (Oi horizon), a 5-cm diameter split-PVC pipe corer was hammered into the soil with a rubber mallet. The nominal depth of sampling was 30 cm but because of obstructions the actual depth of sampling averaged 27 cm. Each core was divided in the field into two depth increments, 0-10 cm (including the Oe and Oa horizon and usually some mineral material) and 10-30 cm (dominantly E and B mineral horizons with varying amounts of organic matter). Samples were transported to the lab for storage at -20 °C until laboratory processing.
Live fine roots were hand sorted from each sample; dead roots were distinguished by their dark color and low tensile strength. For roots in the 0-10 cm depth increment, fine root biomass was estimated from the diminishing root mass recovered during sequential, timed picking intervals. For this approach, we manually extracted roots from soil cores for four intervals of 10 minutes each, and the cumulative biomass extracted over time was predicted by fitting a logarithmic curve. Total fine root biomass was estimated at the point at which the predicted incremental root mass extracted in the next 10-minute time interval was <2% of the cumulative total. On average, this required 6 more picking intervals, representing a time savings of one hour per core. For 10-30 cm samples, which had less root mass, all roots were hand sorted from each core. Sorted roots were washed free of adhering soil on a fine sieve, dried to constant mass at 70 °C and weighed.
Ingrowth cores were installed in 3 mature forest stands (C7, C8, and C9; Table 1) in the Bartlett Experimental Forest. Each of these stands has four 1/4 ha (50 m x 50 m) treatment plots, treated annually beginning in spring 2011, with N (30 kg N/ha/yr as NH4NO3), P (10 kg P/ha/yr as NaH2PO4), N+P, or nothing (an untreated control).
Ingrowth cores were used to measure root growth and foraging for nutrients in mature and young forest stands. In mature stands, we estimated rates of fine root growth in each plot by using ingrowth cores containing soils from the same plot (i.e., C, N, P or NP), and we tested fine root foraging for P or Ca derived from added apatite.
Ingrowth cores were installed in mature stands in early June 2013 in four locations (soil sampling subplots SS-A, SS-B, SS-C, and SS-D) in each treatment plot. At each location, we collected 4 soil cores (5 cm in diameter, 25 cm depth). Soil was separated into three layers: forest floor (FF, consisting of Oe+Oa horizons), 0-10 cm mineral soil and 10- 20 cm mineral soil. Soil was prepared for filling each layer of the core holes in a plot by combining cores of the same layer with extra soil from that layer in the same plot, homogenizing, and passing through 5 mm mesh hardware cloth to remove roots and rocks. Two types of cores were prepared: apatite-treatment cores, each of which contained 6.4 g ground apatite (125 um) mixed evenly into mineral soil layers, and cores with no treatment, which contained soil from the plot with nothing additional added to it. We refilled each core hole with 10 cm of each of the two mineral layers and 5 cm of homogenized forest floor. We inserted a 5-cm diameter ring of hardware cloth in the top to mark the core location.
Mature forest ingrowth cores incubated in situ for 2 years to allow complete root colonization. During this time the cores were fertilized individually with equal amounts of N or P per m2 as the surrounding plot. In August 2015 all cores were extracted by re-coring the marked locations with a smaller diameter (4.5 cm) corer to a depth of 20 cm in the mineral soil. The cores were separated into three layers (forest floor, 0 – 10 cm mineral soil, and 10 – 20 cm mineral soil), transported to the lab, and stored at 4 °C for up to 2 weeks prior to further processing. Fine roots (< 1 mm) diameter were sorted from soils by hand, washed with tap water, and scanned. Fine root length was quantified with the Analyze Skeleton plugin (Niemisto and others 2005) in ImageJ (Schneider and others 2012). Dry root mass was measured after oven drying at 60 °C.
In young stands we estimated rates of fine root growth in each plot by using ingrowth cores containing soils from the same plot (i.e., C, N, P or NP), similar to mature forest stands. We also studied fine root foraging by "transplanting" soils from other treatment plots. Ingrowth cores were installed in young stands (C1 and C2) in October 2017 in the treated buffer zones. Soil cores were removed to 20 cm depth with a 5-cm diameter corer. Three steel rods (0.5 cm diameter) were positioned vertically on the wall of the core holes to assist with later retrieval. In each plot 10 holes were filled with soil taken from the same plot. In control plots additional holes were filled with soil from N, P, and NP treatment plots (10 holes per treatment), and in N and P plots additional holes were filled with soil from plots fertilized with the other nutrient (10 per treatment). Each ingrowth core was fertilized individually with equal amounts of N or P per m2 as the surrounding plot in early June 2018, and cores were collected using a 4-cm diameter corer in October 2018. Harvested ingrowth cores were returned to the laboratory cold and then frozen at -20 °C until processing. Live fine roots of 0-1 mm diameter were collected from each core, cleaned of adhering soil over a fine sieve, dried to constant mass at 70 °C and weighed.
