Ants and Ecosystem Function in Hemlock Removal Experiment at Harvard Forest 2006-2014

Exclosure construction: These sub-plots are located in the buffer-area of Simes plots 1-8. Exclosures are 3 x 3 m squares built from fiberglass roofing (3-m long x 30 cm high). The roofing is embedded 10-cm in the ground, leaving a 20-cm high wall. Corners are screwed to wooden posts and sealed with silicon caulk. Within this 3 x 3-m enclosure, a 2nd enclosure, 2 x 2 m in size, is constructed of 8” (~20-cm) diameter PVC sewer pipe cut in half lengthwise. The pipe is buried 5-cm deep, leaving 15-cm exposed, with the cut-edge (C-profile) pointing outwards. Corners are bolted into right-angle (90-degree) PVC elbows, also cut in half lengthwise, and sealed with silicon caulk. The result is a double-walled barrier separated by a 50-cm space. All monitoring and measurements are taken in the inner 2 x 2 m portion of the sub-plot. Exposed portions of both the fiberglass roofing and the PVC pipe barriers are sprayed with Tanglefoot insect barrier in May and July each year. Disturbance controls: These sub-plots are 2 x 2 m squares to match the working area of the exclosures, and they are located in the buffer area and within 5-15 m of the exclosures. Corners are marked with red plastic pipe (Pex) and the edges were dug and cleared of rocks in the same way we dug the exclosures to embed the PVC pipes. Controls: These sub-plots are also 2 x 2 m squares. In Simes plots 1-6, we co-located these sub-plots with two of the sub-plots used by Kathleen Savage and Eric Davidson to measure soil respiration in the center 30 x 30-m core area (see HF-130). In Simes plots 7 and 8 (the hardwood controls), control sub-plots are located in the buffer area within 5-m of the disturbance controls and the exclosures. Additions: In 2006, 20 dowels (10-cm long, 2 cm diameter, with 1-cm bored hole) and 20 nest boxes (plywood, 20 cm x 10 cm x 2cm with routed grooves) were scattered haphazardly 5-m away from the ant exclosures. Dowels and nest boxes were sampled for ant colonies in 2007 and 2008. Ant species composition and treatment effectiveness: To track changes in ant species richness and abundance, and to simultaneously determine effectiveness of the treatments, two pitfall traps are set in each sub-plot. Pitfall traps are 180-ml, 8-cm diameter urinalysis cups buried in the soil so that the rim of the cup is flush with the soil surface. Three times each summer, during the first weeks of June, July, and August, the cups are opened, filled with 10-ml of water to which dish detergent has been added (1 drop / gallon = 3.89 L) and left to collect ants for 48 hours. After 48 hours, trap contents are collected and cups are re-capped to await the next sample. Ants collected are identified and stored in 95% Ethanol in the Harvard Forest Sample Archive. Capped cups remain in place year-round to avoid repeated disturbance and further perforation of these small experimental units. Soil respiration: Soil respiration is measured in each sub-plot using Kathleen Savage and Eric Davidson’s Li-Cor 6000/6200 back-pack system. Permanent 25-cm diameter PVC collars are embedded in each sub-plot and repeated respiration measurements are made every two weeks during the growing season (May through October). Measurements were made in 2006-2008 (see HF-130). Nitrogen availability: Nitrogen availability is measured as extractable NO3 and NH4 on commercial resin-capsules (Unibest; WECSA, Inc.). Resin capsules are placed at the end of permanently embedded Wecsa-style samplers and are collected and replaced quarterly (January, April, July, October). Nitrogen is extracted and measured using a Lachat Autoanalyzer (see HF-054). Soil temperature: Soil temperature was initially measured using I-buttons, but failure rates more than 25% led to their abandonment in early 2008. We consider as reliable only plot-level soil temperature data from HF-108 and spot-measurements taken during soil respiration measurements. Decomposition rates of standard substrates: Whatman filter paper disks (10-cm) and tongue depressors (20-cm) were placed on the soil suface in flat mesh litter bags. Replicates were collected in 2008, 2009, and 2010, dried, and weighed. Ant samples and vouchers are stored in the Harvard Forest Sample Archive.