Site Selection:
In the fall of 2005, three sites were selected that varied primarily in deciduous shrub abundance, hereafter referred to as low, medium and high shrub sites. Sites were chosen to have similar state factors (climate, relief, parent material, and time) but varied in the abundance of deciduous shrubs (Jenny, 1994). The sites represent a natural gradient of increasing shrub abundance because the same species of deciduous shrubs (Betula nana, and Salix pulchra) are found at all three sites (except S. richardsonii, which is found only at the medium shrub site); however, their percent cover increases from 15 to 94 %. Our sites are within 1 km of each other, and have similar parent material, and time since last glaciation (Itkillik I, deglaciated ca. 60 000 yr), and regional climate, although microclimates vary across sites due to differences in slope and aspect. The low shrub site is located on top of gently rolling hills, while the medium and high shrub sites are located in depressions along water tracks of ephemeral streams fed by spring snowmelt.
Our low shrub abundance site is located in moist acidic tussock tundra where the vegetation consists of approximately equal biomass of graminoids (Eriophorum vaginatum and Carex bigelowii), dwarf deciduous shrubs (B. nana, Vaccinium uliginosum, and S. pulchra), evergreen shrubs (Ledum palustre ssp. decumbens and V. vitis-idea), and mosses (Hylocomium splendens, Aulacomnium turgidum, Dicranum spp., and Sphagnum spp.) (Shaver and Chapin, 1991). In our medium shrub abundance site, vegetation consists of graminoids (primarily C. bigelowii), deciduous shrubs (B. nana, V. uliginosum, S. pulchra and S. richardsonii), and mosses (H. splendens and Dicranum spp.). Our high shrub abundance site has predominantly deciduous shrubs (B. nana, S. pulchra, and some Potentilla fruticosa) with some evergreen or wintergreen shrubs (V. vitis-idaea and Linnaea borealis), forbs (Polygonum bistorta, Petasites frigidus, Stellaria longipes, Valeriana capitata, and Artemisia alaskana), graminoids (Poa arctica, C. bigelowii, and Calamagrostis canadensis), and mosses (Sphagnum spp. and H. splendens).

Snow Manipulation:
To manipulate snow depth, snow fences (1.5 m high and 62 m long) were set up in the fall of 2005 at the low and medium sites to manipulate snow depth. For the high site, the patchy nature of the shrub stands made it necessary to set up two separate snow fences (1.5 m high and 32 m long) in patches with similar shrub composition and density. Our purpose for adding snow was to simulate the amount of snow that might be trapped by deciduous shrubs; therefore, the height of the snow fences was selected to match the maximum shrub height within the region. Fences were oriented E-W, and snow drifts accumulated on the northern side of the fences. Two treatments (control=ambient snow and drift=manipulated snow) were set up at each site. The drift plots were set up 4 m from the fence on the northern side of the fence, because this was the zone of maximum snow accumulation. At the low and medium sites, the control plots were set up on the southern (non-drift) side of the fence, 10 m from the fence at the low site, and 7 m from the fence at the medium site. These buffer zones were left to prevent the control from being exposed to snow trapped by the fence. At the high shrub sites, control plots were located in line with one of the fences, beginning 5 m from its end, and in 3 discontinuous blocks of tall shrubs to the south (control side) of the fence, beginning approximately 15 m from the fence. This arrangement was chosen because the cover of tall shrubs was discontinuous on the southern (control) side of the fences. For all sites, plots on the drift side of the fences were located in the zone of maximum snow accumulation, which was relatively uniform. Within each treatment, 18 2 by 10 m plots, with 1 m buffer strips between, were established. For this study, six plots per treatment (n=6) were randomly assigned to measure N mineralization and nitrification. Remaining plots were used for additional experiments.

Snow addition effects on litter decomposition (common substrate experiment):
To directly test the effect of microclimate and snow addition on litter decomposition rates we incubated the senesced leaves from a common substrate, B. neoalaskana (Sarg.), in the ambient and snow manipulated plots across all three sites. Senesced leaves were collected from trees growing near Fairbanks, AK. Leaves were still attached to the trees but the petiole had already started to abscise. This common substrate was used because the large leaf size and relative abundance allowed us to collect enough material for our study. Leaves were air dried, well mixed, and then subsampled for litter bags. One gram of leaves was sewn into 2 mm mesh bags, 8 x 8-cm in size. Litter bags were incubated beneath the live moss and litter layer starting in early June of 2006. The moss and litter in this system are well mixed so bags were inserted in this layer. Four identical bags were strung together for four separate annual harvests. Bags were placed in six treatment plots in each site (n=6), with three sub-replicates within each plot. Bags were removed in July of 2007, 2008, and 2009 and were kept frozen until they could be processed.
At time of processing, bags were thawed and then gently rinsed with deionized (DI) water to remove soil and loose litter attached to the outside of the bag. All original leaf litter was removed, dried at 45° C for a minimum of 48 hours and weighed. To determine the percent C and N of the litter, samples were ground to a fine powder on a Wiley-mill, with a #40 mesh screen, and then analyzed using an ECS 4010 elemental analyzer (Costech Analytical, Valencia, California, USA). Percent of initial mass remaining was calculated by dividing the incubated mass by the initial mass and multiplying by 100. Percent of initial C (ICR) and initial N remaining (INR) was calculated by the following equation: