Soil Sampling and Field Processing

Soil samples were collected on four dates to characterize the soil microbial biomass carbon (C) pool, extractable soil NO3-, and microbial community composition, starting first in autumn after plant senescence (September 12, 2016), at peak winter snow depth (March 7, 2017), during the snowmelt period (May 9, 2017), and following the complete loss of snow and the start of the plant growing season (June 9, 2017). During snow-free times of the year, soils were collected using a 4 cm diameter soil bulk density corer at 12 plots at the Hillslope and at 3 plots at the Floodplain. The six plots on the Hillslope were encompassed by a representative 80 m × 100 m area that was approximately 100 m upslope from the edge of the Floodplain. Each plot on the Hillslope was 20 m × 20 m and separated by at least 2.5 m. The Hillslope plots were arranged in three pairs that ran parallel to the river. The paired plots will serve as experimental blocks for a future field experiment. The three plots in the Floodplain were similarly 20 m × 20 m and arranged in a 100 m × 30 m transect that was perpendicular to the river and spanned the area from the toeslope-floodplain transition to near the edge of the river. Soil cores were subsampled and split into three discrete depth increments; 0 to 5 cm, 5 to 15 cm, and 15 cm + below the soil surface. These relatively shallow soils are best described as A-horizons at the Hillslope and Floodplain locations. Soil horizon development and differentiation is greatly reduced because of substantial bioturbation from burrowing mammals on the Hillslope and periodic sediment deposition during spring flooding events in the Floodplain.

During periods of winter snow cover (i.e., March and May 2017), snow-pits were dug down to the soil surface at three locations on the Hillslope and one location in the Floodplain in order to sample soils from beneath the snowpack. In each snow-pit, soils were sampled at two adjacent locations separated by more than 1 m using the soil coring method described above. New snow pits were dug in different locations on the Hillslope and Floodplain during snowmelt in May 2017 to avoid repeatedly sampling the same locations that were sampled at peak winter snow depth in March 2017. Thus, during the snow-free time of the year, we sampled and analyzed 180 total soil samples (2 ecosystem types × 15 Plots × 2 Time Points × 3 Depths) and 48 soils total during winter (4 Snow-pits × 2 Time Points × 3 Depths × 2 replicate cores). A ∼10 g subsample from each soil core at each depth was placed immediately on dry ice in the field, frozen, and archived for archaeal, bacterial and fungal community analysis. The remainder of the soil core was allocated to physical and chemical characterization (described below) and was stored at 4°C until further analysis.