Elk and bison carcasses in Yellowstone, USA

This dataset contains all data on which the following publication below is based. Paper Citation: Risch, A.C., Frossard, A., Schütz, M., Frey, B., Morris, A.W., Bump, J.K. (accepted) Effects of elk and bison carcasses on soil microbial communities and ecosystem functions in Yellowstone, USA. (accepted). Functional Ecology doi: ... Methods Study area and study sites This study was conducted in YNP’s Northern Range (NR), located in north-western Wyoming and south-western Montana, USA (~44.9163° N, 110.4169° W). The NR expands over ~1000 km2 and features long cold winters and short dry summers. Grasslands and shrublands dominate the NR that is the home of large migratory herds of bison (winter counts 2017: ~3919 individuals; Geremia, Wallen, & White, 2017) and elk (~5349 individuals) as well as their main predators, approximately five packs of wolves with a total of 33 individuals (Smith et al., 2017). As part of a long-term research program within YNP, wolf predation has been studied since their reintroduction in 1995. For our study, we received ground-truthed coordinates of bison and elk carcasses from winter 2016/17 (November 2016 through April 2017) from the YNP Wolf Project. Between June 20 and July 1, 2017, we visited 24 carcasses in total. At five sites, we could not sample as the carcasses were no longer found. In total we located remains (hairmats, rumen content, bones, teeth) of 19 adult male and female carcasses (7 bison, 12 elk; Supplementary Table 1). Live body weights of adult bison and elk are approximately 730 kg (male bison), 450 kg (female bison), 330 kg (male elk), and 235 kg (female elk, Meagher, 1973; Quimby & Johnson, 1951). The kills and subsequent consumption happened between 34 and 173 days prior to our sampling (hereafter “days since kill”, DSK), for which we accounted in our statistics. Note that wolves and other scavengers consumed the soft tissue of the carcasses quickly, hence, there is close to no soft tissue left for decomposition as compared to an intact body left on the soil surface. The 19 carcass sites covered the extent of YNP’s NR, with both bison and elk carcasses showing similar distributions; elevation ranged from 1703 to 2884 m a.s.l. (Supplementary Fig 1 & Supplementary Table 1). The carcasses were all located in grassland or sage-brush shrubland, with or without sparsely scattered trees, and both bison and elk carcasses showed the same distribution of DSK. At each study site, we selected a reference plot (hereafter “control”) that was of comparable size, slope aspect and vegetation to the carcass location (hereafter “carcass”). The control was at least 10 m away (Danell, Berteaux, & Brathen, 2002; Melis et al., 2007) from the carcass itself to ensure the absence of potential direct and indirect carcass effects (paired design; (Bump, Webster, et al., 2009; Bump, Peterson, et al., 2009). Ecosystem functions and soil properties We randomly collected 50 g of mineral soil from three locations on both control and carcass plots to a depth of 5 cm with sterile techniques and gently mixed the material to obtain a composite sample. Half the soil sample was immediately bagged in plastic bags (whirl packs), stored in a cooler with ice packs (~5 ºC), sieved (2-mm) and frozen within 4-6 hours of collection to assess soil microbial communities. For this purpose, we extracted total genomic DNA from 0.5 g soil using the PowerSoil DNA Isolation Kit (Qiagen, Hilden, Germany). DNA concentrations were measured using PicoGreen (Molecular Probes, Eugene, OR, USA). PCR amplifications of partial bacterial small-subunit ribosomal RNA genes (region V3–V4 of 16S rRNA) and fungal ribosomal internal transcribed spacers (region ITS2) were performed as described previously (Frey et al., 2016). Each sample consisting of 40 ng DNA was amplified in triplicate and pooled before purification with Agencourt AMPure XP beads (Beckman Colter, Berea, CA, USA) and quantified with the Qubit 2.0 fluorometric system (Life Technologies, Paisley, UK). Amplicons were sent to the Genome Quebec Innovation Center (Montreal, Canada) for barcoding using the Fluidigm Access Array technology and paired-end sequencing on the Illumina MiSeq v3 platform (Illumina Inc., San Diego, CA, USA). Quality control of bacterial and fungal reads was performed using a customized pipeline (Supplementary Table 2; Frey et al., 2016). Paired-ends reads were matched with USEARCH (Edgar & Flyvbjerg, 2015), substitution errors were corrected using Bayeshammer (Nikolenko, Korobeynikov, & Alekseyev, 2013) and PCR primers were trimmed (allowing for 1 mismatch, read length -300 bp for 16S and -200 bp for ITS primers) using Cutadapt (M. Martin, 2011). Sequences were dereplicated and singleton reads removed prior to clustering into operational taxonomic units (OTUs) at 97% identity using USEARCH (Edgar, 2013). The remaining centroid sequences were tested for the presence of ribosomal signatures using Metaxa2 (Bengtsson-Palme et al., 2015) or ITSx (Bengtsson-Palme et al., 2013). Taxonomic assignments of the OTUs were obtained using Bayesian classifier (Wang, Garrity, Tiedje, & Cole, 2007) with a minimum bootstrap support of 60% implemented in mothur (Schloss et al., 2009) by querying the bacterial and fungal reads against the SILVA Release 128 (Quast et al., 2013) and UNITE 8.0 (Abarenkov et al., 2010) reference databases for 16S and ITS OTUs, respectively. Abundances of the bacterial 16S rRNA gene and fungal ITS amplicon were determined by quantitative real-time PCR (qPCR) on an ABI7500 Fast Real-Time PCR system (Applied Biosystems, Foster City, CA, USA) as described previously (Frossard et al., 2018). The same primers (without barcodes) and cycling conditions as for the sequencing approach were used for the 16S and ITS qPCR. Three standard curves per target region were obtained using tenfold serial dilutions of plasmids generated from cloned targets (Frey, Niklaus, Kremer, Lüscher, & Zimmermann, 2011). Data were converted to represent mean copy number of targets per gram of soil (dry weight). The other half of the soil sample was bagged in paper, dried to constant weight at 60°C, passed through a 2 mm sieve and analyzed for total C and N concentration with a CE Instruments NC 2100 soil analyzer (CE Elantech Inc., Lakewood NJ, USA). We also collected 20 mature and undamaged leaves of the dominant grass species growing on control and carcass sites, but taxa were not recorded. The plant material was dried at 60°C, finely ground till homogenized and also analyzed to obtain total C and N concentrations. Soil temperature (10 cm depth) was measured with a waterproof digital thermometer (Barnstead International, Dubuque IA, USA) at three locations each at the control and carcass site. Soil moisture (0 – 10 cm depth) was measured with time domain reflectometry (Field-Scout TDR-100; Spectrum Technologies, Plainfield IL, USA) at five randomly chosen points on control and carcass sites. We measured soil respiration at five randomly chosen points at both control and carcass sites with a PP-Systems SRC-1 soil respiration chamber (closed circuit) attached to a PP-Systems EGM-4 infrared gas analyzer (PP-Systems, Amesbury, MA, USA). For each measurement the soil chamber (15 cm high; 10 cm diameter) was tightly placed on the soil surface, after clipping plants to avoid measuring plant respiration or photosynthesis. Measurements were conducted over 120 s. In addition, we assessed the decomposition rates of standardized OM using the cotton strip assay (Latter & Howson, 1977; Latter & Walton, 1988). Cotton cloth tensile strength loss (CTSL) is a measure of decomposition, and an index to express the combined effect of soil microclimatic, physical, chemical and biological properties on decomposition while accounting for OM quality (Latter & Walton, 1988; Risch, Jurgensen, & Frank, 2007; Withington & Sanford Jr., 2007). We placed five 20 cm wide x 13 cm long sheets of 100% unbleached cotton cloth (American Type SM 1/18’’, Warp: 34/1, Weft: 20/1, Weave plain, 29.5 picks/cm warp, 22 picks/cm weft, 237 g/m2; Daniel Jenny & Co., Switzerland;) at each carcass and control site vertically into the soil by making slits with a flat spade to a depth of 12 cm. We inserted each cloth with the spade, and then pushed the slit closed to assure tight contact with the soil. The cloths were retrieved after 18 to 27 days. After retrieval, the cloths were air-dried, remaining soil gently removed by hand, and 1.5 cm wide strips were cut at the 3.5-5.0 cm (top) and the 9-10.5 cm (bottom) soil depth. The strips were equilibrated at 50 % relative humidity and 20°C for 48 hours (climate chamber) prior to strength testing (Scanpro Awetron TH-1 tensile strength tester; AB Lorentzen and Wettre, Kista, Sweden). Cotton rotting rate (CRR) = (CTScontrol - CTSfinal/CTSfinal)1/3 * (365/t), where CTScontrol is the cotton tensile strength of a control cloth and CTSfinal the cotton tensile strength of the incubated sample, t is the incubation period in days. Control cloths were inserted into the ground and immediately retrieved to account for tensile strength loss associated with cloth insertion. We averaged the CRR of top and bottom strips for further analyses as no difference was found between the two. All sampling and cloth insertion took place between June 20 and July 1, 2017, cloths were retrieved between July 17 and 20, 2017. Soil respiration, average CRR, vegetation N concentration and vegetation C:N ratio are defined as ecosystem functions, soil C and N concentration, soil temperature and moisture as soil abiotic properties, and bacterial and fungal richness (number of taxa), diversity (Shannon) and abundance as soil biotic properties. Statistical analyses Univariate analyses for ecosystem functions, soil biotic and abiotic properties We tested whether individual ecosystem functions, soil biotic and abiotic properties differed between carca

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.