Drivers of the microbial metabolic quotient across global grasslands - Risch et al. GEB_data.xlsx

Study sites and experimental design We collected data from 23 sites that are part of the Nutrient Network Global Research Cooperative (NutNet, https://nutnet.umn.edu/). The mean annual air temperature (MAT) across these sites ranged from -4 to 22°C, mean annual precipitation (MAP) from 252 to 1592 mm, and elevations from 6 to 4261 m above sea level (Fig 1a, Supplementary Table S1), hence cover a wide range of climatic conditions under which grasslands occur (Fig 1b). Soil organic C concentrations ranged between 0.8 to 7.8%, soil total N concentrations between 0.1 and 0.6%, and the soil C:N ratio between 9.1 and 21.5. Soil clay content spanned from 3.0 to 35%, and soil pH from 3.4 to 7.6 (Supplementary Table S2). At each site, the effects of nutrient addition and herbivore exclusion were tested via a randomized-block design (Borer et al., 2014). Three blocks with 10 treatment plots each were established at each site, except for the site at bldr.us (only two blocks). Each of these 10 plots was randomly assigned to a nutrient or fencing treatment. An individual plot was 5 x 5 m, divided into four 2.5 x 2.5 m subplots. Each subplot was further divided into four 1 x 1 m square sampling plots, one of which was set aside for soil sampling (Borer et al., 2014). Plots were separated by at least 1 m wide walkways. We collected soil samples from four different treatments for this study: (i) untreated control plots (Control), (ii) herbivore exclusion plots (Fence), (iii) plots fertilized with N, P, K, plus nine essential macro and micronutrients (NPK), and (iv) plots with simultaneous fertilizer addition and herbivore exclusion (NPK+Fence). The experiments were established at different times in the past, with years of treatment different among sites (2 – 9 years since start of treatment; Supplementary Table S1). For the nutrient additions, all sites applied 10 g N m-2 each year as time-release urea; 10 g P m-2 yr-1 as triple-super phosphate; 10 g K m-2 yr-1 as potassium sulfate. A micro-nutrient mix (Fe, S, Mg, Mn, Cu, Zn, B, Mo, Ca) was applied at 100 g m-2 together with K in the first year of treatments but not thereafter. We excluded large vertebrate herbivores (Fence) by fencing two plots, one with and one without NPK additions, within each block. The fences excluded all aboveground mammalian herbivores with a body mass of over 50 g (Borer et al., 2014). At most sites, the fences were 180 cm high, and the fence contained a wire mesh (1 cm holes) for the bottom 90 cm with a 30 cm outward-facing flange stapled to the ground to exclude burrowing animals. Climbing and subterranean animals may potentially still access these plots (Borer et al., 2014). For slight modifications in fence design at a few sites see Supplementary Table S3. Most sites only had wild herbivores, although four sites were also grazed by domestic animals (Supplementary Table S1).

Collection of soil samples, soil microbial respiration, microbial biomass, and other soil properties Each of the 23 sites received a package containing identical material from the Swiss Federal Institute for Forest, Snow and Landscape Research WSL, Switzerland to be used for sampling (Risch et al., 2015; Risch et al., 2019). We collected two soil cores of 5 cm diameter and 12 cm depth in each sampling plot and composited them to measure MR, MBC, and soil chemical properties (see below). An additional sample (5 x 12 cm) was collected to assess soil physical properties. This sample remained within a steel sampling core after collection and both ends were tightly closed with plastic caps to avoid disturbance. All soils were shipped cooled to the laboratory at (Location will be disclosed after manuscript acceptance) within a few days after collection. Soils were sampled roughly 6 weeks prior to peak biomass at each site during 2015 and 2016. To assess MR (CO2 production) in a laboratory incubation experiment we weighed duplicate soil samples (8 g dry soil equivalent) into 50-ml Falcon tubes. No additional substrate (glucose, sugar) was added to these samples. We adjusted the soil moisture of each sample to 60% field capacity. We then placed a 15 ml plastic test tube (Semadeni 1701A) containing 7.25 ml 0.05 M NaOH over each soil sample. The test tube was fixed with a plastic rod so that it was not in contact with the soil sample. The Falcon tubes were then sealed with a screw cap and placed in an incubator under completely dark conditions at 20°C. The CO2 produced by microbial respiration was absorbed by the 0.05 M NaOH. For five weeks we measured the decrease in conductivity within the 0.05 M NaOH solution on a weekly basis with a Multimeter WTW Multi 3410 (WTW GmbH, Germany) and replaced the 0.05 M NaOH with fresh solution. We included Falcon tubes without soil samples in each incubation run as blanks to test if tubes were tight and no CO2 could enter or escape. We calibrated the relationship between conductivity reduction and NaOH absorbed as follows: 400 ml 0.05 M NaOH was placed in a beaker and its conductivity was measured with the multimeter. While stirring, air containing CO2 was blown into the solution for approximately one minute, which reacts with NaOH to form Na2CO3. After this process, conductivity was measured again. We then transferred 7.25 ml of the solution into a smaller beaker and added 1 ml of 0.1 M BaCl2 to precipitate Na2CO3 and then titrated the solution with 0.05 M HCl to determine the remaining NaOH. We then repeated these steps with the remaining solution a total of nine times and plotted the conductivities (y-axis) against the NaOH consumed (x-axis, Supplementary Fig S1). This regression line was used to infer the consumption of NaOH from the conductivity reduction in the incubation experiments and to calculate CO2 evolution during incubation. In addition, we determined the optimum concentration for the NaOH solution in series of preliminary experiments, so that the concentration was not too high to become insensitive, but also not too low so that not all NaOH reacts during incubation. We then calculated MR (mg CO2-C kg dry soil-1 h-1) as total amount of CO2 released over the 5 weeks divided by the duration of the entire incubation in hrs. Soil microbial biomass carbon (MBC; mg C kg soil-1 ) was measured at the beginning of the experiment by measuring the maximal respiratory response to the addition of glucose solution (4 mg glucose per g soil dry weight dissolved in distilled water; substrate-induced respiration method) on approximately 5.5 g of soil (J. P. E. Anderson & Domsch, 1978; Nico Eisenhauer et al., 2018; Scheu, 1992). For this purpose we used an O2-micro-compensation apparatus (Scheu, 1992). More specifically, substrate-induced respiration was calculated from the respiratory response to D-glucose for 10 hr at 20°C. Glucose was added according to preliminary studies to saturate the catabolic enzymes of microorganisms (4 mg g soil-1 dissolved in 400 ml deionized water). The mean of the lowest three readings within the first 10 hrs (between the initial peak caused by disturbing the soil and the peak caused by microbial growth) was taken as maximum initial respiratory response (MIRR; ml O2 kg soil-1 h-1) and microbial biomass (mg C kg soil-1) was calculated as 38 x MIRR (Beck et al., 1997; Cesarz et al., 2022; Thakur et al., 2015). The rest of the composited sample was dried at 65°C for 48 h, ground and sieved (2 mm mesh) to assess the soil pH, mineral soil total C and N and C:N ratio, and mineral soil organic C (Risch et al., 2019). The undisturbed sample was used to assess water holding capacity (WHC), bulk density (BD), and soil texture [sand, silt, clay; methods in (Risch et al., 2019)]. We used the percentage of sand and clay as an indicator of soil texture in this study. MAT (°C), MAP (mm) and temperature of the wettest quarter (°C) were obtained from www.worldclim.com (Fick & Hijmans, 2017; Hijmans, Cameron, Parra, Jones, & Jarvis, 2005). These variables were selected as they were found to be drivers of soil nutrient processes across these sites in earlier studies (Risch et al., 2020; Risch et al., 2019). Mean annual soil temperatures (MAST; °C) for the 0 to 5 cm soil layer were obtained for each site from the SoilTemp maps (J. Lembrechts et al., 2021; J. J. Lembrechts et al., 2022), global gridded modelled products of soil bioclimatic variables for the 1979-2013 period at a 1-km² resolution, based on CHELSA, ERA5 and in-situ soil temperature measurements. Numerical calculations and statistical analyses We calculated MMQ as MR/MBC. We corrected this measure using the average soil temperature of each site (MMQsoil). This temperature correction is necessary as incubation temperatures are usually much higher than site mean annual soil temperatures (see Xu et al. 2017). MMQsoil = MMQ x Q10(MAST – 20)/10, where Q10 was assumed to be 2 (Xu et al. 2017). See Supplementary Fig S2 for comparison of air and soil temperatures across the 23 sites as well as the incubation temperature. Some of the explanatory variables (clay, soil organic C, C:N ratio) were skewed and were thus log-transformed prior to analyses. All continuous explanatory variables were centred and scaled to have a mean of zero and variance of one. To avoid collinearity between them we filtered them using correlation analysis (Supplementary Fig S3). From the variables that were strongly correlated (Pearson’s |r| > 0.70) (Dormann et al., 2013), we selected the ones that allowed us to minimize the number of variables (Supplementary Fig S3). Specifically, soil total N concentration, soil total C concentration, soil sand content and soil bulk density were dropped from the dataset. We then assessed how these edaphoclimatic variables are related to MMQ across our global grasslands. For this, we used linear mixed effects models (LMMs) fitted by maximum likelihood with the lme function in the nlme package (version 3.1-153) (Pinheiro, Bates, DebRoy, & Sarkar, 2021) in R version 3.6.3. (R Core Team, 2019). We used treatment as a fixed effect and plot nested in site as random effects to assess treatment differences in MMQsoil, as well as MR, and MBC. The number of years since the treatment started was included as a fixed effect in all the initial models but was not significant and therefore not retained in the models. To assess how differences in MMQsoil were affected by environmental factors (soil, climatic properties) we again used LMMs. Soil and climatic properties were included as fixed effects and plot nested in site as random effects. We did not include interactions between environmental variables. We then used the MuMin package (Barton, 2018) (version 1.42.1) to select the best models that explained the most variation based on Akaike’s information criterion (AIC; model.avg function). We used the corrected AIC (AICc) to account for our small sample size and selected the top models that fell within 4 AICc units (delta AICc < 4) (Burnham & Anderson, 2002; Johnson & Omland, 2004). We present all our top models rather than model averages. Conditional averages are provided in the Supplementary material. Based on findings from analyses described above and the literature, we developed a conceptual model of direct and indirect relationships between both edaphoclimatic variables and experimental treatments (Supplement Figure S4) to obtain a more holistic approach in understanding how these properties affect MMQsoil. We had data from 23 sites with 272 observations. We tested this model using structural equation modelling based on a d-sep approach (Lefcheck, 2016; Shipley, 2009). We considered those environmental drivers that were included in our top LMMs, namely temperature of the wettest quarter (T.q.wet), soil pH, water holding capacity (WHC) and soil organic C (organic C; Supplementary Figure S4). These factors were allowed to directly affect MMQsoil, and via their interactions with treatments. In addition, treatments were allowed to directly affect MMQsoil. Treatments were included as dummy variables in the model. We tested our conceptual model (Supplementary Fig S4) using the piecewiseSEM package (version 2.0.2; Lefcheck, 2016) in R 3.4.0, in which a structured set of linear models are fitted individually. This approach allowed us to account for the nested experimental design, and overcome some of the limitations of standard structural equation models, such as small sample sizes (Lefcheck, 2016; Shipley, 2009). We used the lme function of the nlme package to model response variables, including site as a random factor. Good fit of the SEM was assumed when Fisher’s C values were non-significant (p > 0.05). For all significant interactions between soil or climate variables and treatments detected in the SEMs, we calculated treatment effect sizes, i.e., the differences in MMQsoil between Control and treatments as log response ratios (LRR) and plotted these values against the climate or soil factors. The LRR were defined as log(Control/Treatment), where treatment was either Fence, NPK or NPK+Fence. To assess which of the LRR-climate or soil property relationships were significant we again used LMMs, in which soil and climatic properties were included as fixed effects and plot nested in site as random effects.

Champ	Valeur
Dernière modification de la donnée	14 août 2024
Dernière modification de la métadonnée	14 août 2024
Créé le	14 août 2024
Format	SERVICE
Identifier (Permalink)	drivers-of-the-microbial-metabolic-quotient-across-global-grasslands.1d2ce620-f7d6-4408-bf61-df123c0936e0
Date de publication	18 février 2023
Date de modification	18 février 2023
Langues	Anglais
URL d'accès	https://www.envidat.ch/dataset/drivers-of-the-microbial-metabolic-quotient-across-global-grasslands/resource/1d2ce620-f7d6-4408-bf61-df123c0936e0
Licence (Conditions d'utilisation)	Non-commercial Allowed / Commercial Allowed / Reference Required
Format	SERVICE
Couverture	-
Documentation	-
Service d'accès	-
Résolution temporelle	-
Byte size	110 080
Datastore active	False
Datastore contains all records of source file	False
Display name	Risch et al. GEB_data.xlsx
Has views	False
Id	f8685451-347c-4ab4-a4c2-f0b9d2a7c3e4
Identifier	drivers-of-the-microbial-metabolic-quotient-across-global-grasslands.1d2ce620-f7d6-4408-bf61-df123c0936e0
Issued	2023-02-18T03:43:03+00:00
Language	['en']
License	https://opendata.swiss/terms-of-use#terms_by
Modified	2023-02-18T04:43:07+00:00
Package id	da90eef6-2aa3-47b0-aa84-2c837782eddf
Position	0
Rights	https://opendata.swiss/terms-of-use#terms_by
State	active
Title	Risch et al. GEB_data.xlsx
Uri	https://www.envidat.ch/dataset/drivers-of-the-microbial-metabolic-quotient-across-global-grasslands/resource/1d2ce620-f7d6-4408-bf61-df123c0936e0

Risch et al. GEB_data.xlsx

Informations additionnelles