Regional differences in soil stable isotopes and vibrational features at depth in three California Grasslands

There is an urgent need to investigate soil carbon dynamics in grasslands in differing precipitation and temperature regimes in light of current and future climate change. In this study, we assessed differences in soil stability, turnover, and organic matter composition along a climatic gradient of California grasslands. We focused on organic matter composition across three California grassland sites, from a dry (~300 mm precipitation/year) and hot regime (MAT: 14.6 ) to a wet (~2160 mm precipitation/year) and cool (MAT: 11.7 ) regime. We determined changes in total elemental and stable isotopic concentrations of soil carbon, nitrogen, δ13C, δ15N, and vibrational features as measured through Diffuse Reflectance Infrared Fourier Transformed Spectroscopy (DRIFTS) to 1m soil depth. We measured two isotopic indicators of soil turnover and stability, a natural carbon isotope approach defined as beta ( ), and the relationship between δ15N and C:N values. We did observe significantly more carbon concentration at our driest site and observed the lowest C:N values (~4) at depth at our wettest site. Carbon isotope indicators suggested the lowest stability at the dry site. Soils at depth (>30cm) at the wettest site had the greatest stability and had the greatest signal of carboxylic acid and polysaccharides. These results suggest differing stabilization mechanisms of organic matter at depth across our study sites. Results from this study suggest that taking into account both depth and regional differences in soil stability, turnover, and organic matter composition would aid with carbon sequestration efforts and aid biodiversity conservation efforts.  

Methods
Study Area 

Soil samples were collected at three annual grasslands: Sedgwick National Reserve (Suttle et al., 2007), University of California Hopland Research and Extension Center and the Angelo Coast Range Reserve in California (fig.1). All three sites are dominated by exotic annual grasses , while forbs are of both native and exotic origin. The dominant vegetation at Angelo is a mix of Aira spp., Bromus spp, and Briza spp. (Foley et al., 2023). At Hopland, it is a mix of Avena spp., Bromus spp., Erodium spp., and Festuca spp. (Foley et al., 2023). At Sedgwick, the dominant vegetation is Avena spp. and Bromus spp. (Foley et al., 2023). At Angelo, soils are part of the Holohan-Hollowtree-Casabonne Complex, are classified as Ultic Haploxeralfs. The parent material is largely gray wacke and mudstone. At Hopland, soils are part of the Yorkville series, and are classified as Typic Argixerolls. Parent material is largely sandstone and shale. Soils at Sedgwick are part of the Salinas series and are classified as Pachic Haploxerolls. Parent material is sandstone and shale. 

Each of these sites is an established reserve site within the University of California system, and as such, has detailed information on vegetation and climatic factors. To assess differences in plant communities, we aggregated all citations associated with plant surveys at each of our study sites (Supplemental table 1). We then aggregated information about whether each species was C3/C4, native/introduced, or is a N fixing or non-N fixing plant. This helped us determine potential plant community impacts on isotopic values. Plant communities in this region are typically dominated by C3 annual grasses.

Experimental Design and Sampling 

Soils were collected to 1m across the three study sites, Angelo, Hopland, and Sedgwick. Samples at Hopland and Sedgwick were collected by hand auger to 1m with 6-7 replicates per site (with each site encompassing approximately 3 ha), wherein we attempted to capture spatial heterogeneity but avoid confounding factors at each site by evenly sampling across similar slope positions (only Hopland had significant relief). At Angelo, samples were collected by Geoprobe due to being part of a different sampling campaign. At Angelo, samples were collected to depth of resistance (approximately 3m) with 4 replicates. Depths greater than 1m are not reported for Angelo in this study. At all sites, samples were collected at consistent 10cm intervals (0-10, 10-20, and so on).

After soils were collected, they were stored in coolers with ice packs in transport and then they were stored in a 4  cold room for approximately 4 months until they could be subsampled and analyzed. Storage took place during the pandemic due to a lack of access to laboratory facilities. When samples could be processed, a subsample was removed from each sample, and air dried for 7 days at room temperature. No carbonates were observed in these samples. Following air drying, the sample was then sieved through a 2mm sieve. A further subsample was taken for ball milling (using a Sample Prep 8000M Ball Mill) to a homogenous particle size.

We collected bulk density at Angelo and Hopland through Geoprobe cores, and calculated carbon stocks with these bulk density estimates. At Angelo, we subsampled each depth increment to estimate water content, and then calculated the dry mass of soil in a 10cm increment. Bulk density was calculated as the mass of the dry >2mm fraction to correct for the impact of rock and root volume on soil carbon and nitrogen stocks (Throop et al., 2012). However, we did not observe a high coarse fraction at Hopland.

Elemental and Isotopic Analyses 

The d13C and d15N values and elemental carbon and nitrogen contents of all samples were measured in the Stable Isotope Ecosystem Laboratory at the University of California, Merced (SIELO). Briefly, samples were weighed into tin capsules and combusted in a Costech 4010 Elemental Analyzer coupled with a Delta V Plus Continuous Flow Isotope Ratio Mass Spectrometer. Carbon and nitrogen isotope compositions were corrected for instrumental drift, mass linearity, and standardized to the international VPDB (d13C) and AIR (d15N) scales using the USGS 41A and USGS 40 standard reference materials. Mean d13C values for reference materials were USGS 40 = -26.4 ± 0.1‰ (n = 173) and USGS 41a = 36.5 ±0.2‰ (n = 87) and corresponding mean d15N values were USGS 40 -4.5 ± 0.1‰ (n = 173) and USGS 41a 47.5 ± 0.1‰ (n = 87). Elemental carbon and nitrogen content were determined via linear regression of CO2 and N2 sample gas peak areas against the known carbon and nitrogen contents of USGS 40, USGS 41a, and Costech acetanilide. All isotope compositions are expressed in standard delta notations.

Diffuse Reflectance Infrared Fourier transform Spectroscopy (DRIFTS)

To characterize the chemical composition of soil C across our study systems, we used diffuse reflectance mid-infrared Fourier Transform spectroscopy (DRIFTS) analyses on bulk soil samples. DRIFTS measures the vibrational frequencies of functional groups in a sample, and allows us to measure the presence of functional groups that are important for organic matter and mineral surfaces. DRIFTS is informative on the abundance of organic and inorganic substances by measuring the excitation of molecular bonds when exposed to infrared radiation (Parikh et al., 2014). We used a Bruker IFS 66v/S Spectrophotometer (Ettlingen, Germany) with a praying Mantis apparatus (Harrick Scientific, Ossining, NY).  Potassium bromide (KBr) was used as a background reference, btu samples were not diluted with KBr. Samples were first dried in a desiccator following homogenization to remove interference from water. Absorption was measured between 4000 and 400 cm-1 averaged over 300 scans with an aperture of 4mm. Functional group assignments are based on Parikh et al. (2014)