Effects of Climate Change on Soil Organic Matter C and H Isotope Composition in a Mediterranean Savannah (Dehesa): An Assessment Using Py-CSIA.

Dehesas are Mediterranean agro-sylvo-pastoral systems sensitive to climate change. Extreme climate conditions forecasted for Mediterranean areas may change soil C turnover, which is of relevance for soil biogeochemistry modeling. The effect of climate change on soil organic matter (SOM) is investigated in a field experiment mimicking environmental conditions of global change scenarios (soil temperature increase, +2-3 °C, W; rainfall exclusion, 30%, D; a combination of both, W+D). Pyrolysis-compound-specific isotope analysis (Py-CSIA) is used for C and H isotope characterization of SOM compounds and to forecast trends exerted by the induced climate shift. After 2.5 years, significant δ13C and δ2H isotopic enrichments were detected. Observed short- and mid-chain n-alkane δ13C shifts point to an increased microbial SOM reworking in the W treatment; a 2H enrichment of up to 40‰ of lignin methoxyphenols was found when combining W+D treatments under the tree canopy, probably related to H fractionation due to increased soil water evapotranspiration. Our findings indicate that the effect of the tree canopy drives SOM dynamics in dehesas and that, in the short term, foreseen climate change scenarios will exert changes in the SOM dynamics comprising the biogeochemical C and H cycles.

Fire effects on C and H isotopic composition in plant biomass and soil: Bulk and particle size fractions.

This work studies carbon (C) and hydrogen (H) isotope composition of plant biomass and soil organic matter (SOM) in an attempt to assess both, changes exerted by fire and possible inputs of charred materials to the soil after a wildfire. Isotope composition of bulk soil, soil particle size fractions and biomass of the dominant standing vegetation in the area (Quercus suber) from Doñana National Park (SW-Spain) were studied by isotope ratio mass spectrometry (IRMS). SOM C isotope composition indicates the occurrence of two SOM pools with different degree of alteration. Coarse soil fractions (>0.5 mm) were found 13C depleted with δ13C values close to those in leaf biomass, pointing to a predominance of poorly transformed SOM. Conversely, fine fractions (<0.1 mm) were found enriched in 13C as corresponds to a more humified SOM. The fire produced no changes in this trend, although a consistent 13C enrichment (c. 1‰) was observed in all soil fractions with decreasing size. Concerning H isotopes, the coarse fractions (>0.5 mm) displayed significant lower δ2H values than the intermediate and fine ones (<0.5 mm), again similar to those in leaf biomass (c. -80‰), whereas the fine fractions were found deuterium (2H)-enriched with significant higher δ2H values (c. 50‰), suggesting physical speciation of H depending on soil particle size. The fire produced a significant 2H depletion (Δ2H c. -10‰) in the finer fractions (<0.1 mm). The study of stable isotope analysis added new information and complements the results obtained by other proxies to better understand the effect of fire on SOM.

Effect of a wildfire and of post-fire restoration actions in the organic matter structure in soil fractions.

The impact of wildfires and of restoration actions on soil organic matter (SOM) content and structure was studied in a soil under pine (Pinus pinea) from Doñana National Park (SW Spain). Samples were collected from burnt areas before (B) and after post-fire restoration (BR) and compared with an unburnt (UB) site. Analytical pyrolysis (Py-GC/MS) was used to investigate SOM molecular composition in whole soil samples and in coarse (CF) and fine (FF) fractions. The results were interpreted using a van Krevelen graphical-statistical method. Highest total organic carbon (TOC) was found in UB soil and no differences were found between B and BR soils. The CF had the highest TOC values and FF presented differences among the three scenarios. Respect to SOM structure, the B soil was depleted in lignin and enriched in unspecific aromatics and polycyclic aromatic hydrocarbons, and in all scenarios, CF SOM consisted mainly of lignocellulose derived compounds and fatty acids. In general, FF SOM was found more altered than CF. High contribution of unspecific aromatic compounds and polycyclic aromatic hydrocarbons was observed in B-FF whereas BR-FF samples comprised considerable proportions of compounds from labile biomass, possibly due to soil mixing during rehabilitation actions. The fire caused a defunctionalisation of lignin-derived phenolics and the formation of pyrogenic compounds. The van Krevelen diagram was found useful to-at first sight-differentiate between chemical processes caused by fire and of the rehabilitation actions. Fire exerted SOM demethoxylation, dealkylation and dehydration. Our results indicate that soil management actions after the fire lead to an increase in aromaticity corresponding to the accumulation of lignin and polycyclic aromatic compounds. This suggests additional inputs from charred lignocellulosic biomass, including black carbon, that was incorporated into the soil during rehabilitation practices.

Ultra-high resolution mass spectrometry of physical speciation patterns of organic matter in fire-affected soils.

Fire is one of the most important modulating factors of the environment and the forest inducing chemical and biological changes on the most reactive soil component, the soil organic matter (SOM). Assuming the complex composition of the SOM, we used an ultra-high resolution mass spectrometry analysis technique to assess the chemical composition and fire-induced alterations in soil particle size fractions (coarse and fine) from a sandy soil in a Mediterranean oak forest at Doñana National Park (Southwest Spain). Electrospray ionization-Fourier transform ion cyclotron resonance mass spectrometry (ESI-FTICRMS) showed that the coarse fraction of soils not affected by fires consisted mainly of polyphenolic compounds consistent with little-transformed SOM and fresh biomass, whereas the fine fraction was enriched in protein and lipid like homologues suggesting microbially reworked SOM. In fire-affected SOM, the coarse fraction contained a high proportion of aromatic compounds, consistent with inputs of charred litter or in situ chemical transformation of the SOM. Analysis of the fine fraction revealed two differentiated chemical families pointing to the existence of two carbon pools; a native microbial-derived moiety composed of lipids and polypeptide compounds, and a secondary, pyrogenic or thermally-altered moiety rich in aromatic compounds. This work represents the first application of ultra-high resolution mass spectrometry to study the chemical composition of SOM in different particle size fractions.

Wildfire effects on lipid composition and hydrophobicity of bulk soil and soil size fractions under Quercus suber cover (SW-Spain).

Soil water repellency (hydrophobicity) prevents water from wetting or infiltrating soils, triggering changes in the ecosystems. Fire may develop, enhance or destroy hydrophobicity in previously wettable or water-repellent soils. Soil water repellency is mostly influenced by the quality and quantity of soil organic matter, particularly the lipid fraction. Here we report the results of a study on the effect of fire on the distribution of soil lipids and their role in the hydrophobicity grade of six particle size fractions (2-1, 1-0.5, 0.5-0.25, 0.25-0.1, 0.1-0.05 and <0.05mm) of an Arenosol under Quercus suber canopy at the Doñana National Park (SW-Spain). Hydrophobicity was determined using water drop penetration time test. Field emission scanning electron microscopy (FESEM) was used to assess the presence and morphology of the inorganic and organic soil components in the particle size fractions. Soil lipids were Soxhlet extracted with a dichloromethane-methanol mixture. Fatty acids (FAs) and neutral lipids were separated, derivatized, identified and quantified by gas chromatography/mass spectrometry and gas chromatography/flame ionization detection. The hydrophobicity values of soil samples and fractions were statistically different (P < 0.05), for both, the unburnt and burnt soils, and particle size fractions. All samples displayed a similar distribution of FAs, straight-chain saturated acids in the C14-C32 range, and neutral lipids (n-alkan-1-ols, n-alkanes), only differing in their relative abundances. Among possible biogeochemical mechanisms responsible for the changes in soil lipids, the observed depletion of long chain FAs (C≥24) in the coarse fraction is best explained by thermal cracking caused by the heat of the fire. The enrichment of long chain FAs observed in other fractions suggests possible exogenous additions of charred, lipid-rich, material, like cork suberin or other plant-derived macromolecules (cutins). Principal component analysis was used to study the relationships between hydrophobicity with soil organic matter and its different components. Extractable organic matter (EOM) and specifically long chain FAs content were positively correlated to soil hydrophobicity. Therefore, the latter could be used as biomarkers surrogated to hydrophobicity in sandy soils.