Dynamics of organic matter molecular composition under aerobic decomposition and their response to the nitrogen addition in grassland soils

Highlights

• The decomposition of labile compounds is regulated by mineral association.

• The metabolism of recalcitrant molecules is controlled by biochemical preservation.

• Molecular composition of the persistent carbon pool is consistent across ecosystems.

• N addition inhibits cumulative respiration and the convergence of SOM composition.

Abstract

Grassland soils store a substantial proportion of the global soil carbon (C) stock. The transformation of C in grassland soils with respect to chemical composition and persistence strongly regulate the predicted terrestrial-atmosphere C flux in global C biogeochemical cycling models. In addition, increasing atmospheric nitrogen (N) deposition alters C chemistry in grassland soils. However, there remains controversy about the importance of mineralogical versus biochemical preservation of soil C, as well as uncertainty regarding how grassland soil C chemistry responds to elevated N. This study used grassland soils with diverse soil organic matter (SOM) chemistries in an 8-month aerobic incubation experiment to evaluate whether the chemical composition of SOM converged across sites over time, and how SOM persistence responded to the N addition. This study demonstrates that over the course of incubation, the richness of labile compounds decreased in soils with less ferrihydrite content, whereas labile compounds were more persistent in ferrihydrite rich soils. In contrast, we found that the richness of more complex compounds increased over the incubation in most sites, independent of soil mineralogy. Moreover, we demonstrate the extent to which the diverse chemical composition of SOM converged among sites in response to microbial decomposition. N fertilization decreased soil respiration and inhibited the convergence of molecular composition across ecosystems by altering N demand for microbial metabolism and chemical interactions between minerals and organic molecules. This study provides original evidence that the decomposition and metabolism of labile organic molecules were largely regulated by soil mineralogy (physicochemical preservation), while the metabolism of more complex organic molecules was controlled by substrate complexity (biochemical preservation) independent to mineral-organic interactions. This study advanced our understanding of the dynamic biogeochemical cycling of C by unveiling that N addition dampened C respiration and diminished the convergence of SOM chemistry across diverse grassland ecosystems.

Introduction

Grassland ecosystems are one of the largest terrestrial carbon (C) reservoirs, covering approximately one-third of the global terrestrial surface and storing ~20% of the global soil C stock (Conant et al., 2001; Scurlock and Hall, 1998). Increased atmospheric nitrogen (N) deposition caused by anthropogenic activities has a critical impact on ecosystem structure and function at the global scale, including on soil C cycling in grasslands (Bai et al., 2010; Gruber and Galloway, 2008). The decomposition and persistence of organic matter (OM) (see Table 1) in grassland soil and how they respond to elevated N inputs have the potential to significantly influence the global C cycle (Scurlock and Hall, 1998).

Decomposition of soil organic matter (SOM) is mainly mediated by microbial processes that rely on extracellular enzymes to break down organic polymers into oligomers and monomers (Garcia-Pausas and Paterson, 2011; Schimel and Schaeffer, 2012; Zhou et al., 2012a). ‘Selective preservation’ of SOM by microbial processes proposed that the labile C pool, including proteins, amino sugars, and carbohydrates of plant and microbial origin, is depleted over time (Feng et al., 2005). What remains is a suite of recalcitrant organic compounds, including lignin, tannin, and condensed aromatic C, that require more energy for microorganisms to decompose (Lehmann and Kleber, 2015; Lützow et al., 2006; Sollins et al., 1996). Such ‘selective preservation’ theory suggests that the microbial community regulates C decomposition, thus determining OM transformations in soil. In contrast, some studies demonstrated the decomposition of recalcitrant compounds could be more rapid than the decomposition of labile compounds as labile compounds could be associated with minerals to become chemically inaccessible from microbial decomposition (Klotzbücher et al., 2011; Lehmann and Kleber, 2015; Lützow et al., 2006).

The soil matrix plays a crucial role in the physicochemical persistence of SOM (Cotrufo et al., 2013; Six et al., 2002). Mineral-associated organic matter (MAOM) is a dominant form of relatively stable SOM, contributing up to 72% of total C in soils and sediments (Fang et al., 2019; Lalonde et al., 2012; Lehmann and Kleber, 2015; Lützow et al., 2006; Wagai and Mayer, 2007; Zhao et al., 2016). The multi-layer zonal structure model of mineral-OM complexes proposed that molecules enriched in hydrophilic C functional groups, fatty acids, and aromatic-rich molecules preferentially persisted to mineral surfaces, the hydrophobic zone, and the kinetic zone, respectively (Kleber et al., 2007). Our empirical study supports that protein- and lipid-like molecules preferentially accumulate in the hydrophobic zone and lignin-like molecules persist in the kinetic zone of the mineral-OM complexes from soil with more ferrihydrite content (Zhao et al., 2020). Thus, these molecular classes are inherently persistent because of their chemical interactions with mineral surfaces (Bahureksa et al., 2021; Kleber et al., 2007; Zhao et al., 2020).

Elevated N inputs to soil has shown significant impacts on soil C decomposition and geochemical interactions (Chen et al., 2020; Huang et al., 2020a; Riggs et al., 2015; Zhang et al., 2014; Zhao et al., 2020). Previous studies found that experimental N addition has negative effects on the decomposition of SOM (Chen et al., 2020; Riggs et al., 2015; Tan et al., 2017). The decrease of SOM decomposition by N addition is likely caused by decreased lignin-degrading enzymes (Fog, 1988; Ramirez et al., 2012) and reduced microbial biomass (Lu et al., 2011; Treseder, 2008). Decreased C decomposition in response to N addition may not always result in the increase of soil C storage, given the large variation of soil C pools among different ecosystems. For instance, Lu et al. (2011) found that N addition resulted in minor changes in soil C storage from forest and grassland ecosystems, but a significant increase in agricultural soils because of relatively large, new C inputs from aboveground production in agricultural ecosystems. Moreover, N addition was found to increase the new C pool but have no impact or decrease old C in soil (Huang et al., 2020a; Hobbie et al., 2012). Thus, elevated N inputs have the potential to influence soil C storage depending on the SOM chemical composition. Meanwhile, previous studies have demonstrated that N addition decreased soil pH across multiple ecosystems (Chen et al., 2020; Riggs et al., 2015; Zhou et al., 2017; Zhao et al., 2020). N addition led to the accumulation of hydrogen ions (H⁺) and nitrate (NO³⁻) due to stimulated nitrification. Consequently, N addition-induced soil acidification potentially may decrease microbial decomposition by suppressing microbial growth (Rousk et al., 2009; Treseder, 2008) and altering microbial community composition (Rousk et al., 2010; Zhou et al., 2017). Soil acidification may also alter chemical interactions at the interface between minerals and OM (Yu et al., 2013; Zhao et al., 2020). Soils acidification resulted in more positive charges on both proteins and ferrihydrite surfaces (Yu et al., 2013), causing the loss of proteins because of disturbed electrostatic interactions. Soil acidification also causes the leaching of base cations (Ca²⁺ or Mg²⁺) by disrupting cation bridging of OM on mineral surfaces (Rowley et al., 2018; Yu et al., 2017). Moreover, decreased pH increases the prevalence of ligand exchange on mineral surfaces by releasing H⁺, resulting in more associations of OM (Kleber et al., 2005; Kögel-Knabner et al., 2008; Lützow et al., 2006). Therefore, N addition disrupted mineral-OM associations through altering chemical interactions, such as electrostatic interactions, H-bonds, and cation bridging (Bahureksa et al., 2021; Kleber et al., 2007; Zhao et al., 2020).

Building off these findings, a wide range of chemical composition of SOM from diverse grassland soils can persist by both chemical recalcitrance and mineral associations, resulting in a predictable pattern of C accumulation, by which grassland soils with high recalcitrant C content and/or short-range order minerals (also referred to as poorly crystalline minerals) accumulate more C. There is still a lack of information on a unifying mechanism of SOM persistence applicable across grassland ecosystems during microbial decomposition processes, and how the SOM persistence mechanism responds to the nutrient addition.

This study aims to test the extent to which the accumulation of molecules reflects persistence of compound classes due to chemical recalcitrance and mineral-OM associations. We hypothesized that in the absence of new C inputs, microbial decomposition of SOM will significantly decrease the richness of labile organic molecules that are bioavailable and vulnerable to decomposition, but increase the richness of organic compounds that are either chemically recalcitrant for microbial metabolism or inaccessible due to the association with minerals through depolymerization of recalcitrant compounds and sorption to mineral surfaces. In addition, we hypothesized that mechanisms for SOM persistence, by both chemical recalcitrance and mineral associations, are consistent across soils, resulting in a convergence of SOM chemistry across soils over the course of the incubation. Moreover, N fertilization will inhibit the molecular compositional convergence of SOM across soils during the incubation due to the reduction of microbial decomposition and the interference of organo-mineral associations. We expect that the reduced microbial decomposition will result from inhibited microbial growth and respiration induced by the N fertilization; meanwhile, the disrupted chemical interactions, such as electrostatic interactions, H-bonds, and cation bridging, in N-fertilized soils will interfere with the formation of organo-mineral associations.