Spatial patterns and environmental functions of dissolved organic matter in grassland soils of China

The study investigates how the composition and functions of soil dissolved organic matter (DOM) vary across China’s extensive grassland ecosystems and how environmental factors, particularly climate, drive these spatial patterns. DOM mediates energy and nutrient transfer among land, water, and atmosphere, undergoing mobilization, deposition, and microbial degradation that influence nutrient availability, microbial activity, and ecosystem carbon balance. Inputs of terrestrial DOM into fluvial systems alter water quality, affect treatment processes, and can lead to toxic and carcinogenic disinfection byproducts. Environmental drivers such as vegetation inputs, soil properties (pH, moisture, metals, clay), and climate (temperature, precipitation) regulate DOM fate. Despite grasslands covering ~40% of China and storing substantial terrestrial carbon, continental-scale assessments of grassland soil DOM geochemistry, environmental controls, and potential implications for ecosystem processes and human health are scarce. The research addresses two questions: (i) What are the compositional features and environmental functions of soil DOM in large-scale grassland ecosystems? (ii) How do environmental factors, especially climate, drive the spatial patterns of DOM at the continental scale?

Prior work shows terrestrial DOM substantially impacts aquatic systems and drinking water treatment by altering organic carbon levels and reactivity, leading to challenges in coagulation, adsorption, oxidation, and membrane processes, and forming potentially toxic disinfection byproducts (THMs, HAAs). Unique characteristics of natural organic matter in local environments have been linked to endemic diseases of unknown etiology (e.g., kidney disease). Soil and climatic factors (pH, moisture, metal oxides, clay, temperature, rainfall) influence DOM bioavailability and transformations. Grasslands hold large soil organic carbon stocks and DOC levels comparable to forests and croplands, yet previous studies in China focused mainly on regional scales. Fluorescent and molecular signatures of DOM (humic-like vs protein-like; molecular weight fractions) are known across ecosystems and relate to persistence and bioavailability. However, comprehensive, large-scale studies connecting DOM geochemistry to ecosystem carbon exchange and potential human health indicators, and identifying the key environmental drivers, have been limited.

Sampling: In May 2021, soils (0–20 cm) were collected from 89 natural grassland sites across 30 provinces in China prior to the wet season to avoid loss of mobile compounds. At each site, five cores were combined into a composite sample; sampling patterns adjusted for grassland extent. Samples were air-dried, ground, and sieved (2 mm). Regions were grouped into seven geographic zones; site positions were summarized along South–North and West–East orientations. DOM extraction and basic chemistry: Water-extractable DOM was obtained by shaking 50 g soil with 500 mL Milli-Q water (24 h, 25 °C), centrifugation (10,000×g), and filtration (0.7 µm GF/F). Extract pH, conductivity, and redox potential were measured. DOC and TDN were determined by TOC analyzer; inorganic N (NH4+, NOx−) by continuous flow analyzer; DON by difference. DBP formation tests: DOM solutions were adjusted to ~3 mg C/L, buffered at pH 8.0, and chlorinated with NaOCl at [Cl2] = 3×[DOC] + 7.6×[NH3]; incubated 72 h at 25 °C in the dark. Residual chlorine was quenched with ascorbic acid. THMs and HAAs were quantified by GC-ECD using US EPA Methods 551.1 and 552.3. DBP formation potential (DBP-FP; mg·kg−1 soil) was calculated relative to soil mass; specific DBP formation potential (SDBP; mg·g−1 DOC−1) normalized to DOC. Optical and compositional characterization: UV-vis spectra provided A254; 3D excitation–emission matrices (EEMs) were analyzed to compute HIX, BIX, FI. PARAFAC identified four fluorescent components (C1–C4), with model validation via OpenFluor comparison. Apparent molecular weight (MW) distributions were obtained by HPSEC and partitioned into VHMW (25–100 kDa), HMW (3.4–25 kDa), MMW (1.2–3.4 kDa), and LMW (<1.2 kDa) fractions. For 14 representative samples, FT-ICR MS (15 T, ESI) with SPE-PPL preconcentration assigned molecular formulas (C,H,O,N,S) to profile compound classes (e.g., lignin-, tannin-, condensed aromatic-, protein-/amino sugar-, lipid-like). Biodegradation incubations: Fourteen DOM samples (seven northern, seven southern) were inoculated with a microbial consortia prepared from fresh soils, incubated at 20 °C in the dark for 31 days with nutrient amendment (C:N:P 106:16:1) and quartz sand for microbial attachment. Biodegradation (%) of DOC, CDOM, FDOM was quantified; one- and two-phase association models estimated rate constants and labile/stable fractions. Environmental and response datasets: Site-specific climate (MAT, MAP, PET, solar radiation), vegetation (LAI, GPP, NPP), soil (SM, clay, pH, CEC, Na, K, Ca, SOC, TN, TP) variables, and ecosystem responses (NEP, Rs) were compiled from public datasets and averaged to annual means. County-level cancer incidence and mortality (2019 China Cancer Registry Annual Report) and provincial tap water TOC were used for association analyses. Statistics and modeling: Wilcoxon tests compared groups; Spearman correlations assessed bivariate relations; regression analyzed spatial gradients (South–North, West–East) and associations with ecosystem carbon and cancer cases. Variation partitioning quantified DOC/DON variance explained by MW and fluorescence fractions. PCA with PERMANOVA assessed regional DOM differences. Random forest models identified predictors of DOM quantity and composition. Partial least-squares path models (PLS-PM) integrated environmental drivers, DOM features, ecosystem carbon exchange, and cancer outcomes; significance via 999 bootstraps and model selection by GoF.

- Storage and composition: Four fluorescent components (C1–C4) were identified; humic-like components (C1–C3) and 1.2–25 kDa fractions dominated. Mean concentrations (mg·kg−1 soil): DOC 199.9 [15.3–804.5], DON 14.3 [0.4–68.4], THM-FP 33.7 [2.6–123.8], HAA-FP 4.3 [0.3–16.9]. DOM quantity correlated with HMW (3.4–25 kDa) and humic-like fractions (P<0.05). Fluorescence and size fractions together explained ~40% of DOC and DON variation. - Two compositional signatures: One signature enriched in humic-like and HMW (3.4–25 kDa) fractions, linked to lignin-, tannin-, and condensed aromatic-like compounds (H/C<1, O/C>0.5). The other enriched in protein-like (C4) and LMW (<1.2 kDa) fractions, associated with amino sugar- and lipid-like compounds (higher H/C, intermediate O/C). Humic-like abundance positively correlated with HMW and HIX; protein-like correlated with LMW/VHMW and BIX (all P<0.05). - Spatial patterns: Northern China exhibited higher HMW and humic-like fractions, higher HIX, and higher DOC, DON, and DBP-FP; southern China had higher LMW/MMW and protein-like (C4), with higher BIX/FI. SDBP (chlorine reactivity) for THMs and HAAs was notably higher in northwestern and Qinghai–Tibetan regions. DOM features showed a pronounced South–North divide and, for some variables (DON, SDBP), a West–East gradient (P<0.05). PCA/PERMANOVA indicated significant geochemical differences among regions (R2=0.28, P=0.001). - Biodegradability: DOM with higher humic-like and HMW fractions (mainly northern) had 21–35% lower biodegradability at day 31 (P<0.01) and >20% higher stable pools (P<0.05) than protein-like/LMW-enriched DOM (mainly southern). Humic-like/HMW and higher O/C with lower H/C were negatively correlated with biodegradation and positively with stable fractions (P<0.05). Protein-like/LMW and compounds with H/C>1 and O/C<0.5 showed the opposite. - Ecosystem carbon exchange: Humic-like and 3.4–25 kDa fractions were negatively associated with soil respiration (Rs) and net ecosystem productivity (NEP) (P<0.05), whereas <1.2 kDa and protein-like fractions were positively associated (P<0.01). DOM quantity was positively related to soil organic matter. - Human health associations: Age-standardized cancer incidence and mortality for nasopharynx, pancreas, and kidney/urinary organs varied with latitude, mirroring DOM spatial composition. Higher LMW and protein-like fractions correlated with greater nasopharyngeal cancer incidence and mortality (P<0.01). Higher HMW and humic-like fractions correlated with pancreatic and kidney/urinary organ cancer cases (P<0.05). THM-FP correlated positively with liver cancer; HAA-FP correlated with pancreas, kidney and urinary organs, and bladder cancers (P<0.05). Provincial tap water TOC correlated with soil DOM (P<0.05). - Environmental drivers and pathways: Random forests identified edaphic SOC and TN as key predictors for DOM quantity (DOC, DON, THM-FP, HAA-FP), while climate parameters primarily explained composition (LMW, HMW, C2, C4). PLS-PM showed climate and soil factors had negative effects on DOM quantity and quality; DOM fluorescence and MW had direct negative effects on Rs and NEP. Combined variables explained 15% (Rs) and 51% (NEP) of variance; for health outcomes, 55% (incidence) and 90% (mortality) were explained, with DOM quantity linked to mortality and fluorescence composition linked to both.

The study demonstrates that grassland soil DOM across China displays a clear South–North compositional divide governed by environmental drivers. Northern grasslands, with lower temperature/precipitation and higher mineral ions, accumulate more humic-like, high-MW, plant-derived compounds that are recalcitrant and less biodegradable, suppressing soil respiration and conserving carbon. Southern grasslands, with warmer, wetter climates and higher productivity and soil moisture, contain more labile, microbially derived proteinaceous and low-MW DOM that is rapidly mineralized, enhancing Rs and aligning with higher ecosystem productivity. These compositional contrasts translate into differential impacts on aquatic systems and drinking water: northern DOM, being more aromatic and persistent, exerts longer-lasting effects and has higher chlorine reactivity in certain regions, thus contributing more to DBP precursor loads; southern DOM, being more labile, promotes carbon outgassing and nutrient availability but has less persistent effects on water treatment. Statistically significant associations between DOM features and specific cancer incidences and mortalities suggest DOM geochemistry in grassland soils could serve as a geographical indicator of endemic cancers, potentially through direct exposure to DOM in source and drinking waters and via DBP formation or interactions with other pollutants. Environmental controls—climate, vegetation, and soil properties—jointly shape DOM quantity and composition, which in turn influence ecosystem carbon exchange and align with spatial cancer statistics. These findings answer the research questions by defining the key compositional signatures and environmental drivers of grassland soil DOM and illustrating their functional implications at a continental scale.

Continental-scale sampling and multi-method characterization reveal that China’s grassland soils store substantial DOM dominated by humic-like and HMW fractions in the north and protein-like and LMW fractions in the south. Composition governs biodegradability and ecosystem carbon exchange, with humic-like/HMW suppressing Rs and NEP and protein-like/LMW enhancing them. DOM geochemistry aligns with spatial patterns of DBP reactivity and exhibits significant statistical associations with specific cancer incidences and mortalities, indicating potential as a geographical indicator of endemic cancers. Environmental drivers, especially climate and soil properties, structure these patterns, and integrated path modeling quantifies their roles. Practically, management to increase plant-derived inputs and stabilize humic components (e.g., reducing overgrazing, vegetation restoration, enhancing clay-associated stabilization) can promote carbon sequestration, while targeted water treatment can mitigate health risks by removing harmful DOM fractions and DBP precursors. Future work should extend observations across seasons, incorporate microbial community dynamics, and mechanistically elucidate the pathways linking terrestrial DOM, drinking water chemistry, and health outcomes.

- The study is observational and relies on statistical associations; it does not establish causality between soil DOM and cancer outcomes. - Sampling was conducted only in the dry season; DOM quantity and composition can vary seasonally, so results reflect dry-season conditions. - Biotic factors, especially microbial community composition and activity, were not explicitly integrated across all sites, though they can influence DOM composition and transformation. - FT-ICR MS molecular characterization was performed on a subset (14 samples), potentially limiting representativeness of molecular-level inferences. - Potential confounders for health associations (e.g., socioeconomic, lifestyle, other environmental exposures) were not exhaustively controlled in ecological analyses.