European Journal of Soil Science最新文献

Precipitation intensity and nitrogen (N) deposition are projected to increase under global change scenarios, and both are expected to affect greenhouse gas (GHG) fluxes. However, the interactive effects of increasing precipitation intensity and N addition on GHG fluxes are still unknown. To address this gap, a mesocosm simulation experiment was conducted to investigate the individual and combined effects of changing precipitation intensity (with a constant event magnitude of 50 mm) and long-term N addition on GHG fluxes. The results revealed that precipitation application triggered a pulse effect on GHG fluxes, with increases up to 876% compared to pre-precipitation levels. The net changes in water-filled pore spaces (Δ WFPS) affected the temporal dynamics of GHG fluxes. Increasing precipitation intensity suppressed cumulative soil respiration, methane uptake, and nitrous oxide fluxes by directly reducing water availability (WFPS) and indirectly suppressing microbial biomass and substrate availability (dissolved organic carbon (DOC) or nitrate N content (NO₃⁻-N)). Furthermore, precipitation application altered the magnitude or direction of GHG flux responses to N addition. Changes in precipitation intensity and N addition had interactive effects on the Δ cumulative soil respiration and Δ cumulative N₂O fluxes, but not on Δ cumulative CH₄ fluxes. Increasing precipitation intensities decreased the Δ DOC content in the unfertilized treatment and increased Δ DOC content in the N addition treatment, thereby interactively affecting Δ cumulative soil respiration. N addition increased the Δ NO₃⁻-N content, influencing the response of Δ cumulative N₂O fluxes to increasing precipitation intensities. Our findings highlight that precipitation intensity regulates grassland GHG with N interactions, providing mechanistic insights to refine climate feedback predictions in ecosystems.

Risk assessment of per- and polyfluoroalkyl substances (PFASs) requires accurate data on their fate in the environment. Current soil studies are generally based on short-term adsorption tests in soil spiked with PFAS, with limited attention to long-term reactions after that spiking (ageing) or to differences in solid–liquid partitioning between spiked and field-contaminated soils (field to spike). This study addressed both effects with a focus on perfluorooctanoate (PFOA), thereby using carrier-free ¹⁴C-labelled PFOA to discriminate the spiked from the field-originating PFOA. Short-term (48 h) adsorption of trace ¹⁴C-labelled PFOA in soils suspended in 0.01 M CaCl₂ indicated linear sorption; the PFOA distribution (K_D) values ranged from 0.2 to 46 L kg⁻¹ (median 2.2 L kg⁻¹) in 91 soil samples and correlated (p < 0.001) mainly with soil organic carbon (r = +0.65). Three soils were incubated up to 6 months after PFOA spiking. The desorption K_D values were only 1.7–2.8-fold higher than 48 h adsorption K_D values; these factors increased by ageing but plateaued 2–4 months after spiking. Field-contaminated soils were collected (n = 21, 0.5–1100 μg PFOA kg⁻¹). The PFOA desorption K_D was almost zero in field-contaminated soils with continuous fresh deposition and in soils with exceptionally high total PFAS concentrations (21000–53,000 μg kg⁻¹), the latter suggesting the formation of micelles facilitating desorption. In most other soils, PFOA desorption K_D values were similar to or maximally 1.6 times higher than corresponding ¹⁴C-PFOA adsorption K_D values measured in the same soils. Data suggest that PFOA adsorption is generally reversible and that small PFOA ageing effects observed in laboratory conditions at trace PFOA levels do not even occur in field conditions.

Accurate estimation and projection of soil organic carbon (SOC) density is crucial for understanding the terrestrial carbon cycle and formulating carbon neutrality strategies. The increasing availability of SOC and related environmental data, coupled with advanced prediction models, has opened new opportunities for improving the accuracy of SOC (kg C m⁻²) predictions using data-driven methods. In this study, we developed a deep learning model TabTransformer_WT, by coupling the weighted mean squared error loss function with TabTransformer, to optimise estimation of surface (0–20 cm, SOC_0–20) and profile (0–100 cm, SOC_0–100) SOC in China. Using SOC observations and multi-source environmental covariates, we evaluated model performance through time-series-based 10-fold cross-validation across four periods (1979–1984, 2000–2004, 2005–2009 and 2010–2014) and compared it with machine learning and deep learning models (RF, SVR, CNN-1D, LSTM, RNN and TabTransformer). Our results indicate that TabTransformer_WT achieved the best prediction accuracy, with R² improvements of 8%–37% for SOC_0–20 and 6%–38% for SOC_0–100, and RMSE reductions of 0.31–1.07 and 0.99–2.39 kg C m⁻², respectively. We applied the model to evaluate historical and future spatiotemporal evolution of SOC_0–20 and SOC_0–100 in China. Historical analysis (1979–2023) showed China's soil acted as a carbon sink with annual growth rates of 45 Tg C year⁻¹ for surface and 33.37 Tg C year⁻¹ for profile SOC. Future projections using CMIP6 data revealed slow SOC accumulation under SSP1-1.9 but decreasing trends under SSP2-4.5 and SSP5-8.5 scenarios, with the 0–100 cm layer experiencing the greatest loss (−30.64 Tg C year⁻¹) under SSP5-8.5. This study provides a feasible method for large-scale SOC estimation and insights into SOC evolution under climate change.

Visible near-infrared reflectance spectroscopy (Vis–NIR) has been widely used in soil organic carbon (SOC) prediction due to its rapid, cost-effective, and non-destructive characteristics. Numerous soil spectral libraries have been used for SOC prediction. However, the growing volume of Vis–NIR spectral data has amplified its complexity, high dimensionality, and nonlinearity, creating significant challenges for traditional analytical models, particularly in terms of feature extraction, prediction accuracy, and generalisation capacity. To address these limitations, we developed a novel hybrid deep learning model that synergistically combines an enhanced convolutional neural network (CNN), a long short-term memory (LSTM) network, and an efficient channel attention (ECA) mechanism, termed the CNN-LSTM-ECA model. The CNN-LSTM-ECA model was evaluated using the LUCAS spectral library. Additionally, the SOC prediction performance of the CNN-LSTM-ECA model was compared against that of the CNN and CNN-LSTM models. To further assess the predictive performance of the model, spectral data specific to France were extracted from the library for validation. The results show that the CNN-LSTM-ECA model significantly outperforms the CNN and CNN-LSTM models in SOC content prediction. Specifically, the proposed model achieved remarkable prediction accuracy with an R² of 0.92 and an RMSE of 25.07 g kg⁻¹ on the validation, representing significant improvements of 10.72% and 7.15% in RMSE compared to the CNN (RMSE = 28.08 g kg⁻¹) and CNN-LSTM (RMSE = 27.00 g kg⁻¹) models, respectively. The model's generalisation capability was further confirmed through additional testing on the French dataset, where it maintained consistent predictive performance (R² = 0.93, RMSE = 24.83 g kg⁻¹). These findings underscore the model's high prediction accuracy and robust generalisation across diverse datasets. This study illustrates that the CNN-LSTM-ECA model significantly improves both accuracy and generalisation in SOC prediction, thereby providing a promising approach for spectral data analysis.

This study investigates the environmental and geochemical controls on forming and transforming acid sulfate (AS) soils along the southern Baltic Sea coast. Field surveys and laboratory analyses were conducted on a series of coastal soil transects located in hydrologically dynamic environments, including abrasive terraces/beaches, micro-cliffs/beach ridges, and organic-rich depressions. The results revealed a high site-specific variability in AS soil properties driven by topographic position, hydrological regime, and sedimentary history. Hypersulfidic materials, indicative of sulfide accumulation under reducing conditions, were found across all geomorphological settings. Geochemical indicators such as field pH, total organic carbon to total sulfur ratio, chloride, and calcium carbonate content proved effective in assessing the soil variability, including acidification potential. Magnetic susceptibility measurements indicated a predominantly natural origin of potentially toxic elements and the absence of technogenic contamination. However, under changing redox conditions, particularly in carbonate-poor soils, the mobilisation of toxic elements such as chromium, nickel, lead, and zinc cannot be excluded, despite their generally low concentrations. Organic matter, derived from both autochthonous and allochthonous sources, played a key role in sulfidisation processes, although the influence of its humification degree on acidification risk remains unclear. Overall, the study highlights the importance of localised environmental controls in AS soil development and provides a methodological framework for identifying similar systems in other coastal plains of the Baltic Sea.

The conversion of pure coniferous plantations to coniferous–broadleaf mixed forests increases the organic carbon (OC) content of soil and aggregates; however, the mechanisms of OC retention through soil aggregation remain inadequately understood. We selectively removed Fe oxides and OC from soil of both poorly aggregated (pure coniferous plantation) and well aggregated (mixed forest) soil systems. The mechanism of particulate organic carbon (POC) and mineral-associated organic carbon (MAOC) sequestration in Fe oxide soil aggregation under broadleaf transformation was studied. The removal of Fe oxides broke the macroaggregates into microaggregates and < silt + clay fractions and revealed the attachment and entanglement effects of plant residues encapsulated by macroaggregates on soil particles, whereas plant residue decomposition maximised the degree of macroaggregate fragmentation (64.8%–100%). These results indicate that POC self-isolates and that the presence of Fe oxides further enhances POC physical occlusion during soil aggregation. The extent of this physical protection provided by Fe oxides follows the order: free Fe (Fe_D) > amorphous Fe (Fe_O) > complex Fe (Fe_P). Specifically, Fe_O and Fe_P promote macroaggregate formation through organic–inorganic complexes (MAOC formation) to enhance POC physical occlusion, whereas Fe_D predominantly forms inorganic–inorganic complexes. Microaggregate formation and MAOC accumulation occurred simultaneously through organic–inorganic interactions with various Fe oxide forms. These processes enhanced soil aggregation and were accompanied by significant accumulation of POC (80.2%–169.8%) and MAOC (41.1%–137.3%) after stand conversion (p < 0.05). These findings indicate that improved soil aggregation capacity mediated by Fe oxides during forest conversion promotes POC and MAOC accumulation through distinct Fe oxide-specific aggregation mechanisms.