Abstract:

Soil total organic carbon (TOC) is a fundamental component of terrestrial ecosystem carbon pools and plays a vital role in the global carbon cycle. In the context of global climate change, even marginal variations in soil carbon stocks can significantly influence atmospheric CO₂ concentrations, thereby exerting important feedbacks to the climate system. Consequently, the dynamic variation, storage mechanisms, and environmental regulation of soil total organic carbon have emerged as major research foci across the interconnected fields of ecology, soil science, and environmental science. The Qilian Mountains constitute a critical ecological barrier in the arid region of Northwest China, performing irreplaceable functions in maintaining regional ecological security, regulating water resources, and conserving biodiversity. Distinguished by pronounced altitudinal gradients and marked spatial heterogeneity in hydrothermal conditions, this region serves as an ideal natural laboratory for studying complex, high-altitude semi-arid mountain ecosystems. As a constructive keystone species within the forest ecosystem of the Qilian Mountains, Picea crassifolia Kom. plays a crucial role in sustaining both the structural stability and functional integrity of this ecologically significant region. Although multi-dimensional research has been conducted on the carbon sink function of the Qilian Mountains ecosystem, systematic studies remain insufficient with respect to the spatial variation patterns, key drivers, and underlying mechanisms of TOC along altitudinal gradients in P. crassifolia forests on the southern slopes. In particular, the spatial distribution characteristics of TOC across different elevation zones in P. crassifolia forests, the existence of consistent altitudinal patterns, and the primary environmental controls remain poorly understood. This knowledge gap hinders a mechanistic understanding of the soil carbon cycle in these alpine montane forests and impedes the development of science-based carbon management strategies for the region. In order to address these aforementioned knowledge gaps, this study was designed to quantify the spatial patterns and elucidate the underlying mechanisms governing TOC along an altitudinal gradient in P. crassifolia forests, aiming to identify the key environmental and edaphic controls on TOC distribution and to advance a mechanistic understanding of regional soil carbon cycling and providing a scientific foundation for assessing the vulnerability of alpine forest carbon stocks to climate-related elevational shifts. To achieve these objectives, this study focused on monospecific P. crassifolia forests on the southern slope of the Qilian Mountains, with sampling sites established across four altitudinal zones (2 900-3 000, 3 000-3 100, 3 100-3 200, and 3 200-3 300 m). At each site, we measured topographic variables, key soil physicochemical properties, and TOC content and density at various soil depths. Gradient analysis was employed to examine the spatial patterns of TOC along the altitudinal gradient and vertical soil profile. Relationships between TOC and environmental variables were then quantified using correlation and redundancy analyses to identify the primary drivers of soil total organic carbon distribution. The results revealed distinct altitudinal and vertical distribution patterns of TOC in the studied Picea crassifolia forests. Regarding altitudinal trends, an analysis of elevation-dependent patterns revealed that TOC content generally followed a unimodal distribution, with mean concentrations spanning from 59.28 g∙kg⁻¹ to 99.85 g∙kg⁻¹ across the gradient. In contrast, TOC density exhibited a distinctly different, concave (often described as trough-shaped) response to increasing elevation, with its mean values calculated to be between 27.41 kg∙m⁻² and 36.76 kg∙m⁻². Regarding layer-specific dynamics, the surface soil layer (0-10 cm) showed the strongest response, with both TOC content and density increasing progressively with elevation, demonstrating the high sensitivity of the surface carbon pool to such gradients; in contrast, TOC content in the subsurface layers consistently exhibited a hump-shaped pattern, characterized by an initial increase followed by a decline at higher elevations. In the 10-20 cm layer, TOC density exhibited a pronounced trough-shaped trend along the altitudinal gradient. Notably, within the 20-50 cm soil layer, TOC density exhibited a consistent triphasic response to increasing elevation, characterized by a pattern of decrease, followed by an increase, and then a subsequent decrease. With respect to vertical profile, across the entire studied elevational gradient, a clear vertical pattern emerged wherein TOC content exhibited a consistent decline with increasing depth, unequivocally demonstrating the phenomenon of surface soil layer enrichment. Correlation analysis identified significant bivariate relationships of TOC with key soil properties. TOC content showed strong positive correlations (p<0.01) with soil total nitrogen, soil saturated water content, and soil total porosity. Conversely, it exhibited strong negative correlations (p<0.01) with soil bulk density and soil specific gravity. TOC density displayed a significant positive correlation with elevation, a finding consistent with the established ecological principle of mountain vertical zonation. Regarding multivariate drivers, redundancy analysis (RDA) revealed that the spatial variation in TOC was jointly explained by soil properties and topographic factors. The key explanatory variables identified were soil total nitrogen, the soil carbon-to-nitrogen ratio, soil specific gravity, elevation, slope, pH, and soil bulk density. Of these, soil total nitrogen, the soil C/N ratio, and soil specific gravity were consistently identified as the dominant factors governing TOC across all soil layers. Most importantly, soil total nitrogen alone accounted for the largest proportion of variance explained in the RDA model, which underscores its role as the key limiting factor for soil carbon sequestration in these alpine forest ecosystems. In conclusion, this study demonstrates that the spatial distribution of TOC in these P. crassifolia forests is jointly governed by soil properties and topographic factors. Key determinants identified were soil total nitrogen, the soil carbon-to-nitrogen ratio, soil specific gravity, elevation, slope, pH, and soil bulk density, with soil total nitrogen emerging as the predominant driver. The influence of these factors exhibited significant vertical stratification across soil layers. Furthermore, elevation, as a composite environmental proxy, was shown to exert an indirect control on TOC spatial patterns primarily through its regulation of local hydrothermal conditions. Our findings advance a mechanistic understanding of how soil carbon stocks in high-altitude, semi-arid mountain ecosystems respond to altitudinal gradients—a critical consideration under ongoing climate change. Consequently, by elucidating the altitudinal distribution patterns and dominant controls of soil total organic carbon, this study provides an empirical scientific foundation for informing regional strategies in ecological management, sustainable development, and climate change adaptation in semi-arid mountain ecosystems.

Key words: soil total organic carbon, elevation gradient, vertical distribution, Qilian Mountains, Picea crassifolia Kom. Forests

摘要：

祁连山是中国西北内陆干旱区重要的生态屏障，其优势树种青海云杉（Picea crassifolia Kom.）林下土壤总有机碳沿海拔梯度的空间分异规律及关键驱动因子尚不明确。该研究旨在揭示上述问题，以深化对区域土壤碳循环过程的理解，并为评估高寒森林土壤碳库对海拔变化的响应能力提供科学依据；以祁连山南坡青海云杉林为研究对象，采用梯度格局法，结合相关性分析与冗余分析，系统解析土壤总有机碳沿海拔及土层的分布特征及其与环境因子的关系。结果表明，青海云杉林土壤总有机碳含量总体随海拔升高呈先增后减趋势，均值变化范围为59.3－99.9 g∙kg⁻¹。各海拔梯度内土壤总有机碳含量随土层加深而递减，0－10 cm土层总有机碳含量和密度均随海拔升高而增加，对海拔响应较显著，其余土层总有机碳含量则呈先增后减趋势。相关性分析和冗余分析表明，土壤总有机碳含量和全氮含量、饱和蓄水量及总孔隙度呈极显著正相关（p<0.01），与容重和土壤相对密度呈极显著负相关（p<0.01）；土壤总有机碳密度与海拔呈正相关。研究得出结论：土壤总有机碳空间分异受土壤属性和地形因素的共同调控，土壤全氮、碳氮比、相对密度、海拔、坡度、pH和容重为主要影响因素，其中全氮贡献率最高，且不同土层驱动因子对土壤总有机碳的主导作用存在垂直分异性。

关键词: 土壤总有机碳, 海拔梯度, 垂直分布, 祁连山, 青海云杉林