<b>Global pattern and drivers of soil </b><b>soluble</b><b> organic nitrogen</b>

Soil soluble organic nitrogen (SON) is a key hub in terrestrial nitrogen cycle, which can be converted to inorganic nitrogen by mineralization and formed from inorganic nitrogen by microbial metabolism. Meantime, SON is also one of the significant sources of nitrogen pollution contributing to eutrophication in water bodies within watersheds due to its high solubility and mobility. Despite its undeniable importance, the global distribution and drivers of SON are still a huge debate due to the scarcity of extensive data. To address this gap, a comprehensive global dataset of SON from 384 published studies was established, which encompassed 5782 subsoil samples (0-30 cm). Machine learning methods were employed to quantify the contributions of environmental factors to SON and subsequently predictions at the global scale. Overall, global SON concentrations varied significantly under various climate, ecosystems, vegetation types, and soil properties and ranged from 0.04 to 1034 (mean 41.36) mg kg^-1. The 82% variation of SON concentrations was explained by the 11 selected variables, among which altitude, bulk density, mean annual precipitation, and parent material were the primary factors. Additionally, the generated map of global SON concentrations revealed a latitudinal pattern in which SON concentrations gradually increased from low to high latitudes. The global SON stocks were estimated to be 2.4 Pg. The database is not only of great significance for understanding the nitrogen cycle in terrestrial ecosystems but also provides valuable data support for the improvement of biogeochemical models.

土壤可溶性有机氮（Soil soluble organic nitrogen, SON）是陆地氮循环的关键枢纽，可通过矿化作用转化为无机氮，也可由无机氮经微生物代谢生成。与此同时，SON凭借高溶解性与迁移性，亦是造成流域内水体富营养化的重要氮污染来源之一。尽管其重要性毋庸置疑，但由于大范围系统性数据的匮乏，全球SON的分布格局及其驱动因子仍存在较大争议。为填补这一研究空白，本研究整合384项已发表研究的数据，构建了一套全面的全球SON数据集，涵盖5782份0~30 cm土层的土壤样品。本研究采用机器学习方法量化了环境因子对SON的贡献度，并据此开展了全球尺度的SON浓度预测。整体而言，不同气候、生态系统、植被类型及土壤属性条件下的全球SON浓度存在显著差异，取值范围为0.04~1034 mg·kg⁻¹，平均值为41.36 mg·kg⁻¹。所筛选出的11个变量可解释82%的SON浓度变异，其中海拔、容重、年平均降水量以及土壤母质为主要驱动因子。此外，生成的全球SON浓度空间分布图揭示了纬度分布规律：SON浓度随纬度从低纬向高纬逐渐升高。全球SON总储量估算为2.4 Pg。该数据集不仅为理解陆地生态系统的氮循环提供了重要理论支撑，同时也为生物地球化学模型的优化提供了宝贵的数据支持。

搜集汇总

背景与挑战

背景概述

该数据集是一个全球土壤可溶性有机氮（SON）的综合数据集，基于384项已发表研究和5782个亚土壤样本（0-30厘米深度），使用机器学习方法分析SON的分布和驱动因素。SON浓度范围广泛（0.04-1034 mg kg-1，平均41.36 mg kg-1），主要受海拔、容重、年均降水量和母质等因素影响，并呈现纬度递增模式，全球SON储量估计为2.4 Pg，对理解氮循环和改进生物地球化学模型有重要意义。