Scholarlink

<p>Soil microbial entropy (qMB) is the proportion of soil elements within microbial biomass, reflecting microbial resource support capacity and nutrient use efficiency. Its spatial pattern provides the basis for assessing soil ecological functions. However, existing studies are largely limited to local scales and lack comprehensive, high-resolution national predictions, with insufficient understanding of the multifactor-driven mechanisms and future dynamics under climate change. In this study, we collected microbial biomass carbon (MBC), nitrogen (MBN), and phosphorus (MBP) data from 1,288 published studies across China and calculated the corresponding qMBC, qMBN, and qMBP. We compared five machine learning models and selected the random forest model with the best predictive performance to map the spatial distribution of qMB. Subsequently, we explored the main drivers of its spatial variation and projected future trends. The results: (1) The mean values of qMBC, qMBN, and qMBP were 2.93%, 4.31%, and 3.58%, respectively. Overall, they followed a similar spatial distribution, each demonstrating considerable spatial heterogeneity. (2) Vegetation was the most important driver of spatial variation in qMB (qMBC, qMBN, and qMBP) and exerted a significant positive effect. (3) In the future, qMBC is anticipated to follow a generally increasing trend, whereas clear trends for qMBN and qMBP are not projected. This study achieves the high-precision, large-scale mapping of qMB. The findings highlight the pivotal role of vegetation in shaping its spatial differentiation, thus offering a key dataset for integrating microbial response mechanisms into global change ecology models, and providing theoretical insights for soil nutrient management.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Vegetation drives the spatial heterogeneity of soil microbial entropy in China: a national-scale machine learning mapping

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月15日

查看全文
  • Shihang Zhang,
  • Bin Yang,
  • Xing Guo,
  • Xiaobing Zhou,
  • Bo Zhu

摘要

Soil microbial entropy (qMB) is the proportion of soil elements within microbial biomass, reflecting microbial resource support capacity and nutrient use efficiency. Its spatial pattern provides the basis for assessing soil ecological functions. However, existing studies are largely limited to local scales and lack comprehensive, high-resolution national predictions, with insufficient understanding of the multifactor-driven mechanisms and future dynamics under climate change. In this study, we collected microbial biomass carbon (MBC), nitrogen (MBN), and phosphorus (MBP) data from 1,288 published studies across China and calculated the corresponding qMBC, qMBN, and qMBP. We compared five machine learning models and selected the random forest model with the best predictive performance to map the spatial distribution of qMB. Subsequently, we explored the main drivers of its spatial variation and projected future trends. The results: (1) The mean values of qMBC, qMBN, and qMBP were 2.93%, 4.31%, and 3.58%, respectively. Overall, they followed a similar spatial distribution, each demonstrating considerable spatial heterogeneity. (2) Vegetation was the most important driver of spatial variation in qMB (qMBC, qMBN, and qMBP) and exerted a significant positive effect. (3) In the future, qMBC is anticipated to follow a generally increasing trend, whereas clear trends for qMBN and qMBP are not projected. This study achieves the high-precision, large-scale mapping of qMB. The findings highlight the pivotal role of vegetation in shaping its spatial differentiation, thus offering a key dataset for integrating microbial response mechanisms into global change ecology models, and providing theoretical insights for soil nutrient management.