首页  |  本刊简介  |  编委会  |  投稿须知  |  订阅与联系  |  微信  |  出版道德声明  |  Ei收录本刊数据  |  封面
沂河流域土壤有机质空间分异及其影响因素
摘要点击 1347  全文点击 73  投稿时间:2024-05-22  修订日期:2024-08-09
查看HTML全文 查看全文  查看/发表评论  下载PDF阅读器
中文关键词  土壤有机质(SOM)  空间分布  影响因素  随机森林  沂河流域
英文关键词  soil organic matter (SOM)  spatial distribution  influencing factors  random forest  Yihe River Basin
作者单位E-mail
李英姿 山东师范大学地理与环境学院, 济南 250358 lyz102930@163.com 
李子君 山东师范大学地理与环境学院, 济南 250358 lizijun@sdnu.edu.cn 
郑树伟 山东师范大学地理与环境学院, 济南 250358  
林锦阔 山东师范大学地理与环境学院, 济南 250358  
杨舒媛 山东师范大学地理与环境学院, 济南 250358  
赵蕾 山东师范大学地理与环境学院, 济南 250358  
张新新 山东师范大学地理与环境学院, 济南 250358  
中文摘要
      准确预测土壤有机质(SOM)的空间分布对于提升耕地生产力、发展现代精准农业、调控土壤碳平衡以及保护生态环境至关重要. 有研究分析了不同区域SOM的水平空间分布特征及其影响因素,但鲜见研究从垂向角度阐述SOM空间分布及其驱动因子. 以北方土石山区典型区域——沂河流域为重点研究区,利用随机森林模型对流域SOM含量的空间分布进行预测,并揭示影响SOM空间分布差异的控制性因素. 结果表明:①所构建的4个土层深度SOM预测随机森林模型决定系数R2分别为0.761、0.651、0.672和0.727,拟合精度R2均达到了0.650以上,对流域SOM空间分布的预测精度较高. ②流域SOM含量的空间分布存在垂向递减特征,0~10、10~20、20~30和30~40 cm土层的ω(SOM)平均值分别为18.62、12.71、9.30和7.14 g·kg-1,属于四、五等级;不同土层SOM含量皆呈现东部和北部高,西部和西南部低的空间格局. ③自然环境因素尤其是土壤属性因素是影响流域SOM空间差异的主控因素,其中全氮(TN)对SOM的空间分布影响最大;人类活动加剧了SOM空间分布的差异化,ω(SOM)存在林地(15.92g·kg-1)>草地(15.81 g·kg-1)>园地(12.42 g·kg-1)>耕地(12.21 g·kg-1)的趋势. 研究结果可为我国北方土石山区土地资源合理利用及土壤质量提高提供科学参考.
英文摘要
      Accurate prediction of the spatial distribution of soil organic matter (SOM) is crucial for improving the productivity of arable land, developing modern precision agriculture, regulating soil carbon balance, and protecting the ecological environment. Previous research has analyzed the horizontal spatial distribution characteristics and influencing factors of SOM in different regions. Still, few studies have examined the spatial distribution of SOM and its drivers from a vertical perspective. Taking the Yihe River Basin, a typical area in the northern soil and rocky mountain region, as the key research area, the random forest model was used to predict the spatial distribution of SOM content in the basin and to reveal the controlling factors affecting the differences in the spatial distribution of SOM. The results showed that: ① The coefficients of determination R2 of the four constructed random forest models for predicting soil depth SOM were 0.761, 0.651, 0.672, and 0.727, respectively, and the fitting accuracies R2 reached greater than 0.650, with a high accuracy in predicting the spatial distribution of SOM in the catchment. ② The spatial distribution of SOM content in the River Basin was characterized by a vertical decrease, with the average values of ω (SOM) in the 0-10, 10-20, 20-30, and 30-40 cm soil layers being 18.62, 12.71, 9.30, and 7.14 g·kg-1, respectively, which belonged to the fourth and fifth classes. The SOM content in different soil layers showed the spatial pattern of high in the east and north and low in the west and southwest. ③ Natural environmental factors, especially soil property factors, were the main controlling factors affecting the spatial variation of SOM in the River Basin, with total nitrogen (TN) having the greatest influence on the spatial distribution of SOM. Human activities have intensified the differentiation of SOM spatial distribution, with a trend of forest land (15.92 g·kg-1)>grassland (15.81 g·kg-1)>orchard (12.42 g·kg-1)>cultivated land (12.21 g·kg-1) for ω (SOM). The study's results can provide scientific references for the rational use of land resources and the improvement of soil quality in northern China's soil and rocky mountain areas.

您是第83070495位访客
主办单位:中国科学院生态环境研究中心 单位地址:北京市海淀区双清路18号
电话:010-62941102 邮编:100085 E-mail: hjkx@rcees.ac.cn
本系统由北京勤云科技发展有限公司设计  京ICP备05002858号-2