ScholarMate
English简体中文繁體中文
客服热线:400-1616-289
登录注册
期刊论文

Mapping field-scale soil organic carbon with unmanned aircraft system-acquired time series multispectral images

Guo, Long; Fu, Peng; Shi, Tiezhu; Chen, Yiyun; Zhang, Haitao; Meng, Ran; Wang, Shanqin*华中农业大学; 武汉大学
Soil and Tillage Research2020-2DOI: 10.1016/j.still.2019.104477
60被引16阅读
引用收藏
分享

摘要

Rapidly and accurately obtaining soil organic carbon (SOC) maps can help understand farmland soil fertility and serves managing mineral fertilizers. The pedogenesis theory provides a fundamental basis for digital soil mapping by environmental factors. However, these soil forming factors may be insensitive to soil properties in low relief areas, such as plains or small-scale farmlands. To overcome this problem, an ingenious method for soil mapping was proposed based on the relationships between vegetative growth and soil fertility. Six multispectral images of a winter oilseed rape (Brassica napus L.) field were captured from October 2017 to March 2018 every month by using an unmanned aircraft system. The spectral bands and three vegetation indices were used as predictors to reflect the growth of rape and to predict SOC. Two regression models, i.e., stepwise regression (STR) and partial least square regression (PLSR), and three machine learning algorithms, namely, back-propagation neural networks (BPNN), extreme learning machine (ELM), and support vector machine (SVM), were applied for digital SOC mapping. The results show that (1) SOC has strong relationships with the spectral bands and vegetation indices in different months; (2) the ratios of performance to interquartile range (RPIQ) of STR, PLSR, SVM, BPNN, and ELM were 2.218, 2.074, 2.168, 2.546, and 2.380 in predicting SOC; (3) the spatial characteristics of SOC were existed between and within fields, but only BPNN and ELM can highlight the spatial heterogeneity of SOC within fields. Therefore, time series multispectral images can provide covariates and overcome the influence of surface vegetation in predicting SOC at the field scale, whereas machine learning algorithms can further extract valuable information from complex variables and improve soil mapping accuracy.

学科与关键词

关键词
Digital soil mappingVegetation indicesMachine learning algorithmsSpatial variation
相关
加载中…

文献加载失败,请重试

重试

发表信息

论文状态
公开发表
期刊
Soil and Tillage Research
发表日期
2020-02
卷期
第 196 卷
页码
104477
DOI
10.1016/j.still.2019.104477
语言
ENGLISH
期刊收录
NSL.SCIE
期刊等级
1区
成果收录
SCIE
资助标注
National Natural Science Foundation of China [41371227]; National Key R&D Program of China [2017YFC0506200]; Basic Research Program of Shenzhen Science and Technology Innovation Committee [JCYJ20170302144323219]; Fundamental Research Funds for the Central Universities [2662016QD032]

作者

  • G

    Guo, Long (第一作者)

    huazhong agr univ

  • F

    Fu, Peng

查看全部 7 位作者收起作者
  • S

    Shi, Tiezhu

    shenzhen univ

  • C

    Chen, Yiyun

    wuhan univ

  • Z

    Zhang, Haitao

    huazhong agr univ

  • M

    Meng, Ran

    huazhong agr univ

  • W

    Wang, Shanqin (通讯)

    huazhong agr univ

登录解锁更多功能

  • 收藏这篇,随时在“我的 · 收藏”找回
  • 认领你的成果,生成推广素材
  • 关注作者获取最新成果
  • 加入项目组、协同写作
注册 / 登录

产品服务

  • 科研之友
  • 创新城
  • 科创云

服务支持

  • 帮助中心
  • 隐私政策
  • 服务条款

联系方式

在线客服:【】
客服热线:400-1616-289
电子邮箱:support@scholarmate.com

关注或下载科研之友

微信二维码
微信公众号
客户端下载二维码
下载客户端
科研成果科研人员科研机构科研动态爱瑞思软件

©2026 深圳市科研之友网络服务有限公司

公安备案图标粤公网安备 44030502000213
粤ICP备 16046710 号粤B2-20110417