欢迎访问《生态学杂志》官方网站,今天是 分享到:

生态学杂志 ›› 2021, Vol. 40 ›› Issue (8): 2341-2347.doi: 10.13292/j.1000-4890.202108.011

• 农田地质高背景重金属污染专栏 • 上一篇    下一篇

区域土壤-水稻籽粒镉耦合关系模型的构建和验证

陈家乐1,唐林茜1,相满城1,张春华2,葛滢1*,陈效民1   

  1. 1南京农业大学资源与环境科学学院, 南京 210095;2南京农业大学生命科学实验中心, 南京 210095)
  • 出版日期:2021-08-10 发布日期:2021-08-12

Model constructions and validations for regional cadmium coupling relationships in soil rice grain.

CHEN Jia-le1, TANG Lin-xi1, XIANG Man-cheng1, ZHANG Chun-hua2, GE Ying1*, CHEN Xiao-min1   

  1. (1College of Resources and Environmental Sciences, Nanjing Agricultural University, Nanjing 210095, China; 2Laboratory Centre of Life Science, Nanjing Agricultural University, Nanjing 210095, China).
  • Online:2021-08-10 Published:2021-08-12

摘要: 为构建区域土壤-水稻籽粒镉(Cd)耦合关系模型,通过文献调研获取369组数据,采用广义线性回归(GLM)、梯度提升机器(GBM)、随机森林(RF)和Cubist等方法,以文献共有的土壤pH和总Cd含量(Soil_Cd)为自变量、水稻籽粒Cd含量(Grain_Cd)为因变量构建模型,并以实测土壤pH、Soil_Cd和Grain_Cd数据验证,分析比较不同模型的预测能力。结果表明:GLM、GBM、RF和Cubist模型的性能接近,其决定系数R2都在0.5左右,但RF模型对实测数据的拟合效果最好(R2=0.534)。因此,基于机器学习的RF模型能在区域尺度合理预测稻米Cd含量。

关键词: 土壤, 水稻, 镉, 耦合关系模型, 机器学习

Abstract: In order to establish regional models for describing soilrice grain cadmium (Cd) coupling relationship, we collected 369 groups of data from literature to construct models, which used generalized linear model (GLM), gradient boosting machine (GBM), random forest (RF) and Cubist methods, with soil pH and total Cd content (Soil_Cd) as independent variables and Cd content in rice grains (Grain_Cd) as the dependent variable. The robustness of those models in the Grain_Cd predictions was evaluated using the measured data of soil pH, Soil_Cd and Grain_Cd. Results showed that GLM, GBM, RF and Cubist models showed similar performance, with all of their coefficients of determination (R2) being around 0.5. The measured Grain_Cd values were best matched to the prediction of the RF model (R2=0.534). Therefore, the RF model, which is based on machine learning, was capable to reasonably predict Cd content in rice grains at the regional scale.

Key words: soil, rice, Cd, coupling relationship model, machine learning.