Spatial prediction of soil calcium carbonate content based on Bayesian maximum entropy using environmental variables
Spatial prediction of soil calcium carbonate content based on Bayesian maximum entropy using environmental variables
复制标题
基于贝叶斯最大熵的环境变量土壤碳酸钙含量空间预测
DOI：
10.1007/s10705-021-10135-8
复制
发表时间：
2021-04
影响因子：
3.1
通讯作者：
Zhenxing Ma
中科院分区：
文献类型：
--
作者：
Mei Shan;Shuang Liang;Hongchen Fu;Xiaoli Li;Yu Teng;Jingwen Zhao;Yaxin Liu;Chen Cui;Li Chen;Hai Yu;Shunbang Yu;Yanling Sun;Jian Mao;Hui Zhang;Shuang Gao;Zhenxing Ma
土壤碳酸钙（CaCO 3）含量是土壤的重要性质。土壤CaCO 3含量的预测是土壤肥力可持续管理的必要条件。在这项工作中，我们试图将环境变量直接和通过回归模型到贝叶斯最大熵（BME）的框架来预测碳酸钙含量。首先，利用多元线性回归（MLR）和地理加权回归（GWR）建立了采样数据与环境变量（包括数字高程模型、pH值、温度、降雨量和潮土）之间的关系。MLR和GWR的预测结果作为软数据，并纳入BME的框架，估计CaCO 3的含量。其次，土壤样品和环境变量相结合，生成概率分布的CaCO 3在未采样点。这些概率分布被用作软数据的BME预测碳酸钙含量。结果表明，GWR方法（r = 0.84，RMSE = 24.0 g kg−1）优于MLR方法（r = 0.73，RMSE = 30.1 g kg−1）。BME-GWR方法优于BME-EV和BME-MLR方法。BME-GWR、BME-EV和BME-MLR方法的r值分别为0.87、0.86和0.82，三种方法的RMSE分别为22.2、23.9和25.2 g kg−1。各方法预测的CaCO 3含量空间分布相似，西南部显著高于东北部。
Soil calcium carbonate (CaCO3) content is an important soil property. The prediction of soil CaCO3content is necessary for the sustainable management of soil fertility. In this work, we attempted to incorporate environmental variables directly and through regression models into the framework of Bayesian maximum entropy (BME) to predict CaCO3content. Firstly, multiple linear regression (MLR) and geographically weighted regression (GWR) were used to establish a relationship between sampling data and environmental variables, including Digital Elevation Model, pH, temperature, rainfall, and fluvo-aquic soils. Prediction results of MLR and GWR served as soft data and were incorporated into the framework of BME to estimate the CaCO3content. Secondly, soil samples and environmental variables were combined to generate probability distributions of CaCO3at unsampled points. These probability distributions were used as soft data for the BME to predict the CaCO3content. The results showed that the GWR method (r = 0.84, RMSE = 24.0 g kg−1) performed better than the MLR method (r = 0.73, RMSE = 30.1 g kg−1). The BME-GWR method outperformed the BME-EV and BME-MLR methods. The r values of BME-GWR, BME-EV, and BME-MLR methods were 0.87, 0.86, and 0.82, respectively, and the RMSEs of the three methods were 22.2, 23.9, and 25.2 g kg−1, respectively. The spatial distribution of CaCO3content predicted by the above methods was similar and significantly higher in the southwest than in the northeast.
登录
查看更多内容
DOI：
10.1109/saintw.2003.1210142
发表时间：
2003-01
期刊：
2003 Symposium on Applications and the Internet Workshops, 2003. Proceedings.
影响因子：
--
作者：
I. Takahashi
通讯作者：
I. Takahashi
影响因子：
8.8
作者：
Sandeep Kumar
通讯作者：
Sandeep Kumar
影响因子：
6.1
作者：
Li, Yong
通讯作者：
Li, Yong
DOI：
--
发表时间：
2000-11
期刊：
--
影响因子：
--
作者：
G. Christakos
通讯作者：
G. Christakos
使用地形指数和分类变量作为辅助信息的土壤有机质空间预测
影响因子：
6.1
作者：
Zhang, Shiwen;Huang, Yuanfang;Shen, Chongyang;Ye, Huichun;Du, Yichun
通讯作者：
Du, Yichun