arrow
返回

Multivariate random forest for digital soil mapping

delete2023-03-01
delete36
delete
OA
AI
S
Stephan van der Westhuizen *
G
G.B.M. Heuvelink
D
David P. Hofmeyr
DOI:10.1016/j.geoderma.2023.116365delete
delete原文链接
delete分享
delete收藏
查看原文
摘要

摘要

En 中文
In digital soil mapping (DSM), soil maps are usually produced in a univariate manner, that is, each soil map is produced independently and therefore, when multiple soil properties are mapped the underlying dependence structure between these soil properties is ignored. This may lead to inconsistent predictions and simulations. For example, soil organic carbon (SOC) and total nitrogen (TN) maps produced independently may show unrealistic carbon-nitrogen (C:N) ratios. In the last decade the production of soil maps with machine learning models has become increasingly popular as these models are able to capture complex non-linear relationships between soil properties and environmental covariates. However, producing soil maps with multivariate machine learning models is still lacking and requires much investigation in DSM. In this paper we present the combined modelling of multiple soil properties with a multivariate random forest (MRF) model. We applied this model to mapping SOC and TN, and we compared it with results of two separate univariate random forest (RF) models. The comparison was done by means of stochastic simulations determined by sampling from the conditional distributions of the soil properties, given the covariates, as estimated by quantile regression forest. The results show that the MRF model is superior in terms of maintaining the dependence structure between SOC and TN, and consequently, is also able to produce more realistic C:N ratios. The models were also compared on the basis of prediction accuracy using commonly used accuracy metrics such as the root mean square error (RMSE). We found that the accuracy of the MRF model (RMSE-SOC = 40.04, RMSE-TN = 2.26, RMSE-CN = 3.58) is comparable to that of the univariate RF models (RMSE-SOC = 39.76, RMSE-TN = 2.26, RMSE-CN = 3.65). We performed the same comparisons between a regression co-kriging model and two separate regression kriging models, and made similar conclusions.
Keyword:
Digital soil mapping
Random forest
Stochastic simulation
Regression co-kriging
Soil organic carbon
AI总结

AI总结

对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。

期刊

Geoderma 封面图
Geoderma
IF:
6.6
论文数:
9.4K
被引数:
5.9W

机构

S
stellenbosch university
学者数:
610
论文数: 295
被引数: 6.8K
W
wageningen university research
学者数:
24
论文数: 8
被引数: 0
引用论文

引用论文

The impact of life expectancy on cost-effectiveness of treatment options for clinically localized prostate cancer
err2023-04-01
err0
PREAI
errAurash Naser-Tavakolian; Abhishek Venkataramana; Brennan Spiegel; Christopher Almario; Paul Kokorowski; Stephen J. Freedland; Jennifer T. Anger; John T. Leppert; Timothy J. Daskivich
err分享
err收藏
Geostatistical prediction and simulation of European soil property maps欧洲土壤性质图的地统计学预测和模拟
err2016-06-01
err32
PREAI
errHeuvelink, Gerard B. M.; Kros, Johannes; Reinds, Gert Jan; De Vries, Wim
err分享
err收藏
SoilGrids 2.0: producing soil information for the globe with quantified spatial uncertaintySoilGrids 2.0: 以量化的空间不确定性为全球生产土壤信息
errSOIL
IF4.3
err2021-06-14
err876
errOAAI
errPoggio, Laura; de Sousa, Luis M.; Batjes, Niels H.; Heuvelink, Gerard B. M.; Kempen, Bas; Ribeiro, Eloi; Rossiter, David
err分享
err收藏
Sampling design optimization for multivariate soil mapping
err2010-03-01
err84
PREAI
errVasat, R.; Heuvelink, G. B. M.; Boruvka, L.
err分享
err收藏
Using deep learning for digital soil mapping
errSOIL
IF4.3
err2019-02-26
err172
errOAAI
errPadarian, Jose; Minasny, Budiman; McBratney, Alex B.
err分享
err收藏
err分享
err收藏
Multi-source data integration for soil mapping using deep learning
errSOIL
IF4.3
err2019-03-22
err79
errOAAI
errWadoux, Alexandre M. J-C; Padarian, Jose; Minasny, Budiman
err分享
err收藏
学者 查看更多内容