当前位置: X-MOL 学术Remote Sens. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Improving the Spatial Prediction of Soil Organic Carbon Content in Two Contrasting Climatic Regions by Stacking Machine Learning Models and Rescanning Covariate Space
Remote Sensing ( IF 4.2 ) Pub Date : 2020-03-29 , DOI: 10.3390/rs12071095
Ruhollah Taghizadeh-Mehrjardi , Karsten Schmidt , Alireza Amirian-Chakan , Tobias Rentschler , Mojtaba Zeraatpisheh , Fereydoon Sarmadian , Roozbeh Valavi , Naser Davatgar , Thorsten Behrens , Thomas Scholten

Understanding the spatial distribution of soil organic carbon (SOC) content over different climatic regions will enhance our knowledge of carbon gains and losses due to climatic change. However, little is known about the SOC content in the contrasting arid and sub-humid regions of Iran, whose complex SOC–landscape relationships pose a challenge to spatial analysis. Machine learning (ML) models with a digital soil mapping framework can solve such complex relationships. Current research focusses on ensemble ML models to increase the accuracy of prediction. The usual ensemble method is boosting or weighted averaging. This study proposes a novel ensemble technique: the stacking of multiple ML models through a meta-learning model. In addition, we tested the ensemble through rescanning the covariate space to maximize the prediction accuracy. We first applied six state-of-the-art ML models (i.e., Cubist, random forests (RF), extreme gradient boosting (XGBoost), classical artificial neural network models (ANN), neural network ensemble based on model averaging (AvNNet), and deep learning neural networks (DNN)) to predict and map the spatial distribution of SOC content at six soil depth intervals for both regions. In addition, the stacking of multiple ML models through a meta-learning model with/without rescanning the covariate space were tested and applied to maximize the prediction accuracy. Out of six ML models, the DNN resulted in the best modeling accuracies, followed by RF, XGBoost, AvNNet, ANN, and Cubist. Importantly, the stacking of models indicated a significant improvement in the prediction of SOC content, especially when combined with rescanning the covariate space. For instance, the RMSE values for SOC content prediction of the upper 0–5 cm of the soil profiles of the arid site and the sub-humid site by the Remote Sens. 2020, 12, 1095; doi:10.3390/rs12071095 www.mdpi.com/journal/remotesensing Remote Sens. 2020, 12, 1095 2 of 26 proposed stacking approaches were 17% and 9% respectively, less than that obtained by the DNN models—the best individual model. This indicates that rescanning the original covariate space by a meta-learning model can extract more information and improve the SOC content prediction accuracy. Overall, our results suggest that the stacking of diverse sets of models could be used to more accurately estimate the spatial distribution of SOC content in different climatic regions.

中文翻译:

通过叠加机器学习模型和重新扫描协变量空间,改善两个不同气候区土壤有机碳含量的空间预测

了解不同气候区域的土壤有机碳(SOC)含量的空间分布将增强我们对由于气候变化而导致的碳得失的认识。但是,对伊朗干旱和半湿润地区对比之下的SOC含量知之甚少,其复杂的SOC与景观关系对空间分析构成了挑战。具有数字土壤映射框架的机器学习(ML)模型可以解决这种复杂的关系。当前的研究集中在整体机器学习模型上,以提高预测的准确性。通常的合奏方法是增强或加权平均。这项研究提出了一种新颖的集成技术:通过元学习模型来堆叠多个ML模型。此外,我们通过重新扫描协变量空间以最大化预测准确性来测试集合。我们首先应用了六个最新的ML模型(即立体主义,随机森林(RF),极端梯度增强(XGBoost),经典人工神经网络模型(ANN),基于模型平均的神经网络集成(AvNNet)以及深度学习神经网络(DNN)来预测和绘制两个区域在六个土壤深度间隔的SOC含量的空间分布。此外,测试了通过元学习模型(带有/不带有重新扫描协变量空间)的多个ML模型的堆叠,并将其应用于最大化预测准确性。在六个ML模型中,DNN具有最佳的建模精度,其次是RF,XGBoost,AvNNet,ANN和Cubist。重要的是,模型的堆叠表明对SOC含量的预测有了显着改善,尤其是与重新扫描协变量空间结合使用时。例如,通过遥感,干旱地区和半湿润地区的土壤剖面的上0-5 cm的SOC含量预测的RMSE值,2020,12,1095;doi:10.3390 / rs12071095 www.mdpi.com/journal/remotesensing Remote Sens。2020,12,1095提议的26种堆叠方法中有2种分别为17%和9%,比DNN模型获得的最佳个体模型要少。这表明通过元学习模型重新扫描原始协变量空间可以提取更多信息,并提高SOC含量预测的准确性。总体而言,我们的结果表明,可以使用不同模型集的堆叠来更准确地估计不同气候区域中SOC含量的空间分布。利用遥感,2020年,12、1095年预测干旱地区和半湿润地区土壤剖面的上部0-5 cm的SOC含量的RMSE值;doi:10.3390 / rs12071095 www.mdpi.com/journal/remotesensing Remote Sens。2020,12,1095提议的26种堆叠方法中有2种分别为17%和9%,比DNN模型获得的最佳个体模型要少。这表明通过元学习模型重新扫描原始协变量空间可以提取更多信息,并提高SOC含量预测的准确性。总体而言,我们的结果表明,可以使用不同模型集的堆叠来更准确地估计不同气候区域中SOC含量的空间分布。利用遥感,2020年,12、1095年预测干旱地区和半湿润地区土壤剖面的上部0-5 cm的SOC含量的RMSE值;doi:10.3390 / rs12071095 www.mdpi.com/journal/remotesensing Remote Sens。2020,12,1095提议的26种堆叠方法中有2种分别为17%和9%,比DNN模型获得的最佳个体模型要少。这表明通过元学习模型重新扫描原始协变量空间可以提取更多信息,并提高SOC含量预测的准确性。总体而言,我们的结果表明,可以使用不同模型集的堆叠来更准确地估计不同气候区域中SOC含量的空间分布。提议的26种堆叠方法中的1095 2种分别为17%和9%,比DNN模型(最佳个体模型)获得的结果要少。这表明通过元学习模型重新扫描原始协变量空间可以提取更多信息,并提高SOC含量预测的准确性。总体而言,我们的结果表明,可以使用不同模型集的堆叠来更准确地估计不同气候区域中SOC含量的空间分布。提议的26种堆叠方法中的1095 2种分别为17%和9%,比DNN模型(最佳个体模型)获得的结果要少。这表明通过元学习模型重新扫描原始协变量空间可以提取更多信息,并提高SOC含量预测的准确性。总体而言,我们的结果表明,可以使用不同模型集的堆叠来更准确地估计不同气候区域中SOC含量的空间分布。
更新日期:2020-03-29
down
wechat
bug