Using Machine Learning Algorithms Based on GF-6 and Google Earth Engine to Predict and Map the Spatial Distribution of Soil Organic Matter Content

Zhishan Ye, Ziheng Sheng, Xiaoyan Liu, Youhua Ma, Ruochen Wang, Shiwei Ding, Mengqian Liu, Zijie Li and Qiang Wang
Additional contact information
Zhishan Ye: College of Resources and Environment, Anhui Agricultural University, Hefei 230036, China
Ziheng Sheng: College of Resources and Environment, Anhui Agricultural University, Hefei 230036, China
Xiaoyan Liu: College of Resources and Environment, Anhui Agricultural University, Hefei 230036, China
Youhua Ma: College of Resources and Environment, Anhui Agricultural University, Hefei 230036, China
Ruochen Wang: Heinz College of Information Systems and Public Policy, Carnegie Mellon University, Pittsburgh, PA 15213, USA
Shiwei Ding: College of Resources and Environment, Anhui Agricultural University, Hefei 230036, China
Mengqian Liu: School of Plant Protection, Anhui Agricultural University, Hefei 230036, China
Zijie Li: Realty Research Center, Nanjing Agricultural University, Nanjing 210095, China
Qiang Wang: College of Resources and Environment, Anhui Agricultural University, Hefei 230036, China

Sustainability, 2021, vol. 13, issue 24, 1-20

Abstract: The prediction of soil organic matter is important for measuring the soil’s environmental quality and the degree of degradation. In this study, we combined China’s GF-6 remote sensing data with the organic matter content data obtained from soil sampling points in the study area to predict soil organic matter content. To these data, we applied the random forest (RF), light gradient boosting machine (LightGBM), gradient boosting tree (GBDT), and extreme boosting machine (XGBoost) learning models. We used the coefficient of determination (R 2 ), root mean square error (RMSE), and mean absolute error (MAE) to evaluate the prediction model. The results showed that XGBoost (R 2 = 0.634), LightGBM (R 2 = 0.627), and GBDT (R 2 = 0.591) had better accuracy and faster computing time than that of RF (R 2 = 0.551) during training. The regression model established by the XGBoost algorithm on the feature-optimized anthrosols dataset had the best accuracy, with an R 2 of 0.771. The inversion of soil organic matter content based on GF-6 data combined with the XGBoost model has good application potential.

Keywords: geospatial modeling; machine learning; predictive mapping; remote sensing inversion; soil organic matter (search for similar items in EconPapers)
JEL-codes: O13 Q Q0 Q2 Q3 Q5 Q56 (search for similar items in EconPapers)
Date: 2021
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2071-1050/13/24/14055/pdf (application/pdf)
https://www.mdpi.com/2071-1050/13/24/14055/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jsusta:v:13:y:2021:i:24:p:14055-:d:706620

Access Statistics for this article

Sustainability is currently edited by Ms. Alexandra Wu

More articles in Sustainability from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().