Predicting and Mapping of Soil Organic Carbon Using Machine Learning Algorithms in Northern Iran

Mostafa Emadi,Amir Mosavi,Ruhollah Taghizadeh-Mehrjardi,Thomas Scholten,Ali Cherati,Majid Danesh

doi:10.3390/rs12142234

Abstract

Estimation of the soil organic carbon (SOC) content is of utmost importance in understanding the chemical, physical, and biological functions of the soil. This study proposes machine learning algorithms of support vector machines (SVM), artificial neural networks (ANN), regression tree, random forest (RF), extreme gradient boosting (XGBoost), and conventional deep neural network (DNN) for advancing prediction models of SOC. Models are trained with 1879 composite surface soil samples, and 105 auxiliary data as predictors. The genetic algorithm is used as a feature selection approach to identify effective variables. The results indicate that precipitation is the most important predictor driving 14.9% of SOC spatial variability followed by the normalized difference vegetation index (12.5%), day temperature index of moderate resolution imaging spectroradiometer (10.6%), multiresolution valley bottom flatness (8.7%) and land use (8.2%), respectively. Based on 10-fold cross-validation, the DNN model reported as a superior algorithm with the lowest prediction error and uncertainty. In terms of accuracy, DNN yielded a mean absolute error of 0.59%, a root mean squared error of 0.75%, a coefficient of determination of 0.65, and Lin’s concordance correlation coefficient of 0.83. The SOC content was the highest in udic soil moisture regime class with mean values of 3.71%, followed by the aquic (2.45%) and xeric (2.10%) classes, respectively. Soils in dense forestlands had the highest SOC contents, whereas soils of younger geological age and alluvial fans had lower SOC. The proposed DNN (hidden layers = 7, and size = 50) is a promising algorithm for handling large numbers of auxiliary data at a province-scale, and due to its flexible structure and the ability to extract more information from the auxiliary data surrounding the sampled observations, it had high accuracy for the prediction of the SOC base-line map and minimal uncertainty.

Highlights

Soil organic carbon (SOC) is central to soil health as it plays a significant role in soil aggregation, water holding capacity, cation/anion exchangeability, and nutrient availability, which promotes plant growth
Further analysis was performed on the log-transformed data; and the predicted SOC values were back-transformed to the original scale
The objective of this study was to determine a reliable algorithm for predicting the SOC contents in Mazandaran province through consideration of six different machine learning (ML) algorithms and using 105 environmental auxiliary variables derived from terrain attributes, remote sensing, and climatic data

Summary

Introduction

Soil organic carbon (SOC) is central to soil health as it plays a significant role in soil aggregation, water holding capacity, cation/anion exchangeability, and nutrient availability, which promotes plant growth. SOC can potentially affect both soil ecosystems and crop productivity due to its several critical roles in soil functioning. Changes in SOC pools induced by soil management and land cover changes affect global warming and, in turn, can significantly influence soil physical, chemical, and biological properties [4,5,6]. The spatial variability of SOC at the field to the regional scale is highly related to the soil forming-factors including the climate (precipitation and temperature); organisms (vegetation and human), relief (terrain attributes), parent materials, and time [8]

Objectives

Methods

Results

Conclusion