Comparing Machine Learning Algorithms for Soil Salinity Mapping Using Topographic Factors and Sentinel-1/2 Data: A Case Study in the Yellow River Delta of China

Abstract

Soil salinization is a critical and global environmental problem. Effectively mapping and monitoring the spatial distribution of soil salinity is essential. The main aim of this work was to map soil salinity in Shandong Province located on the Yellow River Delta of China using Sentinel-1/2 remote sensing data and digital elevation model (DEM) data, coupled with soil sampling data, and combined with four regression models: support vector regression (SVR), stepwise multi-regression (SMR), partial least squares regression (PLSR) and random forest regression (RFR). For these purposes, 60 soil samples were collected during the field survey conducted from 9 to 14 October 2019, corresponding to the Sentinel-1/2 and DEM data. Then we established a soil salinity and feature dataset based on the sampled data and the features extracted from Sentinel-1/2 and DEM data. This study adopted the feature importance of the RF model to screen all features. The results showed that the CRSI index made the greatest contribution in retrieving soil salinity in this region. In this paper, 18 sampling points were used to validate and compare the performance of the four models. The results reveal that, compared with the other regression models, the PLSR model has the best performance (R2 = 0.66, and RMSE = 1.30). Finally, the PLSR method was used to predict the spatial distribution of soil salinity in the Yellow River Delta. We concluded that the model can be used effectively for the quantitative estimation of soil salinity and provides a useful tool for ecological construction.

Evaluating Drought Effects on Soil: Innovative Soil Salinity Monitoring via SAR Data, Sentinel-2 Imagery, and Machine Learning Algorithms in Kerkennah Archipelago

“…Various regression models were used, like polynomial, random forest, linear regression, and exponential regression. Similar recent studies (2023) have highlighted the efficacy of various regression algorithms, like random forest with R 2 = 0.80 [24] and the PLSR model with R 2 = 0.66 [46], for predicting soil salinity in semi-arid and arid regions. Notably, in our study, the exponential regression model, due to its accurate fitting R 2 = 0.75 and low root mean square error (RMSE) = 0.47 ds/m, was selected to predict soil salinity based on ground truth measurements and Sentinel-1 SAR data.…”

“…Approaches like machine learning algorithms show promise in modeling the relationship between remote sensing data and electrical conductivity (EC) by using simple linear regression [19], random forest regression [20], support vector machine [21] and other types of regression models [22]. Despite challenges, these statistical models have shown promising results in salinity estimation using Sentinel-1 images in arid and semi-arid regions [12,[23][24][25][26][27][28]. These advancements pave the way for improved soil salinity estimation and management.…”

A longitudinal analysis of soil salinity changes using remotely sensed imageries

“…Regression analysis is a commonly used statistical tool to study relationships between factors, making it straightforward to analyze multifactor data 54 . This study employs SVR 44 , 55 – 59 , RF 22 , 58 , 60 – 62 , DT 63 , 64 , and the XGBoost 16 , 65 – 68 methods to investigate the association between remotely sensed imagery data and field soil salinity. The Scikit-learn for Python (version 1.3.0) is used to implement these algorithms 69 .…”

Soil Salinity Mapping of Croplands in Arid Areas Based on the Soil–Land Inference Model

“…However, the simulation accuracy of machine learning and regression methods heavily relies on the number of samples available. In situations with a limited number of samples, these models tend to perform poorly [19,86]. On the other hand, the SoLIM is capable of acquiring soil-environment relationship knowledge and inferring soil properties at the watershed or even larger scales [24,87,88], while ensuring a certain level of inference accuracy with a limited number of samples [89][90][91].…”