Inversion of soil salinity and pH in farmland of the Hetao Plain based on Sentinel-2 and explainable machine learning.

ABSTRACTSalinization of soil is one of the most important environmental issues in arid and semi-arid areas. Accordingly, agricultural production and ecological development have been profoundly influenced in these regions. Therefore, it is becoming increasingly important to assess soil salinization and its driving factors. However, soil salinity is difficult to accurately characterize by using single-factor and linear models. Thus, it is necessary to develop a robust modeling technique by integrating multiple biophysical indicators to quantitatively monitor soil salinity. In this paper, the Support Vector Machine (SVM) regression algorithm and Artificial Neural Network (ANN) algorithm were employed to better estimate the soil salinity in the Yanqi Basin, Xinjiang, China. The soil backscattering coefficient (), Groundwater Depth (GD), Salinity Index (SI) and Surface Evapotranspiration (SET) were used as model parameters. was obtained from Sentinel-1A Synthetic Aperture Radar (SAR) data; GD and SI were calculated from Landsat-8 imagery; and SET was obtained from Moderate Resolution Imaging Spectroradiometer (MODIS) evapotranspiration product (MOD16). The performances of SVM and ANN in evaluating the nonlinear relationship between , GD, SI, SET, and soil Electrical Conductivity (EC) were compared. The results showed that, the SVM regression algorithm performs better than ANN algorithm in monitoring soil salinity. The Root Mean Square Error (RMSE) and the coefficient of determination (R2) in the estimation obtained using the SVM regression algorithm were about 2.01 and 0.82, respectively, versus the training data set; the RMSE and R2 were 1.36 and 0.88, respectively, versus the testing data set. The accuracy was significantly higher than that using the ANN algorithm, which obtained an RMSE of 2.20 and R2 of 0.79 versus the training data set, and 2.25 and 0.68 versus the testing data set. The results of this study indicated that about 56.82% of the soil in the study area was affected by different degrees of salinity. It is obvious that SVM regression algorithm has great potential for estimating soil salinity using multi-source remote sensing data.

Enhancing FRP-concrete interface bearing capacity prediction with explainable machine learning: A feature engineering approach and SHAP analysis

Soil salinization has seriously affected agricultural production and ecological balance in the Yellow River Delta region. Rapid and accurate monitoring of soil salinity has become an urgent need. Traditional machine learning models tend to fall into local optimal values during the learning process, which reduces their accuracy. This paper introduces Circle map to enhance the crayfish optimization algorithm (COA), which is then integrated with the regularized extreme learning machine (RELM) model, aiming to improve the accuracy of soil salinity content (SSC) inversion in the Yellow River Delta region. We employed Landsat5 TM remote sensing images and measured salinity data to develop spectral indices, such as the band index, salinity index, vegetation index, and comprehensive index, selecting the optimal modeling variable group through Pearson correlation analysis and variable projection importance analysis. The back propagation neural network (BPNN), RELM, and improved crayfish optimization algorithm–regularized extreme learning machine (ICOA-RELM) models were constructed using measured data and selected variable groups for SSC inversion. The results indicate that the ICOA-RELM model enhances the R2 value by an average of about 0.1 compared to other models, particularly those using groups of variables filtered by variable projection importance analysis as input variables, which showed the best inversion effect (test set R2 value of 0.75, MAE of 0.198, RMSE of 0.249). The SSC inversion results indicate a higher salinization degree in the coastal regions of the Yellow River Delta and a lower degree in the inland areas, with moderate saline soil and severe saline soil comprising 48.69% of the total area. These results are consistent with the actual sampling results, which verify the practicability of the model. This paper’s methods and findings introduce an innovative and practical tool for monitoring and managing salinized soils in the Yellow River Delta, offering significant theoretical and practical benefits.

Soil, a significant natural resource, plays a crucial role in supporting various ecosystems and serves as the foundation of Pakistan’s economy due to its primary use in agriculture. Hence, timely monitoring of soil type and salinity is essential. However, traditional methods for identifying soil types and detecting salinity are time-consuming, requiring expert intervention and extensive laboratory experiments. The objective of this study is to propose a model that leverages MODIS Terra data to identify soil types and detect soil salinity. To achieve this, 195 soil samples were collected from Lahore, Kot Addu, and Kohat, dating from October 2022 to November 2022. Simultaneously, spectral data of the same regions were obtained to spatially map soil types and salinity of bare land. The spectral reflectance of band values, salinity indices, and vegetation indices were utilized to classify the soil types and predict soil salinity. To perform the classification and regression tasks, the study employed three popular techniques in the research community: Random Forest (RF), Ada Boost (AB), and Gradient Boosting (GB), along with Decision Tree (DT), K-Nearest Neighbor (KNN), and Extra Tree (ET). A 70–30 test train validation split was used for the implementation of these techniques. The efficacy of the multi-class classification models for soil types was evaluated using accuracy, precision, recall, and f1-score. On the other hand, the regression models’ performances were evaluated and compared using R-squared (R), Mean Squared Error (MSE), Mean Absolute Error (MAE), and Root Mean Squared Error (RMSE). The results demonstrated that Random Forest outperformed other methods for both predicting soil types (accuracy = 65.38, precision = 0.60, recall = 0.57, and f1-score = 0.57) and predicting salinity (R = 0.90, MAE = 0.56, MSE = 0.98, RMSE = 0.97). Finally, the study designed a web portal to enable real-time prediction of soil types and salinity using these models. This web portal can be utilized by farmers and decision-makers to make informed decisions regarding soil, crop cultivation, and agricultural planning.

The timely and accurate monitoring of regional soil salinity is crucial for the sustainable development of land and the stability of the ecological environment in arid and semi-arid regions. However, due to the spatiotemporal heterogeneity of soil properties and environmental conditions, improving the accuracy of soil salinization monitoring remains challenging. This study aimed to explore whether partitioned modeling based on salinization degrees during both the bare soil and vegetation cover periods can enhance the accuracy of regional soil salinity prediction. Specifically, this study integrated in situ hyperspectral data and satellite multispectral data using spectral response functions. Subsequently, machine learning methods such as random forest (RF), extreme gradient boosting (XGBoost), support vector machine (SVM), and multiple linear regression (MLR) were employed, in combination with sensitive spectral indices, to develop a multi-source remote sensing soil salinity estimation model optimized for different salinization degrees (mild or lower salinization vs. moderate or higher salinization). The performance of this partitioned modeling approach was then compared with an overall modeling approach that does not distinguish between salinization degrees to determine the optimal modeling strategy. The results highlight the effectiveness of considering regional soil salinization degrees in enhancing the sensitivity of spectral indices to soil salinity and improving modeling accuracy. Classifying salinization degrees helps identify spectral variable combinations that are more sensitive to the construction of soil salinity content (SSC) models, positively impacting soil salinity estimation. The partitioned modeling strategy outperformed the overall modeling strategy in both accuracy and stability, with R2 values reaching 0.84 and 0.80 and corresponding RMSE values of 0.1646% and 0.1710% during the bare soil and vegetation cover periods, respectively. This study proposes an optimized modeling strategy based on regional salinization degrees, providing scientific evidence and technical support for the precise assessment and effective management of soil salinization.

Soil salinization is one of the primary factors contributing to land degradation in arid areas, severely restricting the sustainable development of agriculture and the economy. Satellite remote sensing is essential for real-time, large-scale soil salinity content (SSC) evaluation. However, some satellite images have low temporal resolution and are affected by weather conditions, leading to the absence of satellite images synchronized with ground observations. Additionally, some high-temporal-resolution satellite images have overly coarse spatial resolution compared to ground features. Therefore, the limitations of these spatiotemporal features may affect the accuracy of SSC evaluation. This study focuses on the arable land in the Manas River Basin, located in the arid areas of northwest China, to explore the potential of integrated spatiotemporal data fusion and deep learning algorithms for evaluating SSC. We used the flexible spatiotemporal data fusion (FSDAF) model to merge Landsat and MODIS images, obtaining satellite fused images synchronized with ground sampling times. Using support vector regression (SVR), random forest (RF), and convolutional neural network (CNN) models, we evaluated the differences in SSC evaluation results between synchronized and unsynchronized satellite images with ground sampling times. The results showed that the FSDAF model’s fused image was highly similar to the original image in spectral reflectance, with a coefficient of determination (R2) exceeding 0.8 and a root mean square error (RMSE) below 0.029. This model effectively compensates for the missing fine-resolution satellite images synchronized with ground sampling times. The optimal salinity indices for evaluating the SSC of arable land in arid areas are S3, S5, SI, SI1, SI3, SI4, and Int1. These indices show a high correlation with SSC based on both synchronized and unsynchronized satellite images with ground sampling times. SSC evaluation models based on synchronized satellite images with ground sampling times were more accurate than those based on unsynchronized images. This indicates that synchronizing satellite images with ground sampling times significantly impacts SSC evaluation accuracy. Among the three models, the CNN model demonstrates the highest predictive accuracy in SSC evaluation based on synchronized and unsynchronized satellite images with ground sampling times, indicating its significant potential in image prediction. The optimal evaluation scheme is the CNN model based on satellite image synchronized with ground sampling times, with an R2 of 0.767 and an RMSE of 1.677 g·kg−1. Therefore, we proposed a framework for integrated spatiotemporal data fusion and CNN algorithms for evaluating soil salinity, which improves the accuracy of soil salinity evaluation. The results provide a valuable reference for the real-time, rapid, and accurate evaluation of soil salinity of arable land in arid areas.

Effects of land use-land cover on soil water and salinity contents

Investigating the spatial distribution characteristics of the interaction between soil salinity and moisture is crucial in revealing moisture–salinity interaction in semi-arid farmland. The sampling of soil was performed on the second (S1), fifth (S2), eighth (S3), eleventh (S4), and fourteenth (S5) days after the erosive rainfall. The multifractal method was used to analyze spatial distribution parameters of soil moisture and salinity under the different stages. The findings showed that the soil moisture content decreased from 22.44% to 12.73%, while the salinity increased from 0.71 to 1.18 g kg–1 after the rainfall. As the amount of moisture in the soil decreased, the variability in the distribution of moisture initially increased from S1 to S3 and then decreased, while the salinity content also decreased. The spatial distribution of soil moisture and salinity content showed a strong correlation at S3 to S4 (with the relative water content of soil ranging from 0.52 to 0.75), indicating a significant coupling effect in these stages. However, the distribution of soil salinity was not uniform under high moisture content conditions (S1 to S2), as it was leached unevenly by rainfall, and under low moisture content conditions (S5), it precipitated, resulting in a low correlation between the spatial distribution of soil moisture and salinity content. This research has provided insight into the coupling dynamics of soil moisture and salinity content, revealing the mechanisms governing their spatial distribution in dryland agricultural regions.

Abstract. In regions with distinct seasons, soil salinity usually varies greatly by season. Thus, the seasonal dynamics of soil salinization must be monitored to prevent and control soil salinity hazards and to reduce ecological risk. This article took the Kenli District in the Yellow River delta (YRD) of China as the experimental area. Based on Landsat data from spring and autumn, improved vegetation indices (IVIs) were created and then applied to inversion modeling of the soil salinity content (SSC) by employing stepwise multiple linear regression, back propagation neural network and support vector machine methods. Finally, the optimal SSC model in each season was extracted, and the spatial distributions and seasonal dynamics of SSC within a year were analyzed. The results indicated that the SSC varied by season in the YRD, and the support vector machine method offered the best SSC inversion models for the precision of the calibration set (R2>0.72, RMSE < 6.34 g kg−1) and the validation set (R2>0.71, RMSE < 6.00 g kg−1 and RPD > 1.66). The best SSC inversion model for spring could be applied to the SSC inversion in winter (R2 of 0.66), and the best model for autumn could be applied to the SSC inversion in summer (R2 of 0.65). The SSC exhibited a gradual increasing trend from the southwest to northeast in the Kenli District. The SSC also underwent the following seasonal dynamics: soil salinity accumulated in spring, decreased in summer, increased in autumn and reached its peak at the end of winter. This work provides data support for the control of soil salinity hazards and utilization of saline–alkali soil in the YRD.

Abstract. Soil salinity, a significant environmental indicator, is considered one of the leading causes of land degradation, especially in arid and semi-arid regions. In many cases, this major threat leads to loss of arable land, reduces crop productivity, groundwater resources loss, increases economic costs for soil management, and ultimately increases the probability of soil erosion. Monitoring soil salinity distribution and degree of salinity and mapping the electrical conductivity (EC) using remote sensing techniques are crucial for land use management. Salt-effected soil is a predominant phenomenon in the Eshtehard Salt Lake located in Alborz, Iran. In this study, the potential of Sentinel-2 imagery was investigated for mapping and monitoring soil salinity. According to the satellite's pass, different salt properties were measured for 197 soil samples in the field data study. Therefore several spectral features, such as satellite band reflectance, salinity indices, and vegetation indices, were extracted from Sentinel-2 imagery. To build an optimum machine learning regression model for soil salinity estimation, three different regression models, including Gradient Boost Machine (GBM), Extreme Gradient Boost (XGBoost), and Random Forest (RF), were used. The XGBoostmethod outperformed GBM and RF with the coefficient of determination (R2) more than 76%, Root Mean Square Error (RMSE) about 0.84 dS m−1, and Normalized Root Mean Square Error (NRMSE) about 0.33 dS m−1. The results demonstrated that the integration of remote sensing data, field data, and using an appropriate machine learning model could provide high-precision salinity maps to monitor soil salinity as an environmental problem.

Texture features have been consistently overlooked in digital soil mapping, especially in soil salinization mapping. This study aims to clarify how to leverage texture information for monitoring soil salinization through remote sensing techniques. We propose a novel method for estimating soil salinity content (SSC) that combines spectral and texture information from unmanned aerial vehicle (UAV) images. Reflectance, spectral index, and one-dimensional (OD) texture features were extracted from UAV images. Building on the one-dimensional texture features, we constructed two-dimensional (TD) and three-dimensional (THD) texture indices. The technique of Recursive Feature Elimination (RFE) was used for feature selection. Models for soil salinity estimation were built using three distinct methodologies: Random Forest (RF), Partial Least Squares Regression (PLSR), and Convolutional Neural Network (CNN). Spatial distribution maps of soil salinity were then generated for each model. The effectiveness of the proposed method is confirmed through the utilization of 240 surface soil samples gathered from an arid region in northwest China, specifically in Xinjiang, characterized by sparse vegetation. Among all texture indices, TDTeI1 has the highest correlation with SSC (|r| = 0.86). After adding multidimensional texture information, the R2 of the RF model increased from 0.76 to 0.90, with an improvement of 18%. Among the three models, the RF model outperforms PLSR and CNN. The RF model, which combines spectral and texture information (SOTT), achieves an R2 of 0.90, RMSE of 5.13 g kg−1, and RPD of 3.12. Texture information contributes 44.8% to the soil salinity prediction, with the contributions of TD and THD texture indices of 19.3% and 20.2%, respectively. This study confirms the great potential of introducing texture information for monitoring soil salinity in arid and semi-arid regions.

Soil salinity is one of the most brutal environmental factors reducing the surface area and productivity of salt-sensitive cultivated plants. This major problem hurts land cover and agricultural productivity, principal to a decline in soil fertility and quality. The objective of this paper is to study the possibility of using an image of the spectral remote sensing data and field survey to sample soil and map salinity by correlating electrical conductivity (EC) field measurements with soil salinity indices derived from the Landsat7 satellite image. In Habra plain (North-western Algeria), the field’s electrical conductivity was measured during the period from 15 to 19 November 1999. These data were used as ground truth for the correlation analysis with different indices of image band values. Landsat 7 images were used to calculate soil salinity indices such as the Normalized Difference Vegetation Index (NDVI), the Normalized Difference Salinity Index (NDSI), the Soil Adjusted Vegetation Index (SAVI), and the Soil Salinity Index. A statistical analysis of the electrical conductivity (EC1:5 dS/m) and the environmental indices acquired from the Landsat 7 image was performed. Exponential regression was used to find the best indices, which were NDVI, NDSI, SAVI, and SI5, with a (NDVI R2=0.9, NDSI R2=0.9, SAVI R2=0.68, and SI5 R2=0.77) correlation with field truth data. A salinity map of the Habra plain was generated using this index with an acceptable level of accuracy. This study found that Landsat 7 images can effectively monitor soil salinity levels. The results of this study are relevant for agricultural operations, planners, and farmers by mapping and monitoring the soil salinity contamination. Finally, it is concluded that using remote sensing in salinity detection and mapping is highly significant.

Soil salinity content is monitored by embedding a distributed fiber optic sensor underneath soil and actively sensing an acoustic signal’s propagation properties. The system can distinguish soil salinity levels from 0 dS/m to 8 dS/m.

In order to explore the optimal remote sensing salinity monitoring index model for the inversion of soil salinization in the Alar reclamation area, based on the Sentinel-2 images and field measured data, the salinity index 1 (SI1), the normalized difference vegetation index in a green–red band (GRNDVI), the normalized vegetation index of greenness (GNDVI), and the normalized difference vegetation index (NDVI) were selected to construct the remote sensing-based salinization 1 detection index (S1DI) model. Next, the cotton field soil salinization information in the Alar reclamation area was extracted, and the accuracy of the model was verified to obtain the optimal remote sensing salinity monitoring index model. The results show that the overall classification accuracy of the S1DI1 (SI1-GRNDVI), S1DI2 (SI1-GNDVI), S1DI3 (SI1-NDVI), and S1DI4 (SI1-DVI) models for salinity monitoring is 83.35%, 83.10%, 82.96%, and 80.25%, respectively. The S1DI1 model is most suitable for retrieving the degree of soil salinization in the cotton field in the Alar reclamation area, and the S1DI2, S1DI3, and S1DI4 models are also very useful for monitoring soil salinization in the Alar reclamation area. Using the S1DI1 model to invert the soil salinization level of the cotton fields in the Alar reclamation area, it was found that the cotton field in the reclamation area is dominated by non-saline soil, and the light saline soil and moderate saline soil are mainly distributed in the 9th and 12th clusters of the reclamation area. As the S1DI1 model possesses the highest accuracy in extracting the soil salinization information of the cotton fields in the Alar reclamation area, it can be used as a remote sensing salinity 1 monitoring index model for the inversion of the soil salinization of the cotton fields in the reclamation area, which is expected to provide an effective reference value for soil salinization monitoring.

Soil salinization is one of the main causes of land degradation in arid and semi-arid areas. Timely and accurate monitoring of soil salinity in different areas is a prerequisite for amelioration. Hyperspectral technology has been widely used in soil salinity monitoring due to its high efficiency and rapidity. However, vegetation cover is an inevitable interference in the direct acquisition of soil spectra during crop growth period, which greatly limits the monitoring of soil salinity by remote sensing. Due to high soil salinity could lead to difficulty in plants' water absorption, and inhibit plant dry matter accumulation, a method for monitoring root zone soil salinity by combining vegetation canopy spectral information and crop aboveground growth parameters was proposed in this study. The canopy spectral information was acquired by a spectroradiometer, and then variable importance in projection (VIP), competitive adaptive reweighted sampling (CARS), and random frog algorithm (RFA) were used to extract the salinity spectral features in cotton canopy spectrum. The extracted features were then used to estimate root zone soil salinity in cotton field by combining with cotton plant height, aboveground biomass, and shoot water content. The results showed that there was a negative correlation between plant height/aboveground biomass/shoot water content and soil salinity in 0-20, 0-40, and 0-60 cm soil layers at different growth stages of cotton. Spectral feature selection by the three methods all improved the prediction accuracy of soil salinity, especially CARS. The prediction accuracy based on the combination of spectral features and cotton growth parameters was significantly higher than that based on only spectral features, with R2 increasing by 10.01%, 18.35%, and 29.90% for the 0-20, 0-40, and 0-60 cm soil layer, respectively. The model constructed based on the first derivative spectral preprocessing, spectral feature selection by CARS, cotton plant height, and shoot water content had the highest accuracy for each soil layer, with R2 of 0.715,0.769, and 0.742 for the 0-20, 0-40, 0-60 cm soil layer, respectively. Therefore, the method by combining cotton canopy hyperspectral data and plant growth parameters could significantly improve the prediction accuracy of root zone soil salinity under vegetation cover conditions. This is of great significance for the amelioration of saline soil in salinized farmlands arid areas.