A Global Meta-Analysis of Soil Salinity Prediction Integrating Satellite Remote Sensing, Soil Sampling, and Machine Learning

Despite the growing interest among researchers, satellite-based prediction of soil salinity remains highly uncertain. The improvements in prediction accuracy reported in previous studies are usually limited to a single area. We performed a meta-analysis of regional satellite-based soil salinity predictions combined with in situ soil sampling and machine learning. Based on R² and root-mean-square error (RMSE) collected, we evaluated the effects of various features on the model accuracy and established a Bayesian network to evaluate the joint causal effect of multifeatures. Most significant differences were found in soil sampling schemes and characteristics of the study area, including the mean and variability (averaged R² of 0.75 for soil sample sets with lower salinity variation and 0.62 for others) of the salinity, climate type (R² of 0.64 in arid areas and 0.74 in others), soil texture (R² of 0.66 in sandy areas and 0.57 in others), and the interval between sampling date and satellite data acquisition date (R² of 0.53 under the condition of over 15 days and 0.65 in others). Generally, using different satellite data has limited effects on model performance among which Sentinel-2 performed better (R² = 0.72) than Landsat (R² = 0.66). The sampling of subsamples for each sample should focus on their subpixel-scale spatial heterogeneity across satellite data rather than the number of subsamples. It is also necessary to select appropriate vegetation and salinity indices for different satellite data under different vegetation conditions. Among algorithms, random forests (R² = 0.70) and support vector machines (R² = 0.71) performed best.