Riparian buffers are important for several services such as providing high water quality, nutrient recycling, and buffering agricultural production. Accordingly, in this research, we used machine learning to predict soil properties in the forests.
Sixteen models were used for describing the PSD. Results indicated that the models had acceptable accuracy for describing the PSD curves. Importantly, the Jaky model, with only one fitting parameter was able to accurately describe PSD. Therefore, the parameter P in the Jaky model was used for spatial mapping of the PSD curve using DSM techniques.
We use five machine learning techniques-Random Forest, M5P, Reduced Error Pruning Tree, Gaussian Processes, and Pace Regression-under two scenarios to predict soil erodibility. All five algorithms show a positive correlation between the soil erodibility factor and silt, sand, fine sand, bulk density, and infiltration.
Vis-NIR spectroscopic data was combined with DEM-derived topographic data and remote sensing data using an RF model hybridized with a particle swarm optimization algorithm for predicting the spatial variability of soil surficial clay contents, EC, and CCE for the agriculturally-intensive region of Kurdistan, Iran.
A new approach for extrapolating soil information from reference to target areas is proposed in the current research. We evaluated the ability of a semi-supervised learning approach compared to a supervised learning approach for extrapolating soil classes in two areas.
we tested the predictive power of three different ML models, the random forest, the support vector machine, and extreme gradient boosting coupled with the remote sensing data covariates to estimate soil texture data in Germany.
This study tested and evaluated a suite of nine individual base learners and seven model averaging techniques for predicting the spatial distribution of soil properties. Based on the nested-cross validation approach, the results showed that the all seven model averaging techniques performed better than the base learners.
A synthetic resampling approach using the synthetic minority oversampling technique was employed to make a balanced dataset from the original data. Bioclimatic and remotely sensed data, distance, and terrain attributes were used as environmental covariates to model and map soil textural classes.
This study demonstrated the effectiveness of the RF model for predicting the spatiotemporal patterns of SOC content of the oasis and arid-agroecosystem area which the approach may be utilized in other similarly arid conditions. In general, the results showed that the RF model could be used for mapping the spatiotemporal dynamics of SOC content.
This study demonstrates a framework for assessing the impacts of deforestation on the spatial patterns of soils using DSM techniques, which will facilitate effective land use planning and sustainable forest resource management strategies.