Comparison of Ensemble Machine Learning Methods for Soil Erosion Pin Measurements

Abstract

Although machine learning has been extensively used in various fields, it has only recently been applied to soil erosion pin modeling. To improve upon previous methods of quantifying soil erosion based on erosion pin measurements, this study explored the possible application of ensemble machine learning algorithms to the Shihmen Reservoir watershed in northern Taiwan. Three categories of ensemble methods were considered in this study: (a) Bagging, (b) boosting, and (c) stacking. The bagging method in this study refers to bagged multivariate adaptive regression splines (bagged MARS) and random forest (RF), and the boosting method includes Cubist and gradient boosting machine (GBM). Finally, the stacking method is an ensemble method that uses a meta-model to combine the predictions of base models. This study used RF and GBM as the meta-models, decision tree, linear regression, artificial neural network, and support vector machine as the base models. The dataset used in this study was sampled using stratified random sampling to achieve a 70/30 split for the training and test data, and the process was repeated three times. The performance of six ensemble methods in three categories was analyzed based on the average of three attempts. It was found that GBM performed the best among the ensemble models with the lowest root-mean-square error (RMSE = 1.72 mm/year), the highest Nash-Sutcliffe efficiency (NSE = 0.54), and the highest index of agreement (d = 0.81). This result was confirmed by the spatial comparison of the absolute differences (errors) between model predictions and observations using GBM and RF in the study area. In summary, the results show that as a group, the bagging method and the boosting method performed equally well, and the stacking method was third for the erosion pin dataset considered in this study.

Understanding the Drivers of Mobility during the COVID-19 Pandemic in Florida, USA Using a Machine Learning Approach

“…Generally, random forest models are a good choice for regression and classification tasks based on their advantages, e.g., little preprocessing (rescaling or transforming) of the data is required, the modeling can be parallelizable, are compatible with high dimensional data, and are typically robust to outliers and unbalanced data [17]. Comparisons of random forest models with other machine learning algorithms (e.g., linear regression, decision tree, artificial neural network, and support vector machine) for geospatial modeling find that the random forest model performance, in terms of both computation time and prediction accuracy, is generally positive [18,19].…”

DEM- and GIS-Based Analysis of Soil Erosion Depth Using Machine Learning

“…Environmental factors were obtained from various GIS sources such as land use/land cover maps, geological maps, river maps, and road system maps. These factors and four additional factors (% sand, % silt, % clay, and % organic) were previously analyzed in Nguyen et al [22]. However, the four additional factors were removed from this study because they were point data and could not be directly mapped to the entire study area (watershed).…”

“…The knowledge gained from previously grown trees is used to grow new trees in a sequential manner. The GBM model was once used to model soil erosion [22]. It was implemented in this study using R software's "gbm" package.…”

“…where P is the predicted value, O is the observed value, and O is the mean observed value. RMSE was used to compare the difference between the expected values (model outputs) and the observed values (erosion pin measurements) in the two indices, while NSE was used to determine the effectiveness of the model against the average observed value [20][21][22].…”

“…Nguyen et al [20] were the first to create machine learning models from field erosion pin measurements, a critical difference from other ML studies on soil erosion. The analysis was improved and expanded to different ML algorithms, including ensemble learning methods [21][22][23]. However, because some of the environmental factors used in the studies mentioned above were point data, the resulting models could not be directly applied to the entire study area (watershed) without interpolation.…”