Accuracy Assessment of the Planar Morphology of Valley Bank Gullies Extracted with High Resolution Remote Sensing Imagery on the Loess Plateau, China

A Study on Prediction Model of Gully Volume Based on Morphological Features in the JINSHA Dry-Hot Valley Region of Southwest China

Gully erosion is well-developed in the Jinsha dry-hot valley region, which has caused serious soil losses. Gully volume is regarded as an effective indicator that can reflect the development intensity of gully erosion, and the evolutionary processes of gullies can be predicted based on the dynamic variation in gully volume. Establishing an effective prediction model of gully volume is essential to determine gully volume accurately and conveniently. Therefore, in this work, an empirical prediction model of gully volume was constructed and verified based on detailed morphological features acquired by elaborate field investigations and measurements in 134 gullies. The results showed the mean value of gully length, width, depth, cross-section area, volume, and vertical gradient decreased with the weakness of the activity degree of the gully, although the decrease in processes of these parameters had some differences. Moreover, a series of empirical prediction models of gully volume was constructed, and gully length was demonstrated to be a better predictor than other morphological features. Lastly, the effectiveness test showed the model of V = aL^b was the most effective in predicting gully volume among the different models established in this study. Our results provide a useful approach to predict gully volume in dry-hot valley regions.

Evaluation of Three Different Machine Learning Methods for Object-Based Artificial Terrace Mapping—A Case Study of the Loess Plateau, China

Artificial terraces are of great importance for agricultural production and soil and water conservation. Automatic high-accuracy mapping of artificial terraces is the basis of monitoring and related studies. Previous research achieved artificial terrace mapping based on high-resolution digital elevation models (DEMs) or imagery. As a result of the importance of the contextual information for terrace mapping, object-based image analysis (OBIA) combined with machine learning (ML) technologies are widely used. However, the selection of an appropriate classifier is of great importance for the terrace mapping task. In this study, the performance of an integrated framework using OBIA and ML for terrace mapping was tested. A catchment, Zhifanggou, in the Loess Plateau, China, was used as the study area. First, optimized image segmentation was conducted. Then, features from the DEMs and imagery were extracted, and the correlations between the features were analyzed and ranked for classification. Finally, three different commonly-used ML classifiers, namely, extreme gradient boosting (XGBoost), random forest (RF), and k-nearest neighbor (KNN), were used for terrace mapping. The comparison with the ground truth, as delineated by field survey, indicated that random forest performed best, with a 95.60% overall accuracy (followed by 94.16% and 92.33% for XGBoost and KNN, respectively). The influence of class imbalance and feature selection is discussed. This work provides a credible framework for mapping artificial terraces.

Spatial-Temporal Dynamics of the Ephemeral Gully Belt on the Plowed Slopes of River Basins in Natural and Anthropogenic Landscapes of the East of the Russian Plain

Erosion is the leading process of soil degradation on agricultural land. In the spectrum of erosion processes, the most unfavorable for soil degradation are the processes of linear (ephemeral and gully) erosion. An assessment of the dynamics of linear erosion in the intensive farming zone of the European part of Russia (EPR) is relevant due to the lack of generalized data on the development of this type of erosion in the post-Soviet period and also, due to the highest intensity of soil erosion in the ephemeral gully erosion. The development of information technologies and the availability of high-resolution and ultra-high-resolution satellite images make it possible to solve the problems of ephemeral gully erosion belts identification, and also makes it possible to trace the dynamics of development of stream erosion on arable lands over a period characterized by the greatest changes in the climate system and economic conditions in the post-Soviet period (1980s–2010s). The study was conducted on the eastern wing of the boreal ecotone of the Russian Plain within the southern border of these zones of mixed and broad-leaved forests, forest-steppe, and steppe landscapes using the basin approach. For the initial material, satellite images of medium (30 m) and high resolution (0.5–1.5 m) were used in the work. The study used methods of image interpretation such as remote sensing of the earth and geoinformation mapping. For 70 key areas (interfluve spaces of river basins), the study developed a method of geoinformation mapping of the ephemeral gully erosion belt dynamics on arable lands. In the same way, the research developed a system of quantitative indicators characterizing its development on arable slopes. The dynamics of ephemeral gully erosion was evaluated over three-time intervals: the 1980s, 2000s, and 2010s by determining the horizontal dissection (density) and density of ephemeral gully erosion. Over the past 30 years, in the direction from the south of the forest sub-zone to the forest-steppe and steppe landscapes, there was a sharp increase in the horizontal dissection and density of the ephemeral gully network: an average of 4.6 and 10 times, respectively. The ephemeral gully erosion belt advances toward the watershed because of the formation of new erosion in the upper parts of the ephemeral gully networks and its extension, while there is a noticeable reduction in the width of the erosion-weakly active belt-sheet and rill erosion.

Optimized Segmentation Based on the Weighted Aggregation Method for Loess Bank Gully Mapping

The Chinese Loess Plateau suffers severe gully erosion. Gully mapping is a fundamental task for gully erosion monitoring in this region. Among the different gully types in the Loess Plateau, the bank gully is usually regarded as the most important source for the generation of sediment. However, approaches for bank gully extraction are still limited. This study put forward an integrated framework, including segmentation optimization, evaluation and Extreme Gradient Boosting (XGBoost)-based classification, for the bank gully mapping of Zhifanggou catchment in the Chinese Loess Plateau. The approach was conducted using a 1-m resolution digital elevation model (DEM), based on unmanned aerial vehicle (UAV) photogrammetry and WorldView-3 imagery. The methodology first divided the study area into different watersheds. Then, segmentation by weighted aggregation (SWA) was implemented to generate multi-level segments. For achieving an optimum segmentation, area-weighted variance (WV) and Moran’s I (MI) were adopted and calculated within each sub-watershed. After that, a new discrepancy metric, the area-number index (ANI), was developed for evaluating the segmentation results, and the results were compared with the multi-resolution segmentation (MRS) algorithm. Finally, bank gully mappings were obtained based on the XGBoost model after fine-tuning. The experiment results demonstrate that the proposed method can achieve superior segmentation compared to MRS. Moreover, the overall accuracy of the bank gully extraction results was 78.57%. The proposed approach provides a credible tool for mapping bank gullies, which could be useful for the catchment-scale gully erosion process.