Evaluation of different boosting ensemble machine learning models and novel deep learning and boosting framework for head-cut gully erosion susceptibility.

The objective of this study is to assess the gully head-cut erosion susceptibility and identify gully erosion prone areas in the Meimand watershed, Iran. In recent years, this study area has been greatly influenced by several head-cut gullies due to unusual climatic factors and human induced activity. The present study is therefore intended to address this issue by developing head-cut gully erosion prediction maps using boosting ensemble machine learning algorithms, namely Boosted Tree (BT), Boosted Generalized Linear Models (BGLM), Boosted Regression Tree (BRT), Extreme Gradient Boosting (XGB), and Deep Boost (DB). Initially, we produced a gully erosion inventory map using a variety of resources, including published reports, Google Earth images, and field records of the Global Positioning System (GPS). Subsequently, we distributed this information randomly and choose 70% (102) of the test gullies and the remaining 30% (43) for validation. The methodology was designed using morphometric and thematic determinants, including 14 head-cut gully erosion conditioning features. We have also investigated the following: (a) Multi-collinearity analysis to determine the linearity of the independent variables, (b) Predictive capability of piping models using train and test dataset and (c) Variables importance affecting head-cut gully erosion. The study reveals that altitude, land use, distances from road and soil characteristics influenced the method with the greatest impact on head-cut gully erosion susceptibility. We presented five head-cut gully erosion susceptibility maps and investigated their predictive accuracy through area under curve (AUC). The AUC test reveals that the DB machine learning method demonstrated significantly higher accuracy (AUC = 0.95) than the BT (AUC = 0.93), BGLM (AUC = 0.91), BRT (AUC = 0.94) and XGB (AUC = 0.92) approaches. The predicted head-cut gully erosion susceptibility maps can be used by policy makers and local authorities for soil conservation and to prevent threats to human activities.

Aerial and terrestrial-based monitoring of channel erosion, headcutting, and sinuosity.

Headcuts are points of accelerated channel erosion that frequently have ecological consequences. A particularly large and dynamic headcut in southwest Wyoming has affected natural and anthropogenic resources for decades. To better understand and address this issue, we undertook a review of the headcut's upstream retreat, followed by photogrammetric monitoring of the present condition for erosion monitoring. Aerial photography shows the Bitter Creek headcut retreated > 200 m upstream in 68 years (1948-2016) at ~ 1.4 m year-1. Following installation of a concrete slab structure in the mid-1970s, headcut retreat slowed to ~ 0.5 m year-1. Channel sinuosity downstream of the headcut is greater than upstream, which we attribute to the presence of the headcut, given that there are no major changes in valley geometry, geology, or soils through this reach. Both aerial and terrestrial-based image platforms were used to collect stereo imagery and create 3D photogrammetric models of the headcut in 2016. From these two models, we measured soil loss downstream of the headcut at ~ 126 m3 m-1 valley length. Since 1954, soil loss within the channel has been ~ 98 m3 year-1 or ~871 t ha-1 year-1since then. Models created from aerial- and terrestrial-based images differed in volumetric estimates by 2%, indicating that either method could be used for this type of monitoring. The ground-based imagery model showed more detail, especially on vertical and overhanging surfaces, while the aerial imagery model produced a more realistic orthomosaic and efficiently covered a larger area. Ground-based image acquisition took longer and was more costly per unit area, but is an efficient method for small project areas, or areas where aerial imagery cannot be safely or practically acquired. Historical imagery and photogrammetric modeling proved very useful in elucidating stream dynamics associated with this large, dynamic headcut.