Use of TanDEM-X and Sentinel Products to Derive Gully Activity Maps in Kunene Region (Namibia) Based on Automatic Iterative Random Forest Approach

Abstract

Gullies are landforms with specific patterns of shape, topography, hydrology, vegetation, and soil characteristics. Remote sensing products (TanDEM-X, Sentinel-1 and Sentinel-2) serve as inputs into an iterative algorithm, initialized using a micro-mapping simulation as training data, to map gullies in the northwestern of Namibia. A Random Forest Classifier examines pixels with similar characteristics in a pool of unlabeled data and gully objects are detected where high densities of gully pixels are enclosed by an alpha shape. Gully objects are used in subsequent iterations following a mechanism where the algorithm uses the most reliable pixels as gully training samples. The gully class continuously grows until an optimal scenario in terms of accuracy is achieved. Results are benchmarked with manually tagged gullies (initial gully labelled area <0.3% of the total study area) in two different watersheds (408 km 2 and 302 km 2 , respectively) yielding total accuracies of >98%, with 60% in the gully class, Cohen Kappa>0.5, Matthews Correlation Coefficient>0.5, and ROC Area Under the Curve>0.89. Hence, our method outlines gullies keeping low false-positive rates while the classification quality has a good balance for the two classes (Gully/No Gully). Results show the most significant gully descriptors as the high temporal radar signal coherence (22.4%) and the low temporal variability in Normalized Difference Vegetation Index (NDVI) (21.8%). This research builds on previous studies to face the challenge of identifying and outlining gully-affected areas with a shortage of training data using global datasets, which are then transferable to other large (semi-) arid regions.

Classifying types of gully changes with unoccupied aircraft vehicles 3D multitemporal point clouds for training of satellite data analysis in Northwest Namibia

“…However, the latter is less sensitive to modifications in the proportion of training to testing data. These results represent an advance in the zonal characterisation and monitoring of gullies (Vanmaercke et al, 2021), surpassing the mapping of gullies at a specific time using RF (Vallejo‐Orti et al, 2021) and adding value to the quantification of changes to the gully perimeter in large periods (Shruthi et al, 2015).…”

“…Furthermore, extensive research focuses on the development of empirical models to map gully heads at a global and continental scale (Omran et al, 2022; Sidorchuk, 2021; Vanmaercke et al, 2020). Identifying and measuring gullies in large extents at a given time in 2D has been addressed by various authors (Golosov et al, 2018; Shahabi et al, 2019; Vallejo‐Orti et al, 2021). Nevertheless, certain aspects remain understudied and relevant, such as the temporal dimension of gullies and techniques to measure their evolution over long periods, that is, >5 years (Hayas et al, 2017), and at regional scales (Vanmaercke et al, 2021), that is, >1000 ha (Kimura et al, 2009; Morgan, 2005).…”

Gullies and badlands of India: Genesis, geomorphology and land management

“…Busch et al, 2021) or Namibia (e.g. Orti et al, 2021), it is difficult to find value of such gully occurrence mapping studies in India, as state-wise land degradation datasets are already available, with gullies and badlands delineated at a scale of 1:50,000 (NRSC, 2007(NRSC, , 2019.. Therefore, mapping geomorphic characteristics of gully systems/ badlands, such as gully head density (cf.…”

“…Although such data-driven mapping is relevant in countries that have a dearth of data on the spatial extents of gullies and/or badlands, such as Iran (e.g., Arabameri et al, 2020;Rahmati et al, 2017), Ethiopia (e.g., Busch et al, 2021) or Namibia (e.g., Orti et al, 2021), it is difficult to find the value of such gully occurrence mapping studies in India, as state-wise land degradation datasets are already available, with gullies and badlands delineated at a scale of 1:50,000 (NRSC, 2007(NRSC, , 2019. Therefore, mapping geomorphic characteristics of gully systems/badlands, such as gully head density (cf.…”

Gully erosion mapping based on remote sensing data

“…Various sensors and image resolutions have become utilized in the gully's detection and mapping process. For instance, the application of mediumresolution optical systems like ASTER [14][15][16] and Sentinel 17,18 , as well as high-resolution images like UAV 19 . Alternatively, passive technologies such as RADAR 18 and LIDAR [20][21][22][23][24] may be utilized to derive topography data.…”