|
Persistent Identifier
|
doi:10.11588/data/WGAU4Q |
|
Publication Date
|
2024-05-13 |
|
Title
| An Automatic Iterative Random Forest approach to derive gully activity maps in large areas with training data scarcity [Data and Source Code] |
|
Author
| Vallejo-Orti, Miguel (Namibia University of Science and Technology, Namibia; 3D Geospatial Data Processing Group, Institute of Geography, Heidelberg University, Germany) - ORCID: 0000-0002-8464-772X
Winiwarter, Lukas (University of Innsbruck, Austria) - ORCID: 0000-0001-8229-1160
Corral, Eva (University of Huelva, Spain)
Williams, Jack (Risk Analyst at Risk Management Solutions, Inc.)
Bubenzer, Olaf (Institute of Geography, Heidelberg University, Germany) - ORCID: 0000-0002-3199-1156
Höfle, Bernhard (3D Geospatial Data Processing Group, Institute of Geography, Heidelberg University, Germany) - ORCID: 0000-0001-5849-1461 |
|
Point of Contact
|
Use email button above to contact.
Vallejo-Orti, Miguel (3D Geospatial Data Processing Group, Institute of Geography, Heidelberg University, Germany) |
|
Description
| 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 labeled area <; 0.3% of the total study area) in two different watersheds (408 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 receiver operating characteristic 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 (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. |
|
Subject
| Earth and Environmental Sciences |
|
Related Publication
| M. Vallejo Orti, L. Winiwarter, E. Corral-Pazos-de-Provens, J. G. Williams, O. Bubenzer and B. Höfle, "Use of TanDEM-X and Sentinel Products to Derive Gully Activity Maps in Kunene Region (Namibia) Based on Automatic Iterative Random Forest Approach," in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 14, pp. 607-623, 2021 doi: 10.1109/JSTARS.2020.3040284 https://doi.org/10.1109/JSTARS.2020.3040284 |
|
Production Date
| 2020-03-15 |
|
Production Location
| Namibia |
|
Funding Information
| Heidelberg University (Kurt-Hiehle-Foundation))
German Aerospace center (TanDEM-X Science Team): DEM_HYDR2024 |
|
Deposit Date
| 2024-04-01 |
