Journal
REMOTE SENSING
Volume 14, Issue 1, Pages -Publisher
MDPI
DOI: 10.3390/rs14010234
Keywords
lidar; fluvial geomorphology; stream width; remote sensing; deep learning
Categories
Funding
- National Science Foundation [1339015, 1830734]
- U.S. Army Engineer Research and Development Center Cold Regions Research and Engineering Laboratory Remote Sensing/GIS Center of Excellence
- Directorate For Geosciences
- Division Of Earth Sciences [1830734] Funding Source: National Science Foundation
- Division Of Earth Sciences
- Directorate For Geosciences [1339015] Funding Source: National Science Foundation
Ask authors/readers for more resources
This paper utilizes a U-Net CNN to map stream boundaries in the McMurdo Dry Valleys in Antarctica, and explores the impact of topographic features on stream boundary detection. The results show that elevation and slope are the most effective feature combinations, and meandering streams outperform other geometries in terms of prediction performance.
Convolutional neural networks (CNNs) are becoming an increasingly popular approach for classification mapping of large complex regions where manual data collection is too time consuming. Stream boundaries in hyper-arid polar regions such as the McMurdo Dry Valleys (MDVs) in Antarctica are difficult to locate because they have little hydraulic flow throughout the short summer months. This paper utilizes a U-Net CNN to map stream boundaries from lidar derived rasters in Taylor Valley located within the MDVs, covering & SIM;770 km(2). The training dataset consists of 217 (300 x 300 m(2)) well-distributed tiles of manually classified stream boundaries with diverse geometries (straight, sinuous, meandering, and braided) throughout the valley. The U-Net CNN is trained on elevation, slope, lidar intensity returns, and flow accumulation rasters. These features were used for detection of stream boundaries by providing potential topographic cues such as inflection points at stream boundaries and reflective properties of streams such as linear patterns of wetted soil, water, or ice. Various combinations of these features were analyzed based on performance. The test set performance revealed that elevation and slope had the highest performance of the feature combinations. The test set performance analysis revealed that the CNN model trained with elevation independently received a precision, recall, and F1 score of 0.94 & PLUSMN;0.05, 0.95 & PLUSMN;0.04, and 0.94 & PLUSMN;0.04 respectively, while slope received 0.96 & PLUSMN;0.03, 0.93 & PLUSMN;0.04, and 0.94 & PLUSMN;0.04, respectively. The performance of the test set revealed higher stream boundary prediction accuracies along the coast, while inland performance varied. Meandering streams had the highest stream boundary prediction performance on the test set compared to the other stream geometries tested here because meandering streams are further evolved and have more distinguishable breaks in slope, indicating stream boundaries. These methods provide a novel approach for mapping stream boundaries semi-automatically in complex regions such as hyper-arid environments over larger scales than is possible for current methods.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available