River sensing: the inclusion of red band in predicting reach-scale types using machine learning algorithms

ABSTRACT

This study aims to predict channel unit types (CUTs) by combining remotely sensed data with morphological variables using machine learning algorithms (random forest, support vector machines, multiple adaptive regression splines, extreme gradient boosting and adaptive boosting) within the Upper Ogun River Basin, Southwestern Nigeria. In achieving the aim of this study, we identified the most important variable(s) in CUT discrimination using the random forest – recursive feature elimination (RF-RFE). A total of 249 cross-sections across 83 reaches were sampled during the fieldwork. Landsat 8 and Sentinel-1 bands were retrieved for days the fieldwork was carried and mosaiced using the Google Earth Engine platform. The RF-RFE identified five top variables (accuracy: 0.79 ± 0.14; kappa: 0.39) discriminating the CUT as dimensionless stream power, slope, width, wetted perimeter and Band 4. In essence, there is much hope in the use of remote sensing in CUT mapping at the reach scale.

KEYWORDS:

• reach-scale
• remote sensing
• red band
• river sensing
• machine learning and classifications

Editor A. Castellarin; Associate editor A. Domeneghetti

Editor A. Castellarin; Associate editor A. Domeneghetti

Acknowledgements

We thank Late Emeritus Professor Faniran who supervised the PhD thesis of Adeyemi Olusola. Also, we are grateful to the reviewers who took the time to read through this manuscript and provide valuable comments. Finally, appreciation goes to Prof. Gregory Pasternack (UC Davis) who took the time to read through the rough drafts patiently over and over.

Disclosure statement

No potential conflict of interest was reported by the authors.