Coastal Wetland Mapping Using Ensemble Learning Algorithms: A Comparative Study of Bagging, Boosting and Stacking Techniques

• Published in: Remote sensing (Basel, Switzerland) Vol. 12; no. 10; p. 1683
• Main Authors: Wen, Li, Hughes, Michael
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01.05.2020
• Subjects: Algorithms; anthropogenic activities; Australia; Bagging; Classifiers; Coastal ecosystems; Coastal management; coastal wetland; Coastal zone; coasts; Comparative studies; comparative study; Critical components; Datasets; decision support systems; Decision trees; Discriminant analysis; Ecosystem management; Ecosystems; Ensemble learning; Estuaries; fractional cover; Geomorphology; Landscape; landscapes; Learning algorithms; Machine learning; mangrove; Mapping; Performance evaluation; phenology; Quantitative research; Remote sensing; Remote sensing systems; River ecology; Rivers; saltmarsh; Satellites; sea level; Sea level rise; sentinel-2; spatial data; Stacking; Support vector machines; time series analysis; vegetation index; Vegetation mapping; Water depth; watersheds; Wetlands; Australia
• ISSN: 2072-4292; 2072-4292
• DOI: 10.3390/rs12101683

Coastal wetlands are a critical component of the coastal landscape that are increasingly threatened by sea level rise and other human disturbance. Periodically mapping wetland distribution is crucial to coastal ecosystem management. Ensemble algorithms (EL), such as random forest (RF) and gradient boosting machine (GBM) algorithms, are now commonly applied in the field of remote sensing. However, the performance and potential of other EL methods, such as extreme gradient boosting (XGBoost) and bagged trees, are rarely compared and tested for coastal wetland mapping. In this study, we applied the three most widely used EL techniques (i.e., bagging, boosting and stacking) to map wetland distribution in a highly modified coastal catchment, the Manning River Estuary, Australia. Our results demonstrated the advantages of using ensemble classifiers to accurately map wetland types in a coastal landscape. Enhanced bagging decision trees, i.e., classifiers with additional methods to increasing ensemble diversity such as RF and weighted subspace random forest, had comparably high predictive power. For the stacking method evaluated in this study, our results are inconclusive, and further comprehensive quantitative study is encouraged. Our findings also suggested that the ensemble methods were less effective at discriminating minority classes in comparison with more common classes. Finally, the variable importance results indicated that hydro-geomorphic factors, such as tidal depth and distance to water edge, were among the most influential variables across the top classifiers. However, vegetation indices derived from longer time series of remote sensing data that arrest the full features of land phenology are likely to improve wetland type separation in coastal areas.