Mapping wetlands with high spatial and thematic accuracy is crucial for the management and monitoring of these important ecosystems. Wetland maps in New Brunswick (NB) have traditionally been produced by the visual interpretation of aerial photographs. In this study, we used an alternative method to produce a wetland map for southern New Brunswick, Canada, by classifying a combination of Landsat 8 OLI, Alos-1 Palsar, Sentinel-1, and LiDAR-derived topographic metrics with the Random Forests (RF) classifier. The images were acquired in three seasons (spring, summer, and fall) with different water levels and during leaf-off/on periods. The resulting map provides eleven wetland classes (open bog, shrub bog, treed bog, open fen, shrub fen, freshwater marsh, coastal marsh, shrub marsh, shrub wetland, forested wetland, and aquatic bed) plus various non-wetland classes. We achieved an overall accuracy classification of 97.67% and a kappa coefficient of 95.39%. The remained 489 in-situ validation sites were compared to the classified image, the provincial wetland reference map available through Service New Brunswick in 2016, and the new provincial wetland map of 2019. Of these sites, 94.27% were identified in the correct wetland class on the classified image, but only 29.86% of the sites were correctly identified on the 2016 NB provincial reference map. Only 62.17% of the sites were mapped as wetlands on the 2019 NB provincial reference map. Our approach had a better performance than the traditional visual interpretation of air photos. Keywords: Wetland; Optical; SAR, Sentinel-1, Landsat-8, LiDAR, Random Forests, Alos-Palsar.
Zircon - This is a contributing Drupal Theme Design by WeebPal.
