The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLIII-B3-2020
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLIII-B3-2020, 685–692, 2020
https://doi.org/10.5194/isprs-archives-XLIII-B3-2020-685-2020
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLIII-B3-2020, 685–692, 2020
https://doi.org/10.5194/isprs-archives-XLIII-B3-2020-685-2020

  21 Aug 2020

21 Aug 2020

EELGRASS MAPPING IN ATLANTIC CANADA USING WORLDVIEW-2 IMAGERY

D. Forsey1, B. Leblon1, A. LaRocque1, M. Skinner2, and A. Douglas3 D. Forsey et al.
  • 1Faculty of Forestry and Environmental Management, University of New Brunswick, Fredericton (NB), E3B 5A3, Canada
  • 2Stantec Consulting Ltd., 40 Highfield Park Drive 102-40, Dartmouth (NS), B3A 0A3, Canada
  • 3Southern Gulf of St. Lawrence Coalition on Sustainability, Stratford (PEI), C1B 1L1, Canada

Keywords: Eelgrass mapping, Atlantic Canada, WorldView-2, Maximum Likelihood, Random Forests

Abstract. Eelgrass (Zostera marina L.) is a marine angiosperm plant that grows throughout coastal areas in Atlantic Canada. Eelgrass meadows provide numerous ecosystem services, and while they have been acknowledged as important habitats, their location, extent, and health in Atlantic Canada are poorly understood. This study examined the effectiveness of WorldView-2 optical satellite imagery to map eelgrass presence in Tabusintac Bay, New Brunswick (Canada), an estuarine lagoon with extensive eelgrass coverage. The imagery was classified using two supervised classifiers: the parametric Maximum Likelihood Classifier (MLC) and the non-parametric Random Forests (RF) classifier. While Random Forests was expected to produce higher classification accuracies, it was shown not to be much better than MLC. The overall validation accuracy was 97.6% with RF and 99.8% with MLC.