MERIS AO # 516 - Land Cover Mapping at BOREAS Study Area

The authors are grateful for the financial support from research grants provided through GEOmatics for Informed Decisions (GEOIDE), part of the Canadian Networks of Centres of Excellence (NCE).úÒ’ûŒÉ
The objective of this study is to validate the MERIS vegetation land cover classification product. The full resolution MERIS radiance data sets obtained over the BOREAS (Boreal Ecosystem-Atmosphere Study) South Study Area in May and August 2003 were used. The MERIS radiance data were first converted to at-canopy reflectance data, which were compared to CASI data obtained during BOREAS (1994), followed by unsupervised classification performed based on seasonal variation of pigments as inferred from visible and near-infrared spectral bands. Three modified normalized Difference Vegetation Indices (mNDVI), sensitive to relative proportions among pigments and pigment content, and a red-edge spectral parameter, the wavelength at the reflectance minimum (ë0) were used in the unsupervised classification. Accuracy assessments of the derived vegetation classification maps were performed using a forest inventory map provided by the Saskatchewan Environment and Resource Management Forestry Branch-Inventory Unit (SERM-FBIU). The forest vegetation classification using seasonal changes in optical indices (mNDVIs and ë0), derived from the MERIS imagery in May and August revealed a reasonably high overall classification accuracy for all vegetation cover types identified: conifer, mixed stands, and fen. The classification results also demonstrated that classification using reflectance parameters sensitive to pigment absorption outperformed that using reflectance itself and the classification using seasonal information was better than that using information obtained in a single MERIS image, which were consistent with that achieved using multi-date CASI imagery. Implications of MERIS spatial resolution on the data product classification accuracy are discussed by comparison with results obtained with CASI.