Spectral vs. Spatial scaling in low-Arctic tundra

• Main Authors: Beamish, Alison, Heim, Birgit, Chabrillat, Sabine, Coops, Nicholas C
• Format: Conference Object
• Language: unknown
• Published: 2017
• Subjects: Arctic; Tundra
• Online Access: https://epic.awi.de/id/eprint/43772/

Arctic ecosystems are highly heterogeneous with small-scale variations in species composition, microtopography, and surface moisture. Additionally, the low stature of tundra vegetation and lack of major structural changes seasonally creates uncertainties in the application of broadband remote sensing vegetation indices. The goal of this research is to examine the potential of high spectral resolution visible and near infrared (VNIR) spectroscopy to accurately differentiate vegetation communities as well as identify phenological stage and photosynthetic activity. Ground-based reflectance spectra were collected in 8 distinct vegetation communities at early, peak and late season in a low Arctic tundra ecosystem at the Toolik Research station on the Alaskan North Slope. The field-based reflectance spectra were used to simulate spectral data from the upcoming Environmental Mapping and Analysis Program (EnMAP) satellite to examine its applications in Arctic tundra ecosystems. Using an instability index (ISI), a waveband selection algorithm, and absorption feature band depth between 400 and 985 nm, we examined the performance of these two techniques in differentiation vegetation type and identifying biophysical parameters. Results show overall high and consistent waveband selection at major pigment absorption and reflectance features in the visible spectrum and red-edge transition when differentiating phenological stage. We expect that waveband selection when differentiating between vegetation communities will also be high in these spectral regions. Spectral band depth was well correlated to bulk pigment concentrations of chlorophyll and carotenoids. This indicates the potential to use band depth to infer photosynthetic activity. We will also examine the ability of band depth to differentiate vegetation type and phenology. The results of this research supports spectral remote sensing applications (airborne as well as current and future satellite missions) to assess vegetation heterogeneity and biophysical properties of ...