Beamish, A., M. K. Raynolds, H. E. Epstein, G. V. Frost, M. J. Macander, H. Bergstedt, A. Bartsch, S. Kruse, V. Miles, C. M. Tanis, B. Heim, M. Fuchs, S. Chabrillat, I. Shevtsova, M. Verdonen, and J. Wagner. 2020. Recent trends and remaining challenges for optical remote sensing of Arctic tundra vegetation: A review and outlook. Remote Sensing of Environment 246:111872. DOI:10.1016/j.rse.2020.111872.
A systematic review and inventory of recent research relating to optical remote sensing of Arctic vegetation was conducted, and thematic and geographical trends were summarized. Research was broadly categorized into four major themes of (1) time series, including NDVI trends and shrub expansion; (2) disturbance and recovery, including tundra fires, winter warming, herbivory, permafrost disturbance, and anthropogenic change; (3) vegetation properties, including biomass, primary productivity, seasonality, phenology, and pigments; and (4) classification and mapping. Remaining challenges associated with remote sensing of Arctic vegetation were divided into three categories and discussed. The first are issues related to environmental controls including disturbance, hydrology, plant functional types, phenology and the tundra-taiga ecotone, and understanding their influence on interpretation and validation of derived remote sensing trends. The second are issues of upscaling and extrapolation related to sensor physics and the comparability of data from multiple spatial, spectral, and temporal resolutions. The final category identifies more philosophical challenges surrounding the future of data accessibility, big data analysis, sharing and funding policies among major data providers such as national space agencies and private companies, as well as user groups in the public and private sectors. The review concludes that the best practices for the advancement of optical remote sensing of Arctic vegetation include (1) a continued effort to share and improve in situ-validated datasets using camera networks and small Unmanned Aerial Vehicles, (2) data fusion with non-optical data, (3) sensor continuity, consistency, and comparability, and (4) free availability and increased sharing of data. These efforts are necessary to generate high quality, temporally