| Summary: | Sea ice algae stands at the base of polar marine food webs and have an active role in large-scale biogeochemical cycles. Due to the invasive and spatially coarse nature of traditional point sampling techniques, we face an unsettling understanding of ice algal dynamics with associated implications for estimating their overall contribution to marine production and how they respond to environmental changes. Using an algae colonized sea-ice simulation tank, we experimentally assessed the use hyperspectral imaging technology in transmission mode as a new method for capturing sea-ice algae spatial variability at the ice water interface. For the first time, the results highlighted the potential of hyperspectral imaging to capture sea-ice algal biomass variability at unprecedented scales in a non-invasive manner.Even though several challenges are associated with measuring transmitted light through the ice, the extensive amount of spatially and spectrally resolved information acquired through hyperspectral imaging allows to reach beyond the pure discrete biomass distribution estimates and in this study we further explored such possibilities.We inoculated different species of ice algal colonies in our experimental set-up in different concentrations and took a multi-sensor approach to gain a better understanding of the hyperspectral remote sensing capabilities to quantify per-pixel biomass abundance, assess achievable spatial resolutions and detect differences in algae populations among others.Produced algae distribution maps over time were then coupled to detailed ice topography reconstructions derived from digital photogrammetry to assess the capability of the method in spatial change detection context.We conclude discussing the feasibility of deploying such techniques in the field and the next steps to be taken towards future engagement of Unmanned Underwater Vehicles (UUVSs) to increase the spatial coverage of the methodology.
|
|---|