Juan C. Montes-Herrera1, Vonda Cummings2, Stacy Deppeler2, Roberta D’Archino2, Wendy Nelson2, Vanessa Lucieer1
1 Institute for Marine and Antarctic Studies, University of Tasmania, Tasmania 7001, Australia, 2 National Institute of Water and Atmospheric Research, PO Box 14901, Wellington, New Zealand
Correspondence: Juan C. Montes-Herrera, juancarlos.montesherrera@utas.edu.au
Coralline algae are a diverse group of calcifying red macroalgae (Rhodophyta) that exhibit distinct spectral signatures due to their light-harvesting pigments called phycobilins. Phycobilin quantification is a standard process in coralline algae photosynthesis research, yet it is time-consuming, and can only be measured destructively. In this study, we assess the potential of non-invasive hyperspectral imaging (HSI) in the visible spectrum (400-800 nm), to describe relationships between hyperspectral indices and photosynthetic pigment content of Antarctic coralline algae (n = 14). We validated our hyperspectral images with standard DNA barcoding methods and spectrophotometric pigment analysis. All specimens were identified as Tethysphytum antarcticum (Hapalidiales) using the psbA marker. Phycobilin extractions displayed peaks in absorbance at specific wavelengths (494 and 564 nm) reported for Antarctic red algae. Accordingly, dips in the reflectance spectral signature obtained with HSI matched the absorbance peaks and overall pigment content. We developed a semi-automated image analysis workflow, to test several hyperspectral indices and their relationship with pigment content. Our results identified the two indices more effective for tracing pigment content in these samples: (1) the double-derivative spectra (563 nm), which had a coefficient of determination (R2) of 0.7, and (2) the area under the curve (480-520 nm) normalized by max band depth (494 nm), with an R2 of 0.66. Results obtained in this study help lay the basis for using hyperspectral imaging in red seaweed phycobilin estimates both in lab and field settings.