Cyanobacteria are ubiquitous aquatic organisms with a remarkable evolutionary history reaching as far as 1.9 Ga. They play a vital role in ecosystems yet also raise concerns due to their association with harmful algal blooms. Understanding the historical patterns and drivers behind these blooms is crucial for effective ecosystem management. Lake-sediment cores are valuable natural environmental archives, recording the histories of such blooms. Among others, phycocyanin, a pigment specific to cyanobacteria, emerges as a promising biomarker for reconstructing past cyanobacterial bloom events. However, due to the physicochemical properties of phycocyanin, there is no validated method available to extract and measure this pigment from complex sediment matrix. This study explores the applicability of hyperspectral imaging (HSI), a non-destructive technique, as a novel approach for high resolution in-situ detection and quantification of phycocyanin in lake sediments. Our experiments show that phycocyanin can be detected by HSI with an absorption trough at 620 nm (relative absorption band depth, RABD620). We established a semi-quantitative calibration of the spectral index RABD620 by conducting spiking experiments with phycocyanin standard (known phycocyanin mass) on organic-rich and mineral-rich sediments of varying water contents. We also assessed potential interference from chlorophyll a, another photosynthetic pigment, ensuring the reliability of hyperspectral phycocyanin measurements. Our findings demonstrate a significant correlation (R2 ranging from 0.37 to 0.997) between the RABD620 index and associated phycocyanin amounts in organic-rich and minerogenic sediments. This indicates the potential of the spectral index to directly measure in-situ biomarker concentrations on split sediment cores. Although confounding factors such as water and chlorophyll a content can influence the spectral signal, this method offers a rapid and non-destructive approach for studying historical cyanobacterial blooms in sedimentary records. This opens promising grounds for various applications, including ecosystem-health assessment and environmental change monitoring.
