Validation and Application of PHYDOTax In The Santa Barbara Channel

Phytoplankton Functional Types(PFTs) are conceptual groupings of phytoplankton based on physical or functional characteristics. Understanding phytoplankton is essential in our study of how ecosystems function and in monitoring carbon flow. PHYDOTax is a PFT algorithm that discriminates taxon-specific biomass in images from airborne and spaceborne hyperspectral sensors. The PHYDOTax algorithm uses a spectral library and an inverse matrix approach to deconvolve pure phytoplankton spectral end-members from spectra of natural waters. The spectral library used in development was created from phytoplankton taxa found in Monterey Bay, CA and the California Current System (CCS). PHYDOTax has only been validated in Monterey Bay and for only one airborne sensor, Spectroscopic Aerial Mapping System with On-board Navigation (SAMSON). The objectives of this research were to apply PHYDOTax to a different region of the CCS, Santa Barbara Channel (SBC), and to test the usability of PHYDOTax with a different airborne imager, Airborne Visible/Infrared Imaging Spectrometer (AVIRIS). PHYDOTax was modified to accommodate the reduced spectral resolution of AVIRIS, which has fewer wavelengths than the SAMSON imagery that was used to validate the model for Monterey Bay. PHYDOTax's predictions were consistent with cell-count data from whole water samples on June 5, 2013, courtesy of the Plumes and Blooms Cruise (UCSB). PHYDOTax's ability to perform in another area of the CCS shows promise that it may be accurately applied to the west coast of the US. PHYDOTax's ability to perform with lower spectral resolution imagery, suggests that it may it may be robust enough to be down-sampled to multi-spectral resolution inputs. This opens the possibility of applying PHYDOTax to historical (SeaWiFS, MODIS, MERIS), existing (VIIRS, HICO), and future (PACE, GEO-CAPE) sensors to describe temporal trends in phytoplankton distribution and carbon flow in the ocean and to build continuity among the sensor suites.