Antarctic phytoplankton in response to environmental change studied by a synergistic approach using multi- and hyper-spectral satellite data
Final Report Abstract
The PHYSYN project goal was to assess environmental changes in the Southern Ocean on the phytoplankton community by establishing a long-term phytoplankton diversity data set with sufficient spatial and temporal resolution using a combination of satellite and model data input. PFT retrievals from hyper-spectral satellites were explored from GOME-2A and OMI based on earlier studies for the SCIAMACHY sensor. Improved retrieval settings found in the PHYSYN project were also implemented for retrievals on SCIAMACHY data contributing to the successfull development of the synergistic algorithm and a 10 year synergistic PFT product based on SCIAMACHY-PhytoDOAS-PFT and multispectral OC-PFT time series data. Especially challenging was the conversion of the PhytoDOAS PFT retrieval output to a PFT Chla which requires the use of the retrievals of the vibrational Raman scattering (VRS). The studies of the VRS signal resulted in a useful side product, the diffuse attenuation coeffcient Kd. Kd in the blue spectral range was retrieved for a time period of 11 months (Feb 2007 to Dec 2007) for SCIAMACHY, OMI, and GOME-2A. The intercomparison of Kd from the three sensors was the first detailed comparison of oceanic quantities derived from hyperspectral sensors in the blue spectral range. The intercomparison revealed how challenging it is to accurately calibrate hyperspectral sensors reliably to consistently retrieve oceanic quantities. Producing a long time series from these sensors is difficult and requires complex and time-consuming corrections. Further studies are required to achieve this goal. Then data sets were explored and adapted to the Southern Ocean. Cloud and offset corrections were implemented to obtain two time periods of six months of PFT data from GOME-2A in the Southern Ocean (Oct 2010/11 to Mar 2011/12). A synergistic PFT product for the Southern Ocean using multispectral OC-CCI and hyperspectral GOME-2A satellite output (in combination with the multispectral OC-PFT product) was tested. The results stressed the necessity to accurately assess the uncertainty of the single input products and to correct the bias existing due to a mismatch between phytoplankton grouping assumed in PhytoDOAS and OC-PFT algorithms. Our studies on developing PFT products for the Southern Ocean by hyperspectral satellite retrievals of PFT Chla and a synergistic combination of hyperspectral and multispectral satellite information were complemented by PFT-resolved numerical coupled sea-ice - ocean - biogeochemical modeling. The satellite data products developed within the PHYSYN project and available in situ data sets on the Southern Ocean PFTs allowed us to better understand the biogeography of the PFTs and to infer/define the trait requirements for model parameterization. The adjusted model results represented well the in situ and satellite observations, when a) two diatoms were included (heavily and lightly silicified), b) an ability to escape grazing and a high affinity for nutrients were implemented for coccolithophores, and c) two life stages for Phaeocystis were introduced (solitary and colonial form). By introducing these traits we could simulate the observed diatoms vs. coccolithophores dominance in the Great Calcite Belt and distinguish between two predicted groups of Southern Ocean haptophytes (coccolithophores and Phaeocystis). This ability of the biogeochemical model to predict the distribution of both coccolithophores and Phaeocystis will provide benefits when combing model outputs with satellite data products such as OC-PFT Chla for haptophytes in general and PhytoDOAS Chla for coccolithophores.
Publications
- (2016). “Towards improved spatial resolution of hyper-spectral phytoplankton functional type products”, ‘Colour and Light in the Ocean from Earth Observation (CLEO) Relevance and Applications Products from Space and Perspectives from Models’ (Frascati: ESA-ESRIN)
Oelker, J., Dinter, T., Richter, A., Burrows, J. P., and Bracher, A.
- Synergistic exploitation of hyper- and multispectral precursor Sentinel measurements to determine Phytoplankton Functional Types at best spatial and temporal resolution (SynSenPFT). Frontiers in Marine Science 4: 203 (2017)
Losa S., Soppa M. A., Dinter T., Wolanin A., Brewin R. J. W., Bricaud A., Oelker J., Peeken I., Gentili B., Rozanov. V. V., Bracher A.
(See online at https://doi.org/10.3389/fmars.2017.00203) - Investigating the phytoplankton diversity in the Great Calcite Belt: perspectives from hyper- and multispectral satellite retrievals and numerical modelling. Extended Abstract of the XXIV Ocean Optics, Dubrovnik, Croatia (2018)
Losa, S. N, Oelker, J., Soppa, M. A., Dutkiewicz, S., Losch, M., Dinter, T., Rozanov, V. V., Richter, A., Burrows, J. P., Bracher, A.
- Global diffuse attenuation derived from vibrational Raman scattering detected in hyperspectral backscattered satellite spectra. Opt. Express 27, A829-A855 (2019)
Oelker J., Richter A., Dinter T., Rozanov. V. V., Burrows J.P., Bracher A.
(See online at https://doi.org/10.1364/oe.27.00a829) - On modeling the Southern Ocean Phytoplankton Functional Types, Biogeosciences Discuss., 2019
Losa, S. N., Dutkiewicz, S., Losch, M., Oelker, J., Soppa, M. A., Trimborn, S., Xi, H., and Bracher, A.
(See online at https://doi.org/10.5194/bg-2019-289)
