Revising Carbon Uptake Estimates in the European Arctic with a regional satellite algorithm and BGC-Argo data

Revising Carbon Uptake Estimates in the European Arctic with a regional satellite algorithm and BGC-Argo data Aleksandra Cherkasheva, Artur Palacz, Rustam Manurov, Piotr Kowalczuk, Alexandra Loginova, Astrid Bracher 25 November 2nd Ocean Carbon from Space workshop Marine Organic Carbon Atlas (MOCA) Pilot Demonstration for the Arctic Steinberg & Landry 2017 C uptake C export C recycling C burial http://www.ioccp.org/images/D2backgrou ndDoc/IOCR_WG_Report_2021.pdf From: www.sea-quester.eu Contact: Artur Palacz [email protected] Marine Organic Carbon Atlas (MOCA) Pilot Demonstration for the Arctic Steinberg & Landry 2017 C uptake C export C recycling C burial http://www.ioccp.org/images/D2backgrou ndDoc/IOCR_WG_Report_2021.pdf From: www.sea-quester.eu Contact: Artur Palacz [email protected] Greenland Sea Primary Production (PP) algorithm development Morel(1991) Greenland Sea Primary Production (PP) algorithm development Morel(1991) Data used for Greenland Sea adaptation: -Chlorophyll a -Particulate absorption -PI parameters (from Bouman et al. 2018) -Primary production (for validation) Greenland Sea Primary Production (PP) algorithm development Morel(1991) Data used for Greenland Sea adaptation: -Chlorophyll a -Particulate absorption -PI parameters (from Bouman et al. 2018) -Primary production (for validation) Greenland Sea Primary Production (PP) algorithm development Morel(1991) Data used for Greenland Sea adaptation: Example: CHL profile parameterization -1680 data profiles -years from 1957 Cherkasheva et al. (2013) more than 3 depth choice of categories -Chlorophyll a -Particulate absorption -PI parameters (from Bouman et al. 2018) -Primary production (for validation) Validation: PP field data Best performing model setup Cherkasheva et al., 2025 https://doi.org/10.3389/fmars.2024.1491180 Marine Organic Carbon Atlas (MOCA) Pilot Demonstration for the Arctic Steinberg & Landry 2017 C uptake C export C recycling C burial http://www.ioccp.org/images/D2backgrou ndDoc/IOCR_WG_Report_2021.pdf From: www.sea-quester.eu Contact: Artur Palacz [email protected] NCP estimates: main challenges Izett et al., 2024 https://doi.org/10.5194/bg-21-13-2024 Li and Cassar, 2016 doi:10.1002/2015GB005314 NCP estimates: main challenges Method Field Data Needed Temporal scale Cons In Situ Incubations O ₂or DIC changes in light/dark bottles Daily high sampling effort, short -term estimates Oxygen -to-Argon (O₂/Ar) Method O ₂ /Ar ratio, temperature, wind speed Daily to weekly Requires specialized equipment, surface -limited Oxygen Mass Balance (BGC -Argo) O ₂profiles, wind speed, MLD Weekly to annual Sensitive to gas exchange errors Nitrate sensors (BGC -Argo) nitrate profiles, MLD Weekly to annual Ignores regenerated production, weekly to annual estimates (not daily) DIC Mass Balance pCO ₂(converted to DIC), salinity, temperature Seasonal to annual mostly surface -limited, sensitive to gas exchange errors Nitrate Drawdown (from water samples) Nitrate profiles Seasonal to annual Ignores regenerated production (leads to underestimation), could be challenging to estimate winter nitrate levels Triple Oxygen Isotopes Isotopic O ₂samples Daily to weekly Lab processing effort unknown to me, specialized equipment needed Optical Sensors (BGC -Argo) (?) Backscatter, fluorescence Daily Indirect, calibration needed Thorium -234 Method (attached to sinking particles) ²³⁴Th and POC profiles Weekly to monthly Requires radionuclide handling, gives a componet of NCP usually used as an additional method in NCP intercomparison Sediment Traps POC flux data Seasonal to annual Expensive, gives a componet of NCP usually used as an additional method in NCP intercomparison Various sensors on gliders (oxygen, pCO2) same as p.4 Weekly to seasonal same as p.4 link to table Annual Net Community Production (ANCP) from Argo Floats ANCP calculation method following Johnson et al. (2017) Climatological cycles BGC Argo CHL and satellite PP agree well Conclusions PP algorithm is operational with Python codes uploaded online Accuracy of the selected model setups to reproduce the field data in terms of RMSD (RMSD=0.4) is better than in the related Arctic studies (RMSD=0.61-0.67)*. Larger Greenland Sea basin PP annual estimates, seasonal cycle pattern align with BGC Argo CHL NCP algorithm is in progress as function of PP satellite, NCP BGC Argo estimates are in the range of glider studies *Lee et al. (2015), Saba et al. (2011) Cherkasheva et al., 2025 https://doi.org/10.3389/fmars.2024.1491180 Conclusions PP algorithm is operational with Python codes uploaded online Accuracy of the selected model setups to reproduce the field data in terms of RMSD (RMSD=0.4) is better than in the related Arctic studies (RMSD=0.61-0.67)*. Larger Greenland Sea basin PP annual estimates, seasonal cycle pattern align with BGC Argo CHL NCP algorithm is in progress as function of PP satellite, NCP BGC Argo estimates are in the range of glider studies *Lee et al. (2015), Saba et al. (2011) Cherkasheva et al., 2025 https://doi.org/10.3389/fmars.2024.1491180 Knowledge gaps and priorities for next steps Scarce to no data for validation of PP and NCP at polar latitudes Use not only traditional methods (C14), but also alternative (oxygen sensors, BGC floats) Include at least PP and maybe NCP as GOOS Essential Ocean Variable No salinity sensor for half of the BGC float dives (3445 out of 6881) Include it in basic setup Need for Arctic-specific CHL satellite algorithm Zoffoli et al is planned to be soon available on Copernicus, better to use it if you’re working in the Arctic Need for reliable PAR level 3 product Using climatologies or level 2 PAR from EUMETSAT instead Global Observing Ocean System EOV specification sheets