Ocean carbon pumps and export
Click a map for a larger view and the full caption.
How these maps were made. Maps from the published paper (Beckwith et al. 2026) were made by hand in ArcGIS. All other maps, including new additions, were made in QGIS with AI-assisted Python (PyQGIS) scripting.
Ocean carbon pumps and export
Original maps made in QGIS with AI-assisted Python (PyQGIS) scripting, from open data. Each map's sources and licences are listed with it.
- Landschützer et al., MPI-SOM-FFN, NCEI Accession 0160558, doi:10.7289/v5z899n6 (CC BY 4.0).
- Imagery reproduced from the GEBCO_2026 Grid, GEBCO Compilation Group (2026), doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation.
- We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's ESDIS.
- Made with Natural Earth.
- Made in QGIS with AI-assisted Python (PyQGIS) scripting.
Full caption and credits
SHORT CAPTION
Surface-ocean pCO2, delta pCO2 and atmospheric pCO2, July 2024: SOM-FFN v2025. (A) surface-ocean pCO2 (spco2_raw 'surface ocean pCO2 raw', µatm); (B) the product's own delta pCO2 (dco2 'delta pCO2', µatm); its sign convention is not printed in the file or README, so colours are not labelled as uptake or outgassing; (C) the product's own atmospheric pCO2 (atm_co2 'Atmospheric pCO2', µatm), which the README says is derived from the NOAA GML marine boundary layer xCO2 reference, SST and sea-level pressure; it is defined over ocean cells only. All panels are the same single month (printed time 2024.5; the record runs 1982 to 2024.92) of the SOM-FFN v2025 observation-based product on a 1 x 1 degree grid; the surface-ocean pCO2 is a neural-network reconstruction from SOCAT v2024 observations, not raw ship measurements. Nothing was derived: the script did not subtract ocean from air, average months or convert units. A single month carries the seasonal state of that month (northern summer for July); it is not an annual mean. Class breaks are display bins only; land and cells without values are transparent. Made in QGIS.
FULL CREDITS
Data: Jersild, A., Landschuetzer, P., Gruber, N. and Bakker, D.C.E. (2017). An observation-based global monthly gridded sea surface pCO2 and air-sea CO2 flux product from 1982 onward and its monthly climatology (NCEI Accession 0160558). Version 9.9 (SOM-FFN v2025). NOAA National Centers for Environmental Information. Dataset. https://doi.org/10.7289/v5z899n6 [accessed on 2026-11-18]. Method: Landschuetzer, P., Gruber, N., Bakker, D.C.E. (2016). Decadal variations and trends of the global ocean carbon sink. Global Biogeochemical Cycles 30, doi:10.1002/2015GB005359. The pCO2 fields are reconstructed from SOCAT v2024 observations (Bakker et al. 2016, doi:10.5194/essd-8-383-2016) with a two-step neural network (SOM-FFN). Variables shown (July 2024, printed time 2024.5): spco2_raw 'surface ocean pCO2 raw' (units as printed: µatm); dco2 'delta pCO2' (units as printed: µatm); atm_co2 'Atmospheric pCO2' (units as printed: µatm). File globals as printed: institution 'Flanders Marine Institute - University of Maryland College Park', run_id 'SOM-FFN_v2025', creation_date '2025-03-01'. dco2, per the README: Delta pCO2 calculated from the CO2 partial pressure difference between atmosphere and the raw surface ocean partial pressure. The file and README do not print the sign convention (which is subtracted from which). atm_co2, per the README: atmospheric pCO2 in uatm for the air-sea flux calculation, derived from the NOAA ESRL (now GML) marine boundary layer reference dry-air mixing ratio of atmospheric CO2 (https://gml.noaa.gov/ccgg/mbl/), SST (Reynolds et al. 2002) and sea-level pressure (Kalnay et al. 1996), following Dickson et al. 2007. Licence: README: "This product is free to be used. Please cite the data as" the citation above. NCEI metadata for Accession 0160558 (https://www.ncei.noaa.gov/data/oceans/ncei/ocads/metadata/0160558.html) prints: "DATA LICENSE: This dataset is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) License." Ocean bathymetry: shaded-relief imagery from the GEBCO WMS (https://wms.gebco.net/, layer GEBCO_LATEST). Imagery reproduced from the GEBCO_2026 Grid. GEBCO Bathymetric Compilation Group 2026 (2026). The GEBCO_2026 Grid - a continuous terrain model for oceans and land at 15 arc-second intervals. NERC EDS British Oceanographic Data Centre NOC. doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation or any purpose relating to safety at sea. Shown re-toned to a light neutral grey (relief shading kept, colours removed). Imagery: NASA GIBS / EOSDIS Worldview, Blue Marble Next Generation (MODIS, cloud-free mosaic; not time-matched to the map window). We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's Earth Science Data and Information System (ESDIS). Basemap vectors: Natural Earth 1:50m (rivers (ne_50m_rivers_lake_centerlines), lakes (ne_50m_lakes), coastline (ne_50m_coastline), country borders (ne_50m_admin_0_boundary_lines_land), land mask (ne_50m_land)). Made with Natural Earth. Free vector and raster map data @ naturalearthdata.com. Public domain.
HOW THIS FIGURE WAS MADE
Made in QGIS. Generated by the PyQGIS script air_sea_pco2_gradient_v2.py, written for Neil Beckwith's QGIS workflow (QGIS 3.44 LTR on Windows, Python Console editor). This render was exported headlessly in QGIS 3.40.6-Bratislava on 2026-11-18 01:56 (local time). Data route: NCEI HTTPS, HTTP range requests via GDAL /vsicurl/ (HDF5 driver): https://www.ncei.noaa.gov/data/oceans/ncei/ocads/data/0160558/MPI_SOM-FFN_v2025/SOM-FFN_v2025_NCEI_OCADS.nc. Only the chosen month of each variable is read; each slice is cached in the home folder as a 32-bit float GeoTIFF (the file stores 64-bit floats; the map classes are far coarser than that precision). The slice is re-ordered from the file's (lon, lat) south-to-north layout to a north-up raster; no value is modified. Fill values (>= 1e19; the file prints 'Fill Value: 1e20') are transparent. No Earthdata token is needed. Data layers drawn at 60% opacity over the basemap; coastline, borders, rivers and lakes are drawn on top (coastline and borders with thin halos; rivers Natural Earth scalerank <= 5, 0.2 mm muted blue). Ocean basemap: GEBCO shaded relief re-toned to light neutral grey so data colours match the legend. Colour ramps: viridis (spco2_raw); ColorBrewer PuOr (diverging at 0; two palest middle classes omitted) (dco2); cividis (atm_co2), chosen as colour-vision-deficiency-friendly, with visible (not near-white) low classes. Equirectangular EPSG:4326 map; the scale bar is approximate at the equator; north is up.
Surface-ocean pCO2, delta pCO2 and atmospheric pCO2, July 2024: SOM-FFN v2025. (A) surface-ocean pCO2 (spco2_raw 'surface ocean pCO2 raw', µatm); (B) the product's own delta pCO2 (dco2 'delta pCO2', µatm); its sign convention is not printed in the file or README, so colours are not labelled as uptake or outgassing; (C) the product's own atmospheric pCO2 (atm_co2 'Atmospheric pCO2', µatm), which the README says is derived from the NOAA GML marine boundary layer xCO2 reference, SST and sea-level pressure; it is defined over ocean cells only. All panels are the same single month (printed time 2024.5; the record runs 1982 to 2024.92) of the SOM-FFN v2025 observation-based product on a 1 x 1 degree grid; the surface-ocean pCO2 is a neural-network reconstruction from SOCAT v2024 observations, not raw ship measurements. Nothing was derived: the script did not subtract ocean from air, average months or convert units. A single month carries the seasonal state of that month (northern summer for July); it is not an annual mean. Class breaks are display bins only; land and cells without values are transparent. Made in QGIS.
FULL CREDITS
Data: Jersild, A., Landschuetzer, P., Gruber, N. and Bakker, D.C.E. (2017). An observation-based global monthly gridded sea surface pCO2 and air-sea CO2 flux product from 1982 onward and its monthly climatology (NCEI Accession 0160558). Version 9.9 (SOM-FFN v2025). NOAA National Centers for Environmental Information. Dataset. https://doi.org/10.7289/v5z899n6 [accessed on 2026-11-18]. Method: Landschuetzer, P., Gruber, N., Bakker, D.C.E. (2016). Decadal variations and trends of the global ocean carbon sink. Global Biogeochemical Cycles 30, doi:10.1002/2015GB005359. The pCO2 fields are reconstructed from SOCAT v2024 observations (Bakker et al. 2016, doi:10.5194/essd-8-383-2016) with a two-step neural network (SOM-FFN). Variables shown (July 2024, printed time 2024.5): spco2_raw 'surface ocean pCO2 raw' (units as printed: µatm); dco2 'delta pCO2' (units as printed: µatm); atm_co2 'Atmospheric pCO2' (units as printed: µatm). File globals as printed: institution 'Flanders Marine Institute - University of Maryland College Park', run_id 'SOM-FFN_v2025', creation_date '2025-03-01'. dco2, per the README: Delta pCO2 calculated from the CO2 partial pressure difference between atmosphere and the raw surface ocean partial pressure. The file and README do not print the sign convention (which is subtracted from which). atm_co2, per the README: atmospheric pCO2 in uatm for the air-sea flux calculation, derived from the NOAA ESRL (now GML) marine boundary layer reference dry-air mixing ratio of atmospheric CO2 (https://gml.noaa.gov/ccgg/mbl/), SST (Reynolds et al. 2002) and sea-level pressure (Kalnay et al. 1996), following Dickson et al. 2007. Licence: README: "This product is free to be used. Please cite the data as" the citation above. NCEI metadata for Accession 0160558 (https://www.ncei.noaa.gov/data/oceans/ncei/ocads/metadata/0160558.html) prints: "DATA LICENSE: This dataset is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) License." Ocean bathymetry: shaded-relief imagery from the GEBCO WMS (https://wms.gebco.net/, layer GEBCO_LATEST). Imagery reproduced from the GEBCO_2026 Grid. GEBCO Bathymetric Compilation Group 2026 (2026). The GEBCO_2026 Grid - a continuous terrain model for oceans and land at 15 arc-second intervals. NERC EDS British Oceanographic Data Centre NOC. doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation or any purpose relating to safety at sea. Shown re-toned to a light neutral grey (relief shading kept, colours removed). Imagery: NASA GIBS / EOSDIS Worldview, Blue Marble Next Generation (MODIS, cloud-free mosaic; not time-matched to the map window). We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's Earth Science Data and Information System (ESDIS). Basemap vectors: Natural Earth 1:50m (rivers (ne_50m_rivers_lake_centerlines), lakes (ne_50m_lakes), coastline (ne_50m_coastline), country borders (ne_50m_admin_0_boundary_lines_land), land mask (ne_50m_land)). Made with Natural Earth. Free vector and raster map data @ naturalearthdata.com. Public domain.
HOW THIS FIGURE WAS MADE
Made in QGIS. Generated by the PyQGIS script air_sea_pco2_gradient_v2.py, written for Neil Beckwith's QGIS workflow (QGIS 3.44 LTR on Windows, Python Console editor). This render was exported headlessly in QGIS 3.40.6-Bratislava on 2026-11-18 01:56 (local time). Data route: NCEI HTTPS, HTTP range requests via GDAL /vsicurl/ (HDF5 driver): https://www.ncei.noaa.gov/data/oceans/ncei/ocads/data/0160558/MPI_SOM-FFN_v2025/SOM-FFN_v2025_NCEI_OCADS.nc. Only the chosen month of each variable is read; each slice is cached in the home folder as a 32-bit float GeoTIFF (the file stores 64-bit floats; the map classes are far coarser than that precision). The slice is re-ordered from the file's (lon, lat) south-to-north layout to a north-up raster; no value is modified. Fill values (>= 1e19; the file prints 'Fill Value: 1e20') are transparent. No Earthdata token is needed. Data layers drawn at 60% opacity over the basemap; coastline, borders, rivers and lakes are drawn on top (coastline and borders with thin halos; rivers Natural Earth scalerank <= 5, 0.2 mm muted blue). Ocean basemap: GEBCO shaded relief re-toned to light neutral grey so data colours match the legend. Colour ramps: viridis (spco2_raw); ColorBrewer PuOr (diverging at 0; two palest middle classes omitted) (dco2); cividis (atm_co2), chosen as colour-vision-deficiency-friendly, with visible (not near-white) low classes. Equirectangular EPSG:4326 map; the scale bar is approximate at the equator; north is up.
SHORT CAPTION
Surface-ocean pCO2, delta pCO2 and atmospheric pCO2, July 2024: SOM-FFN v2025. (A) surface-ocean pCO2 (spco2_raw 'surface ocean pCO2 raw', µatm); (B) the product's own delta pCO2 (dco2 'delta pCO2', µatm); its sign convention is not printed in the file or README, so colours are not labelled as uptake or outgassing; (C) the product's own atmospheric pCO2 (atm_co2 'Atmospheric pCO2', µatm), which the README says is derived from the NOAA GML marine boundary layer xCO2 reference, SST and sea-level pressure; it is defined over ocean cells only. All panels are the same single month (printed time 2024.5; the record runs 1982 to 2024.92) of the SOM-FFN v2025 observation-based product on a 1 x 1 degree grid; the surface-ocean pCO2 is a neural-network reconstruction from SOCAT v2024 observations, not raw ship measurements. Nothing was derived: the script did not subtract ocean from air, average months or convert units. A single month carries the seasonal state of that month (northern summer for July); it is not an annual mean. Class breaks are display bins only; land and cells without values are transparent. Made in QGIS.
FULL CREDITS
Data: Jersild, A., Landschuetzer, P., Gruber, N. and Bakker, D.C.E. (2017). An observation-based global monthly gridded sea surface pCO2 and air-sea CO2 flux product from 1982 onward and its monthly climatology (NCEI Accession 0160558). Version 9.9 (SOM-FFN v2025). NOAA National Centers for Environmental Information. Dataset. https://doi.org/10.7289/v5z899n6 [accessed on 2026-11-18]. Method: Landschuetzer, P., Gruber, N., Bakker, D.C.E. (2016). Decadal variations and trends of the global ocean carbon sink. Global Biogeochemical Cycles 30, doi:10.1002/2015GB005359. The pCO2 fields are reconstructed from SOCAT v2024 observations (Bakker et al. 2016, doi:10.5194/essd-8-383-2016) with a two-step neural network (SOM-FFN). Variables shown (July 2024, printed time 2024.5): spco2_raw 'surface ocean pCO2 raw' (units as printed: µatm); dco2 'delta pCO2' (units as printed: µatm); atm_co2 'Atmospheric pCO2' (units as printed: µatm). File globals as printed: institution 'Flanders Marine Institute - University of Maryland College Park', run_id 'SOM-FFN_v2025', creation_date '2025-03-01'. dco2, per the README: Delta pCO2 calculated from the CO2 partial pressure difference between atmosphere and the raw surface ocean partial pressure. The file and README do not print the sign convention (which is subtracted from which). atm_co2, per the README: atmospheric pCO2 in uatm for the air-sea flux calculation, derived from the NOAA ESRL (now GML) marine boundary layer reference dry-air mixing ratio of atmospheric CO2 (https://gml.noaa.gov/ccgg/mbl/), SST (Reynolds et al. 2002) and sea-level pressure (Kalnay et al. 1996), following Dickson et al. 2007. Licence: README: "This product is free to be used. Please cite the data as" the citation above. NCEI metadata for Accession 0160558 (https://www.ncei.noaa.gov/data/oceans/ncei/ocads/metadata/0160558.html) prints: "DATA LICENSE: This dataset is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) License." Ocean bathymetry: shaded-relief imagery from the GEBCO WMS (https://wms.gebco.net/, layer GEBCO_LATEST). Imagery reproduced from the GEBCO_2026 Grid. GEBCO Bathymetric Compilation Group 2026 (2026). The GEBCO_2026 Grid - a continuous terrain model for oceans and land at 15 arc-second intervals. NERC EDS British Oceanographic Data Centre NOC. doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation or any purpose relating to safety at sea. Shown re-toned to a light neutral grey (relief shading kept, colours removed). Imagery: NASA GIBS / EOSDIS Worldview, Blue Marble Next Generation (MODIS, cloud-free mosaic; not time-matched to the map window). We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's Earth Science Data and Information System (ESDIS). Basemap vectors: Natural Earth 1:50m (rivers (ne_50m_rivers_lake_centerlines), lakes (ne_50m_lakes), coastline (ne_50m_coastline), country borders (ne_50m_admin_0_boundary_lines_land), land mask (ne_50m_land)). Made with Natural Earth. Free vector and raster map data @ naturalearthdata.com. Public domain.
HOW THIS FIGURE WAS MADE
Made in QGIS. Generated by the PyQGIS script air_sea_pco2_gradient_v2.py, written for Neil Beckwith's QGIS workflow (QGIS 3.44 LTR on Windows, Python Console editor). This render was exported headlessly in QGIS 3.40.6-Bratislava on 2026-11-18 01:56 (local time). Data route: NCEI HTTPS, HTTP range requests via GDAL /vsicurl/ (HDF5 driver): https://www.ncei.noaa.gov/data/oceans/ncei/ocads/data/0160558/MPI_SOM-FFN_v2025/SOM-FFN_v2025_NCEI_OCADS.nc. Only the chosen month of each variable is read; each slice is cached in the home folder as a 32-bit float GeoTIFF (the file stores 64-bit floats; the map classes are far coarser than that precision). The slice is re-ordered from the file's (lon, lat) south-to-north layout to a north-up raster; no value is modified. Fill values (>= 1e19; the file prints 'Fill Value: 1e20') are transparent. No Earthdata token is needed. Data layers drawn at 60% opacity over the basemap; coastline, borders, rivers and lakes are drawn on top (coastline and borders with thin halos; rivers Natural Earth scalerank <= 5, 0.2 mm muted blue). Ocean basemap: GEBCO shaded relief re-toned to light neutral grey so data colours match the legend. Colour ramps: viridis (spco2_raw); ColorBrewer PuOr (diverging at 0; two palest middle classes omitted) (dco2); cividis (atm_co2), chosen as colour-vision-deficiency-friendly, with visible (not near-white) low classes. Equirectangular EPSG:4326 map; the scale bar is approximate at the equator; north is up.
Credits: Landschützer et al., MPI-SOM-FFN, NCEI Accession 0160558, doi:10.7289/v5z899n6 (CC BY 4.0). Imagery reproduced from the GEBCO_2026 Grid, GEBCO Compilation Group (2026), doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation. We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's ESDIS. Made with Natural Earth. Made in QGIS with AI-assisted Python (PyQGIS) scripting.
- Landschützer et al., MPI-SOM-FFN, NCEI 0160558 (CC BY 4.0).
- de Boyer Montégut, C. (2023). Mixed layer depth climatology, SEANOE, doi:10.17882/91774 (CC BY 4.0).
- Imagery reproduced from the GEBCO_2026 Grid, GEBCO Compilation Group (2026), doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation.
- We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's ESDIS.
- Made with Natural Earth.
- Made in QGIS with AI-assisted Python (PyQGIS) scripting.
Full caption and credits
SHORT CAPTION
Global air–sea CO2 flux density (fgco2_clim, January climatology, mol/m2/yr) and mixed-layer depth (mld_dr003, January climatology, m), on GEBCO bathymetry and Blue Marble land imagery. The CO2 flux is transfer across the sea surface, not burial or storage; values and units are shown as the sources print them, with one climatology month and no averaging. The climatology file prints no time axis; step 1 of 12 is read as January (Assumption: steps run January to December). CO2 flux sign convention: Unknown (not printed by the source file or README). Made in QGIS.
FULL CREDITS
Air-sea CO2 flux density: Jersild, A., Landschützer, P., Gruber, N. and Bakker, D.C.E. (2023). An observation-based global monthly gridded sea surface pCO2 and air-sea CO2 flux product from 1982 onward and its monthly climatology (NCEI Accession 0160558). Version 7.7. NOAA National Centers for Environmental Information. Dataset. doi:10.7289/v5z899n6. File MPI_SOM_FFN_v2022_CLIM_NCEI_OCADS.nc, variable fgco2_clim ('CO2 flux clim', units 'mol/m2/yr'), 1° x 1°. [accessed on 2026-11-18]. Method: Landschützer, P., Gruber, N., Bakker, D.C.E. (2016) Decadal variations and trends of the global ocean carbon sink, Global Biogeochemical Cycles 30, doi:10.1002/2015GB005359. "This product is free to be used" with citation (README). Licence: CC BY 4.0 (NCEI data licence for Accession 0160558).
Mixed-layer depth: de Boyer Montégut, C. (2023). Mixed layer depth climatology computed with a density threshold criterion of 0.03kg/m3 from 10 m depth value. SEANOE. doi:10.17882/91774. File mld_dr003_ref10m_v2023.nc, variable mld_dr003 ('Mixed Layer Depth (MLD) estimated through a threshold value of surface potential density equals to +0.03 kg/m3 with reference to the surface value taken at 10 m depth', units 'm'); monthly climatology, 1° x 1°, from profiles January 1970 to December 2021 (Argo and NCEI-NOAA World Ocean Database). Licence CC BY 4.0. [accessed on 2026-11-18]. Method: de Boyer Montégut, C., Madec, G., Fischer, A.S., Lazar, A., Iudicone, D. (2004) Mixed layer depth over the global ocean: An examination of profile data and a profile-based climatology, J. Geophys. Res. 109, C12003, doi:10.1029/2004JC002378.
Ocean bathymetry: shaded-relief imagery from the GEBCO WMS (https://wms.gebco.net/, layer GEBCO_LATEST). Imagery reproduced from the GEBCO_2026 Grid. GEBCO Bathymetric Compilation Group 2026 (2026). The GEBCO_2026 Grid - a continuous terrain model for oceans and land at 15 arc-second intervals. NERC EDS British Oceanographic Data Centre NOC. doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation or any purpose relating to safety at sea.
Imagery: NASA GIBS / EOSDIS Worldview, Blue Marble Next Generation (MODIS, cloud-free mosaic; not time-matched to the map window). We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's Earth Science Data and Information System (ESDIS).
Basemap vectors: Natural Earth 1:50m (rivers (ne_50m_rivers_lake_centerlines), lakes (ne_50m_lakes), coastline (ne_50m_coastline), country borders (ne_50m_admin_0_boundary_lines_land), land mask (ne_50m_land)). Made with Natural Earth. Free vector and raster map data @ naturalearthdata.com. Public domain.
HOW THIS FIGURE WAS MADE
PyQGIS script physics_co2flux_mld_v3.py (QGIS 3.40.6-Bratislava). Data fetched by URL inside the script and cached in ~/qgis_physics_co2flux_mld_cache. One source band (climatology month 1) copied unchanged to GeoTIFF; only the source fill value set to no-data. Classes and colour-blind-safe ramps (ColorBrewer RdBu for CO2 flux, viridis for MLD) chosen for display; CO2 flux values between -0.5 and 0.5 drawn transparent. Equirectangular (EPSG:4326) world map; scale bar approximate at the equator. Made in QGIS.
Global air–sea CO2 flux density (fgco2_clim, January climatology, mol/m2/yr) and mixed-layer depth (mld_dr003, January climatology, m), on GEBCO bathymetry and Blue Marble land imagery. The CO2 flux is transfer across the sea surface, not burial or storage; values and units are shown as the sources print them, with one climatology month and no averaging. The climatology file prints no time axis; step 1 of 12 is read as January (Assumption: steps run January to December). CO2 flux sign convention: Unknown (not printed by the source file or README). Made in QGIS.
FULL CREDITS
Air-sea CO2 flux density: Jersild, A., Landschützer, P., Gruber, N. and Bakker, D.C.E. (2023). An observation-based global monthly gridded sea surface pCO2 and air-sea CO2 flux product from 1982 onward and its monthly climatology (NCEI Accession 0160558). Version 7.7. NOAA National Centers for Environmental Information. Dataset. doi:10.7289/v5z899n6. File MPI_SOM_FFN_v2022_CLIM_NCEI_OCADS.nc, variable fgco2_clim ('CO2 flux clim', units 'mol/m2/yr'), 1° x 1°. [accessed on 2026-11-18]. Method: Landschützer, P., Gruber, N., Bakker, D.C.E. (2016) Decadal variations and trends of the global ocean carbon sink, Global Biogeochemical Cycles 30, doi:10.1002/2015GB005359. "This product is free to be used" with citation (README). Licence: CC BY 4.0 (NCEI data licence for Accession 0160558).
Mixed-layer depth: de Boyer Montégut, C. (2023). Mixed layer depth climatology computed with a density threshold criterion of 0.03kg/m3 from 10 m depth value. SEANOE. doi:10.17882/91774. File mld_dr003_ref10m_v2023.nc, variable mld_dr003 ('Mixed Layer Depth (MLD) estimated through a threshold value of surface potential density equals to +0.03 kg/m3 with reference to the surface value taken at 10 m depth', units 'm'); monthly climatology, 1° x 1°, from profiles January 1970 to December 2021 (Argo and NCEI-NOAA World Ocean Database). Licence CC BY 4.0. [accessed on 2026-11-18]. Method: de Boyer Montégut, C., Madec, G., Fischer, A.S., Lazar, A., Iudicone, D. (2004) Mixed layer depth over the global ocean: An examination of profile data and a profile-based climatology, J. Geophys. Res. 109, C12003, doi:10.1029/2004JC002378.
Ocean bathymetry: shaded-relief imagery from the GEBCO WMS (https://wms.gebco.net/, layer GEBCO_LATEST). Imagery reproduced from the GEBCO_2026 Grid. GEBCO Bathymetric Compilation Group 2026 (2026). The GEBCO_2026 Grid - a continuous terrain model for oceans and land at 15 arc-second intervals. NERC EDS British Oceanographic Data Centre NOC. doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation or any purpose relating to safety at sea.
Imagery: NASA GIBS / EOSDIS Worldview, Blue Marble Next Generation (MODIS, cloud-free mosaic; not time-matched to the map window). We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's Earth Science Data and Information System (ESDIS).
Basemap vectors: Natural Earth 1:50m (rivers (ne_50m_rivers_lake_centerlines), lakes (ne_50m_lakes), coastline (ne_50m_coastline), country borders (ne_50m_admin_0_boundary_lines_land), land mask (ne_50m_land)). Made with Natural Earth. Free vector and raster map data @ naturalearthdata.com. Public domain.
HOW THIS FIGURE WAS MADE
PyQGIS script physics_co2flux_mld_v3.py (QGIS 3.40.6-Bratislava). Data fetched by URL inside the script and cached in ~/qgis_physics_co2flux_mld_cache. One source band (climatology month 1) copied unchanged to GeoTIFF; only the source fill value set to no-data. Classes and colour-blind-safe ramps (ColorBrewer RdBu for CO2 flux, viridis for MLD) chosen for display; CO2 flux values between -0.5 and 0.5 drawn transparent. Equirectangular (EPSG:4326) world map; scale bar approximate at the equator. Made in QGIS.
SHORT CAPTION
Global air–sea CO2 flux density (fgco2_clim, January climatology, mol/m2/yr) and mixed-layer depth (mld_dr003, January climatology, m), on GEBCO bathymetry and Blue Marble land imagery. The CO2 flux is transfer across the sea surface, not burial or storage; values and units are shown as the sources print them, with one climatology month and no averaging. The climatology file prints no time axis; step 1 of 12 is read as January (Assumption: steps run January to December). CO2 flux sign convention: Unknown (not printed by the source file or README). Made in QGIS.
FULL CREDITS
Air-sea CO2 flux density: Jersild, A., Landschützer, P., Gruber, N. and Bakker, D.C.E. (2023). An observation-based global monthly gridded sea surface pCO2 and air-sea CO2 flux product from 1982 onward and its monthly climatology (NCEI Accession 0160558). Version 7.7. NOAA National Centers for Environmental Information. Dataset. doi:10.7289/v5z899n6. File MPI_SOM_FFN_v2022_CLIM_NCEI_OCADS.nc, variable fgco2_clim ('CO2 flux clim', units 'mol/m2/yr'), 1° x 1°. [accessed on 2026-11-18]. Method: Landschützer, P., Gruber, N., Bakker, D.C.E. (2016) Decadal variations and trends of the global ocean carbon sink, Global Biogeochemical Cycles 30, doi:10.1002/2015GB005359. "This product is free to be used" with citation (README). Licence: CC BY 4.0 (NCEI data licence for Accession 0160558).
Mixed-layer depth: de Boyer Montégut, C. (2023). Mixed layer depth climatology computed with a density threshold criterion of 0.03kg/m3 from 10 m depth value. SEANOE. doi:10.17882/91774. File mld_dr003_ref10m_v2023.nc, variable mld_dr003 ('Mixed Layer Depth (MLD) estimated through a threshold value of surface potential density equals to +0.03 kg/m3 with reference to the surface value taken at 10 m depth', units 'm'); monthly climatology, 1° x 1°, from profiles January 1970 to December 2021 (Argo and NCEI-NOAA World Ocean Database). Licence CC BY 4.0. [accessed on 2026-11-18]. Method: de Boyer Montégut, C., Madec, G., Fischer, A.S., Lazar, A., Iudicone, D. (2004) Mixed layer depth over the global ocean: An examination of profile data and a profile-based climatology, J. Geophys. Res. 109, C12003, doi:10.1029/2004JC002378.
Ocean bathymetry: shaded-relief imagery from the GEBCO WMS (https://wms.gebco.net/, layer GEBCO_LATEST). Imagery reproduced from the GEBCO_2026 Grid. GEBCO Bathymetric Compilation Group 2026 (2026). The GEBCO_2026 Grid - a continuous terrain model for oceans and land at 15 arc-second intervals. NERC EDS British Oceanographic Data Centre NOC. doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation or any purpose relating to safety at sea.
Imagery: NASA GIBS / EOSDIS Worldview, Blue Marble Next Generation (MODIS, cloud-free mosaic; not time-matched to the map window). We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's Earth Science Data and Information System (ESDIS).
Basemap vectors: Natural Earth 1:50m (rivers (ne_50m_rivers_lake_centerlines), lakes (ne_50m_lakes), coastline (ne_50m_coastline), country borders (ne_50m_admin_0_boundary_lines_land), land mask (ne_50m_land)). Made with Natural Earth. Free vector and raster map data @ naturalearthdata.com. Public domain.
HOW THIS FIGURE WAS MADE
PyQGIS script physics_co2flux_mld_v3.py (QGIS 3.40.6-Bratislava). Data fetched by URL inside the script and cached in ~/qgis_physics_co2flux_mld_cache. One source band (climatology month 1) copied unchanged to GeoTIFF; only the source fill value set to no-data. Classes and colour-blind-safe ramps (ColorBrewer RdBu for CO2 flux, viridis for MLD) chosen for display; CO2 flux values between -0.5 and 0.5 drawn transparent. Equirectangular (EPSG:4326) world map; scale bar approximate at the equator. Made in QGIS.
Credits: Landschützer et al., MPI-SOM-FFN, NCEI 0160558 (CC BY 4.0). de Boyer Montégut, C. (2023). Mixed layer depth climatology, SEANOE, doi:10.17882/91774 (CC BY 4.0). Imagery reproduced from the GEBCO_2026 Grid, GEBCO Compilation Group (2026), doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation. We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's ESDIS. Made with Natural Earth. Made in QGIS with AI-assisted Python (PyQGIS) scripting.
- Gruber et al. (2019) Science 363, 1193, doi:10.1126/science.aau5153; NCEI 0186034.
- World Ocean Atlas 2023 Vol. 1 Temperature, doi:10.25923/54bh-1613.
- Imagery reproduced from the GEBCO_2026 Grid, GEBCO Compilation Group (2026), doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation.
- We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's ESDIS.
- Made with Natural Earth.
- Made in QGIS with AI-assisted Python (PyQGIS) scripting.
Full caption and credits
SHORT CAPTION
Solubility carbon pump: change in anthropogenic CO2 stored in the ocean, 1994–2007, with sea surface temperature. Physical side: the change in the column inventory of anthropogenic CO2 between 1994 and 2007 (0-3000 m; DCANT_INV01, mol m-2; Gruber et al. 2019) and World Ocean Atlas 2023 sea surface temperature (t_an at 0.0 m, annual 1991-2020, degrees_celsius), on GEBCO bathymetry and Blue Marble land imagery. The inventory is storage of anthropogenic CO2 in the water column, not sediment burial, and is shown as printed (the vertical integration is the authors'). How much of it is due to the solubility pump alone is Unknown (not separated by the source). Showing SST as the driver panel is an Assumption (an interpretive choice); the map computes no relationship between the panels. Values and units as the sources print them; nothing integrated, differenced, averaged or converted by this script. Made in QGIS.
FULL CREDITS
Change in anthropogenic CO2 column inventory, 1994-2007 (0-3000 m): Gruber, N., Clement, D., Carter, B.R., Feely, R.A., van Heuven, S., Hoppema, M., Ishii, M., Key, R.M., Kozyr, A., Lauvset, S.K., Lo Monaco, C., Mathis, J.T., Murata, A., Olsen, A., Perez, F.F., Sabine, C.L., Tanhua, T., Wanninkhof, R. (2019). The oceanic sink for anthropogenic CO2 from 1994 to 2007 - the data (NCEI Accession 0186034). Version 1.1. NOAA National Centers for Environmental Information. Dataset. doi:10.25921/wdn2-pt10. File inv_dcant_emlr_cstar_gruber_94-07_vs1.nc, variable DCANT_INV01 ('Inv C_{ant} Vs101', units 'mol m-2'), 1° x 1°. [accessed on 2026-11-18]. Study: Gruber, N. et al. (2019) The oceanic sink for anthropogenic CO2 from 1994 to 2007, Science 363(6432), 1193-1199, doi:10.1126/science.aau5153. Method: Clement, D. and Gruber, N. (2018) The eMLR(C*) method to determine decadal changes in the global ocean storage of anthropogenic CO2, Global Biogeochem. Cycles 32(4), 654-679, doi:10.1002/2017GB005819. Licence as printed (README): "This product is free to be used." with citation; the NCEI metadata page prints no other licence.
Sea surface temperature: World Ocean Atlas 2023 : sea_water_temperature Annual 1991-2020 1.00 degree; variable t_an ('Objectively analyzed mean fields for sea_water_temperature at standard depth levels.', units 'degrees_celsius') at the printed 0.0 m standard level, annual, 1°. Locarnini, R.A., Mishonov, A.V., Baranova, O.K., Reagan, J.R., Boyer, T.P., Seidov, D., Wang, Z., Garcia, H.E., Bouchard, C., Cross, S.L., Paver, C.R., Dukhovskoy, D. (2023). World Ocean Atlas 2023, Volume 1: Temperature. A. Mishonov Technical Ed., NOAA Atlas NESDIS 89, doi:10.25923/54bh-1613. Accessed via NCEI THREDDS OPeNDAP [accessed on 2026-11-18]. Licence as printed in the file: "These data are openly available to the public. Please acknowledge the use of these data with the text given in the acknowledgment attribute." (the file prints no acknowledgment attribute, so that text is Unknown; cited here with the WOA23 Volume 1 reference above).
Ocean bathymetry: shaded-relief imagery from the GEBCO WMS (https://wms.gebco.net/, layer GEBCO_LATEST). Imagery reproduced from the GEBCO_2026 Grid. GEBCO Bathymetric Compilation Group 2026 (2026). The GEBCO_2026 Grid - a continuous terrain model for oceans and land at 15 arc-second intervals. NERC EDS British Oceanographic Data Centre NOC. doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation or any purpose relating to safety at sea.
Imagery: NASA GIBS / EOSDIS Worldview, Blue Marble Next Generation (MODIS, cloud-free mosaic; not time-matched to the map window). We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's Earth Science Data and Information System (ESDIS).
Basemap vectors: Natural Earth 1:50m (rivers (ne_50m_rivers_lake_centerlines), lakes (ne_50m_lakes), coastline (ne_50m_coastline), country borders (ne_50m_admin_0_boundary_lines_land), land mask (ne_50m_land)). Made with Natural Earth. Free vector and raster map data @ naturalearthdata.com. Public domain.
HOW THIS FIGURE WAS MADE
PyQGIS script physics_solubility_pump_v2.py (QGIS 3.40.6-Bratislava). Data fetched by URL inside the script (NCEI HTTPS file; NCEI THREDDS OPeNDAP ASCII subset) and cached in ~/qgis_physics_solubility_pump_cache. Source grids copied unchanged to GeoTIFF; only the source fill values set to no-data; 0-360 longitudes rolled to -180-180 for display. Classes and colour-blind-safe ramps (ColorBrewer YlGnBu for the inventory, RdYlBu reversed for SST) chosen for display; inventory values <= 0 drawn transparent. Equirectangular (EPSG:4326) world map; scale bar approximate at the equator. Made in QGIS.
Solubility carbon pump: change in anthropogenic CO2 stored in the ocean, 1994–2007, with sea surface temperature. Physical side: the change in the column inventory of anthropogenic CO2 between 1994 and 2007 (0-3000 m; DCANT_INV01, mol m-2; Gruber et al. 2019) and World Ocean Atlas 2023 sea surface temperature (t_an at 0.0 m, annual 1991-2020, degrees_celsius), on GEBCO bathymetry and Blue Marble land imagery. The inventory is storage of anthropogenic CO2 in the water column, not sediment burial, and is shown as printed (the vertical integration is the authors'). How much of it is due to the solubility pump alone is Unknown (not separated by the source). Showing SST as the driver panel is an Assumption (an interpretive choice); the map computes no relationship between the panels. Values and units as the sources print them; nothing integrated, differenced, averaged or converted by this script. Made in QGIS.
FULL CREDITS
Change in anthropogenic CO2 column inventory, 1994-2007 (0-3000 m): Gruber, N., Clement, D., Carter, B.R., Feely, R.A., van Heuven, S., Hoppema, M., Ishii, M., Key, R.M., Kozyr, A., Lauvset, S.K., Lo Monaco, C., Mathis, J.T., Murata, A., Olsen, A., Perez, F.F., Sabine, C.L., Tanhua, T., Wanninkhof, R. (2019). The oceanic sink for anthropogenic CO2 from 1994 to 2007 - the data (NCEI Accession 0186034). Version 1.1. NOAA National Centers for Environmental Information. Dataset. doi:10.25921/wdn2-pt10. File inv_dcant_emlr_cstar_gruber_94-07_vs1.nc, variable DCANT_INV01 ('Inv C_{ant} Vs101', units 'mol m-2'), 1° x 1°. [accessed on 2026-11-18]. Study: Gruber, N. et al. (2019) The oceanic sink for anthropogenic CO2 from 1994 to 2007, Science 363(6432), 1193-1199, doi:10.1126/science.aau5153. Method: Clement, D. and Gruber, N. (2018) The eMLR(C*) method to determine decadal changes in the global ocean storage of anthropogenic CO2, Global Biogeochem. Cycles 32(4), 654-679, doi:10.1002/2017GB005819. Licence as printed (README): "This product is free to be used." with citation; the NCEI metadata page prints no other licence.
Sea surface temperature: World Ocean Atlas 2023 : sea_water_temperature Annual 1991-2020 1.00 degree; variable t_an ('Objectively analyzed mean fields for sea_water_temperature at standard depth levels.', units 'degrees_celsius') at the printed 0.0 m standard level, annual, 1°. Locarnini, R.A., Mishonov, A.V., Baranova, O.K., Reagan, J.R., Boyer, T.P., Seidov, D., Wang, Z., Garcia, H.E., Bouchard, C., Cross, S.L., Paver, C.R., Dukhovskoy, D. (2023). World Ocean Atlas 2023, Volume 1: Temperature. A. Mishonov Technical Ed., NOAA Atlas NESDIS 89, doi:10.25923/54bh-1613. Accessed via NCEI THREDDS OPeNDAP [accessed on 2026-11-18]. Licence as printed in the file: "These data are openly available to the public. Please acknowledge the use of these data with the text given in the acknowledgment attribute." (the file prints no acknowledgment attribute, so that text is Unknown; cited here with the WOA23 Volume 1 reference above).
Ocean bathymetry: shaded-relief imagery from the GEBCO WMS (https://wms.gebco.net/, layer GEBCO_LATEST). Imagery reproduced from the GEBCO_2026 Grid. GEBCO Bathymetric Compilation Group 2026 (2026). The GEBCO_2026 Grid - a continuous terrain model for oceans and land at 15 arc-second intervals. NERC EDS British Oceanographic Data Centre NOC. doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation or any purpose relating to safety at sea.
Imagery: NASA GIBS / EOSDIS Worldview, Blue Marble Next Generation (MODIS, cloud-free mosaic; not time-matched to the map window). We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's Earth Science Data and Information System (ESDIS).
Basemap vectors: Natural Earth 1:50m (rivers (ne_50m_rivers_lake_centerlines), lakes (ne_50m_lakes), coastline (ne_50m_coastline), country borders (ne_50m_admin_0_boundary_lines_land), land mask (ne_50m_land)). Made with Natural Earth. Free vector and raster map data @ naturalearthdata.com. Public domain.
HOW THIS FIGURE WAS MADE
PyQGIS script physics_solubility_pump_v2.py (QGIS 3.40.6-Bratislava). Data fetched by URL inside the script (NCEI HTTPS file; NCEI THREDDS OPeNDAP ASCII subset) and cached in ~/qgis_physics_solubility_pump_cache. Source grids copied unchanged to GeoTIFF; only the source fill values set to no-data; 0-360 longitudes rolled to -180-180 for display. Classes and colour-blind-safe ramps (ColorBrewer YlGnBu for the inventory, RdYlBu reversed for SST) chosen for display; inventory values <= 0 drawn transparent. Equirectangular (EPSG:4326) world map; scale bar approximate at the equator. Made in QGIS.
SHORT CAPTION
Solubility carbon pump: change in anthropogenic CO2 stored in the ocean, 1994–2007, with sea surface temperature. Physical side: the change in the column inventory of anthropogenic CO2 between 1994 and 2007 (0-3000 m; DCANT_INV01, mol m-2; Gruber et al. 2019) and World Ocean Atlas 2023 sea surface temperature (t_an at 0.0 m, annual 1991-2020, degrees_celsius), on GEBCO bathymetry and Blue Marble land imagery. The inventory is storage of anthropogenic CO2 in the water column, not sediment burial, and is shown as printed (the vertical integration is the authors'). How much of it is due to the solubility pump alone is Unknown (not separated by the source). Showing SST as the driver panel is an Assumption (an interpretive choice); the map computes no relationship between the panels. Values and units as the sources print them; nothing integrated, differenced, averaged or converted by this script. Made in QGIS.
FULL CREDITS
Change in anthropogenic CO2 column inventory, 1994-2007 (0-3000 m): Gruber, N., Clement, D., Carter, B.R., Feely, R.A., van Heuven, S., Hoppema, M., Ishii, M., Key, R.M., Kozyr, A., Lauvset, S.K., Lo Monaco, C., Mathis, J.T., Murata, A., Olsen, A., Perez, F.F., Sabine, C.L., Tanhua, T., Wanninkhof, R. (2019). The oceanic sink for anthropogenic CO2 from 1994 to 2007 - the data (NCEI Accession 0186034). Version 1.1. NOAA National Centers for Environmental Information. Dataset. doi:10.25921/wdn2-pt10. File inv_dcant_emlr_cstar_gruber_94-07_vs1.nc, variable DCANT_INV01 ('Inv C_{ant} Vs101', units 'mol m-2'), 1° x 1°. [accessed on 2026-11-18]. Study: Gruber, N. et al. (2019) The oceanic sink for anthropogenic CO2 from 1994 to 2007, Science 363(6432), 1193-1199, doi:10.1126/science.aau5153. Method: Clement, D. and Gruber, N. (2018) The eMLR(C*) method to determine decadal changes in the global ocean storage of anthropogenic CO2, Global Biogeochem. Cycles 32(4), 654-679, doi:10.1002/2017GB005819. Licence as printed (README): "This product is free to be used." with citation; the NCEI metadata page prints no other licence.
Sea surface temperature: World Ocean Atlas 2023 : sea_water_temperature Annual 1991-2020 1.00 degree; variable t_an ('Objectively analyzed mean fields for sea_water_temperature at standard depth levels.', units 'degrees_celsius') at the printed 0.0 m standard level, annual, 1°. Locarnini, R.A., Mishonov, A.V., Baranova, O.K., Reagan, J.R., Boyer, T.P., Seidov, D., Wang, Z., Garcia, H.E., Bouchard, C., Cross, S.L., Paver, C.R., Dukhovskoy, D. (2023). World Ocean Atlas 2023, Volume 1: Temperature. A. Mishonov Technical Ed., NOAA Atlas NESDIS 89, doi:10.25923/54bh-1613. Accessed via NCEI THREDDS OPeNDAP [accessed on 2026-11-18]. Licence as printed in the file: "These data are openly available to the public. Please acknowledge the use of these data with the text given in the acknowledgment attribute." (the file prints no acknowledgment attribute, so that text is Unknown; cited here with the WOA23 Volume 1 reference above).
Ocean bathymetry: shaded-relief imagery from the GEBCO WMS (https://wms.gebco.net/, layer GEBCO_LATEST). Imagery reproduced from the GEBCO_2026 Grid. GEBCO Bathymetric Compilation Group 2026 (2026). The GEBCO_2026 Grid - a continuous terrain model for oceans and land at 15 arc-second intervals. NERC EDS British Oceanographic Data Centre NOC. doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation or any purpose relating to safety at sea.
Imagery: NASA GIBS / EOSDIS Worldview, Blue Marble Next Generation (MODIS, cloud-free mosaic; not time-matched to the map window). We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's Earth Science Data and Information System (ESDIS).
Basemap vectors: Natural Earth 1:50m (rivers (ne_50m_rivers_lake_centerlines), lakes (ne_50m_lakes), coastline (ne_50m_coastline), country borders (ne_50m_admin_0_boundary_lines_land), land mask (ne_50m_land)). Made with Natural Earth. Free vector and raster map data @ naturalearthdata.com. Public domain.
HOW THIS FIGURE WAS MADE
PyQGIS script physics_solubility_pump_v2.py (QGIS 3.40.6-Bratislava). Data fetched by URL inside the script (NCEI HTTPS file; NCEI THREDDS OPeNDAP ASCII subset) and cached in ~/qgis_physics_solubility_pump_cache. Source grids copied unchanged to GeoTIFF; only the source fill values set to no-data; 0-360 longitudes rolled to -180-180 for display. Classes and colour-blind-safe ramps (ColorBrewer YlGnBu for the inventory, RdYlBu reversed for SST) chosen for display; inventory values <= 0 drawn transparent. Equirectangular (EPSG:4326) world map; scale bar approximate at the equator. Made in QGIS.
Credits: Gruber et al. (2019) Science 363, 1193, doi:10.1126/science.aau5153; NCEI 0186034. World Ocean Atlas 2023 Vol. 1 Temperature, doi:10.25923/54bh-1613. Imagery reproduced from the GEBCO_2026 Grid, GEBCO Compilation Group (2026), doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation. We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's ESDIS. Made with Natural Earth. Made in QGIS with AI-assisted Python (PyQGIS) scripting.
- Mouw, C.B. et al. (2016): Global ocean flux rates. PANGAEA, doi:10.1594/PANGAEA.855594 (CC BY 3.0).
- Imagery reproduced from the GEBCO_2026 Grid, GEBCO Compilation Group (2026), doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation.
- We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's ESDIS.
- Made with Natural Earth.
- Made in QGIS with AI-assisted Python (PyQGIS) scripting.
Full caption and credits
Biological carbon pump: measured sinking POC flux near 100 m and near 1000 m
Points are individual measured records of sinking particulate organic carbon (POC) flux from the Mouw et al. (2016) global compilation, mapped at the latitude and longitude each record prints. Colour = the column "POC flux [mg/m**2/day]" exactly as printed (no unit conversion, averaging, summing, ratio, Martin-curve or other depth normalisation). Shape = sampling method as printed in "Type (Data type)": circle = sediment_trap, triangle = th234. Colour classes (viridis, colour-blind-safe) are display breaks only: 1, 5, 10, 25, 50, 100, 250 and above.
Depth selection: records are selected (not altered) by the printed column "Depth water [m] (Sediment trap deployment depth)". Windows (inclusive): near 100 m = 80–120 m; near 1000 m = 800–1200 m. These windows are parameters at the top of the script.
Records shown (counts of records and of distinct printed latitude/longitude pairs):
- Near 100 m: Depth water 80–120 m: 547 records (sediment_trap: 167, th234: 380) at 265 locations; printed deployment/retrieval dates 1984-09-24 to 2009-08-08.
- Near 1000 m: Depth water 800–1200 m: 2583 records (sediment_trap: 2583) at 137 locations; printed deployment/retrieval dates 1979-12-03 to 2012-11-10.
These are discrete measurements from different years, seasons and collection durations, not a synoptic field or climatology. Many locations are time series with several records; overlapping records are not averaged, and higher printed values are drawn on top. Sediment-trap and 234Th records measure flux by different methods and are shown with different shapes; they are not combined.
No th234 record falls in the window "Near 1000 m: Depth water 800–1200 m".
The depth column is labelled in the dataset "Depth water [m] (Sediment trap deployment depth)"; th234 records print their depth in the same column, and the dataset does not print a separate label for what that depth means for a 234Th estimate (Unknown).
Points drawn at their printed coordinates that fall on the Natural Earth 1:50m land polygon (shown as printed, not moved): 10.3, -64.4 (sediment_trap, 11 records, Reference: Montes et al., 2012); 10.3, -64.4 (sediment_trap, 338 records, Reference: Thurnell, 2013, CARIACO Ocean Time Series Data); 46.77, 142.07 (sediment_trap, 20 records, Reference: not printed). These lie at coastlines where the 1:50m land outline and the printed coordinates disagree.
Why this dataset: the Mouw et al. (2016) compilation prints POC flux from both sediment traps and 234Th in one unit (mg/m**2/day) with a printed depth for every record, and is fetched by script as PANGAEA tab-delimited text with no login. Le Moigne (2013, PANGAEA doi:10.1594/PANGAEA.809717; ESSD doi:10.5194/essd-5-295-2013) was checked and is fetchable, but prints "TOC flux" (Carbon, organic, total) rather than POC, and Mouw et al. already include 234Th records, so adding it would risk showing the same measurements twice. Ceballos-Romero et al. (2022) was not verified (no DOI found in a Crossref/DataCite search) and is not used.
Projection: EPSG:4326 (plate carree); graticule every 30 degrees. The scale bar is approximate at the equator (equirectangular world map); north is up.
SOURCES AND CREDITS
- Sinking POC flux: Mouw, C. B.; Barnett, A.; McKinley, G. A.; Gloege, L.; Pilcher, D. (2016) Global ocean flux rates [dataset]. PANGAEA, doi:10.1594/PANGAEA.855594 (in the collection "Global Ocean Particulate Organic Carbon flux merged with satellite parameters", doi:10.1594/PANGAEA.855600). In supplement to: Mouw, C. B., Barnett, A., McKinley, G. A., Gloege, L. and Pilcher, D. (2016) Global ocean particulate organic carbon flux merged with satellite parameters. Earth System Science Data 8, 531-541. doi:10.5194/essd-8-531-2016. Licence: Creative Commons Attribution 3.0 Unported (CC BY 3.0, https://creativecommons.org/licenses/by/3.0/). Fetched from https://doi.pangaea.de/10.1594/PANGAEA.855594?format=textfile. Individual records carry their original source in the dataset's "Reference" column.
- Ocean basemap: shaded-relief imagery from the GEBCO WMS (https://wms.gebco.net/, layer GEBCO_LATEST), reproduced from the GEBCO Grid. GEBCO Bathymetric Compilation Group 2026 (2026). The GEBCO_2026 Grid - a continuous terrain model for oceans and land at 15 arc-second intervals. doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation.
- Land imagery: NASA GIBS, Blue Marble Next Generation (static mosaic, not time-matched). We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's Earth Science Data and Information System (ESDIS).
- Vectors: Natural Earth 1:50m (borders, coastline, rivers, lakes, land mask). Made with Natural Earth. Public domain.
Made in QGIS. Script: bcp_poc_flux_depth_map_v2.py (fetches all inputs by URL; cache in ~/QGISMapCache/bcp_poc_flux).
Points are individual measured records of sinking particulate organic carbon (POC) flux from the Mouw et al. (2016) global compilation, mapped at the latitude and longitude each record prints. Colour = the column "POC flux [mg/m**2/day]" exactly as printed (no unit conversion, averaging, summing, ratio, Martin-curve or other depth normalisation). Shape = sampling method as printed in "Type (Data type)": circle = sediment_trap, triangle = th234. Colour classes (viridis, colour-blind-safe) are display breaks only: 1, 5, 10, 25, 50, 100, 250 and above.
Depth selection: records are selected (not altered) by the printed column "Depth water [m] (Sediment trap deployment depth)". Windows (inclusive): near 100 m = 80–120 m; near 1000 m = 800–1200 m. These windows are parameters at the top of the script.
Records shown (counts of records and of distinct printed latitude/longitude pairs):
- Near 100 m: Depth water 80–120 m: 547 records (sediment_trap: 167, th234: 380) at 265 locations; printed deployment/retrieval dates 1984-09-24 to 2009-08-08.
- Near 1000 m: Depth water 800–1200 m: 2583 records (sediment_trap: 2583) at 137 locations; printed deployment/retrieval dates 1979-12-03 to 2012-11-10.
These are discrete measurements from different years, seasons and collection durations, not a synoptic field or climatology. Many locations are time series with several records; overlapping records are not averaged, and higher printed values are drawn on top. Sediment-trap and 234Th records measure flux by different methods and are shown with different shapes; they are not combined.
No th234 record falls in the window "Near 1000 m: Depth water 800–1200 m".
The depth column is labelled in the dataset "Depth water [m] (Sediment trap deployment depth)"; th234 records print their depth in the same column, and the dataset does not print a separate label for what that depth means for a 234Th estimate (Unknown).
Points drawn at their printed coordinates that fall on the Natural Earth 1:50m land polygon (shown as printed, not moved): 10.3, -64.4 (sediment_trap, 11 records, Reference: Montes et al., 2012); 10.3, -64.4 (sediment_trap, 338 records, Reference: Thurnell, 2013, CARIACO Ocean Time Series Data); 46.77, 142.07 (sediment_trap, 20 records, Reference: not printed). These lie at coastlines where the 1:50m land outline and the printed coordinates disagree.
Why this dataset: the Mouw et al. (2016) compilation prints POC flux from both sediment traps and 234Th in one unit (mg/m**2/day) with a printed depth for every record, and is fetched by script as PANGAEA tab-delimited text with no login. Le Moigne (2013, PANGAEA doi:10.1594/PANGAEA.809717; ESSD doi:10.5194/essd-5-295-2013) was checked and is fetchable, but prints "TOC flux" (Carbon, organic, total) rather than POC, and Mouw et al. already include 234Th records, so adding it would risk showing the same measurements twice. Ceballos-Romero et al. (2022) was not verified (no DOI found in a Crossref/DataCite search) and is not used.
Projection: EPSG:4326 (plate carree); graticule every 30 degrees. The scale bar is approximate at the equator (equirectangular world map); north is up.
SOURCES AND CREDITS
- Sinking POC flux: Mouw, C. B.; Barnett, A.; McKinley, G. A.; Gloege, L.; Pilcher, D. (2016) Global ocean flux rates [dataset]. PANGAEA, doi:10.1594/PANGAEA.855594 (in the collection "Global Ocean Particulate Organic Carbon flux merged with satellite parameters", doi:10.1594/PANGAEA.855600). In supplement to: Mouw, C. B., Barnett, A., McKinley, G. A., Gloege, L. and Pilcher, D. (2016) Global ocean particulate organic carbon flux merged with satellite parameters. Earth System Science Data 8, 531-541. doi:10.5194/essd-8-531-2016. Licence: Creative Commons Attribution 3.0 Unported (CC BY 3.0, https://creativecommons.org/licenses/by/3.0/). Fetched from https://doi.pangaea.de/10.1594/PANGAEA.855594?format=textfile. Individual records carry their original source in the dataset's "Reference" column.
- Ocean basemap: shaded-relief imagery from the GEBCO WMS (https://wms.gebco.net/, layer GEBCO_LATEST), reproduced from the GEBCO Grid. GEBCO Bathymetric Compilation Group 2026 (2026). The GEBCO_2026 Grid - a continuous terrain model for oceans and land at 15 arc-second intervals. doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation.
- Land imagery: NASA GIBS, Blue Marble Next Generation (static mosaic, not time-matched). We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's Earth Science Data and Information System (ESDIS).
- Vectors: Natural Earth 1:50m (borders, coastline, rivers, lakes, land mask). Made with Natural Earth. Public domain.
Made in QGIS. Script: bcp_poc_flux_depth_map_v2.py (fetches all inputs by URL; cache in ~/QGISMapCache/bcp_poc_flux).
Biological carbon pump: measured sinking POC flux near 100 m and near 1000 m
Points are individual measured records of sinking particulate organic carbon (POC) flux from the Mouw et al. (2016) global compilation, mapped at the latitude and longitude each record prints. Colour = the column "POC flux [mg/m**2/day]" exactly as printed (no unit conversion, averaging, summing, ratio, Martin-curve or other depth normalisation). Shape = sampling method as printed in "Type (Data type)": circle = sediment_trap, triangle = th234. Colour classes (viridis, colour-blind-safe) are display breaks only: 1, 5, 10, 25, 50, 100, 250 and above.
Depth selection: records are selected (not altered) by the printed column "Depth water [m] (Sediment trap deployment depth)". Windows (inclusive): near 100 m = 80–120 m; near 1000 m = 800–1200 m. These windows are parameters at the top of the script.
Records shown (counts of records and of distinct printed latitude/longitude pairs):
- Near 100 m: Depth water 80–120 m: 547 records (sediment_trap: 167, th234: 380) at 265 locations; printed deployment/retrieval dates 1984-09-24 to 2009-08-08.
- Near 1000 m: Depth water 800–1200 m: 2583 records (sediment_trap: 2583) at 137 locations; printed deployment/retrieval dates 1979-12-03 to 2012-11-10.
These are discrete measurements from different years, seasons and collection durations, not a synoptic field or climatology. Many locations are time series with several records; overlapping records are not averaged, and higher printed values are drawn on top. Sediment-trap and 234Th records measure flux by different methods and are shown with different shapes; they are not combined.
No th234 record falls in the window "Near 1000 m: Depth water 800–1200 m".
The depth column is labelled in the dataset "Depth water [m] (Sediment trap deployment depth)"; th234 records print their depth in the same column, and the dataset does not print a separate label for what that depth means for a 234Th estimate (Unknown).
Points drawn at their printed coordinates that fall on the Natural Earth 1:50m land polygon (shown as printed, not moved): 10.3, -64.4 (sediment_trap, 11 records, Reference: Montes et al., 2012); 10.3, -64.4 (sediment_trap, 338 records, Reference: Thurnell, 2013, CARIACO Ocean Time Series Data); 46.77, 142.07 (sediment_trap, 20 records, Reference: not printed). These lie at coastlines where the 1:50m land outline and the printed coordinates disagree.
Why this dataset: the Mouw et al. (2016) compilation prints POC flux from both sediment traps and 234Th in one unit (mg/m**2/day) with a printed depth for every record, and is fetched by script as PANGAEA tab-delimited text with no login. Le Moigne (2013, PANGAEA doi:10.1594/PANGAEA.809717; ESSD doi:10.5194/essd-5-295-2013) was checked and is fetchable, but prints "TOC flux" (Carbon, organic, total) rather than POC, and Mouw et al. already include 234Th records, so adding it would risk showing the same measurements twice. Ceballos-Romero et al. (2022) was not verified (no DOI found in a Crossref/DataCite search) and is not used.
Projection: EPSG:4326 (plate carree); graticule every 30 degrees. The scale bar is approximate at the equator (equirectangular world map); north is up.
SOURCES AND CREDITS
- Sinking POC flux: Mouw, C. B.; Barnett, A.; McKinley, G. A.; Gloege, L.; Pilcher, D. (2016) Global ocean flux rates [dataset]. PANGAEA, doi:10.1594/PANGAEA.855594 (in the collection "Global Ocean Particulate Organic Carbon flux merged with satellite parameters", doi:10.1594/PANGAEA.855600). In supplement to: Mouw, C. B., Barnett, A., McKinley, G. A., Gloege, L. and Pilcher, D. (2016) Global ocean particulate organic carbon flux merged with satellite parameters. Earth System Science Data 8, 531-541. doi:10.5194/essd-8-531-2016. Licence: Creative Commons Attribution 3.0 Unported (CC BY 3.0, https://creativecommons.org/licenses/by/3.0/). Fetched from https://doi.pangaea.de/10.1594/PANGAEA.855594?format=textfile. Individual records carry their original source in the dataset's "Reference" column.
- Ocean basemap: shaded-relief imagery from the GEBCO WMS (https://wms.gebco.net/, layer GEBCO_LATEST), reproduced from the GEBCO Grid. GEBCO Bathymetric Compilation Group 2026 (2026). The GEBCO_2026 Grid - a continuous terrain model for oceans and land at 15 arc-second intervals. doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation.
- Land imagery: NASA GIBS, Blue Marble Next Generation (static mosaic, not time-matched). We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's Earth Science Data and Information System (ESDIS).
- Vectors: Natural Earth 1:50m (borders, coastline, rivers, lakes, land mask). Made with Natural Earth. Public domain.
Made in QGIS. Script: bcp_poc_flux_depth_map_v2.py (fetches all inputs by URL; cache in ~/QGISMapCache/bcp_poc_flux).
Credits: Mouw, C.B. et al. (2016): Global ocean flux rates. PANGAEA, doi:10.1594/PANGAEA.855594 (CC BY 3.0). Imagery reproduced from the GEBCO_2026 Grid, GEBCO Compilation Group (2026), doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation. We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's ESDIS. Made with Natural Earth. Made in QGIS with AI-assisted Python (PyQGIS) scripting.
- Contains public sector information licensed under the Open Government Licence v3.0 (BICEP/NCEO export production).
- Derived from ESA Ocean Colour CCI v4.2 data. © ESA 2019. Acknowledge the ESA Climate Change Initiative and the Ocean Colour CCI project.
- Imagery reproduced from the GEBCO_2026 Grid, GEBCO Compilation Group (2026), doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation.
- We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's ESDIS.
- Made with Natural Earth.
- Made in QGIS with AI-assisted Python (PyQGIS) scripting.
Full caption and credits
Global oceanic export production, February 2019 (BICEP/NCEO, EP_Henson)
Map of the BICEP/NCEO monthly oceanic export production variable EP_Henson for February 2019, drawn exactly as stored in the file, in the unit the file prints (mg C m-2 d-1). No unit conversion, averaging, summing or other arithmetic was applied. Colour classes (viridis, colour-blind-safe) are display breaks only. No-data cells are transparent; cells stored as exactly 0 are also drawn transparent. Land imagery is drawn above the data layer, so grid cells show over ocean only. Projection: EPSG:4326 (plate carree); graticule every 30 degrees. The scale bar is approximate at the equator (equirectangular world map); north is up.
Variable description as printed: "Export Production based on Henson et al 2011".
Unit as printed: "mg C m-2 d-1".
Time coverage as printed: 2019-02-01 00:00:00 to 2019-02-28 23:59:59.
Spatial resolution as printed: 9.28 km.
Data coverage as seen in the file: Valid cells in the February 2019 file lie between 78.5°S and 82.0°N; blank ocean has no value in the file. Latitudes are those of the southernmost and northernmost grid rows holding any non-NaN cell. No cause for the gaps is printed in the file attributes, so none is stated here.
Data access: HTTP range read of one variable (GDAL /vsicurl/), full resolution; no striding or resampling of the data grid.
Export depth: Unknown. The file does not print an export depth or horizon for this variable or in its global attributes.
Why this month: February is the default month because, of the twelve 2019 monthly files, it has the most valid EP_Henson grid cells: 5,137,971 cells. Method: for each 2019 monthly file, the non-NaN cells of EP_Henson were counted (a cell count, not a flux value). The June 2019 file, used in the first draft, has 4,557,441 valid cells.
Why this algorithm: the file carries three export estimates (EP_Dunne, EP_Henson, EP_Li) for the same month. EP_Henson is the default because, in the test file (June 2019), it has more valid ocean cells than EP_Dunne (EP_Dunne also needs euphotic depth, Z_eu, per its "Derived from" attribute) and the same valid cells as EP_Li. The three algorithms give different values; only one is shown and none is preferred scientifically by this map. Change VARIABLE at the top of the script to map another.
Why this dataset: the Nowicki, DeVries and Siegel (2022) model output on Figshare (doi:10.6084/m9.figshare.19074521) could not be downloaded by script from the test machine (HTTP 403), so the openly downloadable BICEP/NCEO product was used.
SOURCES AND CREDITS
- Export production: Jönsson, B. F.; Sathyendranath, S.; Kulk, G.; Dall'Olmo, G. (2021). BICEP / NCEO: Monthly global Oceanic Export Production, between 1998-2019 at 9 km resolution (derived from the Ocean Colour Climate Change Initiative v4.2 dataset). NERC EDS Centre for Environmental Data Analysis. doi:10.5285/a6fc730d88fd4935b59d64903715d891. File: BICEP_NCEO_ExportProduction_ESA-OC-L3S-MERGED-1M_MONTHLY_9km_mapped_201902-fv4.2.nc. Variable EP_Henson: "Export Production based on Henson et al 2011"; unit "mg C m-2 d-1"; "Derived from": SST, Primary Production. Algorithm DOI as printed in the file: doi:10.1029/2011gl046735. Dataset description as printed: "Oceanic Export Production Based on Satellite-Derived Properties (see Laws, E. A. (1991), doi:10.1016/0198-0149(91)90059-o. and the cited algorithm references)". Licence: Open Government Licence v3.0 (http://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/). Contains public sector information licensed under the Open Government Licence v3.0.
- ESA: Derived from ESA Ocean Colour CCI v4.2 data. Data provided by the European Space Agency. © ESA 2019. Source chlorophyll as printed in the file (Chl_source_URL): https://catalogue.ceda.ac.uk/uuid/5400de38636d43de9808bfc0b500e863. The European Space Agency (ESA) Climate Change Initiative (CCI) Ocean Colour data are developed and produced through funding and resources from ESA, Copernicus Climate Change Service (C3S) and the NERC EO Data Analysis and AI Service (NEODAAS). OC-CCI reference: Sathyendranath, S. et al. (2019) An ocean-colour time series for use in climate studies: the experience of the Ocean-Colour Climate Change Initiative (OC-CCI). Sensors 19, 4285. doi:10.3390/s19194285. The © ESA year is the data year (2019); the OC-CCI terms (https://artefacts.ceda.ac.uk/licences/specific_licences/esacci_oc_terms_and_conditions.pdf) do not specify a year, so the [year] placeholder in the ESA credit is filled with the data year.
- Ocean basemap: shaded-relief imagery from the GEBCO WMS (https://wms.gebco.net/, layer GEBCO_LATEST), reproduced from the GEBCO Grid. GEBCO Bathymetric Compilation Group 2026 (2026). The GEBCO_2026 Grid - a continuous terrain model for oceans and land at 15 arc-second intervals. doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation.
- Land imagery: NASA GIBS, Blue Marble Next Generation (static mosaic, not time-matched). We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's Earth Science Data and Information System (ESDIS).
- Vectors: Natural Earth 1:50m (borders, coastline, rivers, lakes, land mask). Made with Natural Earth. Public domain.
Made in QGIS 3.40.6-Bratislava. Script: export_production_map.py (fetches all inputs by URL; cache in ~/QGISMapCache/export_production).
Map of the BICEP/NCEO monthly oceanic export production variable EP_Henson for February 2019, drawn exactly as stored in the file, in the unit the file prints (mg C m-2 d-1). No unit conversion, averaging, summing or other arithmetic was applied. Colour classes (viridis, colour-blind-safe) are display breaks only. No-data cells are transparent; cells stored as exactly 0 are also drawn transparent. Land imagery is drawn above the data layer, so grid cells show over ocean only. Projection: EPSG:4326 (plate carree); graticule every 30 degrees. The scale bar is approximate at the equator (equirectangular world map); north is up.
Variable description as printed: "Export Production based on Henson et al 2011".
Unit as printed: "mg C m-2 d-1".
Time coverage as printed: 2019-02-01 00:00:00 to 2019-02-28 23:59:59.
Spatial resolution as printed: 9.28 km.
Data coverage as seen in the file: Valid cells in the February 2019 file lie between 78.5°S and 82.0°N; blank ocean has no value in the file. Latitudes are those of the southernmost and northernmost grid rows holding any non-NaN cell. No cause for the gaps is printed in the file attributes, so none is stated here.
Data access: HTTP range read of one variable (GDAL /vsicurl/), full resolution; no striding or resampling of the data grid.
Export depth: Unknown. The file does not print an export depth or horizon for this variable or in its global attributes.
Why this month: February is the default month because, of the twelve 2019 monthly files, it has the most valid EP_Henson grid cells: 5,137,971 cells. Method: for each 2019 monthly file, the non-NaN cells of EP_Henson were counted (a cell count, not a flux value). The June 2019 file, used in the first draft, has 4,557,441 valid cells.
Why this algorithm: the file carries three export estimates (EP_Dunne, EP_Henson, EP_Li) for the same month. EP_Henson is the default because, in the test file (June 2019), it has more valid ocean cells than EP_Dunne (EP_Dunne also needs euphotic depth, Z_eu, per its "Derived from" attribute) and the same valid cells as EP_Li. The three algorithms give different values; only one is shown and none is preferred scientifically by this map. Change VARIABLE at the top of the script to map another.
Why this dataset: the Nowicki, DeVries and Siegel (2022) model output on Figshare (doi:10.6084/m9.figshare.19074521) could not be downloaded by script from the test machine (HTTP 403), so the openly downloadable BICEP/NCEO product was used.
SOURCES AND CREDITS
- Export production: Jönsson, B. F.; Sathyendranath, S.; Kulk, G.; Dall'Olmo, G. (2021). BICEP / NCEO: Monthly global Oceanic Export Production, between 1998-2019 at 9 km resolution (derived from the Ocean Colour Climate Change Initiative v4.2 dataset). NERC EDS Centre for Environmental Data Analysis. doi:10.5285/a6fc730d88fd4935b59d64903715d891. File: BICEP_NCEO_ExportProduction_ESA-OC-L3S-MERGED-1M_MONTHLY_9km_mapped_201902-fv4.2.nc. Variable EP_Henson: "Export Production based on Henson et al 2011"; unit "mg C m-2 d-1"; "Derived from": SST, Primary Production. Algorithm DOI as printed in the file: doi:10.1029/2011gl046735. Dataset description as printed: "Oceanic Export Production Based on Satellite-Derived Properties (see Laws, E. A. (1991), doi:10.1016/0198-0149(91)90059-o. and the cited algorithm references)". Licence: Open Government Licence v3.0 (http://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/). Contains public sector information licensed under the Open Government Licence v3.0.
- ESA: Derived from ESA Ocean Colour CCI v4.2 data. Data provided by the European Space Agency. © ESA 2019. Source chlorophyll as printed in the file (Chl_source_URL): https://catalogue.ceda.ac.uk/uuid/5400de38636d43de9808bfc0b500e863. The European Space Agency (ESA) Climate Change Initiative (CCI) Ocean Colour data are developed and produced through funding and resources from ESA, Copernicus Climate Change Service (C3S) and the NERC EO Data Analysis and AI Service (NEODAAS). OC-CCI reference: Sathyendranath, S. et al. (2019) An ocean-colour time series for use in climate studies: the experience of the Ocean-Colour Climate Change Initiative (OC-CCI). Sensors 19, 4285. doi:10.3390/s19194285. The © ESA year is the data year (2019); the OC-CCI terms (https://artefacts.ceda.ac.uk/licences/specific_licences/esacci_oc_terms_and_conditions.pdf) do not specify a year, so the [year] placeholder in the ESA credit is filled with the data year.
- Ocean basemap: shaded-relief imagery from the GEBCO WMS (https://wms.gebco.net/, layer GEBCO_LATEST), reproduced from the GEBCO Grid. GEBCO Bathymetric Compilation Group 2026 (2026). The GEBCO_2026 Grid - a continuous terrain model for oceans and land at 15 arc-second intervals. doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation.
- Land imagery: NASA GIBS, Blue Marble Next Generation (static mosaic, not time-matched). We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's Earth Science Data and Information System (ESDIS).
- Vectors: Natural Earth 1:50m (borders, coastline, rivers, lakes, land mask). Made with Natural Earth. Public domain.
Made in QGIS 3.40.6-Bratislava. Script: export_production_map.py (fetches all inputs by URL; cache in ~/QGISMapCache/export_production).
Global oceanic export production, February 2019 (BICEP/NCEO, EP_Henson)
Map of the BICEP/NCEO monthly oceanic export production variable EP_Henson for February 2019, drawn exactly as stored in the file, in the unit the file prints (mg C m-2 d-1). No unit conversion, averaging, summing or other arithmetic was applied. Colour classes (viridis, colour-blind-safe) are display breaks only. No-data cells are transparent; cells stored as exactly 0 are also drawn transparent. Land imagery is drawn above the data layer, so grid cells show over ocean only. Projection: EPSG:4326 (plate carree); graticule every 30 degrees. The scale bar is approximate at the equator (equirectangular world map); north is up.
Variable description as printed: "Export Production based on Henson et al 2011".
Unit as printed: "mg C m-2 d-1".
Time coverage as printed: 2019-02-01 00:00:00 to 2019-02-28 23:59:59.
Spatial resolution as printed: 9.28 km.
Data coverage as seen in the file: Valid cells in the February 2019 file lie between 78.5°S and 82.0°N; blank ocean has no value in the file. Latitudes are those of the southernmost and northernmost grid rows holding any non-NaN cell. No cause for the gaps is printed in the file attributes, so none is stated here.
Data access: HTTP range read of one variable (GDAL /vsicurl/), full resolution; no striding or resampling of the data grid.
Export depth: Unknown. The file does not print an export depth or horizon for this variable or in its global attributes.
Why this month: February is the default month because, of the twelve 2019 monthly files, it has the most valid EP_Henson grid cells: 5,137,971 cells. Method: for each 2019 monthly file, the non-NaN cells of EP_Henson were counted (a cell count, not a flux value). The June 2019 file, used in the first draft, has 4,557,441 valid cells.
Why this algorithm: the file carries three export estimates (EP_Dunne, EP_Henson, EP_Li) for the same month. EP_Henson is the default because, in the test file (June 2019), it has more valid ocean cells than EP_Dunne (EP_Dunne also needs euphotic depth, Z_eu, per its "Derived from" attribute) and the same valid cells as EP_Li. The three algorithms give different values; only one is shown and none is preferred scientifically by this map. Change VARIABLE at the top of the script to map another.
Why this dataset: the Nowicki, DeVries and Siegel (2022) model output on Figshare (doi:10.6084/m9.figshare.19074521) could not be downloaded by script from the test machine (HTTP 403), so the openly downloadable BICEP/NCEO product was used.
SOURCES AND CREDITS
- Export production: Jönsson, B. F.; Sathyendranath, S.; Kulk, G.; Dall'Olmo, G. (2021). BICEP / NCEO: Monthly global Oceanic Export Production, between 1998-2019 at 9 km resolution (derived from the Ocean Colour Climate Change Initiative v4.2 dataset). NERC EDS Centre for Environmental Data Analysis. doi:10.5285/a6fc730d88fd4935b59d64903715d891. File: BICEP_NCEO_ExportProduction_ESA-OC-L3S-MERGED-1M_MONTHLY_9km_mapped_201902-fv4.2.nc. Variable EP_Henson: "Export Production based on Henson et al 2011"; unit "mg C m-2 d-1"; "Derived from": SST, Primary Production. Algorithm DOI as printed in the file: doi:10.1029/2011gl046735. Dataset description as printed: "Oceanic Export Production Based on Satellite-Derived Properties (see Laws, E. A. (1991), doi:10.1016/0198-0149(91)90059-o. and the cited algorithm references)". Licence: Open Government Licence v3.0 (http://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/). Contains public sector information licensed under the Open Government Licence v3.0.
- ESA: Derived from ESA Ocean Colour CCI v4.2 data. Data provided by the European Space Agency. © ESA 2019. Source chlorophyll as printed in the file (Chl_source_URL): https://catalogue.ceda.ac.uk/uuid/5400de38636d43de9808bfc0b500e863. The European Space Agency (ESA) Climate Change Initiative (CCI) Ocean Colour data are developed and produced through funding and resources from ESA, Copernicus Climate Change Service (C3S) and the NERC EO Data Analysis and AI Service (NEODAAS). OC-CCI reference: Sathyendranath, S. et al. (2019) An ocean-colour time series for use in climate studies: the experience of the Ocean-Colour Climate Change Initiative (OC-CCI). Sensors 19, 4285. doi:10.3390/s19194285. The © ESA year is the data year (2019); the OC-CCI terms (https://artefacts.ceda.ac.uk/licences/specific_licences/esacci_oc_terms_and_conditions.pdf) do not specify a year, so the [year] placeholder in the ESA credit is filled with the data year.
- Ocean basemap: shaded-relief imagery from the GEBCO WMS (https://wms.gebco.net/, layer GEBCO_LATEST), reproduced from the GEBCO Grid. GEBCO Bathymetric Compilation Group 2026 (2026). The GEBCO_2026 Grid - a continuous terrain model for oceans and land at 15 arc-second intervals. doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation.
- Land imagery: NASA GIBS, Blue Marble Next Generation (static mosaic, not time-matched). We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's Earth Science Data and Information System (ESDIS).
- Vectors: Natural Earth 1:50m (borders, coastline, rivers, lakes, land mask). Made with Natural Earth. Public domain.
Made in QGIS 3.40.6-Bratislava. Script: export_production_map.py (fetches all inputs by URL; cache in ~/QGISMapCache/export_production).
Credits: Contains public sector information licensed under the Open Government Licence v3.0 (BICEP/NCEO export production). Derived from ESA Ocean Colour CCI v4.2 data. © ESA 2019. Acknowledge the ESA Climate Change Initiative and the Ocean Colour CCI project. Imagery reproduced from the GEBCO_2026 Grid, GEBCO Compilation Group (2026), doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation. We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's ESDIS. Made with Natural Earth. Made in QGIS with AI-assisted Python (PyQGIS) scripting.
- MODIS PIC: NASA OB.DAAC via NOAA CoastWatch.
- Gregor & Gruber, OceanSODA-ETHZ v2025, NCEI 0220059, doi:10.25921/m5wx-ja34.
- Mouw et al. (2016) PANGAEA doi:10.1594/PANGAEA.855594 (CC BY 3.0).
- Imagery reproduced from the GEBCO_2026 Grid, GEBCO Compilation Group (2026), doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation.
- We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's ESDIS.
- Made with Natural Earth.
- Made in QGIS with AI-assisted Python (PyQGIS) scripting.
Full caption and credits
SHORT CAPTION
The Carbonate Counter Pump (CCP), in source terms: marine calcification (the formation of calcium carbonate, CaCO3) is a source of carbon dioxide (Frankignoulle, Canon and Gattuso 1994, "Marine calcification as a source of carbon dioxide: Positive feedback of increasing atmospheric CO2", Limnology and Oceanography 39(2), 458-462, doi:10.4319/lo.1994.39.2.0458). This map shows satellite particulate inorganic carbon (PIC, pic, mol m-3) from MODIS-Aqua (long_name as printed: "Calcite Concentration, Balch and Gordon"); surface calcite saturation state (omega_ca, dimensionless) from OceanSODA-ETHZ; measured PIC flux into sediment traps (PIC flux [mg/m**2/day]) from the Mouw et al. (2016) compilation. It maps where calcite is present, the surface-water carbonate state and measured sinking PIC; it does not show, and no value on it is, a flux of CO2. Panels A and B each show one monthly composite as printed by the provider, for July 2021; trap records span many years and depths and are not time-matched. In panel C, lower PIC flux values are drawn on top of higher ones. Values and units are as printed; nothing was converted, and no ratios (e.g. rain ratios), psi or other derived numbers were calculated. Class breaks are display bins only. Made in QGIS.
FULL CREDITS
A. PIC: NASA Goddard Space Flight Center, Ocean Ecology Laboratory, Ocean Biology Processing Group. MODIS-Aqua Level-3 Mapped Particulate Inorganic Carbon Data, R2018 (4 km, monthly composites); doi:10.5067/AQUA/MODIS/L3M/PIC/2018. Dataset title as printed: "MODIS Aqua, Level-3 SMI, Global, 4km, Particulate Inorganic Carbon, 2003-2022 (Monthly Composite)"; variable pic, long_name "Calcite Concentration, Balch and Gordon", units "mol m-3". Served by NOAA NMFS SWFSC ERD CoastWatch ERDDAP (griddap erdMPICmday), no login; every 12th 4 km cell requested (stride 12). One monthly composite shown, time as printed 2021-07-16T00:00:00Z; only this time slice was fetched (cells with no valid retrieval are blank). Algorithm: Balch et al. (2005), J. Geophys. Res. 110, C07001, doi:10.1029/2004JC002560; Gordon et al. (2001), GRL 28(8), 1587-1590. Licence: NASA Earth Science data and information policy (https://science.nasa.gov/earth-science/earth-science-data/data-information-policy/); free use and redistribution; not intended for legal use. B. omega_ca: Gregor, L. and Gruber, N. OceanSODA-ETHZ: a global gridded data set of the surface ocean carbonate system for seasonal to decadal studies of ocean acidification (NCEI Accession 0220059), product OSETHZ-v2025.OCADS, version v2025.OCADS; NOAA National Centers for Environmental Information, doi:10.25921/m5wx-ja34. Method: Gregor and Gruber (2021), Earth Syst. Sci. Data 13, 777-808, doi:10.5194/essd-13-777-2021. Variable omega_ca, standard_name "calcite saturation state", units "dimensionless" as printed; 1 degree monthly grid. Per the product README, total alkalinity and fCO2 are machine-learning estimates and the other carbonate-system variables, including omega_ca, are calculated from them by the producers with PyCO2SYS; this map does no further calculation. One monthly slice shown, time as printed 14426 days since 1982-01-15 (2021-07-15); only this slice was read. Fetched from https://www.ncei.noaa.gov/data/oceans/ncei/ocads/data/0220059/OceanSODA_ETHZ-v2025.OCADS.01-1982-2024.nc by HTTP range reads, no login. Licence: not printed in the file or README; see the NCEI Accession 0220059 page. C. Sediment-trap PIC flux: Mouw, C.B., Barnett, A., McKinley, G.A., Gloege, L. and Pilcher, D. (2016): Global ocean flux rates [dataset]. PANGAEA, doi:10.1594/PANGAEA.855594; in supplement to Mouw et al. (2016), Global ocean particulate organic carbon flux merged with satellite parameters, Earth Syst. Sci. Data 8, 531-541, doi:10.5194/essd-8-531-2016. Licence: CC-BY-3.0 (as shown on PANGAEA). Field shown: "PIC flux [mg/m**2/day]" as printed, one point per record with a printed value (5870 records at 141 locations); trap deployment depth, dates, trap type, standard deviation and original reference are kept in the layer attribute table. Records overlap at repeat stations; draw order: records are sorted by PIC flux in descending order (QGIS order-by on pic_flux, descending), so lower values are drawn last, on top. Symbols: minimum 1.8 mm, graduated up to 4.5 mm, 0.25 mm dark outline on every symbol. The compilation's original data sources are listed on the PANGAEA page. Ocean bathymetry: shaded-relief imagery from the GEBCO WMS (https://wms.gebco.net/, layer GEBCO_LATEST). Imagery reproduced from the GEBCO_2026 Grid. GEBCO Bathymetric Compilation Group 2026 (2026). The GEBCO_2026 Grid - a continuous terrain model for oceans and land at 15 arc-second intervals. NERC EDS British Oceanographic Data Centre NOC. doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation or any purpose relating to safety at sea. Imagery: NASA GIBS / EOSDIS Worldview, Blue Marble Next Generation (MODIS, cloud-free mosaic; not time-matched to the map window). We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's Earth Science Data and Information System (ESDIS). Basemap vectors: Natural Earth 1:50m (rivers (ne_50m_rivers_lake_centerlines), lakes (ne_50m_lakes), coastline (ne_50m_coastline), country borders (ne_50m_admin_0_boundary_lines_land), land mask (ne_50m_land)). Made with Natural Earth. Free vector and raster map data @ naturalearthdata.com. Public domain.
HOW THIS FIGURE WAS MADE
Made in QGIS. Generated by the PyQGIS script carbonate_counter_pump_v2.py, written for Neil Beckwith's QGIS workflow (QGIS 3.44 LTR on Windows, Python Console editor). This render was exported headlessly in QGIS 3.40.6-Bratislava on 2026-11-18 01:58 (local time). All inputs were fetched by URL inside the script (no logins) and cached in the home folder. Data layers drawn at 80% opacity (points 85%) over the basemap; no-data cells are transparent. Colour ramps: viridis (PIC), plasma (omega_ca), ColorBrewer YlOrRd (steps 4-9, no near-white step), dark outlines (trap PIC flux), chosen as colour-vision-deficiency-friendly. Equirectangular EPSG:4326 world maps; scale bar approximate at the equator; north arrow shown (north is up).
The Carbonate Counter Pump (CCP), in source terms: marine calcification (the formation of calcium carbonate, CaCO3) is a source of carbon dioxide (Frankignoulle, Canon and Gattuso 1994, "Marine calcification as a source of carbon dioxide: Positive feedback of increasing atmospheric CO2", Limnology and Oceanography 39(2), 458-462, doi:10.4319/lo.1994.39.2.0458). This map shows satellite particulate inorganic carbon (PIC, pic, mol m-3) from MODIS-Aqua (long_name as printed: "Calcite Concentration, Balch and Gordon"); surface calcite saturation state (omega_ca, dimensionless) from OceanSODA-ETHZ; measured PIC flux into sediment traps (PIC flux [mg/m**2/day]) from the Mouw et al. (2016) compilation. It maps where calcite is present, the surface-water carbonate state and measured sinking PIC; it does not show, and no value on it is, a flux of CO2. Panels A and B each show one monthly composite as printed by the provider, for July 2021; trap records span many years and depths and are not time-matched. In panel C, lower PIC flux values are drawn on top of higher ones. Values and units are as printed; nothing was converted, and no ratios (e.g. rain ratios), psi or other derived numbers were calculated. Class breaks are display bins only. Made in QGIS.
FULL CREDITS
A. PIC: NASA Goddard Space Flight Center, Ocean Ecology Laboratory, Ocean Biology Processing Group. MODIS-Aqua Level-3 Mapped Particulate Inorganic Carbon Data, R2018 (4 km, monthly composites); doi:10.5067/AQUA/MODIS/L3M/PIC/2018. Dataset title as printed: "MODIS Aqua, Level-3 SMI, Global, 4km, Particulate Inorganic Carbon, 2003-2022 (Monthly Composite)"; variable pic, long_name "Calcite Concentration, Balch and Gordon", units "mol m-3". Served by NOAA NMFS SWFSC ERD CoastWatch ERDDAP (griddap erdMPICmday), no login; every 12th 4 km cell requested (stride 12). One monthly composite shown, time as printed 2021-07-16T00:00:00Z; only this time slice was fetched (cells with no valid retrieval are blank). Algorithm: Balch et al. (2005), J. Geophys. Res. 110, C07001, doi:10.1029/2004JC002560; Gordon et al. (2001), GRL 28(8), 1587-1590. Licence: NASA Earth Science data and information policy (https://science.nasa.gov/earth-science/earth-science-data/data-information-policy/); free use and redistribution; not intended for legal use. B. omega_ca: Gregor, L. and Gruber, N. OceanSODA-ETHZ: a global gridded data set of the surface ocean carbonate system for seasonal to decadal studies of ocean acidification (NCEI Accession 0220059), product OSETHZ-v2025.OCADS, version v2025.OCADS; NOAA National Centers for Environmental Information, doi:10.25921/m5wx-ja34. Method: Gregor and Gruber (2021), Earth Syst. Sci. Data 13, 777-808, doi:10.5194/essd-13-777-2021. Variable omega_ca, standard_name "calcite saturation state", units "dimensionless" as printed; 1 degree monthly grid. Per the product README, total alkalinity and fCO2 are machine-learning estimates and the other carbonate-system variables, including omega_ca, are calculated from them by the producers with PyCO2SYS; this map does no further calculation. One monthly slice shown, time as printed 14426 days since 1982-01-15 (2021-07-15); only this slice was read. Fetched from https://www.ncei.noaa.gov/data/oceans/ncei/ocads/data/0220059/OceanSODA_ETHZ-v2025.OCADS.01-1982-2024.nc by HTTP range reads, no login. Licence: not printed in the file or README; see the NCEI Accession 0220059 page. C. Sediment-trap PIC flux: Mouw, C.B., Barnett, A., McKinley, G.A., Gloege, L. and Pilcher, D. (2016): Global ocean flux rates [dataset]. PANGAEA, doi:10.1594/PANGAEA.855594; in supplement to Mouw et al. (2016), Global ocean particulate organic carbon flux merged with satellite parameters, Earth Syst. Sci. Data 8, 531-541, doi:10.5194/essd-8-531-2016. Licence: CC-BY-3.0 (as shown on PANGAEA). Field shown: "PIC flux [mg/m**2/day]" as printed, one point per record with a printed value (5870 records at 141 locations); trap deployment depth, dates, trap type, standard deviation and original reference are kept in the layer attribute table. Records overlap at repeat stations; draw order: records are sorted by PIC flux in descending order (QGIS order-by on pic_flux, descending), so lower values are drawn last, on top. Symbols: minimum 1.8 mm, graduated up to 4.5 mm, 0.25 mm dark outline on every symbol. The compilation's original data sources are listed on the PANGAEA page. Ocean bathymetry: shaded-relief imagery from the GEBCO WMS (https://wms.gebco.net/, layer GEBCO_LATEST). Imagery reproduced from the GEBCO_2026 Grid. GEBCO Bathymetric Compilation Group 2026 (2026). The GEBCO_2026 Grid - a continuous terrain model for oceans and land at 15 arc-second intervals. NERC EDS British Oceanographic Data Centre NOC. doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation or any purpose relating to safety at sea. Imagery: NASA GIBS / EOSDIS Worldview, Blue Marble Next Generation (MODIS, cloud-free mosaic; not time-matched to the map window). We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's Earth Science Data and Information System (ESDIS). Basemap vectors: Natural Earth 1:50m (rivers (ne_50m_rivers_lake_centerlines), lakes (ne_50m_lakes), coastline (ne_50m_coastline), country borders (ne_50m_admin_0_boundary_lines_land), land mask (ne_50m_land)). Made with Natural Earth. Free vector and raster map data @ naturalearthdata.com. Public domain.
HOW THIS FIGURE WAS MADE
Made in QGIS. Generated by the PyQGIS script carbonate_counter_pump_v2.py, written for Neil Beckwith's QGIS workflow (QGIS 3.44 LTR on Windows, Python Console editor). This render was exported headlessly in QGIS 3.40.6-Bratislava on 2026-11-18 01:58 (local time). All inputs were fetched by URL inside the script (no logins) and cached in the home folder. Data layers drawn at 80% opacity (points 85%) over the basemap; no-data cells are transparent. Colour ramps: viridis (PIC), plasma (omega_ca), ColorBrewer YlOrRd (steps 4-9, no near-white step), dark outlines (trap PIC flux), chosen as colour-vision-deficiency-friendly. Equirectangular EPSG:4326 world maps; scale bar approximate at the equator; north arrow shown (north is up).
SHORT CAPTION
The Carbonate Counter Pump (CCP), in source terms: marine calcification (the formation of calcium carbonate, CaCO3) is a source of carbon dioxide (Frankignoulle, Canon and Gattuso 1994, "Marine calcification as a source of carbon dioxide: Positive feedback of increasing atmospheric CO2", Limnology and Oceanography 39(2), 458-462, doi:10.4319/lo.1994.39.2.0458). This map shows satellite particulate inorganic carbon (PIC, pic, mol m-3) from MODIS-Aqua (long_name as printed: "Calcite Concentration, Balch and Gordon"); surface calcite saturation state (omega_ca, dimensionless) from OceanSODA-ETHZ; measured PIC flux into sediment traps (PIC flux [mg/m**2/day]) from the Mouw et al. (2016) compilation. It maps where calcite is present, the surface-water carbonate state and measured sinking PIC; it does not show, and no value on it is, a flux of CO2. Panels A and B each show one monthly composite as printed by the provider, for July 2021; trap records span many years and depths and are not time-matched. In panel C, lower PIC flux values are drawn on top of higher ones. Values and units are as printed; nothing was converted, and no ratios (e.g. rain ratios), psi or other derived numbers were calculated. Class breaks are display bins only. Made in QGIS.
FULL CREDITS
A. PIC: NASA Goddard Space Flight Center, Ocean Ecology Laboratory, Ocean Biology Processing Group. MODIS-Aqua Level-3 Mapped Particulate Inorganic Carbon Data, R2018 (4 km, monthly composites); doi:10.5067/AQUA/MODIS/L3M/PIC/2018. Dataset title as printed: "MODIS Aqua, Level-3 SMI, Global, 4km, Particulate Inorganic Carbon, 2003-2022 (Monthly Composite)"; variable pic, long_name "Calcite Concentration, Balch and Gordon", units "mol m-3". Served by NOAA NMFS SWFSC ERD CoastWatch ERDDAP (griddap erdMPICmday), no login; every 12th 4 km cell requested (stride 12). One monthly composite shown, time as printed 2021-07-16T00:00:00Z; only this time slice was fetched (cells with no valid retrieval are blank). Algorithm: Balch et al. (2005), J. Geophys. Res. 110, C07001, doi:10.1029/2004JC002560; Gordon et al. (2001), GRL 28(8), 1587-1590. Licence: NASA Earth Science data and information policy (https://science.nasa.gov/earth-science/earth-science-data/data-information-policy/); free use and redistribution; not intended for legal use. B. omega_ca: Gregor, L. and Gruber, N. OceanSODA-ETHZ: a global gridded data set of the surface ocean carbonate system for seasonal to decadal studies of ocean acidification (NCEI Accession 0220059), product OSETHZ-v2025.OCADS, version v2025.OCADS; NOAA National Centers for Environmental Information, doi:10.25921/m5wx-ja34. Method: Gregor and Gruber (2021), Earth Syst. Sci. Data 13, 777-808, doi:10.5194/essd-13-777-2021. Variable omega_ca, standard_name "calcite saturation state", units "dimensionless" as printed; 1 degree monthly grid. Per the product README, total alkalinity and fCO2 are machine-learning estimates and the other carbonate-system variables, including omega_ca, are calculated from them by the producers with PyCO2SYS; this map does no further calculation. One monthly slice shown, time as printed 14426 days since 1982-01-15 (2021-07-15); only this slice was read. Fetched from https://www.ncei.noaa.gov/data/oceans/ncei/ocads/data/0220059/OceanSODA_ETHZ-v2025.OCADS.01-1982-2024.nc by HTTP range reads, no login. Licence: not printed in the file or README; see the NCEI Accession 0220059 page. C. Sediment-trap PIC flux: Mouw, C.B., Barnett, A., McKinley, G.A., Gloege, L. and Pilcher, D. (2016): Global ocean flux rates [dataset]. PANGAEA, doi:10.1594/PANGAEA.855594; in supplement to Mouw et al. (2016), Global ocean particulate organic carbon flux merged with satellite parameters, Earth Syst. Sci. Data 8, 531-541, doi:10.5194/essd-8-531-2016. Licence: CC-BY-3.0 (as shown on PANGAEA). Field shown: "PIC flux [mg/m**2/day]" as printed, one point per record with a printed value (5870 records at 141 locations); trap deployment depth, dates, trap type, standard deviation and original reference are kept in the layer attribute table. Records overlap at repeat stations; draw order: records are sorted by PIC flux in descending order (QGIS order-by on pic_flux, descending), so lower values are drawn last, on top. Symbols: minimum 1.8 mm, graduated up to 4.5 mm, 0.25 mm dark outline on every symbol. The compilation's original data sources are listed on the PANGAEA page. Ocean bathymetry: shaded-relief imagery from the GEBCO WMS (https://wms.gebco.net/, layer GEBCO_LATEST). Imagery reproduced from the GEBCO_2026 Grid. GEBCO Bathymetric Compilation Group 2026 (2026). The GEBCO_2026 Grid - a continuous terrain model for oceans and land at 15 arc-second intervals. NERC EDS British Oceanographic Data Centre NOC. doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation or any purpose relating to safety at sea. Imagery: NASA GIBS / EOSDIS Worldview, Blue Marble Next Generation (MODIS, cloud-free mosaic; not time-matched to the map window). We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's Earth Science Data and Information System (ESDIS). Basemap vectors: Natural Earth 1:50m (rivers (ne_50m_rivers_lake_centerlines), lakes (ne_50m_lakes), coastline (ne_50m_coastline), country borders (ne_50m_admin_0_boundary_lines_land), land mask (ne_50m_land)). Made with Natural Earth. Free vector and raster map data @ naturalearthdata.com. Public domain.
HOW THIS FIGURE WAS MADE
Made in QGIS. Generated by the PyQGIS script carbonate_counter_pump_v2.py, written for Neil Beckwith's QGIS workflow (QGIS 3.44 LTR on Windows, Python Console editor). This render was exported headlessly in QGIS 3.40.6-Bratislava on 2026-11-18 01:58 (local time). All inputs were fetched by URL inside the script (no logins) and cached in the home folder. Data layers drawn at 80% opacity (points 85%) over the basemap; no-data cells are transparent. Colour ramps: viridis (PIC), plasma (omega_ca), ColorBrewer YlOrRd (steps 4-9, no near-white step), dark outlines (trap PIC flux), chosen as colour-vision-deficiency-friendly. Equirectangular EPSG:4326 world maps; scale bar approximate at the equator; north arrow shown (north is up).
Credits: MODIS PIC: NASA OB.DAAC via NOAA CoastWatch. Gregor & Gruber, OceanSODA-ETHZ v2025, NCEI 0220059, doi:10.25921/m5wx-ja34. Mouw et al. (2016) PANGAEA doi:10.1594/PANGAEA.855594 (CC BY 3.0). Imagery reproduced from the GEBCO_2026 Grid, GEBCO Compilation Group (2026), doi:10.5285/4f68d5c7-45eb-f999-e063-7086abc036fa. Not to be used for navigation. We acknowledge the use of imagery provided by services from NASA's Global Imagery Browse Services (GIBS), part of NASA's ESDIS. Made with Natural Earth. Made in QGIS with AI-assisted Python (PyQGIS) scripting.