UEx-Eddies: a biogeochemical long lived mesoscale eddy trajectories for studying air-sea CO2 fluxes

Changelog v0-1 – Preliminary release of the dataset before manuscript submission Introduction Mesoscale eddies are globally known to affect the physical, chemical and biological properties of the oceans, compared to their surrounding environment. These features are associated with radii on the order 100 km, lifetimes from a few days to multiple years, and can transit ocean basins transporting distinct water masses across them (Chelton et al., 2011; Pegliasco et al., 2022). Eddies generally fall into two categories; (1) anticyclonic and (2) cyclonic. Anticyclonic eddies are associated with high pressure centres, anticlockwise rotation in the Northern Hemisphere, warmer sea surface temperatures (SST), and a depression of isopycnals (and downwelling in the eddy core). Whereas cyclonic eddies are generally the opposite; low pressure centres, clockwise rotation in the Northern Hemisphere, cooler SSTs, and an elevation of isopycnals (and upwelling in the eddy core). During their lifetimes, these eddies can alter the air-sea CO2 exchange through their modification of the ocean and atmospheric properties. As the CO2 solubility in seawater is highly temperature sensitive, the fCO2 (sw) in anticyclonic eddies would theoretically be elevated and therefore the features may act as a weaker CO2 sink or stronger CO2 source compared to the surrounding environment. Conversely the opposite may be true for cyclonic eddies, with reduced fCO2 (sw), and therefore capacity to act as a stronger CO2 sink. Despite the abundance of mesoscale eddies, previous studies generally investigate singular eddies (Chen et al., 2007; Jones et al., 2017; Pezzi et al., 2021) or a regional subset of eddies (Ford et al., 2023; Orselli et al., 2019; Song et al., 2016) and their effect on the air-sea CO2 flux. The regional or global cumulative effect of eddies on the air-sea CO2 flux is still under investigation. Ford et al. (2023), used a Lagrangian tracking approach and suggested that long-lived (lifetimes greater than one year) mesoscale eddies enhanced the air-sea CO2 flux in the South Atlantic Ocean by ~0.05 Tg C yr-1 (~0.08%). Guo and Timmermans (2024) used a spatial and timeseries decomposition to extract the mesoscale flow impact on the air-sea CO2 fluxes globally, and estimate a small integrated effect on the order of Tg C yr-1 (compared to global ocean uptake of ~2.9 Pg C yr-1). However, this may include mesoscale signals not related to mesoscale eddies. In this dataset we provide a global dataset of long lived (lifetimes greater than one year) mesoscale eddies (N = 5996) and their associated air-sea CO2 fluxes tracked in a Lagrangian mode between 1993 and 2022. The methodology refines the approach described in Ford et al. (2023), using a global neural network as submitted to the Global Carbon Budget (UExP-FNN-U) to estimate the fCO2 (sw) (Ford et al., 2024; Friedlingstein et al., 2025). We prioritise the use of climate quality satellite data records (Embury et al., 2024; Mears et al., 2022a; Sathyendranath et al., 2019) within the analysis. The uncertainties on the air-sea CO2 fluxes are systematically assessed following the work of Ford et al. (2024). These refinements provide a robust foundation to studying the modulation of air-sea CO2 flux by mesoscale eddies, with a systematic uncertainty budget. Data records The data are provided as individual netCDF files labelled with the AVISO+ META3.2 DT all sat eddy number. Anticyclonic eddies are found in the ‘n_anticylonic’ folder, and cyclonic eddies are within the ‘n_cyclonic’ folder. For each eddy the netCDF file contains all the variables, which comprise of daily timeseries as well as monthly timeseries (these are denoted by their dimensions). Initially the data from the AVISO+ eddy trajectory product have been included within the netCDF, and comprises the location (longitude, latitude), size (radius and polygon shape denoting the eddy) and eddy properties (amplitude, observation flag). Environmental parameters (see Table 1 for list) for the eddy are provided at daily resolution as statistics including mean, median, standard deviation, interquartile range, total number of samples (total number of pixels), valid number of samples (number of pixels that have a value). These statistics are provided for the eddy region (denoted with ‘_in’) and for outside the eddy within three eddy radii (denoted with ‘_out’). For example the European Space Agency Climate Change Iniativie SST daily means inside the eddy have names ‘cci_sst_in_mean’ and outside the eddy ‘cci_sst_out_mean’. Table 1: List of environmental data Parmeter Dataset Source Reference Sea Surface Temperature CCI-SST (v3) (Good & Embury, 2024) (Embury et al., 2024) Sea Surface Salinity CMEMS GLORYS12V1 (CMEMS, 2021) (Jean-Michel et al., 2021) Mixed Layer Depth CMEMS GLORYS12V1 (CMEMS, 2021) (Jean-Michel et al., 2021) Wind Speed CCMP v3.1 (Mears et al., 2022b) (Mears et al., 2022a) Chlorophyll-a OC-CCI (v6) (Sathyendranath et al., 2023) (Sathyendranath et al., 2019) Variables are then aggregated into monthly means as the mean of the daily median (this is to reduce the effects of outliers in the daily data). These are denoted with the term ‘month_’ added to the start of the variable name, for example ‘month_cci_sst_in_median’. Additionally for some variables an anomaly is calculated with respect to a monthly 1 degree climatology from 1985 to 2022. These variables have been appended with ‘_anom’ at the end of the variable name, for example, ‘month_cci_sst_in_median_anom’. These monthly timeseries were then provided to the UExP-FNN-U neural network approach for estimating fCO2 (sw) – see Ford et al. for more information. The fCO2 (sw) are provided for within and outside the eddy, and for the physics only UExP-FNN-U as well as a secondary version with chlorophyll-a as input. The physics only UExP-FNN-U fCO2 (sw) within the eddy for example can be found in ‘month_fco2_sw_in_physics’. The biological version changes ‘_physics’ for ‘_bio’. The UExP-FNN-U also comes with uncertainty estimates for three components: (1) the network uncertainty (‘_net_unc’), (2) parameter uncertainty (‘_para_unc’) and (3) evaluation uncertainty (‘_val_unc’). These are combined in quadrature to provide the total fCO2 (sw) uncertainty (‘_tot_unc’). The total fCO2 (sw) uncertainty for the UExP-FNN-U physics only within the eddy has a variable name ‘month_foc2_tot_unc_in_physics’. The air-sea CO2 fluxes are provided for both variants of the UExP-FNN-U as well as within and outside the eddy. These are provided in variables named, for example ‘flux_in_physics’, with a corresponding total uncertainty denoted as ‘flux_unc_in_physics’. These have been converted to a CO2 flux that accounts for the number of days in the month, and the area of eddy and are denoted by variable ‘flux_in_physics_areaday’ (note there is no flux uncertainty variable for this intermediate stage). Finally these are cumulatively added to provide the CO2 flux over the eddy lifetime, denoted ‘flux_in_physics_areaday_cumulative’. These cumulative CO2 fluxes are paired with a total uncertainty denoted by ‘flux_unc_tot_in_physics_areaday_cumulative’. The individual components of the flux uncertainties are provided for the fluxes and the cumulative fluxes, considering temporal correlations where applicable. Please see the netCDF file for all the components covered (or Ford et al. (2024)). Additional variables are provided within the netCDF files, which have metadata to indicate the data within them. This is true of all the variables described in this document – please see the netCDF files for all available fields and usage information. Additional Information Please contact Daniel J. Ford ([e-mail]) if there are any questions. How to cite these data Please cite this dataset DOI, as well as the underlying manuscript (Ford et al., in prep) Acknowledgements DJF and JDS were supported by funding from the European Space Agency under the projects ‘Satellite-based observations of Carbon in the Ocean: Pools, Fluxes and Exchanges’ (SCOPE; 4000142532/23/I-DT) and ‘Ocean Carbon for Climate’ (OC4C; 3-18399/24/I-NB). GHT and VK were supported by The Atlantic Meridional Transect is funded by the UK Natural Environment Research Council through its National Capability Long-term Single Centre Science Programme, Climate Linked Atlantic Sector Science (grant number NE/R015953/1). The Surface Ocean CO₂ Atlas (SOCAT) is an international effort, endorsed by the International Ocean Carbon Coordination Project (IOCCP), the Surface Ocean Lower Atmosphere Study (SOLAS) and the Integrated Marine Biosphere Research (IMBeR) program, to deliver a uniformly quality-controlled surface ocean CO₂ database. The many researchers and funding agencies responsible for the collection of data and quality control are thanked for their contributions to SOCAT. References Chelton, D. B., Schlax, M. G., & Samelson, R. M. (2011). Global observations of nonlinear mesoscale eddies. Progress in Oceanography, 91(2), 167–216. https://doi.org/10.1016/j.pocean.2011.01.002 Chen, F., Cai, W. J., Benitez-Nelson, C., & Wang, Y. (2007). Sea surface pCO2-SST relationships across a cold-core cyclonic eddy: Implications for understanding regional variability and air-sea gas exchange. Geophysical Research Letters, 34(10). https://doi.org/10.1029/2006GL028058 CMEMS. (2021). Copernicus Marine Modelling Service global ocean physics reanalysis product (GLORYS12V1). Copernicus Marine Modelling Service [Data Set]. https://doi.org/10.48670/moi-00021 Embury, O., Merchant, C. J., Good, S. A., Rayner, N. A., Høyer, J. L., Atkinson, C., et al. (2024). Satellite-based time-series of sea-surface temperature since 1980 for climate applications. Scientific Data, 11(1), 326. https://doi.org/10.1038/s41597-024-03147-w Ford, D. J., Tilstone, G. H., Shutler, J. D., Kitidis, V., Sheen, K. L., Dall’Olmo, G., & Orselli, I. B. M. (2023). Mesoscale Eddies Enhance the Air‐Sea CO 2 Sink in the South Atlantic Ocean. Geophysical Research Letters, 50(9), e2022GL102137. https://doi.org/10.1029/2022GL102137 Ford, D. J., Blannin, J., Watts, J., Watson, A. J., Landschützer, P., Jersild, A., & Shutler, J. D. (2024). A Comprehensive Analysis of Air‐Sea CO2 Flux Uncertainties Constructed From Surface Ocean Data Products. Global Biogeochemical Cycles, 38(11), e2024GB008188. https://doi.org/10.1029/2024GB008188 Ford, D. J., Shutler, J. D., Sheen, K. L., Tilstone, G. H., & Kitidis, V. (in prep). UEx-Eddies: a biogeochemical long lived mesoscale eddy trajectories for studying air-sea CO2 fluxes. Friedlingstein, P., O’Sullivan, M., Jones, M. W., Andrew, R. M., Hauck, J., Landschützer, P., et al. (2025). Global Carbon Budget 2024. Earth System Science Data, 17(3), 965–1039. https://doi.org/10.5194/essd-17-965-2025 Good, S. A., & Embury, O. (2024). ESA Sea Surface Temperature Climate Change Initiative (SST_cci): Level 4 Analysis product, version 3.0 [Application/xml]. NERC EDS Centre for Environmental Data Analysis. https://doi.org/10.5285/4A9654136A7148E39B7FEB56F8BB02D2 Guo, Y., & Timmermans, M. (2024). The Role of Ocean Mesoscale Variability in Air‐Sea CO 2 Exchange: A Global Perspective. Geophysical Research Letters, 51(10), e2024GL108373. https://doi.org/10.1029/2024GL108373 Jean-Michel, L., Eric, G., Romain, B.-B., Gilles, G., Angélique, M., Marie, D., et al. (2021). The Copernicus Global 1/12° Oceanic and Sea Ice GLORYS12 Reanalysis. Frontiers in Earth Science, 9(July), 1–27. https://doi.org/10.3389/feart.2021.698876 Jones, E. M., Hoppema, M., Strass, V., Hauck, J., Salt, L., Ossebaar, S., et al. (2017). Mesoscale features create hotspots of carbon uptake in the Antarctic Circumpolar Current. Deep-Sea Research Part II: Topical Studies in Oceanography, 138, 39–51. https://doi.org/10.1016/j.dsr2.2015.10.006 Mears, C., Lee, T., Ricciardulli, L., Wang, X., & Wentz, F. (2022a). Improving the Accuracy of the Cross-Calibrated Multi-Platform (CCMP) Ocean Vector Winds. Remote Sensing, 14(17), 4230. https://doi.org/10.3390/rs14174230 Mears, C., Lee, T., Ricciardulli, L., Wang, X., & Wentz, F. (2022b). RSS Cross-Calibrated Multi-Platform (CCMP) monthly ocean vector wind analysis on 0.25 deg grid, Version 3.0 [Data set]. https://doi.org/10.56236/RSS-uv1m30 Orselli, I. B. M., Kerr, R., Azevedo, J. L. L. de, Galdino, F., Araujo, M., & Garcia, C. A. E. (2019). The sea-air CO2 net fluxes in the South Atlantic Ocean and the role played by Agulhas eddies. Progress in Oceanography, 170(2018), 40–52. https://doi.org/10.1016/j.pocean.2018.10.006 Pegliasco, C., Delepoulle, A., Mason, E., Morrow, R., Faugère, Y., & Dibarboure, G. (2022). META3.1exp: a new global mesoscale eddy trajectory atlas derived from altimetry. Earth System Science Data, 14(3), 1087–1107. https://doi.org/10.5194/essd-14-1087-2022 Pezzi, L. P., de Souza, R. B., Santini, M. F., Miller, A. J., Carvalho, J. T., Parise, C. K., et al. (2021). Oceanic eddy-induced modifications to air–sea heat and CO2 fluxes in the Brazil-Malvinas Confluence. Scientific Reports, 11(1), 10648. https://doi.org/10.1038/s41598-021-89985-9 Sathyendranath, S., Brewin, R. J. W., Brockmann, C., Brotas, V., Calton, B., Chuprin, A., 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(19). https://doi.org/10.3390/s19194285 Sathyendranath, S., Jackson, T., Brockmann, C., Brotas, V., Calton, B., Chuprin, A., et al. (2023). ESA Ocean Colour Climate Change Initiative (Ocean_Colour_cci): Version 6.0, 4km resolution data [Data set]. NERC EDS Centre for Environmental Data Analysis. https://doi.org/10.5285/5011D22AAE5A4671B0CBC7D05C56C4F0 Song, H., Marshall, J., Munro, D. R., Dutkiewicz, S., Sweeney, C., McGillicuddy, D. J., & Hausmann, U. (2016). Mesoscale modulation of air-sea CO2 flux in Drake Passage. Journal of Geophysical Research: Oceans, 121(9), 6635–6649. https://doi.org/10.1002/2016JC011714