F. Schütte et al.: Hidden vortices: near-equatorial low-oxygen extremes driven by high-baroclinic-mode vortices 139
ii. Low-DO events are related to subsurface intensified
submesoscale coherent vortices: we found 66% of
open ocean low-DO events to be related to subsurface-
intensified submesoscale coherent vortices, whith an-
ticyclonic eddies appearing to dominate. The vertical
structure of these vortices is characterized by high baro-
clinic modes (modes 4 to 10), and they are confined to
the upper 250 m. In situ velocity observations revealed
an average radius of 34km, which is well below the first
baroclinic Rossby radius of deformation (O(100 km)),
5ut agrees well with Rossby radii of the higher baro-
clinic modes 4 to 10 (34 to 13km at 9° N). Despite the
small length scales, the Rossby number of the vortices
is below 1, assigning them to the dynamical range of
mesoscale variability.
ii. Origin and life time: the vortices most likely origi-
nate from the eastern boundary. They can propagate far
into the open ocean with a propagation speed of 1.8-
4.9kmd”!, reaching a life time of more than half a year
Äät took around 100 to 500d to propagate the 550 km
distance towards 23° W). This is much longer than cur-
rently considered possible, given the highly dynamical
area and the proximity to the equator. Model simula-
ons even show a life time of up to 1.5 years. Cyclonic
eddies with low-oxygen cores were less frequent than
anticyclonic eddies. Cyclonic eddies were not found in
ship sections along 23° W, but in the minority of all
low-DO extreme events from moored observations at
L1°N/21° W.
iv. Impact of the vorticies on DO and biogeochemistry:
near-equatorial anticyclonic vortices have unexpectedly
long lifetimes and strongly isolate their low-PV cores
from surrounding water. This can create a DO deficient
zone, due to enhanced primary production on top and
remineralization (DO decrease of 0.16 umol kg”! a7!
for the simulated anticyclonic vortices), accompanied
oy elevated nitrate levels in the eddy core.
Detection of near-equatorial vortices with remote sens-
ing satellites: near-equatorial vortices are hardly de-
‚ectable by conventional satellite altiımetry observa-
tions, which precludes a backtracking of these eddies.
New observations are desirable to verify whether the
new SWOT mission can capture such HBV, although
a strong surface signal is not expected due to the mainly
zubsurface structure (also supported by the model
Subsurface coherent vortices in the near-equatorial ocean
have been so far overlooked in driving DO deficient zones.
The long-lived vortices appear unexpectedly quite regularly
given theoretical considerations and are able to generate hy-
poxic regimes in the open ocean, which may have local-
ized effects habitats, biodiversity and biogeochemical cy-
cling. They are typically not tracable in satellite products,
https:/doi.org/10.5194/os-22-119-2026
which makes a collocation of satellite data with in-situ obser-
vations (CTD-O, Argo profiles, moored observations) hardly
possible. The comparatively coarse resolution of satellite
observations might instead lead to a wrong collocation of
the subsurface low-DO events with larger surface intensified
mesoscale structures nearby. The mechanisms for the gen-
eration of these near-equatorial low-DO eddies remain an
open question. So far, we here identified a potential source
region and provided a first insight about the dynamics (life
time, baroclinicity, isolation) of these eddies. A more com-
prehensive investigation from high resolution ocean circula-
tion models — coupled to biogeochemistry — would shed light
onto the generation. Further, the study of the temporal evo-
lution of dominant vertical baroclinic modes throughout the
eddies’ life cycle would contribute to a better understand-
ing of the eddy dynamics and stability. Moreover, the inter-
disciplinary view on changes in biogeochemical processes
would increase the understanding about the impact on bio-
geochemistry. The in-situ tracking and observation of these
eddies over their life cycle is challenging, but would provide
key information to validate the simulation of these eddies.
Data availability. The assembled shipboard measurements (27 re-
search cruises) and moored data used in this paper are available and
collected at https://doi.org/10.1594/PANGAEA.987397 (Schütte et
al., 2025). The used satellite altimetry data is provided by Marine
Copernicus (https://marine.copernicus.eu, last access: 15 Septem-
ber 2025) can be downloaded at https://doi.0org/10.48670/moi-
00148 (E.U. Copernicus Marine Service Information, 2024).
The used gridded climatological hydrography and oxygen from
ıhe World Ocean Atlas 2023 (WOA223), is available at NOAA
under: https://doi.org/10.25921/va26-hv25. The model data and
material that support the findings of this study are avail-
able through GEOMAR at https://hdl.handle.net/20.500.12085/
21ab7868-c112-4173-ada4-75b86f098f18 (Frenger, 2025).
Supplement. The supplement related to this article is available on-
line at https://doi.org/10.5194/os-22-1 19-2026-supplement.
Author contributions. Conzeptualization: FS, JH, PB, Data cura-
tion: JH, IF, FS, Formal analysis and methodology: JH, IF, MS,
FS, AB, AFD, Funding acquisition: PB, FS, JH, Writing — origi-
nal draft: JH, FS, IF, Writing — review and editing: FS, JH, IF, AB.
AFD. MS. PB.
Competing interests. The contact author has declared that none of
the authors has any competing interests.
Disclaimer. Publisher’s note: Copernicus Publications remains
neutral with regard to Jurisdictional claims made in the text, pub-
lished maps, institutional affiliations, or any other geographical rep-
resentation in this paper. While Copernicus Publications makes ev-
Ocean Sci... 22. 119-143. 2026
