Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an...
Transcript of Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an...
![Page 1: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/1.jpg)
Stirring of ecological landscapes: from 2D to 3D dynamicsF. d'Ovidio
LOCEAN-IPSL, Paris
Work in collaboration with: A. Soccodato, A. Della Penna, C. Cotté, S. De Monte, S. Alvain, C. Guinet, S. Blain
![Page 2: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/2.jpg)
Objectives of this presentation
- Examples of multisatellite data analysis for eco-biogeochemical applications
- Emergence of 3D dynamics: noise or signal?
Outline
- Estimation of biogeochemical budgets by merging multisatellite data
- Hotspots of trophic interactions
- hotspots of biodiversity
![Page 3: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/3.jpg)
Time
MesoscaleEddies
SubmesoscaleFilaments
Large-scaleBoundary currentsUpwelling systems
months Yearsdays/weeks
~ 1-10 km
~ 100 km
~1000 km
A scale-dependent classification of the ocean circulation
Ex.: patterns of oilspills
Phytoplankton ecological timescales (doubling times, bloom,..) are days/weeksResonance between physics and biology at the submesoscale
Ex.: Gulf stream
Cyclons/anticyclons
Key spatial scale: limit of current global circulation models and global observational dataLink between the surface and the ocean interior
Horizontal stirring
![Page 4: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/4.jpg)
Ecological patchiness
100 km
1000 kmLarge scale variability
North Atlantic(1998-2003, mean)
Meso/submeso scale patterns
Northeast Atlantic(daily images)
SeaWiFS chlorophyll
mg/m30.1 0.3 1
(Sub-)mesoscale variability (~10 km, days/weeks) has the same tracer contrasts as the large scale (~1000 km, months/years) variability!
Results of both horizontal and vertical dynamics
Dataset 1: SeaWiFS
![Page 5: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/5.jpg)
Part I: Stirring and biogeochemistry
![Page 6: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/6.jpg)
Abraham et al., Nature, 2000
Iron was released in nutrient-rich, low-Chlorophyll waters:
Main results
- Adding iron stimulated a planktonic bloom
- The bloom was strongly affected by stirring
Open question:
- Does the bloom result in carbon export?
→ follow the phytoplanktonic patch from bloom to algal death
→ Estimate stirring contribution
The SOIREE experiment (1999)
![Page 7: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/7.jpg)
→ follow the phytoplanktonic patch from bloom to algal death: 40-day experiment
→ Estimate stirring contribution: relied on altimetry-derived diagnostics
The EIFEX experiment (February 2004)
POC at 350m
Smetacek et al., Nature 2012
![Page 8: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/8.jpg)
→ follow the phytoplanktonic patch from bloom to algal death: 40-day experiment
→ Estimate stirring contribution: relied on altimetry-derived diagnostics
The EIFEX experiment (February 2004)
POC at 350m
Smetacek et al., Nature 2012
![Page 9: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/9.jpg)
→ follow the phytoplanktonic patch from bloom to algal death: 40-day experiment
→ Estimate stirring contribution: relied on altimetry-derived diagnostics
Open question:
What happens in natural fertilization?
The EIFEX experiment (February 2004)
POC at 350m
Smetacek et al., Nature 2012
![Page 10: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/10.jpg)
The KEOPS2 experiment (end 2011)
→ follow a naturally fertilized patch
→ Estimate stirring contribution and sources of iron
![Page 11: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/11.jpg)
Disentangling iron sources and circulation patterns with altimetry
The KEOPS2 experiment (end 2011)
![Page 12: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/12.jpg)
Disentangling iron sources and circulation patterns with altimetry
The KEOPS2 experiment (end 2011)
![Page 13: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/13.jpg)
Iron concentration from altimetry data
Age from the plateau (days) µ mol/m2Iron concentration
The age of water parcels can be converted in iron concentration by assuming a constant iron concentration over the plateau, an exponential scavenging dynamics and geostrophic dynamics for the mixed layer. KEOPS2 iron measurements are coherent with this approach.
![Page 14: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/14.jpg)
Some conclusions (Satellite-derived stirring and biogeochemistry)
- Stirring patterns can be reconstructed accurately from altimetry
- “Virtual sensors” (e.g., Kerguelen iron concentration) can be constructed by gluing multisatellite data with theoretical tools
- Can we really neglect the 3D dynamics?
![Page 15: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/15.jpg)
Ocean color
A
Ocean color, threshold
Altimetry model, global product
The KEOPS2 experiment (end 2011)
![Page 16: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/16.jpg)
Ocean color
A
Ocean color, threshold
Altimetry model, global product Altimetry model, regional product
The KEOPS2 experiment (end 2011)
![Page 17: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/17.jpg)
Stirring patterns and Chl plume
In Spring (11/nov/2011):
- Geostrophic (altimetry) velocities are very consistent with the hypothesis that blooming waters are water coming from the plateau.
- Adding Ekman velocities leads to largely overstimate the plume extension.
Geostrophic velocitiesGeostrophic+Ekman velocities
![Page 18: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/18.jpg)
I. Stirring patterns and Chl plume
In summer (11/dec/2011):
- Geostrophic (altimetry) velocities undersetimate the plume extension!
- Adding Ekman velocities agrees with Chl maps.
Geostrophic velocitiesGeostrophic+Ekman velocities
![Page 19: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/19.jpg)
Possible scenario:
1. There is an Ekman layer with a NE strong component.
2. Winter/spring time: the mixed layer is deeper than the Ekman layer: tracers diluted in the mixed layer have a geostrophic circulation, close-to-surface buoys at fixed depth have a geostrophic+Ekman trajectories.
3. late-spring/summer time: the mixed layer approaches the Ekman layer, and tracers also are structured by geostrophy+Ekman
Supporting observations:
1. Chl map consistent with geostrophy in early spring, with geostrophy+Ekman in late spring
2. Drifters show a strong NE drift.
3. SST images indicate stratification between november and december (next slides)
This may explain why altimetry agrees with Chl but not with drifter trajectories
Mixed layer
Ekman layer
Winter/spring time Late-spring/summer time
![Page 20: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/20.jpg)
Part 2: Stirring and marine ecology
![Page 21: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/21.jpg)
Phytoplankton biogeography
Alan Longhurst, Ecological Geography of the Sea, Academic Press (1998)
![Page 22: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/22.jpg)
Phytoplankton biogeography in models
![Page 23: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/23.jpg)
Remotely sensed biogeography of dominant phytoplankton types in the global ocean
█ nanoeukaryotes█ synechococcus█ prochlorococcus█ diatoms█ phaeocistis█ Coccolithophores
Algorithm PHYSATAlvain et al. (2004)
High spatiotemporal resolutionbut
Only provides the dominant phytoplanktonic type
![Page 24: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/24.jpg)
Remotely sensed biogeography of dominant phytoplankton types in the global ocean
█ nanoeukaryotes█ synechococcus█ prochlorococcus█ diatoms█ phaeocistis█ synechococcus
Algorhytme PHYSATAlvain et al. (2004)
ZOOM IN
Chl annual cycle
Spring bloom
26/11/2001 Chl
![Page 25: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/25.jpg)
A case-studystirring at the confluence of Malvinas' and Brazil currents
Satellite SSHMesoscale currents
Tracer initialized with a longitudinal gradient
Effect of stirring by advection ofsynthetic tracersVelocity field
Tracer initialized with a latitudinal gradient
Filaments induced by stirring appear when a passive tracer is advected.However, the spatial structures may depend on the tracer initial conditions
![Page 26: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/26.jpg)
A case-studystirring at the confluence of Malvinas' and Brazil currents
Satellite SSHMesoscale currents
SSH-derived Lyapunov exponents
Stirring diagnosticVelocity field
Stirring diagnostics like the Lyapunov exponents permits to detect stirring structures independently of specific initial conditions (in this case, fronts induced by stirring).
![Page 27: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/27.jpg)
A case-studystirring at the confluence of Malvinas' and Brazil currents
Satellite SSHMesoscale currents
SSH-derived Lyapunov exponents
Stirring diagnosticVelocity field
The Lyapunov exponent in particular identifies (sub-)mesoscale fronts which agree remarkably well (~10 km) with the boundaries of biogeochemical tracers like Chlorophyll
Satellite Chl
Stirred tracer
![Page 28: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/28.jpg)
What is the effect of stirring on phytoplanktonic communities?
PHYSAT types (SeaWiFS)
Diatoms Nanoeukariotes Phaeocystis Coccolithophorides Synechoccocus
Novel satellite products go beyond Chl concentration. PHYSAT estimates dominant planktonic types and unveils a patchy structure of the dominant community. What is the origin of this ecological patchiness?
![Page 29: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/29.jpg)
Superposition between PHYSAT types and Lyapunov exponents:
Frontiers among dominant types agree with Lyapunov lines→ stirring drives patchiness of dominant types
d'Ovidio et al., PNAS 2010
![Page 30: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/30.jpg)
Trying to reproduce the community structure by advecting subtropical and supolar environmental patches
![Page 31: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/31.jpg)
Trying to reproduce the community structure by advecting subtropical and supolar environmental patches
Observed from SeaWiFS Advected with altimetry(3 months)
![Page 32: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/32.jpg)
Trying to reproduce the community structure by advecting subtropical and supolar environmental patches (i.e., passive scalar)
Observed from SeaWiFS Advected with altimetry(3 months)
EXISTENCE OF FLUID DYNAMICAL NICHESTransport barriers enclose fluid dynamical niches with lifetimeslong enough for phytoplakton competition to occur
d'Ovidio et al. PNAS 2010
![Page 33: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/33.jpg)
Working hypothesis: biodiversity is generated by mesoscale patchiness plus submesoscale mixing
Area-based Shannon Index: τ=−Σ p i log pi
Patchiness of dominant types and diversity?
.
.
![Page 34: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/34.jpg)
Validating the area-based biodiversity proxy with the MIT Darwin model
Soccodato et al., in preparation
![Page 35: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/35.jpg)
Satellite map of diversity proxy: comparing with Tara data
and Tara stationsDe Monte et al. ISME J. 2013
![Page 36: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/36.jpg)
Tara
Satellite Model
Comparison of Tara-Oceans Shannon index and remote sensing proxy
47 surface stations, area-based Shannon index from Physat reanalysis (montly average, filtered)
![Page 37: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/37.jpg)
Tittensor et al. Nature 2010
Relation with temperature
Total oceanic biodiversity from OBIS
Worms et al. Science 2009Rutherford et al. Nature 1999
![Page 38: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/38.jpg)
Relation with SST and comparison with in situ data
Biodiversity
The satellite-based biodiversity proxy follows a compact relation with SST predicted by biodiversity theories and well in agreement with AMT biodiversity data.
![Page 39: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/39.jpg)
Comparison with predators biodiversity
biodiversity index from satellite
![Page 40: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/40.jpg)
Conclusions
Satellites typically see the 2D surface of the ocean.
The ocean surface however is part of a 3D system.
For a single satellite detector, 3D signal is present, but it is typically the result of many confounding effects. It is usually treated as a noise over the 2D information, or simply neglected.
Mutltisatellite data analysis (with in situ) may allow to go one step further and disentangle some of the 3D sources
Some possible products:
- Estimation of vertical velocities
- Mixed layer/Ekman layer depth
- proxies of ecological features- hotspots of biodiversity- hotspots of trophic interactions- biomass density for intermediate trophic levels?
![Page 41: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/41.jpg)
![Page 42: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/42.jpg)
![Page 43: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/43.jpg)
During autumn:winter, seals appear to track the drifting water mass that has bloomed
Cotté et al.submitted
![Page 44: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/44.jpg)
![Page 45: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/45.jpg)
Cotté et al.submitted
![Page 46: Stirring of ecological landscapes: from 2D to 3D dynamics · Possible scenario: 1. There is an Ekman layer with a NE strong component. 2. Winter/spring time: the mixed layer is deeper](https://reader033.fdocuments.us/reader033/viewer/2022052611/5f08ca447e708231d423bbda/html5/thumbnails/46.jpg)
Seals seem to follow and forage on transport (Lyapunov) fronts and SST fronts, especially during the post-bloom season