A multiproxy approach to timeseries fluxes and coretop sediments: A possibility to deconvolve temper

1
A multi-proxy approach to time-series fluxes and
coretop sediments A possibility to deconvolve
temperature and salinity in open ocean
environments?

• NIOZ group Ulrike Fallet, Isla Castaneda, Sanne
  Vogels, Geert-Jan
• Brummer, Stefan Schouten

2
I. Research objectives

• 1. Establish temperature relationships in
  planktonic foraminifera with Mg/Ca and d18O
• 2. Verify these with satellite remote sensing SST
  and in situ T-S from moored CTDs
• Is it possible to eliminate salinity in this
  equation?
• Salinity gradient in the Mozambique Channel is
  0.4
• Calculation of salinity depends on accuracy of
• satellite remote sensing
• Mg/Ca and d18O measurements
• Intraspecies variation
• 3. Determine foraminifera fluxes to the channel
  floor
• 4. Apply results to coretop foraminifera to
  calibrate proxies

3
II. Research objectives

• 5. Compare Mg/ Ca and d18O from planktonic
  foraminifera with
• Coral proxies (Sr/Ca) off the coast of Madagascar
• Organic matter proxies (Tex86, Uk37)
• Downward particulate matter fluxes
• Eddy transport through the channel (nutrient
  distribution by frontal upwelling, nutrient
  transport from coastal margins to the open
  channel, etc.)

4
Study area The Mozambique Channel

• Anti-cyclonic eddies passing through the channel
  at 70 days frequency
• Eddies induce frontal upwelling and might bring
  nutrients to the surface
• Enhanced foraminifera production in cooler
  frontal zones?
• Implications for recorded temperature signal?

5
(No Transcript)
6
Satellite remote sensing SST quality and
averaging problems
7
Foraminifera fluxes to the channel floor

• SST red
• Ruber black
• Trilobus blue
• Ruber has highest production rates in January and
  February (SST 28 30C)
• Trilobus in June (SST 26 27C)
• Application for coretop sediments!

8
First results on Mg/Ca from trapped planktonic
foraminifera
Satellite remote sensing has an error of about
0.5 C

• Foraminifera follow SST but show a continuous
  offset of about /- 1.0C

TEX86 varies little having a slight
anti-correlation with SST (Does eddy velocity
keep OM in suspension? d13Corg correlates well
with eddy passage.)
9
Mg/Ca from trapped planktonic foraminifera

• Black ruber
• Blue trilobus
• Ruber and trilobus as surface dwellers follow SST
  but show a distinct offset
• Temperatures were calculated with calibrations
  from Elderfield (ruber) and Nuernberg (trilobus)

10
Paired Mg/Ca and d18O in coretop sediment

• Equations used Kim and ONeill for d18O and
  Nuernberg for Mg/Ca
• Ruber has been corrected for vital effect with
  results from Somalia
• Only slight range in Mg/Ca (lt0.5mmol/mol 2C)
  corresponds to relatively large range in d18O (1
  5 C )
• Black line linear fit
• Thin black line SST calculated into Mg/Ca and
  d18O with Nurnberg and Kim and ONeill equations

11
Cleaning artefacts or intraspecies variation?

• Accuracy of NIOZ ICP-MS matches the Cambridge
  ICP-OES (less than 0.1 mmol/mol offset)
• Cambridge method too rigorous for fragile trap
  samples?
• Preferential dissolution?

12
Conclusions

• SST data
• Satellite remote sensing has an error of 0.5 C
• Varies locally by up to 4C
• Eddy passage produces frontal upwelling
  (nutrients)
• Collecting period of 3 weeks is too long for this
  fast changing system
• Mg/Ca cleaning procedure
• Pre-cleaning step that removes OM from fluxes
  sufficient for fragile sediment trap samples???
• Comparison ICP-MS (NIOZ) ICP-OES (Cambridge)
• Matching is very high (within 0.5C)
• Precision is very high (lt 0.5 C)

13
Conclusions

• Salinity in the open ocean (Mozambique Channel)
  varies only slightly (34.8 35.1)
• dutertrei from coretops (no seasonality because
  sub-surface dweller with vital effect correction)
  only leaves a very small offset which could be
  attributed to salinity)
• Needs to be verified with time-series fluxes

Mg/Ca
d18O with vital effect
d18O
Temperature C after Nuernberg
14
Outlook

• Finish Mg/Ca measurements on time-series fluxes
• Analyse time-series fluxes for d18O
• Paired Mg/Ca - d18O can be used to check for
  salinity effect
• Obtain satellite remote sensing temperatures that
  have been quality checked and calculate
  tri-weekly averages for trap location
• Use SST and altimetry data to estimate frontal
  upwelling in eddies (temperature difference?)