INTRODUCTION INTRODUCTION
The European Project for Ice Coring in Antarctica (EPICA) provid
The European Project for Ice Coring in Antarctica (EPICA) provided deep ice cores from two drilling sites in Antarctica. The Droed deep ice cores from two drilling sites in Antarctica. The Dronning Maud Land (DML) core at [00nning Maud Land (DML) core at [00°°0404’’E, 75E, 75°°0000’’S ], representing climate variability of the Southern Atlantic OS ], representing climate variability of the Southern Atlantic Ocean, and the Dome Concordia (DC) core at cean, and the Dome Concordia (DC) core at [123
[123°°21'E, 7521'E, 75°°06'S ], mainly reflecting changes of the Indic part of the South06'S ], mainly reflecting changes of the Indic part of the Southern Ocean (figure 1). Both cores reach back to at least the coldern Ocean (figure 1). Both cores reach back to at least the coldstage MIS 6. Ion chemistry measurements allow conclusions on source strength and transport efficiency of sea salt, bio productistage MIS 6. Ion chemistry measurements allow conclusions on source strength and transport efficiency of sea salt, bio production compounds, on compounds, dust, volcanoes etc. In the framework of the EPICA project, the
dust, volcanoes etc. In the framework of the EPICA project, the aerosol chemistry is analyzed by ion chromatography (IC) in relaaerosol chemistry is analyzed by ion chromatography (IC) in relatively high resolution at both cores. This poster will give deeptively high resolution at both cores. This poster will give deeper insight in the components originating from biological producter insight in the components originating from biological productivity methanesulfonate (MS) and ivity methanesulfonate (MS) and non sea salt sulfate (nss
non sea salt sulfate (nss--SOSO44). As so far only for the DC core an independent depth age scale). As so far only for the DC core an independent depth age scalehas been established, the cores were synchronized by matching chas been established, the cores were synchronized by matching conspicuous peaks and dips in the dust records. The dust record wonspicuous peaks and dips in the dust records. The dust record was chosen because of the identical source region as chosen because of the identical source region (Patagonia) for both sites. The shown records are resampled to a
(Patagonia) for both sites. The shown records are resampled to aresolution of 100 years, as this can be sustained for large parresolution of 100 years, as this can be sustained for large parts of the cores. First comparisons show synchronous temporal evots of the cores. First comparisons show synchronous temporal evolutions of chemical components in both cores. lutions of chemical components in both cores.
One of the few significant discernible differences can be seen i
One of the few significant discernible differences can be seen in the MS record of both cores concerning concentration levels, vn the MS record of both cores concerning concentration levels, variability and phasing. As MS measured in ice cores is subject tariability and phasing. As MS measured in ice cores is subject to diverse influences, the interpretation of this observed deviato diverse influences, the interpretation of this observed deviations is challenging. For MS, those ions is challenging. For MS, those deviations could be caused by regional differences in MS product
deviations could be caused by regional differences in MS production, or differences in accumulation and dust deposition, implyinion, or differences in accumulation and dust deposition, implying changing MS fluxes or postdepositional losses. g changing MS fluxes or postdepositional losses.
The surprisingly constant flux of nss
The surprisingly constant flux of nss--SOSO44at DC is in contradiction to the often mentioned iron fertilizaat DC is in contradiction to the often mentioned iron fertilization hypothesis of the Southern Oceans (SO) biosphere. An unchantion hypothesis of the Southern Oceans (SO) biosphere. An unchanged productivity of the SO during the last Glacial would have aged productivity of the SO during the last Glacial would have alarge imprint on existing models of the carbon cycle. large imprint on existing models of the carbon cycle.
However, the record of the nss
However, the record of the nss--SOSO44flux from DML shows an increased during the LGM. In this posterflux from DML shows an increased during the LGM. In this posterthe influence of transport and source strength of nssthe influence of transport and source strength of nss--SOSO44measured at DML is estimated.measured at DML is estimated.
Variability in marine biogenic species in the EPICA ice cores during the last 150’000 years: Effects of aerosol deposition or bio productivity?
F. Fundel(1),
F. Fundel(1), H. Fischer(1)H. Fischer(1), , B. Twarloh (1), U. Ruth (1), E.W. Wolff (2), G. B. Twarloh (1), U. Ruth (1), E.W. Wolff (2), G. LittotLittot(2), R. Mulvaney (2), M. de Angelis (3), M. Hansson (4) U. (2), R. Mulvaney (2), M. de Angelis (3), M. Hansson (4) U. JonsellJonsell(4), M. Hutterli (5), P. Kaufmann (5), U. (4), M. Hutterli (5), P. Kaufmann (5), U. FedererFederer(5), F. Lambert (5), F. Lambert (5), J.P. Steffensen (6), R. Udisti (7), S.
(5), J.P. Steffensen (6), R. Udisti (7), S. BecagliBecagli(7), E. Castellano (7), M. (7), E. Castellano (7), M. SeveriSeveri(7), C. (7), C. BarbanteBarbante(8) and (8) and VaniaVaniaGaspariGaspari(8)(8) Alfred Wegener Institute for Polar and Marine Research, Bremerha
Alfred Wegener Institute for Polar and Marine Research, Bremerhaven, Germany, (2) British Antarctic Survey, Cambridge, UK, (3) ven, Germany, (2) British Antarctic Survey, Cambridge, UK, (3) LaboratoireLaboratoirede de GlaciologieGlaciologieet et GeophysiqueGeophysiquede de ll’’EnvironnementEnvironnement, , GrenobleGrenoble, , France, (4) Stockholm University, Sweden, (5) University of Bern
France, (4) Stockholm University, Sweden, (5) University of Bern, Switzerland, (6) University of Copenhagen, Denmark, (7) Univer, Switzerland, (6) University of Copenhagen, Denmark, (7) University of Florence, Italy, (8) University of Venice, Italy sity of Florence, Italy, (8) University of Venice, Italy ffundel@awi
ffundel@awi--bremerhaven.de Phone: +49(0)471bremerhaven.de Phone: +49(0)471--48314831--13451345
MS MS
The MS concentration records from DML and DC (figure 2) show onl
The MS concentration records from DML and DC (figure 2) show only few similarities. They differ in y few similarities. They differ in concentration level, long and short term variability and phasing
concentration level, long and short term variability and phasing. The poor correlation of both MS records . The poor correlation of both MS records suggests other processes than changes in source strength to be r
suggests other processes than changes in source strength to be responsible for the observed signal. Most esponsible for the observed signal. Most likely loss processes are worth to be considered.
likely loss processes are worth to be considered.
At DC, the utilization of loss processes as main influence for t
At DC, the utilization of loss processes as main influence for the observed variation in MS is a plain sailing. he observed variation in MS is a plain sailing.
This can be clearly seen when comparing the DC MS record to the
This can be clearly seen when comparing the DC MS record to the Cl/Na ratio. RCl/Na ratio. Rööthlisberger et al., 2005 have thlisberger et al., 2005 have shown the dependency of the Cl/Na ratio on postdepositional loss
shown the dependency of the Cl/Na ratio on postdepositional lossfor the past 45kyrs. This loss is especially for the past 45kyrs. This loss is especially strong in periods of low dust levels, where the risen acidity of
strong in periods of low dust levels, where the risen acidity ofthe ice supports volatilization of HCl. In these the ice supports volatilization of HCl. In these periods, the Cl/Na ratio is far below the standard sea water rat
periods, the Cl/Na ratio is far below the standard sea water ratio (SWR) of 1.79. However, during the LGM, the io (SWR) of 1.79. However, during the LGM, the Holocene Optimum (9
Holocene Optimum (9--11kyrs BP) and MIS 5.5, the Cl/Na ratio close to or above the SW11kyrs BP) and MIS 5.5, the Cl/Na ratio close to or above the SWR indicates no loss of R indicates no loss of HCl. The good correlation of Cl/Na and MS (figure 4) suggests th
HCl. The good correlation of Cl/Na and MS (figure 4) suggests the same mechanism of loss responsible for MS e same mechanism of loss responsible for MS as well as for HCl at DC. During the warm stages, higher accumul
as well as for HCl at DC. During the warm stages, higher accumulation might prevent loss of HCl, whereas ation might prevent loss of HCl, whereas during LGM the high dust levels improve fixation of HCl. There,
during LGM the high dust levels improve fixation of HCl. There, the correlation of MS and Cl/Na breaks down the correlation of MS and Cl/Na breaks down as well, and some original biogenic signal might be left over.
as well, and some original biogenic signal might be left over.
At DML, under present conditions, loss of MS exists as well, as
At DML, under present conditions, loss of MS exists as well, as shown by Weller et al., 2004. Since the Cl/Na shown by Weller et al., 2004. Since the Cl/Na ratio is above the SWR for the entire record, loss of MS might h
ratio is above the SWR for the entire record, loss of MS might have been not significantly increased during low ave been not significantly increased during low accumulation periods as well. This is also supported by the loss
accumulation periods as well. This is also supported by the lossestimation given by Weller et al., 2004, that estimation given by Weller et al., 2004, that predicts loss of more than 100% and thus clearly can not be vali
predicts loss of more than 100% and thus clearly can not be valid in glacial times. Further on, besides long d in glacial times. Further on, besides long term trends, there is no correlation of MS and the accumulation
term trends, there is no correlation of MS and the accumulation rate existing in glacial times (figure 3). The dust rate existing in glacial times (figure 3). The dust fixation mechanism, that works well at DC, is not possible to ex
fixation mechanism, that works well at DC, is not possible to explain all of the MS variability in glacial times as plain all of the MS variability in glacial times as well. At the Antarctic Warm Events A2
well. At the Antarctic Warm Events A2--A5 low levels of nssA5 low levels of nss--Ca can account for the observed MS levels yet dust Ca can account for the observed MS levels yet dust fixation fails in explaining the high MS levels during the other
fixation fails in explaining the high MS levels during the otherA events. Here, the Cl/Na record, used as A events. Here, the Cl/Na record, used as transport efficiency indicator, might give a clue. According to
transport efficiency indicator, might give a clue. According to Weller et al., 2004, no loss of HCl is observable at Weller et al., 2004, no loss of HCl is observable at DML at recent conditions. This probably holds for the Glacial as
DML at recent conditions. This probably holds for the Glacial aswell, indicated by an Cl/Na above the SWR. well, indicated by an Cl/Na above the SWR.
The formation process of HCl described by Legrand and Delmas, 19
The formation process of HCl described by Legrand and Delmas, 1985 thus allows to interpret a period of 85 thus allows to interpret a period of Cl/Na close to the SWR as time of efficient transport and vice v
Cl/Na close to the SWR as time of efficient transport and vice versa (figure 4). The observed Cl/Na during A ersa (figure 4). The observed Cl/Na during A events suggest an inefficient transport and might thus explain t
events suggest an inefficient transport and might thus explain the relatively low MS levels at A2he relatively low MS levels at A2--A5. A5.
SUMMARY SUMMARY
The MS records from DML and form DC are not The MS records from DML and form DC are not interpretable in terms of biogenic productivity of interpretable in terms of biogenic productivity of the SO. DC MS is clearly influenced by loss the SO. DC MS is clearly influenced by loss processes. The loss is especially pronounced in processes. The loss is especially pronounced in times of low dust and low accumulation rates.
times of low dust and low accumulation rates.
DML MS shows an influence on dust DML MS shows an influence on dust concentrations as well, but not all variability can concentrations as well, but not all variability can be explained by this. Additionally an effect of be explained by this. Additionally an effect of transport efficiency on MS is proposed here.
transport efficiency on MS is proposed here.
This is supported by the anti
This is supported by the anti--correlation of MS correlation of MS and the Cl/Na ratio which can be used as and the Cl/Na ratio which can be used as indicator for transport efficiency in DML.
indicator for transport efficiency in DML.
The constant nss
The constant nss--SOSO44flux observed at DC is flux observed at DC is probably the result of an unchanged bio probably the result of an unchanged bio productivity during all ages in the Indic sector of productivity during all ages in the Indic sector of the SO. At DML, where the Cl/Na ratio can be the SO. At DML, where the Cl/Na ratio can be used as transport efficiency indicator, the effect used as transport efficiency indicator, the effect of transport on the enhanced level of nss of transport on the enhanced level of nss--SOSO44
during the LGM is roughly estimated to account during the LGM is roughly estimated to account for a 10% increase. The remaining 40%
for a 10% increase. The remaining 40%
increase from Holocene to LGM might be the increase from Holocene to LGM might be the effect of enhanced bio
effect of enhanced bio--productivity in the Atlantic productivity in the Atlantic sector of the SO. The proximity to the source of sector of the SO. The proximity to the source of fertilizing dust, Patagonia, might be the reason fertilizing dust, Patagonia, might be the reason why the biological source in the
why the biological source in theAtlantic shows Atlantic shows an increase whereas the Indic biosphere does an increase whereas the Indic biosphere does not.
not.
Figure 2:
Figure 2:Oxygen Isotopic ratio, MS, nss-Oxygen Isotopic ratio, MS, nss-SOSO4,4,and nss-and nss-Ca concentrations as well as the Cl/Na Ca concentrations as well as the Cl/Na ratio from DML and DC in centennial resolution. The horizontal l
ratio from DML and DC in centennial resolution. The horizontal line shows the standard sea ine shows the standard sea water ratio of Cl/Na. The dotted vertical lines roughly mark the
water ratio of Cl/Na. The dotted vertical lines roughly mark theAntarctic warm events, labeled Antarctic warm events, labeled by A1
by A1--A7. Interglacials and transitions are shaded grey, thick lines sA7. Interglacials and transitions are shaded grey, thick lines show the 2000 years low how the 2000 years low pass filtered records. Volcano events in nss
pass filtered records. Volcano events in nss--SOSO44are removed.are removed.
Figure 5:
Figure 5:Accumulation rates, MS and non sea salt SOAccumulation rates, MS and non sea salt SO44fluxes from DML and DC in centennial fluxes from DML and DC in centennial resolution. The DML accumulation rate was estimated by thermo dy
resolution. The DML accumulation rate was estimated by thermo dynamical model based on namical model based on δ
δ1818O. Volcano events in nssO. Volcano events in nss--SOSO44were removed.were removed.
nss-nss-SOSO44 Nss
Nss--SOSO44is a conservative chemical species in Antarctic ice cores. Effeis a conservative chemical species in Antarctic ice cores. Effects of loss or diffusion are not known or cts of loss or diffusion are not known or of low order only. Therefore the fluxes of nss
of low order only. Therefore the fluxes of nss--SOSO44, that are representative for concentrations in the air at low , that are representative for concentrations in the air at low accumulation sites, can be interpreted in terms of changes in th
accumulation sites, can be interpreted in terms of changes in the source. Cosme et al., 2005 found at least e source. Cosme et al., 2005 found at least 90% of nss
90% of nss--SOSO44to be derived from DMS and thus from the Southern Oceans biosphere. This makes nssto be derived from DMS and thus from the Southern Oceans biosphere. This makes nss--SOSO44
a reliable indicator for changes in the SO bio a reliable indicator for changes in the SO bio--productivityproductivity The DC record of the nss
The DC record of the nss--SOSO44flux shows a surprisingly constant level throughout the last glflux shows a surprisingly constant level throughout the last glacial cycle (figure acial cycle (figure 5). In DML, during the glacial maxima and between A4 and A5, th
5). In DML, during the glacial maxima and between A4 and A5, the nsse nss--SOSO44flux is increased by flux is increased by approximately 50% compared to the rest of the record.
approximately 50% compared to the rest of the record.
The constant flux of nss
The constant flux of nss--SOSO44measured at DC suggests an unchanged productivity of the Indic measured at DC suggests an unchanged productivity of the Indic Ocean Ocean biosphere. A balancing effect of meridional transport efficiency
biosphere. A balancing effect of meridional transport efficiencyand source strength appears feasible as well and source strength appears feasible as well but not uniformly supported by atmospheric transport models.
but not uniformly supported by atmospheric transport models.
At DML an effect of additional nutrient supply, due to the proxi
At DML an effect of additional nutrient supply, due to the proximity to the Patagonian source of aeolian dust mity to the Patagonian source of aeolian dust might be responsible for the higher LGM flux, as well as an addi
might be responsible for the higher LGM flux, as well as an additional effect of transport efficiency, as tional effect of transport efficiency, as supported by the Cl/Na ratio. An estimate of the strength of sou
supported by the Cl/Na ratio. An estimate of the strength of source and transport effects at DML is performed rce and transport effects at DML is performed with an simple one dimensional transport model described in the
with an simple one dimensional transport model described in the right box. To quantify the transport effect, the right box. To quantify the transport effect, the Cl/Na ratios found by Legrand and Delmas, 1985 on a traverse fro
Cl/Na ratios found by Legrand and Delmas, 1985 on a traverse from Dumont Dm Dumont D’’Urville to DC were used (figure Urville to DC were used (figure 6). The mean Holocene Cl/Na ratio compared to the mean LGM ratio
6). The mean Holocene Cl/Na ratio compared to the mean LGM ratioresults in a reduction of 14% of transport results in a reduction of 14% of transport time from Holocene to LGM. Assuming a constant source strength o
time from Holocene to LGM. Assuming a constant source strength of nssf nss--SOSO44in the Holocene and in the LGM in the Holocene and in the LGM (mean fluxes were used as input parameter here), 10% of the incr
(mean fluxes were used as input parameter here), 10% of the increase of nssease of nss--SOSO44air concentration can be air concentration can be explained by changes in transport. As this is still below the me
explained by changes in transport. As this is still below the measured nssasured nss--SOSO44flux at DML, an additional flux at DML, an additional increase of source strength (40% of the Holocene level) is sugge
increase of source strength (40% of the Holocene level) is suggested here.sted here.
Figure 4:
Figure 4: Scatter Scatter plot of detrended plot of detrended MS and Cl/Na MS and Cl/Na from EDC and from EDC and EDML. The colors EDML. The colors show the actual show the actual accumulation rate.
accumulation rate.
The vertical line The vertical line
shows the
shows the
standard sea standard sea water ratio of water ratio of Cl/Na.
Cl/Na.
References:
Cosme, E., et al., Origin of dimethylsulfide, non-sea-salt sulfate, and methanesulfonic acid in eastern Antarctica, J.
Geophys. Res., 110, 2005.
Legrand, M., and R. J. Delmas, Formation of HCl in the Antarctic atmosphere, J. Geophys. Res.,1988.
Röthlisberger, R. et al., Limited dechlorination of sea-salt aerosols during the last glacial period: Evidence from the European Project for Ice Coring in Antarctica (EPICA) Dome C ice core, J. Geophys. Res., 2003.
Weller, R., et al., Postdepositional losses of methane sulfonate, nitrate, and chloride at the European Project for Ice Coring in Antarctica deep-drilling site in Dronning Maud Land, Antarctica, J. Geophys. Res., 2004.
Figure 1:
Figure 1:The drilling sites DML and DC in Antarctica The drilling sites DML and DC in Antarctica with the source region of deposited aerosol marked with the source region of deposited aerosol marked roughly in blue (DML) and red (DC)
roughly in blue (DML) and red (DC)
90oW
0o
90oE
180oW 70oS
60oS
50oS EDML
EDC
-100 0 100 200 300 400 500 600
-6 -4 -2 0 2 4 6 8 10 12
y=0.02(+-0.01) *x - 2.2(+-2.7)
Distance from coast [km]
Cl [ppb] / Na [ppb]
Figure 6:
Figure 6:Cl/Na ratio with distance from coast on a traverse to DC. The uCl/Na ratio with distance from coast on a traverse to DC. The used data are sed data are taken from Legrand and Delmas, 1985. The lines show the linear t
taken from Legrand and Delmas, 1985. The lines show the linear trend with confidence rend with confidence intervals (95%) for the expected Cl/Na ratio. The red square sho
intervals (95%) for the expected Cl/Na ratio. The red square shows the composite of 12 ws the composite of 12 snow pits from DML, the green square shows the mean Holocene rat
snow pits from DML, the green square shows the mean Holocene ratio derived from the io derived from the EDML core. Both are not included in the regression.
EDML core. Both are not included in the regression.
1 dimensional Transport Model 1 dimensional Transport Model
( ) 0exp
air t air
C C t τ
= − Cair0= atmospheric concentration at source
( ) atmospheric concentration at time t
air t
C =
atmospheric residence time (7days) τ =
Input parameters (units arbitrary):
Input parameters (units arbitrary):
( ) 400
air t Holocene
C =
0 0 820
air Holocene air LGM
C =C =
Holocene 5 t =days
LGM Holocene14%
t =t ⋅ Cair t LGM( ) =440
Output:
Output:
0 5 10 15 20 25 30 35
1 2 3 4 5 6 7
detrended MS [ppb]
detrended accumulation [w.e./yr]
E DML E DC
Figure 3:
Figure 3: Scatter plot Scatter plot of linearly detrended of linearly detrended accumulation rates accumulation rates
and MS
and MS
concentrations form concentrations form EDC and EDML. The EDC and EDML. The shown data are taken shown data are taken from the last Glacial from the last Glacial only (17 only (17--115kyrs 115kyrs BP).
BP).
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
0 5 10 15 20 25 30
detrended C l-[pp b]/N a+[p pb]
Age BP [kyrs]
detrended MS [ppb]
2 4 6 8 10 12 ED ML 14 ED C
Loss + Transport -
MIS 1 T1 EDML MIS 5e T2
EDC
3.0 2.0 -+Cl/Na1.0
15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Age BP [kyrs]
80 60 40 20 0
nss-Ca++ [ppb] 200
150 100 nss-SO450 --[ppb]
25 20 15 10 5 0
MS [ppb]
-56 -52 -48 -44 -40
δδδδ18181818Ο [Ο [Ο [Ο [οοοο////οοοοοοοο]]]]
A1 A2A3A4 A5 A6 A7
1000 800 600 400 200 0 J nss-SO4-- [ng cm-2 yr-1]
150 140 130 120 110 100 90 80 70 60 50 40 30 20 10 0
Age BP [kyrs]
100
80 60
40
20 0 J MS [ng cm-2 yr-1] 10
8 6 4 2
Acc. [cm w.e./yr]
MIS 1 T1 MIS 5e T2
EDMLEDC
A1 A2 A3A4 A5 A6 A7
