Figure 1 - Illimani ice core site in Eastern Bolivian Andes, South America. Distance from Illimani...
1
Figure 1 - Illimani ice core site in Eastern Bolivian Andes, South America. Distance from Illimani to La Paz is about 40 Km. •Two ice cores were drilled in June 1999 by a French-Swiss team at the summit of Nevado Illimani (1637’S, 6746’W, 6350 m asl, Figure 1). The cores were drilled to the bedrock, with total depths of 137m and 139m, corresponding to about 18,000 years (Knüsel et al., 2003; Ramirez et al., 2003), and transported frozen to European laboratories. This work shows the results obtained for the French core. •Several analyses were done: electrical conductivity, gamma spectrometry, D and 18 O measurements. The uppermost 50m of the French core was cut in 744 sub-samples in a cold room at -15°C. These 744 sub-samples, corresponding to about 80 years from 1919 to 1999, were analysed for insoluble aerosol content, major ions by ion chromatography and multielemental concentration by ICP-MS (Correia et al., 2003). Alexandre Correia 1,2 ([email protected]); Robert Delmas 2 ; Rémi Freydier 3 ; Jefferson C. Simões 4 ; Jean-Denis Taupin 5 ; Bernard Dupré 3 and Paulo Artaxo 1 1 Institute of Physics, University of São Paulo, R Matão Trav R 187, São Paulo, SP 05508-900, Brazil; 2 LGGE, CNRS and Université Joseph Fourier, Grenoble, France; 3 UMR5563, LMTG, CNRS and Université Paul Sabatier, Toulouse, France; 4 Institute of Geosciences, Federal University of Rio Grande do Sul, Brazil; 5 LGGE, IRD, Grenoble, France EGS-AGU-EUG Joint Assemblies 06-11 April 2003, Nice, France Ice cores are important tools for reconstructing past environmental conditions, from climatic variations to atmospheric chemistry. Several snow and ice cores from polar regions were studied in last decades, especially from Greenland and Antarctica. Pioneering work by Thompson and co-workers (e.g. Thompson, et al., 1985) showed the importance of high- altitude tropical ice cores, with different features and challenges compared to polar ice cores. In general, tropical ice cores sites are closer to moisture and aerosol sources than polar cores. The aerosol load of tropical ice is therefore much higher than in polar regions. Measured impurities may come from a variety of sources, either natural or anthropic in origin, presenting high spatial and temporal variability of atmospheric concentrations and climate-related physicochemical characteristics. Most Andean ice core studies aimed at paleoclimate reconstructions, for which the focus rests over a wide time scale covering several millennia, or over a few elemental concentration profiles. This work explores a high temporal resolution (i.e. sub-seasonal) ice core record extracted from the Andes, covering about 80 years during 20 th century. This ice core was studied under several analytic techniques, among which the determination of elemental concentration profiles of 45 chemical species, ranging from Li to U, by Inductively Coupled Plasma-Mass Spectrometry (ICP-MS). •About 80% of annual precipitation at Illimani occurs during austral summer (DJF), originating from the Amazon Basin to the East (Vuille, 1999). •Winter precipitation also comes from Amazon, as convective disturbances that reach the site and break against the prevailing westerly flow (Grootes et al., 1989). •Recent works show that El Niño can control precipitation over the Andes, even tough the humidity source is the Amazon Basin (Bradley et al., 2003; Hoffmann et al., 2003). •The ICP-MS analyses of the 744 sub-samples provided elemental concentration profiles of aerosol composition for 45 elements ranging from Li to U, from 1919 to 1999. The Lithium concentration profile is shown in Figure 2. •Lithium is associated with salt lakes in the Altiplano, the World’s largest reservoir for this element (Risacher, 1992). The strong seasonality in Figure 2 is a common feature of most elemental concentrations measured. It is attributed to yearly precipitation regimes at the site, with high concentrations during dry austral winter and low concentrations during wet austral summer, according to stable isotopes measurements. Figure 2 – Lithium concentration profile and seasonal classification scheme exemplified for a profile section. •Measurement of 45 chemical species by ICP-MS in 744 samples during the 20th century allowed studying concentration profiles at high temporal resolution. •During dry season high aerosol loadings are observed, with low crustal enrichment factors. Wet season conditions favour identification of other sources by removing local aerosol by wet deposition processes. •A chemical characterization for the background soil dust aerosol deposited during 20 th century was derived from enrichment factor plateau levels at high aerosol loading conditions. References Bowen, H. (1979) Environmental Chemistry of the Elements. Bradley, R.S. et al. (2003) Geophys. Res. Let. 30(4), doi: 10.1029/2002GL016546. Correia et al. (2003) Atm. Phys. Chem., [accepted]. Grootes, P.M. et al. (1989) J. Geophys. Res., 94, 1187-1194. Hoffmann, G. et al. (2003) Geophys. Res. Let., 30(4), doi: 10.1029/2002GL014870. Knüsel, S. et al. (2003) J. Geophys. Res., 108(D6), doi: 10.1029/2001JD002028. Ramirez, E. et al. (2003) Earth Plan. Sci. Let., [submitted]. Risacher, F. (1992) , Sci. Géol., Bull., 45, 3-4, 135-214. Thompson et al. (1985) Science, 229, 971- 973. Vuille, M (1999) Int. J. Climatol., 19, 1579-1600. •The samples were classified as belonging to dry, wet, or transition seasons, as exemplified in Figure 2. This classification scheme takes all measured elemental concentrations into account. •Figure 3 shows the average crustal enrichment factors for all measured elements, during dry and wet seasons. Reference is Aluminum in mean soil composition (Bowen, 1979). In Figure 3a one has the heavily to moderately enriched elements, while Figure 3b shows lightly enriched elements. •During wet season one has low elemental concentrations and high enrichment factors, with an opposite situation during the dry season. Figure 3 – Average crustal enrichment factors for (a) heavily to moderately enriched elements, and (b) lightly enriched elements. •Aerosol removal processes during the wet season limit the impact of local soil dust aerosol, allowing the identification of enriched elements associated with anthropic sources (e.g. Cu, Zn, Cd). Moderately enriched elements may be associated with mixed aerosol sources (e.g. salars, biogenic aerosol from Amazon Basin). •Figure 4 shows an example of the relationship between enrichment factors and the reference tracer (Al) concentration, which in this case is a proxy for local soil dust load. Figure 4a is for La, and Figure4b is for Cu. •La is a rare-earth element originated from soil dust, while Cu may originate from mining activities. Figure 4 – Seasonal enrichment factors and soil dust tracer concentration for (a) La and (b) Cu. •La enrichment factors in Figure 4a oscillate about mean soil composition ( i.e. they are close to unity), irrespectively of soil dust load. •In contrast, Figure 4b shows Cu enrichment across four orders of magnitude. As the samples get more heavily loaded on aerosol ( i.e. high Al concentration), Cu enrichment factors tend toward mean soil dust values, as the local aerosol outweighs the influence of other sources. This pattern is an indication of the existence of other non-soil dust aerosol sources, which grow in importance as the aerosol loading decreases in the samples. Acknowledgments This work was supported by CNPq grant 142.683/98-1 and CAPES grant BEX 045899-6 •For each element, histograms of enrichment factors reveal the plateau levels as peaks in the lowest histogram bins. Figure 5 illustrates this procedure for Cd. •These plateau levels of enrichment factors for high aerosol loadings can be seen as effective elemental ratios to Aluminum (i.e. soil dust tracer). A renormalization of these plateau values to unity gives the effective composition of the local background aerosol. Figure 5 – Enrichment factor plateau level for Cd. The histogram peak defines the plateau level marked as the red line. •Most elements show intermediate behaviors between those depicted for La and Cu. Usually they attain a plateau level of lowest enrichment factors in the high aerosol load end of the graph. This plateau is not always unitary since local soil composition may differ from the global mean value.
-
Upload
buck-booker -
Category
Documents
-
view
215 -
download
1
Transcript of Figure 1 - Illimani ice core site in Eastern Bolivian Andes, South America. Distance from Illimani...
Figure 1 - Illimani ice core site in Eastern Bolivian Andes, South America. Distance from Illimani to La Paz is about 40 Km.
•Two ice cores were drilled in June 1999 by a French-Swiss team at the summit of Nevado Illimani (1637’S, 6746’W, 6350 m asl, Figure 1). The cores were drilled to the bedrock, with total depths of 137m and 139m, corresponding to about 18,000 years (Knüsel et al., 2003; Ramirez et al., 2003), and transported frozen to European laboratories. This work shows the results obtained for the French core.
•Several analyses were done: electrical conductivity, gamma spectrometry, D and 18O measurements. The uppermost 50m of the French core was cut in 744 sub-samples in a cold room at -15°C. These 744 sub-samples, corresponding to about 80 years from 1919 to 1999, were analysed for insoluble aerosol content, major ions by ion chromatography and multielemental concentration by ICP-MS (Correia et al., 2003).
Alexandre Correia1,2 ([email protected]); Robert Delmas2; Rémi Freydier3; Jefferson C. Simões4; Jean-Denis Taupin5; Bernard Dupré3 and Paulo Artaxo1
Alexandre Correia1,2 ([email protected]); Robert Delmas2; Rémi Freydier3; Jefferson C. Simões4; Jean-Denis Taupin5; Bernard Dupré3 and Paulo Artaxo1
1Institute of Physics, University of São Paulo, R Matão Trav R 187, São Paulo, SP 05508-900, Brazil; 2LGGE, CNRS and Université Joseph Fourier, Grenoble, France; 3UMR5563, LMTG, CNRS and Université Paul Sabatier, Toulouse, France; 4Institute of Geosciences, Federal University of Rio Grande do Sul, Brazil; 5LGGE, IRD, Grenoble, France
EGS-AGU-EUG Joint Assemblies06-11 April 2003, Nice, France
Ice cores are important tools for reconstructing past environmental conditions, from climatic variations to atmospheric chemistry. Several snow and ice cores from polar regions were studied in last decades, especially from Greenland and Antarctica. Pioneering work by Thompson and co-workers (e.g. Thompson, et al., 1985) showed the importance of high-altitude tropical ice cores, with different features and challenges compared to polar ice cores. In general, tropical ice cores sites are closer to moisture and aerosol sources than polar cores. The aerosol load of tropical ice is therefore much higher than in polar regions. Measured impurities may come from a variety of sources, either natural or anthropic in origin, presenting high spatial and temporal variability of atmospheric concentrations and climate-related physicochemical characteristics.
Most Andean ice core studies aimed at paleoclimate reconstructions, for which the focus rests over a wide time scale covering several millennia, or over a few elemental concentration profiles. This work explores a high temporal resolution (i.e. sub-seasonal) ice core record extracted from the Andes, covering about 80 years during 20th century. This ice core was studied under several analytic techniques, among which the determination of elemental concentration profiles of 45 chemical species, ranging from Li to U, by Inductively Coupled Plasma-Mass Spectrometry (ICP-MS).
•About 80% of annual precipitation at Illimani occurs during austral summer (DJF), originating from the Amazon Basin to the East (Vuille, 1999).
•Winter precipitation also comes from Amazon, as convective disturbances that reach the site and break against the prevailing westerly flow (Grootes et al., 1989).
•Recent works show that El Niño can control precipitation over the Andes, even tough the humidity source is the Amazon Basin (Bradley et al., 2003; Hoffmann et al., 2003).
•The ICP-MS analyses of the 744 sub-samples provided elemental concentration profiles of aerosol composition for 45 elements ranging from Li to U, from 1919 to 1999. The Lithium concentration profile is shown in Figure 2.
•Lithium is associated with salt lakes in the Altiplano, the World’s largest reservoir for this element (Risacher, 1992). The strong seasonality in Figure 2 is a common feature of most elemental concentrations measured. It is attributed to yearly precipitation regimes at the site, with high concentrations during dry austral winter and low concentrations during wet austral summer, according to stable isotopes measurements.
Figure 2 – Lithium concentration profile and seasonal classification scheme exemplified for a profile section.
•Measurement of 45 chemical species by ICP-MS in 744 samples during the 20th century allowed studying concentration profiles at high temporal resolution.
•During dry season high aerosol loadings are observed, with low crustal enrichment factors. Wet season conditions favour identification of other sources by removing local aerosol by wet deposition processes.
•A chemical characterization for the background soil dust aerosol deposited during 20th century was derived from enrichment factor plateau levels at high aerosol loading conditions.
ReferencesBowen, H. (1979) Environmental Chemistry of
the Elements.Bradley, R.S. et al. (2003) Geophys. Res. Let. 30(4), doi: 10.1029/2002GL016546.
Correia et al. (2003) Atm. Phys. Chem., [accepted].
Grootes, P.M. et al. (1989) J. Geophys. Res., 94, 1187-1194.
Hoffmann, G. et al. (2003) Geophys. Res. Let., 30(4), doi: 10.1029/2002GL014870.
Knüsel, S. et al. (2003) J. Geophys. Res., 108(D6), doi: 10.1029/2001JD002028.
Ramirez, E. et al. (2003) Earth Plan. Sci. Let., [submitted].
Risacher, F. (1992) , Sci. Géol., Bull., 45, 3-4, 135-214.
Thompson et al. (1985) Science, 229, 971-973.Vuille, M (1999) Int. J. Climatol., 19, 1579-1600.
•The samples were classified as belonging to dry, wet, or transition seasons, as exemplified in Figure 2. This classification scheme takes all measured elemental concentrations into account.
•Figure 3 shows the average crustal enrichment factors for all measured elements, during dry and wet seasons. Reference is Aluminum in mean soil composition (Bowen, 1979). In Figure 3a one has the heavily to moderately enriched elements, while Figure 3b shows lightly enriched elements.
•During wet season one has low elemental concentrations and high enrichment factors, with an opposite situation during the dry season.
Figure 3 – Average crustal enrichment factors for (a) heavily to moderately enriched elements, and (b) lightly enriched elements.
•Aerosol removal processes during the wet season limit the impact of local soil dust aerosol, allowing the identification of enriched elements associated with anthropic sources (e.g. Cu, Zn, Cd). Moderately enriched elements may be associated with mixed aerosol sources (e.g. salars, biogenic aerosol from Amazon Basin).
•Figure 4 shows an example of the relationship between enrichment factors and the reference tracer (Al) concentration, which in this case is a proxy for local soil dust load. Figure 4a is for La, and Figure4b is for Cu.
•La is a rare-earth element originated from soil dust, while Cu may originate from mining activities.
Figure 4 – Seasonal enrichment factors and soil dust tracer concentration for (a) La and (b) Cu.
•La enrichment factors in Figure 4a oscillate about mean soil composition (i.e. they are close to unity), irrespectively of soil dust load.
•In contrast, Figure 4b shows Cu enrichment across four orders of magnitude. As the samples get more heavily loaded on aerosol (i.e. high Al concentration), Cu enrichment factors tend toward mean soil dust values, as the local aerosol outweighs the influence of other sources. This pattern is an indication of the existence of other non-soil dust aerosol sources, which grow in importance as the aerosol loading decreases in the samples.
AcknowledgmentsThis work was supported by CNPq grant 142.683/98-1 and CAPES grant BEX 045899-6
•For each element, histograms of enrichment factors reveal the plateau levels as peaks in the lowest histogram bins. Figure 5 illustrates this procedure for Cd.
•These plateau levels of enrichment factors for high aerosol loadings can be seen as effective elemental ratios to Aluminum (i.e. soil dust tracer). A renormalization of these plateau values to unity gives the effective composition of the local background aerosol.
Figure 5 – Enrichment factor plateau level for Cd. The histogram peak defines the plateau level marked as the red line.
•Most elements show intermediate behaviors between those depicted for La and Cu. Usually they attain a plateau level of lowest enrichment factors in the high aerosol load end of the graph. This plateau is not always unitary since local soil composition may differ from the global mean value.
