Production rate and concentration of 10Be in the rain and soil.
Climate signals from 10Be records in marine environments have been studied for last two decades (Aldahan et al., 1997, Bourlès et al., 1989, Christl et al., 2003, Eisenhauer et al., 1994, Horiuchi et al., 2000, 2001, Kim and Nam, 2010, Knudsen et al., 2008, McHargue et al., 2010, McHargue and Donahue, 2005). Understanding of regional climate signals is feasible through not only 10Be but also 9Be from the sediment. This is because 9Be is terrigenous origin while 10Be signal is affected by climatic condition and production at the top of atmosphere. Recent study from the East Sea of Korea (05-GCRP-21) indicated that climate signals from 10Be records of Korean marine sediment are generally representing global climate variations during warm and cold periods from Last Glacial Maximum to Holocene and also MIS (marine isotope stage) 6 to Eemian. The 10Be records of the East Sea are well compared with those from the oxygen isotopic record of this marginal sea (Kim and Nam, 2010). During the warm interglacial periods the 10Be concentrations per sediment mass have significantly increased while during the cold glacial periods those have decreased (Aldahan et al., 1997, Eisenhauer et al., 1994). This result also shows that a vivid record of 10Be/9Be indicates a significant increase of 10Be at a time of 120 kyr, which might be an indication of the paleomagentic excursion.
Interestingly, it was found that the 10Be concentrations per 1g sediment of this region were about 30% lower than other 10Be records of largely open marine environment. We also found that 10Be concentrations of the Blake Outer Ridge were similar to those from Korean marine sediments (McHargue et al., 2000). Two study areas are located nearby large continents: the East Asia (the East Sea) vs North America (the Atlantic). This could be caused by sediment inflow to the marine environment which is close enough to the continent. Therefore, local marine environmental influence is revealed through the beryllium isotopes. Both cases would have similar climatic signals due to their geographical locations nearby continent. The lower 10Be concentrations for these regions could be also involved in ocean current and circulation. Relatively deep sea water of these regions may not be well mixed rapidly with the surface water and old sea water with relatively lower 10Be concentration remains in the sediment records. For this chapter, we investigated climatic signals from Be isotope records of the East Sea of Korea and the Mendeleev Ridge of the Arctic Ocean and compared with the records from the Blake Outer Ridge studied by McHargue et al., 2000. In addition, global 10Be records of marine sediments for various regions will be briefly discussed. This chapter would provide a new insight guide into understanding climate signals through 10Be records of various marine environments.
2. Beryllium isotopes in terrestrial environments
2.1. Cosmic ray induced 10Be production rate
Production rates of cosmogenic isotopes depend on geomagnetic latitude, altitude, and flux of incoming cosmic rays to the earth (Lal, 1988). The Earth’s geomagnetic field deflects incoming cosmic rays and has an effect on the production rate of
Production rates as a function of both latitude and altitude have been studied in the past. For a few decades, models from Lal and Peters, 1967 and Lal, 1988, have been widely used for the scaling factors for production rates of
2.1.1. Production rate in the atmosphere
10Be is produced in the atmosphere by nuclear interactions with oxygen and nitrogen (Peters, 1955, Goess and Phillips, 2001). The intensity of the cosmic ray flux depends on galactic and solar sources, and modulation by the heliomagnetic and geomagnetic fields. Both 10Be is produced by spallation reactions in the atmosphere, and then 10Be is well mixed up (Ueikkila et al., 2009) and removed from the atmosphere by precipitation scavenging of aerosol particles to land and sea. Eventually, these nuclides are deposited within ocean sediment. The 10Be concentration at 10.7 km of stratosphere and at 19.2 km in the tropospheric concentration are known to be 7 x 106 atoms/m3, and 1.3 x 107 atoms/m3, respectively. The global average 10Be production rate is found to be (1.21±70) x 106 atoms/cm2/yr (Monaghan et al., 1985). Estimates of the 10Be production rate derived from measurements on ice cores, lake sediments, and deep-sea sediments range from 0.35 x 106 atoms/cm2/yr to 1.89 x 106 atoms/cm2/yr. (Monaghan et al., 1985). Castagnolie et al., 2003 demonstrated reconstruction of the modulation parameter M from the open solar magnetic flux proposed by Solanki et al., 2000, and experimental values calculated from the GCR spectra measured with balloons and spacecraft are compared well with 10Be concentration measured at the Dye3 ice core, assuming constant accumulation rate during the period of 1810-1997. The production rate of 10Be ranged from 0.015 to 0.025 atoms/cm2/s (Castagnolie et al., 2003).
The precipitation onto the surface of the Earth and the deposition of 10Be in soils is influenced by climate. In turn, the production of 10Be in the atmosphere is influenced by the magnetic dipole field of the Earth to which it is inversely related. This relationship between the production of 10Be and the geomagnetic field has been shown by the correlation between the variations of 10Be and those of the me paleo-asuredinclination data of the dipole field in sediments (Frank et al., 1997, Frank,2000, Masarik et al., 2001, Laj et al., 2000), and the concentrations of 10Be in marine sediments and the measured paleointensity (Carcaillet et al., 2004, McHargue and Donahue, 2005).
The influence of climate on the deposition of 10Be, otherwise is problematic for interpretations of the cosmic-ray flux, in itself is a worthy subject for study. For example, variations in the deposition rates of 10Be and sediments affect the 10Be/9Be ratio due to the uneven mixing of the two isotopes in the hydrological cycle. That, 10Be, produced largely in the atmosphere, is transported to the surface of the earth by rain and dry precipitation to the sea. In contrast, 9Be, derived from terrigenous materials, is transported to the sea largely by rivers, and to a lesser extent by atmospheric deposition.
|USA continent||(1.38±0.36) ~(3.96 ± 0.35) x 106 atoms/cm2/yr||rain||Monagahan et al., 1985|
|USA Hawaii||1.9 x 103 ~ 8.94 x 104 atoms/g||rain||Monagahan et al., 1985|
|Tropical region||Avg. 1.53 x 104 atoms/g||rain||Monagahan et al., 1985|
|Illinois, USA||2 ~7 x 107 atoms/g||soil||Brown et al., 1989|
|Japan, Kikari Is||(0.80 ~ 7.17) x 109 atoms/g||soil||Maejima et al., 2005|
|Japan, Kikari Is||(2.0 ~ 3.5) x 106 atoms/cm/yr||rain||Maejima et al., 2005|
|New Zealand||(2.1 ~ 2.9) x 104 atoms/g||rain||Graham et al., 2003|
|Korea, Masanri||(0.67 ~ 1.47) x 108 atom/g||soil||Kim et al., 2011a|
|India||(0.43~3.34) x 107 atoms/l||rain||Somayajulu et al., 1984|
|Global||(1.21 ±0.70) x 106 atoms/cm2/yr||rain||Monagahan et al., 1985|
2.1.2. 10Be in land surface
Precipitation was collected for approximately one year during 1980 at seven localities in the continental U.S.A. (Monagahan et al., 1985). The 10Be flux ranged from (1.38±0.36) x 106 to (3.96±0.35) x 106 atoms/cm2/yr (Monagahan et al., 1985). In the case of Hawaii, 10Be concentration ranges from 1.9 x 103 to 8.94 x 104 atoms/g in rain water. The mean 10Be deposition rate in temperate latitude is determined to be 1.53 x 104 atoms/g in rain water (20% error). Monagahan et al., 1985 indicated that the concentration of 10Be in surface soils and river sediments varies between 107 and 109 10Be atoms/g soil with the modal concentration of 10Be lying between 4 x 108 and 6 x 108 atom/g. The relationship between annual rainfall and 10Be deposition rate is plotted to be linearly proportional to each other (Maejima et al., 2005). This study shows that 10Be fluxes (cm/day) for the two rain collection sites are relatively higher during a collection period of January 22 to April 22 than other collection period (Maejima et al., 2005).
Seasonal variations for 7Be and 10Be concentration in Tokyo and Hachijo-Island during a period of 2002 to 2003 were similar to each other. The peak value for 7Be and 10Be concentration appeared in April and October, respectively. Especially, in April when stratosphere-troposphere exchange occurs, peak values for the atomic ratio 10Be/7Be appeared. Low 7Be and 10Be concentrations and the atomic ratio of 10Be/7Be appeared in summer, July to August. Because the composition of the aerosol of Tokyo was almost same to the nearby soil, it is considered that Tokyo was strongly influenced by re-suspended soil contamination. Yamagata et al., 2005 indicated that using Al concentration in the aerosols, the enrichment of 10Be concentration by re-suspended soil contamination was estimated to be about 30%.
In the case of Southern Hemisphere, Graham et al., 2003 demonstrated 7Be and 10Be fluxes at 36 to 45ºS were determined to be (1.2~14) x 107 atoms/kg and (2.1~2.9) x 107 atoms/kg, respectively. These results are similar to those for rain sampled at mid-latitude sites across the USA from 1986 to 1994. The annual 7Be and 10Be flux rates are ~15 and ~27 x 109 atoms/m2, respectively, at the northern sites of Leigh and Gracefield, and are significantly lower at ~9 and ~19 x 109 atoms/m2, respectively, at the southern site of Denidin, because of the lower average rainfall there. Graham et al., 2003 indicated that 7Be/10Be in New Zealand ranged 0.47 to 0.61 and this is significantly lower than the ratio in USA (0.69~0.78). This is due to re-suspended dust to the primary atmospheric 10Be in the rain sample in New Zealand. Interestingly, the ratio of 7Be/10Be at three sites are 0.70 (Leigh), 0.65 (Gracefield) and 0.50 (Dunedin). These results suggest an overall reduction in the 7Be/10Be ratio from north to south, due to increasing residence time for Be isotopes in the atmosphere above New Zealand. The mean residence time for 7Be and 10Be in the atmosphere above New Zealand range from 77 to 109 days and are lower in the summer than the winter due to transfer of older stratospheric air to the tropopause in late spring-early summer (Graham et al., 2003).
Maejiam et al., 2005 demonstrated that 10Be concentrations of six soil samples on the raised coral reef terraces of Kikari Island, southwest Japan ranged from 0.80 to 7.17 x 109 atoms/g. The annual deposition rate of 10Be from the atmosphere to Kikari Island from 2000 and 2002 ranged from 2.0 to 3.5 atoms/cm2/y. The minimum absolute age was calculated from the inventory of meteoric 10Be in the soil, and the annual deposition rates of 10Be are ranged from 8 to 136 kyr (Maejima et al., 2005). A 36 cm of soil depth profile from the Roberts Massif, Antarctica was studied to obtain the age of soil by Graham et al., 1997. The sampling site is located in the edge of the nearby East Antarctic Ice Sheet at an altitude of 2700 m. This site is considered to have been ice-free for an extremely long period of time, of the order of several million years. The results of Graham et al., 1997 determined its minimum soil age of 12 million years which is much older than other 40Ar/38Ar dating result of 8 million years for volcanic deposit, Scoria associated with soils on the tills laid down by the Meserve Glacier, Antarctica.
2.1.3. 10Be in the ocean
Using radiocarbon, the sedimentation rates during glacial periods and deglacial periods for the western Arctic Ocean were found to be 0.5 cm/kyr and 1-2 cm/kyr, respectively (Darby et al., 1997). A recent study shows that the concentration of 10Be in the authigenic fraction of the sediment normalized to the total sediment mass is indirectly correlated to the oxygen isotope curve (McHargue and Donahue, 2005). For example, a low 10Be/9Be ratio in sediments would imply that terrestrial source of 9Be has increased compared to the more oceanic 10Be. Correlation of 10Be with δ18O recorded in marine sediment from the Blake Outer Ridge (DSDP site 72) shows a climatic effect on the 10Be record in addition to cosmogenic effects. Age-corrected 10Be variability in the sediment cores studied in Aldahan et al., 1997 and the oxygen isotope stratigraphy with the climatic stages numbered from 1 to 10, is generated from Aldahan et al., 1997. 10Be from sediments of the Arctic Ocean covering the past 350 kyr shows well defined trends of Be isotopes coincident with interglacial/glacial climatic cycles and demonstrates that the sedimentation rates are higher during glacial periods and lower generally due to low sedimentation/accumulation rate during interglacial periods (Aldahan et al., 1997).
The 10Be records of four sediment cores, taken along a transect from the Norwegian Sea via the Fram Strait to the Arctic Ocean, demonstrate that high 10Be concentration are related to interglacial stages and that sediment sequences with low 10Be concentration are related to glacial stages. This study confirms that the sharp contrast of high and low 10Be concentrations at climatic stage boundaries are an independent proxy for climatic and sedimentary change, and can be applied for 10Be stratigraphic dating of sediment cores (Eisenhauer, 1994).
Also, Carcaillet et al., 2004 produced high resolution authigenic 10Be/9Be records over the last 300 kyr from sedimentary cores off the Portuguese coast. Comparison of 10Be/9Be and benthic δ18O records from the two cores suggested that dipole moment lows may be associated with the end of interglacial episodes, and have a quasi-period of 100 kyr (Carcaillet et al., 2004). In a recent study, McHargue and Donahue, 2005 showed a strong correlation between 10Be and oxygen isotope stages from the Blake Outer Ridge in the Atlantic Ocean. This relationship between climate and 10Be deposition suggest that 10Be could be used, in addition to, or as proxy for δ18O in the studies of the climatic influences on marine sedimentation.
Other considerations are the carbonate flux in sediment which is strongly correlated to 10Be flux, and carbonate-free sediments from which δ18O is difficult to obtain from foraminifera. In addition, the two isotopes of beryllium, as stated above, are source dependent, thus the relationship of 10Be to 9Be in the sediments is a function of the relative contributions from atmospheric and terrestrial sources, and their mixing time in the sea.
Paleomagnetic intensity obtained from deep sea cores is well described in recent publications (Valet, 2001, Guyodo et al., 2000, Guyodo and Richter, 1999). However, it was found that climatic influence on 10Be deposition can be significant, obscuring those variations from its production in the atmosphere, and thus must be addressed (Frank et al, 1997, Kok, 1999). Scavenging corrected 10Be records compared to calculated variation of the global 10Be production based on paleomagnetic intensity records (Christl et al., 2003) (Figure 2) show the variation of 10Be flux in each location with general agreement of inversely proposal to the paleomagentic intensity from Mazaud et al., 1994, Yamazaki and Iokyr et al., 1994, and Guyodo and Valet, 1999.
2.2. 10Be chemistry
Generally, to extract authigenic beryllium isotopes from sediments, the procedure of Bourles et al., 1989 is used. About one gram of sediment is leached in a solution of 25% acetic acid and hydroxlyamine-HCl to separate the “authigenic” fraction of the sediment from the “terrigenous” fraction. Most samples had more than 1 g of dry sediment; however, in the case of less than 1 g, two or three neighbouring samples were combined for the analysis. When 10Be is normalized to the mass of the authigenic fraction, it should more accurately reflect its concentration in ocean water than 10Be normalized to the total mass of the sediment (McHargue and Donahue, 2005, McHargue et al., 2010). This fraction is mostly composed of exchangeable ions, carbonates, and Fe-Mn hydroxides. Two aliquots of the leachate are prepared, one for the elemental analysis with ICP-MS/ICP-AES, and one for the preparation of AMS samples.
Figure 3 shows the flow chart of 10Be chemistry to extract authigenic beryllium from sediment. This chemistry includes two steps of purification procedures using perchloric acid and nitric/hydrochloric acid. These steps are important to extract authigenic brylllium isotopes. Sometimes, this step is repeated to remove unwanted organic materials. When the unwanted organic materials are not completely removed, the residue sample is often difficult to dissolve in weak acidic solution for ICP analysis. This also causes a further problem in the step of Be separation using Na-EDTA. The concentration level for Be is mostly in ppb range; therefore, Be analysis was performed using ICP-MS. For AMS, the Be fraction is precipitated as Be(OH)2 and combusted to BeO. 10Be/9Be ratios for chemical blank are found to be less than ~ 1 x 10-14 with 2 mg of 9Be carrier. 9Be and other elements can be measured by ICP-MS and ICP-AES.
2.3. The climate signal of 10Be from nearby a continent
The signal of 10Be from the East Sea in the Pacific Ocean and the Black Outer Ridge in the North Atlantic Ocean may give similar climatic influence because of its proximity to continents (Kim and Nam, 2010, McHargue et al., 2000). The depths of basins where sediment core collected were about 3,700 m and 3,818 m in the East Sea and Blake Outer Ridge, respectively. Because both cores were collected in the basin of the ocean, we might expect water circulation could be weaker. This could allow rather older waters can remain at the bottom of the basin. In this case, we may expect lower concentration of 10Be in sediment compared to samples collected from other open seas (Bourlès et al., 1989, Knudsen et al., 2008). Also, the influence from the continent would be similar in both regions. The terrestrial origin of 10Be over glacial/deglacial time period may similarly appear.
As shown in Figure 4, three locations were examined with respect to 10Be concentration as a function of time and any related proxies for each site. Mostly, maximum 10Be concentrations in various marine sediment samples appear to be above 1 x 109 atoms/g sediment (Bourlès et al., 1989, Knudsen et al., 2008). Both the East Sea and Blake Outer Ridge, 10Be concentrations are reached at 8 x 108 atoms/g. This value is at least 30 percent lower than the most maximum value of 10Be in each marine sediment core. Also, when 9Be is investigated with 10Be, 9Be signal may be another indicator as a signal of sediment input from the land to the offshore. In the case of the study of the East Sea, the 9Be values also show similar trend to those of 10Be. his shows that both warmer periods of the Holocene and the Eemian,
wetter and warmer climate influenced 9Be to be transported from the land to the ocean. 10Be is transported from the both land and atmosphere. The signal of 10Be/9Be especially stands for lack of either lack of 9Be transport or higher production rate of 10Be, possibly associated with paleomagnetic intensity. In Figure 5, A, B, and C are associated with relatively higher 10Be/9Be compared to neighbouring 10Be/9Be values. A could be partially due to lowered value of 9Be; B and C could be due to production rate of 10Be. This study shows high production rate of Be at 15.5 and near 120 kyr. Also, lower 10Be production rate is shown at 130.6 kyr during MIS 6 (Figure 5) (Kim and Nam, 2010). These three points can be identified easily with 10Be/9Be rations. In the Figure 5, the regions associated with climatic influence are clearly shown as the 10Be/9B ratios to be within the value between 2 and 3. This observation could be useful in future analysis. Based on Wagner et al., 2000, lower paleomagnetic intensities are associated with the ages at 1.5, 2.5, 4.0, and 6.5 kyr (Christl et al., 2003). Therefore, B could be likely involved in higher production rate of 10Be at 1.5 kyr (Kim and Nam, 2010) (Figure 5).
Figure 6 shows the 10Be concentration and M/Mo as determined from measured NRM/ARM of core CH88-10P with respect to depth and time. This figure shows that 10Be concentration is inversely proportional to the relative paleomagnetic intensity. The production rate of 10Be occurred at about 40 and 65 kyr. The peak values of 10Be/9Be reached at maximum at about 40 kyr. This time period is named as Laschamp paleomagnetic excursion where the expected 10Be reaches at a maximum value. These paleomagnetic excursions are well compared with GRIP records.
A recent investigation on 10Be and 9Be from the Mendeleev ridge in the Arctic Ocean shows that 10Be record at 75 kyr reveals production rate decrease evidently for at least 35 kyr of duration (Kim et al., 2011b). At this time the paleomagnetic intensity is found to be at maximum (Figure 2) (Christl et al., 2003; Flank et al., 1997). Interestingly, the values of magnetic susceptibility (Guyodo and Valet, 1996) obtained from a lake (Lac du Bouchet, France) are high as well as δ18O. Also, 9Be is relatively high which stands for a warm climate. The results of this study confirm that 10Be reveals predominantly paleomagnetic features over the δ18O at the extreme point of paleomagnetic intensity. This situation brings us to have precaution in a misuse of 10Be as a climatic tracer. This study confirms the fact that the 10Be record for climatic tracer, comparison with 9Be is essential. When both beryllium isotopes behaves similarly, the pattern of Be can be used to determine whether the record of Beryllium isotopes is associated with colder or warmer climate based on their consistent concentration trend. The total authigenic 10Be and 9Be (9Be>>10Be) can be referred as 9Be because of their amount ratio in terrestrial environment. The 9Be can be used as another climatic indicator like Sr, Ca, opal, TOC. The study at The Mendeleev Ridge confirms that 9Be generally has a positive correlation with opal, TOC, δ13Corg and negative correlation with CaCO3 (Nam unpublished) (Figure 7). General trend of 9Be clearly show the anti-correlation between 9Be and Ca or Sr (Boulrès et al., 1989, Kim et al., 2011b). Therefore, we can conclude that 10Be has a positive correlation with δ18O which gives 10Be to be used as a climate indicator, however, this is only true when 9Be reveals similar climatic pattern with 10Be. Both 10Be and 9Be show lower concentration at a cold/dry climate and higher concentration at a warm/wet climate period (Kim and Nam, 2010).
2.4. Current problem and future research
A number of investigations show that there has been positive correlation between oxygen isotope and 10Be concentration (Aldahan et al., 1997). Also, a positive correlation between oxygen isotopes and paleomagnetic intensity and also magnetic susceptibility is investigated (Carcaillet et al., 2004). During the Holocene, paleomagnetic intensity was gradually increased since the time of Laschamp excursion. This confirmed that production rate of 10Be at present is the lowest value since Laschamp excursion. However, the 10Be values recent years are higher than the 10Be concentration and the trend of 10Be is similar to that of δ18O value. This implies that 10Be is closely related to climatic and temperature variation. Because of this contradictory fact, confining the cause of climate change using nuclides which sun’s activity related became important.
Although there have been a number of investigation on climate study using above parameters, obscurity in finding the cause of climate change is still remain. The relationship among production rate of 10Be, paleomagenetic intensity, and Sun-climate connection was studied (Sharma, 2002). This study estimated changes in 10Be production rate and the geomagnetic field intensity, variations in solar activity were calculated for the last 200 kyr., and confirms that the production of 10Be in the Earth’s atmosphere depends on the galactic cosmic ray influx that is affected by the solar surface magnetic activity and the geomagnetic dipole strength. However, large variations in the solar activity are evident. The marine δ18O record and solar modulation are strongly correlated at the 100 kyr timescale. This proposes that variation in solar activity control the 100 kyr glacial-interglacial cycles. Sharma, 2002 suggested that the 10Be production rate variations may have been under-estimated during the interval between 115 kyr and 125 kyr, and may have biased the results (Sharma, 2002).
Usoskin et al., 2004 indicated that the reconstructed sunspot record exhibits a prominent period of about 600 years, in agreement with earlier observations based on cosmogenic isotopes. Also, there is evidence for the century scale Gleissberg cycle and a number of shorter quasi-periodicities whose periods seem to fluctuate on millennium time scale. This invalidates the earlier extrapolation of multi-harmonic representation of sunspot activity over extended time intervals and the present high level of sunspot activity is unprecedented on the millennium time scale (Usoskin et al., 2004). Accepting solar forcing of Holocene and galacial climatic shift implies that the climate system is far more sensitive to small variation in solar activity than generally believed. In order to fully understand how sensitive climate really is for variations in solar activity, we need to look for additional evidence and to quantify such evidence, both in paleorecords and in observations of present climate with models to estimate climate change in the future (Geel et al., 1999).
As a climate indicator, 10Be has been frequently investigated because of its property associated with rainfall, dust fallout and its production mechanism in the atmosphere by cosmic-rays. Similar patterns of 10Be and δ18O, magnetic susceptibility records show climatic influence, however, 10Be is incorporated with the production rate which is inversely proportional to the paleomagenetic intensity. The cyclic orbital forcing effect toward 10Be, δ18O, and paleomagenetic intensity are connected, 10Be signal is clearly mixed with climatic component and earth’s paleo magnetic strength. Scrutinizing authigenic 10Be with 9Be, a region either climatic or production related zone can be evidently identified by looking at the 10Be/9Be ratios. Investigation of 9Be is another useful tracer to examine climatic influence of marine environments with other multi-proxies which have positive or anti-correlated with 9Be. Marine environments like the East Sea of Korea, Blake Outer Ridge, and Mendeleev Ridge are associated with significant terrigeous input during deglacial periods. Examining both beryllium isotopes together with other multi-proxy stratigraphy will provide understanding the pattern of environmental change at various glacial/interglacial events much more evidently.
This study is partially supported by Korean-IODP at Korea Institute of Geoscience and Mineral Resources and also funded by K-Polar (PP11070) at Korea Polar Research Institute.