Ferrobasalts from the Central Indian Ocean Basin

We report the occurrence of ferrobasalts recovered from the Central Indian Ocean Basin crust generated at the Southeast Indian Ridge during a phase of moderate to fast spreading accretion (～110–190?mm/yr, full rate).The rocks are rich in plagioclase, FeO* (13–19%), and TiO₂ (2.27–2.76%), poor in olivine and MgO (3.44–6.20%), and associated with topographic highs and increased amplitude magnetic anomalies corresponding to chrons A25 and A24. We suggest that secon dary eruptions from ancient N-MORB magma, which may have been trapped at a shallow depth in a horizon of neutral buoyancy, could have produced the ferrobasalts.     

Long-term variations of surface and intermediate waters in the southern Indian Ocean along 32°S

Variations of water properties in surface and intermediate layers along 32°S in the southern Indian Ocean were examined using a 50-year (1960–2010) time series reproduced from historical hydrographic and Argo data by using optimum interpolation. Salinity in the 26.7–27.3σ_θ density layer decreased significantly over the whole section, at a maximum rate of 0.02 decade⁻¹ at 26.8–26.9σ_θ, for the 50-year average. Three deoxygenating cores were identified east of 75°E, and the increasing rate of apparent oxygen utilization in the most prominent core (26.9–27.0σ_θ) exceeded 0.05 ml l⁻¹ decade⁻¹. The pycnostad core of Subantarctic Mode Water (SAMW) and the salinity minimum of Antarctic Intermediate Water shifted slightly toward the lighter layers. Comparisons with trans-Indian Ocean survey data from 1936 suggest that the tendencies found in the time series began before 1960. Interestingly, cores of many prominent trends were located just offshore of Australia at 26.7–27.0σ_θ, which is in the SAMW density range. Spectrum analysis revealed that two oscillation components with time scales of about 40 and 10 years were dominant in the subsurface layers. Our results are fairly consistent with, and thus support, the oceanic responses in the southern Indian Ocean to anthropogenic climate change predicted by model studies.     

Dissolved and labile particulate Zr,Hf, Nb,Ta, Mo and W in the western North Pacific Ocean

Dissolved and labile particulate Zr, Hf, Nb, Ta, Mo and W were determined at stations K1 (51°N, 165°E), K2 (47°N, 160°E), KNOT (44°N, 155°E) and 35N (35°N, 160°E) in the western North Pacific Ocean. A portion of seawater for dissolved species (D) was passed through a 0.2 μm Nuclepore filter and acidified to pH 2.2 with HCl and HF. A portion of seawater for acid-dissolvable species (AD) was acidified without filtration. Labile particulate (LP) species is defined as AD minus D, which represents a chemically labile fraction of particulate species. D-Zr, Hf and Ta increase with depth, Nb shows a slight depletion in surface water, whereas Mo and W have a conservative vertical profile. The concentration range of D-Zr, Hf, Nb, Ta and W is 31–275, 0.14–0.95, 4.0–7.2, 0.08–0.29 and 40–51 pmol kg⁻¹, respectively, whereas that of Mo is 97–105 nmol kg⁻¹. LP-species of Zr, Hf and Ta account for 10–14% of AD in average and increase up to 25% below 4000 m, whereas those for Mo and W are negligible. In contrast, LP-Nb shows maxima (up to 27%) in surface water. We also found that D-Zr/Hf, Nb/Ta and Mo/W mole ratios generally increase in the order continental crust < river water < coastal sea < open ocean.     

Late Pleistocene channel–levee development on Monterey submarine fan, central California

 Much of the modern upper (proximal) Monterey fan is a channel–levee complex, the Upper Turbidite Sequence (UTS), that was deeply eroded after the channel breached a volcanic ridge to reach a deeper base level. Ages of sediment samples collected with the ALVIN submersible from the deepest outcrop within the channel–levee system, 390 m below the adjacent western levee crest, indicate that the UTS deposits accumulated at ≥1 m ka^-1 during the last 500 ka. Neogene and Early Pleistocene sediment accumulation on the fan prior to the UTS was much slower (<0.03 m ka^-1), and underlying turbidite systems(?) had substantially different morphologic expression(s). Received: 10 February 1998 / Revision received: 6 July 1998     

Nutrient regeneration at bottom after a massive spring bloom in a subarctic coastal environment,Funka Bay,Japan

Nutrient regeneration and oxygen consumption after a spring bloom in Funka Bay were studied on monthly survey cruises from February to November 1998 and from March to December 1999. A high concentration of ammonium (more than 4 μmol l⁻¹) was observed near the bottom (80–90 m) after April. Phosphate and silicate gradually accumulated and dissolved oxygen decreased in the same layer. Salinity near the bottom did not change until summer, leading to the presumption that the system in this layer is semi-closed, so regenerated nutrients were preserved until September. Nitrification due to the oxidation of ammonium to nitrate was observed after June. Nitrite, an intermediate product, was detected at 4–7 μmol L⁻¹ in June and July 1999. Assuming that decomposition is a first order reaction, the rate constant for decomposition of organic nitrogen was determined to be 0.014 and 0.008 d⁻¹ in 1998 and 1999, respectively. The ammonium oxidation rate increased rapidly when the ambient ammonium concentration exceeded 5 μmol L⁻¹. We also performed a budget calculation for the regeneration process. The total amount of N regenerated in the whole water column was 287.4 mmol N m⁻² in 4 months, which is equal to 22.8 gC m⁻², assuming the Redfield C to N ratio. This is 34% of the primary production during the spring bloom and is comparable to the export production of 25 gC m⁻² measured by a sediment trap at 60 m (Miyake et al., 1998).     

Hydrocarbon gases in sediments and mud breccia from the central and eastern part of the Mediterranean Ridge

 Microbiologically produced hydrocarbon gases (HCGs) in normal pelagic sediments have lithologically controlled distribution – their content increases sharply in organic-rich sapropels. Generally, the HCG content is low (<10000 nl/l), The amount of heavy HCGs is high (up to 60%, unsaturated HCGs dominate the saturated ones, and ethylene is prevalent. The same features were found in sediment and mud breccia from inactive mud volcanoes. Thermogenic HCG was determined in active mud volcanoes and in a high salinity fluid vent. They are characterized by: high methane concentration, δ¹³C(CH₄) =−37.1 to −57.8%, essential ethane contents (2–7%), absence of unsaturated HCG, and the prevalence of iso-over n-butane.     

The distribution of chlorophyll a in the tropical eastern Indian Ocean in austral summer

To study the effect of hydrographic factors on the spatial distributions of chlorophyll a (Chl a), an investigation was carried out in the tropical eastern Indian Ocean (80 -100 E along 7 S, and 7 -18 S along 80 E) in December 2010. The fluorescent method was used to obtain total Chl a and size-fractioned Chl a at the 26 stations. The results show that surface Chl a concentration averaged at (0.168 ± 0.095) mg/m 3 s.d. (range: 0.034-0.475 mg/m 3 ), concentrations appeared to be higher in the west for longitudinal variations, and higher in the north for latitudinal variations. Furthermore, the surface Chl a concentration was lower (0.034-0.066 mg/m 3 ) in the region to the south of 16 S. There was a strong subsurface Chl a maximum layer at all stations and the depth of the Chl a maximum increased towards to the east and south along with the respective nitracline. The spatial variation of Chl a was significant: correlation and regression analysis suggests that it was primarily affected by PO 3 4 , N(NO 3 -N+NO 2 -N) and temperature. Size-fractionated Chl a concentration clearly showed that the study area was a typical oligotrophic open ocean, in which picophytoplankton dominated, accounting for approximately 67.8% of total Chl a, followed by nanophytoplankton (24.5%) and microphytoplankton (7.6%). The two larger fractions were sensitive to the limitation of P, while picophytoplankton was primarily affected by temperature.     

Coral extension rates in the NW Indian Ocean I: reconstruction of 20th century SST variability and monsoon current strength

To date, only a few coral proxy studies have investigated coral growth as an indicator of climate variability. This study presents the first extension-rate record (Porites lutea) from the Maldives (NW Indian Ocean), inferred from skeletal δ¹⁸O chronology for the lagoon of Rasdhoo Atoll (4°N/73°W) in the central area of the Maldives, influenced by the Indian monsoon. The record spans 90 years over the period 1917–2007. The mean annual extension over this period was 9.9 mm/year, and an increase of annual extension rates until 1990 by 3 mm/year can be explained by a rise of 0.7°C in sea surface temperature (SST) in this region. After 1990, the extension rates do no continue increasing, possibly due to ecological stress caused by progressive ocean warming and acidification. The correlation between annual extension rates and SSTs is thus significant and strong in the lower part of the record until 1955 (r = +0.69, p < 0.0001), but weaker thereafter (r = +0.44, p < 0.001). The extension rates yield a distinct interannual variability of 3–4 years, caused by interannual SST fluctuations driven by the El Ni?o-Southern Oscillation. A variability of 8–9 years is likely driven by SST variations endemic to the Indian Ocean. Spectral peaks between 18–19 years and 6–7 years cannot be explained by SST fluctuations, but by variations in the strength of the SW monsoon currents. It is suggested that during phases of stronger monsoon activity, the coral sacrificed coral extension in favor of a denser, more robust skeleton. The geomorphology of the atoll may strengthen the potential of this new coral archive to track climate variability.     

Assessment of nitrogen loading from the Kiso-Sansen Rivers into Ise Bay using stable isotopes

Concentrations of particulate organic nitrogen (PN), dissolved inorganic nitrogen (DIN), and their nitrogen isotope ratios (δ ¹⁵N) in the Kiso-Sansen Rivers were determined from monthly observations over the course of a year to assess variations in the form and sources of riverine nitrogen discharged into Ise Bay. The δ ¹⁵N values of NO₃ ⁻ observed in the Kiso-Sansen Rivers showed a logarithmic decreasing trend from 8 to 0‰, which varied with the river discharge, indicating mixing between point sources with high δ ¹⁵N and non-point sources with low δ ¹⁵N. The influence of isotope fractionation of in situ biogeochemical processes (mainly DIN assimilation by phytoplankton) on δ ¹⁵N of NO₃ ⁻ was negligible, because sufficient concentrations of NH₄ ⁺ for phytoplankton demand would inhibit the assimilation of NO₃ ⁻. A simple relationship between river discharge and δ ¹⁵N of NO₃ ⁻ showed that the fraction of total NO₃ ⁻ flux arising from point sources increased from 4.0–6.3% (1.1–1.8 tN day⁻¹) during higher discharge (>600 m³ s⁻¹) to 30.2–48.3% (2.6–4.1 tN day⁻¹) during lower discharge (<300 m³ s⁻¹). Riverine NO₃ ⁻ discharge from the Kiso-Sansen Rivers can explain 75% of the variations in surface NO₃ ⁻ at the head of Ise Bay over the year.     

Helium Isotopes in Pacific Waters from Adjacent Region of Honshu, Japan

We have measured helium isotopic ratios of thirty-seven Pacific water samples from various depths collected in adjacent regions of Honshu, Japan. The ³He/⁴He ratios vary significantly from 0.989 R _atm to 1.208 R _atm where R _atm is the atmospheric ratio of 1.39 × 10⁻⁶. The mid-depth (750–1500 m) profile of ³He/⁴He ratios at ST-1 located Northwestern Pacific Ocean east of Japan (Off Joban; 37°00′ N, 142°40′ E) is significantly different from that at ST-2 of the Northern Philippine Sea south of Japan (Nankai Trough; 33°07′ N, 139°59′ E), suggesting that these waters were separated by a topographic barrier, the Izu-Ogasawara Ridge. Taking ³He/⁴He data of the Geosecs expeditions in the western North Pacific, an extensive plume of 15% excess ³He relative to the air may be traced at ST-1 over 12,000 kilometers to the northwest of the East Pacific Rise where the mantle helium may originate. The 20% excess found at ST-2 may be attributable to the additional source of the subduction-type mantle helium in the Okinawa Trough. A 15% excess of ³He has also been discovered at a depth of about 1000∼1500 m at ST-3 adjacent to Miyakejima Island (33°57′ N, 139°22′ E) and ST-4 of Sagami Bay (35°00′ N, 139°22′ E). It is confirmed that mid-depth all over the western North Pacific water is affected by the mantle helium with a high ³He/⁴He ratio. This revised version was published online in July 2006 with corrections to the Cover Date.     

Characteristics of the ratio of dissolved cadmium to phosphate in subtropical coastal waters of Ishigaki Island,Okinawa, Japan

The ratio of dissolved cadmium (Cd) to phosphate (PO₄) in the subtropical coastal area of Ishigaki Island, Okinawa, Japan, was investigated. Twenty vertical seawater samplings were carried out once a month from May 2008 to January 2010. In order to examine how the Cd/PO₄ ratio in seawater varies with the oceanographic conditions (i.e., the water temperature–salinity characteristics), the water masses at the study sites were classified into two types: group 1 with a water temperature of >25°C and a salinity range of 34.0–34.4, and group 2 with a water temperature of <25°C and a salinity of >34.4. A different phytoplankton assemblage was observed in each water mass defined. Different Cd/PO₄ ratios were obtained for the two water mass types, due to the differences between the types in terms of the environmental conditions such as the water temperature–salinity (T–S) characteristics and phytoplankton assemblages, as well as possible variations in the concentrations of dissolved iron, zinc, manganese, and CO₂ in seawater in each water mass.     

Multidisciplinary investigations exploring indicators of gas hydrate occurrence in the Krishna–Godavari Basin offshore, east coast of India

We report some main results of multidisciplinary investigations carried out within the framework of the Indian National Gas Hydrate Program in 2002–2003 in the Krishna–Godavari Basin offshore sector, east coast of India, to explore indicators of likely gas hydrate occurrence suggested by preliminary multi-channel seismic reflection data and estimates of gas hydrate stability zone thickness. Swath bathymetry data reveal new evidence of three distinct geomorphic units representing (1) a delta front incised by several narrow valleys and mass flows, (2) a deep fan in the east and (3) a WNW–ESE-trending sedimentary ridge in the south. Deep-tow digital side-scan sonar, multi-frequency chirp sonar, and sub-bottom profiler records indicate several surface and subsurface gas-escape features with a highly resolved stratification within the upper 50 m sedimentary strata. Multi-channel seismic reflection data show the presence of bottom simulating reflections of continuous to discrete character. Textural analyses of 76 gravity cores indicate that the sediments are mostly silty clay. Geochemical analyses reveal decreasing downcore pore water sulphate (SO₄ ²⁻) concentrations (28.7 to <4 mM), increasing downcore methane (CH₄) concentrations (0–20 nM) and relatively high total organic carbon contents (1–2.5%), and microbial analyses a high abundance of microbes in top core sediments and a low abundance of sulphate-reducing bacteria in bottom core sediments. Methane-derived authigenic carbonates were identified in some cores. Combined with evidence of gas-escape features in association with bottom simulating reflections, the findings strongly suggest that the physicochemical conditions prevailing in the study area are highly conducive to methane generation and gas hydrate occurrence. Deep drilling from aboard the JOIDES Resolution during 2006 has indeed confirmed the presence of gas hydrate in the Krishna–Godavari Basin offshore.     

Helium and oxygen isotope analyses of hydrothermal fluids from the East Pacific Rise between 17°S and 19°S

 Hydrothermal vent fields south of the Garret Fracture zone were sampled for the isotope composition of helium and oxygen ([¹⁸O]H₂O/[¹⁶OH₂O). The helium isotopes end-member (³He / ⁴He=8.3×R _a and [⁴He]≈1.2–2.4×10^-5 cm³ STP g^-1) is quite similar to other known hydrothermal sites pointing to the homogeneous helium composition of the upper mantle. The δ¹⁸O end-member value (δ¹⁸O≈0.5–0.6‰) confirms previous suggestions from other sites and from isotope modeling, that hydrothermal fluids are slightly enriched in ¹⁸O relative to the ocean as a result of water–rock interactions at high temperature. Received: 11 December 1995/Revision received: 20 December 1996     

Temporal variability and characterization of physicochemical properties in the neritic area of Sagami Bay,Japan

Seasonal and interannual variations in physicochemical properties were investigated in the neritic area of Sagami Bay, Kanagawa, Japan, from December 2000 to December 2005. Physicochemical properties (i.e. temperature, salinity, density, dissolved oxygen and dissolved inorganic nutrient concentration) revealed clear seasonal variations, which were similar to each other during all 5 years. Temperature, salinity and dissolved inorganic nutrients showed rapid, drastic variations within a few days and/or weeks. These variations are related to sea levels, principally due to the shifting effects of the Kuroshio Current axis: they were strongly affected by the Kuroshio Water and other waters, when sea level difference was greater than ca. 35 cm and lower than ca. 15 cm, respectively. Temperature difference (DF _T ) increased with sea level difference, and the difference of salinity and dissolved inorganic nutrients (NH₄ ⁺-N, NO₃ ⁻+NO₂ ⁻-N, NH₄ ⁺+NO₃ ⁻+NO₂ ⁻-N, PO₄ ³⁻-P and SiO₂-Si) increased and decreased with DF _T , respectively. All these correlations are significant. Total dissolved inorganic nitrogen (N), phosphate (P) and silicate (Si) revealed seasonal variations in the ranges of 0.57–16.08, 0.0070–0.91 and 0.22–46.38 μM, respectively. From the regression equations between these elements allowed the following relation to be obtained; Si:N:P = 14.8:13.4:1. Dissolved inorganic nutrients were characterized by Si and/or P deficiency, especially in the upper layer (0–20 m depth) during summer. Single and/or combined elements are discussed on the basis of potential and stoichiometric nutrient limitations, which could restrict phytoplankton (diatom) growth as a limiting factor.     

Coral extension rates in the NW Indian Ocean II: reconstruction of 20th century Indian monsoon strength and rainfall over India

To date, coral proxy studies have proved largely unsuccessful in reconstructing temporal variability of the Indian monsoon system in the NW Indian Ocean. In a recent publication in this journal, Storz and Gischler demonstrated that extension rates in corals can be used to fill this gap. Those authors found a link between decadal and interannual variations of the SW monsoon current velocities and Porites lutea extension rates in the lagoon of Rasdhoo Atoll (4°N/73°W) in the central Maldives for the period 1917–2007. This sister paper shows that this extension-rate record can be used to reconstruct decadal variations of summer monsoon rainfall over peninsular India and adjacent areas. The amount of monsoon rainfall and current velocities during summer are both affected by the strength of the Indian monsoon system. Assessments of coral extension rates and the mean May–September All India Monsoon Rainfall index, a measure of monsoon strength over the Indian subcontinent, revealed a significant spectral coherence within periods of 18–19 and 6–7 years. Almost 92% of variance is shared between both time series in the former band, and 85% in the latter. A correlation between the 5-year running mean extension rate and rainfall records from the Western Ghats (grid of 73–76°E/13–15°N) is significant especially for 1958–2006 (r = −0.82, p < 0.05). These findings imply that coral growth characteristics can serve as a new marine archive to reconstruct past variations of the Indian monsoon system on interannual to decadal timescales.     

Simple coprecipitation method combined with low-background γ-spectrometry: Determination of 7Be, 137Cs, 210Pb, and radium and thorium isotopes in small-volume coastal water samples

Application of a coprecipitation method combined with low-background γ-spectrometry allowed the isolation of ⁷Be, ¹³⁷Cs, ²¹⁰Pb, and radium and thorium isotopes together with BaSO₄, Fe(OH)₃, or AMP/Cs in 18 L of coastal water. This simple method showed high chemical yields of 77–91% (mean, 85%) for ²¹⁰Pb and radium isotopes, 87–92% (mean, 91%) for 7Be and thorium isotopes, and 98% for ¹³⁷Cs. The activity of the nuclides showed good reproducibility with standard deviations of 1–8%, while the deviation of activity in particle-reactive ²¹⁰Pb (32%) and ⁷Be (21%) in filtered water samples was exceptionally large.     

Phytoplankton pigment change as a photoadaptive response to light variation caused by tidal cycle in Ariake Bay,Japan

Underwater light environment and photosynthetic accessory pigments were investigated in Ariake Bay in order to understand how change of the pigments occurs in response to the tidal-induced changes in underwater light conditions. We hypothesize that phytoplankton increases photo-protective pigments and decreases light-harvesting pigments under higher light condition in the mixed layer caused by tidal cycle. Contribution rates of non-phytoplankton particles (a _nph (400–700)) for light attenuation coefficient (K _d ) was highest (32–85%), and those of phytoplankton particles (a _ph (400–700)), dissolved organic matter (a _g (400–700)) and water were 6–32, 6–21 and 5–23%, respectively. Mean K _d was higher during the spring tide (0.55 ± 0.23 m⁻¹) than the neap tide (0.44 ± 0.16 m⁻¹), and the K _d difference was caused by the substances resuspension due to the tidal current. In contrast, ratios of photo-protective pigments (diadinoxanthin and diatoxanthin) per chlorophyll a ((DD+DT)/Chl a) were higher during the neap tide (0.10 ± 0.03 mg mg-Chl a ⁻¹) than the spring tide (0.08 ± 0.03 mg mg-Chl a ⁻¹). And there was significant positive correlation between (DD+DT)/Chl a and mean relative PAR in the mixed layer ($ \overline {I_{mix} } $ \overline {I_{mix} } ). Moreover, there was significant negative correlation between ratios of light-harvesting pigments (fucoxanthin) per Chl a (Fuco/Chl a) and $ \overline {I_{mix} } $ \overline {I_{mix} } . These results suggested that phytoplankton in Ariake Bay increase photo-protective pigments and decrease light-harvesting pigments in the higher light condition of less turbid, shallower mixed layer during neap tide than spring tide.     

Time variability of the total methane content in the atmosphere over the vicinity of St. Petersburg

The results of measuring the methane content in the entire atmospheric thickness over the St. Petersburg region are given for 1991–2007. It is shown that, within this period, the mean annual cycle of the total methane content is characterized by its maximum values in December–January and its minimum values in June–August when the annual-cycle amplitude amounts to ∼3.6%. In this case, the annual variations in the total methane content may differ significantly from the mean annual cycle obtained in some years. A statistically significant linear trend of the total CH₄ content has not been revealed for 1991–2007. The obtained values of the linear-trend index have opposite signs in the winter and summer months (positive for January 0.6 ± 0.2%/year and February 0.4 ± 0.2%/year and negative for July 0.3 ± 0.2%/year and August 0.2 ± 0.1%/year). This fact suggests the tendency for an increase in the amplitude of the annual cycle of the total CH₄ content. The results of a spectral analysis of a series of data on the total CH₄ content show that, for 1991–2007, the following harmonics are pronounced with a confidence of 95%: 12 months (annual harmonic), 32 months (quasi-biennial oscillations), and 55 months (4.5 years), which are also pronounced in the series of meteorological parameters and total ozone content.     

Oxygen Utilization and Organic Carbon Remineralization in the Upper Water Column of the Pacific Ocean

As a part of the JGOFS synthesis and modeling project, researchers have been working to synthesize the WOCE/JGOFS/DOE/NOAA global CO₂ survey data to better understand carbon cycling processes in the oceans. Working with international investigators we have compiled a Pacific Ocean data set with over 35,000 unique samples analyzed for at least two carbon species, oxygen, nutrients, chlorofluorocarbon (CFC) tracers, and hydrographic parameters. We use these data here to estimate in-situ oxygen utilization rates (OUR) and organic carbon remineralization rates within the upper water column of the Pacific Ocean. OURs are derived from the observed apparent oxygen utilization (AOU) and the water age estimates based on CFCs in the upper water and natural radiocarbon in deep waters. The rates are generally highest just below the euphotic zone and decrease with depth to values that are much lower and nearly constant in water deeper than 1200 m. OURs ranged from about 0.02–10 μmol kg⁻¹yr⁻¹ in the upper water masses from about 100–1000 m, and averaged = 0.10 μmol kg⁻¹yr⁻¹ in deep waters below 1200 m. The OUR data can be used to directly estimate organic carbon remineralization rates using the C:O Redfield ratio given in Anderson and Sarmiento (1994). When these rates are integrated we obtain an estimate of 5.3 ± 1 Pg C yr⁻¹ for the remineralization of organic carbon in the upper water column of the Pacific Ocean. This revised version was published online in July 2006 with corrections to the Cover Date.     

Late-Quaternary variations in clay minerals along the SW continental margin of India: evidence of climatic variations

Down-core variations in illite, chlorite, smectite and kaolinite (the major clays) in two ¹⁴C-dated cores collected along the SW continental margin of India show that illite and chlorite have enhanced abundance during 20–17, 12.5, 11–9.5, and 5–4.8 ka b.p., whereas smectite accumulation is higher between 17 and 12.5, and after 9 ka b.p. The climate may have been predominantly arid at 17 (20–17), 12.5, 10.5 (11–9.5), and 4.8 ka b.p. The first three dates correspond to the last glacial maximum, Bolling-Allerod, and Younger Dryas events, respectively. The SW monsoon was variable between 17 and 15 ka b.p., and it was more stable and intense after the Younger Dryas until about 6 ka b.p. Received: 2 December 1999 / Revision accepted: 11 April 2000