An alkenone (
U
K
037
) quasi-annual sea surface temperature
record (A.D. 1440 to 1940) using varved sediments from
the Santa Barbara Basin
Meixun Zhao
a,*, Georey Eglinton
b,c, Gareth Read
b, Arndt Schimmelmann
daDepartment of Earth Sciences, Dartmouth College, Hanover, NH 03755, USA
bBiogeochemistry Centre, Department of Earth Sciences, University of Bristol, Bristol BS8 1RJ, UK cMarine Chemistry and Geochemistry, Woods Hole Oceanographic Institution, MA 02543, USA
dDepartment of Geological Sciences, Indiana University, Bloomington, IN 47405-1403, USA
Received 9 February 1999; accepted 13 March 2000 (Returned to author for revision 13 June 1999)
Abstract
Analyses for C37alkenones (U K0
37) and total organic carbon (TOC) in a varved core from the depocenter of the Santa Barbara Basin (3413.50N, 120010W, 590 m water depth) have been made on an approximately 1±2 year basis from A.D. 1440 to 1940. TheUK0
37-based sea surface temperature (SST) record oscillates around 15.5C with an amplitude of less than 3C at approximately interannual, decadal, and centennial frequencies. Historical El NinÄo/Southern Oscilla-tion (ENSO) events are matched only partially by warmerUK0
37SSTs. Three records: our SST, a published Paci®c coral
d18O, and a published tree-ring temperature anomaly from the California valleys, show no evidence of colder tem-peratures during the Little Ice Age (ca. 1500 to 1840 A.D.). Signi®cantly, there is a strong centennial-scale cyclicity in theUK0
37 SST record which approaches 1C in amplitude. TOC burial and alkenone ¯uxes are largely controlled by sedimentation rate as expressed in varve thickness.#2000 Published by Elsevier Science Ltd. All rights reserved.
Keywords:Molecular stratigraphy; Alkenones; SST;UK0
37; Santa Barbara Basin; Varved sediments; Little Ice Age; ENSO; Centennial
scale oscillation
1. Introduction
Continuous, high resolution climate records of the past few hundred years are needed for predictive mod-eling, but few have been reported (e.g. Martinson et al., 1995; Overpeck et al., 1997; Mann et al., 1998). Inter-pretation of the processes responsible for 20th century warming will bene®t from the availability of more records on a decadal to centennial scale. This paper reports a quasi-annual resolution record for the last 500 years ofUK0
37sea surface temperature (SST) for the Santa Barbara Basin (SBB), o southern California.
The Little Ice Age (LIA) is commonly known as a centennial-scale cold climate event, mostly documented in the Northern Hemisphere, that started ca. 1500 A.D. and terminated just before the acceleration of anthro-pogenic activities (Grove, 1988). Although terrestrial records of LIA climatic variability have been well-documented (Grove, 1988; Bradley and Jones, 1993; Mann et al., 1998), the extent of oceanic changes during the same period is less well known (Dunbar et al., 1994; Keigwin, 1996). A key parameter in any paleo-reconstruction relating climate and oceanography is SST, since heat transfer between ocean and atmosphere is fundamental to the generation of climate. The El NinÄo/Southern Oscillation (ENSO) phenomenon (Phi-lander, 1990; Diaz and Markgraf, 1992) in the Paci®c has dramatically highlighted the importance of changes in SST, since it is clear that ENSO eects and
0146-6380/00/$ - see front matter#2000 Published by Elsevier Science Ltd. All rights reserved.
P I I : S 0 1 4 6 - 6 3 8 0 ( 0 0 ) 0 0 0 3 4 - 6
www.elsevier.nl/locate/orggeochem
* Corresponding author. Tel.: 603-646-2150; fax: +1-603-646-3922.
telecommunications extend from southeast Asia beyond the Americas to Europe, Africa, India and worldwide (Kiladis and Diaz, 1989; Philander, 1990; Charles et al., 1997; Neelin and Latif, 1998).
The surface water circulation is, to a ®rst approxima-tion, westwards in the east Paci®c both north and south of the equator (Fig. 1). The ENSO events occur irregu-larly over periods of 2±11 years, and are characterized by a general warming of the tropical east Paci®c waters of roughly one year duration. Thus, the waters o Peru during strong/very strong El NinÄo events, such as those of 1982 and 1997, show strong positive SST anomalies
of up to 5C as compared with the non-El NinÄo years. Further to the west and north in the vicinity of the Galapagos Islands, the warmer equatorial waters exhibited somewhat smaller SST anomalies of ca. 3C. In the Northern hemisphere along the California coast, the SST anomalies are even smaller, for example 1±2C for the 1982 and 1997/8 ENSO (Rasmusson and Hall, 1983; Neelin and Latif, 1998). Much socio-economic interest attaches to the ENSO climatic pattern. Thus, unusual weather events have been documented inter-mittently for the east Paci®c rim since the initial Spanish occupation (Quinn, 1992). However, more continuous
proxy records are available in the sediments o the California coast, where the varved deposits in the SBB provide a particularly valuable source of information (Schimmelmann and Lange, 1996).
1.1. The Santa Barbara Basin
The SBB lies within a highly productive coastal region of the Paci®c Ocean (Soutar and Crill, 1977; Schimmel-mann and Tegner, 1991). Its high productivity con-tributes to high sedimentation rates (3±5 mm yearÿ1in recent years) and high organic carbon content (2±5%). Oxygen depletion in the bottom water of the depocenter (Sholkovitz and Gieskes, 1971) inhibits bioturbation by macrobenthos and allows for the preservation of milli-meter-scale laminae (Soutar and Crill, 1977; Reimers et al., 1990; Thunell et al., 1995). Past variations in paleo-ceanographic parameters, such as SST, upwelling strength, productivity and intensities of the California Current, could have been registered on an almost annual basis in the sediment record.
SST in the SBB and the Southern California Border-land region in general are in¯uenced by: (1) the colder and relatively nutrient-rich, southward-¯owing Cali-fornia Current (Fig. 1), which is strong during spring and summer; (2) the warm and nutrient-poor north-ward-¯owing Davidson Current (Fig. 1), which is strong during winter (Hickey, 1979); (3) the seasonal (mostly spring and summer) upwelling of the cold and nutrient-rich Paci®c Intermediate Waters, which is driven by the northerly trade winds caused by the North Paci®c high (Thunell and Sautter, 1992). Recent shipboard, buoy and satellite data (e.g. Winant and Hendershott, 1996) show that cold water from the North mixes extensively with the warm coastal waters, creating cyclonic ¯ow within the Basin and numerous eddies. Due to the complex and variable interaction of the currents and the seasonal upwelling, productivity in the SBB surface waters shows a large seasonal variation while SST varies from 10 to 18C (cf. Kennedy and Brassell, 1992b). Instrumental SSTs recorded in the SBB by the CalCOFI studies do not show a well-de®ned seasonal pattern, although satellite temperature estimates and sea level heights reveal a more consistent seasonal cycle in the SBB (Harms and Winant, 1998). During typical ENSO years, the California coastal region experiences wea-kened upwelling, accompanied by a deeper thermocline, while the warm Davidson current extends northwards and the ¯ow of the cold California current slows, resulting in a generally warmer SST and lower pro-ductivity (Hayward et al., 1994).
1.2. Molecular stratigraphy-U37K 0
SST
Recent developments in biomarker geochemistry have included molecular tools suitable for high resolution
climatic reconstruction (Brassell, 1993). The C37 long-chain unsaturated ketones (alkenones) are lipid bio-markers whose biological source is known only from a narrow range of marine microalgae belonging to the class of Haptophyceae (Prymnesiophyceae), of which Emiliania huxleyi is an important contemporary mem-ber (Volkman et al., 1980a,b; Marlowe et al., 1990). The
UK0
37 index [C37:2/(C37:2+C37:3)], based on the ratio of the C37:2and C37:3alkenones, has been widely used to reconstruct past SST (for example, Brassell et al., 1986; Prahl and Wakeham, 1987; Eglinton et al., 1992; Prahl et al., 1995; Rosell-Mele et al., 1995; Zhao et al., 1995a; MuÈller et al., 1997; Villanueva et al., 1998) and the abundances and accumulation rates of the alkenones have been used to estimate haptophyte productivity (e.g. Prahl et al., 1993; Villanueva et al., 1995). Small sample requirements and the ability to achieve high sample throughput with automation of analytical procedures make SST molecular stratigraphy especially suitable for high resolution paleoclimate applications. This techni-que has been applied for the estimation of modern and last glacial surface waters of the California Current sys-tem (Doose et al., 1997), and for the reconstruction of SBB SST by Kennedy and Brassell (1992a,b) and by Herbert et al. (1998), with an approximate annual reso-lution record from ca. 1915 to 1990. Additionally, lower resolution SBB UK0
37 data for the Holocene have been published by Herbert et al. (1995) and Hinrichs et al. (1997) using ODP Hole 893 samples.
The present paper extends and further develops our preliminary report (Zhao et al., 1995b) of a UK0
37 SST record for a site in the depocenter of the SBB, back into the 15th century. We present 1±2 year resolution mole-cular records ofUK0
37 SST and alkenone content for the period A.D. 1440 to 1940 and compare them with par-allel data of TOC and varve thickness for the same sample set. We have evaluated the variations of theUK0
37 SST record in the context of past ENSO in¯uences and of the oceanic response to the LIA in the southern California Borderland. The TOC and alkenone ¯uxes are evaluated as indicators of paleoproductivity/pre-servation.
2. Methods
2.1. Samples
1990). The sediment in the acrylic tubes was then extru-ded and sectioned with a stainless spatula, layer by layer (1±5 mm). In most cases, distinct dark/light dierences marked the boundaries between layers, which were then followed closely to minimize contamination of adjacent layers. One varve normally comprised a dark, clay-rich winter layer together with the adjoining light, carbo-nate/opal-rich biogenic layer. This sampling procedure ensured almost annual resolution. Each slice was then documented as containing one to a few varves by com-parison with the pre-scanned sediment X-radiograph. Each layer was then freeze-dried and thoroughly mixed for sub-sampling prior to geochemical analysis. Freeze-drying was done immediately after sectioning of samples to minimize the oxidation of sul®des. The samples were then stored under nitrogen until the sulfur study was completed (Schimmelmann and Kastner, 1993). The samples were later stored dry under air at room tem-perature (20±25C), prior to analyses for TOC and alkenones. The detailed procedure for TOC analysis is given in Schimmelmann and Tegner (1991). The cores were dated using varve-counts and correlation of varve thickness versus dendrochronological data (Schimmel-mann et al., 1990; Schimmel(Schimmel-mann and Lange, 1996). The samples used have been set against this scale as A.D. 1440 to 1940. The samples used in this study did not include thin gray ``turbidites'' or recognizable ¯ood deposits (Schimmelmann and Lange, 1996). These sedi-ment intervals were removed from the stratigraphic fra-mework since they are regarded as eectively instantaneous deposition events. Dating and calcula-tions across the bioturbatedMacomainterval (ca. 1840) (Schimmelmann et al., 1992) assumed a constant sedi-mentation rate. Samples were not obtainable from our core after the 1941 varve due to heavy prior sampling.
2.2. Alkenones
Aliquots of 0.2 to 0.3 g of sediment (dry weight) were used for alkenone analyses. Sulfur was removed from the samples by adding ca. 0.2 g of `activated' copper granules to the sediment before extraction. The copper had been activated with 0.5±2.0 M HCl for 1 h, rinsed with double distilled water, methanol (MeOH) and dichloromethane (DCM) before drying with N2. After adding an internal standard (C36n-alkane), the samples were extracted (three times) with ultra-clean solvents (CH2Cl2/CH3OH, 3/1 by vol.). Extractions were carried out using a semi-automated BenchMate robot work-station (Zymark Corp.), which handled the weighing of sediments, addition of internal standard, and the addi-tion and transfer of liquid phases. The total extracts were puri®ed by solid phase extraction using silica gel cartridges (500 mg Bond Elute, Varian), using the BenchMate. Solid phase extraction eliminates polar and large molecules, to avoid the injection of total extracts
from the very organic-rich SBB samples which cause the deterioration of the resolution of the gas chromato-graphy (GC) columns. The cleaned extracts were deri-vatized with 40 ml of N,O-bis(trimethylsilyl)tri¯uoro acetamide (BSTFA) and alkenone contents were mea-sured using a Varian 3400 GC ®tted with an auto-sampler and a 50 m capillary column (Chrompack CP-Sil 5, 0.32 mm i.d., 0.25 mm ®lm, H2 carrier gas). GC data were acquired and processed using VG Minichrom software. Abundances of individual compounds were calculated by comparing their peak areas with those of the internal standard (C36n-alkane). SST was calculated from theUK0
37 index using the Prahl et al. (1988) calibra-tion: SST (C)=(UK0
37ÿ0.039)/0.034. Alkenone contents (per dry weight) are expressed as the total of C37:2and C37:3. A few of the sediment samples lacked sucient amounts of sediment for alkenone analysis. The analy-tical error is about0.3C in SST. In the SBB extracts, an unknown compound (possibly the C36methyl ester) often co-elutes with, or elutes very close to, the C37:3 alkenone peak, under some chromatographic condi-tions. Since integration of this peak with the C37:3peak is undesirable (inclusion results in a slightly lower SST estimate), we took special care to make sure that these peaks were completely separated and that the integra-tion of the alkenone peaks was accurate. The alkenone data will be submitted to the World Data Center-A for Paleoclimatology in Boulder, CO.
2.3. Burial ¯ux calculation
The burial ¯ux of the alkenones (ng cmÿ2yearÿ1) was calculated from contents in dry sediment (corrected for salt content) and from varve thickness data. The same procedure was also used to calculate TOC burial ¯ux (Schimmelmann and Tegner, 1991). Since each sample can contain sediment deposited over several months to over 3 years, single point ¯ux calculations can be mis-leading. This problem has been minimized by smoothing the data over ®ve samples, where stated.
3. Results
Varve thickness (Fig. 2A; Schimmelmann and Lange, 1996) varied from 1 to 5 mm per year. The thickness varied between 1 to 3 mm from A.D. 1440 to 1640 and then increased to 5 mm and stayed generally high until 1730, but with major high frequency oscillations. The next 120 years are noted for thinner varves. Varve thickness was relatively large (3 to 4 mm) from 1860 to 1940, but also with major variations.
near 1475, 1600, 1760, 1840, 1870 and 1910 A.D. These TOC oscillations seem to be characterized by higher values followed by a sudden drop and a gradual return to reach the next maximum. From the mid-1700s to the
mid-1800s, the oscillations decrease in frequency and amplitude.
Alkenone content (Fig. 2D) varied by a factor of 7 over the 500-year period, with a maximum of about 8mg gÿ1 and a minimum of 1 mg gÿ1. The record reveals many oscillations on interannual, decadal and cen-tennial timescales.
UK0
37 SST (Fig. 2F) varied by 3C (14±17C) over the 500-year period from A.D. 1440 to 1940. Of the 499 points, 83% fall in the range of 15 to 16C, though numerous small SST oscillations on the order of 1±2C occur over 3±4-year intervals. SST is generally lower around the turns of the centuries, while the higher SSTs are approximately midway.
The TOC burial ¯ux (Fig. 2C) and the alkenone bur-ial ¯ux (Fig. 2E) are similar in general trend, except for the interval from 1840 to 1860. They resemble closely the plot of varve thickness (Fig. 2A), unlike the original content data (Fig. 2B and D). This underlines the dom-inance of the varve thickness in the burial ¯ux calcula-tion, at least qualitatively.
4. Discussion
4.1. U37K 0
SST in the Santa Barbara Basin
The UK0
37 SST technique has been used sucessfully in the Eastern Paci®c region in a number of studies. How-ever, problems with the interpretation of the signi®cance of small dierences in the UK0
37 values include the obvious ones of seasonality (the timing of the cocco-lithphore bloom), the depth of maximum production, and the nutrient availability. A recent paper by Epstein et al. (1998) based on isothermal culture experiments with a single clone has shown that extreme nutrient variation can aectUK0
37 SST by several degrees. Very recently, Herbert et al. (1998) have published a substantive evaluation of the method using a latitudinal transect (23 to 40N) along the Californian Margin, using core top samples, which included some from the SBB. They have convincingly demonstrated that UK0
37 data derived by the standard GC procedures can be used with con®dence for SST's in the range of 12 to 23C in this region. Using the Prahl et al. (1988) cali-bration, theirUK0
37 SST estimates correlate linearly with modern mean annual SST of the mixed layer (less that 30 m depth). The estimatedUK0
37 SST for the SBB (ca. 15C) is in agreement with that compiled by the World Ocean Atlas for the SBB (15.3C; Levitus, 1994). Thus,
UK0
37 measurements on individual varves from laminated sediments may be expected to give an average tempera-ture for each quasi annual interval. Indeed, Kennedy and Brassell (1992a,b) had already presentedUK0
37 SST results for a core from the SBB for the period 1915 to 1988, and a second record from almost the same site
Fig. 2. Geochemical time-series of varved sediments from the Santa Barbara Basin with ca. 1±2 year resolution, A.D. 1440 to 1940. (A) Annual varve thickness (mm, Schimmelmann and Lange, 1996); (B) TOC content (% of dry, salt-free weight, Schimmelmann and Tegner, 1991); (C) a 5-point moving
aver-age of TOC burial ¯ux (mgC cmÿ2yearÿ1); (D) the alkenone
content (mg gÿ1 dry, salt-free weight); (E) a 5-point moving
average of alkenone burial ¯ux (ng cmÿ2yearÿ1); (F)UK0
37 sea
surface temperature (SST,C). The thick line in Fig. 2F is a
50-point (ca. 50-year) moving average of SST to show the cen-tennial scale oscillations. Data gaps result either from the pre-sence of turbidites and ¯ood deposits, or are due to insucient sample for individual measurements. Dashed intervals in ¯ux calculations bridge gaps that result from the exclusion of gray turbidites and ¯ood deposits that do not constitute regular varves. These layers were not included in the burial ¯ux calcu-lation, because they contained unusual amounts of terrigenous materials in ¯ood-related pulses not relevant to the SBB ocea-nography. F, ¯ood layer of 1605, T, turbidite of 1738, M,
covering a similar time period has now been described in the paper by Herbert et al. (1998) TheUK0
37 SST data in the present paper are also derived from a core from this site, but for a much longer time span (A.D. 1440 to 1940). Comparing the data for the 20th century in each of the three records (Table 1 and Fig. 3) gives us some measure of the reliabilility ofUK0
37 SST molecular strati-graphy based on varve counting: all these studies used the same equation: SST (C)=(UK0
37ÿ0.039)/0.034. In Fig. 3C, there is only a short time interval (A.D. 1919 to 1941) where all three sets ofUK0
37 SST values are available. It is evident that there are small dierences in average SST between the three records, with ours being
almost 0.5C higher than those of Herbert et al. (1998) and almost 1C higher than those of Kennedy and Brassell (1992a,b). These dierences may have their ori-gin in analytical methodology and sampling site loca-tion. The records despite being quite dierent in detail, do show some general similarities. Small dierences may be caused by the diculties inherent in varve chronol-ogy, such as the dissection and counting procedures. The accuracy of varve dating is 2 years above 1840 (the Macoma Event; Schimmelmann et al., 1992), but as much as 5 years below 1840 (Schimmelmann and Lange, 1996). The three curves display a warm max-imum around 1941, which is a well-known ENSO event.
Table 1
20th centuryUK0
37SST core records for the Santa Barbara Basina
Cores Location Age range
(A.D.)
UK0
37 SSTb
(C)
C37alkenones
(mg gÿ1)
References
SBB 88-9, 10a, 10b 34140N, 120010W 1915±1988 14.3 6 Kennedy and Brassell, 1992a,b
SBBX-1 34130N 120030W 1918±1995 14.8 4.5 Herbert et al., 1998
SABA 87-1, 87-2, 88-1 3413.60N 120010W 1900±1941 15.3 4 Zhao et al., 1995b and this paper
a Water depth is 590 m for all cores. SST is calculated using the following equation: SST (C)=(UK0
37ÿ0.039)/0.034.
b These are the average SSTs for the common period: 1919±1941.
Fig. 3. Comparison of sea surface temperature records. (A) instrumental SST record from Puerto Chicama, Peru, A.D. 1936±1981 (Shen et al., 1992); (B) the instrumental SST anomalies from the western coast of USA, A.D. 1900±1985 (Schimmelmann et al., 1995).
(C) the Santa Barbara BasinUK0
37 SST (solid line: this paper using SST scale on the left, A.D. 1900±1941; dotted line: data from
Kennedy and Brassell (1992a,b) using SST scale on the right, A.D. 1914±1982, dashed line: data from Herbert et al. (1998), using SST
scale on the right, A.D. 1919±1985); the two SST scales are used to show the rough parallelism of theUK0
37 SST records, despite the
systematic oset of 1C, which may have its origin in analytical methodology and sampling site location. Documented strong and very
The situation appears to be similar for the period 1942 to 1985 where only the Kennedy and Brassell (1992a) and Herbert et al. (1998) records can be compared. Once again, the Herbert et al. data are systematically warmer than the Kennedy and Brassell data by 0.3 to 0.5C.
The SBBUK0
37 SST records for the 20th century are of special interest in terms of historical ENSO events. Although the ENSO events are characterized by a gen-eral warming of the tropical east Paci®c waters, not every event has been registered in the SST o the Cali-fornia Borderland. This can be seen by comparing the instrumental SST record of 1936 to 1982 from Puerto Chicama o Peru (Fig. 3A), which has been demon-strated to record ENSO accurately (Shen et al., 1992), with the West Coast instrumental SST anomalies (Bar-nett, 1985; Schimmelmann et al., 1995) (Fig. 3B). Although the two instrumental SST records show some common features, major dierences do exist. For example, the 1965 and the 1972 strong ENSO events were both recorded with a SST maximum at Puerto Chicama, but with SST minima in the West Coast SST record. An increase in upwelling or the strengthening of the cold California Current could have accounted for the lower West Coast SST. Recent studies of the circu-lation patterns of the Santa Barbara Channel show that this region is in¯uenced by cold cyclonic eddies that spin o the California Current and are not necessarily asso-ciated with ENSO or La NinÄa events. The AVHRR satellite images clearly show the variable, complex and especially changing nature of SST maps for this region (Winant and Hendershott, 1996; Harms and Winant, 1998). Nevertheless, the availability of the varved sedi-mentary record renders this a prime site for the study of annual to decadal climate variability.
Kennedy and Brassell (1992a,b) concluded that there was a reasonable correlation of major, historically recorded 20th century ENSO events with increasedUK0
37 SST for the relevant varves in the SBB sedimentary record. However, while the strong 1972 ENSO event matches with a UK0
37 SST maximum in their record for the SBB (Fig. 3C), the West Coast SST anomalies (Fig. 3B) and the CalCOFI SST records show a minimum (Kennedy and Brassell, 1992b; Schimmelmann et al., 1995). Again, in the record of Herbert et al. (1998), there is no major change ofUK0
37 SST around 1972. In making their comparison with instrumental SST (Scripps Pier data set), Herbert et al. (1998) remarked that after 1965 the two records correlate poorly, except during the strong ENSO event of 1982±1983.
The Kennedy and Brassell (1992a) paleo-SST recon-struction prompted the present study ofUK0
37SST record for the SBB back in time beyond the LIA. We have compared our SST record with Quinn's (1992) ENSO events for the period of 1840 to 1920, because con-®dence in varve chronology is higher above 1840. For
this period, only ®ve of the 12 strong/very strong historical ENSO events coincide with signi®cant temperature increases within1 year. The other ENSO events have no apparent SST responses. There are several possible reasons for these discrepancies. Firstly, a small mis-match between an ENSO event and a SST maximum can simply be due to errors in the chronologies of the records. Secondly, the historical ENSO record is based mostly on South American historical data (Quinn, 1992). The warming eect of tropical waters often takes several months to 1 year to become apparent in the SBB. Thirdly, not every ENSO event would be expected to cause a SST increase along the west coast, and indeed, the strong 1965±1966 and 1972±1973 El NinÄo events did not (Schimmelmann et al., 1995). Finally, the present result is from one site only. We do not know how dependent the SST record is on the precise location of the site, bearing in mind the strong SST gradient within the basin (Harms and Winant, 1998). In the Santa Monica Basin, two cores 15 km apart aorded
UK0
37 records which diered by as much as 4C. Gong and Hollander (1999) ascribed the SST dierences to selective oxidative degradation of the alkenone pair in the oxic bottom sediments compared with the anoxic sediments at the basin depocenter. However, the spatial complexity and variability of the thermal structure of the waters in this region may also play an important role.
There is a more general, biologically- and ecologi-cally-related problem inherent in attempting to match ENSO events with alkenone SST records. This is the uncertainty in the precise meaning of theUK0
37 SST esti-mates in relation to the timing and location in the water column of the alkenone production. Thus, in upwelling areas, coccolithophorids do not necessarily bloom at the time of maximum upwelling. Coccolithophorid produc-tion usually does not coincide with that of the diatoms, which are typically the main source of marine organic matter. In fact, coccolithophorids may bloom just after the main bloom crashes (Mitchell-Innes and Winter, 1987). The ecological situation will re¯ect competition for nutrients, as well as other factors. But, in any case the temperature recorded by the alkenone production will not necessarily be that of the time of the main upwelling. This biological aspect to the use ofUK0
periods and found high alkenone ¯uxes in both fall and spring. The UK0
37 SST values correlated well with the seasonal temperatures of the major production depth (50 m for spring and 30 m for fall). However, Ohkouchi et al. (1999) have suggested that in the central Paci®c Ocean, the depth of alkenone production is mainly controlled by the response of the haptophyte producers to the nutrient supply (e.g. nitrate) from deeper waters. Hence, at these sites, the UK0
37 SST measured in the underlying sediments may re¯ect the temperature of thermocline waters signi®cantly deeper than the surface mixed layer. We conclude that alkenone SST estimates are not readily related to a single parameter, such as the water temperature at a speci®c depth and season. In eect, they will re¯ect some time and spatial averaging of the complex pattern of ocean surface temperatures created by insolation, upwelling cells, current move-ments and wind strength. This lack of a facile one-to-one relationship must apply to other proxies as well, but medium to longer term trends (10 to 100 years) can still be useful climate records despite the short-term noise (Weaver et al., 1999).
The above comparisons show that, although theUK0
37 SST record is of some value in ENSO reconstruction, SST maxima are not conclusive measures of ENSO events. Indeed, correlation between UK0
37 SST maxima and individual paleo-ENSO records for much older sediments is even more problematic, due to the increased uncertainties in the true dates of the varves and of the historically-reconstructed ENSO events. Indeed, McCarey et al. (1990) remarked in their study of a core o the Peru margin, that while the UK0
37 SST signals of individual ENSO events were substantially attenuated in the sediments, a cluster ofUK0
37 SST max-ima would indicate a period of frequent and intense ENSO activity, e.g. the warm period of 1865±1885, which is also seen in our record.
4.2. U37K 0
SST record during the Little Ice Age
Recent paleoclimate reconstructions indicate that the Little Ice Age was not global (Bradley and Jones, 1992a, 1993; Mann et al., 1998). Most regions experienced colder periods, but there is geographic variability (Bradley and Jones, 1993). The late 1500s and the early 1600s were cold in both North America and Europe, but this interval was very warm in China and Asia. The most profound cooling period in China occurred in the mid-1600s (1650±1670), but was coeval with one of the warmest decades in North America. Although the unsmoothed SBB UK0
37 SST for individual varves has varied by as much as 3C during the last 500 years (Fig. 2F), there is no recognizable ``cold period'' associated with the Little Ice Age, the average UK0
37 SST for each century being rather consistant (ca. 15.5C). Thus, the dierence in average SST between the coldest century
(18th) and the warmest century (19th) is only 0.2C. This region's oceanic response to global LIA cooling appears to have been minimal. Rather, it seems quite clear from the 50-year averaged SST record (Fig. 2F) that the dominant features are centennial scale oscilla-tions involving decadal scale cooling and warming. On the decadal timescale, the warmest decade (1651±1660) is 1.6C warmer than the coldest one (1899±1908) (Table 2). As far as the sub-decadal SST variability is con-cerned, the relatively warmer 18th century is character-ized by rather stable SST, while the 16th, 17th and the 19th centuries exhibit largerUK0
37 SST oscillations. We observe similar long-term trends in the Santa Barbara BasinUK0
37 SST (Fig. 4B) and in the annual air temperature anomalies from the California Valleys (Fritts, 1991) (Fig. 4C); for example, the major warm-ings around 1655 and 1865. These similar trends are suggestive of forcing mechanisms involving ocean-atmosphere interaction. However, there are signi®cant dierences in both the timing and amplitude of the temperature changes in the two records. For example, the major cooling in theUK0
37 SST starting from ca. 1870 reached a temperature minimum around 1903, while the tree-ring air temperature minimum was around 1913. The age uncertainty of 2±5 years for the SBB samples (Schimmelmann and Lange, 1996) may partially explain this time oset, while dierent response times of the systems to the internal or external forcing may also be important (Stuiver and Braziunas, 1993). Although ocean and atmospheric systems are coupled, it is not unusual for the ocean to witness warming or cooling events a few years after the atmospheric expression (Crowley and Kim, 1996). In addition, the SBB SST is strongly in¯uenced by upwelling events and these are not necessarily recorded in a tree-ring record. Also, volcanic eruptions often cause short-term cooling of the atmosphere, whereas the ocean's thermal inertia limits short-term SST excursions (Portman and Gutzler, 1996).
The UK0
37 SST record for the SBB is also compared with the sea surface temperature proxy record recon-structed using coral d18O values from Urvina Bay, Galapagos Islands in the Eastern Equatorial Paci®c (Fig. 1 and 4A; Dunbar et al., 1994). Corald18O values during the Holocene mainly record SST, with more negative d18O values indicating warmer SST. Again, there is no clear evidence of LIA cooling in the coral record, but there is some correlation between the long term trends in the SBB UK0
The spectral output (10±1000 years) forUK0
37 SST in the SBB, in terms of decadal and centennial cities, is shown in Fig. 5. For the subdecadal periodi-cities, the major peaks are 3±4 and 6 years (Zhao et al.,
1995b). These peaks are within the frequency band of the ENSO variability (Anderson, 1992; Dunbar et al., 1994). On the decadal scale, periodicities of 12±13 and around 20 are signi®cant. For the centennial scale, SST has a strong peak in the region of 100 year periodicity.
Several systems are widely-held to be candidates for climate forcing mechanisms, including the Earth's internal dynamics, coupled atmosphere-ocean interac-tions, and astronomical factors such as solar activity (Rind and Overpeck, 1993; Wallace, 1995; Schimmel-mann et al., 1998). Several authors (Anderson, 1992; Friis-Christensen and Lassen, 1991; Crowley and Kim, 1993, 1996; Lean et al., 1995) have proposed that sun-spot activity is the most likely driving force for terres-trial climate cycles at the decadal level (e.g. 11 years), although the forcing mechanism remains obscure. Indeed, the Maunder minimum (A.D. 1645±1705) in sunspot numbers (Foukal, 1990), which has been pro-posed as one of the causes for the LIA in Europe and
Table 2
The eight warmest and coldest 10-yearUK0
37 SST averages from
1440±1940 in the Santa Barbara Basin
Warmest period Average
SST (C)
Coldest period Average
SST (C)
1651±1660 16.3 1899±1908 14.7
1511±1520 16.1 1585±1594 15.1
1864±1873 16.0 1887±1896 15.1
1771±1780 15.9 1821±1830 15.2
1759±1768 15.9 1455±1464 15.2
1739±1748 15.9 1604±1613 15.3
1491±1500 15.8 1801±1810 15.3
1810±1819 15.7 1682±1691 15.3
Fig. 4. Comparison of three temperature proxy records A.D. 1600 to 1940. (A) corald18O (%, PDB) record from Urvina Bay in the
Eastern tropical Paci®c Ocean (Dunbar et al., 1994). More negative values correspond to warmer SST; (B)UK0
37 sea surface
tempera-ture (C) from the SBB. (C) annual atmospheric temperature anomaly expressed as departure from the 1901±1970 mean instrument
North America, is almost coeval with one of the warm-est 50 years in our SBB SST record (A.D. 1631±1680; Fig. 2F). However, other workers (Latif and Barnett, 1994; Schlesinger and Ramankutty, 1994; Chang et al., 1997) have claimed that there is a decadal variation in tropical ocean-atmosphere interactions that could be responsible for SST oscillations in the tropical Atlantic and Paci®c. Large volcanic eruptions have also been recognized as a major cause of short-term atmospheric and continental cooling (Hammer et al., 1980; Bradley and Jones, 1992b; Bria et al., 1998; de Silva and Zie-linski, 1998). At this stage, with the results from one long-term, quasi-annual SST record, we cannot dis-criminate between the eects of local variability in the physical oceanography of the Santa Barbara Channel (Harms and Winant, 1998) and those of the more regional to global processes outlined above. This situa-tion will be improved by more paleo-SST records from the SBB.
4.3. Preservation records of TOC and alkenones
The sedimentary content of TOC and of individual organic compounds is in¯uenced by a variety of bio-geochemical factors, notably primary productivity (Eppley and Peterson, 1979), water column reminer-alization, sedimentation rate and bottom-water oxygen concentration (MuÈller and Suess, 1979; Pedersen and Calvert, 1990; Hedges and Keil, 1995; Hartnett et al., 1998). Nevertheless, the contents and burial ¯uxes (mass accumulation rates) of TOC have been used as measures of total marine productivity (MuÈller and Suess, 1979; Stein et al., 1989; Sarnthein et al., 1992) and those of alkenones as proxies for haptophyte productivity (Prahl et al., 1993; Villanueva et al., 1995). The SBB alkenone content pro®le (Fig. 2D) shows large variations from 1 to 8 mg gÿ1 (dry sediment), similar to the early 20th century record of Kennedy and Brassell (1992a,b). TOC (Fig. 2B) and alkenone contents show some similarity in general trends over the last 500 years. The burial ¯ux data for TOC (Fig. 2C) and for alkenones (Fig. 2E) correlate better with each other, but not well with the content pro®les. This is understandable, since they are calculated using varve thickness, which can account for a major part of the variances of both data sets.
With regard to the relationship between SST and TOC content in an upwelling region, the prevailing hypothesis is that an increase in upwelling, with accom-panying decrease in SST, would cause an increase in surface productivity and hence in sedimentary TOC. A direct visual comparison of the 10 year smoothed records for these two parameters (Fig. 6) indicates some parallelism in long-term trends, together with several major disagreements (e.g. ca. 1650 and 1850 A.D.). However, anX±Yplot ofUK0
37SST versus TOC (Fig. 7B) does not reveal any meaningful relationship. One could
argue that burial ¯ux is a better measure of surface production than content, since it is calculated using annual varve thickness and hence takes account of the eects of sedimentation rate and of terrigenous dilution. If the TOC burial ¯ux is mainly controlled by primary productivity, then it should have a reasonably good inverse correlation with SST, since both are in¯uenced by the upwelling of cold and nutrient-rich waters: stronger upwelling would induce higher TOC burial ¯ux and lower SST. However, comparison of these records does not support this hypothesis (Fig. 2). For example, the TOC ¯ux maximum from 1640 to 1680 AD is cor-related with a SST maximum and anX±Yplot of SST vs. TOC burial ¯ux shows no meaningful relationship (Fig. 7C). Thus, primary productivity does not seem to be the main factor controlling TOC burial ¯ux to the SBB sediments. Similarly, there is no signi®cant corre-lation between SST and alkenone burial ¯ux, supporting the results of Kennedy and Brassell (1992b). However, Herbert et al. (1998) observed, qualitatively, that alke-none content appeared to correlate inversely with UK0
37 SST for the time span of 1919 to 1994, but our longer term data (Fig. 7A) show no relationship.
Hartnett et al. (1998) have suggested that oxygen exposure time, which eectively incorporates environ-mental variables other than primary productivity, con-trols sedimentary organic matter (OM) preservation. Oxygen exposure time is determined mainly by the bot-tom water O2concentration and the sedimentation rate. For the SBB, near-bottom water O2is very low (<0.2 ml/l) at present (Sholkovitz and Gieskes, 1971; Reimers et al., 1990), but no direct measure of past O2 con-centrations exists. However, varve thickness, which records the annual sedimentation rate, is another factor
Fig. 5. Power spectral analysis ofUK0
37 SST between A.D. 1440
Fig. 6. Comparison ofUK0
37 SST (C) vs. TOC content (%) for the period A.D. 1445 to 1935. Data are smoothed over 10 years to
reveal underlying trends.
Fig. 7. X±Yplots of the 5-year means of the selected geochemical time series for the period A.D. 1440±1940. (A)UK0
37 SST (C) vs.
alkenone content (mg gÿ1); (B)UK0
37 SST (C) vs. TOC contents (%); (C)U
K0
37 SST (C) vs. TOC burial ¯ux (mgC cmÿ2yearÿ1); (D)
which should be considered in relation to the preserva-tion of OM in the SBB. The dark laminae (pre-dominantly terrigenous detrital clays) are formed during the winter months, when increased precipitation and river runo transport more terrigenous materials to the basin and the supply of biogenic materials is less, due to decreased productivity (HuÈlsemann and Emery, 1961; Thunell et al., 1995). Conversely, the light laminae (mostly silica from diatoms) are formed during the spring/summer months when the supply of biogenic debris re¯ects upwelling-induced surface productivity. If bottom-water O2concentration is constant, an increase in varve thickness will reduce oxygen exposure time and therefore increase TOC burial ¯ux. Thus, a simpleX±Y plot of annual varve thickness vs. TOC burial ¯ux (Fig. 7D) gives a moderately good correlation (r2=0.56), suggesting that, indeed, varve thickness is a quantita-tively signi®cant factor in controlling TOC preservation in the SBB, in accord with our earlier qualitative obser-vation based on Fig. 2A and C. There is an element of circularity here in that varve thickness is a dominant factor in calculating burial ¯ux, but nevertheless the result in Fig. 7D re¯ects the sedimentary mechanism controlling TOC burial ¯ux.
Increased varve thickness can result from an increased supply of terrigenous material due to more river runo, an increased production of biogenic mate-rials due to enhanced upwelling, or a combination of both processes. For an increase in surface productivity, there is more OM supply to the sediment/water interface which will cause more deep-water O2 depletion, and more supply of biogenic inorganic materials to increase varve thickness. These two processes work together to reduce the O2 exposure time of the OM and hence increase preservation of TOC. The major increases in TOC burial ¯uxes from 1680 to 1740 and from 1880 to 1910 seem to be such situations, since lower SSTs may indicate increased upwelling. An increase in varve thickness caused by higher terrigenous ¯ux would also increase the supply of terrigenous OM by small amounts. Any increase in TOC burial ¯ux would be largely the result of a shortened time of exposure to oxygen caused by a higher rate of sedimentation. The higher TOC burial ¯uxes for the periods of 1640±1670 and 1850±1880 may be such examples, since these peri-ods are characterized by warmer SST (less-upwelling) and higher rainfalls in California.
We thus propose, in accord with the work of Hartnett et al. (1998), that TOC burial ¯ux in the SBB is mainly controlled by sedimentation rate, which in turn controls oxygen exposure time. Studies of the behavior of other marine productivity proxies, such as diatom ¯uxes, are needed, together with any links between varve thickness, local run-o and precipitation in the catchment region. A means to estimate paleo-bottom water O2 concentra-tion would also be of great value.
5. Conclusions
Our results illustrate the bene®ts for molecular strati-graphy which accrue from the analysis of individual varves in a sediment core. Coupled with the enhanced ability for rapid micro-analysis resulting from partial automation, such cores make it possible to derive mole-cular proxy records of approximately annual resolution, appropriate for studies of climate change, especially in the Holocene.
We have reached the following general conclusions for the 1440±1940 sedimentary record obtained from the depocenter of the Santa Barbara Basin. We emphasize that these are based on a single core record and that their relevance for the whole basin needs to be deter-mined by studies of additional cores.
1. The UK0
37 SST record displays interannual (3±4 year) and interdecadal (12±13 year) variability of 1 to 3C, which may be caused in part by ENSO-type ocean-atmospheric interactions. However, individual increases in SST do not correlate well with historical ENSO records, probably as a con-sequence of variable interplay of the oceanic cur-rents in the basin.
2. There is a strong centennial-scale cyclicity in the
UK0
37 SST record which approaches 1C in ampli-tude.
3. There was no signi®cant cooling in SST during the Little Ice Age period, in agreement with a coral record from the Equatorial Eastern Paci®c (Dun-bar et al., 1994) and a tree ring record from the California Valleys (Fritts, 1991).
4. The TOC and alkenone burial ¯uxes do not cor-relate well with the SST record, but do corcor-relate well with the varve thickness. We suggest that the TOC burial ¯uxes in the SBB re¯ect preservation eciency dominated by oxygen exposure time as controlled by sedimentation rate (cf. Hartnett et al., 1998).
Acknowledgements
grateful for the helpful comments from Drs. John Volkman and Elisabeth Sikes and an anonymous reviewer.
Associate EditorÐJ. Volkman
References
Anderson, R.Y., 1992. Long-term changes in the frequency of occurrence of El NinÄo events. In: Diaz, H.F., Markgraf, V. (Eds.), El NinÄo: Historical and Paleoclimatic Aspects of the Southern Oscillation. Cambridge University Press, Cam-bridge, pp. 193±200.
Barnett, T.P., 1985. Long term trends in surface temperature over the oceans. Monthly Weather Review 112, 303±312. Bentaleb, I., Grimalt, J.O., Vidussi, F., Marty, J.-C., Martin,
V., Denis, M. et al., 1999. The C37alkenone record of
sea-water temperature during seasonal thermocline strati®cation. Marine Chemistry 64, 301±313.
Bradley, R.S., Jones, P.D., 1992a. Climate since A.D. 1500. Routledge, London.
Bradley, R.S., Jones, P.D., 1992b. Records of explosive volca-nic eruptions over the last 500 years. In: Bradley, R.S., Jones, P.D. (Eds.), Climate Since A.D. 1500. Routledge, London, pp. 606±622.
Bradley, R.S., Jones, P.D., 1993. Little Ice Age's summer tem-perature variations: their nature and relevance to recent glo-bal warming trends. The Holocene 3, 367±376.
Brassell, S.C., 1993. Applications of biomarkers for delineating marine paleoclimatic ¯uctuations during the Pleistocene. In: Engel, M.H., Macko, S.A. (Eds.), Organic Geochemistry. Plenum Press, New York, pp. 699±753.
Brassell, S.C., Eglinton, G., Marlowe, I., Sarnthein, M., P¯au-mann, U., 1986. Molecular stratigraphy: a new tool for cli-matic assessment. Nature 320, 129±133.
Bria, K.B., Jones, P.D., Schweingruber, F.H., Osborn, T.J., 1998. In¯uence of volcanic eruptions on Northern Hemi-sphere summer temperature over the past 600 years. Nature 393, 450±455.
Chang, P., Ji, L., Li, H., 1997. A decadal climate variation in the tropical Atlantic Ocean from thermodynamic air-sea interactions. Nature 385, 516±518.
Charles, C.D., Hunter, D.E., Fairbanks, R.G., 1997. Interac-tion between the ENSO and the Asian monsoon in a coral record of tropical climate. Science 277, 925±928.
Crowley, T.J., Kim, K.-Y., 1993. Towards development of a strategy for determining the origin of decadal-centennial scale climate variability. Quaternary Science Reviews 12, 375±385.
Crowley, T.J., Kim, K.-Y., 1996. Comparison of proxy records of climate change and solar forcing. Geophysical Research Letters 23, 359±362.
Diaz, H.F., Markgraf, V., 1992. El NinÄo: Historical and Paleoclimatic Aspects of the Southern Oscillation. Cam-bridge University Press, CamCam-bridge p.476.
Doose, H., Prahl, F.G., Lyle, M.W., 1997. Biomarker tem-perature estimates for modern and last glacial surface waters
of the California Current system between 33 and 42N.
Paleoceanography 12, 615±622.
Dunbar, R.B., Wellington, G.M., Colgan, M.W., Glynn, P.W., 1994. Eastern Paci®c sea surface temperature since 1600
A.D.: Thed18O record of climate variability in GalaÂpagos
corals. Paleoceanogrphy 9, 291±315.
Eglinton, G., Bradshaw, S.A., Rosell, A., Sarnthein, M., P¯aumann, U., Tiedemann, R., 1992. Molecular record of secular sea surface temperature changes on 100-year time-scales for glacial terminations I, II and IV. Nature 356, 423± 426.
Eglinton, T. et al., in press. Collected Papers from Workshop
on Alkenone-based Paleoceanographic Inicators. G3(special
issue) (www.g-cubed.org).
Eppley, R., Peterson, B.J., 1979. Particulate organic matter ¯ux and planktonic new production in the deep ocean. Nature 282, 677±680.
Epstein, B.L., D'Hondt, S., Quinn, J.G., Zhang, J., Hargraves, P.E., 1998. An eect of dissolved nutrient concentrations on alkenone-based temperature estimates. Paleoceanography 13, 122±126.
Foukal, P.V., 1990. The variable Sun. Scienti®c American 262, 26±33.
Friis-Christensen, E., Lassen, K., 1991. Length of the solar cycle: an indicator of solar activity closely associated with climate. Science 254, 698±700.
Fritts, H.C., 1991. Reconstructing Large-Scale Climatic Pat-terns from Tree-Ring Data: A Diagnostic Analysis. The University of Arizona Press, Tucson, p. 286.
Gong, C., Hollander, D.J., 1999. Evidence for dierential degradation of alkenones under contrasting bottom water oxygen conditions: implication for paleotemperature recon-struction. Geochimica et Cosmochimica Acta 64, 405±411. Grove, J.M., 1988. The Little Ice-age. Methuen, London, p. 498. Hammer, C.U., Clausen, H.B., Dansgaard, W., 1980. Green-land ice sheet evidence of post-glacial volcanism and its cli-matic impact. Nature 288, 230±235.
Harms, S., Winant, C.D., 1998. Characteristic patterns of the circulation in the Santa Barbara Channel. Journal of Geo-physical Research 103, 3041±3065.
Hartnett, H.E., Keil, R.G., Hedges, J.I., Devol, A.H., 1998. In¯uence of oxygen exposure time on organic carbon pre-servation in continental margin sediments. Nature 391, 572± 574.
Hayward, T.L., Mantyla, A.W., Lynn, R.J., Smith, P.E., Chereskin, T.K., 1994. The fate of the California Current in 1993±1994. CalCOFI Report 35, 19±35.
Hedges, J.I., Keil, R.G., 1995. Sedimentary organic matter preservation: an assessment and speculative synthesis. Mar-ine Chemistry 49, 81±115.
Herbert, T.D., Yasuda, M., Burnett, C., 1995. Glacial±inter-glacial sea-surface temperature record inferred from alke-none unsaturation indices, site 893, Santa Barbara Basin. Proceedings of the Ocean Drilling Program, Scienti®c Results 146, 257±264.
Herbert, T.D., Schuert, J.D., Thomas, D., Lange, C., Wein-heimer, A., Peleo-Alampay, A. et al., 1998. Depth and sea-sonality of alkenone production along the California margin inferred from a core top transect. Paleoceanography 13, 263± 271.
160-kyr record of alkenone-derived sea-surface temperatures from Santa Barbara Basin sediments. Naturwissenschaften 84, 126±128.
HuÈlsemann, J., Emery, K.O., 1961. Strati®cation in recent sediments of Santa Barbara basin as controlled by organisms and water character. Journal of Geology 69, 279±290. Keigwin, L.D., 1996. The Little Ice Age and Medieval Warm
Period in the Sargasso Sea. Science 274, 1504±1508. Kennedy, J.A., Brassell, S.C., 1992a. Molecular records of
twentieth century El NinÄo events in laminated sediments from the Santa Barbara basin. Nature 357, 62±64.
Kennedy, J.A., Brassell, S.C., 1992b. Molecular stratigraphy of the Santa Barbara basin: comparison with historic records of annual climate change. Organic Geochemistry 19, 235±244. Kiladis, G., Diaz, H., 1989. Global climatic anomalies
asso-ciated with extremes in the Southern Oscillation. Journal of Climate 2, 1069±1090.
Latif, M., Barnett, T.P., 1994. Causes of decadal climate variability over the North Paci®c and North America. Sci-ence 266, 634±637.
Lean, J., Beer, J., Bradley, R., 1995. Reconstruction of solar irradiance since 1610: implications for climate change. Geo-physical Research Letters 22, 3195±3198.
Levitus, S., 1994. Climatological Atlas of the World Ocean (NOAA Prof. Pap.13). US Goverment Printing Oce, Washington, DC.
Mann, M.E., Bradley, R.S., Hughes, M.K., 1998. Global-scale temperature patterns and climate forcing over the past six centuries. Nature 392, 779±787.
Marlowe, I.T., Brassell, S.C., Eglinton, G., Green, J.C., 1990. Long-chain alkenones and alkyl alkenoates and the fossil coccolith record of marine sediments. Chemical Geology 88, 349±375.
Martinson, D.G., Bryan, K., Ghil, M., Hall, M.M., Karl, T.R., Sarachik, E.S. et al., 1995. Natural Climate Variability on Decade to Century Time Scales. National Academy Press, Washington.
McCarey, M.A., Farrington, J.W., Repeta, D.J., 1990. The organic geochemistry of Peru margin surface sediments: I. a
comparison of the C37 alkenone and historical El NinÄo
records. Geochimica et Cosmochimica Acta 54, 1671±1682. Mitchell-Innes, B.A., Winter, A., 1987. Coccolithophores: a
major phytoplankton component in mature upwelled waters of the Cape Peninsula, South Africa in March, 1983. Marine Biology 95, 25±30.
MuÈller, P.J., Suess, E., 1979. Productivity, sedimentation rate and sedimentary organic carbon in the oceansÐI. Organic carbon preservation. Deep-Sea Research 26, 1347±1362. MuÈller, P.J., Cepek, M., Ruhland, G., Schneider, R.R., 1997.
Alkenone and coccolithophorid species changes in late Qua-ternary sediments from the Walvis Ridge: implications for the alkenone paleotemperature method. Palaeogeography, Palaeoclimatology. Palaeoecology 135, 71±96.
Neelin, J.D., Latif, M., 1998. El NinÄo dynamics. Physics Today 52, 32±36.
Ohkouchi, N., Kawamura, K., Kawahata, H., Okada, H., 1999. Depth ranges of alkenone production in the central Paci®c Ocean. Global Biogeochemical Cycles 13, 695±704. Overpeck, J., Hughen, K., Hardy, D., Bradley, R., Case, R.,
Douglas, M. et al., 1997. Arctic environmental change of the last four centuries. Science 278, 1251±1256.
Pedersen, T.F., Calvert, S.E., 1990. Anoxia vs. productivity: what controls the formation of organic-carbon-rich sedi-ments and sedimentary rocks? American Association of Pet-roleum Geologists Bulletin 74, 454±466.
Philander, S.G., 1990. El NinÄo, La NinÄa, and the Southern Oscillation. Academic Press, San Diego p.289.
Pickard, G.L., Emery, W.J., 1982. Descriptive Physical Ocea-nography, 4th Ed. Pergamon, New York.
Portman, D.A., Gutzler, D.S., 1996. Explosive volcanic erup-tions, the El NinÄo-Southern Oscillation, and U.S. climate variability. Journal of Climate 9, 17±33.
Prahl, F.G., Wakeham, S.G., 1987. Calibration of unsaturation patterns in long-chain ketone compositions for palaeo-temperature assessment. Nature 320, 367±369.
Prahl, F.G., Muehlhausen, L.A., Zahnle, D.L., 1988. Further evaluation of long-chain alkenones as indicators of paleo-ceanographic conditions. Geochimica et Cosmochimica Acta 52, 2303±2310.
Prahl, F.G., Collier, R.B., Dymond, J., Lyle, M., Sparrow, M.A., 1993. A biomarker perspective of prymnesiophyte productivity in the Northeast Paci®c-Ocean. Deep-Sea Research 40, 2061±2076.
Prahl, F.G., Pisias, N., Sparrow, M.A., Sabin, A., 1995.
Assessment of sea-surface temperature at 42N in the
Cali-fornia Current over the last 30,000 years. Paleoceanography 10, 763±773.
Quinn, W.H., 1992. A study of southern oscillation-related cli-matic activity for A.D. 622±1900 incorporating Nile River ¯ood data. In: Diaz, H.F., Markgraf, V. (Eds.), El NinÄo: His-torical and Paleoclimatic Aspects of the Southern Oscillation. Cambridge University Press, Cambridgepp, pp. 119±149. Rasmusson, E.M., Hall, J.M., 1983. El NinÄo: The great
Equa-torial Paci®c Ocean warming event of 1982±1983. Weath-erwise 36, 166±175.
Reimers, C.E., Lange, C.B., Tabak, M., Bernhard, J.M., 1990. Seasonal spillover and varve formation in the Santa Barbara basin, California. Limnology and Oceanography 35, 1577± 1585.
Rind, D., Overpeck, J., 1993. Hypothesized causes of decade-to-century-scale climate variability: climate model results. Quaternary Science Reviews 12, 357±374.
Rosell-MeleÂ, A., Eglinton, G., P¯aumann, U., Sarnthein, M.,
1995. Atlantic core-top calibration of the Uk
37index as
sea-surface palaeotemperature indicator. Geochimica et Cosmo-chimica Acta 59, 3099±3107.
Sarnthein, M., P¯aumann, U., Ross, R., Tiedemann, R., Winn, K., 1992. Transfer functions to reconstruct ocean palaeo-productivity: a comparison. In: Summerhayes, C.P., Prell, W.L., Emeis, K.C. (Eds.), Upwelling Systems: Evolution since the Early Miocene. Geological Society Special Publica-tion No. 64.. The Geological Society, London, pp. 411±427. Schimmelmann, A., Kastner, M., 1993. Evolutionary changes
over the last 1000 years of reduced sulfur phases and organic carbon in varved sediments of the Santa Barbara Basin, California. Geochimica at Cosmochimica Acta 57, 67±78. Schimmelmann, A., Lange, C.B., 1996. Tales of 1001 varves: a
Schimmelmann, A., Tegner, M.J., 1991. Historical
oceano-graphic events re¯ected in13C/12C ratio of total organic
car-bon in laminated Santa Barbara Basin sediment. Global Biogeochemical Cycles 5, 173±188.
Schimmelmann, A., Lange, C.B., Berger, W.H., 1990. Climati-cally controlled marker layers in Santa Barbara basin sedi-ments, and ®ne-scale core-to-core correlation. Limnology and Oceanography 35, 165±173.
Schimmelmann, A., Lange, C.B., Berger, W.H., Simon, A., Burke, S.K., Dunbar, R.B., 1992. Extreme climatic condi-tions recorded in Santa Barbara basin laminated sediments: the 1835±1840 Macoma event. Marine Geology 106, 279±299. Schimmelmann, A., Zhao, M., Kuhn, G., Tegner, M.J., 1995. Sea surface temperature and paleo-El NinÄo events in Santa Barbara Basin, AD 1840±1940. In: Isaacs, C.M. and Tharp, V.L. (Eds.), Proceedings of the Eleventh Annual Paci®c Cli-mate (PACLIM) Workshop, 19±22 April 1994. Interagency
Ecological Program, Technical Report 40, California
Department of Resources, pp.93-105.
Schimmelmann, A., Zhao, M., Harvey, C.C, Lange, C.B., 1998. A large California ¯ood and correlative global climatic events 400 years ago. Quaternary Research 49, 51±61. Schlesinger, M.E., Ramankutty, N., 1994. An oscillation of the
global climate system of period 65±70 years. Nature 367, 723±726.
Shen, G.T., Cole, J.E., Lea, D.W., Linn, L.J., McConnaughey, T.A., Fairbanks, R.G., 1992. Surface ocean variability at Galapagos from 1936±1982: Calibration of geochemical tra-cers in corals. Paleoceanography 7, 563±588.
Sholkovitz, E.R., Gieskes, J.M., 1971. A physical±chemical study of the ¯ushing of the Santa Barbara Basin. Limnology and Oceanography 16, 479±489.
Sicre, M.-A., Ternois, Y., Miquel, J.-C., Marty, J.-C., 1999. Alkenones in the Northwest Mediterranean sea: interannual variability and vertical transfer. Geophysical Research Let-ters 26, 1735±1738.
de Silva, S.L., Zielinski, G.A., 1998. Global in¯uence of the AD 1600 eruption of Huanaputina, Peru. Nature 393, 455± 458.
Soutar, A., Crill, P.A., 1977. Sedimentation and climatic pat-terns in the Santa Barbara basin during the 19th and 20th centuries. Geological Society of America Bulletin 88, 1161± 1172.
Stein, R., ten Haven, L.H., Littke, R., Rullkotter, J., Welte, D.H., 1989. Accumulation of marine and terrigenous organic carbon at upwelling site 658 and nonupwelling sites 657 and 659: Implications for the reconstruction of paleoenviron-ments in the eastern subtropical Atlantic through late Cen-ozoic times. Proceedings of the Ocean Drilling Program, Scienti®c Results 108, 361±385.
Stuiver, M., Braziunas, T.F., 1993. Sun, ocean, climate and
atmospheric 14CO
2: an evaluation of causal and spectral
relationships. The Holocene 3, 289±305.
Thunell, R., Sautter, L.R., 1992. Planktonic foraminiferal faunal
and stable isotopic indices of upwelling: a sediment trap study in the San Pedro Basin, Southern California Bight. In: Summerhayes, C.P., Prell, W.L., Emeis, K.C. (Eds.), Upwel-ling Systems: Evolution since the Early Miocene. Geological Society Special Publication No. 64. The Geological Society, London, pp. 411±427.
Thunell, R.C., Tappa, E., Anderson, D.M., 1995. Sediment ¯uxes and varve formation in Santa Barbara Basin, oshore California. Geology 23, 1083±1086.
Villanueva, J., Grimalt, J.O., Labeyrie, L., Cortijo, E., Vidal,
L., Cros, L. et al., 1995. Geochemical assessment of C37
ketone sedimentary ¯uxes as indicators of productivity var-iations in the glacial±interglacial oceans. In: Grimalt, J.O., Dorronsoro, C. (Eds.), Organic Geochemistry. Donostia, San Sebastian, Spain, pp. 809±812.
Villanueva, J., Grimalt, J.O., Cortijo, E., Vidal, L., Labeyrie, L., 1998. Assessment of sea surface temperature variation in the central North Atlantic using the alkenone unsaturation
index (Uk0
37). Geochimica et Cosmochimica Acta 62, 2421±
2427.
Volkman, J.K., Eglinton, G., Corner, E.D.S., Forsberg, T.E.V., 1980a. Long-chain alkene and alkenones in the
mar-ine coccolithophoridEmiliania huxleyi. Phytochemistry 19,
2619±2622.
Volkman, J.K., Eglinton, G., Corner, E.D.S., Sargent, J.R.,
1979b. Novel unsaturated straight-chain C37±C39methyl and
ethyl ketones in marine sediments and the coccolithophore
Emiliana huxleyi. In: Douglas, A.G., Maxwell, J.R. (Eds.), Advances in Organic Geochemistry. Pergamon, New York, pp. 219±227.
Wallace, J.M., 1995. Natural and forced variability in the cli-mate record. In: Martinson, D.G., Bryan, K., Ghil, M., Hall, M.M., Karl, T.R., Sarachik, E.S. et al., (Eds.), Natural cli-mate variability on decade to century time scales. National Academy Press, Washington, pp. 199±208.
Weaver, P.P.E., Chapman, M.R., Eglinton, G., Zhao, M., Rutledge, D., Read, G., 1999. Combined coccolith, for-aminiferal and biomarker reconstruction of paleoceano-graphic conditions over the last 120 ka in the northern
North Atlantic (59N, 23W). Paleoceanography 14, 336±
349.
Winant, C.D., Hendershott, M.C., 1996. Winds and surface circulation in the Santa Barbara Channel. EOS, Transac-tions. American Geophysical Union, 76, Ocean Science Sup-plement, OS 181.
Zhao, M., Beveridge, N.A.S., Shackleton, N.J., Sarnthein, M., Eglinton, G., 1995a. Molecular stratigraphy of cores o Northwest Africa: sea surface temperature history over the last 80 ka. Paleoceanography 10, 661±675.
