Mono Lake excursion recorded in sediment of the Santa Clara Valley, California



[1] Two intervals recording anomalous paleomagnetic inclinations were encountered in the top 40 meters of research drill hole CCOC in the Santa Clara Valley, California. The younger of these two intervals has an age of 28,090 ± 330 radiocarbon years B.P. (calibrated age ∼32.8 ka). This age is in excellent agreement with the latest estimate for the Mono Lake excursion at the type locality and confirms that the excursion has been recorded by sediment in the San Francisco Bay region. The age of an anomalous inclination change below the Mono Lake excursion was not directly determined, but estimates of sedimentation rates indicate that the geomagnetic behavior it represents most likely occurred during the Mono Lake/Laschamp time interval (∼45–28 ka). If true, it may represent one of several recurring fluctuations of magnetic inclination during an interval of a weak geomagnetic dipole, behavior noted in other studies in the region.

1. Introduction

[2] Distinct geomagnetic signatures have been found worldwide within intervals of constant polarity such as the Brunhes Normal Polarity Chron. Magnetic field directions typically remain within about 20° of the mean for a given area and any significant departures from this range are considered anomalous. Such anomalous directions, or excursions, are generally brief, lasting from about 500 years to perhaps 3–5 thousand years (see, for example, Gubbins [1999]). Magnetization directions in sediment and igneous rocks recording an excursion may document a complete polarity reversal, but more often do not. Such excursions can be used as time markers if they can be uniquely identified, but several factors have complicated this effort. Although unusual magnetization directions have been found in paleomagnetic records from many areas, subsequent study of some supposed excursions have proven that the anomalous directions are due to physical disturbances within sedimentary sequences, chemical alteration, remagnetization effects, or perhaps some other non-geomagnetic process (see, for example, Verosub [1975]). Another problem arises because many of the suspected excursions occur in sedimentary sequences where the age of the event cannot be determined directly by any of the absolute dating methods and must be estimated indirectly. Thus, a single geomagnetic excursion can appear to occur at different times in separate geologic sequences, due to differences in sedimentation rates, and be considered separate excursions. In this report we provide additional information on the Mono Lake excursion, which is one of the youngest previously reported for the Brunhes Chron.

2. Late Pleistocene Excursions

[3] One of the earliest found and better-documented excursions in the Brunhes Chron is the Laschamp, in part because it was recorded in volcanic rocks. Two reversed-polarity lava flows were discovered near Laschamp and Olby in the Chaîne des Puys volcanic province, France, by Bonhommet and Babkine [1967]. Bonhommet and Zähringer [1969] considered the lava flows to represent an interval of reversed polarity within the Brunhes Normal Polarity Chron, and estimated that this interval ended between 20,000 and 8,000 years ago. Following the discovery of the Laschamp excursion, Denham and Cox [1971] undertook a paleomagnetic study of the Wilson Creek Formation near Mono Lake, California, to search for additional evidence of the excursion. They assigned an age range of 30,400–13,300 years B.P. to the part of the formation they studied by using available 14C ages determined on ostracodes within two water-laid rhyolitic ash layers [Lajoie, 1966]. Their study showed that a large, rapid, and counter-clockwise excursion of paleomagnetic directions occurred at an estimated 24,600 years B.P. [Denham and Cox, 1971; Denham, 1974]. Although this anomalous field behavior occurred in the same general time frame as that reported for the Laschamp excursion, the fact that a full reversal in direction was not recorded apparently led the authors to conclude that the Laschamp had not been recorded at Mono Lake. The large deviation from the expected axial dipole field direction (42°) was considered by Denham and Cox [1971] to be within the expected range of geomagnetic secular variation. Although an age difference between the anomalous magnetic directions at the two localities could not then be demonstrated, most workers came to accept the occurrence of two separate excursions. Subsequent study of both excursions, described below, seems to support this difference.

2.1. Laschamp Excursion

[4] Revised ages obtained using thermoluminescence and 14C methods on sediment baked by the Laschamp lava flows [Huxtable et al., 1978; Gillot et al., 1979], and K-Ar, 40Ar/39Ar, and 230Th/238U methods on the flows themselves [Condomines, 1978; Hall and York, 1978; Gillot et al., 1979] indicated that the age of this excursion probably is in the range of 45–40 ka. By the late 1970's, a dozen or more possible correlatives to the Laschamp excursion had been suggested worldwide. Most of these occurrences were in sedimentary sequences, however, and could not be directly dated, resulting in a wide range of estimated ages.

[5] Intermediate-polarity directions were found by Kristjansson and Gudmundsson [1980] in tuffs and lava flows of the Reykjanes Peninsula, Iceland, that were in the same general age range as the Laschamp excursion. Again, because the directions found did not indicate a complete reversal or correspond to any of the magnetic directions exhibited by the Laschamp lavas, Kristjansson and Gudmundsson [1980] considered the Icelandic occurrence to represent a separate event, which they referred to as the “Skalamaelifell excursion.” Additional field work in Iceland by Levi et al. [1990] identified the same excursion at other sites, and their compilation of radiometric ages for the Icelandic lavas (42.9 ± 7.8 ka) and the Chaîne des Puys flows (46.6 ± 2.4 ka) shows that both excursions are statistically indistinguishable in age and thus are records of the same geomagnetic behavior occurring at about 45 ka.

2.2. Mono Lake Excursion

[6] Extensive sampling of the Wilson Creek Formation at four additional sites along Wilson Creek, northwest of Mono Lake, by Liddicoat and Coe [1979] provided more details about the Mono Lake excursion and the period of time immediately preceding it. They found that the counter-clockwise rotation of the excursion reported by Denham and Cox [1971] was preceded by a larger, clockwise rotation of direction. Liddicoat and Coe [1979] accepted the established age range of the sequence for consistency with the earlier study, but cautioned that the 14C ages were obtained on ostracodes [Lajoie, 1966] and thus were of unproven reliability. They suggested that the true age could be as much as several thousand years older. Despite the fact that the sequence could be older and the anomalous directions thus closer in age to the Laschamp excursion, Liddicoat and Coe [1979] considered the two to be separate and suggested that the Mono Lake excursion was a feature of the nondipole field. Interpolation and extrapolation of many new tephrachronologic and 14C dates from the Mono Lake and Carson Sink (Nevada) areas [Lund et al., 1988; Liddicoat, 1992] indicate that the age of this excursion is approximately 28 ka. The preponderance of evidence, therefore, is that the Mono Lake and Laschamp are indeed separate excursions, with Mono Lake being approximately 15,000 years younger.

3. Santa Clara Valley Drilling Program

[7] Several monitoring wells are being drilled in alluvial sediment of Santa Clara Valley in a collaborative research effort between the U.S. Geological Survey and the Santa Clara Valley Water District. Results from this program are intended to provide a better basis for assessment of earthquake hazards and management of the groundwater system. Thick sedimentary units deposited during Quaternary time mantle much of this region, including estuarine deposits of San Francisco Bay and broad alluvial plains surrounding the Bay. Thus the sediment encountered is heterogeneous, ranging from coarse gravel and sand to silt and clay. Descriptions of the Quaternary geology, geohydrology and water-quality conditions in the southern Bay region have been provided by Helley et al. [1979], Knudsen et al. [2000], Fio and Leighton [1994], and Leighton et al. [1994].

[8] Paleomagnetic results described in this report were obtained from the first of the research wells drilled under this program, which was named the Coyote Creek Outdoor Classroom (CCOC) after the intended future use of the site [Hanson et al., 2002]. The CCOC drill hole penetrated 308 meters of alluvial fan and stream deposits in the northern part of Santa Clara Valley (Figure 1) that range in age from early Pleistocene (?) to Holocene. Sediment cores were taken at various intervals within the well for a total of 61.8 m of core in the 308-m well.

Figure 1.

Map of the Santa Clara Valley showing CCOC and other new groundwater monitoring wells in their geologic context. Areal geology from Wentworth et al. [1998], Brabb et al. [1998], and Knudsen et al. [2000]; faults modified from these and Jachens et al. [2002].

4. Methods

[9] Seventy-six cylindrical samples, approximately 18.5 cm3 in volume, were taken from individual cores for paleomagnetic study [Mankinen and Wentworth, 2003] after each core had been split longitudinally into “working” and “archival” halves. The cores were oriented only with respect to stratigraphic top, thus permitting magnetic inclination to be determined, but not declination/azimuth. Relative declinations within individual core segments could be determined, however, where internal deformation had not occurred because all samples were taken from the same face of the split core. Samples were taken near the center of the split half to avoid deformation occurring along the core margins and care was taken to avoid any other deformation that could significantly affect the paleomagnetic inclinations. Grain sizes ranging from clay to median sand were sampled and all were found to provide reliable results [Mankinen and Wentworth, 2003].

[10] Natural remanent magnetization (NRM) of each sample was measured using a superconducting magnetometer housed in a magnetically shielded room. Progressive alternating-field (AF) demagnetization experiments were performed using a three-axis tumbling demagnetizer [Doell and Cox, 1967] that was modified to accommodate the large 18.5-cm3 specimens. Doell and Cox [1967] recognized that the demagnetizer could impart a spurious component of magnetization along its innermost rotation axis. This rotational remanent magnetization (RRM, Wilson and Lomax, [1972]) is particularly prevalent in sediments with low magnetic stability. To eliminate the effects of RRM, samples were demagnetized twice at each increment of alternating field above the point where unsystematic behavior was first suspected (generally 20 mT or higher). For the second of these demagnetization pairs, the long axis of the cylindrical sample was reversed 180° with respect to the innermost tumbler axis and the data were averaged using the method of Hillhouse [1977].

5. Results

[11] Paleomagnetic results from all cored intervals within the CCOC drill hole are described in Mankinen and Wentworth [2003] and are available at Results are summarized briefly here and listed in Table 1 (from Mankinen and Wentworth [2003]). The geometric mean NRM intensity for all samples was found to be 20.8 mA/m, and values between 49.7 and 8.71 mA/m are within 1 standard deviation of the mean. Progressive AF demagnetization experiments were performed on selected specimens throughout the cored intervals. All remaining specimens were demagnetized at a minimum of three alternating-field values (5, 10, 15 mT) to confirm their stability and direction. Representative magnetic inclinations were determined by fitting least squares lines [Kirschvink, 1980] to three or more vector endpoints of the magnetic component isolated during demagnetization. These magnetic inclinations were then compared with the inclination that would be produced at the sampling site by a geocentric axial dipole (56.7°). Fifty-six percent of the samples have inclinations within 5° of the expected value, and 77% are within ±10° (Figure 2). Inclinations deviating more than 25° from the expected field are anomalous with respect to the rest of the sample population. Excluding five anomalous samples [Mankinen and Wentworth, 2003], the geometric mean inclination was determined to be 55.4° and statistically indistinguishable from that expected (range of 1 s.d. is from 47.3° to 65.0°). Inclination flattening due to compaction that is sometimes found in sedimentary sequences (see, for example, Deamer and Kodama, [1990]) does not appear to be a factor in the CCOC drill hole. Neither does there appear to be any undetected sediment disturbance that may have biased the results. Based on the stability of magnetization and the statistical parameters for the entire data set, Mankinen and Wentworth [2003] concluded that the sediment obtained from the CCOC drill hole provides an accurate recording of the geomagnetic field. Two intervals within the top 40 meters of the drill hole (Figure 3) that have recorded anomalous inclinations are described below.

Figure 2.

Histogram showing deviation of magnetic inclination (ΔI) of samples throughout the 308-m depth of the CCOC drill hole from that expected at the latitude of the site [Mankinen and Wentworth, 2003].

Figure 3.

Inclination record for cored intervals in the top 40 meters of the CCOC drill hole. Maximum angle of deviation (MAD) [Kirschvink, 1980] for the inclinations shown range from 0.7° to 6.6° (average = 1.8°). Clear areas in lithologic log denote intervals not cored. Tic marks to the right of the lithologic log indicate levels where paleomagnetic samples were taken. Shaded areas show expected normal range of inclinations [Mankinen and Wentworth, 2003].

Table 1. Paleomagnetic Data From the Coyote Creek Outdoor Classroom Drill Holea
CoreSampleDepth, mInclination, degPolarityDelta-I, degM.A.D., degIntensity, mA/m
  1. a

    Drill hole location: 37.337 N; −121.868 W. Inclination is inclination of magnetic vector, positive downward; Polarity: N, normal; R, reversed; I, intermediate; Delta-I is deviation from expected inclination at the drill site; M.A.D., maximum angular deviation [Kirschvink, 1980]; Intensity is intensity of the natural remanent magnetization (NRM).


5.1. Core 17

[12] Anomalous magnetic inclinations were first encountered in core 17, which was collected from a depth of 25.57 to 24.69 meters. This core consisted of 0.35 m of silt and clayey silt overlain by 0.53 m of medium-grained sand and gravel. No internal deformation was apparent, allowing us to determine relative declinations in addition to inclination. A sample from the lower part of core 17 (at 25.32 m) has a magnetic inclination of 38.4° (Figure 4a), which is within the normal range of secular variation for the area. The magnetic vector then swings through an angular distance of 57° to an inclination of 12.9°, just eight centimeters higher (25.24 m). At a depth of 25.21 m, the magnetic vector has moved another 35.3° of arc to an inclination of 19.9°. At this point, the magnetic inclination retains an anomalous orientation, but the overlying sediment was too coarse to sample and determine the total stratigraphic interval over which shallow inclinations are expressed. A sample from core 16, taken 68 cm above the uppermost sample in core 17, has normal inclination. Because the changes in magnetic inclination and relative declination of three paleomagnetic samples from core 17 are large and seem to be serially correlated (Figure 4a), the shallow inclinations are interpreted as recording a geomagnetic excursion.

Figure 4.

Magnetization directions of samples from (a) core 17 and (b) core 21. Solid circles are directions on the lower hemisphere of an equal area projection. Because cores from the CCOC drill hole were not azimuthally oriented, all declinations within each core are relative to one another, but cannot be correlated between cores. Each core was rotated about a vertical axis to produce an apparent northerly declination.

5.2. Core 21

[13] The second interval recording anomalous inclinations was found in core 21, which was collected at a depth of 32.27 to 30.78 meters. This core consists of silty clay to very fine silty sand. One of 5 samples taken from core 21 (Figure 3) has a magnetic inclination that is 30° shallower than expected and near the limit that Mankinen and Wentworth [2003] used to define an intermediate polarity. Because this inclination is at the lower limit of normal, it is possible that it may be approaching the maximum extent of normal geomagnetic secular variation rather than representing a true excursion. On the other hand, the extreme swing of the total magnetic vector (Figure 4b), similar to that seen in core 17, strongly suggests that another excursion has been recorded.

6. Discussion

[14] Because the first anomalous inclinations encountered in the CCOC drill hole occur at such shallow depths (∼25 m-core 17), they most likely were acquired during the Mono Lake/Laschamp time interval as described above. Organic fragments (bark and/or woody twiglets or roots) within black clayey silt between our two intermediate polarity paleomagnetic samples in core 17 yielded an age of 28,090 ± 330 years B.P. (uncalibrated radiocarbon years). The uncertainty given here is three times the quoted laboratory error as recommended by the International Study Group [1982]. This age compares well with the uncalibrated age of 28,620 ± 300 radiocarbon years B.P. determined by Benson et al. [2003] for a volcanic tephra layer (Ash #15) occurring near the midpoint of the Mono Lake excursion at the type locality. The ages of Ash #15 and other tephra in the Pyramid Lake basin, western Nevada, were estimated using 14C ages determined on organic carbon contained in bracketing lacustrine sediment [Benson et al., 2003]. The close agreement in ages between both studies clearly indicates that the anomalous inclinations in core 17 are a record of the Mono Lake excursion. The counter-clockwise direction of rotation (Figure 4a) further indicates that we have detected the younger part of this excursion as characterized by Liddicoat and Coe [1979].

[15] Using the 28.1 ka age for core 17 yields a sedimentation rate of approximately 90 cm/ka for the upper part of the CCOC drill hole. Using the same sedimentation rate, the anomalous inclination in core 21 occurred at about 35 ka. This rate is probably too slow, however, because an unconformity (and hiatus) occurs at the base of the Holocene-age sediment, 2.7 m above core 17 [Hanson et al., 2002], making the 35-ka age an overestimate. Mankinen and Wentworth [2003] estimated Pleistocene deposition rates of 37 and 52 cm/ka for the full depth of the drill hole, based on alternative correlations for an anomalous inclination interval occurring at 305 m. Using these much slower rates, the age of this shallow inclination in core 21 is estimated to be 45 or 40 ka, respectively. All three estimates are indications that the geomagnetic behavior in core 21 most likely occurred during the Mono Lake/Laschamp time interval rather than being a reflection of an older excursion such as the Blake (∼120 ka) or Pringle Falls (∼190 ka).

6.1. Comparison With Other Records

[16] The shallow paleomagnetic inclinations from the upper part of CCOC in the Santa Clara Valley indicate at least two anomalous episodes occurring over an interval lasting some tens of thousands of years. Such a finding is not surprising because most excursions seem to have occurred during periods of very low dipole intensity, intervals that probably lasted some tens of thousands of years in comparison to the typical few hundred years duration of an excursion. In particular, the compilation of absolute paleointensity data by Mankinen and Champion [1993] is permissive evidence that extremely low field strengths existed at the time the Laschamp flows were extruded, and may have lasted until sometime after about 30 ka. They also noted that the strength of the geomagnetic field seems to have been about 35% weaker than the Holocene average over the entire interval between 45 and 10 ka. Although many have suggested that the Holocene geomagnetic field was unusually strong compared with long-term averages, a field intensity of ∼8 x 1022 Am2 seems to have persisted at least throughout Quaternary time [Otake et al., 1993; Selkin and Tauxe, 2000]. Periods of low dipole intensity older than the Laschamp/Mono Lake interval have been determined only by relative paleointensity measurements (e.g., Guyodo and Valet [1999]) that are not calibrated with absolute values. Nevertheless, the overall decrease in field strength during these older intervals is presumed to be comparable in magnitude to that determined for the 40–10 ka interval.

[17] An extended period of low intensities in the 40–10 ka time frame is supported by relative paleointensity records such as SINT-200 [Guyodo and Valet, 1996], but others such as the NAPIS-75 [Laj et al., 2000] show more discrete highs and lows over the same interval. Even in the NAPIS-75 record, the average intensity over most of the 45–10 ka interval is well below normal. In any case, when the strength of the dipole field is weak, non-dipole fields will predominate and unusual field directions can be expected, perhaps repeatedly, as those features wax and wane in strength, or drift geographically [e.g., Merrill and McElhinny, 1983]. Indeed, additional records showing multiple excursions over the same recent interval of time have been reported from other localities in the eastern Pacific region.

[18] Levi and Karlin [1989] provided paleomagnetic results from a continuous and rapidly deposited sedimentary section in a core from the Gulf of California obtained by Deep Sea Drilling Project (DSDP) Leg 64. This core has a detailed record of geomagnetic variations over the past 60,000 years based on δ18O stratigraphy and counts of annual varves. “Noisy” geomagnetic behavior was apparent in sediment deposited between about 50 and 20 ka, which Levi and Karlin attributed to a reduced dipole moment during this time interval. Recurring fluctuations of magnetic inclination were recorded during this part of the sediment record, producing three intervals with anomalous inclinations. Levi and Karlin [1989] correlated the anomalous inclination event between about 51 and 49 ka with the Laschamp excursion, that between about 29 and 26 ka with the Mono Lake excursion, and the youngest, about 3–5 ka later, with the younger (Summer Lake I) of two excursions recorded at Summer Lake [Negrini et al., 1994].

[19] Lund et al. [1988] previously reported similar recurring waveforms in the Wilson Creek beds near Mono Lake, apparently initiated by the Mono Lake excursion. They reported four recurrences of the excursion waveform recorded in that area, each with diminishing amplitudes, and only the initial inclinations were considered anomalous. Negrini et al. [1994] describe a repeating waveform near an earlier excursion (∼190-ka Pringle Falls/Summer Lake II) recorded in Lake Chewaucan deposits of the Summer Lake basin, southern Oregon. However, the section recording the anomalous Summer Lake I inclination, which they consider to represent the Mono Lake excursion, is greatly attenuated due to an unconformity at the oxygen isotope stage 6/5e boundary [Negrini et al., 1994]. The Mono Lake excursion has also been reported from two localities in Pleistocene Lake Lahontan sediment, northwestern Nevada [Liddicoat, 1992]. The excursion is poorly expressed at those localities, however, and no additional information about this period of time is presented.

[20] Excursions and related geomagnetic behavior in the Mono Lake/Laschamp time frame may also have been encountered in three drill holes on the island of Hawaii, although there the dating is less certain. Two lava flows with negative magnetic inclinations and one with horizontal inclination encountered in drill hole HSDP were considered to represent geomagnetic excursions [Holt et al., 1996]. The shallowest excursion, recorded by flow number 23, occurs slightly above a volcanic ash layer yielding a radiocarbon date of 38.6 ± 0.6 ka and thus is correlated with the Mono Lake excursion. We note that flow number 32 below the ash layer also records a very shallow inclination, but cannot be sure whether it may represent behavior most closely associated with the excursion in flow 23 or the next older excursion recorded by flows 39.5, 40, and 42. Based on a 40Ar/39Ar age of 132 ± 32 ka for flow 43, Holt et al. [1996] consider the negative inclinations in the latter 3 flows to represent the Blake excursion. Three geomagnetic excursions were detected in subaerially erupted lavas (0–787 m.) encountered in drill hole SOH1 [Teanby et al., 2002]. Based on various lines of evidence, they estimated that these excursions occurred at 40 ka, 35 ka, and 20 ka. Three geomagnetic excursions also were detected in the upper 675 m of drill hole SOH4 [Laj et al., 2002], with estimated ages of 41 ka, 34 ka, and 18 ka. Importantly, absolute paleointensity measurements on samples from drill holes HSDP [Laj and Kissel, 1999], SOH1 [Teanby et al., 2002], and SOH4 [Laj et al., 2002] show intensities about 40% below average over an interval that ranges from about 70 ka to 20 ka [Laj et al., 2002].

[21] Anomalous geomagnetic behavior occurring repeatedly in the Mono Lake/Laschamp time frame has also been reported from widely separated geographical localities including, for example, the Arctic Ocean [Nowaczyk and Baumann, 1992; Nowaczyk and Knies, 2000] and western Norway [Mangerud et al., 2003]. We believe, however, that attempts at long-range correlation of such excursions should be avoided. If most excursions do occur during periods of very low dipole intensity, the magnetic field at geographically distant sites may or may not exhibit anomalous behavior depending on whether the site is in proximity to a strong non-dipole source. As a result, any concurrent similarities in geomagnetic behavior are likely to be coincidental. While excursions can be expected at many localities, the chance that any two will be entirely synchronous seems unlikely, particularly if the weak dipole window is some tens of thousands years long. For these reasons, we restrict our discussion to geomagnetic behavior in our immediate region. In this context, the term “region” is used in a general sense because the actual dimensions of a given region will depend on the size of any nearby nondipole feature. Compare, for example, the size of the nondipole low near central Africa with the high centered over Mongolia in the geomagnetic nondipole field for 1945 [Bullard et al., 1950].

6.2. Corrected Age of the Mono Lake Excursion

[22] The radiocarbon age we present above (28,090 ± 330 radiocarbon years B.P.) is in excellent agreement with the age presented by Benson et al. [2003] and demonstrates that the Mono Lake excursion was indeed recorded by the sediment in the CCOC drill hole. However, a number of factors have been shown to affect atmospheric 14C concentrations through time and require ages determined by the 14C method to be calibrated. Radiocarbon ages have been well calibrated to about 13,000 years B.P. by comparison with tree ring and glacial-varve chronologies, and extended to about 20,000 years using 234U-230Th ages on coral reefs [Stuiver et al., 1998; Bard et al., 1998]. Bard [1998] presented a method to extrapolate the coral reef data to older time frames. His approximation yields a calendar age of 32,840 years B.P. for the Mono Lake excursion in the CCOC drill hole.

[23] Voelker et al. [1998] correlated the paleoclimatic δ18O record in the annual-layer-counted timescale of the Greenland GISP2 ice core with the marine record in sediment cores from the Iceland and Norwegian Seas. Comparison of their planktic 14C dates with the GISP2 ice core provided many calibration points showing large fluctuations in 14C concentrations, which they attempted to correlate with the then available geomagnetic model of atmospheric Δ14C changes [Laj et al., 1996]. Recently, Voelker et al. [2000] correlated the peaks in 14C production in their marine record with corresponding peaks in 10Be and 36Cl in the ice core, and compared these with a stacked record of relative geomagnetic paleointensity (NAPIS-75) from the North Atlantic [Laj et al., 2000]. Peaks in the production of the cosmogenic nuclides should coincide with minima in the Earth's geomagnetic intensity, and those in the range of 33.5–34.5 and 40.3–41.7 calendar years BP were attributed to the Mono Lake and Laschamp excursions, respectively [Voelker et al., 2000; Laj et al., 2000]. A third peak in Δ14C was observed at about 38 ka without an apparent corresponding paleointensity minimum in NAPIS-75. 36Cl data from the GRIP ice core [Wagner et al., 2000; Beer et al., 2002] indicate that the two peaks attributed to Mono Lake and Laschamp excursions occurred at about 32 and 40 ka.

[24] Despite some discrepancies between timescales derived from the two Greenland summit ice cores, the ∼32–34 ka age range attributed to the Mono Lake excursion from analyses of those records is in general agreement with the 32.8 ka age for the Mono Lake excursion in the CCOC drill hole as determined using the Bard [1998] method. We prefer, however, to use the Bard approximation rather than relying on any long-range correlation such as to intensity minima in the North Atlantic. The 32.8 ka age may be subject to some adjustment in the future as more 14C calibration points become available for ages older than about 25 ka. In support of our preference, we note that marked directional changes occur simultaneously with the minimum values of relative paleointensity in five of six, stacked NAPIS-75 records within the interval that Laj et al. [2000] correlate with the Laschamp excursion. On the other hand, only one of these records (PS2644, Voelker et al. [1998] and Laj et al. [2000]) gives any indication of a directional change near the time attributed to the Mono Lake excursion. Thus, although possible, it is far from certain that the directional changes we see in drill hole CCOC occurred during the intensity minima indicated by the NAPIS-75 record. This finding highlights our concern about long-range correlation of such brief geomagnetic behavior.

[25] Kent et al. [2002] recently provided new radiocarbon and 40Ar/39Ar ages from the Wilson Creek Formation near Mono Lake. Their 14C analyses on carbonate from 11 stratigraphic horizons yielded an uncalibrated age of 32 ka, which was considered to be a minimum age. Although there was a great deal of scatter in ages determined by the 40Ar/39Ar method for the ash layers in the section, Kent et al. [2002] determined a maximum age of 49.9 ± 0.8 ka for Ash #15 near the midpoint of the excursion recorded by these sediments. Based on their analysis of various data, Kent et al. concluded that the excursion most likely occurred between about 41 and 38 ka and thus was a record of the Laschamp rather than a younger excursion (the Mono Lake). Because the ash layers in the section are near or below the lower limit of the 40Ar/39Ar method, and because contamination has resulted in different populations of sanidine being present [Kent et al., 2002], these determinations provide only marginal age constraints. The 14C ages reported by Kent et al. [2002], however, are much older even though they were performed on materials similar to those used for other determinations in the Mono Lake and Carson Sink areas. Benson et al. [2003] described possible sources of error in dating inorganic carbon and concluded that some of the assumptions made by Kent et al. [2002] were incorrect, resulting in ages that are most likely overestimates of the time of deposition.

7. Conclusions

[26] Our results show that the younger part of the Mono Lake excursion occurred at 28,090 ± 330 radiocarbon years B.P. (32.8 ± 0.3 ka, corrected according to Bard [1998]). Because the two intervals of anomalous inclination recognized in the CCOC cores are stratigraphically separated by more than 6 m, it seems unlikely that the clockwise rotation of direction seen in core 21 (Figure 4b) reflects the older part of the Mono Lake excursion (compare with Liddicoat and Coe [1979, Figure 3]). It is therefore tempting to conclude that the anomalous inclination seen in core 21 is a record of the Laschamp excursion. We are reluctant to make such a correlation, however, in the absence of a reliably determined age, and because of uncertainties in long-range correlations. Using a relatively rapid sedimentation rate for the shallowest parts of the CCOC drill hole (see above), the anomalous inclination in core 21 may have occurred at about 35 ka, which is somewhat younger than the age commonly associated with the Laschamp. The much slower rates estimated for the Pleistocene yield ages ranging from 45 to 40 ka, in excellent agreement with radiometric ages determined for the Laschamp. Without a precise age, we also cannot rule out the possibility that the anomalous inclination in core 21 occurred at the same time as the third maximum in 14C concentration in the GISP2 ice core (∼38 ka-[Voelker et al., 2000]), perhaps recording a new, previously unreported excursion. This latter possibility is dependant, however, upon the proximity of a strong non-dipole source at that time.

[27] Excursions have been correlated to the Laschamp in most geographic regions worldwide but with a wide range of estimated ages. The span of ages may occur, as we have suggested, because anomalous paleomagnetic directions are not entirely synchronous from region to region. To avoid confusion when there is a lack of well-determined ages, the behavior might simply be reported as having occurred during the Mono Lake/Laschamp weak geomagnetic dipole interval. When a related excursion has been dated, such as the Mono Lake excursion, that name should be used in a regional sense. The same scheme can be used for older periods, because many of the proposed excursions worldwide seem to occur in clusters at approximately the same times. We believe that simplifying how we view geomagnetic excursions would be beneficial; the current situation with twenty or more named excursions in the Brunhes Chron makes any attempt at correlation extremely difficult.


[28] We appreciate the thoughtful reviews by J. T. Hagstrum, J. C. Liddicoat, S. P. Lund, A. M. Sarna-Wojcicki, N. Teanby, and Associate Editor, A. Jackson. Samples were processed by J. P. McGeehin at the USGS 14C laboratory in Reston VA, and dated at the Center for Accelerator Mass Spectometry, Lawrence Livermore National Laboratory, CA, all under the auspices of the USGS Climate History/Hazard Program.