ARTICLE IN PRESS
Quaternary Science Reviews 25 (2006) 507–525
Evaluation of the Olduvai subchron in the Orce ravine (SE Spain). Implications for Plio-Pleistocene mammal biostratigraphy and the age of Orce archeological sites L. Giberta,, G. Scottb, C. Ferra`ndez-Can˜adellc a
Department Enginyeria Minera i Recursos Naturals, Universitat Polite`cnica de Catalunya, Farinera 2, 08211 Castellar del Valle`s, Barcelona, Spain b Berkeley Geochronology Center, 2455 Ridge Road, Berkeley, CA 94709, USA c Departament d’Estratigrafia, Paleontologia i Geocie`ncies Marines, Facultat de Geologia, Universitat de Barcelona, Martı´ Franque`s s/n, E-08028 Barcelona, Spain Received 9 June 2003; accepted 4 March 2005
Abstract The Barranco de Orce (BO) section in the Baza basin (SE Spain) exposes several fossiliferous layers (O-1 to O-7) with Plio–Pleistocene micro- and macromammals. Biostratigraphic and magnetostratigraphic data from this and other sections in the basin have been extensively used to calibrate the Plio–Pleistocene chronology based on mammal biozonations. Because of its stratigraphic and geographic proximity, the BO section has also been used to date the paleontological and archeological sites of Barranco Leo´n, Fuentenueva-3 and Venta Micena. This study shows that the BO section crosses a mega-landslide that produces partial repetitions of the sedimentary sequence. The seven fossiliferous layers are actually the repetition of only two (O-6 and O-7) which are found in situ in the upper part of the ravine. New paleomagnetic results demonstrate the presence of Reverse magnetization throughout this section, contradicting the Normal event previously reported and assigned to the ‘Olduvai’ subchron (C2n). Published and new magnetostratigraphic data show that all archeological and paleontological sites in the Orce area are within a Reverse magnetochron, presumably C1r.2r (late Matuyama). The use of BO in the magnetobiostratigraphical calibration of the Pliocene/Pleistocene boundary for western Europe is not advised. r 2005 Elsevier Ltd. All rights reserved.
1. Introduction The Guadix-Baza basin, in the Betic ranges (SE Spain), exposes a thick sequence of Plio–Pleistocene continental deposits, rich in fossil mammal sites. A recent paper by Alba et al. (2001) has shown, using Pliocene and Pleistocene data from this and other basins, that the Neogene mammalian fossil record of the Iberian Peninsula is very complete, more than 75% at the specific level, and 90% at the generic one. Because of its continuous sedimentary record from the early Pliocene to the Middle Pleistocene, together with the Corresponding author.
E-mail addresses:
[email protected] (L. Gibert), gscott@ bge.org (G. Scott),
[email protected] (C. Ferra`ndez-Can˜adell). 0277-3791/$ - see front matter r 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.quascirev.2005.03.006
abundance of rich fossil sites, the Guadix-Baza basin has been proposed as a parastratotype area for the Pliocene–Pleistocene boundary in continental sediments (Aguirre, 1997b). After a few preliminary studies in the 1960s and 1970s, the associations of both micro- and macrofauna from the Baza basin were studied in the last two decades. These studies have produced an extensive literature on micromammal systematics, biostratigraphy, and Neogene faunal replacements (e.g. Agustı´ , 1984, 1986a,b, 1990, 1991, 1998; Agustı´ et al., 1986, 1987a–c, 1989, 1993, 2001; Alberdi et al., 1989; Agustı´ and Moya`-Sola`, 1991, 1992, 1998; Marchetti and Sala, 2001). Micromammal associations have also been used to calibrate magnetostratigraphic studies developed in this region (Oms et al., 1994, 1996, 1999, 2000a,b; Agustı´
ARTICLE IN PRESS 508
L. Gibert et al. / Quaternary Science Reviews 25 (2006) 507–525
et al., 1997; Garce´s et al., 1997; Oms, 1998). A significant effort was made to establish micromammal biostratigraphic criteria to recognize the Plio–Pleistocene boundary (e.g. documenting the first occurrence of the Pleistocene marker Allophaiomys pliocaenicus), and to integrate the biostratigraphical data into the stratigraphic and magnetostratigrahic framework (Agustı´ , 1984, 1986a,b, 1991, 2001; Agustı´ et al., 1986, 1987a,b,c, 2001; Agustı´ and Moya`-Sola`, 1991, 1992). Part of the data used in these magnetobiostratigraphical papers comes from the Barranco de Orce (BO) section. Three of the calibrated boundary sections for Neogene mammals of western Europe proposed by Agustı´ et al. (2001) are in the Guadix-Baza basin. The BO section was specifically designated by these authors as the continental boundary section for the Tertiary/ Quaternary mammalian faunas for western Europe (Table 1 in Agustı´ et al., 2001). Additionally, biostratigraphic and magnetostratigraphic data from the BO section have also been used to discuss paleoclimatic events in paleoenvironmental studies of the lacustrine sequence (e.g. Agustı´ and Julia`, 1990; Anado´n et al., 1994). However, these efforts to extend the biostratigraphic and magnetostratigraphic data from the Barranco the Orce sites (O-1 to O-7) into neighboring sections, and to use the data in the calibration of the micromammal chronobiostratigraphy for the Plio– Pleistocene boundary have been frustrated by numerous changes in taxonomic assignments, biostratigraphical zonations and magnetostratigraphical interpretations. In this paper, we present a significant revision of the stratigraphy of the BO section, as corrected by structural analysis of a large landslide present in this area. We include new paleomagnetic data that change the previous magnetostratigraphic interpretation of the BO section and its correlation different sites in this part of the Baza basin. Also, we present a summary of the extensive micromammal literature from the BO section and argue that the low number of specimens together with the unrecognized repetition of the BO fossiliferous layers has led to misinterpretations of faunal associations and, therefore, biozones.
2. Geological setting Neotectonic thrusting and uplift created and isolated the Guadix-Baza Basin from the sea in the late Miocene (Este´vez and Sanz de Galdeano, 1983; Sanz de Galdeano and Vera, 1992). This large basin consists of two sub-basins, Guadix in the SW and Baza in the NE, separated by the Jabalco´n mountain. In the Guadix basin the deposits are basically alluvial with minor influence of Palustrine (Viseras, 1991; Vera et al., 1994). The Baza basin was mainly infilled with evaporitic lacustrine sediments in the central area and palustrine
deposits in the margins, having less influence of alluvial deposits. The studied area is located in the NE sector of the Baza basin near the town of Orce (Fig. 1). This area exposes near 100 m of late Pliocene and Quaternary alluvial and palustrine deposits. The sediments of Orce area (Gibert, L. et al., 1998, 1999b) include the archeological sites of Venta Micena, Barranco Leo´n-5 and Fuentenueva-3, which have yielded human remains and lithic artefacts, together with a rich mammal association (Campillo, 1989, 2002; Gibert, J. et al., 1994, 1998, 1999a,b, 2001; Borja et al., 1997; Tobias, 1998; Roe, 1995). The studied BO section follows the Orce ravine, with its head located in the Jurassic Sierra de Orce (Subbetic zone of the Betic range) and cutting 50 m of Neogene sedimentary succession.
3. Structure of the Orce ravine A preliminary study about the disrupted structure of the Orce ravine and its biostratigraphical interpretations was provided in 1995 (Gibert, L. et al., 1995, 1999a). In 1997, when the first paleomagnetic results of this section were published, the authors wrote: ‘‘Definitive sampling was carried out throughout the section, but disruptions caused by a large number of faults led us to consider only those outcrops free from tectonic displacements. Only the topographically uppermost part of the Orce section was therefore studied, where the O-6 and O-7 localities are located’’ (Agustı´ et al., 1997). The Orce ravine crosses a lobe morphology 1.5 km long and 0.5 km wide, which develops northwards into the main E–W Ve´lez valley (Figs. 2 and 3). Except for the uppermost part, the materials of the entire ravine show different types of structures that reveal movement of the strata due to a mega-landslide. The main listric fault, which separates the in situ southern deposits from the slump lobe, can be followed for more than 1.5 km between the adjacent Fuentecica and Leo´n ravines (Fig. 2B). The whole lobe is crossed by auxiliary faults separating different structural blocks. Antithetic faults are more common in the proximal (southern) areas where the movement initially stopped. Many vertical faults appear to change laterally into subhorizontal faults (Fig. 4). Antiform structures are produced between the auxiliary antithetic and synthetic faults. Different detachment surfaces can be identified in this complex. The most important is associated with the main fault, developed on a basal layer consisting of plastic red siltstone about 10 m thick (Fig. 4A). These alluvial siltstones are overlain by rigid lacustrine limestones with intercalations of dark, uncemented, fine detrital material that now form the slide blocks. The less rigid layers between the limestones are the zones of numerous secondary, low angle detachment surfaces usually parallel to the stratification (Fig. 4B–D). These
ARTICLE IN PRESS L. Gibert et al. / Quaternary Science Reviews 25 (2006) 507–525
509
Fig. 1. Location of the Orce ravine in the Baza basin in south-east Spain.
surfaces are found at the bottom of the slide blocks, which appear undeformed though they can show internal displacement along other minor plastic layers. The basal layers of the blocks normally correspond to a granular media, where a dark fine matrix includes heterometric fragments of adjacent lacustrine strata that were included into the granular media during the displacement (Fig. 5A,B). Such granular layers should be considered as fault breccias, and not as conglomerates. The fabric and structure in these basal layers indicate a fluidization of the material during the slump (Fig. 5C) (see Anders et al. (2000) for similar examples). These multiple fluidized basal layers played an important role in the rapid movement of the landslide.
4. Origin of the landslide Distribution of landslides is largely controlled by bedrock geology (Kawabata and Mokudai, 2002) and earthquakes are the main factor in triggering landslides (Malamud et al., 2002). One major earthquake can generate many contemporaneous landslides, as occurred in Taiwan in 1999 (Liao et al., 2002) and Alaska in 2002.
Tectonic activity in the Baza basin continues from the Miocene to the present, producing earthquakes in the neighboring towns, for example: 4.8 on the Richter scale in Orce-Galera in 1964, and another of 3.1 in Baza in 2001 (Andalusian Institute of Geophysics, 2002). The geological context of the Orce ravine landslide suggests that it originated as a single, gravity-driven event, probably induced by a nearby earthquake (Fig. 6). Many other landslides which affect the same Pleistocene deposits in the Can˜ada de Vele´z and Rı´ o Orce valleys could be contemporaneous to the BO landslide.
5. The age of the landslide In the Late Pleistocene the Baza basin became exorreic (Vera et al., 1994), and began an episode of large-scale erosion and down cutting that continues today. A narrow E–W valley (Ve´lez valley) was developed in the region of the modern BO. The landscape development of rigid, resistant capping carbonatic units, underlain by layers of lutites and clays, combined with the seismic activity in the region facilitates the occurrence of landslides. The dimensions
ARTICLE IN PRESS 510
L. Gibert et al. / Quaternary Science Reviews 25 (2006) 507–525
Fig. 2. (A) View of the slide lobe from the other side of the Ve´lez valley. (B) Aerial view of the studied area showing the lobular morphology of the slumped surface. Altitude curves are separated by 10 m. The rectangle shows the amplified area on Fig. 3.
ARTICLE IN PRESS L. Gibert et al. / Quaternary Science Reviews 25 (2006) 507–525
511
Fig. 3. Detailed aerial view of the Orce ravine showing the location of the paleomagnetic samples (Agustı´ et al., 1997, this paper), the paleontological sites (O-1 to O-7), the faults, and the measured bedding. The Figure also shows the location of the cross section in Fig. 4.
ARTICLE IN PRESS 512
L. Gibert et al. / Quaternary Science Reviews 25 (2006) 507–525
Fig. 4. (A) Main detachment surface over the red alluvial clay of Venta Micena cycle. (B–D) Three examples of antithetic faults, one of them (D) showing subhorizontal movement.
and the geometries of these landslides indicate that they were produced after the valley was well developed to near its current depth. Small ravines, such as BO now erode the BO lobe, produce small alluvial fans at their mouths. These ravines appear to be pre-existing or at least partially developed before the landslide occurred. The present morphology of the Ve´lez valley, with a large meander around the toe of the BO landslide, indicates that the position of the modern streambed has been substantially modified. Taken together, these features indicate that the landslide occurred relatively recently during the latest Pleistocene–Holocene. If we use the activity classification for landslides in southern Spain (Mather et al., 2002) the BO landslide can be classified between dormant-old and dormant-mature (inactive) presenting the following features (Figs. 2 and 3):
The cause of the movement is still identifiable but not likely to re-occur. The main scarp is dissected and vegetated.
Vague lateral margins without lateral drainage. The internal morphology is smooth with undulating topography and a normal stream pattern.
These types of landslides have an estimated age of around 10,000 years (Mather et al., 2002).
6. Paleomagnetic analysis In this paper we suggest a different magnetostratigraphy, therefore age, for the BO section. Our initial reason for recollecting the upper part of the published magnetostratigraphy (Agustı´ et al., 1997) was to refine the sequence of two short polarity events immediately underlying the apparent Olduvai subchron; short events not reported elsewhere in the world. These novel events, if substantiated, would have been a valuable correlation tool for our project of detailed lithostratigraphic correlation and development of the Baza basin.
ARTICLE IN PRESS L. Gibert et al. / Quaternary Science Reviews 25 (2006) 507–525
Fig. 5. (A) Granular media on a basal layer including large (412 cm) fragments of different beds. (B) Internal detachment surfaces through dark clay beds within rigid lacustrine limestones. Note the small intraclasts included in a dark clay matrix and small normal fault with about 10 cm of displacement. This layer corresponds to O-2. C: Flame structure showing fluidification inside the minor detachment layers.
Nine samples were collected in 1998, including samples at both the O-6 and O-7 paleontological sites. Blocks (about 10 cm on a side) were collected from beds that showed the least oxidation/weathering or
513
iron-hydroxide staining. Part of each block was sawn into three to seven specimen cubes (about 8 cm3), sanded, and cleaned with compressed air. Measurements were made with a three-axis cyrogenic magnetometer, and demagnetization made in a non-inductive furnace (+/3nT), all enclosed within a room-sized magnetostatic shield. Arguing for a change in the polarity assignment requires some explanation. This argument is made more difficult since the rocks in question are, in general, poor paleomagnetic recorders. An additional difficulty is that the section is in flat-lying young rocks, which produces a coincidence between Plio–Pleistocene Normal directions and modern Normal directions. These factors combine to confuse a simple interpretation of the upper BO magnetostratigraphy. In addition, we will argue that previous polarity interpretations based on the sign of the mean virtual geomagnetic pole (VGP) can be improved and superceded. A paleomagnetic measurement can be viewed as the composite resultant vector or vector sum of the remanent components carried by the specimen. Laboratory demagnetization is an attempt to selectively remove some of these components, especially those more recently acquired. An example is thermal demagnetization to 1301, which exceeds the Curie point of goethite (Ozdemir and Dunlop, 1996). Fig. 7 shows the subtracted vector directions (difference between the 251 and 1301 vectors). The mean of these directions is coincident, as expected, with the ambient (or Holocene) field direction. Ideally, further demagnetization would remove other remanent components, isolating and revealing the characteristic (primary) remanence. Such ideal behavior is not the case in this study and is uncommon in many magnetostratigraphic studies (see Turner, 2001). We were unable to isolate a simple single component magnetization above 1301, with either a Normal or a Reverse direction. What we did find were multicomponent magnetization directions that were consistent mixtures of Normal and Reverse directions. Multicomponent magnetizations are combinations of two or more remanent vectors. Each vector is a product of an event that affected iron-bearing minerals and/or the grain size of iron minerals. Common events are DRM (depositional/early diagenetic remanence), CRM (chemical precipitation/alteration of iron-bearing minerals), TRM (thermal heating and cooling) and VRM (viscous magnetization acquired through long exposure to a magnetic field). The three expected events in this study are DRM, VRM and a weathering CRM (CwRM). There will be two expected directions from this combination of magnetizations, which will depend on the polarity during deposition. This DRM can be in either a Normal or a Reverse direction. The VRM and the weathering CwRM are directly related with time of surface exposition and will be directed toward the
ARTICLE IN PRESS 514
L. Gibert et al. / Quaternary Science Reviews 25 (2006) 507–525
Fig. 6. Reconstruction of the landslide showing the identified faults and structures, the location of the paleontological sites and the different alluvial (dark grey) and lacustrine (light grey) units.
modern Normal direction. In the case of the Normal directed DRM, in combination with any amount of VRM+CwRM (Normal), the resultant will be in a Normal direction. In the case of the Reverse directed DRM in combination with any amount of Normal VRM/CwRM, the resultant will be in a Reverse direction or in an intermediate direction as larger proportions of Normal vectors are combined. It is this later case of multi-component intermediate directions that we observe in specimens from the in situ BO section. Specimens were thermally demagnetized at six temperatures between 1301 and 3201. In general, at the highest temperatures the specimen directions became
erratic (or the susceptibility increased), limiting the value of increasingly higher temperature demagnetization (see Holm and Verosub, 1988). Two samples (nine specimens) were inconsistent in direction and magnitude during all phases of demagnetization. Seven samples (29 specimens) consistently moved away from the Normal direction upon demagnetization (Fig. 8). The initial subtracted vector (251 to 1301) was directed toward a Normal direction (Fig. 7), as expected for the removal of the remanence carried by goethite and VRM (Ozdemir and Dunlop, 1996; Dunlop and Ozdemir, 2000). In general, subtracted vectors from temperature steps between 1301 and 3201 were subequal mixtures of Normal and Reverse directions (Fig. 9). Therefore, no
ARTICLE IN PRESS L. Gibert et al. / Quaternary Science Reviews 25 (2006) 507–525
515
(A)
Fig. 7. Vector directions related to Goethite and young viscous magnetizations (VRM), based on subtracted vectors between 251 and 1301. Site means and alpha 95 circles of confidence for Barranco de Orce and Barranco Leo´n. A is the modern, long-term average dipole field; P is the present magnetic field.
(B) Fig. 9. Ratio of Normal to Reverse components based on thermal demagnetization. All vectors are deconvolved into two components based on N and R model directions (101 from 000/57, and 201 from 180/57, in the plane of the data). Starting with the components from the highest temperature vector, the components are added from sequentially lower temperature subtracted vectors. Heavy line at 1.0 indicates equal loss of N and R during demagnetization. An example is shown of a typical Reverse specimen with a Normal overprint which was successfully removed (sample from a lower stratigraphic position) (A) Specimens from paleontological site Orce-7 (sample OR.04) and a site 1.0 m above. (B) Specimens from paleontological site Orce-6 (sample OR.01) and a site 0.3 m above.
Fig. 8. Orthogonal demagnetization diagram for a specimen from the paleontological site ORCE-6. Removal of a Normal directed magnetization is shown.
specimen revealed a single component direction of Reverse or Normal polarity. The variable mixing ratios found within (and between) sampled horizons reflect the variation in preservation of the Reverse magnetization and the relative amount of alteration or remagnetization in the modern Normal field. The presence of a Reverse directed magnetization in samples throughout the BO section indicates deposition during a Reverse magnetochron. Most samples from this and the previous study also indicate the presence, in variable amounts, of a Normal directed magnetization which we would argue is of modern origin. The previous study assigned a Normal polarity to any sample with a mixing ratio where the Normal exceeded the Reverse in magnitude. The use of this VGP method would assign a Reverse Polarity if 51% of the remanence is Reverse and
assign Normal polarity if 49% of the remanence is Reverse. This method of polarity assignment appears arbitrary, but more importantly avoids an evaluation or explanation of the presence of the Reverse remanent magnetization in designated Normal samples. Another argument against using VGPs for magnetostratigraphy is that the term is misleading in two ways. Firstly, a VGP is intended as a determination of the geomagnetic field direction. Mixtures of magnetic vectors from different ages will produce an artefact, not a feature of the ancient magnetic field. Secondly, the variation between specimens in the same sampled horizon is the result of variations in the relative amounts of the ancient and modern vectors and is not a feature indicating secular variation of the ancient geomagnetic field. As a substitute for the VGP method, we suggest plotting the angular departure from the expected Normal field direction: D (Hoffman, 1984) toward the antipodal Reverse direction (see Figs. 10 and 11). We conclude from the presence of mixed polarity directions that the DRM or early diagenetic remanence
ARTICLE IN PRESS 516
L. Gibert et al. / Quaternary Science Reviews 25 (2006) 507–525
Fig. 10. Stratigraphic position, paleomagnetic direction, and polarity for the in situ part of Barranco de Orce. Solid diamond symbols are data from Agustı´ et al. (1997). Circles with arrow are data from this report: demagnetized specimen directions at the maximum angular distance ðDÞ from the axial dipole field (Hoffman, 1984), circles with point are samples with no coherent magnetic direction.
is of Reversed polarity (Fig. 10). The in situ fossilbearing sites O-6 and O-7 are of Reversed polarity. Combined with the previous work, it is concluded that all of the in situ BO sediments were deposited during the Matuyama Reversed Chron. The presence of the ubiquitous Normal polarity overprint derives from a combination of vectors whose origin are modern, arising from viscous magnetic realignment and near surface alteration of magnetic minerals. It should be noted that no samples were collected higher in the section owing to the pervasive iron-hydroxide staining and more subdued topography. Apparently these features also limited the sampling of Agustı´ et al. (1997). We suggest that this
uppermost part of the section is an older, pre-landslide surface with a longer weathering history. The section that was sampled appears to be a younger surface that represents the slightly eroded headwall of the landslide scarp. Two unpublished paleomagnetic studies were made in the displaced strata of the lower part of Orce ravine (Se´mah, 1985; Oms, 1998), in which both Reverse and Normal sequences were described. Other studies, mainly biostratigraphic ones, also used samples from the lower part of section (e.g. Agustı´ , 1984, 1986a,b, 1991; Agustı´ et al., 1986, 1987a,b; Agustı´ and Moya`-Sola`, 1992).
ARTICLE IN PRESS L. Gibert et al. / Quaternary Science Reviews 25 (2006) 507–525
517
Fig. 11. Stratigraphic position, paleomagnetic direction, and polarity for the Barranco Leo´n quarry site. Solid diamond symbols are data from Oms et al. (2000b). Circles with arrow from this report: demagnetized specimen directions at the maximum angular distance ðDÞ from axial dipole field direction (Hoffman, 1984).
6.1. Barranco Leo´n The same collection technique as in BO was used at the west wall of nearby Barranco Leo´n quarry. Twelve samples were collected, also in 1998. Samples from the lower 3 m (six samples, 21 specimen cubes, 10 cm3 each) are reported here as an extension of the section published by Oms et al. (2000b). A slightly more detailed demagnetization procedure was attempted than for the correlative BO samples. Alternating field (AF) demagnetization was performed on all specimens at peak fields of 4 and 10 mT. This followed from the work of Dunlop and Ozdemir (2000) showing that AF demagnetization could remove part of a VRM. All specimens were also subjected to eight thermal demag-
netization steps between 801 and 3101. However, multicomponent magnetizations similar to those of BO were found. Five of the six samples clearly have a Reverse polarity component (Fig. 10). These results extend the Reversed polarity section (Oms et al., 2000b) downward by 3 m. Thus, all the fossil and lithic artefact-bearing strata at Barranco Leo´n (Gibert, J. et al., 1992, 1998, 2001; Roe, 1995) were deposited during the Matuyama Reverse Chron. 6.2. Magnetic susceptibility For specimens from both sections, the low field magnetic susceptibility was measured after each thermal demagnetization step, primarily to monitor
ARTICLE IN PRESS 518
L. Gibert et al. / Quaternary Science Reviews 25 (2006) 507–525
mineralogical changes that might occur during heating. One sample from each section showed an increase in susceptibility between 280 and 320 1C. At the BO section, our lowest sample OR.01 (which is the O-6 paleontological site) increased from 0.95 to 2.20 105SI. At the Barranco Leo´n section, sample BL.05 increased from 0.73 to 1.15 105SI. A probable mineralogical change that explains these increases in susceptibility involves non-magnetic sulfide minerals (e.g. pyrite) altering to ferric oxide/hydroxide minerals. The original rock color is olive gray (5Y 4/1) for OR.01 and light olive gray (5Y 6/1) for BL.05 (Munsell color numerical designations). After heating, the color of both samples became light brownish gray (5YR 6/1) with streaks of moderate reddish orange (10R 6/6). Samples with extremely low susceptibility also correlated with poor paleomagnetic performance. From BO, the two samples with the lowest susceptibility (0.002, 0.095 SI) were inconsistent in direction and magnitude thus no polarity determination was made. Likewise in Barranco Leo´n, the only sample without a polarity determination had the lowest magnetic susceptibility (0.35 SI).
7. Discussion 7.1. Magnetochron assignment An assignment of early Matuyama age (C2r,1r) was made for the lower 9.5 m of the in situ BO section (Agustı´ et al., 1997), thus a simple expansion of this magnetozone would be justified. Further reinforcement for an early Matuyama assignment has been added by Agustı´ et al. (2001), who designated the BO section as the boundary section for the MN17/MmQ1 boundary for western Europe. In direct conflict with this, Oms et al. (2000b) assigned a late Matuyama age (C1r.2r) to the adjacent Barranco Leo´n section. This conflict is surprising since these two sites are only 500 m apart (Fig. 2), can be correlated and considered contemporaneous on geologic grounds (Fig. 12). These conflicting studies from adjacent sections did not attempt to resolve the choice in age between mid-late Pliocene or Early Pleistocene. While this report modifies and extends the magnetostratigraphy of both sections, these paleomagnetic data can support either age assignment, but not different ages for the sections. To help clarify this situation, our current research is attempting to find an authentic Normal magnetozone higher or lower in the sedimentary sequence. The ambiguity of having the modern Normal direction coincide in coincidence with Plio–Pleistocene Normal direction, means that paleomagnetic field tests must accompany future claims of finding the Olduvai or Jaramillo magnetochrons. The
fold test and the facies test are both possible in this part of the Baza basin. Waiting for new paleomagnetic results, the mammalian fossil record (see below) supports a late, post-Olduvai, Matuyama age for the BO and the archeological sites. 7.2. Micromammal chronobiostratigraphy Mammal fossils are normally found at isolated sites, usually in fluviatile and lacustrine deposits, or in cave deposits. These occurrences have greatly influenced the mammalian biostratigraphy, in the sense that mammal zones are not true biostratigraphic units but punctual associations of faunas or local faunas defined in reference localities. These local associations are chronologically ordered by the evolutionary stages of taxa or by faunal replacements (‘‘faunal events’’) (e.g. Fahlbusch, 1976; Azzaroli, 1983; Azzaroli et al., 1988). A further difficulty is the correlation between micro- and macromammal faunal events, because these are not always found together. During the last few decades, effort has been made to link these faunal events and reference localities with absolute dating, and with chronostratigraphic stages defined in marine sequences (see Lindsay and Tedford (1990) for a historical review). The continuous Plio– Pleistocene sedimentary record in the Guadix-Baza basin, with its abundant fossil mammalian record, offers a possibility to recognize faunal replacements, date their absolute age, establish a stratigraphical range of faunal associations and correlate to reference localities in Europe as well as to standard stages defined in marine deposits. One main goal is to calibrate mammalian events to the Plio–Pleistocene boundary. The Plio– Pleistocene boundary (base of the Pleistocene Series) was formally defined by a Global Standard Stratotype Section and Point (GSSP) at the base of a claystone unit conformably overlying the sapropelic bed ‘‘e’’ in the Vrica section in Calabria (southern Italy) by the Subcommission on Quaternary Stratigraphy of the INQUA Commission on Stratigraphy at the 27th Session in Moscow, 1984 (Aguirre and Pasini, 1985; Bassett, 1985; Pasini and Colalongo, 1997). The boundary is defined biostratigraphically by LADs and FADs (Last, First Appearance Data) of calcareous nanoplankton and planktonic foraminiferal taxa (Pasini and Colalongo, 1997), but its correlation potential includes astrocyclostratigraphy, magnetostratigraphy, and marine oxygen isotope stratigraphy (Rio et al., 2000). The Vrica GSSP lies just below the top of the Olduvai Subchron (C2n Subchron of Cande and Kent, 1995) (Zijderveld et al., 1991; Pasini and Colalongo, 1997). Thus the importance of identifying this subchron in the Baza basin, with mammalian fossil associations would establish a correlative mammalian zones to the defined Plio–Pleistocene boundary.
L. Gibert et al. / Quaternary Science Reviews 25 (2006) 507–525
Fig. 12. Correlation of the sections with paleomagnetic data, together with Fuentecica section. Sections of Barrranco Leo´n and Fuentenueva-3 contain archeological sites. Paleomagnetic results show Reverse polarity and are consistent with the stratigraphic data. The correlation shows the occurrence of Allophaiomys pliocaenicus. Observe the lack of exposure in the lower Barranco de Orce.
ARTICLE IN PRESS 519
ARTICLE IN PRESS 520
L. Gibert et al. / Quaternary Science Reviews 25 (2006) 507–525
The BO section has been extensively used in biostratigraphy, both to characterize micromammal associations and biozones and to calibrate their relative and absolute chronology using paleomagnetic data (Agustı´ , 1984, 1986a,b, 1991, 1998; Agustı´ et al., 1986, 1987a–c, 1997, 2001; Agustı´ and Moya`-Sola`, 1991, 1992, 1998; Oms et al., 1994, 1996, 1999, 2000a,b). Data from these papers have then been used by other authors to discuss the chronology of Plio–Pleistocene mammalian Iberian and European local faunas as well as the timing of human dispersal into Europe (e.g. Sese´ and Sevilla, 1996; Aguirre, 1997a,b; Aguirre et al., 1997; Martı´ nez et al., 1997; Gibert, J. et al., 1998, 1999a,b). On micromammalian biostratigraphic grounds, the earliest Pleistocene is recognized in Europe by the first occurrence (FAD) of the arvicolid Allophaiomys pliocaenicus (Chaline, 1977, 1997). There has not been a specific relationship established to radiometric dating. Instead, the earliest Pleistocene has been discussed using data from the BO section, which has yielded only a few specimens of A. pliocaenicus in O-3, 4 and 7. Based on different biostratigraphical and magnetostratigraphical interpretations (e.g. Olduvai subchron being located at either O-2 or at O-7), the published first occurrence of A. pliocaenicus in the Baza basin has been interpreted either as prior to (Agustı´ et al., 1997) or posterior to (Agustı´ , 1984; Agustı´ et al., 1987a) the Plio–Pleistocene boundary. Successive magnetostratigraphic interpretations of the BO section were accompanied by different interpretations of the faunal content, identifying up to three different faunal assemblages in BO 1–7 sites (Agustı´ et al., 1987a,b,c), with three populations of A. pliocaenicus: a basal one in O-1 and O2, a middle one with ‘‘primitive’’ forms in O-4 and O-7, and an upper one with ‘‘evolved’’ forms in O-3. Different interpretations were made of the populations of A. pliocaenicus from O-7, either as ‘‘very similar’’ (e.g. Agustı´ et al., 1987a, 1997) or as ‘‘primitive’’ (e.g. Oms et al., 2000a) with respect to the population from the Venta Micena site (Fig. 13). The form of A. pliocaenicus from O-4 was considered much more primitive than that from O-7 and close to the population of the Beftia-2 locality (Agustı´ et al., 1987a–c, p. 79). However, in a latter paper (Agustı´ and Moya`-Sola`, 1998) it was the population from O-7 the one considered primitive and close to that from Beftia. Based on these biostratigraphical and magnetostratigraphical interpretations, the BO section has been suggested as a boundary section to define MN 17 to MQ1 (late Villanyian/early Biharian mammal ages of Europe) based on the FAD of Mimomys ostramosensis (Table 1 in Agustı´ et al. (2001); also see the discussion of the significance of this species at site O-2 in Agustı´ (1986b) and Agustı´ et al. (1986)). However, in previous paper (Oms et al., 2000a) these authors state that the presence of this species in the Guadix-Baza basin ‘‘can no longer be accepted’’.
As an overview, biostratigraphic studies are built on a sequence of observations and interpretations. The first step deals with taxonomy and systematics, which are open to subjective interpretations and need statistically representative populations. Micromammals, and especially arvicolid rodents, show a clear gradualistic evolution (e.g. Chaline and Laurin, 1986; Chaline et al., 1993), and are subject to different taxonomic interpretations (e.g. species assigned to either Allophaiomys, Arvicola or Euphaiomys by different authors, Sese´ and Sevilla, 1996). Another step is the recognition of faunal associations and ranges of taxa. This requires detailed stratigraphic studies, but also depend on the previous systematic and phylogenetic interpretations. The next step is the definition of biozones and their correlation with other biozonations to obtain a relative chronology taking into account possible biogeographic variations, dispersals and endemisms. Finally, these defined biozones can be calibrated with absolute datings to establish an absolute chronology. Biostratigraphic studies in the BO section show deficiencies in all aspects: systematics, stratigraphy, biozonation and absolute datings. Published systematics are based on scarce specimens, usually less than five, which do not allow the recognition of population variability and thus lack statistically precise species characterization and assignment. This feature can be seen in the above-mentioned differing interpretations of a population of A. pliocaenicus as either ‘‘primitive’’ or ‘‘evolved’’ with respect to the population from Venta Micena. The low number of specimens together with the unrecognized repetition of the BO fossiliferous layers by landsliding led to misinterpretations of faunal associations and therefore, biozones. For example, M. ostramosensis was identified in the ‘‘lower’’ strata (O-1 to O-3) and A. pliocaenicus in the ‘‘upper’’ strata (O-4 to O-7), resulting in two biozones in the BO section (Agustı´ , 1984). Based on this latter species, Agustı´ (1986a, 1991) and Agustı´ et al. (1986, 1987b) proposed an initial biozone in the Quaternary (MnQ-1, biozone of M. ostramosensis) previous to the inmigration of A. pliocaenicus, which posteriorly was located in the late Pliocene (Agustı´ and Moya`-Sola`, 1992). Accounting for landsliding, the sites in the BO ravine correspond to repetitions of only two layers, originally separated by less than 3 m, thus the faunal association for all seven sites should be similar. Actually, M. ostramosensis and A. pliocaenicus are found together in the type locality of M. ostramosensis (Chaline, 1997, 1986). Recently, it has been stated that M. ostramosensis no longer occurs in the Guadix-Baza basin (Oms et al., 2000a), although the M. ostramosensis biozone (Mm Q1) was maintained and redefined as the Mimomys oswaldoreigi zone. Summing up all the successive biostratigraphical interpretations (Agustı´ , 1984 to Agustı´ et al., 2001), the faunal association from the BO section has been assigned to a
ARTICLE IN PRESS L. Gibert et al. / Quaternary Science Reviews 25 (2006) 507–525
521
Fig. 13. Successive biochronostratigraphical interpretations proposed for the paleontological sites of the Orce section. Note the shifts of the Barranco de Orce sites within the successive scales. BL: Barranco Leo´n; FN: Fuentenueva; O: Barranco de Orce; VM: Venta Micena.
total range of four biozones, Mn17 to MmQ3 (Fig. 14). We suggest that this range might be a reflection of biodiversity at this location. According to the present study, the BO sites O-1 to O5 are displaced segments of O-6 and O-7, which are close together (o3 m) in a continuous sequence. Additionally, sections at BO, Venta Micena (Scott and Gibert, 1999) and Barranco Leo´n can all be placed within a single reversed polarity interval. Lacking a polarity boundary, the BO section cannot be used to define an absolute chronology to the micromammal biozones. Neither can
the BO section be directly useful in stating the specific relationship of such biozones to the Plio– Pleistocene boundary. If the populations of A. pliocaenicus from these sites are actually the first occurrence of this species in the Baza basin, then it does not coincide with the end of the Olduvai Normal subchron. Further studies will be needed to identify the Olduvai subchron in this part of the Baza basin and to confirm the first occurrence of A. pliocaenicus and its stratigraphic and chronologic relationship with the Plio–Pleistocene boundary.
ARTICLE IN PRESS 522
L. Gibert et al. / Quaternary Science Reviews 25 (2006) 507–525
marker for the Early Pleistocene in Europe/Eurasia (Chaline, 1987, 1997; Aguirre et al., 1997). According to Gabunia et al. (2000a), among the late Villafranchian mammalian faunas of western Europe, the faunal association from Venta-Micena is the most similar to that from Dmanisi, which has been dated in 1.7 Ma (Gabunia et al., 2000b, 2001). The absence of A. pliocaenicus in Dmanisi indicates a very latest Pliocene or earliest Pleistocene age for that site (Gabunia et al., 2001). Owing to the similarity in the faunal association, together with the presence of A. pliocaenicus, the age of Venta Micena and the Orce ravine are slightly younger than Dmanisi and thus corresponds to the post-Olduvai part (C1r.2r) of the Matuyama chron. Fig. 14. Synthesis of the biozonations (Agustı´ et al., 1986–2001) and the differing biostratigraphical interpretation of the micromammal faunal associations from the BO section ‘‘sites’’. The ‘‘sites’’ are actually repetitions of two beds separated less than 3 m. Summing up all successive biostratigraphical interpretations, the faunal association from the BO section has been assigned to a total range of four biozones, Mn 17 to MmQ3.
7.3. The age of Orce archeological sites The polarity determination for the Orce ravine has significance since it is located near archeological sites: Barranco Leo´n-5, Venta Micena and Fuentenueva-3, which have yielded lithic artefacts with evidence of anthropic action as well as fragmentary human remains (Borja et al., 1997; Gibert, J. et al., 1998, 2001; Tobias, 1998). These archeological sites are found in the same subhorizontal palustrine sequence as exposed in the Orce ravine. Because of the proximity of these sites, especially BL-5 (Figs. 1, 2), and their similar stratigraphic position (Fig. 12), the previous paleomagnetic data from the BO section (Agustı´ et al., 1997) had been used to date the archeological sites. The previous paleomagnetic interpretation of O-7 and higher strata as normal polarity (Agustı´ et al., 1997) led to an erroneous dating of BL-5, VM, and FN-3, as being located in the Olduvai subchron (Gibert, J. et al., 1998, 1999a,b). The incorporation of the new paleomagnetic results from the BO section indicates that the correlative archeological sites are located within a period of Reverse polarity. This new interpretation of the BO section is confirmed by the Reversal polarity paleomagnetic results from the adjacent site BL-5 (Fig. 12) (Oms et al., 2000b, this paper). Precise dating of the archeological sites cannot be made from the available paleomagnetic data, neither can it be made from a comparison with other European sites with similar faunal assemblages owing to the lack of radiometric dates. The most significant fossil species in the BO and BL sites is the arvicolid Allophaiomys pliocaenicus (Agustı´ et al., 1987b), which is considered a
8. Conclusions The Barranco de Orce (BO) section crosses a landslide. Sites O-1, O-2, O-3, O-4 and O-5 are located in slid blocks and are repetitions of two fossiliferous layers: O-6 and O-7, which remain in their original stratigraphic position at the top of the section. There is no clear evidence for the presence of the Olduvai Normal subchron in the BO section. Previously identified Normal polarities are apparently missinterpretations of weathering and viscous remagnetization resulting from modern processes. The paleontological sites from the Orce ravine were deposited during the Matuyama Reverse time period. The neighboring section at Barranco Leo´n, including all fossiliferous and lithic artefact bearing strata, is a lateral equivalent section, also of Reverse polarity. Considering that the fauna of the Dmanisi site (Georgia) is no older than 1.7 my, and its faunal association is slightly older than that from Venta Micena, the first occurrence of Allophaiomys pliocaenicus in the Baza basin must occur during a Reverse period of time in the postOlduvai, late Matuyama subchron. The presence of human bones and lithic artefacts in similar stratigraphic horizons (Venta Micena, Fuentenueva-3, Barranco Leo´n) points to this conclusion. The use of BO in the magnetobiostratigraphical calibration of the Pliocene/Pleistocene boundary for western Europe is not advised.
Acknowledgments We wish to thank Dr. C. Swisher for his encouragement of this detailed magnetostratigraphy sampling program and Dr. Josep Gibert for his discussion and useful comments on the geology and biostratigraphy of the Orce region. This study was funded in part by The Earthwatch Institute.
ARTICLE IN PRESS L. Gibert et al. / Quaternary Science Reviews 25 (2006) 507–525
References Aguirre, E., 1997a. Human evolution in the Plio–Pleistocene interval. In: Van Couvering, J.A. (Ed.), The Pleistocene Boundary and the Beginning of the Quaternary. Cambridge University Press, Cambridge, pp. 129–137. Aguirre, E., 1997b. The Pliocene–Pleistocene transition in the Iberian Peninsula. In: Van Couvering, J.A. (Ed.), The Pleistocene Boundary and the Beginning of the Quaternary. Cambridge University Press, Cambridge, pp. 169–177. Aguirre, E., Pasini, G., 1985. The Plio–Pleistocene boundary. Episodes 8 (2), 116–120. Aguirre, E., Vangengeim, E.A., Morales, J., Sotnikova, M.V., Zazhigin, V., 1997. Plio–Pleistocene mammal faunas: an overview. In: Van Couvering, J.A. (Ed.), The Pleistocene Boundary and the Beginning of the Quaternary. Cambridge University Press, Cambridge, pp. 114–128. Agustı´ , J., 1984. Biostratigrafı´ a de los depo´sitos Plio–Pleistocenos de la depresio´n Guadix-Baza (Prov. Granada). Paleontologia i Evolucio´ 18, 13–18. Agustı´ , J., 1986a. Synthe`se biostratigraphique du Plio–Ple´istoce`ne de Guadix-Baza, province de Granada, sud-est de l’Espagne. Geobios 19 (4), 505–510. Agustı´ , J., 1986b. Continental mammal units of the Plio–Pleistocene from Spain. Memorie della Societa` Geologica Italiana 31, 167–173. Agustı´ , J., 1990. The Miocene rodent succession in eastern Spain: a zoogeographical appraisal. In: Lindsay, E.H., Fahlbusch, V., Mein, P. (Eds.), European Neogene Mammal Chronology, NATO ASI Series, Series A, Life Sciences, vol. 180. Plenum Press, New York, pp. 375–404. Agustı´ , J., 1991. The Allophaiomys complex in southern Europe. Geobios 25 (1), 133–144. Agustı´ , J., 1998. A review of the Late Pliocene to Early Pleistocene arvicolid evolution in Spain. Paludicola 2 (1), 8–15. Agustı´ , J., 2001. The Guadix-Baza basin (SE Spain) and the chronostratigraphy of the Plio–Pleistocene in Europe: INQUA, SEQS and EuroMam Workshop. Paleontologia i Evolucio´ 32–33, 5–6. Available at: http://archweb.leidenuniv.nl/EuroMam/frames/ 99euromworkshop.htm. Agustı´ , J., Julia`, R., 1990. Paleoclimatic inferences from the Plio–Pleistocene continental sequence of the Guadix-Baza basin (Spain). Pale´obiologie continentale 16, 269–279. Agustı´ , J., Moya`-Sola`, S., 1991. Spanish Neogene mammal succession and its bearing on continental biochronology. Newsletters Stratigraphy 25, 91–114. Agustı´ , J., Moya`-Sola`, S., 1992. Mammalian dispersal events in the Spanish Pleistocene. Courier Forschung.-Institut Senckenberg 153, 69–77. Agustı´ , J., Moya`-Sola`, S., 1998. The Early Pleistocene mammal turnover in Spain: evidence against an ‘End-Villafranchian’ event. Mededelingen Nederlands Instituut voor Toegepaste, Geowetenschapp.en TNO 60, 513–520. Agustı´ , J., Moya`-Sola`, S., Pons-Moya`., J., 1986. Venta Micena (Guadix-Baza Basin, South-Eastern Spain): its place in the Plio–Pleistocene mammal succession in Europe. Estrato da Geologica Romana 25, 33–62. Agustı´ , J., Moya`-Sola`, S., Martı´ n-Sua´rez, E., Marin, M., 1987a. Faunas de mamı´ feros en el Pleistoceno inferior de la regio´n de Orce (Granada, Espan˜a). Paleontologia i Evolucio´, Memo`ria Especial 1, 73–86. Agustı´ , J., Moya`-Sola`, S., Pons-Moya, J., 1987b. La sucesio´n de mamı´ feros en el Pleistoceno inferior de Europa: Proposicio´n de una nueva escala Biostratigra´fica. Paleontologia i Evolucio´ Memo`ria Especial 1, 287–295. Agustı´ , J., Arbiol, S., Martı´ n-Sua´rez, E., 1987c. Roedores y lagomorfos (Mammalia) del Pleistoceno inferior de Venta Micena
523
(depresio´n de Guadix-Baza, Granada). Paleontologia i Evolucio´, Memo`ria Especial 1, 95–107. Agustı´ , J., Moya`-Sola`, S., Martı´ n-Sua´rez, L., 1989. Review of the late Miocene–early Pliocene mammalian faunas from eastern Spain. Bolletino della Societa` Paleontologica Italiana 28 (2–3), 155–160. Agustı´ , J., Galobart, A., Martı´ n-Sua´rez, E., 1993. Kinslangia gusii sp. nov., a new arvicolid (Rodentia) from the Late Pliocene of Spain. Scripta Geologica 103, 119–134. Agustı´ , J., Oms, O., Garce´s, M., Pare´s, J.M., 1997. Calibration of the late Pliocene–Early Pleistocene transition in continental beds of the Guadix-Baza Basin (SE Spain). Quaternary International 40, 93–100. Agustı´ , J., Cabrera, L., Garce´s, M., Krijgsman, W., Oms, O., Pare´s, J.M., 2001. A calibrated mammal scale for the Neogene of western Europe. State of the art. Earth-Science Reviews 52, 247–260. Alba, D.M., Agustı´ , J., Moya`-Sola`, S., 2001. Completeness of the mammalian fossil record in the Iberian Neogene. Paleobiology 27 (1), 79–83. Alberdi, M.T., Alcala´, L., Azanza, B., Cerden˜o, E., Mazo, A.V., Morales, J., Sese´, C., 1989. Consideracions biostratigra´ficas sobre la fauna de vertebrados fo´siles de la cuenca de Guadix-Baza. In: Alberdi, M.T., Bonnadonna, F.P. (Eds.), Geologı´ a y Paleontologı´ a de la cuenca de Guadix-Baza. Trabajos sobre Neo´geno-Cuaternario 11, Museo Nacional de Ciencias Naturales, Consejo Superior de Investigaciones Cientı´ ficas, Madrid, pp. 347–355. Anado´n, P., Utrilla, R., Julia`, R., 1994. Paleoenvironmental reconstruction of a Pleistocene lacustrine sequence from faunal assemblages and ostracode shell geochemistry, Baza Basin, SE Spain. Palaeogeography, Palaeoclimatology, Palaeoecology 111, 191–205. Andalusian Institute of Geophysics. 2002. University of Granada (Spain), Website, http://www.ugr.es/iag/iagpds.html. Anders, M.H., Aharonov, E., Walsh, J.J., 2000. Stratified granular media beneath large slide blocks: implications for mode of emplacement. Geology 28 (11), 971–974. Azzaroli, A., 1983. Quaternary mammals and the ‘end-Villafranchian’ dispersal event—a turning point in the history of Eurasia. Palaeogeography, Palaeoclimatology, Palaeoecology 44, 117–139. Azzaroli, A., De Giuli, C., Ficcarelli, G., Torre, D., 1988. Late Pliocene to early mid Pleistocene mammals in Eurasia, faunal succession and dispersal events. Palaeogeography, Palaeoclimatology, Palaeoecology 66, 77–100. Bassett, M.G., 1985. Towards a ‘‘common language’’ in Stratigraphy. Episodes 8 (2), 87–92. Borja, C., Garcı´ a-Pacheco, M., Garcı´ a Olivares, E., Scheuenstuhl, G., Lowenstein, G., 1997. Immunospecificity of albumin detected in 1.6 million-year-old fossils from Venta Micena in Orce, Granada, Spain. American Journal of Physical Anthropology 103, 433–441. Campillo, D., 1989. Study of the Orce man. In: Gibert, J., Campillo, D., Garcı´ a Olivares, E. (Eds.), Los restos humanos de Orce y Cueva Victoria. Publicacions de l’Institut de Paleontologı´ a Dr. M. Crusafont, Sabadell. Diputacio´ de Barcelona, pp. 187–220. Campillo, D., 2002. El cra´neo de Orce. Ed. Bellaterra, Barcelona. Cande, S.C., Kent, D.V., 1995. Revised calibration of the geomagnetic polarity timescale for the Late Cretaceous and Cenozoic. Journal of Geophysical Research 100 (B4), 6093–6095. Chaline, J., 1977. Les e´ve´nements remarquables de l’histoire Plio– Ple´istoce`ne des Campagnols dans l’he´misfe`re nord, essai de corre´lation avec la limite Plio–Ple´istoce`ne e´tablie dans le niveaux marins d’Italie. Giornale di Geologia 41, 123–129. Chaline, J., 1986. Continental faunal units of the Plio–Pleistocene of France. Memorie della Societa` Geologica Italiana 31, 175–183. Chaline, J., 1987. Arvicolid data (Arvicolidae, Rodentia) and evolutionary concepts. Evolutionary Biology 21, 237–310. Chaline, J., 1997. Biostratigraphy and calibrated climatic chronology of the Upper Pliocene and Lower Pleistocene of France. In: Van
ARTICLE IN PRESS 524
L. Gibert et al. / Quaternary Science Reviews 25 (2006) 507–525
Couvering, J.A. (Ed.), The Pleistocene Boundary and the Beginning of the Quaternary. Cambridge University Press, pp. 178–182. Chaline, J., Laurin, B., 1986. Philetic gradualism in a European Plio–Pleistocene Mimomys lineage (Arvicolidae, Rodentia). Paleobiology 12 (2), 203–216. Chaline, J., Laurin, B., Prunet-Lecompte, P., Viriot, L., 1993. Morphological trends and rates of evolution in arvicolids (Arvicolidae, Rodentia): towards a punctuated equilibria/disequilibria model. Quaternary International 19, 27–41. Dunlop, D.J., Ozdemir, O., 2000. Effects of grain size and domain state on thermal demagnetization tails. Geophysical Research Letters 27, 1311–1314. Este´vez, A., Sanz de Galdeano, C., 1983. Ne´otectonique du secteur central des Chaˆines Be´tiques (Basins de Guadix-Baza et de Grenade). Revue de Ge´ologie Dynamique Geo´graphie Physique 24, 23–24. Fahlbusch, V., 1976. Report on the International Symposium on Mammalian Stratigraphy of the European Tertiary. Newsletter Stratigraphy 5, 160–167. Gabunia, L., Vekua, A., Lordkipanidze, D., 2000a. The Environmental context of early human occupation in Georgia (Transcaucasia). Journal of Human Evolution 34, 785–802. Gabunia, L., Vekua, A., Lordkipanidze, D., Swisher, C.C., Ferring, R., Justus, A., Nioradze, M., Tvalchrelidze, M., Anto´n, S.C., Bosinski, G., Joris, O., de Lumley, M.A., Majsuradze, G., Mouskhelishvili, A., 2000b. Earliest Pleistocene cranial remains from Dmanisi, Republic of Georgia: taxonomy, geological setting, and age. Science 288, 1019–1025. Gabunia, L., Anto´n, S.C., Lordkipanidze, D., Vekua, A., Justus, A., Swisher, C.C., 2001. Dmanisi and dispersal. Evolutionary Anthropology 10, 158–170. Garce´s, M., Agustı´ , J., Pare´s, J.M., 1997. Late Pliocene continental chronology from the Guadix-Baza basin (Betics, Spain). Earth and Planetary Science Letters 146, 677–687. Gibert, J., Sa´nchez, Malgosa, A., Martı´ nez, B., 1994. De´couvertes de restes humains dans les gisements d’Orce (Granada, Espagne). Comptes Rendus de L Academie des Sciences Paris Serie II 319, 963–968. Gibert, J., Gibert, L., Maestro, E., Iglesias, A., 1998. Two oldowan assemblages in the Plio–Pleistocene deposits of the Orce region, SE Spain. Antiquity 72, 17–25. Gibert, J., Albaladejo, S., Gibert, L., Sa´nchez, F., Ribot, F., Gibert, P., 1999a. The oldest human remain of the Orce region. Human Evolution 14 (1–2), 3–19. Gibert, J., Campillo, D., Garcı´ a-Olivares, E., Walker, M., Ferra´ndez, C., Borja, C., Malgosa, A., Sa´nchez, F., Ribot, F., Gibert, L., Albaladejo, S., Iglesias, A., Gibert, P., 1999b. Contribution a l0 etude des premiers peuplements de l0 Europe occidental: l0 app.ort des reserches sur le Plio–leistocene d0 Orce et Cueva Victoria (Espagne). Jahrbuch des Ro¨misch-Germanischen Zentralmuseums Mainz 46, 39–62. Gibert, J., Gibert, L., Ferra`ndez-Canyadell, C., Iglesias, A., Gonza´lez, F., 2001. Venta Micena, Barranco Leo´n-5 and Fuentenueva-3: three archeological sites in the early Pleistocene deposits of Orce, south-east Spain. In: Milliken, S., Cook, J. (Eds.), A Very Remote Period Indeed. Papers on the Palaeolithic presented to Derek Roe. Oxbow Books, Oxford, pp. 144–152. Gibert, L., Albaladejo, S., Gibert, J., 1995. Estratigrafı´ a del Barranco de Orce. Abstracts of the International Conference on Human Paleontology, Orce 1995. Museo de Prehistoria y Paleontologia de Orce, Granada, pp. 145–147. Gibert, L., Maestro, E., Gibert, J., 1998. Neogene–Quaternary lacustrine-alluvial deposits in the Baza Basin (SE Spain). Abstracts of the 15th International Sedimentological Congress, Alicante, pp. 378–380.
Gibert, L., Albaladejo, S., Gibert, J. 1999a. Estratigrafia del Barranco de Orce. In: Gibert, J., Ribot, F., Sa´nchez, F., Gibert, L. (Eds.), The hominids and their environment in the middle and lower Pleistocene of Eurasia. Proceedings of the International Conference on Human Paleontology, Orce 1995. Museo de Prehistoria y Paleontologı´ a de Orce, pp. 145–147. Gibert, J., Iglesias, A., Maillo, A., Gibert, L., 1992. Industrias liticas en el Pleistoceno Inferior de la regio´n de Orce. Proyecto OrceCueva Victoria. Presencia Humana en el Pleistoceno Inferior de Granada y Murcia. 219–240. Ed. museo de Prehistoria de Orce. Granada. Gibert, L., Maestro, E., Gibert, J., Albaladejo, S., 1999b. Plio–Pleistocene deposits of the Orce region (SE Spain): geology and age. In: Gibert, J., Ribot, F., Sa´nchez, F, Gibert, L. (Eds.), The Hominids and Their Environment in the Middle and Lower Pleistocene of Eurasia. Proceedings of the International Conference on Human Paleontology, Orce 1995. Museo de Prehistoria y Paleontologı´ a de Orce, pp. 127–144. Hoffman, K.A., 1984. A method for the display and analysis of transitional paleomagnetic data. Journal of Geophysical Research 89, 6285–6292. Holm, E.J., Verosub, K.L., 1988. An analysis of the effects of thermal demagnetization on magnetic carriers. Geophysical Research Letters 15, 487–490. Kawabata, D., Mokudai, K. 2002. Effects of bedrock geology and landform on landslide distribution. Landslides: New Monitoring Techniques and Models. AGU Fall Meeting San Francisco, Abstract Book, p. F557. Liao, C., Liao, H., Lee, C., 2002. Statistical analysis of factors affecting landslides triggered by the 1999 Chi-Chi earthquake, Taiwan. Landslides: New Monitoring Techniques and Models. AGU Fall Meeting San Francisco, Abstract Book, p. F556. Lindsay, E.H., Tedford, R.H., 1990. Development and application of land mammal ages in North America and Europe, a comparison. In: Lindsay, E.H., Fahlbusch, V., Mein, P. (Eds.), European Neogene Mammal Chronology, NATO ASI Series, Series A, Life Sciences, vol. 180. Plenum Press, New York, pp. 601–624. Malamud, B.D., Turcotte, D.L., Guzzetti, F., 2002. Landslides and earthquakes. Landslides: New Monitoring Techniques and Models. AGU Fall Meeting San Francisco, Abstract Book, p. F557. Marchetti, M., Sala, B., 2001. The middle Pleiocene–early Pleistocene faunas from Guadix-Baza basin in a biochronological context of Western Europe: a proposal. Paleontologia i Evolucio´ 32–33, 59–64. Martı´ nez, B., Turq, A., Agustı´ , J., Oms., O., 1997. Fuente Nueva-3 (Orce, Granada, Spain) and the first human occupation of Europe. Journal of Human Evolution 33, 611–620. Mather, A.E., Griffihs, J.S., Stokes, M., 2003. Anatomy of a fossil landslide from the Pleistocene of SE Spain. Geomorphology, in press. Oms, O., 1998. Magnetoestratigrafia i litoestratigrafia a la conca de Guadix-Baza i altres punts del Neo`gen continental de les Serralades Be`tiques. Ph.D. Thesis, Universitat Auto`noma de Barcelona, Spain. Oms, O., Garce´s, M., Pare´s, J.M., Agustı´ , J., Anado´n, P., Julia`, R., 1994. Magnetostratigraphic characterization of a thick Lower Pleistocene lacustrine sequence from the Baza Basin (Betic Chain, Southern Spain). Physics of the Earth and Planetary Interior 85, 173–180. Oms, O., Dinare`s-Turell, J., Pare´s, J.M., 1996. Resultados paleomagne´ticos preliminares de la seccio´n plio-pleistocena de Fuente Nueva (cuenca de Guadix-Baza, Cordilleras Be´ticas). Revista de la Sociedad Geolo´gica de Espan˜a 9 (1–2), 89–95. Oms, O., Dinares-Turell, J., Agustı´ , J., Pares, J.M., 1999. Refinements of the European mammal biochronology from the magnetic polarity record of the Plio–Pleistocene (Guadix-Baza Basin, SE Spain). Quaternary Research 51 (1), 94–103.
ARTICLE IN PRESS L. Gibert et al. / Quaternary Science Reviews 25 (2006) 507–525 Oms, O., Agustı´ , J., Gaba`s, M., Anado´n, P., 2000a. Lithostratigraphical correlation of micromammal sites and biostratigraphy of the Upper Pliocene to Lower Pleistocene in the Northeast GuadixBaza Basin (southern Spain). Journal of Quaternary Science 15 (1), 43–50. Oms, O., Pare´s, J.M., Martı´ nez-Navarro, B., Agustı´ , J., Toro, I., Martı´ nez-Ferna´ndez, G., Turq, A., 2000b. Early human occupation of Western Europe: paleomagnetic dates for two paleolithic sites in Spain. Proceedings of the National Academy of Sciences 97, 10666–10670. Ozdemir, O., Dunlop, D.J., 1996. Thermoremanence and Neel temperature of Goethite. Geophysical Research Letters 23, 921–924. Pasini, G., Colalongo, M.L., 1997. The Plio–Pleistocene boundarystratotype at Vrica, Italy. In: Van Couvering, J.A. (Ed.), The Pleistocene Boundary and the Beginning of the Quaternary. Cambridge University Press, Cambridge, pp. 15–45. Rio, D., Castradori, D., Van Couvering, J., 2000. The Pliocene/ Pleistocene boundary-stratotype at Vrica (Calabria, Italia) survived the last challenge. Newsletter 7, 19–21 Available at: http:// www.geo.uu.nl/sns/Newsletters/Newsletter7/Newsletter7.html. Roe, D.A., 1995. The Orce basin (Andalucia, Spain) and the initial Paleolitic of Europe. Oxford Journal of Archeology 14, 1–12. Sanz de Galdeano, C., Vera, J.A., 1992. Stratigraphic record and paleogeographical context of the Neogene basins in the Betic Cordillera, Spain. Basin Research 4, 21–26. Scott, G.R., Gibert, L1. 1999. Evaluation of the Olduvai Subchron in the Orce Region. Workshop on The Guadix-Baza Basin (Andalu-
525
cia Spain) and the chronostratigraphy of the terrestrial Plio– Pleistocene in Europe, INQUA, SEQS and EuroMam Workshop. Orce, Granada, 27–29 May, 1999, pp. 11–12. Se´mah, F., 1985. Analyses pale´omagne´tiques dans la depression de Guadix-Baza (Grenade Espagne). Unpublished. Institut de Paleontologia ‘‘M.Crusafont’’, Sabadell, Barcelona. Sese´, C., Sevilla, P., 1996. Los micromamı´ feros del Cuaternario peninsular espan˜ol: cronostratigrafı´ a e implicaciones biostratigra´ficas. Revista Espan˜ola de Paleontologı´ a, N1 Extraordinario, 278–287. Tobias, P.V., 1998. Some comments on the case for early Pleistocene hominids in south-eastern Spain. Human Evolution 13 (2), 91–96. Turner, G.M., 2001. Toward an understanding of the multicomponent magnetization of uplifted Neogene marine sediments in New Zealand. Journal of Geophysical Research 106, 6385–6397. Vera, J.A., Rodrı´ guez-Ferna´ndez, J., Guerra Mercha´n, A., Viseras, C., 1994. Documents et Travaux, Institut Ge´ologique Albert-deLapparent (IGAL) 14, 1–17. Viseras, C., 1991. Estratigrafia y sedimentologia del relleno aluvial de la cuenca de Guadix (Cordilleras Be´ticas). Tesis Doctoral, Universidad de Granada. Zijderveld, J.D.A., Hilgen, F.J., Langereis, C.G., Verhallen, P.J.J.M., Zachariasse, W.J., 1991. Integrated magnetostratigraphy and biostratigraphy of the upper Pliocene–lower Pleistocene from the Monte Singa and Crotone areas in Calabria (Italy). Earth and Planetary Science Letters 107, 697–714.