449

Journal of Paleolimnology 21: 449–460, 1999. © 1999 Kluwer Academic Publishers. Printed in Belgium.

Late Glacial to Early Holocene environmental adjustment in the Mediterranean semi-arid zone of the Salines playa-lake (Alacante, Spain)* S. Giralt1, F. Burjachs1, J. R. Roca2 & R. Julià1 1 Institute of Earth Sciences ‘Jaume Almera’, (CSIC). Lluís Solé i Sabarís, s/n. E-08028, Barcelona, Spain (E-mail: [email protected], [email protected] & [email protected]) 2 Department Microbiology and Ecology, Universitat de Valéncia, E-46100 Burjassot (Valéncia, Spain) (E-mail: [email protected]) Key words: Mediterranean environment, paleolimnology, paleoecology, Early Holocene, saline lake

Abstract The transition from the Late Glacial to the Early Holocene in the endorheic Salines sequence, which is characterized cyclical sedimentation, occurs between 5.50 and 2.85 m depth. From 5.50–3.50 m depth the cycles are composed of a centimetre alternation of layers of dolomitic marls and gypsarenites and from 3.50–2.85 m depth by the alternation of calcitic marls and calcarenites. Pollen, biotic assemblages and geochemistry provide evidence of a gap with respect to the new hydrological conditions that characterized the beginning of the Holocene. Mesic pollen taxa increased their percentages at the beginning of the Holocene, indicating climate improvement, which coincides with the 14C radiocarbon age of 10,000 years BP. The first biotic remains (gastropods, ostracods and foraminifers) found in this sequence appeared later, at 3.80 m depth, which corresponds to 9,500 years BP, whereas the mineralogical change occurred at 3.50 m depth, which corresponds to 9,000 years BP. The advanced adaptation of the vegetation and biotic aquatic assemblages with respect to the mineralogical response corresponds to a process of a gradual increase in water availability into the lacustrine system. During the Boreal, the calcitic cycles reached their maximum thickness, suggesting a more continuous water input. This assumption has also been corroborated by the expansion of the mesic pollen taxa and the occurrence of biota taxa which depend on a permanent water body for their development. The multiproxy approach in paleoclimate scenarios is an essential tool for understanding the ecosystem adjustment during climate changes. Our results demonstrate an interval of 1000 years between the vegetal and the mineralogical reaction.

Introduction Data on the last deglacial process and on the new climate conditions which characterize the Holocene constitutes one of the most exciting aspects of the paleoclimate models, mainly for the circum-Atlantic lands (Lowe & Members, 1995; for a general revision). Between 15,000 and 10,000 years BP (a comparable time would be from the Pyramid epoch till today) the northern hemisphere underwent considerable climate *This paper was presented at the 7th International Symposium on Palaeolimnology (1997), held at Heiligkreuztal, Germany

changes (i.e., a fall of 6 °C in the mean temperature has been reported for the YD event (Wansard, 1996)). These climate changes, which were mainly triggered by North Atlantic processes, brought about considerable adjustments in the ecosystems. These adjustments have been supported by paleontological evidence (Duplessy et al., 1981, 1992; Fairbanks, 1989; Bard et al., 1994; Sarnthein et al., 1995). However, areas removed from the Atlantic seem to be less sensitive to these climate events (Thompson et al., 1995). Thus, the propagation or teleconnections of these climate changes and their interferences with other climate events (such as the ITCZ northward shifts (Gasse & Van Campo,

450 1994)) as well as the chronology and the kinematics of community modifications warrant further investigation. The Iberian Peninsula, located in southern European and exposed to Atlantic influences, was regarded as the eastern border of the polar front during the Last Glacial Maximum (CLIMAP, 1984; Ruddiman & McIntyre, 1981) and the Late Glacial climate changes are well recorded in lands adjacent to the Atlantic (Watts, 1985; Pons & Reille, 1988; Montserrat, 1992; Pérez-Obiol & Julià, 1994; Allen et al., 1995; Peñalba et al., 1997). However, towards the southeast of the Iberian Peninsula, which is affected by the arid Mediterranean climate, the deglacial phases or chronozones as defined by Mangerud et al. (1974) are less evident. Some authors even cast doubt on their existence in Mediterranean areas. This particular behaviour suggests a more complex pattern, which demands explanation. This paper seeks to account for the time-gaps due to the inertial ecosystem response to these rapid climate changes by correlating the different proxy responses in an arid Mediterranean environment, which has never been affected by glacial activity.

Description of the studied site The Salines playa-lake extends to the foot hills of the Salines (1,240 m a.s.l.) and Cabrera ranges, which belong to the eastern end of the Alpine Betic Chain in the Iberian Peninsula (Figure 1). These ranges are primarily formed by Mesozoic limestones, dolostones and gypsum. During the Paleogene and Neogene, continental and marine sediments were deposited over the Mesozoic bedrock. Large alluvial fans made up of soil horizons (brown and calcrete soils) and detrital slope debris developed between these ranges and the lake. The Salines playa-lake, which covers an area of approximately 1.6 km2, has a catchment area of about 71 km2. The maximum water volume that the lake is capable of holding is about 113 hm3. The mean annual precipitation is 350 mm and the mean temperature is about 14°C with abrupt daily and seasonal fluctuations. The annual evapotranspiration rate measured in pan evapometers is about 1500 mm. Hence, the Salines area has been considered to have a xeric Mediterranean climate (Sánchez-Toribio et al., 1990). The Salines playa-lake is fed by runoff and groundwater, and the lake was used as a salt-mineral resource mainly in the nineteenth century. Later, it underwent a fall in its water level due to groundwater exploitation,

which has resulted in the current dryness. Nevertheless, on occasions, it refills during rainy periods, when the water chemistry records a predominance of sodium, magnesium and calcium among the cations and sulphate and chloride among the anions.

Methods Three cores (SAL-1 (22.60 m long), SAL-2 (4.74 m long) and SAL-3 (46.77 m long)) were drilled in the central part of the Salines Lake in January, 1993, and a 46.77 m continuous core, 10 cm in diameter, was obtained in PVC pipes using a wire corer. The cores were stored in a cool room (+2°C) prior to analysis before being split longitudinally for description and sampling. A synthetic profile was built bearing in mind the magnetic susceptibility and the lithological correlation of these three cores. The data presented here correspond to the interval extending from 2.85–5.50 m depth. A total of 128 samples were taken from the central part of the cores for mineralogical investigations. The samples were weighed, dried at 60°C for approximately 48 h, and weighed again in order to determine their water content. The samples were ground by hand using a small agate mill. X-ray diffractions were performed with an automatic Siemens D-500 x-ray diffractometer: Cu kα radiation, 40kV, 30 mA and graphite monochromator. The quantification of the different mineral species followed the standard procedure (Chung, 1974a, 1974b). On the basis of a large number of measurements, replicate analyses for contiguous samples indicate a precision of ±1%, whereas in extreme cases, a precision of about 3%. The mineralogical zones were established using the Edwards & Cavalli-Sforza chord distance method of TILIA software (Grimm, 1987). Continuous samples for thin sections were taken from the central part of the core. 3 × 2 trapezoidal petrographic thin sections (30 µm in thickness), with an overlapping of 1 cm at each end were obtained after freeze-drying and balsam-hardening. The thickness of each cycle occurring in the interval from 2.85–5.50 m depth was measured in several parts of the thin section under a petrographic microscope. Pollen and spores were extracted by physico-chemical treatment using about 10 g of dry sediment with the aid of the Goeury & de Beaulieu (1979) technique, modified by Burjachs (1991). A mean of 543 pollen grains was counted. The basic sum used for calculating the percentages includes all taxa excluding hygrophytes and hydrophytes. The mesic tree category includes

451

Figure 1. Geographical location and geological setting of the Salines site and hombrothermical diagram.

452 Acer, deciduous Quercus, Corylus, Tilia, Fraxinus, Alnus, Ulmus and Salix taxa, and the xerophytic tree category comprises evergreen Quercus and OleaPhillyrea taxa. The shrub pollen category is made up of Cistaceae, cf. Erica, Buxus, Chamaerops, Coriaria, Rhamnus and Thymelaeaceae taxa. For biota analysis, 15 ml of wet sediment samples were dried at 60°C and weighed. After treatment with hydrogen peroxide, the samples were washed and filtered through a 200 µm-mesh sieve. The remaining sediment and fossils were dried and weighed again. Subsamples were taken and counted at 15–40 × in a reticulate (1 cm square) box (9 × 13 cm) using a stereomicroscope. Ostracods and other biotic remains, such as gastropods and diatoms were identified. The time scale was framed using 5 uncalibrated AMS radiocarbon ages obtained from different organic materials (Table 1). According to these dates, the Lateglacial to early Holocene transition is recorded between 2.85 and 5.50 m depth of this sequence.

Results X-ray diffractions Eleven mineral phases were found: calcite (0–70% wt), high-magnesian calcite (0–40% wt), dolomite (0–70% wt), calcian dolomite (0–12% wt), calcian magnesite (0–3% wt), aragonite (0–60% wt), celestine (0–4% wt), gypsum (0–93% wt), quartz (1–16% wt), illite (2–25% wt) and halite, not shown in the diagram (Figure 2). The calcian magnesite has been recently described as an identifiable discrete mineral phase in depositional saline lake environments (Queralt et al., 1997). Facies assemblages The core sediments are marls and silty sands. Large diagenetic gypsum crystals cut across the primary sedimentary structures in several places. Table 1. Uncallibrated AMS 14C of the Salines core Drill

Prof. Age (m) (years BP)

Material

Reference Number

SAL-2 SAL-3 SAL-1 SAL-1 SAL-2

1.75 7,660 ± 50 2.71 8,570 ± 70 3.44 8,810 ± 60 4.30 10,120 ± 60 5.10 11,540 ± 110

Charcoal Wood Pollen Charcoal Pollen

Beta-70900 CAMS-11925 Beta-62417 CAMS-6513 Beta-67374 CAMS-9766 Beta-70899 CAMS-11924 Beta-90850

The studied part of the sequence is mainly composed of a cyclical sedimentation pattern with a centimetre alternation of two facies: marls and fine sands. The upper and lower parts of the studied part are massive, without any visible sedimentary structure. Marl facies Two main subfacies can be differentiated in this part of the studied sequence on the basis of their mineralogical composition: the first, between 4.70 and 3.50 m, where dolomite is the main carbonate, and the second, located between 3.50 and 2.85 m depth, where the mineral phases of the calcareous carbonates are dominant (Figure 3). Usually, the dolomite marl layers are grey in colour and present a micrite texture under an optical microscope. In fact, the size of the grains composing these layers is about 2 µm under an electronic microscope. These micritic grains can constitute bigger aggregate particles. These marls are massive, but near the top, parallel or domic laminated structures can be found. Remains of vegetal fibres and millimetre vertical cracks are common inside this lamination. The thickness of these marls ranges between 0.05 and 4.55 cm. The contact between the marls and the upper sands is gradual, whereas the contact with the lower ones is abrupt. The colour of the calcareous marls changes progressively from grey, at the bottom, to a dark olive green at the top. These marls are formed by a mixture of carbonates (mainly calcite, high-magnesian calcite, dolomite and aragonite). They have a pelletoid texture, with abundant bioclasts, such as gastropods and ostracods. Some of these organisms have geopetal sediment. Vegetal remains and charcoal fragments are also found. The thickness of the calcareous marl layers ranges between 0.1 and 9 cm, and the contacts between the upper and lower sands are relatively abrupt. Silty sand facies Two subfacies can be differentiated on the basis of their mineralogical composition: the first, located between 4.70 and 3.38 m depth, is mainly composed of gypsum, while the second, from 3.38–2.85 m depth, is made up of carbonate particles. Gypsum subfacies The gypsum sandy facies can also be subdivided into two microstructures on the basis of the morphology and

Figure 2. X-ray diffraction diagram of the studied section (2.85– 4.70 m) of the Salines sequence. The mineral phases are expressed in percentages. The two zones has been established according to the cluster results.

453

Figure 3. 3-mean running average of the deviation from the mean cycle thickness, and the mineralogical spectra of the calcitic and dolomitic cycles. Dotted frame represents the sandy facies while the short dashed line frame represents the marl facies.

454

455 organization of the crystals. The first, located between 4.70 and 4.10 m depth, is formed by heterometric (between 50 and 300 µm), euhedral or subhedral crystals of gypsum. Usually, these sandy layers are organized in a coarse-tail grading structure, with welldeveloped tail-twins at the top. The second microstructure of sandy gypsum is mainly constituted by heterometric (50–250 µm), anhedral or subhedral, and rounded crystals, without an internal structure. These layers are located between 3.99 and 3.38 m depth, and their thickness ranges between 0.06 and 1.53 cm.

(Kromer & Becker, 1992). The second group is composed of the other radiocarbon dates (11,500 ± 110, 8,570 ± 70 and 7,660 ± 60 years BP). Although the first group of radiocarbon dates does not allow us to establish accurately the timing of the different events recorded in the Salines sequence these dates do provide a guideline for the sedimentation rate model. The sedimentation rate model for the Salines lake (Figure 4) was obtained using the best fitted adjustment taking into account all the radiocarbon dates. Pollen

Calcarenite subfacies This sandy subfacies is mainly composed of detrital carbonate grains, well-rounded and well-sorted, and bioclasts, such as microforaminifera, gastropods and ostracods. Gypsum grains and charcoal particles are also common. Distribution of the sedimentary facies The marl and sandy facies described above present a vertical distribution throughout the sequence. Three main facies sequences can be differentiated. The first sequence, located between 4.70 and 4.06 m depth, is composed of dolomite marl layers, at the bottom, which progressively changes to the first gypsum sandy subfacies, at the top. The thickness of the sedimentary cycles making up this part of the sequence is usually below the mean cycle thickness. Only punctual cycles are thicker than the average. The second sequence, from 4.06–3.50 m depth, is formed by dolomitic marl layers, at the bottom, and by both gypsum sandy subfacies at the top, although the second gypsum subfacies predominates. The thickness of the cycles presents important oscillations. The third sequence, which is dominated by the calcareous marl and the calcarenite facies, occupies the upper part of the studied profile. Each cycle shows a thickness exceeding the mean cycle thickness. Chronology An accurate chronological framework was established in order to interpret the paleoclimate reconstructions. Radiocarbon dates from the Salines sequences can be subdivided into two main groups: the first group includes the radiometric dates situated in the 14C plateaux. The 10,120 ± 60 date is located in the 10,000 plateau (Ammann & Lotter, 1989; Stuiver et al., 1993) while the 8,810 ± 60 date is situated in the 8,800 plateau

From 2.85–5.50 m depth, three main palynological zones were established in accordance with the percentages of the most important ecological taxa (Figure 5). The first zone (B1A), located between 5.50 and 4.30 m depth, is characterized by low arboreal pollen values (approx 30%). The AP is dominated by Pinus, although Cupressaceae taxa and xerophytic mesic trees progressively increase their percentages. In the shrub taxa, Ephedra distachya-type predominates in the lower part of this zone whereas Ephedra fragilis-type is the main taxa in the upper part. The herbs are mainly composed of Chenopodiaceae and Artemisia. Ruppia is also found in the upper part of zone B1A. This palynological assemblage can be interpreted as an open-steppic landscape with sparse pines and Cupressaceae, and rare copses of xerophytic taxa. The local vegetation is mainly dominated by halophyte taxa. The second zone, labelled A2B, is between 4.30 and 3.15 m depth. It is characterized by the replacement of Chenopodiaceae and Artemisia taxa by Poaceae and Lygeum. The AP percentages show a tendency to increase. Throughout this zone Lygeum, Cupresaceae, mesic and xerophytic taxa reveal a relay pattern that suggests fluctuating environmental conditions. The proliferation of Cupressaceae and the regional wood of xerophytic and mesic taxa suggest the presence of a park landscape. In the upper zone (A2A), from 3.15–2.80 m depth, the tendency of AP percentages to increase persists. The arboreal pollen is characterized by a fall in Cupressaceae values, by stabilization of mesic tree percentages, and by an increase in Pinus and xerophytic trees. The Pistacia shrub reaches its maximum values. This zone corresponds to a park landscape, with regional xerophytic and mesic woody and pine forests. This vegetal assemblage reflects a well established semiarid Mediterranean environment.

456

Figure 4. Sedimentarion rate model established according to the 14C AMS datations (continuous line) fitted in accordance with a polynomial adjustment (discontinuous line).

Biotic assemblages From 5.50–2.85 m depth, two main biota zones were distinguished (Roca & Julià, 1997). Only cysts of the anostracan Artemia salina, diatoms, such as Nitzschia rostellata, and decapoda eggs (cf. Palaemonidae) were found in the first zone located between 5.50 and 3.80 m depth. The second zone, between 3.80 and 2.85 m depth is characterized by a highly fluctuating biotic content, composed of the ostracod Cyprideis torosa, gastropods belonging to the genus Hydrobia, the foraminifer Ammonia becarii, and charophytes (Lamphrothamium sp. and Chara minor). The intervals with a low density of biotic assemblages are either barren or only contain forms resistant to desiccation such as cysts of the anostracean Artemia salina. Discussion Endorheic lakes constitute depositional systems that are highly sensitive to the hydrological balance (Garrels &

Mackenzie, 1967; Hardie & Eugster, 1970; Dargam & Depetris, 1996). Our results show that in dry periods, when the Salines Lake undergoes a retraction, the brines precipitate sulphates and chlorures, whereas, during infilling periods, detrital minerals, such as quartz and illite are common and diluted waters precipitate calcite. Thus, the lacustrine ecosystem of Salines seems to be governed by this sedimentological dynamics and the cycles are a clear response to the expansion and retraction of the lacustrine environment. Likewise, the ecosystem response reflected by the different proxies used in our reconstruction allow us to obtain more details about water availability dynamics. The major event recorded at 3.50 m and evidenced by the abrupt mineralogical change from dolomite and gypsum to calcite and by the variation in the cyclical pattern marks a hydrological change in the lacustrine basin, resulting in an increase in water availability and relative stabilization of the water supply. A similar trend of changes is observed in aquatic biotic assemblages and pollen spectra (Figure 5). However, these proxies indicate a number of prior changes in the

Figure 5. Synthetic pollen profile of the Salines site. See the text for details.

457

Figure 6. Mineralogy, pollen, cycle thickness and total biota content during the Late Glacial to Early Holocene transition in the Salines sequence.

458

459 scenario, which could have an important predictive value. In dry Mediterranean environments, the main constraint in vegetal productivity is water availability with the result that pollen data reflect the composition and structure of the vegetal landscape. In Salines an increase in pollen percentage of mesic trees is recorded at 4.25 m depth, which corresponds to 10,000 years BP in accordance with the sedimentation rate model. The first biota remains are located at 3.80 m depth, coinciding with an initial increase in cycle thickness. These facts coincide with a date modeled by the sedimentation rate of 9,500 years BP. The advanced response of the vegetation and biotic aquatic assemblages versus the abrupt mineralogical change corresponds to a gradual increase in water availability in the lacustrine system (Figure 6). First of all, the vegetation reflects an increase in moisture, on a regional scale, approx 1000 years prior to the abrupt mineralogical change. Next, lacustrine biotic communities which require a relative permanent water body (Hermann, et al. 1983) developed 500 years before the mineralogical change. This scenario is also corroborated by an increase in the thickness cycles. After this abrupt mineralogical change occurring at 3.50 m depth (9000 years BP, according to the sedimentation rate model), vegetation, biotic assemblages, mineral phases and cycle thickness show a similar hydrological pattern. The expansion of mesic pollen taxa, the presence of lacustrine biotic remains and thick calcitic cycles suggest a phase with a predominant infilling process with a more permanent water body. This behaviour lends support to the previous observations suggesting that Boreal chronozone in southern Spain corresponds to the most humid period during the Holocene (Julià et al., 1994). In conclusion, the climate change at the start of the Holocene is more complex in bordering areas such as Salines than in Atlantic seaboard zones.

Acknowledgements The authors are deeply indebted to Prof. G. Seret (Université Catholique de Louvain) for performing the drilling and the thin sections. We would also like to thank George von Knorring for improving the final version of the manuscript. Financial support was provided by UE project (EV5V-CT91-0037) and CICYT project (CLI95-1905-03).

References Allen, J. R. M., B. Huntley & W. A. Watts, 1995. The vegetation and climate of northwest Iberia over the last 14000 yr. J. Quat. Sci. 11: 125–147. Ammann B. & A. F. Lotter, 1989. Late-Glacial radiocarbon- and palynostratigraphy on the Swiss Plateau. Boreas 18: 108–126. Bard, E., M. Arnold, J. Mangerud, M. Paterne, L. Labeyrie, J. Duprat, M.-A. Mélières, E. Sonstegaard & J.-C. Duplessy, 1994. The North Atlantic atmosphere-sea surface 14C gradient during the Younger Dryas climatic event. Earth Planetary Sci. Lett. 126: 275–287. Burjachs, F., 1991. Palinologia dels dòlmens de l’Alt empordà i dels dipòsits quaternaris de la cova de l’Arbreda (Serinyà, Pla de l’Estany) i del Pla de l’Estany (Olot, Garrotxa). Evolució del paisatge vegetal i del clima des de fa més de 140,000 anys an NE de la Península Ibèrica. Universitat Autònoma de Barcelona. CLIMAP, 1984. The last Interglacial Ocean. Quat. Res. 21: 123– 224. Chung, F. H., 1974a. Quantitative Interpretation of X-ray Diffraction Patterns of Mixtures. I. Matrix-Flushing Method for Quantitative Multicomponent Analysis. J. Applied Crystallogr. 7: 519– 525. Chung, F. H., 1974b. Quantitative Interpretation of X-ray Diffraction Patterns of Mixtures. II. Adiabatic Principle of X-ray Diffraction Analysis of Mixtures. J. Applied Crystallogr. 7: 526– 531. Dargam, R. M. & P. J. Depetris, 1996. Geochemistry of waters and brines from the Salinas Grandes basin, Córdoba, Argentina. II. Gypsum dissolution-calcite precipitation, and brine evolution. International. J. Salt Lake Res. 5: 81–101. Duplessy, J.-C., G. Delibrias, J. L. Turon, C. Pujol, & J. Duprat, 1981. Deglacial warming of the northeasten Atlantic ocean: Correlation with the paleoclimatic evolution of the european continent. Palaeogeogr. Palaeoclimatol. Palaeoecol. 35: 121– 144. Duplessy, J.-C., L. Labeyrie, M. Arnold, M. Paterne, J. Duprat & T. C. E. van Weering, 1992. Changes in the surface salinity of the North Atlantic Ocean during the last deglaciation. Nature 358: 485–488. Fairbanks, R. G., 1989. A 17,000-year glacio-eustatic sea level record: Influence of glacial melting rates on the Younger Dryas event and deep-ocean circulation. Nature 342: 637–642. Garrels, R. M. & F. T. Mackenzie, 1967. Origin of the chemical composition of some springs and lakes. In Gould R. F. (ed), Equilibrium concepts in natural water systems. Advances of Chemistry Series. American Chemical Society, American Chemical Society: 222–242. Gasse, F. & E. Van Campo, 1994. Abrupt post glacial climate events in West Asia and North Africa monsoon domains. Earth Planetary Sci. Lett. 126: 435–456. Goeury, C. & J. L. de Beaulieu, 1979. Á propos de la concentration du pollen á l’aide de la liqueur de Thoulet dans les sédiments minéraux. Pollen et Spores XXI: 239–251. Grimm, E., 1987. A Fortran 77 Program for Stratigraphically Constrained Cluster Analysis by the Method of Incremental Sum of Squares. Computers Geosci. 13: 13–35. Hardie, L. A. & H. P. Eugster, 1970. The evolution of closed-basin brines. Mineralog. Soc. Am. Spec. Paper 3: 273–290.

460 Herman, P. M. J., C. Heip, & G. Vranken, 1983. The production of Cyprideis torosa JONES, 1850 (Crustacea, Ostracoda). Oecologia 58: 326–331. Julià, R., J. F. W. Negendank, G. Seret, A. Brauer, F. Burjachs, Ch. Endres, S. Giralt, J. M. Parés & J. R. Roca, 1994. Holocene climatic change and desertification in the Western Mediterranean region. Terra Nostra 1: 81–84. Kromer B. & B. Becker, 1992. Tree-ring 14C Calibration at 10,000 BP. In: Bard E. & W. S. Broecker (eds.) The Last Deglaciation: Absolute and Radiocarbon Chronologies I 2: 3-11. Lowe, J. J. & Members, 1995. Palaeoclimate of the North Atlantic seaboards during the last Glacial/Interglacial transition. Quat. Int 28: 51–61. Mangerud J., S. T. Anderson, B. E. Berlung & J. J. Donner, 1974. Quaternary stratigraphy of Norden, a proposal for terminology and classification. Boreas 3: 109–128. Montserrat, J. M., 1992. Evolución glaciar y postglaciar del clima y la vegetación en la vertiente sur del Pirineo: Estudio palinológico., Monografías del Instituto Pirenaico de Ecología. C.S.I.C. 147 pp. Peñalba, M. C., M. Arnold, J. Guiot, J. C. Duplessy & J. L. de Beaulieu, 1997. Termination of the Last Glaciation in the Iberian Peninsula Inferred from the Pollen sequence of Quintanar de la Sierra. Quat. Res. 48: 205–214. Pérez-Obiol, R. & R. Julià, 1994. Climatic Change on the Iberian Peninsula Recorded in a 30,000-Yr Pollen Record from Lake Banyoles. Quat. Res. 41: 91–98. Pons, A. & M. Reille, 1988. The Holocene and upper Pleistocene pollen record from Padul (Granada, Spain): A new study. Palaeogeography, Palaeoclimatol. Palaeoecol. 66: 243–263. Queralt, I., R. Julià, F. Plana & J. L. Bischoff, 1997. An hydrous Ca-bearing magnesium carbonate in recent saline lake

sediments (Salines Lake, Spain). Am. Mineralog. 82: 812– 819. Roca J. R. & R. Julià, 1997. Late Glacial and Holocene climatic changes and desertification expansion based on biota content in the Salines lacustrine sequence, Southeastern Spain. Geobios 36: 823–830. Ruddiman, W. F. & A. McIntyre, 1981. The North Atlantic ocean during the last deglaciation. Palaeogeogr. Palaeoclimatol. Palaeoecol. 35: 145–214. Sánchez-Toribio, M. I., F. López-Bermúdez & F. Amor-García, 1990. Análisis de EPAN regional: Mapas de evaporación. Manejo del riego. Fichas Agronómicas. Programa de asesoramiento en riegos. Consejería de Agricultura, Ganadería y Pesca de la Región de Murcia., Consejería de Agricultura, Ganadería y Pesca de la Región de Murcia. Sarnthein, M., E. Jansen, M. Weinelt, M. Arnold, J. C. Duplessy, H. Erlenkeuser, A. Flatøy, G. Johannessen, T. Johannessen, S. Jung, N. Koc, L. Labeyrie, M. Maslin, U. Pflaumann & H. Schulz, 1995. Variations in Atlantic surface ocean paleoceanography, 50-80°N: A time slice record of the last 30,000 years. Paleoceanography 10: 1063–1094. Stuiver M., A. Long, R. Kra, & J. M. Devine, 1993. Radiocarbon 35(1). Thompson, L. G., E. Mosley-Thompson, M. E. Davis, P. N. Lin, K. A. Henderson, J. Cole-Dai, J. F. Bolzan, K. B. Liu, 1995. Late Glacial Stage and Holocene Tropical Ice Core Records from Huascarán, Peru. Science 269: 46–50. Wansard, G., 1996. Quantification of paleotemperature changes during isotopic stage 2 in the La Draga continental sequence (NE Spain) based on Mg/Ca ratio of freshwater ostracods. Quat. Sci. Rev. 15: 237–245. Watts, W. A., 1985. A long pollen record from Laghi di Monticchio, southern Italy: A preliminary account. J. Geolog. Soc. London 142: 491–499.