The Iberian peninsula

Figure 2.3 shows the Iberian peninsula.

2.2.3.a CONTEMPORARY CLIMATE

The climate of the Iberian peninsula is decided by various factors. In the first place, its position on the southwestern flank of Europe, between 36° N and 44° N, causes it to be influenced periodically by the high sub-tropical and the high polar pressure zones. It lies between the Atlantic Ocean and the Mediterranean Sea and is an extension of the Euro-Asian continent; consequently, it forms a passageway for air masses, either of sub-tropical marine and continental (Sahara) or of marine polar and Mediterranean origin (seldom of polar continental origin).

The special mountainous morphology of the peninsula (average altitude of the Spanish Meseta 660 m above mean sea level) also influences the climate. The high Meseta obstructs the free passage of the Atlantic air masses of the westerlies system. Though the valleys of the large transversal rivers enable air masses to penetrate inland, they do not reach the Mediterranean coastal areas.

In general, it can be said that the mountainous character and the size of the peninsula minimize the regulating influence of the seas surrounding the peninsula and thus contribute to its continental type climate, which is distinguished by high and low temperatures during the summer and winter, respectively.

The climate in the southern Iberian peninsula is similar to that found in Sicily, especially in the Andalusia plain. The average temperature is 12-13 °C in January and 27-28 ° C in July; annual precipitation is 500 to 700 mm. Along the length of the Mediterranean coast (the "Spanish Levant"), precipitation is less

Iberian Peninsula Map
Fig. 2.3. Map of the Iberian peninsula.

(300-350 mm per year). These characteristics show that this region is a transitional area towards the hot and dry regions of northern Africa (Capel Molina, 1981).

Rodo et al. (1997), analyzing data from the Iberian peninsula, Balearic Islands and northern Africa, have found a relationship between the variations in seasonal rainfall during the present century and the North Atlantic oscillation (NAO) and the ENSO. While the NAO influences most of the peninsula except its eastern part, the opposite was found for the ENSO.

2.2.3.b PALEO-GEOMORPHOLOGY

A study of the terraces of the Jarama river, a tributary of the Tajo river, east of Madrid (Alonso and Garzon, 1994) found evidence for a process of river incision in the transitional period between the Late Pleistocene and the Holocene. After the incision stage, a period of stability occurred, long enough to allow the development of a soil of green clay rich in organic matter and the establishment of a forest of phreato-phitic species (such as Alnus glutinosa and Ulmus spp.). In one sector, the wood was dated and was found to be from 6kaBP. (This piece of wood may be reworked.) In another sector, the date was 3kaBP. A conglomerate showing an erosive base and bearing evidence of multi-episodic infilling, which apparently indicates high and low regimes of flow, overlaid this layer. The conglomerate itself was also overlaid by silts or clays, as well as by lenticular layers of conglomerates and sands. This sequence represents a meandering river of medium-to-low sinuosity, with a bed load composed mainly of gravels. The date of the middle part of the gravel sequence was found to be 2.4 ka BP, while the age of a clay layer from a secondary channel on top of the sequence dates from0.4kaBP.

In conclusion, the history of the Jarama river system during the Holocene can be divided into three stages.

1. An incision stage during the transitional Pleistocene-Holocene period;

2. A stable soil-forming phase prior to 6 ka and lasting to 3 kaBP;

3. An alluvial aggredation phase from c. 2.5 ka to 0.4kaBP.

On the basis of similar incision stages in Italy and Greece, Alonso and Garzon (1994) suggest that the incision is a result of human activity, namely rapid forest felling and resultant alluvation. I am more inclined to suggest a climatic reason, as will be discussed below.

Goytre and Garzon (1996) analyzed historical data of floods of the Jucar river, flowing to the Mediterranean south of Valencia, from the fourteenth century to the present. The frequency curve shows a period of high frequency starting at 1720 AD and reaching a maximum at 1800 AD coming to a low c. 1820 AD and then again reaching a peak in 1870 AD and 1890 AD. They correlated the flood frequency and size curve with precipitation data available since 1860 AD. The increase in flood frequency and size from the second half of the eighteenth century till the beginning of the twentieth century also coincided with an increase in winter flooding.

Zazo et al. (1996) analyzed data from studies of sedimentolog-ical processes in karstic systems, pollen time series and morpho-sedimentological studies from the Iberian littoral zone. They concluded that the Mediterranean region had a cold dry climate during the last glacial period, while the Cantabrian region was cold to cool and humid. During the Younger Dryas cold period, the climate was dry but not cold in the Mediterranean region and cool and humid, but less than today, in Cantabria. To the west of the Gibraltar Strait, the climatic conditions were humid and temperate, both during the last glacial maximum and during the Younger Dryas.

The sea level changes along the southern coasts of the Iberian peninsula during the Holocene have been explored rather intensively (Goy et al., 1996; Lario et al., 1995). The main finding is that most of the coasts of southern Spain were influenced by a process of uplift (or sea retreat) during the upper part of the Holocene, which caused spit-bar formation. During the lower part, from 10 ka BP, the sea was rising, reaching its maximumlevel at c. 6.4kaBP. This was followed by the retreat of the sea and a prograding spit-bar system developed in the areas from which the sea moved back. The first phase of progradation, starting after the c. 6.4kaBP maximum, lasted until c. 4.5kaBP, reaching its maximum at c. 4. ka BP, when a gap in sedimentation occurred. This was followed by a period of retreat of the sea and the formation of spit-bars. A pronounced gap of progradation, accompanied by a pronounced phase of modification of littoral dynamics and evidence for changes of wind direction, from mainly easterlies and north easterlies to westerlies, occurred between 3 ka and 2.75 ka BP. Sea-level retreat and progradation continued until 1.2kaBP, when there was a gap of sedimentation and a period of sea-level rise, which extended to c. 0.7kaBP. Estuaries had a greater fluvial than marine influence at c. 1.0kaBP. From then, a period of retreat continued, with an extraordinary increase in coastal progradation in the littoral zones, reaching its maximum c. 0.5 kaBP.

Two Holocene transgressive phases have been detected along the Cantabrian coast of northern Spain (Altuna et al., 1993). The lower transgression occurred before 5.8kaBP and the upper one after 4.9kaBP. During the transgressive phases, coarse material was deposited on the sandy beach material, while during the regression of the sea, coastal dunes and freshwater environments developed. On these flourished a deciduous forest suited to a temperate and humid climate.

A geomorphological study was carried out in the central Ebro valley by Soriano and Calvo (1987). It revealed two periods of intensive deposition, one post-Roman and pre-Visigoth, and the other post-Medieval. The authors suggest that processes of accumulation of gravel were connected with colder climates, while periods of incision associated with warmer climates.

Another investigation of the middle Ebro river system was carried out by Stevenson et al. (1991). This investigation also extended to the saline marshes (saladas) of the region. It was based on pollen and geochemical analyses of cored material from the saladas, as well as geomorphological investigations. Unfortunately, dated samples are scarce. The general conclusions drawn by Stevenson et al. (1991) are that erosional phases conformed with aridization phases and vice versa. They observed three main periods of "cut and fill": a major erosion phase of the Lower and Middle Holocene, followed by an aggradation phase that started with alluvial material and continued with fine-grained and clay material (Cerezuela unit). Unfortunately, Stevenson et al. (1991) give only one 14C date in their section of a hearth found in the mid part of the exposed section, which is 3815 ± 80 BP. On the basis of the similarity in pollen assemblage found at the depth of 150 to 210 cm in Salada Pequenia, they try to make a correlation to the top layers of the exposed section of Cerzuela unit and claim that it is of the Iberian period and is evidence of a humid climate. This correlation, not based on 14C dating, is rather questionable. However, in addition to this interpretation by Stevenson et al., quite a lot of information can be gained from the pollen and chemical data from the outcrop of the Cerzuela unit at Alcaniz. The layers above the dated one, namely after 3.8kaBP, show an abrupt increase in the pollen of Artemisia spp., Chaenopodiaceae and Gramineae, and a slight increase in Pinus spp., while at the same time, there is an increase in sodium (unfortunately only cations and no anions were analyzed). This may point to a warmer period in which evaporation exceeded precipitation and inflow to the marsh. This agrees with the conclusion from many other investigations, as will be discussed below, that a period of warming up started c. 4kaBP. The increase in Pinus pollen may suggest that the precipitation in this region did not fall below that which is crucial for the existence of a Pinus habitat, while the warmer climate promoted the expansion of this species.

Evidence for the Holocene glacial episode, attributed to the Little Ice Age, was found in the central and Aragonean Pyrenees (Serrano Candas and Agudo Garrido, 1988).

2.2.3.c PALYNOLOGICAL TIME SERIES

Carrion and Dupre (1996) investigated the Late Quaternary vege-

tational history atNavarres, eastern Spain (Fig. 2.4a). The area lies

Present

1000 bp

2000 bl

3000 bp-

4000 bp

5000 bl

6000 bp-

7000 bp

9000 bp-

lo.ooo bp:

Pollen lime series from Castillo de Caltarava

Garcia Anton et al. (1986)*

Levant

PINUS PISTAC1A

ARBOREAL POLLEN

Pollen tune senes from Padul 3

Based on data from Pons and Reille (1988) QUERCUS

Pollen time series from Cantabria

Based on data from Sallas (1992)

Vegetational history at Navanés

Carrion and Dupré(1996)*

PINUS

QUERCUS 90 IOC»

History

Levant

PINUS PISTAC1A

ARBOREAL POLLEN

Pollen tune senes from Padul 3

Based on data from Pons and Reille (1988) QUERCUS

PINUS

QUERCUS 90 IOC»

History

Present

1000 bp

2000 bl

3000 bp-

4000 bp

5000 bl

6000 bp-

7000 bp

9000 bp-

lo.ooo bp:

Iberian Arboreal

Presenl -

1000 bp

2000 bp

3000 bp-

4000 bp-

5000 bp

6000 bp-

7000 bp

8000 BP.

9000 bp-

n.ooo bp

Pollen lime series from Almeria, Spain Based on data from

AR1DUSEUROMKD (1997) __

Levant

History Paleo-climate

PINUS

"J

OLEA

ARTEMISIA

Pollen time series from Minorca, Balearic Isles, Spain Based on data from ARlDUSKUROMfiD (1997)

PINUS E. QUERCUS

Levant

History Paleo-climate

PINUS

"J

OLEA

ARTEMISIA

PINUS E. QUERCUS

Presenl -

3000 bp-

4000 bp-

5000 bp

6000 bp-

8000 BP.

9000 bp-

n.ooo bp

Fig. 2.4. Pollen time series for the Iberian peninsula. 1 Adjusted to scale and streamlined (3-5) points by the running average method.

in the lower meso-Mediterranean belt and has a dry to sub-humid climate. The mean annual temperatures are 15-16 °C and mean annual rainfall is 550 mm. In the assemblage dated as Upper Pleistocene, they found the recognized European sequence of upper Pleniglacial, from 16ka to 15kaBP, characterized by a high percentage of Artemisia spp. (about 19%) while Pinus spp. it is only 50%. This means that continental climatic conditions prevailed during the colder periods of the Pleistocene. This period was followed by the Bolling-Allerod warm period, in which the Artemisia percentage declined sharply (~3%) while the Pinus percentage increased (to 77-89%). From c. 11 ka to 10 kaBP, one can see the influence of the cold Younger Dryas episode, which led to the strengthening of the continental conditions. This is evidenced by an increase in Artemisia spp. (up to 14%), Chenopo-diaceae and Ephedra spp., while Pinus spp. decreased (to 6773%). With the beginning of the Holocene, c. 10kaBP, the climate became warmer. This led to a decline in the percentage of the Artemisia, a relative increase in Pinus (to 67-84%) and the appearance of evergreen Quercus spp. (2-3%). At c. 5.5 kaBP, there was a decrease in Pinus (to 46-61%) while the Quercus spp. increased (reaching up to 14%). At 4kaBP, Quercus reached 32%, while the Pinus percentage was 23-53%. This continued until c. 3 ka BP.

Carrion and Dupre (1996) discussed the reasons for the expansion of the oaks and their replacement of pines and spruces after c. 5 ka BP. They suggest that this change in flora was more a climate change indicator rather than an indication of any impact of human intervention, although, generally speaking, the anthropogenic factor could have had local influence. The data presented by Carrion and Dupre (1996) are shown in Fig. 2.4.

If these authors are correct in believing that climate was the main cause for the changes in the pollen assemblages, one can conclude from the palynological data (even though the dating of the core was sparse) that, c. 5.8 ka BP, the climate was cold, getting warmer towards c. 5 kaBP and again colder towards 4.5 kaBP. It then warmed up again c. 4kaBP.

Carrion and Dupre (1996) correlated their data with those of Menendez-Amor and Florschutz (1961), who analyzed a cored section in the same region. Dating was also sparse for this core, but continuing with the same assumption, one can conclude that c. 6.2kaBP there was a rather warm period. Assuming a similar rate of deposition for the lower part of the section as for its upper part, one can conclude that there was a cold period c. 7.5 kaBP, while between 8 ka and 9 ka BP it was warm, after a cold period about 10 kaBP.

Another study providing an indicator of climate changes during the lower part of the Holocene in the northeastern part of the Iberian peninsula was by Perez-Obiol and Julia (1994), who studied the pollen record in a core from Lake Banyoles, northeast of Barcelona. The record, which started c. 31 kaBP, showed an in-

terstadial event between 30 ka and 27 ka BP, a Pleniglacial Period with minor oscillations that ended abruptly c. 14 ka BP. There is also the Younger Dryas event c. 12kaBP. The pollen diagram indicates that the ameliorating climatic conditions led the deciduous Quercus spp. to flourish from c. 11 kaBP onwards. As Quercus flourished so Pinus spp. diminished, indicating a climate change. The lack of dating in this study makes detailed conclusions regarding climate changes rather difficult, but by correlating the time scaled section for Pinus (calculated by polynomial interpolation on the basis of 14C and U/TH dates) with that of Quercus, one can draw very general conclusions with regard to the climate changes in this region during the lower half of the Holocene. This period started with a cold climate stretching from 10 ka to 9 ka BP (with a short warm interval sometime around 9.5 BP), followed by a gradual warming up, which reached its maximum around 8kaBP. A profound cold period, when Quercus falls and Pinus surges, reached its maximum around 6 ka BP, with an intermediate warm phase c. 5 ka BP.

Pons and Reille (1988) presented a detailed picture of the impact of climate changes on the natural environment in the southern part of Spain for the period from before the last glacial period to the mid Holocene (Fig. 2.4a). The site of the borings is a peat bog at Padul, situated in a long and vast tectonic valley at the eastern foot of the Sierra Nevada. The bog was most probably fed rather constantly by groundwater. The Oldest Dryas, c. 15 ka BP, was marked by an increase in steppe species (Juniperus) at the expense of Pinus, which indicates dry continental conditions. This was followed by a warmer period when Quercus and Pistachia spp. became more abundant. The Younger Dryas cold phase (between dated layers of 12 ka and 10 ka BP, with a hiatus in between) was clearly characterized by a reduction in Quercus and Pistachia and an increase in Pinus. Artemisia and Juniperus also become more abundant: evidence for steppe conditions. At c. 10 ka BP, the warmer climate brought a recurrence of Quercus ilex and Pistachia, while Pinus decreased. At c. 8.5kaBP, there was again a decrease in Pinus, but at the same time there was also a decrease in Quercus spp. and some decrease in Pistachia, in the total arboreal flora and in the Poaceae. By comparison, fern spores showed an abrupt increase, while Juniperus and Artemisia spp. practically disappeared. Although Pons and Reille (1988) considered this to be of local significance, I would suggest the possibility of some kind of a climatic anomaly, for example higher temperatures and the strengthening of summer precipitation. Between c. 8.2ka and 8kaBP, there is an abundance of Quercus with the almost total absence of steppe species, which suggested to Pons and Reille "... that optimal postglacial thermic and humid conditions were then prevailing". The reduction in Q. ilex, Q. suber and Pistachia, together with the increase in Pinus, was seen by Pons and Reille as reflecting conditions of more open regional forest formation, while I consider it as a sign of a colder phase. Just after 8 ka BP, there was another increase in Quercus spp., Pistachia and Olea, while Pinus slightly decreased, which implies a Mediterranean climate. During the period immediately following, 7.5 ka to 6.5kaBP, Olea, Pistachia and evergreen Quercus spp. decreased, while Pinus was more abundant, which may betoken a colder climate (although deciduous oak increased). At 6.5 kaBP, Olea returned, Quercus and Pistachia increased, while Pinus decreased. This trend reversed c. 6 ka BP.

Changes in the ratio of Quercus and Pistachia to Pinus can also be found in the palynological time series from cores at Castillo de Caltarava, in the sediments in the marshes close to the Guadiana river (Garcia Anton et al., 1986; Fig. 2.4a). In the general section, one can see that there are changes in the ratio of Pinus to Pistachia. These changes are climatic indicators because, as shown by Pons and Reille (1986), high comparative levels of Pinus are characteristic for the glacial periods, and vice versa, for the Granada region. The scarcity of 14C dating does not permit a good correlation with the other curves, but from the dates available, one can see that c. 6.2kaBP, there was a high ratio of Pinus to Pistachia, which might correspond to a colder climate. This was followed by a short period of decrease in the ratio, which may correspond to a warmer phase. At c. 5 ka BP, there was an increase in the ratio of Pinus to Pistachia, which may represent a cold and humid period, and at c. 4 ka BP, there was a strong decrease in Pinus and increase in Pistachia, which may represent a warm and dry period. Approaching the layer dated c. 1.7kaBP, there was another increase in Pinus, which was interrupted about 2 ka BP (not dated) by a layer of about 10 cm devoid of pollen. I would suggest that this is equivalent to the period of high precipitation during the early Roman period in which the marshes turned into a flowing river, which caused the oxidation of the sediments. At c. 0.5 ka BP, there was an increase in Pinus and reduction in Pistachia, which most probably can be correlated with the Little Ice Age.

Dupre et al. (1996) reconstructed the paleo-environment of the lake of San Benito in central eastern Spain, near Valencia. They found a clear predominance of pines during the Upper Pleistocene when lagoonal conditions prevailed. Up to the Middle Holocene, fluvio-aluvial conditions prevailed. At the top of the section, c. 1.4kaBP, the pollen profile showed the regeneration of the tree cover, mainly Quercus, which the authors believed was the result of the decline of intensive agriculture after the fall of the Roman Empire.

Ruiz Zapata et al. (1996) reconstructed the paleo-climates of the central western part of the Iberian peninsula from an east-west cross section based on four cores drilled in intra-mountain depressions. Two of the cores were drilled in glacio-lacustrine deposits, while two comprised periglacial deposits. Pollen assemblages for the central zone showed that, between 8 ka and 7 ka BP, Betula showed an increasing trend, while Pinus was more or less stable and Quercus diminished, disappearing c. 7kaBP. Quercus reappeared c. 4.5 kaBP. At this time, Pinus reached a minimum, while Betula reached a peak and from then on diminished, until it disappeared altogether at 2.3 kaBP. At 4.5 kaBP, Pinus was at a minimum; it reached a peak at c. 3.2kaBP, with a certain minimum at 1.7kaBP and another between 0.8 ka and 0.5 kaBP. Olea appeared c. 2.5 ka BP, disappeared c. 2 ka BP and reappeared c. 1.7kaBP, to disappear again for a short while between 1.5ka and 1.3kaBP. From this period on, Olea spp. increased. In the eastern part of the cross section, Pinus spp. were at a maximum at c. 2 ka BP, decreased c. 1.3 kaBP, showeda short peak c. 1.2kaBP and slowly decreased to a minimum c. 0.5 ka and c. 0.25 ka BP. By comparison, Quercus spp. were at a minimum in the eastern part of the section from 1.4 ka to c. 1 ka BP and reached a maximum c. 0.7kaBP.

A multidiscipline investigation was carried out on the deposits in a cave near the seashore on the central eastern coastline of Spain (Badal et al., 1993). The sediments, about 3.5 m thick, contained artifacts of the Iberian Neolithic and Bronze Ages. The lowermost layer was dated at 7540 ± 140 BP, but this seems to be too early a date as the archaeological remains of Ceramic Neolithic I A, are only 7 ka years old. The other 14C dates more or less corresponded with the archaeological findings. The pollen assemblage was rather poor in arboreal and rich in non-arboreal taxa. The major tree component was Pinus, followed by Quercus (Q. ilex and Q. faginea). The interchange between these trees seems to be a climatic indicator, according to Dupre et al. (1996), who divided the section into four zones. Zone A (350-260 cm) contained little tree pollen, mainly Pinus, and was considered as created during a dry period. Zone B (255-185 cm) was rich in Quercus pollen and was considered to be the most humid part of the sequence. Zone C (175-115 cm), in which Pinus replaces Quercus, was considered to be influenced by human factors. However, the same reason was given for the reciprocal trend in zone D, where Quercus replaces Pinus. As already discussed, I am more inclined to believe that the interchange between Pinus and Quercus (and Pistachia) is a reaction to climate change, rather than reflecting anthropogenic impact.

As past of a multidiscipline program (ARIDUSEUROMED, 1997), a group of Spanish palynologists have investigated the pollen paleo-records for the coast of Almeria, in southeastern Spain, which is semi-arid, with an annual mean precipitation of c. 250-350 mm and mean annual temperature of 18-21 °C.

In a core from the salt marsh of San Rafael in this region, the layers were mostly clay to 10.75 m, peat from 10.75 to 11.75 m, clay at 11.75-14.50 m, argillaceous slime 14.50-18.00 m and gravel at the bottom layer to the depth of 19.00 m. The group studying this core interpreted the pollen assemblages in the following way.

Layers deposited between 18 ka and 15 ka BP (bottom to 18.00 m) contained mainly arboreal pollen, composed of Olea, deciduous and non-deciduous Quercus and Pinus, evidence of a relatively warm and humid environment. From 15 ka to 7kaBP (18.00-13.50 m), the assemblage showed a decline in arboreal pollen and an increase in steppe type vegetation. There are indications that during this period there was a decrease in temperature as well as a possible decrease or change in the distribution of the precipitation. From 7ka to 4.5kaBP (13.50-6.50m), there was a decline in steppe vegetation together with an increase in arboreal pollen and certain shrub taxa. This shows a climate optimum, most probably warm and humid. From 4.5 kaBP upwards, there were indications of the encroachments of arid conditions. This was evidenced by an increase in steppe vegetation and a decrease in arboreal pollen. Deciduous Quercus disappeared and evergreen Quercus and Olea declined. The uppermost part of the core, presumably from c. 0.1 ka to 0.2kaBP, showed an increase in arboreal pollen, including Pinus, Quercus and Olea, and a decrease in Artemisia, which may be a result of anthropogenic as well as natural processes.

The investigation of the paleo-biotic assemblage and lithology of cored sediments from the bottom of Laguna de Medina, near Cadiz in southwestern Spain, gave a detailed paleo-environmental history of the lower part of the Holocene (Reed et al., 2001). These cores suggest that there was a dry phase from c. 8 ka to 7.2 ka BP and a humid period from c. 6.9kato c. 6.7kaBP (calibrated (cal.): 14C dates given as "absolute" dates by calibrating either locally, with varved lake deposits or tree rings, or generally, with a general calibration curve).

Based on climate changes observed in other parts of the Mediterranean, and considering that the climate of the Almeria coast is Mediterranean, I would suggest a somewhat different conclusion with regard to the climate changes during the Holocene. It is possible that the peak in steppe flora, especially Artemisia, and the strong decrease in arboreal flora, especially Pinus and deciduous Quercus, between c. 10 ka and 7 ka BP was not caused by a decrease in temperature but rather by a relatively warm and dry period. When there was an increase in arboreal flora and decrease in steppe flora (7ka to c. 4.5 kaBP), the region enjoyed a mainly cooler and more humid climate. This interpretation is supported by the accumulation of peat layers and higher charcoal percentage in the layers deposited during this period. At c. 4.5 BP, there was a considerable increase in temperature, which caused another increase in the steppe flora and decrease in arboreal flora.

In the same report (ARIDUSEUROMED, 1997), one can also find palynological data from the island of Minorca. The climate of the southern part of the island, where the core was taken, is Mediterranean semi-arid, with an annual precipitation of 450 mm. The four to five summer months are dry. The bottom of the section dated to c. 8 ka BP and the floral assemblage was characterized by the dominance of Buxus and Corylus (taxa that are absent today in Minorca) and high values of Juniperus, Ephedra and Quercus (deciduous and evergreen). High values of Typha indicated the presence of coastal marshes. At c. 6 ka BP, there was a pronounced change: Olea appeared while Buxus and Corylus declined. Some time before 5kaBP, Buxus and Juniperus disappeared, while Olea increased to reach a peak c. 4.5 kaBP, after which it decreased a little. In the uppermost zone, starting at 4.5kaBP, Pistachia increased, and there was a rise in Chenopodiaceae, which is evidence for the spread of coastal marshes, most probably because of a rise in sea level.

A diagram summarizing the total arboreal pollen frequency of the Minorcan site shows a rather low frequency in the lower part, from c. 8 ka to c. 7 ka BP, followed by an increase between c. 6 ka and 5 kaBP and a strong decrease c. 4.5 kaBP.

A rather recent palynological study in south-central Spain (Carrion et al., 2001) indicated the ecological changes during the Holocene over a region that spreads along the boundary between semi-arid plateau and mountain environments. Pinus dominated from c. 9.7 ka to 7.5 ka BP (cal.), which the authors believed was a consequence of the relatively dry climate and natural fires. From c. 7.5 ka to 5.9 kaBP (cal.), there was a moderate invasion by Quercus as a result of increasing moisture and temperature. From c. 5.9 ka to 5.0 ka BP (cal), Pinus was replaced by deciduous Quercus, as well as by Corylus, Betula, and Alnus, etc. From c. 5.0ka to 1.9kaBP (cal), the Mediterranean type of forest including Artemisia dominated. From c. 1.9ka to 1.1 kaBP, Pinus became dominant and from 1.6kaBP onwards, human impact became influential. Although it is difficult to correlate this section with the Mediterranean climate change timetable because each one of the intervals expands over more than one division of the Holocene, the impact of climate and not human activity on the vegetation during most of the Holocene is stressed by the authors.

Yll et al. (1995) produced a synthesis of the history of the vegetation landscape of the eastern part of the Iberian peninsula and Balearic Islands during the Holocene. They found two fundamental points of climatic impact on the natural environments that defined periods of accentuation of aridity. These occurred at 6 ka and at 4kaBP.

Ruiz Bustos (1995) investigated the climatic conditions during the Last Ice Age by comparing the mammal remains from this age with the mammal population of the present. His main conclusion was that the Spanish climate was cold and dry during the Wurm. He cites Riquelem Cantal (1994), who investigated eight Bronze Age sites in southern Spain dating from the lower half of the second millennia BC (i.e., between 5 ka and 4.5 ka BP, equivalent to the Early Bronze Age of the Levant). The faunal assemblage indicated a cold and dry climate during this period.

Information with regard to changes in aridity during the Holocene has been obtained through the investigation of 13C/12C ratios in grain cereals collected in archaeological sites in Catalonia (northeastern Spain) and Andalusia (southeastern Spain) (Araus et al., 1997). Lower values of d12C %o, which show higher water use efficiency, resulting from reduced water supply and/or increased temperature, were consistently found in eastern Andalusia than in Catalonia throughout the period ranging from Neolithic to Iron Age. According to these authors, this shows that Andalusia was drier throughout this period. The data for Catalonia were available from c. 6.5 kaBP (Iberian Neolithic) and showed a gradual decrease of values of d12C % and a tendency to more arid conditions extending through the periods for which samples were available: 4.2kaBP (Iberian Chalcolithic Bronze) to 2.5kaBP (Iberian Iron Age). There is a small increase at 0.8 kaBP, and a strong decrease towards the present.

Sallas (1992) used a pollen time series in Cantabria to analyse climate changes during the Holocene (Fig. 2.4a). He distinguished three main climatic phases: a cold and dry period from 10.2ka to 7 ka BP, a warm dry period from 7 ka to 5 ka BP and a colder period from 5 ka to the present. Using the Pinus to Quercus ratios as a climate change indicator, one can suggest a more detailed division of the Holocene based on his data than that suggested by Sallas (1992; see his Fig. 5). This more detailed division should not be overemphasized as control of dating intervals is lacking. However, it is worthwhile noting that the highest ratios of Pinus to Quercus at c. 7 ka BP and at 5 ka BP may be evidence for a colder climate. Sallas's data appear to indicate a warm phase starting c. 4.5 kaBP and continuing until c. 3.5 ka BP, followed by a colder phase from c. 3kaBP to 2.3kaBP.

An evaluation of the changes in precipitation during the last 20 ka years was carried out by Igor Parra (1994), who investigated the pollen assemblages and oxygen isotopes of two marine cores, one in the Mediterranean off the coast of southeast Spain (SU 8103) and the other in the Atlantic (SU8113) off the southern coast of Spain, northeast of the Straits of Gibraltar, and two continental cores, one on the northeastern coast of Catalonia and the other on the western part of the island of Majorca. The paly-nological analysis of the cores was based mainly on the variations of the four species of Quercus and on the comparison of the abundance of each of these types with other pollens, mainly Pinus, Cedrus, Artemisia and Ephedra. Data from the data bank of pollen carried by air between Catalonia and Andalusia was used to determine the ratio of pollen as a function of climate and precipitation. The ratios of Pinus to Quercus were mainly used. These findings, correlated with the data from the cores, gave rise to the following conclusions. According to the Atlantic core, the most humid periods were at 12 ka, 9 ka and 3 ka BP, while according to the Mediterranean core, they occurred c. 12 ka, 9 ka and 7kaBP. The continental cores indicated that the most humid periods were c. 5 ka and 3 ka BP, while there was a pronounced period of aridization around 4kaBP. Parra (1994) recognized an anthropogenic influence from c. 6 ka BP but maintained that it did not influence the imprint of climate change on the pollen ratios.

By comparison Van den Brink and Janssen (1985) did interpret the pollen data obtained from a core of a small pond at an altitude of 1600 m in the Serra de Estrela in Portugal in anthropogenic terms. However, they did find that the Quercus-Betula forest was destroyed by fire at 4.3 kaBP and replaced by heath, while c. 3.2kaBP, there was a temporary regeneration of Betula, corresponding to climate changes suggested by other authors at other sites. It is possible, therefore, that their anthropogenic conclusions are overemphasized. The recent destruction of the forest at 0.85kaBP may well have been connected with human activities.

A more balanced approach, with regard to the influence of climate change versus human activity for the same region, is demonstrated by van der Knaap and van Leeuwen (1995). They found the following five stages.

1. From c. 10.4ka to 8.7kaBP, the late glacial steppe changed into a xerothermic forest under warm and rather dry conditions.

2. From c. 8.7 ka to 5.7 ka BP, the climate became moister and cooler, and from c. 8.2kaBP, the forest changed from xero-thermic to mesothermic. The anthropogenic activity started to show but played a minor role.

3. From c. 5.7 to 3.2BP, the area covered by forests was hardly affected by human activity, yet local overgrazing with soil erosion started c. 4.5kaBP.

4. From c. 3.2 ka to 1 ka BP, large-scale deforestation occurred, with regeneration phases in response to human activity.

5. From 1 ka BP to the present, anthropogenic activities have caused the forest to disappear. Human activity is claimed to be the cause of the influx of pine pollens.

Although this analysis is less anthropogenic than that of Van den Brink and Janssen (1985), it is nevertheless likely that climatic, rather than anthropogenic, reasons should be considered for the deforestation c. 4.5kaBP (i.e., a warm period) and for the pine increase c. 0.5 BP (Little Ice Age).

Although it is clear that much better chrono-stratigraphy of the Holocene of the Iberian peninsula would have been obtained with more 14C and other types of dating, some conclusions can still be reached by following well-established guide stratigraphic horizons.

One clearly demarkated phase in various pollen time series is the chronozone of c. 4 ka BP. The sections in Padul 3 (Pons and Reille, 1988; Fig. 2.4.a) and Cala en Porter in Minorca (ARIDUSEU-ROMED, 1997; Fig. 2.4b) terminate at this period, while the profile in San Rafael in Almeria (ARIDUSEUROMED, 1997; Fig. 2.4b) shows an abrupt reduction in Quercus and Olea, and an increase in Artemisia. In the section at Castillo de Caltrava (Garcia Anton et al., 1986; Fig. 2.4a), one can see a reduction in Pinus and total arboreal pollen, with a relative increase of Quercus.

Similarly, the profile in Cantabria (Sallas, 1992; Fig. 2.4.a) shows an abrupt decrease of Pinus and a certain increase in Quercus. Investigation of the middle Ebro river system (Stevenson et al., 1991) showed that, in the layers immediately after 3.8 ka BP, there was an abrupt increase in the pollen of Artemisia, Chaenopodi-aceae and Gramineae, and a slight increase in Pinus. At the same time, there was an increase in the sodium content of the salina, all of which most probably indicate a warmer period.

Information on special Christian ceremonies shows that during the years 1675 to 1715 AD, namely Late Maunder Minimum, which belongs to the Little Ice Age, there were prayers regarding too much precipitation; that is, there were too frequent passages of low-pressure systems over the peninsula (Barriendos, 1997).

2.2.3.d REGIONAL CORRELATION

Geomorphological data show sea-level changes along the southern coasts of the Iberian peninsula. The first phase of progradation, starting after the c. 6.5kaBP pollen maximum, lasted until c. 4.5kaBP and reached its maximum at c. 4kaBP, when a gap in sedimentation occurred (Goy et al., 1996).

One may conclude that there was indeed a severe change to a warmer climate, spelling dryness, at 4kaBP, at least in the southeastern part of the Iberian Peninsula. This change was global, as it coincided with a high sea level.

This is also the opinion of Jose S. Carrion Garcia (personal communication, 1997), who thinks that increased dryness could have provoked many of the peat sites to stop peat formation. This possibility was also noticed in the thesis by Parra (1994). Perhaps the Azores high pressures had moved latitudinal and, while the southwest was still affected by the southwesternly storms, their influence in the southeast became negligible.

If 4 ka BP is taken as a correlation chronozone, then, despite the scarcity of dating, the re-examination of the Iberian time series suggests that the three millennia prior to this chronozone contained longer colder and less-arid periods. Such a climate was also seen c. 3 ka BP, that is, after the warm period that started c. 4 ka BP came to an end. One can also claim that the climate was cooler c. 2 ka BP, as the section at Castillo de Caltrava (Garcia Anton et al., 1986; Fig. 2.4a) showed an increase in Pinus and decrease in Pistachia. As mentioned earlier, the influence of the Little Ice Age can also be discerned. A more detailed stratigraphy of the Holocene of the Iberian peninsula cannot be attempted until more dated profiles are available. In general, it can be said that the changing ratios of pollen, as can be seen in the rather detailed pollen time series, indicate many more changes in climate than those envisaged by the Blytt-Sernander division of the Holocene. However, much more detailed dating is required before a new division of the Holocene in the Iberian Peninsula could be considered.

Correlating the impact of the climate changes during the Holocene along the western shores of the Mediterranean (Goy et al., 1996) with those to the east, one can see that the influence of the most pronounced changes was similar. Thus, during the part of the Neolithic when the climate was warm, the sea level came up, while during the Chalcolithic and EB, it mainly receded. The MB warm period was marked by an ingression of the sea, which reached a climax at 4 ka BP. During the Lower Iron Age, which was cold, there was again a regression, the same being true for the Roman period and the Little Ice Age. However, during medieval times (the Arab period, which was warm), the sea advanced inland.

The question, which has to be answered now is, what was the impact of these changes on the hydrological cycle? To answer this question, it is first important to confirm that the changes observed in the time series are indeed a result of climate changes, rather than anthropogenic. Second, it is important to confirm the climatic interpretation of the palynological time series with regard to changes in temperatures and humidity, for example the pine/oak ratio, and then its chronological correlation with the geomorpho-logical observations. Needless to say, the scarcity of dates in the profiles in the Iberian peninsula makes this rather difficult and also makes it hard to draw parallels with the continental section of the Levant for verification. Taking these problems as constraints rather than obstacles, it is suggested that the most obvious key horizon be chosen, in this case the 4kaBP chronozone, which was prominent in both the Iberian peninsula and the Levant. As this was a conspicuously warm period, one can draw conclusions about its impact on the hydrological cycle and, in general, one can conclude that opposite phenomena occurred during a cold period. As most of the geomorphological data discussed in this chapter lead to the conclusion that there is evidence of a drier climate c. 4kaBP, one can reach the general conclusion that warm periods during the Holocene had a negative effect on the hydrological cycle, while a colder climate brought more rains and humidity. It should be noted that there was a difference between the climate of the Pleistocene and that of the Holocene. In the Pleistocene, most of the lberian peninsula was cold and dry during glacial periods, while interglacial warm periods were more humid.

Ayala-Carcedo and Iglesiaz Lopez (1997) used a general circulation model to investigate the possible impact of climate change on the hydrological resources of Spain for the year 2060, taking into account a predicted rise of 2.5 °C in the average annual temperature. They estimated that precipitation will be reduced by 6-8% in the northern part of Spain (except the most northwestern provinces of Calicia and the Basque country, where it will be reduced by 2%) and 9-17% in the southern part. In general, this will reduce the surface and subsurface hydrological resources. The change will be affected by a high degree of annual variability. The major impact will be in the southern part of Spain.

Consequently, in spite of the scarcity of precise dates for past samples, it can be concluded that during the past, and, therefore, future warm periods, the south and especially the southwestern part of the Iberian peninsula would become much drier, while other areas might become somewhat drier.

0 0

Post a comment