Lake-level changes and fire history at Lagunillo del Tejo (Spain) during the last millennium: Climate or humans?

Abstract

Hydroclimatic variability is expected to be affected by global warming in the Mediterranean region where climate, fire and human activities are known to be interdependent. The latter is examined here for the past millennium by studying paleoenvironmental indicators from a sedimentary sequence at Lagunillo del Tejo (Cuenca, central Spain). Inferred changes in fire activity and lake levels are based on records of macrocharcoals and plant/algal macrofossils, respectively, and are compared with independent information on climate change and socio-economical transformations from historical and meteorological records. The results show that there is an obvious climatic forcing behind the lake-level changes recorded at Lagunillo del Tejo, and a good correlation between the periods of high fire activity/frequent fires and low lake level/drought conditions. The reconstructed fire regime may therefore be natural (climate-induced), but can also be explained by important socio-economical events/changes, including wars and the introduction of the Transhumance practices (ad 1273). There is a good chronological agreement between lowest fire activity and high lake levels (c. ad 1600–1800), concurrent with the late ‘Little Ice Age’ (LIA) and the collapse of the Transhumance system. We propose that periods of drought favored both natural and human-induced fires during the ‘Medieval Climatic Anomaly’ (around ad 1200), at the start of the LIA (around ad 1400), in the middle of the LIA (sixteenth century) and during the entire nineteenth century. This record is an example of long-term interplay between climate changes and human activities and its impact on environmental changes such as fire regimes.

Keywords

climate fire history human impact Iberian Peninsula lake-level change macroscopic charcoal plant-macrofossil

Introduction

Global warming is expected to have severe effects on hydroclimatic variability, generally increasing precipitation at high latitudes and reducing global rainfall at low latitudes (Intergovernmental Panel on Climate Change (IPCC), 2007). Higher frequencies in the occurrence of droughts, together with extreme summer temperatures, may cause changes in land use to counteract the impact of climate change in extremely dry or humid periods (Viglizzo et al., 1997). The Iberian Peninsula, with its latitudinal position and complex orography, tends to be a territory with high spatial/temporal climatic heterogeneity. These geographic conditions, together with the long-term interactions between natural and human-induced changes, have shaped the Mediterranean area during the last millennia. Burning of woodland has long been used by humans to convert natural ecosystems into pasture, to produce charcoal, or as a war strategy. According to the oldest historical records, we know that Mediterranean landscapes have burnt regularly; today, 90% of the burned areas of Europe lie within the Mediterranean region (Chuvieco, 2009). However, little is known about the timing and causes behind past changes in fire activity in the Iberian Peninsula. The most detailed studies of past changes of fire activity and regimes in the European-Mediterranean region have been carried out in Italy (e.g. Colombaroli et al., 2008; Vannière et al., 2008). Recent studies indicated that fire was also used by humans in high altitude areas such as the Pyrenees and the Alps to destroy the forest and provide open areas for farming (Bal et al., 2011; Colombaroli et al., 2008; Galop et al., 2002; Tinner et al., 1998, 2005). Most studies in the Iberian Peninsula have focused on the effect of fire on vegetation (Carrión and Geel, 1999; Carrión et al., 2003; Peñalba, 1994; Sánchez-Goñi and Hannon, 1999; Santos et al., 2000) rather than on the relationship between fire and hydroclimatic variability or human activities. Therefore, there is a need of looking into long-term fire history to resolve such inquiries as:

Is there any climate–fire relationship in the recent past?

Were fires in the past driven by humans rather than by climate?

To answer these questions we have studied a sedimentary record recovered in Lagunillo del Tejo, a small karstic lake near Cuenca (central Spain). This lake has already proved to contain valuable evidence of past rainfall changes inferred from sedimentary pigments, diatoms and cladocera (Romero-Viana et al., 2009). The aim of this new study was to further examine the patterns of humidity changes inferred from a reconstruction of the lake-level changes at Lagunillo del Tejo and to look into possible relationships with the fire history at the site as inferred from macrocharcoal analysis. We also compared the results with historical, archaeological and meteorological information to disentangle natural (climatic) from anthropogenic factors as fire forcing. A major goal of the study was to evaluate the possible relationships between local fire history and independent evidences of past climate change and human factors, in an attempt to better understand the impact of the interplay of climate and human activity on fire regimes during periods of significant climate change such as the last millennium with two major anomalies, i.e. the ‘Medieval Climate Anomaly’ (MCA) and the ‘Little Ice Age’ (LIA).

Material and methods

Study site

Lagunillo del Tejo is a closed doline lake situated at 3 km from the village Cañada del Hoyo in the Cuenca Mountains (southwestern zone of the Iberian System, Spain). It is a circular sinkhole fed by groundwater and subject to water-level fluctuations, with significant changes in lake-surface area (Romero-Viana et al., 2009) (Figure 1). In early 1980, the maximum depth of the lake was 12 m, and its diameter 85 m (Vicente and Miracle, 1984); since then, it never reached again such a high water level. In the driest years (e.g. 1995, 2008) the lake was 4 m deep with 48 m diameter. The lake is monomictic in the years with high water level, thermally stratified from May to early autumn, with an anoxic zone developing in summer, whereas it is not stratified and no anoxic zone develops in the years of low water level (<5 m). It is a hard-water lake with conductivities around 400–600 μS/cm, increasing slightly in the bottom when the lake is stratified. The pH ranges from 8 to 9 in the epilimnion, being around 7 in the bottom waters when the lake is stratified. When the water level is high, we have observed that the littoral zone is colonized by two rings of macrophytes, the inner ring dominated by Potamogeton pectinatus and Myriophyllum spicatum, and the outer ring by Ranunculus sp., Chara spp. and Polygonum amphibium. At low water levels, the littoral zone consists of a single macrophyte ring composed mainly of P. pectinatus, while the outer macrophyte ring dries up (Romero-Viana et al., 2009). Cirujano (1995) also described similar observations; in 1991, when the lake level was relatively high, he recorded three Chara species (C. fragilis, C. muscosa and C. desmacantha) and Polygonum amphibium forming a ring near the shore, and a ring of Potamogeton pectinatus and Myriophyllum spicatum inwards. In 1992 the lake level was lower than in 1991, and there was only a single ring with P. pectinatus, M. spicatum, and some P. amphibium.

Figure 1.

Location of the study site (Lagunillo del Tejo, Spain), and the cattle track net (including Cañada Real de Andalucía). Water-level fluctuations of Lagunillo del Tejo between 2007 and 2009 are illustrated by three pictures. When water depth was high (5.5 m in 2007 and 7 m in 2009), the lake had two rings of macrophytes, while only the inner macrophyte ring was present when water depth was low (4 m in 2008)

The present climate is continental-Mediterranean with sharp daily and seasonal temperature fluctuations. The mean monthly temperatures for the coldest month (January) and the warmest month (July) are 4.3°C and 25°C, respectively. The annual mean precipitation is 525 mm, May being the most rainy month and August the driest month (mean values of the 1950–2003 data series from Agencia Estatal de Meteorología).

Sediment coring and chronology

A 40 cm long sediment core (CN-4) was retrieved in Lagunillo del Tejo on 18 September 2009 using a gravity corer of 6 cm diameter for plant macrofossil and macrocharcoal analyses. The core was recovered at 30 m from the shore, at a depth of 5.5 m, close to the actual inner boundary of the Potamogeton pectinatus ring (Figure 1). At the time of coring, the maximum depth of the lake was 5.7 m and its diameter was 50 m. On 15 March 2008, several cores were taken from the deepest part of the lake using a Russian peat corer. One of these cores, CN-3bis, was used to analyze the isotopes ²¹⁰Pb and ¹³⁷Cs with the aim to verify the chronological model established from a previous core, CN-2, taken in 2004 and described in Romero-Viana et al. (2009). ²¹⁰Pb and ¹³⁷Cs measurements were made by direct gamma assay (Appleby et al., 1986) at the Valencia University Environmental Radioactivity Laboratory. Ages were calculated by the constant initial concentration model (CIC). The cores CN-4 (macrocharcoal and plant macrofossil analysis) and CN-3bis were correlated using changes in lithological characteristics (alternation of dark and light layers) and in loss-of-ignition (LOI of 1 cm³ dry sample burned for 6 h at 460°C, APHA-AWWA-WEF, 1992) (Figure 2). Terrestrial plant and charcoal macroremains from the cores CN-3 and CN-4 were used for accelerator mass spectrometry (AMS) ¹⁴C measurements performed at the Poznan Radiocarbon Laboratory (Poland). Calibrated ages were calculated using OxCal version 4.1. (Ramsey, 1995). All dates in the text below are given in calibrated years (Table 1).

Figure 2.

(A) LOI (loss-on-ignition, in % of dry sediment) records from Lagunillo del Tejo’s cores CN-3 and CN-4. (B) Comparison between the radionuclide activities (Bq/kg) of ²¹⁰Pb and ¹³⁷Cs in the core CN-3. (C) Age/depth model for core CN-4 based on all dates available from CN-3 and CN-4, i.e. the calibrated ¹⁴C dates (Table 1) and the ¹³⁷Cs and ²¹⁰Pb dates

Table 1.

Dates from Lagunillo del Tejo: ¹³⁷Cs and ²¹⁰Pb dates from core CN-3 (in years BP and ad), and ¹⁴C dates from core CN-4 (in uncalibrated ¹⁴C years BP and calibrated years ad (age range and mean). The calibration of the ¹⁴C dates was carried out using the computer program OxCal version 4 (Ramsey, 1995, 2001)

Core	Depth CN-3 (cm)	Depth CN-4(cm)	¹³⁷Cs, ²¹⁰Pb dates (yr BP), uncalibrated ¹⁴C dates (¹⁴C yr BP)	¹³⁷Cs, ²¹⁰Pb dates (yr ad), calibrated ¹⁴C age ranges and mean (yr ad)	Material
CN-3	7.5	5.7	−4	1954	Gyttja ¹³⁷Cs
CN-3	17.5	13.5	100	1850	Gyttja ²¹⁰Pb
CN-3	36-37.5	27.5-28.5	345±30	1406-1640; 1520±117	Pine cone scale
CN-3	43-44	32.5-33.5	705±30	1250-1310; 1280±30	Big fragment of wood
CN-4	44-45	33-34	735±30	1223-1294; 1260±70	Pine cone scale
CN-4	45-46	34-35	790±30	1206-1280; 1240±40	Charcoal
CN-4	47-48	36-37	885±30	1116-1219; 1170±52	Little fragments of wood

Plant macroremain and charcoal analyses

The core CN-4 was subsampled into continuous 1 cm thick sediment slices, and 20 cm³ sediment from each slice were used for plant/algae macrofossil and charcoal analyses. These subsamples were sieved through 0.5 mm and 0.25 mm mesh sizes (Berglund and Ralska-Jasiewiczowa, 1986). Plant macroremains were sorted and counted under a binocular microscope. They were identified using atlases and keys of fruits and seeds (e.g. Beijerinck, 1976; Schoch et al., 1988), the reference collection of the Laboratory of Paleoecology at the School of Natural Sciences, Linnaeus University (Sweden), and a reference collection from the study site. Macroscopic charcoal fragments were counted and divided into two fractions (from 0.25 to 0.5 mm and >0.5 mm) in an attempt to separate local from regional fires (Tinner and Hu, 2003).

The plant/algae macrofossil remains (Figure 3) include leaves (Myriophyllum sp. only), seeds, fruits and oospores of algae (Chara spp. and Nitella spp.). Below, the terms ‘plant macrofossil data’ or ‘plant-macrofossil record’ always refer to remains of both algae and plants for the sake of simplicity. The 22 taxa were grouped into four categories, i.e. trees, upland plants, marsh plants, and aquatics. The results are given in number of remains per volume and plotted against depth and age. The diagrams in Figures 3 and 4 were made using the computer programs Tilia and Tilia graph (Grimm, 1987, 1990, 2004). Plant macrofossil assemblage-zones are based on depth-constrained cluster analysis of their concentrations, as implemented by CONISS in the Tilia program. The total sums of square (dissimilarity measures obtained from the CONISS analysis based on the total record of plant and algae remains) of 300 and 200 were chosen to identify the major zones and subzones, respectively.

Figure 3.

Plant macrofossil record from Lagunillo del Tejo, core CN-4, expressed in number of remains per 20 ml. The charred plant remains (seeds) are not presented in this diagram, but in Figure 4. The local macrofossil assemblage-zones (MZ) are based on a depth-constrained cluster analysis (CONISS) of the total plant macrofossil record as implemented by the computer program Tilia (Grimm, 1987, 1990). The lithostratigraphical layers (R-O, see description in the text) are indicated in the right column for comparison. Levels interpreted as phases of low lake-level are shown with grey shadow

Figure 4.

Record of charred plant macroremains (in presence/absence per 20 ml) and macrocharcoals (number per 20 ml) from Lagunillo del Tejo, core CN-4. The number of bud scales and taxa richness (number of taxa found per 20 ml) are shown for comparison. Local plant macrofossil assemblage zones, lithostratigraphical units (O-R), and phases of low lake-level (grey shadow) as in Figure 3

Results

Lithology and chronological model

The sediment records from Lagunillo del Tejo show conspicuous alternating dark and light bands (Figure 2A) corresponding to reduced (‘R’, high content of organic matter) and oxidized (‘O’, low content of organic matter) zones, respectively, as described earlier by Romero-Viana et al. (2009). We propose a correlation between the CN-3b and CN-4 sediment cores based on the assumption that these ‘R’ and ‘O’ zones, and the changes in LOI, are synchronous. It suggests an average compression factor of 1, 3 (length CN-3/length CN-4) for core CN-4, which might be due to differences in coring techniques, i.e. CN-4 is a gravity core, while CN-3 was extracted with a Russian peat corer without compression.

The ²¹⁰Pb and ¹³⁷Cs activity profiles from core CN-3bis (Figure 2B) show a high level of concordance with those obtained by Romero-Viana et al. (2009) from core CN-2. The ²¹⁰Pb activity shows an exponential decline with depth in the upper zone, reaching very low values coinciding with the change to zone O5, and a rise in zone O5. Application of the CIC model to the undisturbed part of the ²¹⁰Pb profile yields an average sedimentation rate of about 1.09 mm/yr in the CN-3bis core. This corresponds to c. 0.87 mm/yr in the compressed core CN-4.

The ¹³⁷Cs activity (Figure 2B) shows also a decline within the upper zone, with slight increases at 3 cm and 7 cm. All ¹³⁷Cs activity is concentrated in the upper 7.5 cm in core CN3, (at the end of R5), which would correspond with the onset of weapons testing in ad 1954. The application of the CIC model to the ²¹⁰Pb activity is in agreement with the proposed ¹³⁷Cs chronology (see above). Insignificant traces of ¹³⁷Cs were detected in layers preceding the 7.5 cm level, but they can be attributed to diffusion. The chronology for the CN-4 core is based on an age–depth model using the ²¹⁰Pb and ¹³⁷Cs dates obtained from core CN-3bis (by correlation of the cores CN-4 and CN-3bis as described above) and five AMS dates of terrestrial plant material from both cores (Table 1; Figure 2C). The age–depth model is a linear interpolation of the calibrated ¹⁴C dates and between the oldest ²¹⁰Pb date (ad 1850) and the youngest calibrated ¹⁴C date (ad 1520 ± 117). According to this age model, the bottom of the profile is dated to c. ad 1100, and the sediment accumulation rate increases gradually through time, from the bottom to the upper part of the sequence (from c. 0.26 mm/yr to 0.87 mm/yr). The chronology is somewhat uncertain between ad 1520 and ad 1850 because of the lack of additional dates, and in the bottom, below ad 1170 (37–40 cm). However, the interpolation between the ¹⁴C and ²¹⁰Pb chronologies is reasonable given the regular shape of the overall age/depth model. A more gradual increase in accumulation rates between ad 1520 and ad 1850 would produce slightly younger ages between 14 and 27 cm of only a few years (maximum 20 years younger around 21 cm). The uncertainty of the chronology below 37 cm has more serious implications because the accumulation rate might be lower than assumed and the bottom of the core much older than estimated, i.e. a bottom age of ad 900 or less is not impossible. The oldest ¹⁴C age (ad 1116–1219) dates the R2/O3 boundary, which suggests that the O3 layer might well correspond to the MCA. The uncertain bottom chronology does not allow to decide whether the layer R2 preceeds the MCA or belongs to it.

Plant macroremain and charcoal analyses

The plant macrofossil record is characterized by the dominance of oospores of the algae Chara spp. and seeds of the aquatic plant Ranunculus subgenus Batrachium (Figure 3). However, the amount of remains from these two taxa varies significantly through the profile, along with the representation of the less common taxa remains. The description of the zones and subzones is presented in Table 2.

Table 2.

Description of the macrofossil assemblage zones (MZ1–MZ5) from Lagunillo del Tejo (see macrofossils diagrams Figure 3)

Zones MZ	Depth (cm)	Zone description	Subzones	Depth (cm)	Subzone description
1	32-40	Chara dominant, Nitella common; regular occurrence of Juncus acutus and Lycopus europaeus	1a	34-40	Large numbers of upland and marsh plants, few Myriophyllum spicatum and Alisma Plantago-aquatica
		32-33 cm: maximum values of Chara and Ranunculus subg. Batrachium	1b	32-34	Peaks of Chara, Nitella, R. subg. Batrachium, Potamogeton pectinatus
1/2	32	Decrease in the the total number of plant/algae remains
2	21-32	Low values of Chara and Ranunculus subg. Batrachium; regular occurrence of Caryophyllaceae and Cruciferae	2a	24-32	31–32 cm: occurrence of Pinus, Umbelliferae, Verbana officinalis, Lycopus europaeus, Polygonum lapathifolium, Juncus acutus
					32–28 cm: few taxa except Chara, Nitella, R. subg. Batrachium, Cyperaceae and Caryophyllaceae
					24–28 cm: low values of Cyperaceae, new occurrence of Cruciferae and Typha latifolia
			2b	21-22	Peak of Chara, regular occurrence of R. subg. Batrachium, findings of Pinus and Juncus acutus
2/3	21	Sharp decrease in values of Chara and Ranunculus subg. Batrachium
3	7-21	Low values of Chara	3a	13-21	High values of R. subg. Batrachium
		Regular occurrences of Cruciferae			16–21cm: occurrence of Nitella, low values of Cruciferae
		Some findings of Pinus, Caryophyllaceae, Cyperaceae, Papaver rhoeas, Juncus acutus, Typha latifolia			14–16 cm: very low values of R. subg. Batrachium; high values of Cruciferae, Cyperaceae, Typha latifolia; occurrences of Pinus, Caryophyllaceae, Ranunculus reptans, Lycopus europaeus, Cyperaceae, Typha latifolia
					13-14 cm: Peak of R. subgr. Batrachium; occurrences of Cruciferae, Ranunculus reptans and Lycopus europaeus
			3a/3b	13	Sharp decrease in R. subg. Batrachium and Cruciferae
			3b	7-13	Low values of R. subg. Batrachium, high values of Cruciferae 9-11 cm; occurrences of Pinus, Juniperus, Caryophyllaceae, Chenopodiaceae, Cyperaceae, Juncus acutus, Typha latifolia
3/4	7	Increase in Chara
4	1-7	High values of Chara and Ranunculus. subg. Batrachium, regular occurrences of Typha latifolia; occurrences of Myriophyllum spicatum	4a	4-7	Middle high values of Chara, high values of Typha latifolia 4-5 cm
			4b	2-4	Very high values of Chara
			4c	1-2	Middle high values of Chara,
4/5	1	Decrease in values of Chara and Ranunculus subg. Batrachium
5	0-1	Occurrence of Juncus acutus only

The record of macrocharcoal fragments is continuous through the profile (Figure 4). However, the fragments are more abundant in the bottom (27–40 cm) and upper (11–16 cm) parts of the core, corresponding to the periods c. ad 1200–1500 (a charcoal fragment was ¹⁴C dated, Table 1) and ad 1800–1900, respectively. This general trend is found in both charcoal fractions (250–500 µm and >500 µm). Nevertheless, the largest fraction occurs more regularly than the smaller one. High values of bud scales and taxon richness are also related to charcoal abundance. This fact is especially evident for the budscales in the bottom levels, whereas the taxon richness is also related to LOI (Figures 2 and 4).

Discussion

Lake-level fluctuations and climate change

Plant macrofossils are one of the important proxies of past changes in lake-levels and, by inference, humidity (e.g. Dieffenbacher-Krall and Halterman, 2007; Digerfeldt, 1986; Gaillard and Birks, 2008; Gaillard and Digerfeldt, 1991; Hannon and Gaillard, 1997; Harrison and Digerfeldt, 1993; Street-Perrot and Harrison, 1985). In this study, we use the recent neo-ecological behavior of the lake (Figure 1) as a modern analogue to help the interpretation of the plant macrofossil record in terms of past lake-level changes. In March 2008 the maximum depth of the lake was 4 m and a single 5–6 m wide ring of macrophytes with a dominance of P. pectinatus was found between the shore and a depth of 2 m. The shoreline was marked by a white 0.4 m wide fringe of dried macrophytes, fibers of stems and leaves, mainly from P. pectinatus. In June 2009, the maximum water depth was 7 m and two rings of macrophytes were observed. A 5–6 m broad outer ring with P. amphibium, Chara spp. and Ranunculus subgenus Batrachium occurred from the shore to a depth of 2 m, and an inner 7 m broad ring (6–13 m from the shore) reached a depth of 4 m and consisted of P. pectinatus with an important proportion of Myriophylum spicatum, mainly in its outer edge (Figure 1). Besides this recent information, the impoverishment of the outer ring during low lake levels has been recorded at Lagunillo del Tejo for the last 30 years (Vicente and Miracle, 1984 and unpublished observations by these senior authors). In particular, we noted that only perennial species colonized the lake during the dry years, especially during years without spring rains, because the absence of a flooded edge at this season prevented the development of the annual Chara spp. that reproduces solely from spores.

Based on the observations described above, we propose that past plant macrofossil assemblages from sediment cores at Lagunillo del Tejo might indicate major lake-level changes. Given the present relationship between lake-level and shore/aquatic vegetation at the study site and other similar nearby lakes, we assume that high numbers of remains from Chara spp. and Ranunculus subgenus Batrachium in the sediment testify on high lake levels, while the occurrence of remains from ruderal and annual plants (e.g. Caryophyllaceae, Chenopodiaceae, Cruciferae and Ranunculus reptans) in conjunction with low values of Chara spp. indicate low lake levels. The competitive, marginal plant communities (such as Cyperaceae, Juncus acutus and especially Typha latifolia) seem to take advantage of lake-level changes as they occur in most of the record and are particularly abundant during periods of lake-level change – either from high to low, or from low to high – probably as part of a shore vegetation succession in space and time. The record does not register changes in the occurrence of Potamogeton pectinatus and Myriophyllum spicatum that could be assigned to high lake-levels with certainty. This is most probably because of the location of the studied core. A core from a deeper location would probably contain a better record of the ‘inner ring’ (Figure 1). However, we want to point out that the few findings of Myriophyllum and Potamogeton are from periods with inferred high lake levels (Figure 3).

The inferred lake-level changes are based on the data presented in Figure 3, but are also shown in Figures 4 and 5 for comparison with the macrocharcoal record (Figure 4) and other data, such as information on climate change in the study region (Figure 5). In Figure 5, the well-known climatic anomalies of the last millennium, i.e. the ‘Medieval Climate Anomaly’ (MCA, ad 950–1250) and the ‘Little Ice Age’ (LIA, ad 1400–1700) follow the definition by Mann et al. (2009).

Figure 5.

Compilation of the macrocharcoal record and lake-level reconstruction (low levels in grey) from Lagunillo del Tejo (first panel), synthetic pollen and charcoal records from Lake La Cruz (same region, following Julia et al., 1998) (second panel), and evidences from historical documents on socio-economical events and transformations, and climatic changes (periods of droughts and increased rainfall) (third panel). The macrofossil zones (Figure 3), the lithostratigraphical R-O layers (see description in text) and the ‘Medieval Climate Anomaly’ and ‘Little Ice Age’ (according to Mann et al., 2009) are indicated in the two right columns for comparison

The plant-macrofossil record suggests relatively high lake levels until c. ad 1350 and a gradual reduction of the water volume from c. ad 1400 to ad 1600. There is indication of two periods of slightly lower lake levels around ad 1150 and ad 1250 that fall into the MCA. However, the plant macrofossil record between 34 and 40 cm is ambiguous and its interpretation in terms of lake level not straightforward (Figure 3; see also further discussion of this part of the profile below). In contrast, the record above 34 cm starts with clear indications of high lake level c. ad 1300–1350 followed by a successive decrease in lake-levels, with particularly low levels around ad 1400 and c. ad 1550–1600. Interestingly, Roberts et al. (2004) found a general trend of maximum wetness in the Mediterranean basin c. ad 1250–1400, which agrees well with our record. The first lower lake-level c. ad 1400 belongs either to the pre LIA or to the early LIA, and the second one c. ad 1550–1600 falls in the middle of LIA. Historical documents mention that the Fuenteseca Waterwheel (150 km from the lake) did not operate ad 1504, 1515 and 1525 because the main irrigation ditch was dry (Rubio-Dobón and Melendo, 2005). The same authors describe the periods ad 1530–1571 and ad 1580–1589 as especially dry in an area situated 100 km north of our study site. Furthermore, we know that a pilgrimage to the FuenSanta de Villel Sanctuary was undertaken by young people from Cella ad 1556 in order to beg for water. The low lake levels inferred from the macrofossil record c. ad 1500–1600 compare well with these historical records, and imply that the sixteenth century was locally very dry. There are also known changes in precipitations in central-eastern Spain during the seventeenth century, with a minimum of rainfall ad 1610 and a maximum ad 1660 (Toledo and Zaragoza; Rodrigo and Barriendos, 2008). Similarly, the inferred high lake levels c. ad 1625–1800 are particularly high c. ad 1625–1725, i.e. during the late LIA.

From the middle of the eighteenth century, there is indication of successively lower lake-levels (not shown in Figures 3 and 5; indicated by increases in Cruciferae and Juncus acutus) that might correspond to the droughts reported from both northern and southern Spain ad 1750 (Rodrigo and Barriendos, 2008; Rodrigo et al., 1999). The time period c. ad 1800–1930 is characterized by two episodes with low lake levels, around ad 1820 and c. ad 1875–1930, of which the latter might have been the lowest. The two episodes are separated by a high lake-level phase c. ad 1850 that might be compared with the recorded period of high pluviosity at the meteorological station of Madrid in the middle of the nineteenth century. Roberts et al. (2004) describe a general trend in the Mediterranean Basin towards cooler and drier conditions c. ad 1700–1850, which might be compared with the decrease in lake-level indicated in our plant macrofossil record from c. ad 1750, with particularly low levels c. ad 1820.

Around ad 1900, macrofossil remains were very few, which was also the case of pigment markers, diatoms and cladocera in a previous study (Romero-Viana et al., 2009); the latter was ascribed to high erosion during the construction of the agricultural terraces in the lake basin at the beginning of the twentieth century. The dominance of macroremains from marsh, upland and tree taxa also indicates the lowest lake level of the entire record. From ad 1950 to 1980 the lake-level rose again, which is supported by Vicente and Miracle (1984) who reported high levels in 1980–1981. The decrease of Chara spp. in the uppermost layer may correspond to the known lake-level drop in the decade of 1990.

According to Romero-Viana et al. (2009), the water-level fluctuations in the sink hole caused two sedimentary situations in the record of Lagunillo del Tejo. Oxidized ‘O’ layers (O1–O5) were interpreted as the consequence of rapid water-level lowering that caused an increase of erosion from the lake shores and higher accumulation rate of coarse material into the lake sediments, while reduced ‘R’ layers were assumed to indicate a rise in lake level, leading to stratification of the water column and high values of organic matter content. We find that the lake-level reconstruction based on the plant-macrofossil record agrees well with the O-R layers from c. ad 1550 onwards, i.e. layers O4 and O5 correspond to times of low lake levels, while layers R4 and R5 relate to high lake levels. The agreement is less good for the bottom part of the core. The second of the two first episodes of slightly lower lake-level does fall into layer O3, but the first one belongs to the end of R2. As discussed above (Chronology section), R2 might either be of MCA or pre-MCA age. Moreover, the plant-macrofossil record is ambiguous and the sample resolution is low. We suggest that the sediments between c. 34 and 38 cm as a whole may represent a time of lower lake-levels corresponding to the MCA. However, this conclusion is very tentative.

Layer R3 corresponds to a period characterized by several lake-level fluctuations with the highest levels at the beginning of the layer, and the lower lake levels around ad 1400 and from c. ad 1550. As discussed above, periods of drought are documented in historical sources during several periods of the sixteenth century (end of layer R3). The decrease in lake-level around ad 1750 is another example of fluctuation within a R layer (R4 in this case). This implies that the R-O layering at Lagunillo del Tejo is a good proxy of the major trends in long-term lake-level changes. But the shorter-term fluctuations might be overseen in the R-O stratigraphy, while plant-macrofossil records may catch details better. As demonstrated above, several of the lake-level fluctuations found within R or O layers are confirmed by historical data or meteorological records.

Lagunillo del Tejo has no inlet or outlet, and there is no indication of present or past human intervention on the hydrology of the lake. The good correspondence between the inferred lake-level changes, the historical data on periods of droughts and high pluviosity during the last 500 years, and the known general climate history of the last millennium (i.e, the warm MCA, and the cold LIA with its Maunder Minimum) – implies that the past lake-level changes at Lagunillo del Tejo were forced by climate change, i.e. changes in humidity (precipitation minus evaporation). Therefore, we assume that the inferred low lake levels are related to periods of drought, while the high lake-levels are related to wet conditions.

Fire history and socio-economic changes

The record of macroscopic charcoal (both fractions) (Figure 4) indicates that fires occurred at Lagunillo del Tejo during several periods of the last 900 years, in particular between c. ad 1170 and ad 1525 (with peaks of charcoals around ad 1200, ad 1400, and ad 1500), and between c. ad 1800–1900. It is tempting to assign the first period of frequent fires (c. ad 1200) and the low fire activity between c. ad 1550–1800 to the MCA and LIA, respectively, and the high fire activity around ad 1400 and c. ad 1800–1900 to periods of repeated droughts. However, human activities were significant in the study region during the entire last millennium and need to be examined to attempt disentangling climate from human activities as forcing of fire events.

The first peak of charcoal dated to c. ad 1175–1210 (period of assumed low lake-level) may be related to wars of the ‘Reconquest’ period, since the city of Cuenca was conquered by Christians in ad 1177. Burning the forest to avoid ambushes was one of the most common ways that Christians used to regain lands from the Muslims. At that time, the Arabic ‘atalaya’ tower of Cañada del Hoyo (3 km from the lake) was converted by the Christians into a castle to defend the frontier of the Castilla Kingdom that ran through the study area (Lozano-Sahuquillo, 2002).

The main cause of fire activity from the end of the thirteenth century onwards was probably due to a major land-use change in Castilla, i.e. a prominent upsurge in sheep raising (‘Mesta’) based on transhumance. It required that forests were burnt in order to get grass for the cattle. During this period, transhumance used a system of drover roads called vias pecuarias o cañadas. The Cañada del Hoyo village was situated in one of the two branches of the ‘Real Cañada de Andalucía’ that was established in ad 1273 (historical sources). The ‘Real Cañada de Andalucía’ was 37 ha large, 5 km long and 75, 22 m wide. The importance of this shepherding and transhumance in the area was such that Cañada del Hoyo’s castle was inhabited by 80 sirs, three vassals and one lieutenant (Lozano-Sahuquillo, 2002). Close to lakes, the paths were generally widened to provide resting and watering places for the sheep. The ‘Mesta’ was also very important during the sixteenth century (also characterized by low lake levels and droughts), but declined thereafter. Besides being an indicator of low lake level, the abundance of Cruciferae in the plant-macrofossil record c. ad 1550 might indicate a decrease of grazing around the Lagunillo del Tejo and the collapse of the ‘Mesta’. There is also a sharp decrease in charcoal values after ad 1525 followed by low values until c. ad 1800.

Several studies, have shown that macroscopic charcoal reflect local fires, while microscopic charcoal ( < 250 μm) would represent fires at a more regional spatial scale (e.g. Briles et al., 2008; Clark, 1988; Long et al., 1998; Tinner et al., 1998). Although this relationship does not always apply (Clark et al., 1989, Innes et al., 2004; Pitkanen et al., 1999), we suggest that the maxima of large-size charcoal fragments around ad 1400 and ad 1500 may indicate important forest fires in the lake’s immediate surroundings due to transhumance close to the study site. The small-size charcoal fragments are particularly abundant around c. ad 1200 which may indicate that fires occurred in a relatively large area because of the wars of the Reconquest in the region. Nevertheless, the occurrence of charred seeds indicate that fires may also reached the lake’s shore.

Peaks of charcoal between c. ad 1200 and 1500 were also detected by Julià et al. (1998) in the sedimentary sequence of La Cruz lake (300 m from Lagunillo del Tejo) (Figure 5). Moreover, the pollen assemblages in La Cruz lake exhibit a progressive increase of Pinus and a related reduction of herbs and cereals during the same period (Burjachs, 1996; Julià et al., 1998) (Figure 5). These changes were interpreted in terms of land-use transformation, from agriculture to sheep raising. The authors explained the increase in pine pollen by a larger transport of pine in the open, grazed areas, and the decrease in herb pollen by less flowering because of grazing.

The upper peak of charcoal (c. ad 1800–1850), might be related to the implantation of the castle of Cañada del Hoyo during the ‘Carlists wars’ of the nineteenth century. The burning of Carboneras and surrounding sites during the battle of 1839 was one of the important events of the 1st Carlist war. However, this period also coincides with national programs (‘desamortization’) aiming at transforming the ecclesiastic, nobles’ and public lands into productive parcels, including burning of woodlands. Further, the nineteenth century was characterized by burning of stubble and the production of charcoal, coinciding with the industrial era. The migration from Moya (Cuenca) of people who practiced burning of oak for a living led to the foundation of Carboneras village (‘carboneras’ means charcoal sites) situated 15 km from Lagunillo del Tejo. The exploitation and commercialization of this charcoal led to the subsequent development of the village (Esteso-Carnicero, 2004). The population of both villages Cañada del Hoyo and Carboneras increased during the last part of the nineteenth century and the beginning of the twentieth century when many residents were dedicated to the production of charcoal. Therefore, we propose that the two most probable socio-economical explanations for the charcoal record of the nineteenth century are (1) the more or less distant woodland fires due to the transformation of lands into more productive areas, and (2) the production of charcoal.

Conclusions

Changes in both fire activity and lake levels may be human- and/or climate-induced (e.g. Gaillard and Digerfeldt, 1991; Olsson et al., 2010). As discussed above, lake-level changes at Lagunillo del Tejo were shown to be primarily climate-induced. Therefore, the lake-level reconstruction provides a proxy record of regional climate change that can be directly compared with the fire record, which makes it possible to examine the relationship between fire activity and climate change.

If the prevailing climate conditions were relatively dry and facilitated fire by increasing desiccation and inflammability of plant material, humans may have been at the origin of higher fire frequency and intensity, either they used fire intentionally, or set fires by accident. Therefore, under dry climate conditions, it may be difficult to separate human-induced from climate-induced fires. In such circumstances, fire is a product of the interplay between climate and human activities rather than a consequence from either one factor or the other.

There is a strikingly good chronological agreement between the time of lowest fire activity (c. ad 1550–1800), high inferred lake-levels/wet climatic conditions (c. ad 1600–1800), and the R4 layer (ad 1650–1800). These phases are all within the LIA (ad 1400–1700). The record of Lagunillo del Tejo also exhibits a relatively good correspondence between periods of low lake level/drought and times of high fire activity (possibly O3, around ad 1200; around ad 1400; and the first part of O5, c. ad 1800–1850).

The first peak of charcoal (around ad 1200) covers the R2/O3 boundary and is preceeded and succeeded by periods of lower lake levels. However, as stressed above, the plant-macrofossil record is not clear and the chronology uncertain. The entire period c. ad 1150–1250 might as well be considered as a time of lower lake levels corresponding to the MCA. The peak of charcoal falls within the MCA and might be related to dry climatic conditions. The latter does not rule out a possible impact of human activities, such as the wars of the Reconquest Period or/and the introduction of the transhumance practices (Figure 5).

The charcoal peaks around ad 1400 and ad 1525 belong to zone R3, a period of assumed high lake levels. However, the plant macrofossil data indicate two phases of low lake-levels. The first one (ad 1400) correlates with a charcoal peak, the second (ad 1550–1600) just succeeds the second charcoal peak. The documented droughts during the sixteenth century may have favored both fire and low lake-levels (Figure 5). Moreover, the transhumance practices of the ‘Mesta’ may also be at the origin of intentional or accidental fires around ad 1400 and ad 1525. Although zone O4 is assumed to indicate lower pluviosity and correlates relatively well with the recorded low lake-levels and historical documents (Figure 5), there is no sign of fires at that time.

The last evidence of frequent fires is dated to the period c. ad 1800–1900 that correlates well with zone O5 and two periods of particularly low lake levels (c. ad 1820 and ad 1890). This is also a time of wars, innovation, land-use transformation and industrial revolution, all of which certainly had a profound impact on the fire regime in the region. The two periods of more severe drought obviously favored fires (human induced or natural) locally.

All in all, there is an obvious climatic forcing behind the lake-level changes registered in the sediments of Lagunillo del Tejo. The study suggests that periods of drought have favored both natural- and human-induced fires locally and in the region during several periods of the last millennium, in particular at the end of the MCA (around ad 1200), at the start of the LIA (around c. ad 1400), in the middle of the LIA (sixteenth century), and during the entire nineteenth century. The period with the least inferred fire activity and the collapse of the ‘Mesta’ clearly falls within the coldest part of the LIA (Maunder Minimum, ad 1645–1710).

The evidence of the effect of socio-economic transformations and climate change during the Middle Ages at Lagunillo del Tejo is in good agreement with the results obtained in the nearby lake La Cruz (Julià et al., 1998). The record of Lagunillo del Tejo is a nice example of the effects of the complex interplay between climate change and human activities (socio-economical shifts) on environmental changes such as changes fire regimes, which in turn will influence the vegetation, flora and fauna, the overall landscape, and last but not least, living conditions for humans.

Human activities have been so important during the last centuries that it is often impossible to separate natural from anthropogenic signals in the palaeoecological record, which is a feature of this time in most parts of the world (e.g. Brönnimann et al., 2008; Marlon et al., 2010). Nonetheless, major climate shifts such as the MCA and LIA have certainly been an important forcing behind many environmental changes and major socio-economical transformations such as the introduction of transhumance practices in mountain areas during the warm MCA and the collapse of the system probably related to the particularly cold conditions during the Maunder Minimum (end of LIA).

Footnotes

Acknowledgements

CLB thanks all colleagues at the School of Natural Sciences, Linnaeus University (formely University of Kalmar), Sweden, for their kind hospitality during her stay in Sweden. We are very grateful to Dr J.L. Ferrero for ²¹⁰Pb and ¹³⁷Cs analyses, to Dr B. Valero-Garces for lending us the gravity corer. We are also very grateful to two anonymous reviewers for their very valuable comments and suggestions that greatly helped to improve an earlier version of the manuscript.

CLB acknowledges the MICINN for her FPU scholarship. MJG acknowledges the Faculty of Natural Sciences and Technology at Linnaeus University, Kalmar, Sweden, for its financial support. This study was financed by the MICINN:CGL2005-04040/BOS project and the CGL2009-06772-E/BOS grant to EV.

References

APHA-AWWA-WEF (1992) Standard Methods for the Examination of Water and Wastewater. Washington DC: American Public Health Association.

Appleby

Nolan

Gifford

Goffrey

Oldfield

Anderson

. (1986) ²¹⁰Pb dating by low background gamma counting. Hydrobiologia 143: 21–27.

Bal

Pelachs

Pérez-Obiol

Julià

Cunill

(2011) Fire history and human activities during the las 3300 cal yr BP in Spain’s Central Pyrenees: The case of the Estany de Burg. Palaeogeography, Palaeoclimatology, Palaeoecology 300: 179–190.

Beijerinck

(1976) Zadenatlas der Nederlandsche flora. Amsterdam: Backhuys & Meesters.

Berglund

Ralska-Jasiewiczowa

(1986) Handbook of Holocene Palaeoecology and Palaeohydrology. Chichester/New York: J Wiley.

Briles

Whitlock

Bartlein

Higuera

(2008) Regional and local controls on postglacial vegetation and fire in the Siskiyou Mountains,northern California,USA. Palaeogeography, Palaeoclimatology, Palaeoecology 265: 159–169.

Brönnimann

Vogler

Staehelin

Stolarski

Hansen

(2008) Climate variability and extremes during the past 100 years. In: Brönnimann

Luterbacher

Ewen

Diaz

Stolarski

Neu

(eds) Advances in Global Change Research. Springer, 129–140.

Burjachs

(1996) La secuencia palinologica de La Cruz (Cuenca, España). Estudios Palinologicos, 31–36.

Carrión

Geel

(1999) Fine-resolution Upper Weichselian and Holocene Palynological record from Navarrés (Valencia, Spain) and a discussion about factors of Mediterranean forest succession. Review of Palaeobotany and Palynology 106: 209–236.

10.

Carrión

Sánchez-Gómez

Mota

Riker

Chaín

(2003) Holocene vegetation dynamics, fire and grazing in the Sierra de Gádor, southern Spain. The Holocene 13: 839–849.

11.

Cirujano

(1995) Flora y vegetación de las lagunas y humedales de la provincia de Cuenca. Madrid: Junta de Comunidades de Castilla-La Mancha, CSIC, Real Jardín Botánico.

12.

Clark

(1988) Particle motion and the theory of charcoal analysis: Source-area, transport, deposition and samplig. Quaternary Research 30: 67–80.

13.

Clark

Merkt

Müller

(1989) Post-glacial fire, vegetation, and human history on the northern alpine forelands, south western Germany. Journal of Ecology 77: 897–925.

14.

Colombaroli

Vannière

Chapron

Magny

Tinner

(2008) Fire–vegetation interactions during the Mesolithic at Lago dell’Accesa Italy. The Holocene 18: 679–692.

15.

Chuvieco

(2009) Earth Observations and Wildland Fires in Mediterranean Ecosystems. Dordrecht: Springer, 251 pp.

16.

Dieffenbacher-Krall

Halterman

(2007) The relationship of modern plant remains to water depth in alkaline lakes in New England, USA. Journal of Paleolimnology 24: 223–229.

17.

Digerfeldt

(1986) Studies on past lake-level fluctuations. In: Berglund

(ed.) Handbook of Holocene Palaeoecology and Palaeohydrology. New York: John Wiley & Sons, 127–143.

18.

Esteso-Carnicero

(2004) Documentos para la historia: Carboneras de Guadazaón (Cuenca). Ayuntamiento de Carboneras de Guadazaón (eds), Cuenca, 208 pp.

19.

Gaillard

Birks

(2008) Paleolimnological applications. In: Elias

(ed.) Encyclopedia of Quaternary Science. Elsevier, 2337–2356.

20.

Gaillard

Digerfeldt

(1991) Palaeohydrological atudies and their contribution to palaeoeological and palaeoclimatic reconstructions. In: Berglund

(ed.) The Cultural Landscape During 6000 Years in Southern Sweden – The Ystad Project. Ecological Bulletin 41: 275–282.

21.

Galop

Vannière

Fontugne

(2002) Fires and human activities since 4500 BC on the northern slope of the Pyrénées recorded in the peat bog of Cuguro (Central Pyrennes). In: Thiebault

(ed.) Charcoal Analysis, Methodological Approaches, Palaeocological Results and Wood Uses. BAR International Series 1063, 1043–1051.

22.

Grimm

(1987) CONISS. A FORTRAN 77 program for stratigraphically constrained cluster analysis by the method of incremental sum of squares. Computers and Geosciences 13: 13–35.

23.

Grimm

(1990) TILIA 2.0. Software. Springfield: Illinois State Museum.

24.

Grimm

(2004) Tilia Graph v. 2.0.2, Software. Springfield: Illinois State Museum, Research and Collection Center.

25.

Hannon

Gaillard

(1997) The plant macrofossil record of past lake-level changes. Journal of Paleolimnology 18: 15–28.

26.

Harrison

Digerfeldt

(1993) European lakes as palaeohydrological and palaeoclimatic indicator. Quaternary Science Reviews 12: 223–248.

27.

Innes

Blackford

Simmons

(2004) Testing the integrity of fine spatial resolution palaeoecological records: microcharcoal data from near-duplicate peat profiles from the North York Moors, UK. Palaeogeography, Palaeoclimatology, Palaeoecology 214: 295–307.

28.

Intergovernmental Panel on Climate Change (2007) Cambio climático 2007: Informe de sínteis. Contribución de los Grupos de trabajo I,II,III al Cuarto Informe de evaluación del Grupo Intergubernamental de Expertos sobre Cambio Climático. Ginebra: Cambridge University Press.

29.

Julià

Burjachs

Dasí

Mezquita

Miracle

Roca

. (1998) Meromixis origin and recent trophic evolution in the Spanish mountain lake la Cruz. Aquatic Sciences 60: 279–299.

30.

Long

Whitlock

Bartlein

Millspaugh

(1998) A 9000-year fire history from Oregon Coast Range, based on a high resolution charcoal study. Canadian Journal of Forestry Research 28: 774–787.

31.

Lozano-Sahuquillo

(2002) Cañada del Hoyo: su personalidad histórica. Cartagena: Lozano-Sahuquillo (ed.), 224 pp.

32.

Mann

Zhang

Rutherford

Bradley

Hughes

Shindell

. (2009) Global signatures and dynamical origins of the Little Ice Age and the Medieval Climate Anomaly. Science 27: 1256–1260.

33.

Marlon

Cui

Gaillard

M-J

McWethy

Walsh

(2010) Humans and fire: Consequences of anthropogenic burning during the past 2 ka. PAGES News 18: 80–82.

34.

Olsson

Gaillard

M-J

Lemdahl

Greisman

Lanos

Marguerie

. (2010) A continuous record of fire covering the last 10,500 calendar years from southern Sweden – The role of climate and human activities. Palaeogeography, Palaeoclimatology, Palaeoecology 291: 128–141.

35.

Peñalba

(1994) The history of the Holocene vegetation in northern Spain from pollen analysis. Journal of Ecology 82: 815–832.

36.

Pitkanen

Lehtonen

Huttunen

(1999) Comparison of sedimentary microscopic charcoal particle records in a small lake with dendrochronological data: Evidence for the local origin of microscopic charcoal produced by forest fires of low intensity in Eastern Finland. The Holocene 9: 559–568.

37.

Ramsey

(1995) Radiocarbon calibration and analysis of stratigraphy. The OxCal program. Radiocarbon 37: 425–430.

38.

Ramsey

(2001) Development of the radiocarbon calibration program. Radiocarbon 43: 355–363.

39.

Roberts

Stevenson

Davis

Cheddadi

Brester

Rosen

(2004) Holocene climate, environmental and cultural change in the circum-Mediterranean region. In: Battarbee

. (eds) Past Climate Variability through Europe and Africa. Dordrecht: Springer, 343–361.

40.

Rodrigo

Barriendos

(2008) Reconstruction of seasonal and annual rainfall variability in the Iberian Peninsula (16th–20th centuries) from documentary sources. Global and Planetary Change 63: 243–257.

41.

Rodrigo

Esteban-Parra

Pozo-Vazquez

Castro-Diez

(1999) A 500-year precipitation record in Southern Spain. International Journal of Climatology 19: 1233–1253.

42.

Romero-Viana

Miracle

López-Blanco

Cuna

Vilaclara

García-Orellana

. (2009) Sedimentary multiproxy response to hydroclimatic variability in Lagunillo del Tejo (Spain). Hydrobiologia 631: 231–245.

43.

Rubio-Dobón

Melendo

JDV

(2005) Estudio de la evolución de régimen hidrológico en zonas húmedas drenadas: los humedales del Cañizar (Provincia de Teruel, España). Investigaciones geográficas 38: 47–63.

44.

Sánchez-Goñi

Hannon

(1999) High altitude vegetational pattern on the Iberian Mountain Chain (North-Central Spain) during the Holocene. The Holocene 9: 39–57.

45.

Santos

Vidal-Romaní

Jalut

(2000) History of vegetation during the Holocene in the Courel and Queixa Sierras, Galicia, northwest Iberian Peninsula. Journal of Quaternary Science 15: 621–632.

46.

Schoch

Pawlik

Schweingruber

(1988) Botanische Makrorestes. Bern: Haupt.

47.

Street-Perrot

Harrison

(1985) Lake levels and climate reconstruction. In: Hecht

(ed.) Paleoclimate Analysis and Modeling. New York: Wiley, 291–340.

48.

Tinner

(2003) Size parameters, size class, distribution and area–number relationship of microscopic charcoal: Relevance for fire reconstruction. The Holocene 13: 499–505.

49.

Tinner

Conedera

Ammann

Gäggeler

Gegye

Jones

. (1998) Pollen and charcoal in lake sediments compared with historical documented forest fires in southern Switzerland since ad 1920. The Holocene 8: 31–42.

50.

Tinner

Conedera

Ammann

Lotter

(2005) Fire ecology north and south of the Alps since the last Little Ice Age. The Holocene 15: 1214–1226.

51.

Vannière

Colombaroli

Chapron

Leroux

Tinner

Magny

(2008) Climate versus human-driven fire regimes in Mediterranean landscapes: The Holocene record of Lago dell’Accesa (Tuscany, Italy). Quaternary Science Reviews 27: 1181–1196.

52.

Vicente

Miracle

(1984) Distribution od photosynthetic organism in a temporal stratified karstic pond near Cuenca, Spain. Verhandlungen International Verein Limnologie 22: 1704–1710.

53.

Viglizzo

Roberto

Lertora

Gay

Bernardos

(1997) Climate and land-use change in field-crop ecosystems of Argentina. Agriculture, Ecosystems & Environment 66: 61–70.