Late Holocene climate variations and anthropogenic influence recorded in a sediment core from Pozzines du Renoso, Corsica, France.

(1)

1

Late Holocene climate variations and anthropogenic influence recorded in a

sediment core from Pozzines du Renoso, Corsica, France.

Martine Brunsting Bachelor Thesis Earth Sciences

University of Amsterdam Under supervision of:

Dr. Crystal McMichael (University of Amsterdam)

Dr. Brahimsamba Bomou and Dr. Christopher Castellani (University of Corsica) Figure 1: Renoso Lake

(2)

2

Abstract

The Mediterranean region is one of the most sensitive areas for climate change. A better

understanding of the interactions between human activity and climate in this area is therefore of special interest. The Island of Corsica, located in the western part of the Mediterranean, possesses many natural archives in its high altitude lakes where climatic fluctuations are recorded in lake sediments since the last glacial maximum. A 36 cm sediment core was taken from the Renoso Lake situated in the Regional Natural Park of Corsica and analysed at one centimetre resolution.

Multiproxy analyses were performed at the University of Corsica, the University of Lausanne and the University of Amsterdam, using geochemical analyses (Total Organic Carbon, phosphorus, nitrogen contents) and mineralogical analyses (X-Ray Diffraction, grain-size distribution) in order to

reconstruct the paleoenvironmental conditions during the late Holocene and evaluate anthropogenic influence since the last decades. High terrestrial inputs as well as prevailing larger grain sizes were found, indicating high energetic conditions. These findings are closely related to specific site

conditions and catchment basin characteristics, which highly influence climate at small spatial scales in the area. Human influence slightly increases around the lake which, was mainly due to increased tourism.

(4)

4

Introduction

Although the Holocene (11,500 cal yr B.P. to the present) is characterized by a relatively stable climate compared to the strong fluctuations of the previous glacial periods, significant variations in climate can be distinguished during this epoch (Ramrath, 1999). Whereas previous climate variations were only due to natural variability, emergence of civilizations from the early Holocene impacted natural energy balances influencing more recent climate variations (Hass, 1995). In the last hundred years, anthropogenic influence has strongly increased and is widely accepted to be an important factor of global climate change.

The Intergovernmental Panel on Climate Change reported in 2007 that the Mediterranean region is one of the most sensitive areas for climate change. Model simulations predict a precipitation decrease exceeding 25-30%, increased occurrence of hot extremes and annual warming exceeding 4-5 °C (Giorgi, 2008). Within the area, the west Mediterranean islands are an obstacle to eastward-moving cyclones and they are a site capable of recording climate variations through stable isotopes (Van Geldern et al., 2014). Besides, these islands possess many undisturbed natural archives in their high altitude lakes, providing complete sediment sequences (Rivier, 1996). Therefore, Mediterranean islands were recently chosen for palaeoclimate reconstructions including hydrogen and oxygen isotopes analysis in several water bodies on Corsica (Van Geldern et al., 2014), lake sediment

deposits analysis on Tuscany and Sicily (Magny et al., 2007, 2011), marine sediment records (Rodrigo-Gamiz et al., 2011; Toucanne et al., 2012; Martinez-Ruiz et al., 2015), or in the speleothems records in Sardinia (Antonioli et al., 2003), and in Italy (Zanchetta et al., 2014).

However, temperature records for the past centuries, which can contribute to a better understanding of how future climate change will affect the system, are sparse for this region (Szymczak et al., 2014). Szymczak et al. (2014) recently reconstructed the climate of the last 600 years using the multi-parameter of black pinewood growing in relationship with stable isotopic data. Nevertheless, the lack of other climate proxies in the area stays problematic and prevents the further understanding of the past climate variability in this area.

Lacustrine laminated sediments provide some of the most refined records of climate changes affecting continental areas since the Last Glacial Maximum (Oldfield, 1977; Kelts & Talbot 1990). Whereas deep-sea sediment deposits are often disturbed by high-energy movements, sediment deposits taken from lakes can provide continuous records without hiatus. Besides, lacustrine sediments enable us to distinguish between global and local climate change effects since they are both directly influenced by human impact and global climate change (Kelts & Talbot, 1990).

Mechanical, chemical and biological conditions of high altitude lakes vary drastically between warm and cold periods. Hence, sediment records of these lakes provide clear indications on climate variability (Garcon et al., 2012). For example, slight changes in temperature influence the acid-base balance in the lake(Sommaruga-Wograth, 1997) which again influences primary productivity and thus eutrophication levels. Climate also determines vegetation cover within the catchment basin, which influences the erosion rates and thus sedimentation.

This study aims to find out whether a sediment core from the Renoso Lake shows evidence for recent human-induced climate change and/or supports theories of recent substantial drying in the

(5)

5

Mediterranean. This will be done by analyzing mineralogical and geochemical content on a lacustrine sediment core from the Renoso Lake (Corsica, France). Similar surveys have not yet been conducted in this area and can provide valuable information about anthropogenic activity and its impacts on the environment, in particular during the last century.

Geological & Climatic context

The island of Corsica is situated close to the European continent in the western Mediterranean Basin (see figure 2), at a distance of 80 km from the Italian coast and 160 km from the French coast. The area is located in the transition zone between the west wind drift and the subtropical high-pressure belt, which naturally leads to domination of descending motions and thussevere dry periods during summer (Toucanne et al., 2012). Many researchers state that this makes the Mediterranean extra susceptible for the prospected strong temperature increase and precipitation decrease; drought stress will be intensified and lead to a northward extension of dry periods (Giorgi, 2008; Jacobeit et al., 2005; Magny et al., 2011) . The present climate in Corsica is characterized by these subtropical Mediterranean-type conditions (Reille, 1997) with dry and warm summers (May to September) and

oceanic-type temperate, wet winters (October-April). Geology As becomes clear in figure 3, Corsica can be divided into five different geological

structures. A large part of the East coast is made up of Quaternary deposits. The North mainly exists out of schists with a small

sedimentary basin next to Cap Corse. The adjacent strip (marked black in figure 3) consist of sedimentary basins. In the South there is a small calcareous Miocene formation which is part of the Bonifacio Basin. However, the largest part of the island exists out of the so called ‘Hercyninan Corsica’, which is where the Renoso lake is situated. This part essentially consists of granite, rhyolite and other plutonic rocks (Reille, 1997) and makes up the highest mountains of the island. The action of glaciers in this area during the Last Glacial Maximum led to the formation of steep mountainous relief and numerous lakes like the Renoso Lake (Kuhleman, 2004). The presence of U-shaped valleys, cirques and moraines widely reveal glacial impact, independent from differences in lithology (Kuhleman, 2004). Whereas fluvial and glacial erosion deeply incised the valleys, ridges suffered low periglacial

overprint and erosion as can been seen in Figure 4 (ibid.). Figure 3: Schema of the simplified Geological structures of Corsica (Reille, 1997)

Figure 2: Corsica and its location in the Mediterranean area

(6)

6

The lake is situated at an elevation of 1,900 metres and was formed during the LGM (Gauthier, 1984). The lake is In a late state of hydrological succession which means that the largest part of the lake has been filled in with sediments forming a spongy peat landscape. Little ponds and rivulets, called pozzines, cut through the peat formations and are the only remaining water surfaces (Gauthier, 1984 ). Water input comes from melting water and precipitation.

Vegetation

Corsican vegetation varies from grasslands to dense forests and is mainly determined by altitude (Reille, 1997). In the In the supra-mediterranean vegetation belt, the summer drought is shorter than in lower zones which is favourable to species such as P. nigra ssp. laricio and Quercus pubescens. In cooler areas, mesophilous species such as Alnus cordata, Ostrya carpinifolia and Castanea sativa are commonly found. In the thermo- and mesomediterranean zones, Q. ilex is dominant, often mixed with high shrubs (Gamisans et al., 1995).

Corsican forest dynamics are mainly determined by altitude, species interaction but also human influence (ibid.). Betula spatula is a heliophilous species which participates in dynamics leading to the formation of climax forests as it can germinate and develop easily in deforested areas (Reille, 1997). The species gradually disappears when the forest closes and colonizes very rapidly openings in the Fagus, Abies or P. nigra ssp. laricio forests, gradually forming a pre-forest stage (ibid.).

At higher altitudes the treeline is often replaced by a specific kind of Mediterranean shrubs called

maquis including Arbutus unedo, Erica scoparia, Pistacia lentiscus, Olea europaea (Gamisans et al.,

1995). When certain tree species disappear, this maquis re-establishes and can be considered as climax vegetation. Human action is often accountable for the openings in forested areas (Reille, 1997).

Figure 4: Google Earth imagery of Renoso Lake and its catchment basin showing glacial impacts. 4a shows cirque formations, 4b shows glacial incisions, 4c shows an U shaped valley right next to Renoso Lake and 4d shows Renoso Lake itself that could also be classified as a U shaped valley. Furthermore glacial polish is visible in all four images.

(7)

7 Figure 6: Google Earth imagery of Renoso Lake

dated November 2003

Figure 7: Google Earth imagery of Renoso Lake dated December 2013

Figure 5: Pozzines. 5a: Cross section of pozzines; 5b top view of pozzines; 5c pozzines de Renoso.

Between 1600 and 2200 meter altitude acid peat formations studded with small pockets filled with water from snow melt and precipitation are found. They are called pozzines in the Corcican language. Although these formations are considered as typically Corsican, they also exist in other European mountain ranges as well as in North Africa and Central Asia. These formations mainly consist of hygrophilous grasslands on completely or partially filled in glacial lakes (Gamisans et al., 1995). They are usually dominated by Graminaea, Cuperacaea and Juncacaea and sometimes with patches of Sphagum. Lake Renoso, is one of the numerous pozzine formations found on Corsica representing a late stage of hydrological succession of a glacial lake.

The catchment basin is fairly green and covered with low shrubs and grasses. Imagery from Google Earth indicates rapid changing vegetation cover as a result of filling in of the pozzines which provides fertile soils for many grass species (see figures 5 & 6). In the 2013 imagery more shrubs are visible and especially the soils surrounding the pozzines are now covered with vegetation whereas in 2003 during the same season, bare soil is exposed around the pozzines.

The catchment basin does not include any industries or settlements. However, tourism on the island is growing fast and starts to leave its traces even at remote sites such as Renoso Lake. A small refuge and a camping site near the pozzines are welcoming more tourists with the year (personal

(8)

8

Climate

Besides the Mediterranean climate that characterizes especially the coastal areas in Corsica, local processes induced by the complex physiography of the region and the presence of a large body of water (the Mediterranean Sea) affect local climate conditions. Temperature on the island is strongly determined by altitude, while precipitation shows a more diverse pattern with the north-eastern part of the mountains being somewhat drier than the southern and western coastal mountain ranges (Reille, 1999). As the Renoso lake is situated at a high altitude, the temperatures are generally lower than on the rest of the island and its situation at the windward side causes higher precipitation and more frequent high energy events than on lower altitudes. Furthermore, coastlines and vegetation cover of the region are well known to modulate regional climate signal at small spatial scales

(Kuhleman, 2004).Alpert et al. (2006) also suggest that anthropogenic and natural aerosols of central European, African and Asian origin could possibly influence Mediterranean climate characteristics. Interactions of all these processes at a wide range of spatial and temporal scales result in a variety of climate types and great spatial variability within the area (Kuhleman, 2004).

At the study site, during winter, snow is the main form of precipitation with a continuous snow layer from December to April from 1500 m a.s.l. upwards with a very short or even non-existent summer drought (ibid.).

Methodology

Sampling

A gravity corer was made by the technical staff of the University of Corsica. It consists of a tubular system with a valve, weighted with a mass, which was let down in the water and penetrated into the sediment. The valve allows the water to escape during the descending of the corer and during the rising it allows a suction effect, which catches the sediment in the tube.

A 36 cm long sediment core was taken at the outflowing part of the Renoso lake as can be seen in figure 8. After transportation to the laboratory of the University of Corsica, a small hole was made in the lower part of the tube to let exceeding water escape. After one day of drying, the core was split using an electrical circular saw to cut the tube and a thin wire was used to cut the sediment without smearing. After this, the core was precisely logged and imaged. Each centimetre of the sediment was taken out of the tube using a knife and a spoon and put in a synthetic box without cover. The samples were dried in an oven at 40 °C during 24 hours.

Analyses

One half of the core was used for grain size analysis. The different size fractions (sizes 63 um to 1 mm) were separated using a series of vertically stacked sieves. This column of nested sieves was placed in a mechanical shaker device using the humid mode. After shaking, each size fraction will be dried and weighted in order to construct a grain size distribution diagram.

The other half was used for geochemical analyses. The samples were crushed using a mechanical agate crusher for geochemical analyses.

(9)

9

Total phosphorus analyses were also conducted at the University of Lausanne using the ascorbic acid molybdate blue method (Eaton et al., 1995). An amount of 100 mg for each sample powder was mixed with 0.5 ml of MgNO3 (1 M) into decontaminated glass bottles in order to degrade organic

matter. After drying in oven at 100°C, the samples were heated in a furnace at 550°C for 2h30. After cooling, 10 ml of HCl (1M) was added to each sample and bottles were placed in ultrasonic disaggregation bath in order to liberate the phosphorus from the sediment matrix. After 14h of constant shaking, the samples were filtered and diluted by ten times. 90 μl of molybdate mixing reagent and ascorbic acid will be added into 3 ml of each sample solution. The intensity of the blue colour, depending to the phosphorus concentration, was quantified with an UV/Vis Perkin Elmer Lambda 25 spectrophotometer. The concentration of PO4 in mg/L was obtained by calibration with

internal standard solutions with a precision better than 5%.

Total organic carbon (TOC) and nitrogen (N) abundances were determined with a CHNS Elemental Analyser Thermo Finnigan Flash EA 1112 on decarbonated and oven-dried bulk sediment samples. The samples will be heated up to 900°C and the combustion products were extracted into a

chromatographic column, in which they were converted into simpler components. The components were measured by a thermal conductivity detector and recalculated in wt.%. Analytical precision and accuracy were determined by replicate analyses and by comparison with Organic Analytical Standard composed of purified DLMethionine, and are better than 0.1% (1σ) and 0.01% (1σ) for carbon and nitrogen determinations, respectively. The amount of TOC were recalculated in wt.% by taking into account the amount of CaCO3 wt.% measured on the same samples.

C/N ratios were calculated to determine the source of organic material found in the sediments. Terrestrial plants have a C/N Ratio larger than 10 whereas aquatic plants have a C/N Ratio between 4 and 10 (Meyers and Teranes, 2003). However, degradation within the lake may alter C/N ratios slightly which could cause bias (ibid.).

(10)

10

Results

The sediment record is clearly laminated showing units of fairly large grain sizes alternated with units containing finer grains sizes rich in organic material in the lower part of the sediment record. The larger grains are lighter coloured and give a good

representation of the bedrock in the catchment area which is mainly granite. Little twigs and plant material can easily be distinguished in the darker coloured areas. The clearest laminations are visible at 17 cm , 18-19 cm, 24-25cm and 33-34 cm (see figure).

In the upper part of the record, larger grains are visible all through and they are well sorted not showing any clear laminations. The colour of the upper part is slightly lighter than the lower part.

Grain size

The grain size study shows that the distribution of the size fractions is not constant (fig ). As already became clear from the log and imagery, the largest grain size fractions are well represented in this record.

The

laminations are clearly recognizable in the grain size distribution.

Larger grain sizes are generally interpreted to indicate higher energy conditions. These seem to prevail during the interval alternated by small peaks in smaller grain sizes which are usually deposited in a lower energy condition. Figure 3 suggests that lower energy conditions prevail from 1- 3 cm, 16 to 18 cm and 21 to 27 cm as these show higher values for smaller grain fractions.

Figure 11: Results of geochemical analysis showing curves for C, N, C/N and P and the correspondence of peaks with laminations in the sediment sequence. Red lines point out the correspondence of large grain size laminations with low values of organic compounds and high C/N values.

Figure 10: Grain size distribution Figure 9: Sediment log, clear laminations of larger grains at 17cm, 18-19 cm, 24-25 cm and 33-34 cm.

(11)

11

The carbon measurements contain both organic and inorganic carbon.However, as XRD analyses

didn’t show any peaks in calcite (CaCO3), the carbon values can be assumed to only contain organic carbon. The C values range between 1.76 and 28.13 with a mean of 10.21 and are high especially in the lower part of the record with peak at 32, 30, 27 and 23 cm of respectively 28.13, 26.82, 24,2 and 19,74 percent.

N values range between 0.04 and 1.35 percent with a mean of 0.48. There are 3 peaks recognizable that parallel those of the C values at 30, 27 and 23 cm of respectively 1.35, 1.3 and 1.15 percent.

Logically following from these high C values and low N values, C/N values are very high ranging between 18.773 and 49.52 percent. A slightly increasing trend can be seen. The phosphorous curve first parallels with the carbon and nitrogen curves but in the upper part it shows an increase whereas the other organic compound more or less stabilize. The peaks of C, N and P are systematically

synchronous but inversely correlated with lowest levels of C/N. P values range from 182 ppm to 843 ppm with a mean of 497 ppm.

Two different units could be interpreted from these results where all organic compounds in the lower part of the sequence seems to follow a similar variable pattern and the second upper part where C and N values seem to stabilize but where phosphorous shows a steady increase. The laminations with larger grain sizes correspond with the peaks in organic compounds.

Discussion

Climatic tendencies

The clear laminations of larger grain sizes at 25, 23 and 17 cm as can been seen in figure 9 suggest the occurrence of high energy events. These larger grain sizes correspond with peaks for C/N values (see figure 9), thus high terrestrial inputs, which supports this theory. The prevailing larger grain sizes also indicate moderate to high precipitation which is not surprising as the lake is situated at the windward site of the mountain range.

Several other studies in the area confirmed a reduction of precipitation and increased warming over the last decades (Giorgi, 2008; Jacobeit et al., 2005, Magny et al., 2011). A decrease in high

precipitation events could be interpreted from the grain size analysis as clear laminations with larger grains sizes are not present in the upper part of the sediment record and the amount of larger grain size slightly decreases. However, this could also be explained by the higher vegetation cover that diminishes the erosion effect of such precipitation events. As C/N values are high in the upper part, and sediments thus mainly come from terrestrial sources, the latter one seems more valid.

Corsican precipitation reconstruction of Szymczak et al. (2014) shows the highest number of wet summers during the last 50 years, pointing to a trend towards warmer and wetter summer conditions. The large grain sizes and high C:N found as can been seen in figure 10, supports that theory. However it is very hard to actually compare the data and put them into context as the sediment record was not dated.

The prevailing larger grain sizes in this area differs from another similar study done in the same area at Nino Lake (Passera, 2016) and can partially be explained by the shape of the catchment basin which is very steep and narrow. The Nino lake is situated about 160 meters lower than the Renoso

(12)

12

lake and its catchment basin is more stretched out. The Renoso lake catchment basin is steeper and detritus that erodes from bedrock therefore only travels a short and steep distance before being deposited in the lake resulting in fairly larger grain sizes. The angular shape of the grains also suggest short sedimentation distances.

It must be noted however that organic material has not been removed from the samples resulting in small aggregates of silica and organic matter which may have contributed to a slightly higher weight for the largest fractions. The location of the core also influences the grain sizes that were found. The core was taken in a shallow part of a pozzine. On such a location, coarser sediments more rapidly accumulate than at a deeper part of a lake.

C/N Ratio

Very high C/N ratios were found ranging between 18.7 and 49.5 indicating the deposition of terrestrial plant derived organic matter with limited input of algal derived organic matter. Values slightly increase from deeper layers up to the higher part. This data suggests that terrestrial plants dominated through all of the interval and continue to increase in abundance. This is not surprising for a lake that is in a late stage of hydrological succession. The slowly filling in of the lake with sediments, provides fertile peat soils for terrestrial plants. This theory is also supported by the imagery found on Google Earth showing increase in vegetation cover as discussed in the climatic context (page 6). Furthermore, as the water level drops, primary production decreases which naturally results in a higher C/N ratio. Drops in C/N values as seen at 25 and 34 cm in the lower part and frequently in the upper part (see figure 11) indicate periods of high primary productivity and thus a higher level of eutrophication. However, terrestrial inputs dominate though all of the record.

Some peaks with even higher values can be distinguished in the sediment record. Higher C values mean higher influx of eroded soils and terrestrial plant material, which could indicate higher precipitation. The peaks with even higher values, at 25 and 13 cm, could indicate sediment inputs from surrounding forests during a more humid time interval. The peaks also correspond with a peak in large grain sizes, which supports the theory of higher precipitation at that time interval.

Human influence

Phosphorous indicates the level of eutrophication, and primary productivity as well as the preservation of organic matter. In this area, the source of eutrophication is mainly from human activity and therefore P values are used as an indicator of human influence. P values are higher in the

lower part of the sediment core, steeply decrease at 19 cm and from there on slightly increase again reaching the highest peak in the sequence at 3 cm. As, the samples are not dated, it is very difficult to interpret this line. However, it is noticeable that P inputs first parallel the other organic compound inputs (C & N), at 18cm all values sharply decrease after which C and N values more or less stabilize but P values sharply increase. This could indicate higher human impacts in the latter part of the sediment record which corresponds with growing tourism in this area.

Figure 11: Phosphorous levels at Renoso Lake and at Nino Lake.

(13)

13

When comparing the P values with the data from a similar study in the same area and at the same time at Nino Lake (personal communication with Anais Passera, 2016), the trend more or less corresponds showing similar values fluctuating between 250 and 750 ppm in the lower part followed by a steady increase from the middle section upwards (see figure 11). At Nino Lake this increase is a lot sharper, which can be explained by the fact that it is an easier accessible site for tourists and also horseback riding is offered in this area whereas the Renoso Lake is more difficult to reach and is thus frequented by less tourists.

The Nino Lake shows lower C/N values and smaller grain sizes, which can be explained by the different catchment basin but also because the Nino lake is in an earlier state of hydrological succession.

Influence of local site conditions

An interesting finding of this study is the high importance of local site conditions on the outcome of multi proxy sediment analysis. In paleoclimatic studies, effects of local site conditions should

therefore not be underestimated. The Renoso Lake and the Nino Lake are both Corsican high altitude lakes situated in ‘Hercynian Corsica’ but grain sizes and organic compound found in their sediment records differ greatly. These differences can be explained by the differently shaped catchment basins, different positioning within a mountain range, different vegetation cover in their catchment basins, slightly different altitude and the different hydrological succession stage of the lake itself. Interactions of all these factors at a wide range of temporal scales result in high variability of climatic conditions at small spatial scales.

Recommendations

This was an explorative research and the project on Corsican lakes will be continued. For this continuation it is recommended to take multiple cores at different locations because different size fractions are likely to accumulate at different depths in a lake. Also, a longer sequence as well as dating of the sequence will make it a lot easier to interpret the results and put them into the context of past research.

In further research, grain shape and composition of quartz/feldspar/mica’s could be included to better understand structural maturity and provenance of the sediments. This could help to better understand the precipitation regime.

Conclusion

A 40cm sediment sequence with an 1cm resolution was analysed using a mineralogical and geochemical approach as to analyse the climate variability and anthropogenic impact of the last century.

The Renoso lake is situated at a highly dynamic site and responds rapidly to climatic changes. High terrestrial inputs characterize the sediment record, which is explained by the late hydrological succession state of the lake. Moderate precipitation is alternated with high energy events with higher precipitation that resulted in higher deposition of terrestrial plant derived organic matter from surrounding forests. These events only occurred in the lower unit of the sediment sequence.

Human influence slightly increased over the last years. This pattern is also visible in the results from a similar study at Nino Lake. This finding corresponds with the increased tourism in the area.

(14)

14

An interesting finding of this study is that the interaction of different local factors, such as vegetation cover, hydrological succession stage and steepness of catchment basin, highly influence climate at small spatial scales. As the Renoso Lake is situated in a very steep catchment basin, grain sizes are naturally larger than in sediments from the same region with more gradual catchment basins. Also, the late succession state of the lake explains the high C:N ratio. This study therefore underlines the high importance of local site conditions on the outcome of paleo climatic studies as well as the high climate variability at small spatial scales on Corsica.

Acknowledgements

I thank Chrystal McMichael, Brahimsamba Bomou, Cristopher Castellani and Will Gosling for

feedback and suggestions. Assistance in the field by Anaïs, Topher and Batti was greatly appreciated. I would like to extend my thanks to the staff of the laboratory of the Hydrology department of the University of Corsica providing me with all resources needed and to Brahimsamba Bomou and the Geological department of the University of Lausanne for geochemical analyses. A last thanks goes to Anaïs for tackling this project with me and reading my work and to my parents for encouragement and support throughout my studies.

(15)

15

References

Brewer, S., Alleaume, S., Guiot, J., & Nicault, A. (2007). Historical droughts in Mediterranean regions during the last 500 years: a data/model approach.Climate of the Past, 3(2), 355-366.

Gamisans, J., & Jeanmonod, D. (1995). La flore de Corse: Bilan des connaissances, intérêt patrimonial et état de conservation. Ecologia mediterranea, 21(1-2), 135-148.

Garçon, M., Chauvel, C., Chapron, E., Faïn, X., Lin, M., Campillo, S., ... & Charlet, L. (2012). Silver and lead in high-altitude lake sediments: Proxies for climate changes and human activities. Applied

Geochemistry, 27(3), 760-773.

Gauthier A., Roche B., Frisoni G.F. (1984). Contribution à la connaissance des lacs d’altitude de la Corse. Parc Naturel Régional de la Corse. Rapport de 221 p.

Giorgi, F., & Lionello, P. (2008). Climate change projections for the Mediterranean region. Global and

planetary change, 63(2), 90-104.

Hass, H. C., & Kaminski, M. A. (1995). Change in atmospheric and oceanic circulation reflected in North Sea sediments during the late Holocene. Zbl. Geol. Paläont, 51-65.

Jacobeit, J., & Dünkeloh, A. (2005). The Eastern Mediterranean transient in relation to atmospheric circulation dynamics. Hydrogeologie and Umwelt, 33, 1-8.

Kelts, K., & Talbot, M. (1990). Lacustrine carbonates as geochemical archives of environmental change and biotic/abiotic interactions. In Large Lakes (pp. 288-315). Springer Berlin Heidelberg. Kuhlemann, J., Frisch, W., Székely, B., Dunkl, I., Danišik, M., & Krumrei, I. (2004). Würmian maximum glaciation in Corsica. Austrian Journal of Earth Sciences, 97, 68-81.

Leroux, A., Bichet, V., Walter-Simonnet, A. V., Magny, M., Adatte, T., Gauthier, É., ... & Baltzer, A. (2008). Late Glacial-Holocene sequence of Lake Saint-Point (Jura Mountains, France): Detrital inputs as records of climate change and anthropic impact. Comptes Rendus Geoscience,

340(12), 883-892.

Magny, M., Vannière, B., Calo, C., Millet, L., Leroux, A., Peyron, O., ... & Tinner, W. (2011). Holocene hydrological changes in south-western Mediterranean as recorded by lake-level fluctuations at Lago Preola, a coastal lake in southern Sicily, Italy. Quaternary Science Reviews, 30(19), 2459-2475.

Martinez-Ruiz, F., Kastner, M., Gallego-Torres, D., Rodrigo-Gámiz, M., Nieto-Moreno, V., & Ortega-Huertas, M. (2015). Paleoclimate and paleoceanography over the past 20,000 yr in the Mediterranean Sea Basins as indicated by sediment elemental proxies. Quaternary Science

Reviews, 107, 25-46.

Meyers, P. A. (2003). Applications of organic geochemistry to paleolimnological reconstructions: a summary of examples from the Laurentian Great Lakes.Organic geochemistry, 34(2), 261-289. Oldfield, F. (1977). Lakes and their drainage basins as units of sediment-based ecological study.

Progress in Physical Geography, 1(3), 460-504.

(16)

16

Ramrath, A., Sadori, L., & Negendank, J. F. (2000). Sediments from Lago di Mezzano, central Italy: a record of Lateglacial/Holocene climatic variations and anthropogenic impact. The Holocene,

10(1), 87-95.

Reille, M., Gamisans, J., BEAULIEU, J. L., & Andrieu, V. (1997). The late‐glacial at Lac de Creno (Corsica, France): a key site in the western Mediterranean basin. New Phytologist, 135(3), 547-559.

Reille, M., Gamisans, J., Andrieu‐Ponel, V., & De BEAULIEU, J. L. (1999). The Holocene at Lac de Creno, Corsica, France: a key site for the whole island. New Phytologist, 141(2), 291-307. Rivier, B. (1996). Lacs de haute altitude: méthodes d'échantillonnage ichtyologique (Vol. 11). Editions

Quae.

Rodrigo-Gámiz, M., Martínez-Ruiz, F., Jiménez-Espejo, F. J., Gallego-Torres, D., Nieto-Moreno, V., Romero, O., & Ariztegui, D. (2011). Impact of climate variability in the western Mediterranean during the last 20,000 years: oceanic and atmospheric responses. Quaternary Science Reviews,

30(15), 2018-2034.

Solomon S, Qin D, Manning M, Chen Z, Marquis M, Averyt KB, Tignor M, Miller HL, editors. Climate change 2007. The physical science basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge (UK): Cambridge University Press.

Sommaruga-WÖgrath, S., Koinig, K. A., Schmidt, R., Sommaruga, R., Tessadri, R., & Psenner, R. (1997). Temperature effects on the acidity of remote alpine lakes. Nature, 387(6628), 64-67.

Szymczak, S., Joachimski, M. M., Bräuning, A., Hetzer, T., & Kuhlemann, J. (2012). A 560 yr summer temperature reconstruction for the Western Mediterranean basin based on stable carbon isotopes from Pinus nigra ssp. laricio (Corsica/France). Climate of the Past, 8(5), 1737-1749.

Toucanne, S., Jouet, G., Ducassou, E., Bassetti, M. A., Dennielou, B., Minto'o, C. M. A., ... & Mulder, T. (2012). A 130,000-year record of Levantine Intermediate Water flow variability in the Corsica Trough, western Mediterranean Sea. Quaternary Science Reviews, 33, 55-73.

Van Geldern, R., Kuhlemann, J., Schiebel, R., Taubald, H., & Barth, J. A. (2014). Stable water isotope patterns in a climate change hotspot: the isotope hydrology framework of Corsica (western

Mediterranean). Isotopes in environmental and health studies, 50(2), 184-200. environmental and

Late Holocene climate variations and anthropogenic influence recorded in a sediment core from Pozzines du Renoso, Corsica, France.