Paleolimnology of Iberian Lakes: the 2004 LRC-IPE survey
Iberian lacustrine sediment records: responses to past and recent global changes in the Mediterranean region
Blas L. Valero-Garcés and Ana Moreno
Pyrenean Institute of Ecology, Spanish National Research Council, Apdo 13034, E-50080 Zaragoza, Spain
Keywords: Paleolimnology, Iberian Peninsula, Holocene, Paleoclimate, Human impact, Mediterranean
Introduction
In spring 2004, the Pyrenean Institute of Ecology (Spanish Research Scientific Council, IPE- CSIC) and the Limnological Research Center (LRC), University of Minnesota, USA, organized the first paleolimnological field expedition to the Iberian lakes. Coring was facilitated by use of a platform and a Kullenberg coring system. Doug Schnurrenberger, Mark Shapley and Anders Noren from the LRC were responsible for technical and logistical aspects of the coring expedition. The IPE – CSIC selected the lakes using location and accessibility as criteria. Kerry Kelts, the former director of the LRC, was the inspiration for this campaign. Unfortunately, Kerry did not live to see the success of the expedition. He passed away on February 8th, 2001 at the age of 54, after a long and courageous battle with Hodgkins Disease. Kerry played a major role in the development of limnogeological studies in Spain during the 1980s and 1990s. He came to Spain in 1985 to attend the 6th International Association of Sedimentologists meeting in Lleida and again the next year to teach a short course in Barcelona. Even after becoming ill, Kerry participated actively in the first two meetings of the Southern European Group of the European Lake Drilling Program, in Zaragoza (1999) and Lisbon (2000). We discussed the Iberian lakes expedition with Kerry many times. Finally, in 2004, it became a reality. This collection of papers is a tribute to Kerry’s vision.
Most paleolimnological studies in Spain prior to the 2004 expedition had been conducted on peat bogs, shallow water bodies, dry lakes or even outcrops. For the first time, as part of this expedition, long sediment cores were recovered from relatively deep lakes (11-50 m water depth) in a variety of climatic areas and geographic regions. During one month, the scientific team traveled more than 6,000 km, surveyed eight lakes and recovered 200 m of sediment core. The selected lakes were: (1) Estanya and Montcortés in the Pre-Pyrenean Range, (2) Caicedo de Yuso - Arreo in the western Ebro Basin, (3) Enol in the Cantabrian Mountains, (4) Sanabria in northwestern Spain, (5) El Tobar and Taravilla in the Iberian Range, and (6) Zoñar in the Guadalquivir River Basin (Fig. 1). Cores were shipped to the LRC where initial core descriptions, including magnetic susceptibility measurements, were done in fall 2004. This “core library” represents the best terrestrial archive of climate variability on the Iberian Peninsula. A multidisciplinary team has been working on the cores for the last few years, studying sedimentological, geochemical and biological variables. Several papers from Estanya (Morellón et al. 2008, 2009, 2010), Taravilla (Valero-Garcés et al. 2006; Moreno et al. 2008) Enol (Moreno et al., 2010) and Zoñar (Martín–Puertas et al. 2008, 2009, 2010, 2011) have been published. Three PhD theses using those lake records have been defended (Martín-Puertas 2008; Morellón 2009; Corella 2011) and two more are in progress. The sequences have been studied using a multidisciplinary approach and the chronologies are well constrained with AMS-14C, 210Pb and 137Cs dates, and in some cases (Arreo, Montcortès), varve counting. This Iberian dataset, as a whole, provides high-resolution reconstructions of past climate and environmental variability on the Iberian Peninsula at several timescales: millennial since the last glacial maximum (Enol, Sanabria, Estanya), centennial during the Holocene (all records) and decadal resolution during the last few centuries (Enol, Sanabria, Arreo). In addition, they show a comprehensive picture of recent environmental and climate changes, i.e. information needed to implement policies to manage and preserve these aquatic systems.
The number of extant lakes in Spain is relatively small, compared with other countries, but they occur in a variety of geographic, climatic and ecologic settings (Alonso 1998). There are three main lake types, distinguished by their mode of origin: 1) lakes in the mountains (Cantabrian, Pyrenees, Central System, Iberian Range) originated by glacial activity, 2) karstic lakes that owe their origin to exokarstic activity (travertine damming as in Taravilla, Ruidera Lakes in the Guadiana River Basin) or to dissolution of evaporites or carbonates (many lake systems such as Banyoles, Montcortès, Arreo, El Tobar, Cañada del Hoyo, Zoñar), and 3) ephemeral, saline lakes in the Ebro, Duero and Tajo River Basin, that arose from a combination of wind erosion and karstic processes. Paleolimnological studies in Spain started in the late 19th century in some small ephemeral lakes (Martín-Donayre 1873; Calderón 1888; Hernández Pacheco 1900). Martín-Donayre (1873) had described the extensive mudflats of Gallocanta Lake (“the terrain surrounding the lake is extremely muddy”) as well as the correlation between the amount of salt extracted from La Salineta Lake (Zaragoza Province) and annual rainfall. In the early 20th century, the two largest lakes in Spain, Sanabria in Zamora Province (Ciria and Vinent 1908; Taboada 1913), and Banyoles in Girona Province (see detailed references in Julià 1977), were surveyed. In spite of these early works, a limnological tradition was never established. Only later, with the studies of Ramón Margalef, including his classic book “Limnología” published in 1983, did pioneering limnological work in Spain really begin.
For decades, the Spanish scientific community did not have the technical capability to retrieve cores from relatively deep lakes, and most paleolimnological studies in the region were restricted to short cores. The exception was the IPE floating platform, developed during the 1980s (Montserrat Martí, 1992) and used successfully in glacial Lake Tramacastilla (Jalut et al. 1992) and karstic Lakes Estanya (Wansard et al. 1998) and La Cruz (Julià et al. 1998). An integrated approach to Quaternary lake basin studies, including multiple cores, seismic surveys and multidisciplinary techniques, was not undertaken until the last decade. Recently, paleolimnological research on Iberian lakes has increased, as is evident in the number of presentations at the last few Paleolimnology Symposia and Limnogeological Congresses. A country with a tradition in recent lake ecology and pre-Quaternary limnogeology, has seen a new synergy, with expanded paleolimnological study of extant lakes. During the 2004 expedition, IPE-CSIC researchers provided the scientific basis for selecting coring sites, and scientists involved in the study of the lakes participated in the coring operations. Members of the community who joined the expedition were involved in the study of the cores, and came from many disciplines. The idea of a special JOPL issue on the Iberian lakes started with the IV Limnogeology Meeting, held in Barcelona in 2007. We now have a collection of papers that showcase the results of the 2004 expedition.
The new records presented here enabled us to address some fundamental questions about paleoenvironmental and paleoclimate evolution on the Iberian Peninsula, and to identify similarities and differences between Iberian records and those in northern Europe. In particular, these Iberian sequences demonstrate:
i) Maximum extent of alpine glaciers occurred prior to the global Last Glacial Maximum (LGM). Long cores from glacial lakes (Enol, Sanabria) have proglacial sediments older than 20 ka (the global LGM), demonstrating that the maximum local glacier extent in the Cantabrian and NW mountains occurred earlier than the LGM. After several decades of debate about possible early deglaciation in the Pyrenees and the mountains of southern Europe, these data support a distinctive history for Mediterranean alpine glaciers all over the Iberian Peninsula (IP).
ii) There were large hydrological changes in the IP associated with glacial, late glacial and Holocene climate phases. The Iberian region is key to answering some questions about climate variability and abrupt changes because of its location at middle latitudes, with strong Atlantic and subtropical influences. In particular, sediment sequences from karstic lakes show great hydrological sensitivity to past climate changes. Alpine and coastal lakes fail to record such large hydrological variations during the late glacial and Holocene. For example, the long record from Estanya (Morellón et al. 2008) shows desiccation phases at the onset of the Holocene, an arid “Mystery Interval,” and relatively humid phases during the LGM. All the records show large hydrological variability during the last 4,000 years. The closed-basin nature of some of these lakes probably amplifies the response to changes in moisture balance, but this enabled us to identify long, arid periods during the late glacial and short-term drought episodes during the late Holocene, events for which we have detailed and robust chronologies.
iii) Large changes in lake water balance occurred during historic times, i.e. since the Roman Period, with marked regional variability. In southern Spain, the Iberian–Roman period is the most humid phase during the last 4,000 years, but the northern sequences do not record this period as the most humid. Every record shows evidence for increased aridity during a period roughly synchronous with the Medieval Climate Anomaly (MCA), and colder and more humid conditions after the 15th century, coinciding with the Little Ice Age. The record from the LIA shows a complex internal paleohydrological structure and a strong link with solar irradiance is suggested by the coherence between periods of more positive water balance and phases of reduced solar activity in some records (Estanya, Taravilla). Changes in winter precipitation and dominance of NAO negative phases would account for wet LIA conditions in western Mediterranean regions. This well-established pattern in the IP has an antiphase relation with similar records in the eastern Mediterranean, which has led to the hypothesis of a “Mediterranean Oscillation” and illustrates the complexity of regional variability in the Mediterranean during the last few millennia (Roberts et al. 2011).
iv) There has been profound human impact on the lakes for the last millennium. Most lakes and wetlands in Spain have been exploited for water and other natural resources since the Neolithic. Almost all records show a large increase in sedimentation rate and sediment delivery to the lake, synchronous with medieval settlement, deforestation and farming of the watersheds. Sedimentation rates of 3 mm yr-1 during the last 2,000 yrs are common, and values of ~10 mm yr-1 over the last century were recorded in several lakes. Study of cores from the 2004 expedition show three main recent impacts on Spanish karstic lakes: i) sedimentological, as a consequence of increased sedimentation rates, reflecting an increase of farming activities and also increased flood events during the LIA; ii) chemical, a result of eutrophication caused by fertilizers; and iii) hydrological, caused by human water consumption.
v) paleolimnological studies are key to ecological restoration efforts in Mediterranean lakes. All of the studied aquatic ecosystems receive some protection and government agencies have defined plans for conservation and restoration. Little, however, was known of the lake dynamics at timescales longer than a few years. Monitoring data are very recent and historical or documentary records are scarce. Long-term data on limnological variables in Spanish lakes are rare. Short cores from several lakes, dated with 210Pb and 137Cs have allowed comparison of geochemical proxies with monitoring data in Enol, Arreo and Sanabria. The results, although promising, underline the complexity of such an approach and the need to conduct further research on how the climate signal is transferred to and preserved in the sediments. Paleolimnological studies demonstrate that lakes in the Mediterranean region display large natural variability that is not captured by monitoring. They also illustrate that the limnological systems have been strongly affected by humans and that modern, and even historical environments, do not reflect pristine conditions. These studies provide the tools to evaluate the relative significance of human versus climatic factors in lake hydrology and watershed changes during historic times, and help us model and predict the effects of the current period of rapid climate change in Mediterranean lakes.
The papers
In this special issue, we present papers that came out of study of cores retrieved during the 2004 LRC-IPE survey in six lakes: Enol, Sanabria, Arreo, Estanya, Montcortès and Zoñar (Fig. 1).
Moreno et al. (2011) describe the Holocene sequence from Lago Enol, a high-altitude, karstic and glacial lake in the Cantabrian Mountains. In this record, the Younger Dryas event (13,500-11,600 cal yr BP) is well characterized as a cold and arid phase. The Holocene displays three phases: a warm early Holocene (11,600-8,700 cal yr BP), a more arid middle Holocene (8,700-4,650 cal yr BP) and a return to humid conditions during the late Holocene (4,650-2,200 cal yr BP).
Three papers investigate the Montcortés sequence (Eastern Pre-Pyrenean range). Corella et al. (2011a) identify relatively shallower lake levels during the middle Holocene (6,000-3,500 cal years BP) and deeper environments, with deposition of varves, since then. Increased carbonate production and lower clastic input occurred during the Iberian-Roman Period, the Little Ice Age, and the middle 20th century. Episodes of higher clastic input to the lake correlate with the MCA (690-1460 AD), the last phase of the LIA, and the maximum human occupation in the late 19th and early 20th centuries (1870-1950 AD). Large gravitational deposits during the Holocene and historic times are likely related to earthquake activity. Scussolini et al. (2011) analyze the diatom record for the last 5,340 years. The record shows a conspicuous alternation between Cyclotella comta and Cyclotella cyclopuncta, reflecting changes in trophic state, and a succession of centric and pennate species indicative of the hydrologic balance of the lake. The diatom assemblages identify a period of increased productivity and likely lower lake levels prior to 3,850 BP. Lake level recovers thereafter, and remains high during the Iberian and Roman Epochs. Lower lake levels are recorded during Medieval times and relatively higher lake levels during the LIA and afterwards. Lastly, Rull et al. (2011) report on vegetation changes of the last millennium, identifying the influence of climate and human activities. Conifer forests were intensely burned at the beginning of feudal times (AD 1000) and replaced by farm fields, meadows and pastures. Changes in the Mediterranean vegetation characterized the warm MWA and the LIA cooling in the 15th century. Forest recovery began around AD 1500, coinciding with wetter climate, but fluctuated in association with human pressure in the watershed.