web dirweb dir Bookmark and Share |
Eos Newspaper

Contact:

Christy Hanson
Marketing & Advertising Manager
Phone: +1 202 777 7536
Fax: +1 202 777 7478
Toll Free: +1 800 966 2481
(North America only)
E-Mail: advertising@agu.org

Supplementary material to “Reconstruction of Past Mediterranean Climate”

Ricardo García-Herrera, Departamento de Física de la Tierra II, Facultas de Físicas, Universidad Complutense de Madrid, Spain, rgarciah@fis.ucm.es; Jürg Luterbacher, National Centres of Competence in Research, Climate and Institute of Geography, Climatology and Meteorology, University of Bern, Switzerland; Piero Lionello, Dipartimento Scienza dei Materiali, University of Leche, Leche Italy; Fidel González-Rouco, Departamento de Física de la Tierra II, Facultas de Físicas, Universidad Complutense de Madrid, Madrid; Pedro Ribera, Departamento de Sistemas Físicos, Químicos y Naturales, Universidad Pablo de Olavide, Seville, Spain; Xavier Rodó, Parc Cientific de Barcelona, Universitat de Barcelona, Barcelona, Spain; Christoph Kull, Past Global Changes (PAGES) International Project Office, Bern, Switzerland and Advisory Body on Climate Change (OcCC), Bern, Switzerland; and Christos Zerefos, National Observatory of Athens, Athens.

Citation:
García-Herrera, R. et al. (2007), Reconstruction of Past Mediterranean Climate, Eos Trans. AGU, 88(9), 111. [Full Article (pdf)]

First MEDCLIVAR Workshop on Reconstruction of Past Mediterranean Climate: Unexplored Sources of High-Resolution Data in Historic Time

MEDCLIVAR (http://www.medclivar.eu/) is a program aiming to coordinate and promote research on different aspects of Mediterranean climate. The main MEDCLIVAR goals include reconstruction of past climate, description of patterns and mechanisms characterizing space-time variability, coupled climate model/empirical reconstruction comparisons, seasonal forecasting, and identification of the forcing responsible for the observed changes. The program has been endorsed by CLIVAR (Climate Variability and Predictability project) and is funded by the European Science Foundation (ESF). The first MEDCLIVAR workshop was held in Carmona, Spain, 8–11 November 2006, at the Pablo de Olavide University facilities. The main purpose of the meeting was to identify sources of early instrumental data and natural and documentary climate proxies that had not been previously explored and/or identified and could be relevant for the reconstruction of the Mediterranean climate or weather extremes covering the past millennia. A main focus was on weather and climate information with high temporal (annual or higher) and spatial resolution as well as the potential for past climate estimates from low-resolution proxies covering the past tens of thousands to hundreds of thousands of years.

The first session reviewed the availability of documentary sources in the larger Mediterranean region, with reports from different collections poorly or not analyzed to date, such as Arabic documentary records, Jesuit reports or ship logbooks, information on the availability of documentary sources which provide mostly from Italy, France, and Spain. In the remaining adjacent regions the availability of documentary records has not been systematically evaluated and is poorly known. However, some early instrumental records from those unexplored regions can be found in several archives and libraries from national weather services.

The second session dealt with natural proxies from the Mediterranean Sea and land areas such as tree ring chronologies from western Iberia and lake calcite laminae thickness in eastern Iberia, which show potential for use as NAO proxies. New proxies from the Mediterranean Sea including vermetids, nontropical corals, and deep-water corals can cover various environments and periods of time within the Late Quaternary. The potential of speleothems from Israel to explain climate variations over the past millennia, the use of varved lake sediments from Turkey, the availability of seasonally resolved climate records in mollusc shells from Gibraltar, and a variety of different approaches using alkenones in order to obtain past sea surface temperature records were also discussed.

The third session introduced some examples of paleoclimate modeling and blending of paleoclimate reconstructions and modeling approaches. Some of the most important results in model-data comparisons for the Last Glacial Maximum within the Paleoclimate Modeling Intercomparison Project (PMIP) and PMIP2 were discussed jointly with the implications of using different model resolutions in this context.

NCAR Community Climate System Model (CCSM) Version 2 model integrations illustrated the behavior of cyclone activity during the simulated late Maunder Minimum (1645–1715) and in simulations with future climate change scenarios. Also, integrations with the ECHO-G coupled model extending from the past millennium and continuing to 2100 A.D. under climate change forcing scenarios evidenced that the simulation of different climate variables over the Mediterranean can show different degrees of sensitivity to changes in external forcing. Precipitation and dynamics related variables are more bound to internal variability, while temperature shows a more clear response to external forcing and the potential for direct model-proxy comparisons. Finally, a comparison of tree-ring-based reconstructions of the Palmer Drought Severity Index (PDSI) for the 1350–2000 A.D. period over the Mediterranean with the ECHO-G and HadCM3 simulations was presented. The PDSI series were diagnosed to be reliable at intermediate to low frequencies after 1500 A.D. and characterize dry periods in the 16th and early 17th centuries in the west Mediterranean and wet ones from the late 17th to the early 20th century, in agreement with regional reconstructions using independent climate proxies.

The main conclusions of the workshop may be summarized as follows:

  1. The Mediterranean area shows a high number of documentary proxies for the past 1000–500 years. However, the density is not enough to reproduce with high spatiotemporal resolution the different Mediterranean climates so far. The highest coverage of proxy information is reached in the northwestern and central parts of the region (Spain-France-Italy).
  2. There are numerous documentary collections with potential to build climatic reconstructions that have been identified in the above mentioned countries and over the sea, but they have not been fully explored or analyzed due to their huge size and very limited funding availability. Almost no information is currently available from the North African and Eastern Mediterranean areas and parts of the Balkans.
  3. Despite the fact that there are several initiatives at national and subnational scales, there is not a common strategy to search and abstract the archives from a basin-scale perspective.
  4. The distribution of documentary and natural proxies is also dependent on the target climate parameter (temperature, precipitation, drought, etc.). Currently, more precipitation and drought related information are available.
  5. Exercises comparing documentary and natural proxies at similar places should improve quality of both types of data sets, but they are scarce in the region. One of the main difficulties is the lack of comparability (resolution, time-space coverage, uncertainty estimates) among the different data sets.
  6. Proxies are imperfect records of climate and need some processing to be used as prolongation of instrumental data. The uncertainties associated with proxies should be considered in comparisons of different proxy-based climate reconstructions or with model simulations.
  7. Model-data comparisons in Last Glacial Maximum studies are possible and have shown improvements in the past few years, mostly associated with the use of coupled atmosphere-ocean models and better boundary (ice) conditions. Increases in resolution seem to partially improve agreement between simulations and observations, particularly in temperature. However, more efforts are needed in numerical experiment design, boundary conditions, abundance of proxies, climate sensitivity issues, cross-validation procedures, etc.
  8. Simulated and reconstructed changes in cyclone activity for the late Maunder Minimum suggest a first agreement in indicating increases in precipitation over the Mediterranean. Comparison of these results with other climate evidence (e.g., ship logs) and other general circulation models is advisable.
  9. Model simulations of the past millennium and proxy records should be compared carefully. Over the Mediterranean, different model simulations show similar temperature changes over the past millennium, whereas they indicate quite different behavior in precipitation and associated dynamics. This suggests that the dynamically related variables such as precipitation are mostly subject to internal variability and less to external forcing. This should be taken into consideration in model-proxy comparisons with the purpose of model validation or mechanism assessment.
  10. The uncertainty in forcing estimates (solar and volcanic variability) is an issue that bounds general circulation model simulations of any resolution and consequently affects the simulation of Mediterranean climate.
  11. There is potential for enhancing resolution over specific Mediterranean areas using dynamical downscaling. Simulation at the regional scales is even more bound to model internal variability, which can hamper interpretations of direct comparisons between regional simulations and reconstructions.
  12. Some systematic comparisons have already been carried out between tree-ring-based reconstructions of PDSI over the past millennium and model simulations. Extended exercises integrating natural and documentary proxies as well as a variety of climate models would be required.

Model simulations are a useful tool when used as a pseudoreality in which proxy reconstructions can be tested. PDSI reconstructions have been tested; results suggest a loss of variance in PDSI reconstructions when pseudoproxies are noise contaminated but also present a reliable representation of intermediate frequencies.

The workshop was cosponsored by the European Science Foundation, PAGES, the Spanish Ministry of Education and Science, the National Observatory of Athens, and the University Pablo de Olavide of Spain. It was attended by 64 researchers from 15 countries, 15% being Ph.D. students, who agreed on the following recommendations:

  1. To build a catalogue of documentary, early instrumental, and natural archives available in the larger Mediterranean region.
  2. To build a database to support this catalogue and promote data interchange.
  3. To build a network working toward improving documentary and natural data availability, improving homogeneity and quality of early instrumental data, and promoting multiproxy comparisons and the subsequent integration of documentary, natural proxies, and model outputs.
  4. To make climate reconstructions and the understanding of past climate change an interdisciplinary approach incorporating climatologists, historians, geologists, statisticians, modelers, archaeologists, and scientists from neighboring fields. Special emphasis should be made in estimating the associated uncertainties.
  5. To make different reconstruction methods account for different time and space scales and characterization of the data.
  6. To increase the number of modeling exercises of past Mediterranean climate.
  7. To define carefully the dominant climatic variables recorded by the proxies. Reconstructions should be done on the basis of mechanisms relating proxies and climate and not only of statistical correlations.
  8. To do comparisons of simulations and climate reconstructions of the past millennium with care considering the fact that precipitation and dynamically related variables are largely subject to internal variability.
  9. To produce ensembles of simulations with different initial conditions and various models that would allow for a better discrimination of internal from forced variability at regional scales.
  10. To encourage the development of temperature climate reconstructions over the Mediterranean including marine proxies (corals, sediments, etc.) that would allow comparison with high-resolution simulations.
  11. To accomplish pseudoproxy studies oriented to the validation and improvement of reconstruction techniques.

AGU galvanizes a community of Earth and space scientists that collaboratively advances and communicates science and its power to ensure a sustainable future.