The 10th Symposium on Global Change Studies

4.2
LONG-TERM VARIABILITY OF THE NORTH ATLANTIC OSCILLATION (NAO)

Charles W. Stockton, Univ. of Arizona, Tucson, AZ; and M. F. Glueck

The recent persistent high positive phase of the North Atlantic Oscillation Index series has raised the issue of the uniqueness of this variability. When placed in the context of the recorded surface pressure records, from which the Index is computed, this anomalous period is the most persistent and highest of record, extending from about 1973 to 1995. This has resulted in an extended dry period in Morocco with as much as 35% reduction in runoff from major river basins and serious reduction in cereal crop production. We are utilizing the nearly 1000 year long tree ring database from Morocco, in conjunction with selected tree ring data from Finland, to reconstruct the NAO Index series. Our objective is to provide a 1000 year long reconstruction. Cool season precipitation in coastal Morocco is highly dependent on the state of the NAO Index with correlation coefficients ranging from -0.68 at Agadir and Kenitra to near zero at the interior station of Ouarzazate. NAO Index values computed from Gibraltar and Southwest Iceland surface pressure records provide slightly higher correlations than do the values computed from either the Azores or Lisbon. Tree ring series from Morocco, Cedrus atlantica, are also highly sensitive to cool season rainfall with correlations ranging as high as 0.63. A total of eight out of 23 tree ring series show correlations in the range of 0.5 to 0.6. These data, along with the more temperature sensitive Finland series, are used in a least squares calibration to develop a reconstruction equation that we use to reconstruct the long-term NAO Index series from the 800 to 1000 year long tree ring series. The total variance accounted for is moderate for the entire frequency range but the low frequency range appears to be duplicated quite well. Preliminary reconstructions suggest that the recent extended period of high values in the NAO Index series may not be unique.

The 10th Symposium on Global Change Studies