WORK TO BE COMPLETED:

We propose to reconstruct the past temporal variability and patterns of large scale tropical Atlantic SST, using long-lived Atlantic corals. This will be instrumental to better understand West African climate variability by exploring the statistical relationships between SST pattern variabilities, ITCZ, land climate, and corals. In particular, we wish to see how well the coral proxy indices replicate the relationships seen in the more conventional SST data sets. We can then use those relationships together with the longer coral records to infer the west African climate (from rainfall data archives and Gulf of Guinea runoff) much farther back in time and to answer some key questions regarding the secular variation of that variability. Successful extraction of longer proxy records will enable us (and others) to further explore: (1) the variations of tropical Atlantic SST statistics (including the demonstrated association of SST with short term climate variations in NW Africa and NE Brazil), the tendency for SST north and south of the equator to behave antisymmetrically (dipole), and the effects of these variations in SST behavior on land climate (especially rainfall over NW Africa); and (2) provide more temporal degrees of freedom for multidecadal SST variations by extending coral proxies farther back in time than our inadequate instrumental records reach. Further, the Cape Verde Islands are in a location subject to the path of African dust traveling across the Atlantic. The impacts on corals reefs of this area have not been assessed. The skeleton of these corals along with the annual bands may provide a record of harmattan winds and the included dust driven from the Sahara Desert.

FOCUS AND SCIENTIFIC OBJECTIVES:

This proposal will capitalize on two related features for reconstructing climate: 1) demonstrated associations between Atlantic SST patterns and land climate, and 2) the existence of long-lived corals in key locations with good potential for reconstructing the temporal variability of those patterns. We seek to develop paleoclimatic records from the skeletons of annually banded corals growing in two regions of the eastern Atlantic: the Cape Verde Islands off NW Africa/Sahel and the equatorial Gulf of Guinea west of Lagos (Fig. 1). The Gulf of Guinea is embedded within a characteristic interannual variability pattern of SSTA in the equatorial and tropical South Atlantic (SATL) (Fig. 2 upper). The SSTAs of the Cape Verde Islands are similarly embedded in the characteristic pattern for the tropical North Atlantic (NATL) (Fig. 2 lower). The coral-inferred SSTAs of both Cape Verde and Gulf of Guinea should be reflective of both the large scale NATL and SATL indices of SSTA, as well as their correlated land climate associations (Fig. 3a and 3b).

Our eventual aim is to reconstruct ocean climate history from these Atlantic corals over periods of 100’s of years, emphasizing the time before historical climate records. We expect to develop statistical relations between the present SSTA variability and the contemporaneous skeletal records from corals of the Gulf of Guinea and Cape Verde using stable isotope, Sr and Mg concentrations, and coral skeletal growth parameters. With such information, we hope to extend the temporal coverage of SST patterns allowing assessment of how they have been related to long term variations in precipitation, especially in NE Brazil and NW Africa

RATIONALE FOR AFRICAN AND ATLANTIC CLIMATE AND CORALS:

Our rationale for examining Atlantic corals is that they can help us reconstruct much of the tropical Atlantic SSTA, which we will take as synonymous with tropical Atlantic ocean climate, over time scales longer than those covered by the instrumental record. Prior studies of Pacific corals have similar rationale: they help to understand the ENSO-dominated ocean climate on time scales that transcend the instrumental record. It certainly appears that the statistics of ENSO-related climate variability (e.g., El Nino event intervals) are nonstationary on scales of decades to centuries (e.g., Enfield and Cid, 1991). Hence, analyses based on a few decades of instrumental observations in the twentieth century can be misleading. The potential contributions of corals are great for understanding the longer-term variability of climate relationships. The same is likely true for the Atlantic SSTA. The NATL and SATL averages (or their coral proxies) are direct predictors for the land climates of NW Africa and NE South America. Coral proxies (of NATL) from the Cape Verde Islands offer further predictability for Central America, the Caribbean and northern South America (Enfield and Alfaro, 1999; Giannini et al., 2000), especially when combined with the counterpart proxies of other investigators from corals in the eastern Pacific.

Back to NCRI Projects Page