Reconstructing Past Climates

As demonstrated in my post on 4^th April, the investigation of fossilised reefs allows the reconstruction of historic sea level fluctuations. This is possible due to the known ecological conditions required by certain species of coral. Similarly, it is possible to investigate other aspects of climate, such as air temperature, sea surface temperature, precipitation and salinity, by using coral reefs as a palaeoclimate proxy indicator. Since the 1970s palaeo-climatologists have realised the potential of coral reefs to monitor climate variations. Coral growth rates, density bands and isotopic composition respond to changes in climate, and can therefore be used to reconstruct past climates (Dunbar and Wellington, 1981). Dunbar and Wellington’s paper assessed the stable oxygen isotopic response to seasonal changes in temperature, salinity and growth rate of branching corals. The coral was stained at certain points in time so that visible markers facilitated the measurement of growth rates and isotopic profiles. By measuring ocean temperature, salinity and oxygen isotopic composition of seawater over the period of a year it was possible to link these changes with observed variations in the coral composition and growth rate. They conclude that variations in oxygen isotopes in coral reefs record seasonal temperature changes, as can be used to explain variations in salinity. Low growth rates were closely associated with colder temperatures and greater cloud cover.




The stable oxygen isotope technique employed by Dunbar and Wellington (1981) has been used extensively to reconstruct past climates. Evans et al (2002) reconstructed Pacific sea surface temperatures from 1607-1990 by assessing stable oxygen isotope (δ¹⁸O) in coral. Sea surface temperatures were derived from oxygen isotope time series taken from 12 tropical Pacific Ocean, Indian Ocean and Red Sea sites. By comparing this data with similar attempts and historical data, it was possible to reconstruct sea surface temperatures in the Nino 3.4 region of the Pacific, the region where sea surface temperatures have the most significant impact on shifting rainfall patterns. Figure 1 presents the results of their reconstruction, indicating the number of El Nino Southern Oscillation (ENSO) warm events, where a sea surface temperature anomaly greater than 0.5°C was inferred from their investigation of coral reefs. 




Figure 1: Coral reconstruction of the number of ENSO warm events in NINO 3.4, 1607-1990 (Evans et al, 2002).

A similar technique was employed in a study by Urban et al (2000), in order to reconstruct ENSO variability over the past 155 years.  Negative oxygen isotope anomalies can be found in central western Pacific corals when El Nino events have increased sea surface temperatures and enhanced rainfall. In contrast, La Nina events (characterised by cool and dry conditions in this region) produce positive oxygen isotope anomalies. Coral from the Maiana Atoll was sampled and the bimonthly oxygen isotope record was used to infer climatic conditions. Figure 2 presents results from this paper, showing the strong correlation between oxygen isotope records in the coral and the instrumental sea surface temperature record. This indicates the potential of this method to reconstruct past climate.




Figure 2: Tropical Pacific variability from coral and instrumental data, 1950-1995 (Urban et al, 2000).

 

A review of present knowledge of tropical palaeoclimates derived from coral reef investigation was compiled by Gagan et al (2000). This article reinforces the benefits that the growing network of coral oxygen isotope records has for paleoclimatology. Constructing climate records that go back to the last deglaciation reveals previously unknown climate variability, such as ENSO cycles that exist on times scales of decades to centuries. However, there are still uncertainties involved with this method that stem from a lack of knowledge regarding the processes that control isotope changes within corals, and the specific climate mechanisms that produce these changes. A multi-proxy approach, combining records from ice cores, tree rings and sediment records is encouraged to ensure high-resolution paleoclimatology.