Scientists from the Australian Institute of Marine Science are taking the first steps towards studying the detailed climate of the tropics prior to the period when comprehensive instrument measurements began. Inside coral skeletons they have found techniques of interpreting the patterns of wet-dry seasons for many centuries past. This new knowledge is already providing climate researchers from all over the world with a way of extending their understanding of the Earth's weather history before it was measured with man-made instruments.

Corals grow by adding seasonal layers of calcium carbonate (limestone), extracted from the surrounding seawater, to progressively enlarge their skeletons. In rounded massive-type colonies (like the commonly-understood "brain" coral, especially those of the genus Porites), the seasonal layers of material differ in density between summer and winter. These massive coral domes of Porites are the most widespread reef building corals on the planet and they contain their own regular and consistent calendar which extends back over centuries in many cases and, in rare examples, perhaps as long as a thousand years. X-ray pictures of slices of these skeletons (or in the case of very large corals, slices of cores drilled vertically from the living surface) allow counting of the annual summer/winter density bands. Scientists can thus work the age of sections of skeleton.

In the mid 1980's Dr Peter Isdale and his colleagues at the Australian Institute of Marine Science discovered a remarkable way that corals incorporate special substances from the water surrounding them as they grow. These substances, which are complex chemicals dissolved in the seawater, are laid down with the skeleton material as it deposits its layers, week by week, year by year, and century by century.

Annual runoff reconstruction from the early 1600's to the present time.

Of particular interest to hydrologists, climatologists and meteorologists is the way in which corals growing within about 20 kilometres of the coastline can trap river-borne organic compounds. Most of these are humic compounds derived from the decayed remains of land plants. These particular compounds fluoresce, giving off visible light when exposed to short wave ultraviolet light. The substances in the corals can be seen under ultra-violet light as irregularly-spaced yellow-green bands superimposed on the density calendar in the corals. The brightness of the fluorescence is so great that measurements of the skeletal concentration of the humic compounds can be derived by simply measuring the fluorescence. This concentration is related to the discharge of adjacent rivers, since it is believed that most tropical coastal catchments contain an almost inexhaustible reservoir of these water-soluble compounds. River discharges are valuable statistics to the meteorologist or climatologist because they can represent climatic "summaries" of large river catchment areas.

Following the Australian lead, many scientists from other tropical countries have now found a very good correlation between skeletal fluorescence, river discharge and rainfall in corals growing within the influence of the outflow of some major and minor tropical rivers. They have used this knowledge to examine the climate of several areas of the tropical Indo-Pacific each year back to the early 1600's. The level of detailed information about past wet seasons is quite surprising. In the case of some fast growing corals adjacent to very seasonal rivers, events as close as one month apart can be seen in the fluorescence record. By using appropriate mathematical models, the seasonal rainfall patterns for a region can be described. In addition, fluorescent bands in ancient fossil corals have revealed that much wetter conditions existed several thousand years ago in areas such as the Sinai, now desert.

The patterns of monsoon rainfall in the last three to four hundred years in some tropical areas, as revealed by the corals, holds some surprises. Those periodic changes from wetter to drier seasons on a ten-year scale which have been known about from instrument records for some time are now illustrated by the coral record to lie within much longer-scale alternating periods of activity. In addition, the corals show us that the extreme events of the last 100 years are not as great as some which are further back in time. In some places the longest drought known from instrument records is less than half the length of those which occurred there more than 200 years ago. A similar story is obtained from the corals concerning the number of consecutive very wet seasons.

Current interest in the uniqueness of the recent widespread drought in Eastern Australia has prompted an investigation of the frequency of extended dry periods in the region using the fluorescent banding proxy record of wet season intensity. The corals from North-east Australia point; to a return period of about 30 years for dry spells similar to the recent one, but a frequency of about one in 350 years for decade-length droughts. This knowledge impacts directly on the policy settings of planners who are attempting to insulate rural activities from the disastrous effects of persistent rainfall failure.

In summary, the corals are telling us that the instrument-based "snapshot" we have of our tropical climate can be improved by the coral record to more competently plan our human activities in the tropics.

For further information:

e-Mail: Bruce Parker
Marine Geochemistry Group.

tel: 077 534260
fax: 077 712138