Most people realise that intense cyclones, like Ingrid (March 2005) or Larry (March 2006), can cause severe damage to people and infrastructure once they make landfall; a cyclone is a lethal cocktail of high winds, heavy rainfall and flooding.
What is less obvious is that the heavy seas generated by high winds during cyclones can also devastate underwater environments, such as coral reefs. Over time, repeated damage from cyclones can affect the ecology of a reef, especially when combined with other types of damage such as overfishing or pollution. So, it makes sense to limit damage from human activities, which are preventable, in reef areas that experience the most frequent impacts from cyclones, which are not preventable.
To do this, we need to map the location of cyclones over time as well as patterns in the damage they caused. Since 1910, hundreds of cyclones have tracked through Queensland's Great Barrier Reef. The paths taken by these past cyclones are readily mapped based on data routinely collected by the Australian Bureau of Meteorology.
However, only a few field surveys have been done over that time to map the damage they caused to reefs. Evidence of cyclone damage on land is dramatic and easily mapped. Cyclone-generated waves can cause equally dramatic damage to reefs. During a severe cyclone, it is not uncommon to see three metre wide coral heads broken from the reef and thrown up into shallow water, or entire fields of branching corals broken.
Still, it can be quite challenging to actually locate, much less map, the most severe damage across a group of reefs from a given cyclone. One reason for this is that the vulnerability of individual reefs to damage depends on many highly variable factors. The type, size, and age of corals growing in a particular spot makes a difference, as does the amount of shelter given by nearby reefs, islands or the mainland.
This means that one part of a reef may be completely devastated from a cyclone and need centuries to recover, while another part, located less than a kilometre away, may be completely untouched. Because carrying out an underwater survey is time consuming and expensive, without some hint about where to target search efforts, researchers can easily miss the most damaged reef areas.
The distance of a reef from a cyclone's path can give a crude hint of whether or not damage was possible; high wind and wave energy is generally concentrated within a zone about 100 km wide on either side of the path. However, to be certain that all damaged areas are found would require observing every part of each reef located within this zone. This could require visiting thousands of sites across hundreds of reefs. It's altogether too expensive and time consuming to be feasible.
Luckily, it is possible to further refine this 'target zone' where damage is most likely to have occurred. The intensity, speed of movement and direction of a cyclone frequently changes as it moves along its path. All of these factors affect the distance to which high wind conditions, and thus the possibility of heavy seas, will extend from the path. Mapping the spread of wind speeds and wave heights during a cyclone is rarely possible directly, however. Only a limited number of wind gauges and wave buoys are in place along the Queensland coast and they often fail during the extreme conditions that occur.
To help predict land areas at most risk from cyclone damage, meteorologists have long used equations that estimate cyclone wind speed at given distances from the path. These equations use basic information about the cyclone that is routinely collected by the Australian Bureau of Meteorology, using radar and satellite imagery. This would typically include its position in latitude and longitude, intensity, and direction and speed of movement.
By solving these equations in a GIS, a raster grid of wind speeds across an entire large region such as the Great Barrier Reef can be estimated quickly for each position of the cyclone along its path. These wind speed grids can then be combined to map the maximum wind conditions that occurred across the region over the entire cyclone.
A team from the Australian Institute of Marine Science and the University of Wollongong led by Katharina Fabricius used this information to design an extensive field survey - 490 sites spread across 33 reefs to map the location of damage to reefs from cyclone Ingrid.
The aim was to visit examples of reefs found both inside and outside the four wind zones on either side of the path. Care was taken to also visit reefs located on the inner, middle and outer shelf reefs. From this, we can test how useful the maximum wind zones are in predicting levels of damage.
Preliminary results suggest that the severity of damage generally increases with maximum wind speed. No severe damage, for example, was found outside the category 3 to 4 wind zone. However, some reefs within this zone were undamaged. Further analysis of the data may help us better explain the complex range of factors that account for the difference in reef vulnerability to wave damage that cause this.
Once this is complete, we have the opportunity of using the results to predict the location of damage to reefs located offshore of Innisfail from cyclone Larry. Fabricius will lead a field survey, in July 2006, to test the quality of these predictions.
Wind zones like the ones shown above for cyclone Ingrid have already been created by the author for all of the 85 cyclones that passed through the Great Barrier Reef from 1969 to 2003. The results of the cyclone Ingrid and Larry damage surveys will be used to interpret these wind zone maps to better predict the frequency and timing of cyclone damage across the region over the recent past. This information can then be used by relevant government agencies to help control human activities to protect the Reef from permanent damage.
This paper is based on one of the largest damage surveys completed, in the wake of cyclone Ingrid in May 2005. The survey was carried out by a team of reef ecologists from the Australian Institute of Marine Science - Katharina Fabricius, Terry Done, Scott Burgess and Tim Cooper - and the author, Marji Puotinen, a GIS specialist. She is a senior lecturer in GIS at the School of Earth and Environmental Sciences, University of Wollongong. The address for correspondence is marji@uow.edu.au.
