UNEP Logo Appropriate Technology for Sewage Pollution Control in the Wider Caribbean Region

Report Contents
Report in Word Format

Caribbean Environment Programme Technical Report #4 1989 All CEP Technical Reports

Appendix 9.
The Effects of Hurricane Gilbert on Coral Reefs at Discovery Bay

By: Jeremy D. Woodle
Discovery Bay Marine Laboratory
Discovery Bay, St. Ann, Jamaica


A hurricane is a violent environmental disturbance, tightly constrained in space and time. Its "footprint" of extreme impact may be only a few score miles across, so that its track can be represented by a line drawn on a map of the Caribbean. Wherever it strikes, it passes on with in a matter of hours. To a human observer, it seems like a very rare, extreme event, unpredictable in its occurrence and movement, highly localised, and contrasting sharply with a background of more benign conditions. On a longer time-scale, however, hurricanes are common and ubiquitous; a map of Caribbean tropical storms and hurricanes for the last hundred years is black with their tracks (Neumann, et al., 1987). Thus, despite their small size and brief duration, any point within the hurricane belt is subject to their influence. The temporal structure of that influence depends upon the time scale of other processes affected by storms. Thus, in relation to ecological processes of reef growth or sedimentation, on a time scale of hundreds or thousands of years, hurricanes can be regarded as continuous force. On such a scale, it may be possible to distinguish different intensities of that force due to differences in hurricane frequency in space and time. On shorter time scales, the occurrence of hurricanes is irregular. Their influence on processes measured on a time scale of the same order as the interval between hurricanes, such as the generation time of living organisms, is better understood in terms of the time elapsed since the previous storm.

This preamble will help in understanding the impact of Hurricane Gilbert on Jamaican coral reefs. After Hurricane Charlie (1951), Jamaica enjoyed thirty years free from the impact of large hurricane-generated waves. All kinds of corals flourished, but those which occupied space by rapid growth were especially successful. Acropora palmata (elkhorn) and A. cervicornis (staghorn) dominated large areas in extensive thickets. Rapid occupation of space was achieved by a slender branching morphology; strong enough to resist routine wave energies, but fragile under extreme conditions. In August 1980, Hurricane Allen passed close to the eastern and northern shores of Jamaica and wrought catastrophic damage on their coral reefs (Woodley, 1980). Branching corals were smashed, some massive corals were toppled or shattered, softer organisms like sea-fans and sponges were ripped up, and all were bombarded with fragments and scoured by resuspended sand ( Woodley, et al., 1981). The recovery of reefs to their former luxuriance had not occurred by September 1988, when Hurricane Gilbert struck. Thus, although the physical impact of waves on the north coast was comparable to that of Hurricane Allen, the damage to reef organisms was not as spectacular, because the time elapsed since the previous hurricane was so short.

Injury to Reef Organisms near Discovery Bay

My own observations have been limited to the central north coast, and I will not speculate about other areas. But at Discovery Bay, the reef condition is now approximately what it was after Hurricane Allen. Massive corals on reef spurs that remained erect after Hurricane Allen (eg. 93% of Montastrea annularis at -10 m) mostly (97%) withstood Hurricane Gilbert. In reef channels, survival was less good: 35% in 1980, 56% in 1988. Acropora cervicornis, which had begun to recover at some locations, was completely smashed again. The rubble created by Hurricane Allen (A. palmata slabs from 0-7 m, A. cervicornis sticks from 7-22 m) had become cemented together by crustose calcareous algae and the process known as submarine cementation (Land & Goreau, 1970). Under Hurricane Gilbert, the A. palmata slabs were re-mobilized, scrubbed clean and re-distributed. So were the shallower stretches of A. cervicornis rubble; cemented frameworks in deeper water remained intact, although scoured by sediment. Remobilization of the rubble substratum had serious consequences for corals (and other organisms) that had settled on it since 1980: opportunistic species such as Porites astreoides, P. porites, Agaricia agaricites and Madracis mirabilis. Many sea-fans, sea-whips and sponges were overthrown or broken, and piles of rotting corpses accumulated in channels and chutes on the deeper fore-reef, especially at the sill that separates the fore-reef slope from the vertical deep fore-reef at about -55m.

Physical Effects

Resuspended sediment, macerated tissues and terrestrial runoff greatly reduced underwater visibility (and thus light penetration) after the storm, and it took a couple of weeks to return to normal (visibility after two days, 3-  m; three days, 6 m; four days, 10 m; nine days, 15 m). High organic loading in deposited sediments was evident for days or weeks in the blackening due to sulphate reduction under the anaerobic conditions brought on by decomposition.

The waves generated by Hurricane Gilbert may not have been as high as those of Hurricane Allen, but their physically destructive impact underwater may have been greater, judging from changes seen in reef structures and sediments. I attribute this to the difference between the tracks of the two hurricanes. Hurricane Allen passed along the north coast, about thirty miles offshore, rapidly moving West by North. At any point along the shore, the direction of the incoming waves changed rapidly, and the period of maximum impact was brief (Kjerfve, et al., 1986). Hurricane Gilbert came overland from Kingston on a track converging with the line of the north coast. At Discovery Bay, hurricane force winds blew onshore for several hours from an approximately consistent direction (North North-east), before veering as the storm passed by, only a few miles to the south. Abrasive rubble and sediment were flung back and forth in the same direction for hours. Linear scarification of the West Fore-reef at Discovery Bay is clearly evident even now, three months later.

Quantities of sediment were removed from the reef terraces. Some was dumped onshore as rubble ramparts or floods of sand, but most was carried downslope; not directly, but North North-east, aligned with the major waves. At 20-25 m, therefore, reef lobes to the east of channels suffered encroachment by sand. Sand clearly flowed down the fore-reef slope and some of it will have passed through the intermittent chutes, off the terrace to the island slope below. On the shallow terrace (3-15 m), small sand channels tributary to the major chutes, have been swept clean. Hardgrounds, apparently representing a Pleistocene basement (L.S. Land, personal communication), have been re-exposed.

At West Rio Bueno, a shallow terrace gives way to a vertical cliff at only -9 in. That terrace supports dense reef growth between deep, narrow sand channels. Hurricane Gilbert caused far more erosion here than did Hurricane Allen. Not only were the sand channels flushed clean, but the sides of same channels were torn out, removing corals and revealing older reef fabric. On the cliff itself, where Hurricane Allen caused little damage to plating corals at -10 and -20 m, Hurricane Gilbert removed most of them (T.P. Hughes, personal communication).


Since 1983, when a natural epidemic almost eliminated the important herbivorous sea-urchin Diadema antillarum (the long-spined black sea-egg; Lessios, 1984; Hughes, et al., 1985), free-living algae have proliferated on Jamaican reefs (Liddell & Ohlhorst, 1986; Hughes, 1987). Unrestrained by grazing (since herbivorous fish populations have been depleted by fishing), these plants grew more quickly than corals and were out-competing them for space. Small corals were smothered, and larger corals were being slowly overgrown from around their edges. Hurricarne Gilbert scoured the reefs and removed most of this algal growth, giving a brief respite from competition to the surviving corals. But algae were the first obvious colonisers of the bare spaces created by the storm; a fine green turf in shallow water, and carpets of the red alga Liagora throughout the terrace. The brown alga Dictvota also grew quickly in the first three months.

Fish and Fishing

Immediately after the storm, individual fishes showed changes in behaviour similar to those recorded after Hurricane Allen; loss of territories, unusual shoaling, behaviour. In the longer term, Hurricane Gilbert, like Hurricane Allen, will have reduced the carrying capacity of the reefs by reducing their three-dimensional complexity, although the change is less this time. Some fishermen reported increased trap catches after the storm. This could have been due to seasonal movements: to the fact that reef cover was reduced and traps offered more shelter, as well as being more visible; and to the fact that the number of traps in use was greatly reduced. Most fishermen had sufficient warning of the storm only to make their boats secure. Fishtraps shallower than about -25 in were destroyed, while deeper ones were swept downslope to lodge at the sill or drop over the edge.


Hurricane Gilbert was a very severe storm which generated waves of great destructive power. They had a major impact on reefs of the central north coast, which would have been even more catastrophic had not the damage already been done by the close passage of the equally severe storm, Hurricane Allen, in 1980. As it is, Hurricane Gilbert has reset the clock of reef recovery; they are now in a very similar condition to what they were after Allen.


Hughes, T.P. 1987. Herbivory on coral reefs: community structure following mass mortalities of sea urchins. J. Expt. Mar. Biol. Ecol., 113; 39-59.

Hughes, T.P., Keller, B.D., Jackson, J.B.C. & Boyle, M.J. 1985. Mass mortality of the echinoid Diadema antillarum Philippi in Jamaica. Bull. Mar. Science, 36; 377-384.

Kjerfve, B., Magill, K.E., Porter, J.W. & Woodley, J.D. 1986. Hindcasting of hurricane characteristics and observed storm damage on a fringing reef, Jamaica, West Indies. J. Mar. Res. 44; 119 - 148.

Land, L.S. & Goreau, T.F. 1970. Submarine lithification of Jamaican reefs. J. Sediment. Petrol., 40; 457-462.

Lessios, H.A., Robertson, D.R. & Cubit, J.D. 1984. Spread of Diadema mass mortality through the Caribbean. Science, 226; 335-337.

Liddell, W.D. & Ohlhorst, S.L. 1986. Changes in benthic 6ommunity composition following the mass mortality of Diadema at Jamaica. J. Expt. Mar. Biol. Ecol., 95; 271-278.

Neumann, C.J., Cry, C.W., Caso, E.L. & Jarvinen, B.R. 1978. Tropical cyclones of the North Atlantic Ocean. NOAA, Washington, D.C.

Woodley, J.D. 1980. Hurricane Allen destroys Jamaican coral reefs. Nature (Lond.), 287; 387.

Woodley, J.D. and 19 others. 1981. Hurricane Allen's impact on Jamaican coral reefs. Science, 214; 749-755.


Received: 08 December 1988


Top of Page

Chapter 1. | Chapter 2. | Chapter 3. | Chapter 4. | Chapter 5. | Acknowledgements | References | Appendices 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11

Report Contents

Last Updated: