For purposes of our discussion, these controls are divided into three groups, based on the scale at which they are most important. Macro-scale controls are those that can operate globally or at least over very large areas. Most important among these are tectonics and sea-level rise and fall. Micro-scale processes are those that exert an influence within an individual reef or reef system. These include the topography upon which the reef formed (antecedent topography), light level, local nutrient input, salinity variations and sedimentation patterns. Meso-scale processes fall between the two. They are generally most important in determining the variability in reefs across a single basin. Wave energy and regional temperature patterns head this category. It should be kept in mind throughout the following discussion that these scale-based groupings are somewhat artificial and that most of the controls discussed below exert some influence at all scales.
Based on his examination of Pacific atolls, Darwin 1842 argued that the world-wide distribution of reefs could be explained solely by understanding the tectonics that were involved. He proposed that Indo-Pacific fringing reefs, barrier reefs and atolls represented an evolutionary continuum tied to the subsidence of a volcanic core at the nucleus of the system Figure 7.37 . He further argued that fringing reefs like those in the eastern Caribbean Sea, formed in areas of tectonic stability and that barrier reefs and atolls were reflective of significant subsidence.
While Darwin's ideas about the genetic sequence of fringing reefs to atolls in the Pacific were later confirmed by deep cores through Enewetak and Bikini Atolls, Ladd & Schlanger, 1960 our growing knowledge of global tectonics has shown that Darwin's ideas about subsidence as a universal and sole control of reef type were flawed. Contrary to his prediction, the fringing reefs of the eastern Caribbean exist is an extremely volatile tectonic setting. Conversely, barrier reefs often occur along trailing continental margins that, by comparison, are tectonically stable. Nevertheless, tectonics does exert a strong influence on large-scale patterns of reef development, and must be integrated into any realistic "reef model."
We now understand that reef development is the result of not a single controlling variable, but rather a combination of them. Vail, et al. 1977 cite basin-to-basin similarities in the character of sedimentary sequences and the sea-level histories that they infer as evidence for a decipherable pattern of world-wide ecstasy. By identifying and averaging the patterns common to depositional sequences from many different ocean basins through time, "seismic stratigraphy" has been used to construct a record of global sea-level change through time. Haq, et al., 1987 Careful analyses of the variability of this pattern from site to site can be used to infer the superimposed effects of local tectonics. The term relative sea level refers to the change in sea level at any one locality resulting from the combined effect of the glacio-eustatic rise and fall of the world ocean (related to the melting and freezing of polar ice) and local tectonic motions.
Keep up, catch up or give up: the reef responds - Reefs can be generally classified into one of three groups . Neumann & Macintyre, 1985 Keep-up reefs have maintained their crests at or near sea level throughout their history. The reef interior is dominated by organisms typically found in shallow water. Catch-up reefs were initially outpaced by rising sea level, but have subsequently caught up, usually after the rate of sea-level rise slowed. Their lower (older) section is dominated by deeper-water organisms, and is overlain by shallower-water types that reflect shoaling of the reef.
The geometry of the reef system, when viewed in cross section, will reflect the relative ability of reef accretion to match the rate of sea-level rise. Catch-up reefs or keep-up reefs that accrete at a rate equal to that of sea-level rise will build vertically. Once reaching sea level, many reefs also build seaward to accommodate the imbalance between the volume of carbonate being added to the system and the accommodation space being created by rising sea.
Give-up reefs are those that, for whatever reason, simply stopped accreting. The condition most commonly cited as the cause for a particular reef giving up is a sudden rise in sea level that is faster than the reef can match. When this happens, the reef is either gradually left behind (i.e. "drowned") or else " backsteps " to a shallower and more shoreward location. Ancient examples occur in the Devonian reefs of the Canning Basin of AustraliaPlayford, 1980 and western Canada. Viau, 1983
Over the past decade, the interest in "give-up" reefs has been heightened due to the reminder by Schlager 1981 that most shallow-water reefs are capable of accreting at rates greater than even the most rapid sea level rise. He reconciled this "paradox" by proposing that a sudden degradation of local oceanographic conditions may be at fault. These "inimical" waters can be created by suspended sediment along an updrift platform, Adey, et al., 1977 elevated nutrient levels Veron, 1986 or sudden change in the ambient temperature regime. Glynn, 1984; Hudson, 1981
This problem has particular significance in recent discussions about global warming. It has been proposed that, due to increasing levels of "greenhouse gases" in the atmosphere, a dramatic increase in the rate of sea-level rise may occur by the end of the century. Will present-day reefs be able to keep pace with this rise or will they be left behind, changing the character of nearshore shelves and banks dramatically?
Recognition of a particular type of accretionary pattern can provide valuable information on the factors that have shaped a particular reef and its associated environs. The unraveling of depth-related sequences within individual reefs can provide insight into the sea-level history of a particular area. Conversely, within areas where the likely patterns of sea-level change are known, deviations from the expected suite of facies or organisms outlined above can provide useful clues about other factors that might be in control. These principles can be applied to both modern and ancient reefs alike.
In the Hawaiian Archipelago, Grigg 1982 proposed that the limit of 29° N for atoll formation is the result of both temperature and subsidence . Away from the hot spot that is responsible for the Hawaiian chain, the volcanic islands are gradually subsiding. In warmer water, reef accretion can offset this subsidence and the reefs have kept up with sea level. Toward the north, however, the rate of carbonate production drops gradually with falling temperatures. North of Grigg's "Darwin Point", carbonate production has slowed to the point where the reefs can no longer keep up with the effects of subsidence and they have drowned.
Along the 2500-km long Great Barrier Reef of Australia, cores reveal a record of gradual climatic shift from subtropical to tropical conditions over the past 30 million years. Davies, et al., 1989 This is a result of the northward (i.e. toward the Equator) motion of the Australasian plate in response plate tectonics. As a result of rising temperature, the benthic community and the sediments that they produce have changed gradually through time.
It is important to note that most corals exist near their upper thermal limits. Therefore, even a slight increase in tropical temperatures in the future could have a significant impact on the distribution of corals in the tropics. Temperatures only a few degrees above normal can result in the expulsion of algal symbionts ( bleaching ) that make an important contribution to their metabolic budget. The 3-4o C rise in temperature in the Pacific associated with the 1982-83 El Niņo Event has been linked to widespread bleaching off the western coast of Panama Glynn, 1984 and the eventual devastation of the coral community.
At the other end of the temperature spectrum, cold water that upwells off western Panama may similarly limit the occurrence of corals. Glynn & Stewart, 1973 In the northern Florida Keys, Walker, et al., 1982 proposed that the growth rates and distribution of Montastraea annularis were in large part controlled by the periodic influx of cold water pushed out from Florida Bay during the passage of major cold fronts.
Storms also play an important role in determining reef character. Three primary hurricane tracks exist in the Caribbean Sea. 1982 One of these passes to the south of the Greater Antilles and two pass to the north. Based on these tracks and the regional distribution of total wave energy, Caribbean reefs can be roughly divided into three types that correspond closely to those originally identified by Adey and Burke. 1977
Areas of high prevailing wave energy and frequent hurricane disturbance (Type I: i.e. Windward Islands) are characterized by algal ridges, comprised of storm-generated piles of broken Acropora palmata bound and capped by thick crusts of coralline algae. High wave energy on a day-to-day basis discourages grazing by fishes that would inhibit the accumulation of thick algal crusts under calmer conditions. High wave-energy appears to be the most important factor in the distribution of both Caribbean and Indo-Pacific algal ridges.
Areas of moderate-to-high prevailing wave energy but infrequent disruption by storms (Type II: i.e. northern St. Croix) are dominated by branching Acropora palmata. From this, it would appear that the important distinction is the lack of frequent destruction by passing storms.
Areas of low prevailing wave energy and frequent storm disruption (Type III: i.e. northern Bahamas) are dominated by open pavements with only scattered coral cover. Frequent storm disruption combined with little inhibition of grazing between storms discourages the formation of either the thick algal ridges of Type I reefs or the abundant, branching cover of Type II reefs.
In the Indo-Pacific, both wave energy and sedimentation play similar roles in reef zonation. As in the Caribbean region, algal ridges dominate the high-energy margins of open-Pacific atolls. Along the Great Barrier Reef, the reef-crest and forereef communities change from the high-energy reefs in clearer water near the shelf edge and in the Coral Sea to reefs subjected to higher levels of sedimentation in calmer waters near land.
In addition to older reefs, remnant topography can be related to the edges of tectonic blocks, cemented sand bars, fossilized dunes and even ancient river deltas. One common controlling factor is a process called karsting (see Purdy 1974 for a review) . During episodes of lowered sea level, limestone strata are dissolved by rainwater. Remnant highs often result from the combined influence of weathering and pre-existing reefs. Because reefs are generally reinforced by syndepositional cements formed within the their interstices, the resulting mass is more resistant to weathering than are the muddier and less-cemented sediments of the platform interior. They serve as focal points for recruitment once the area is reflooded by rising sea level.
Montastraea annularis possesses the ability to change its shape in response to more or less light. Grauss & Macintyre, 1982 In the Indo-Pacific region, other corals have shown similar morphologic plasticity. While recent studies have shown that there may be at least three sub-species of Montastraea annularis, the demonstrated ability of individual colonies to change shape in response to artificially changed depth or light level supports the importance of this plasticity to a wide variety of conditions. It should come as no surprise that these more adaptable species are the most common among those recovered in cores from modern reefs.
Shading is probably the most important of all the sediment-related effects. Because it is more subtle than smothering, however, its effects are difficult to quantify. Despite an impressive body of literature (for a review, see Hubbard 1986), little quantitative information exists on the specific responses of reef organisms to sediment loading. The frequency with which coral reefs succumb to turbid water from dredging projects or increased runoff are totally inconsistent with laboratory experiments that have documented surprising tolerance by corals to high doses of sediment over short periods of time. Taylor & Saloman, 1978; Rogers, 1983 The obvious factor here is the effect of long-term exposure. Sediment traps on reefs can be used to measure the sediment influx over a year and the rate of growth of Montastraea annularis. Even lower levels of stress can gradually wear down the reef's defenses if allowed to persist long enough. This accounts for the lack of well-developed reefs along the downwind flanks of most platforms and the increasing number of post-mortem autopsies of reefs killed by dredging or upland clearing.
The level of suspended matter in the water column has a direct effect on light penetration. The greater depth at which corals are found in the open Pacific (>100m) likely reflects clearer water relative to that along the Great Barrier Reef or throughout the Caribbean. Turbidity ("cloudiness" of the water column) exertes a strong control on the depths at which coral zones change, presumably in response to decreased ambient light levels due to suspended solids in the water. Reduced light levels can suppress coral-growth rate, impact natural zonation patterns and induce wholesale mortality if allowed to persist. Morelock, et al., 1979; Hubbard, et al., 1986 The change in percent of cover on individual reefs in relation to the amount of fine-grained suspended sediment can be a practical guide to species resistance to sediment influx . Acevedo, et al., 1981
Abrasion by moving sediment can cause substantial damage to coral tissue, especially during storms. Hubbard, 1992 Even under less-energetic conditions, sediment scour can exclude head corals from the reef crest as regular abrasion inhibits their growth or kills smaller colonies.
Excess sedimentation can also discourage recruitment by coral larvae. Morelock, 1986 discussed the importance of substrate type in larval recruitment in Puerto Rico. Roy and Smith 1971 proposed that on Fanning Island, the increased vulnerability of young corals to sediment damage was a more important factor than sediment covering available space.
The role of nutrient inhibition in coral reefs is multi-faceted. On the organism level, it has been proposed that high phosphate levels in the water can effectively shut down the calcification mechanism ("phosphate poisoning" of Simskiss 1964). At the community scale, higher nutrient levels tend to favor sponges Wilkinson, 1987 and algae , Steneck, 1986 which can out compete corals for space and prevent larval settling. Once the coral dies, higher levels of nutrient availability favor grazing and infestation by infaunal borers such as Cliona spp., which will progressively destroy the remaining skeleton and can totally remove any record of the original organism.
By understanding the response of reefs to natural stress, we can better equipped to predict and mitigate damage related to development. Much of this information will come from detailed experimentation by biologists. However, even the most carefully designed experiment is incapable of addressing one factor that is emerging as perhaps the most important of all - time. Over the past decade, better than half of the Acropora palmata and much Acropora cervicornis in the Caribbean has died off , probably a result of "White Band Disease." Gladfelter, 1982 and hurricanes . In the mid 1980's, the sea urchin Diadema antillarum underwent an unprecedented population crash throughout the region. This grazer is an important regulator of algal turfs that compete with corals for space on the reef. More recently, widespread coral bleaching and other disturbing changes in reef communities noted by the scientific community have spurred congressional hearings on the "health" of the world's reefs.
A critical question in all of this is whether these represent a sudden, man-induced decline or whether we are simply on the downside of a natural "boom-and-bust" cycle that occurs over a period of decades or centuries. Recent concerns over global warming only serve to reinforce the need to understand the long-term responses of coral reefs to large-scale physical-oceanographic forcing functions.
The answers to these temporal questions may be no further away than the interior of the Holocene reefs which underwent a similar warming 10,000 years ago. Placing recent events into the context of change on a scale of decades to centuries also demands a greater commitment to long-term monitoring, which is unfortunately a low priority in a society that demands quick fixes and short-term solutions. Both should be priorities for future research.