Initial reaction to the tsunami by the international community was that it had a large potential to damage coastal and nearshore ecosystems and resources. At small spatial scales, tsunamis were considered to have the potential to directly affect fish by displacing or washing them ashore as occurs in storms (Walsh 1983). Locals reported many small fish washed ashore at Palau Weh (Allen 2005b).
Post-tsunami, environmental damage was expected to be extensive with potentially irreversible destruction of benthic reef habitats (coral mortality, alterations in physical structure), as well as immediate loss of living coastal resources such as fish, lobsters and crabs (IUCN 2005c; Wilson et al. 2006). This was expected to have serious implications for fisheries, though it was unlikely that the pelagic environment had been severely affected (IUCN 2005c). Declines in the abundance of fish following extensive depletion of hard coral are common (Booth et al. 2002; Jones et al. 2004) though there can be a significant time lag between the loss of habitat and a reduction in fish numbers. Pratchett et al. (2006) detected no change four months after bleaching in the abundance of obligate corallivorous fishes, despite a 90 percent decline in coral cover. They suggested that it may take longer than this for the fish to starve or relocate. It should be noted therefore that a low abundance of fish immediately after the tsunami could be the result of conditions prevailing before the waves hit.
A range of mechanisms on how tsunamis could affect coastal resources and ecosystems and the factors that can modify their effects has been identified (Table 2). The direct mechanisms that can damage resources and ecosystems include wave action, smothering by sedimentation and land-based debris, mechanical damage by land-based debris and uplift/submergence. Waves can also directly dislodge attached benthic organisms and move mobile organisms significant distances inland. In rapid assessments in Sri Lanka, many species of nearshore and estuarine fish populations were subjected to mass mortality and washed into inland areas. Freshwater fish species inhabiting low-saline lagoons were reported to have died due to increased salinity (IUCN 2005a). Wave action was specifically found to dislodge, break and move corals. Sediments washed off the land smothered corals and there were many cases of corals being covered by land-based materials swept from damaged settlements. Allen (2005a) suggested that the debris picked up by the waves increased their destructive nature. Corals in Simeulue Island were lifted 2 metres out of the water and completely lost, while on the other side of the island they were submerged by the same amount (ICRI 2005; Phongsuwan and Tun 2005; IUCN 2005e; IUCN et al. 2005; LIPI 2006).
The direct effects of waves, sediments and mechanical damage were patchy and their intensity and extent were modified by a range of factors (Table 2). Most agree that distance from the earthquake epicentre, the presence of inlets and headlands, the presence, extent and gradient of the continental slope, shoreline elevation, the presence of dunes and other vegetation, and density of habitation and infrastructure seem to explain most of the variation (Gibbons et al. 2005; ICRI 2005; UNEP and WCMC 2006). In particular, tsunami waves were considered extremely sensitive to details of nearshore and coastal topography (Gibbons et al. 2005; Rudi and Fadli 2005). Narayan et al. (2005) reported that in India the width of the continental shelf played a major role in the pattern of tsunami damage (less width = more damage) and that wave reflection and interference from Sri Lanka and Maldives created areas of wave doubling and cancelling, explaining the highly patchy nature of damage.
After the initial impacts of tsunami waves, more medium- to long-term feedback effects can lead to further damage to nearshore resources and their environments. Changes in nearshore water characteristics, including temperature and nutrients could be expected to lead to algal blooms or anoxia events because new sources of nutrients have been mobilized. Changes in topography caused by sediments being deposited, or eroded, including the opening of normally closed lagoons in Sri Lanka could be expected to alter water currents. Previous damage to ecosystems was expected to leave them more vulnerable to tsunami damage and less likely to recover. New sources of pollution from damaged sewage and drainage systems could impose new stresses on ecosystems recently affected by the tsunami. Finally, feedback damage might include changes in fishing effort and other human uses of resources and ecosystems. Temporary reduction in fishing caused by loss of boats and fisherfolk, increases resulting from emergency fishing and sudden possibly permanent changes in fishing practices prompted by the introduction of new gear and new boats from donors were all expected to play a role.
According to Campbell et al. (2006) the damage by the tsunami is very different to that observed following large storms. Cyclone damage to reefs is also patchy (Woodley et al. 1981) but it is unusual for shallow reefs to escape damage over large scales (Hughes and Connell 1999). Baird et al. (2005) hypothesized that in cyclones, energy is concentrated near the surface and diminishes with depth, while for tsunami waves the entire water column is in motion. They suggested that for corals the initial run-down of the tsunami, along with the first wave of the tsunami train, excavated unconsolidated substrata from around the bases of unattached colonies, making them susceptible to displacement when inundated by the subsequent waves.
