Large-scale manta tow reef surveys were conducted on 14 offshore and 18 inshore reefs up to 70 km away from the cyclone's path in May 2005, 6-7 weeks after cyclone Ingrid had crossed. On each reef, a front-reef and back-reef site and, where possible, a northward-facing site were surveyed by conducting 6 two-minute manta tows. A total of 490 transects on 82 sites was surveyed on 32 study reefs.Estimates included: Benthic cover (% of available hard substratum of hard corals, macroalgae, and octocorals); Fragility (of the remaining reef framework, rated 0-5, where 0=extremely fragile and 5=few breakable structures); Depth of damage (m).The following measures of disturbance were also recorded:Framework removed (of the top layer of the reef framework); Dislodgement of massive corals (broken off at their bases, often rolled onto their sides); Coral breakage (corals with parts broken off and loose coral fragments); Scarring by debris (on living coral surfaces attributable to surge-borne debris); Octocorals torn (gorgonian fans or bases of Sinularia colonies stripped of their tissues); Algal bloom (generally cyanophycea, filamentous diatoms or chlorophyta covering extensive proportions of disturbed reef surface); Sediment transport (shifted downslope or horizontally, often burying part or whole coral colonies). All disturbance categories were rated on a scale from 0 to 5, including half-steps (0=0%; 1=up to 5%; 2=6-20%; 3=21-50%; 4=51-80%; 5=80%).Fine-scale before-and-after video transect surveys were carried out on 4 inshore reefs south of the cyclone path - MacDonald Reef (32 km distance to the path), Hay Island (46 km), Wilkie Island (54 km), and Hannah Island (67 km). Video transects were used to assess benthic cover and taxonomic richness of hard corals, octocorals, and macroalgae in May 2005 and compared to previous video transects conducted in February and October 2004. The video analyses distinguished 46 genera of hard corals and 16 taxonomic groups of octocorals. Growth forms of hard coral were recorded as: branching corals (included foliose forms, 13 genera, Acropora, Pocillopora, Montipora, Echinopora); massive corals (19 genera, Poritidae, Faviidae, Goniopora, and Mussiidae); unattached (4 fungiid genera).Along the first 10 m of the same transects, the densities and taxonomic richness of hard coral and octocoral recruits (approximately 0.5-5 cm in diameter) were assessed within a 0.3-m wide belt; identification was mostly to genus level. Recruit densities had been determined with the same methods at these sites 4 years earlier (April 2001).

To assess in detail how the types and extent of cyclone damage on coral reefs varied along a wind gradient ranging from 22 to 46 m s-1, across the shelf, with reef aspect, and with depth.

The vulnerability of coral reefs to storm damage is likely related to the robustness and fragility of reefs, which varies according to location, coral community type, and successional stage of coral development. On the Great Barrier Reef, the main spatial factors determining vulnerability of a particular locale are its position across the continental shelf and its location within a reef.Ingrid was a Category 4 cyclone. It was strong yet small in diameter, creating a clearly defined gradient in wind speeds, and it crossed in a straight line through a region that had not been affected by major disturbances for several decadesReefs: 12101, 12104, 12127, 13040, 13050, 13056, 13074, 13116, 13117, 13120, 13121, Binstead, Blanchard, Bow, Celebration, Creech, Ellis, Fife, Hannah, Hay, Heath, MacDonalds, Magpie, Morris, Night, Osborne, Sherrard, Tijou North, Tijou South, Wilkie, Wye, Zenith.