An Analysis of Artificial Reef Fish Community Structure along the Northwestern Gulf of Mexico Shelf: Potential Impacts of "Rigs-to-Reefs" Programs

Abstract

Artificial structures are the dominant complex marine habitat type along the northwestern Gulf of Mexico (GOM) shelf. These habitats can consist of a variety of materials, but in this region are primarily comprised of active and reefed oil and gas platforms. Despite being established for several decades, the fish communities inhabiting these structures remain poorly investigated. Between 2012 and 2013 we assessed fish communities at 15 sites using remotely operated vehicles (ROVs). Fish assemblages were quantified from standing platforms and an array of artificial reef types (Liberty Ships and partially removed or toppled platforms) distributed over the Texas continental shelf. The depth gradient covered by the surveys (30-84 m) and variability in structure density and relief also permitted analyses of the effects of these characteristics on fish richness, diversity, and assemblage composition. ROVs captured a variety of species inhabiting these reefs from large transient piscivores to small herbivorous reef fishes. While structure type and relief were shown to influence species richness and community structure, major trends in species composition were largely explained by the bottom depth where these structures occurred. We observed a shift in fish communities and relatively high diversity at approximately 60 m bottom depth, confirming trends observed in previous studies of standing platforms. This depth was also correlated with some of the largest Red Snapper captured on supplementary vertical longline surveys. Our work indicates that managers of artificial reefing programs (e.g., Rigs-to-Reefs) in the GOM should carefully consider the ambient environmental conditions when designing reef sites. For the Texas continental shelf, reefing materials at a 50-60 m bottom depth can serve a dual purpose of enhancing diving experiences and providing the best potential habitat for relatively large Red Snapper.

Fig 1. Map of ROV Survey Sites.

Artificial reef sites (red stars) and platforms (yellow circles) surveyed with ROVs between 2012 and 2013. Bathymetric countours are indicated by gray lines in 20 m increments. Inset map indicates location of survey sites relative to Gulf of Mexico.

Fig 2. Species-Accumulation Plot.

Randomly permutated species accumulation curve based on fish assemblages observed from remotely operated vehicle surveys. (Michaelis-Menton S _max = 61.12).

Fig 3. Vertical bar chart of mean species richness by structure type.

Sample size (i.e., number of surveys) is represented by the number at the bottom of each bar. Error bars represent standard errors. Letter designations (A, B) correspond to statistically different groups based on Tukey’s pair-wise comparisons tests (α = 0.05).

Fig 4. Scatter plots of species richness by structure height (A) and bottom depth (B).

Various structure types are categorized by symbol and color. Non-linear curve fits are superimposed on each plot with best fit (black line) and 95% confidence bands (red line).

Fig 5. Map of pie charts depicting the indices of abundance of five federally managed species.

Data are shown separately for 2012 (A) and 2013 (B): GS = Gray Snapper (gray), GA = Greater Amberjack (yellow), GT = Gray Triggerfish (cyan), RS = Red Snapper (red), and VS = Vermilion Snapper (pink).

Fig 6. Non-metric multidimensional scaling (NMDS) plot of fish assemblages documented by remotely operated vehicle surveys.

Bottom depths over overlain on symbols, with types and colors representing various structure types: topple (green triangle), cutoff (blue triangle), ship (blue square), deck (red diamond), and standing platform (gray cross). Percent similarity bubbles (from CLUSTER) are overlain at 20, 40, and 60%.

Fig 7. CLUSTER diagram showing linkages of assemblages by structure type and bottom depth.

Linkages shown are based on Bray-Curtis similarity matrix. Values next to symbols represent bottom depths (meters).

Fig 8. Bubble scatter plot of Red Snapper Mean Total Length (TL) by bottom depth.

Circle size is scaled to the number of individuals contributing to the mean. A second order polynomial (quadratic) curve is fit to the data to demonstrate a potential non-linear relationship with bottom depth.