Analyzing bycatch to better understand natural fish communities

Monk, M.H.; Powers, J.E.; Brooks, E.N. Spatial patterns in species assemblages associated with the northwestern Gulf of Mexico shrimp trawl fishery. Marine Ecology Progress Series 519: 1-12, 2015. doi: 10.3354/meps11150

A picture is worth a thousand words

Schutte 1.26

Why we care

Bycatch is any organism that is unintentionally caught as a result of harvesting a different target organism. Shrimp trawling is notorious for its high amount of bycatch- this type of fishing has the highest discard:catch ratio of any fishery considered in Alverson et al. 1994. Think about how small the typical shrimp you eat is, then think about the size of the net necessary to trap such a small, wily organism. When these fine-mesh nets are dragged through the water, anything bigger than a shrimp is liable to be trapped as well. There have been recent improvements in escape mechanisms built into shrimp nets (like Turtle Exclusion Devices), but these measures are a topic for another post.

Per weight, shrimp trawlers harvest from 3 to 15 times as much bycatch as they do shrimp. The process of hauling in the nets and sorting through the catch takes too long to save non-target species caught with the shrimp- creatures like sharks, rays, and other fish die before they are thrown back into the water. Maximizing the use we get from bycatch, while it doesn’t address the problem of bycatch itself, can at least mitigate the impact of harvesting these non-target species. Only in some places is bycatch used at all (non-target species can be eaten by humans or used in animal feed rather than thrown back in the ocean). This article details another use for non-target species caught as bycatch.

Monk and her colleagues describe the community composition of bycatch from trawls along the Gulf of Mexico’s northwestern coastline. From that information they identified potential “indicator species”- by definition, monitoring just these individual species can indicate the health of the entire ecosystem in which they live. Indicator species are generally characterized by their sensitivity to ecosystem changes. For example, aquatic insect larvae sensitive to oxygen levels will be the first to leave a stretch of a stream if oxygen levels vary from the norm. Keeping track of such sensitive species allows earlier detection of ecosystem change compared to tracking a more robust species that tolerates a wide range of conditions.


The shrimp statistical zones in the northwestern Gulf of Mexico that were analyzed for bycatch community composition.

The authors examined data collected by SEAMAP (the Southeast Area Monitoring and Assessment Program), which conducts surveys along the Texas and Louisiana coastlines in the Gulf of Mexico. Trawls were conducted for scientific purposes only, eliminating any biases inherent in analyzing data collected from fishing vessels. Data was analyzed by season: summer surveys were conducted in June and July from 1982 – 2008 and fall surveys were conducted in October and November from 1986 – 2007. Survey results were further analyzed according into the Gulf’s “shrimp statistical zones” (geographic delineations used by managers to divide the gulf into smaller areas that are more easily overseen).

For those of you interested in the detailed math behind the calculations, please check out the paper for a complete description of the calculations they used when compiling data across years and when analyzing pooled data. The authors do a great job describing their analyses.

If you would rather have a quick summary of their statistical methods, here you go: the authors scaled catches by the effort that resulted in that catch so different trawl times, for example, would not affect their results. They then combined species catches across years. Their analyses not only evaluated whole communities (rather than just one species at a time), they also identified potential indicator species.


Summer and fall fish communities were statistically different and were therefore analyzed independently. There were four distinct community areas indicated by summer trawls, while fall surveys yielded only three distinct communities- one zone’s community that was distinct in the summer was not during the fall.

The four distinct fish communities identified during summer surveys and example indicator species from each community.

There are several fish that could be used as indicator species in this part of the Gulf of Mexico. Species were identified as good indicators of ecosystem condition if they were very abundant but only in one distinct fish community.  Each distinct fish community shared at least one indicator species between summer and fall surveys. Fish that could be used included Atlantic bumper, Atlantic croaker, Atlantic cutlassfish, brown shrimp, dwarf goatfish, longspine porgy, and northern white shrimp.

The authors emphasize that the fish communities they analyzed may differ from the actual fish communities caught as bycatch during shrimp trawls in a few ways. Most notably, data on the potential indicator species identified in this study should be compared with bycatch data from shrimp trawls. Species important in both this study and in bycatch datasets would best serve as ecosystem indicators when analyzing bycatch data.

Environmental conditions, particularly the summer hypoxic zone locations, likely explain the difference in fish communities between summer and fall trawls. Different fish species have different sensitivities to low oxygen areas, and therefore are more or less able to tolerate being in close proximity to hypoxic waters. Life history (specifically, location preferences at different life stages) also influences which species were caught in SEAMAP surveys. Different species exhibit varying affinities for habitats over which trawls were conducted depending on fish breeding status and age.

Knowing where fish communities differ along the Gulf of Mexico coast is the start to effective monitoring and preservation of essential ecosystem services. Not only does this study examine a data set of impressive size conducted over a number of decades, but it also identifies species that could be used by themselves to indicate the health of the entire fish community found in a particular area. I am very excited to see these statistical techniques applied to data collected from actual commercial shrimp trawls. Comparing commercial trawl bycatch data to these SEAMAP results could tangibly advance ecosystem management efforts, which is a great use for thousands of animals that would otherwise be destined for discard.

Leave a Reply

Your email address will not be published.