In this section
- The Oceans
- The World Fisheries
- International Cooperation
- Fishing Technology
- Cruise Ship Pollution
Modern fishing technology was built to catch as many fish as possible as efficiently as possible, and it is very effective at doing this. Unfortunately, it achieves this effectiveness at the expense of the health of ocean ecosystems. The three main problems of current fishing technology are: the destruction of the ocean floor and environs, ghost fishing, and bycatch.
Destruction of the ocean floor and environs
The main cause of this part of the problem is a fishing method called bottom trawling. Bottom trawling is usually done by one or two fishing vessels with a large net that is dragged along the bottom for a few hours at a speed of three or four knots. Target species vary widely, and can include orange roughy, hoki, ling, hake, and squid (Starfish, 2003). There are several types of bottom trawls: dredging, beam trawls, and demersal otter trawls. Each of these methods involves scraping the ocean floor with a rigid metal frame, causing massive damage (FishOnline, 2007). Some of the newest equipment is even worse; "rockhopper" equipment, which is used in areas that have rough terrain, rolls weighty wheels over the sea floor, crushing anything in their way (Safina). Unfortunately, bottom trawling is the world's most lucrative fishing method, and so is quickly replacing much more ocean-floor-friendly methods like hook-and-line and trapping (Gabriel, 2005; Safina, p. 4).
Bottom trawls can dig up to several inches into the seabed, disrupting the bottom habitat and the animals that live there, including unique structures made by living creatures. Trawls kill marine life, destroy food sources and shelters, and endanger young fish and future generations of ocean fauna (Safina, 5). Lab studies of the relationship between sea floor composition and predation showed that more complex habitats like rocks, rather than simple habitats like sand or mud, gave prey fish like young cod more time to escape their predators (Safina). Some areas of the oceans are trawled as many as 8 times per year. Each trawl pass kills between 5 and 20% of the marine life on the sea floor, so that even a single year's trawling can completely destroy the bottom life (Safina, p. 7-8). Deeper areas are affected even more; a study by Watling reveals that "certain bottom communities may need as much as a century" to recover (Safina, p. 18).
Dragging metal frames along the ocean floor also kicks up a cloud of sediment. While this does make the nets more efficient, as the dirt helps bring fish into the net, it also causes numerous problems. The increased amount of particles suspended in the water can diminish light levels and stifle bottom-dwelling inhabitants (Jones, 1992). Trawl gear can also redistribute vertical layers of sediment, mixing organic material into the water and creating anaerobic conditions that can kill scallop larvae (Jones, 1992).
The dragging metal frame of bottom trawlers indirectly kills significant quantities of ocean life, including commercial fish such as cod. Is it not, then, in the interest of everyone to find a way to fish without destroying the ocean floor?
A bottom trawling net with rock-hopping gear that leaves a path of destruction.
Fishing nets and traps today are made of durable polymer fibers, built to last. While this seems great at first, this durability can kill millions of fish and other organisms. When fishing nets or traps are lost due to storms or negligence, they actually continue to catch fish (Gabriel, 2005). And thanks to those polymer fibers, they can keep catching fish or crabs or other life for months or years. To make matters worse, many traps and nets become self-baiting: fish become trapped in the gear and die, other fish come to feed on the dead fish, become trapped themselves, and continue the cycle until the net becomes completely full (Matsuoka, 2005). This is called "ghost fishing," and it is probably the most frustrating problem plaguing the fishing industry today. Hundreds or thousands of fish or crustaceans can be caught in a single net, and the fish aren't even used in any way; they are completely wasted. According to Laist (1996), fish deaths caused by ghost fishing may be up to 30% as large as annual landings in some areas. Some countries, such as Sweden, Poland, New Zealand, and the United States, have already instated gear retrieval programs to try to address the issue of ghost fishing, but more, and more universal, measures will be needed if we want to completely solve the problem (Brown and Macfadyen, 2007).
Every fishing method has potential to catch and kill non-target fauna, called bycatch. It is by far the most widespread of the technology-related propblems (Gabriel, 2005).
Most commercial fishing methods today involve dragging an enormous net through vast amounts of water. Inevitably, these nets will catch fish other than those species the fishermen want. In 2005 alone, 7.3 million tons of commercially viable fish were discarded into the ocean (ICES 2005). This would be fine, if the fish were able to survive to breed or be caught later. However, despite federal regulations that bycatch be returned to the ocean as "unharmed as possible," most fish thrown back are either already dead or die shortly afterwards (Turning a Blind Eye, p. 1-3). Ironically, such discards are often the result of regulations that are trying to limit overfishing by placing quotas on the amount or minimum size of fish a boat can bring in (ICES, 2005).
High mortality rates for non-target fish species can change the ecology of an area by changing food web relationships, altering predator-prey interactions, and destroying the environment. In the long term, bycatch can lead to overfishing, decreased productivity, and reduction in the size of the total catch (Turning A Blind Eye, p. 5). Almost 1,000 marine mammals, many of which are from critically endangered species, die every day after becoming tangled in fishing equipment. The largest threat facing marine animals is the possibility of being caught as bycatch, rather than pollution or collisions with ships (Verrengia, p. 4).
Clearly, bycatch is a problem that must be addressed if we are to live in a world of sustainable fisheries.
Photograph from NOAA.
Bycatch includes everything from sand dollars to sea turtles.
Basics of Population Tracking
Population tracking is not limited to fish. In it simplest form, population tracking is the collection of basic information, such as density and distribution, about a population. There are several methods for population tracking; the most common are mark-recapture sampling, distance sampling, and index sampling.
Mark-recapture sampling involves capturing an individual, marking it in some manner (typically with a unique tag), and then releasing it back into the wild. When samples are taken later, the ratio of tagged individuals to untagged individuals can be used to create estimates of population size and density. Distance sampling involves taking visual counts of individuals while traveling along a predetermined path; the population size and density can be estimated using algorithms that account for differences in visibility between individuals close to the path and those further away. Finally, index sampling involves measuring the density of the effects of individuals (such as footprints for bears or nests for birds), and then extrapolating that data to get total population and distribution estimates (Elzinga, Salzer, Willoughby, & Gibbs).
Current State of Fish Tracking
Today, both scientists and commercial fishermen track fish. While both use sonar for biomass estimates, fishermen use the estimates to "catch fish more effectively, while scientists use \[the data\] to study fish distribution and estimate stock abundance" (ICES, 2006). Scientists also use larger-scale methods to survey fish populations, such as trawl surveys and tagging projects.
Trawl surveys are large-scale research efforts that can provide scientists with data on the types of fish, numbers of fish, and characteristics of fish in a given region. They have immense value; NOAA trawl surveys, for example, yield information for nearly 200 species (NOAA).
Trawl surveys are generally short in duration; NOAA uses a thirty-minute trawl (NOAA). ICES data indicates that trawl surveys are generally better with shorter durations than with longer durations, as the short duration gives a more precise point assessment and reduces the number of individual fish that are collected, reducing the work load on the scientists conducting the survey (ICES, 2005).
While there are many variations in trawl survey design, most consist of a short bottom trawl using commercial trawling gear. Following a predetermined course, the trawler makes repeated short trawls, analyzing the collected sample after each trawl. All of the fish in the trawl are counted, measured, and weighed, and then a small subsample is selected to be dissected to reveal age, diet, and health information (NOAA).
Research by ICES indicates that the distribution of fish species is dependent on depth and seafloor characteristics (ICES, 2005). Thus, an optimal survey would take into account depth and seafloor sediment type.
While trawl surveys produce large amounts of data on the distribution of fish, they cannot determine fish migratory patterns. Such information can be provided by a tagging project (Swain & Caradine, 1960). Tagging surveys allow scientists to record locations for capture and release, as well as vital data, such as length and weight, about each fish at both capture and release. The scientists can then use the data to document migration paths (Schwarz, Schweigert, & Arnason, 1993).
Tags come in two main classes: passive and active. There are many examples of passive tags; widely used varieties include pieces of plastic with identifying numbers that get attached to the fish and simple identifying paint marks on fish (Swain & Caradine, 1960). Another form of passive tags, known as PITs, is a bit more complex. PITs, or passive integrated transponders, are passive tags similar to RFID tags that have an embedded unique identification number and can be read electronically (Roussel, Haro, & Cunjak, 2000). Active tags, or tags that require a power source, are much more advanced than passive tags. Active tags currently deployed in some studies are capable of recording both water temperature and depth (Godo & Michalsen, 2000). These tags have advanced in recent years; some can now detect distance to the seafloor while others are capable of uploading data to satellites while the fish are in the ocean (Block, Dewar, Farwell, & Prince, 1998).
Problems with Today's Methods
Today's research surveys have some serious drawbacks. The sonar used cannot distinguish between fish of different species, making it useless for giving solid advice in a mixed-stock environment. Trawl surveys cannot determine migration paths accurately; they can only show that there are concentrations of fish in different places. Finally, tagging is very labor intensive; it requires fishermen to extract the tag from the fish and then send it in to a central data processing facility (Godo & Michalsen, 2000). This results in increased work for fishermen and a delay in the time from capture to data recording.
The Math behind the Method
Mathematics comes into play in two important areas of fish tracking: sample design and data analysis. Samples must be designed so that enough fish are collected over a large enough area to get an accurate representation of true fish populations. When analyzing the resulting data, there are several methods for estimating population parameters; the most appropriate method must be chosen.
Studies have shown that samples should be designed extensively, not intensively (Pennington & Volstad, 1994). Essentially, this means that many small samples should be taken instead of fewer, larger samples. Though reducing the sample size for each sampling location will reduce the total sample size, it will increase the effective sample size, which is more useful (ICES, 2005). Effective sample size takes into account that fish caught together tend to be more similar than the whole population; thus, more samples will result in more diverse samples (ICES, 2005).
The geographic distribution of samples is just as important as sample size and number. When designing the initial survey for any region, or if fish distributions in the target area are known to be random, a stratified random sample should be used (ICES, 2005). For areas in which fish distributions tend to follow trends, systematic samples give the best data (ICES, 2005).
There are several ways to estimate population parameters. The most basic method involves multiplying the average density of fish found in the survey by the total range of the stock; other methods relate the age of fish at capture to total biomass with complex statistical methods (NOAA; Pennington & Volstad, 1994). Estimated population parameters can then be applied to a logistic population model so that estimates of maximum sustainable yield may be made (Jensen, 1975).