Stay informed: Sign up for eNews Subscribe

Our Dying Planet An Ecologist's View of the Crisis We Face

  • by Peter Sale (Author)
  • September 2011
  • First Edition
  • Paperback
    $29.95,  £24.00
    $29.95,  £24.00
  • Title Details

    Rights: Available worldwide
    Pages: 360
    ISBN: 9780520274600
    Trim Size: 6 x 9
    Illustrations: 12 b/w photographs, 8 line illustrations

Read Chapter 1


We have always been fishermen. Fishing extends far back into our human past, and as our last remaining hunter-gatherer activity it ties us to that past in a tangible way. We capture wild aquatic organisms for personal use and to trade with other people. Most, but not all, of these fishery products are used for food. A trout fisherman on a Scottish stream who ties his own flies and approaches his sport with a quasi-religious fervor may have very little in common with the Malaysian peasant who fossiks at low tide for edible shellfish and crabs to feed her family, but both are part of fishing. So too are the giant multinational corporations, with their fleets of factory trawlers, their thousands of miles of longlines and nets, and their flash-frozen tuna air-lifted from deck to jet to Japan. Fishing is a vast global enterprise with a sophisticated array of technology and millions of people all engaged in extracting aquatic organisms from rivers, lakes, and oceans, trading them around the world, and consuming them in many different ways.

While we have always fished, we seem also to have usually overfished, leading to the reduction and sometimes the loss of formerly valuable fishery resources. Until recently, the consequences of such overfishing were generally local and temporary. Now, for the first time in human history, we face the possibility of widespread, essentially permanent collapse of the most important fisheries around the globe. Our continuing tendency to overfish is surprising given the investment in fisheries management around the world; on the surface, it does not seem to be that difficult to manage catch so that it does not exceed the capacity of the fished populations to supply.

Coastlines throughout the world provide scattered evidence of ancient fishing successes in the form of aboriginal middens-large piles of shells, bones, or other debris resulting from the capture, butchering, and presumably eating of the catch over many days and years by members of past cultures. In many middens, the size of the shells and bones varies with depth, with the largest buried deep in the oldest layers and the smallest occurring in the younger layers near the surface. This is evidence of ancient overfishing. The sizes of organisms being caught declined over time because, in all probability, the fishing was sufficiently intense to lower the life expectancy of the fished species. They lived less long, on average, before being caught and no longer reached the sizes they had in past years. And while we cannot tell this from examining a midden, it is very likely that as they became smaller over time, the animals being caught also became less common and harder to catch. Ancient fishermen often overfished and at some point had to search for new fishing grounds.

Most fish are remarkably fecund animals, producing thousands, sometimes millions, of eggs in a lifetime. Most also reproduce through external fertilization so that the female matures a clutch of eggs within her body before mating with a male. Most fish also grow in size throughout life, and because eggs take up space in the body cavity before they are laid, older, larger females are markedly more fecund than youngsters. Many fish also live for decades if not caught first. These life-history characteristics mean that fish populations can be remarkably productive, able to replenish their numbers rapidly following a decline in population size. The relative lack of parental care and the small size of newly hatched fish ensure that fish breeding success is strongly dependent on environmental conditions and a certain amount of luck-a fish population of a given size can produce an enormous cohort of young fish one year and far fewer the next.

Fishing is currently big business and vital to our food supply. The United Nations Food and Agriculture Organization (FAO) reports that, according to 2008 data, fishing provides 15.3 percent of the animal protein needs of the human population worldwide, or about 16.7 kg of fish per person per year. Commercial fishing directly employs about 44 million people and brings about 92 million metric tons of product to market every year, while aquaculture provides an additional 51.7 million metric tons. These fishery products are worth about US$91.2 billion and $78.8 billion per year, respectively, and the international trade in fishery products exceeds $92 billion per year. Adding in so-called illegal, unreported, and unregulated catches and the fish caught by recreational fishermen and by artisanal fishermen to feed their families around the world further increases the total tonnage of fishery species captured to 130 million metric tons per year.

The FAO also reports that the total world catch has been declining at the rate of about 0.7 million metric tons per year since about 1988, despite increases in fishing efforts. Globally, fishing is still very big business, but fisheries are failing to provide as they used to.

The decline in total commercial catch is one of several signs that our tendency to overfish is pushing us up against firm limits and that future catches may become far less bountiful than they have been. In this chapter we will look at fishing, sustainable fishing, and overfishing. We'll get into the science behind fisheries management and the reasons why management so often fails. I hope you'll become more aware of what takes place to make those fish available in grocery stores, and that you will appreciate the need to fish much more sustainably than we currently do. Along the way I will also touch on the more general problem of overuse of natural resources.

The Cod Fishery of the Northwest Atlantic

When John Cabot returned to England in 1497, he brought with him tales of plentiful Atlantic cod (Gadus morhua) of such size and abundance that catching these fish was simple. There were even claims that the fish were so abundant as to impede the progress of ships. Southern Europeans came to refer to the lands Cabot had found as Baccalaos, from the Spanish bacalao, the cod. The Portuguese had commenced fishing for cod off Newfoundland by 1501, followed shortly by the French and Basques. This commercial cod fishery was to last for almost five hundred years.

Initially, fishing off Newfoundland was an entirely ship-based operation. Ships sailed from Europe in the spring, fished intensively, salting down the catch in barrels, and returned home to the markets. But early on, the British, French, and Basques established shore camps where they could land the catch, salt it, and air-dry it. It was then packed dry for transport to the European markets; as a lighter product it was more economical to ship and, as a less heavily salted product, it was preferred by the public. To this day, many Europeans along the Mediterranean coast prefer dried, salted cod to the fresh product.

The colonization of Newfoundland was a direct consequence of the growing commercial fishery. Initially it involved the construction of seasonal dwellings for the people who worked the fishery processing the catch for shipment home. Gradually, seasonal dwellings became year-round homes as investments in real property began to require guarding it through the winter, but it was always the cod industry sustaining the development. Wars in Europe altered the overall fishing effort and the countries involved in the trade, and periodically the cod stocks failed, but on average the overall harvest grew year by year. As early as 1683, the problem of "overcapacity" was recognized by the Colonial Office in London-an excess demand for fish had fueled development of excess capacity (too many ships and nets) to catch them, and fishery stocks were failing.

Farther south, the Gulf of Maine cod fishery was "discovered," and Cape Cod named, by the crew of the Concord, a British ship sent to the New World to hunt for supplies of sassafras in 1602. Fishing vessels followed soon after, using the shores and the offshore islands of the gulf as suitable fish-drying sites. Fishing, and a Europe-based industry using seasonal dwellings on suitable shorelines, was well established by the time the first colonial settlements were being established in New England in 1620. However, the industry quickly became an American-based one, as local populations took up fishing, first in their immediate vicinity, and later in larger vessels venturing as far afield as the Grand Banks and northern Newfoundland. This was in contrast to the situation farther north, where the local Newfoundland and Nova Scotia populations operated inshore fisheries from smaller vessels and left the offshore fishery on the Grand Banks and the Labrador coast to be operated by larger vessels whose home ports were mostly in Europe. By the start of the eighteenth century, the Grand Banks fishery included vessels from England, France, Spain, and Portugal along with vessels from New England.

The cod trade grew so important that it became a vital source of foreign exchange for the developing American and Canadian colonies. It was incorporated into a profitable transatlantic trade in which the vessels that shipped dried cod to Europe returned with African slaves for the West Indies and southern American colonies, stocking up with sugar and salt in the West Indies before moving again to the fishing grounds of New England and the Grand Banks. Simultaneously, some vessels shipped the lower-quality fish south to feed the slaves in the West Indies and transported sugar back to Europe.

In these early days, fishing was done by hand-line from the decks of the vessel. Beginning in the nineteenth century, however, new methods were developed. Cod seines, gill nets, and cod traps were used to a limited extent in coastal waters, and small dories began to be carried by the offshore vessels so that hand-liners could spread out over a wider area to fish. By the 1850s, longlines with hundreds of hooks began to replace hand-lining in the offshore fishery, but it was at the start of the twentieth century, with the arrival of trawling, that fishing methods made a major advance in effectiveness.

The otter trawl was introduced to the U.S. Atlantic seacoast in 1908 but was not used in Newfoundland waters until 1935. An otter trawl consists of a large baglike net that can be dragged across the seafloor, with two large otter boards, or doors, mounted on the towing lines at the ends of the trawl's wings-the outer corners of its mouth. The doors can be as big as garage or barn doors and may each weigh 1,000 kg in commercial trawls that have mouths 100 meters wide. The doors are rigged so that hydrodynamic forces tend to move them outward, spreading the wings and pulling the mouth of the net open. Floats or kites lift the headline of the net to keep the mouth open vertically, and the footline is weighted and protected in various ways to keep the net in close contact with the substratum. The otter trawl proved to be very efficient at catching cod and other groundfish, and trawling became the principal method of capture in this fishery.

In addition to the introduction of trawling technology, the twentieth century saw increased use of steam and diesel power, of refrigeration and flash-freezing, and of long-distance rapid transport to market by truck, train, and plane. The result was that the Northwest Atlantic fishery was presented with an ever-expanding market and the temptation to continue to expand the fishing effort to supply the demand.

So what do we see when we look at the catch of cod? Detailed examination of the early fishery, region by region, reveals many examples of stock declines and resulting poor catches, but the solution was simply to expand to new fishing grounds. For example, a failure of the southern and southeastern inshore Newfoundland fishery in 1715 provided the impetus for expansion to the northeastern Newfoundland shore and for a progressive expansion of fishing on the Grand Banks. And with each shift to more distant fishing grounds there was a shift toward larger vessels and more fishing effort to cover the additional costs. The growth in the catch proceeded as the area being fished expanded, as technology advanced, and as markets opened up. By 1765, the total catch for Newfoundland, the Grand Banks, Georges Bank, and coastal waters was about 180,000 metric tons, supporting a brisk trade with Europe and the West Indies. Catches declined during the American War of Independence but then recovered. By the mid-1800s, the total catch of cod from the Northwest Atlantic was about 200,000 metric tons, but it increased further, reaching 260,000 by the early 1870s. By 1895, the Northwest Atlantic cod fishery was landing 420,000 metric tons, and it continued at about this level, fluctuating between 400,000 and 700,000 metric tons, through to the Second World War. By 1955, the catch had reached about 1,000,000 metric tons, and it peaked at about 1,900,000 metric tons in 1968. Thereafter, catches declined progressively, to about 500,000 metric tons in 1975 and 80,000 metric tons in 1990. The Canadian government closed the northern cod fishery in 1992 and all groundfishing in Canadian Atlantic waters in 1993. Since then, cod stocks have shown minimal recovery. A commercial fishery that had provided enormous economic and nutritive benefits over five hundred years was finished.

From the commencement of commercial fishing, there were local declines or outright failures in the cod fishery. With hindsight, it's possible to see that in a situation in which anyone with the funds to secure a vessel could join the fishery, there was always a tendency to overfish local cod stocks. In the 1600s and 1700s, fishing was restricted to those locations that were near to land or home ports. When fishing yield declined in those locations, it was possible to travel to new locations. The result was that diminished stocks often had a chance to recover, while the fishery was sustained commercially by turning to previously unfished stocks. However, once the fishery grew so large that all fishable locations in the region were being fished, the tendency to overfish still reduced stocks, but there was nowhere else for the fishing effort to go.

If fishermen were not inventive and had continued using hand-lines from relatively small boats, it is possible that the catch of cod would never have grown to the size it did, and the collapse of the 1990s would not have occurred. But that is not the nature of fishing. Fishermen are wily predators, always looking to innovate to capture their prey faster and more economically.

The collapse of the cod fishery provides three clear lessons. First, there is a profound difference between the local failures that occurred from time to time during the early years of the fishery and the final overall collapse. Second, the combination of growing demand and improving technology led to ever-expanding effort and ever-growing yield up until the eventual collapse. Third-but not evident from the information I've provided so far-the fishing effort acted in concert with other factors to bring about the decline in cod populations. To fully understand what happened, it is necessary to move beyond a focus on effort and catches to examine the myriad factors that determine how abundant a population of fish will be and how fishing changes that. To do this we have to dip into theory. It's not particularly complicated theory, so bear with me.

Effects of Fishing on Fish Populations

Logic dictates that populations of fish (or other species) grow when more fish are born than are dying, and they decline when more fish die than are being born. Ideally, a population will remain at constant size if each female produces, on average, the number of offspring needed to ensure that exactly two of them will reach adulthood and breed in their turn. (Two are required because in most species of fish, as in other animals, 50 percent of offspring are males.) That a female cod spawns millions of eggs each year and can live up to twenty more years after reaching maturity at five or six years tells us that very few hatched cod eggs grow up to become adult spawning cod. There are lots of things that happen to kill cod, nearly always well before they reach sexual maturity. Only one of these is fishing, which principally kills older fish.

From the perspective of the fish, fishing is just one more form of predation-one more challenge in its struggle to survive and reproduce. When fishing commences on a previously unfished population, it increases the chance of mortality, with the result that fish live, on average, less long before they die. In addition, fishing is a size-selective form of predation that tends to have the greatest impacts on the larger and older members of the population. While Atlantic cod can live for twenty-five years or more, by the early 1990s fishing was so intense that most cod were being caught before they were seven years old.

Because of these basic facts, there are several consequences of starting to fish a population. First, because animals tend to die younger, the population tends to become smaller than it was before, because each individual is present for a shorter period of time. Second, because the animals tend to die younger, they have fewer seasons after reaching sexual maturity in which to spawn-two or three seasons versus as many as twenty seasons in the case of cod. The result is that each successful fish (one that reproduces at least once) produces fewer offspring over its (shortened) lifespan. Furthermore, because fish are more fecund when they are older, the actual reduction in the production of offspring is substantially greater than the reduction in the number of spawning seasons might suggest.

Given these simple facts, how do we manage a fishery so that it can be maximally profitable without leading to the decline and extinction of the fished population? Managers have relied traditionally on three factors that may make it possible for fishing to increase predation on a population without wiping it out. These are density dependence, the storage effect, and the relationships among cost, catch, and effort in a fishery. The first two are aspects of how populations grow; the third is an economic relationship in the fishing activity.

Density dependence is central to ideas concerning the regulation of numbers in a population and for that reason has featured importantly in the history of ecology. It is also central to the simplest ecological model of population growth-the logistic model. As already noted, the pattern of growth of any population is determined by the pattern of births and deaths within it. Both birth and death rates depend upon the average age and condition of the individuals that make up the population, and by convention we speak of a per capita rate of increase, meaning "the rate per individual at which the population grows." Each member of the population requires food, shelter, and other resources in order to survive, grow, and potentially reproduce. When a population is small relative to its available resources, its individuals are likely in good condition-well fed, growing at maximal rates, healthy. They should possess a relatively high life expectancy and should produce offspring at a higher rate than individuals of a larger and denser population. The high per capita rate of increase causes that population to grow, but, following the logistic model, as the population grows the individuals will begin to experience shortages of resources such as food or shelter space. These shortages will cause the individuals to grow more slowly, be less fit overall, produce fewer offspring, and die at a younger age on average. As a result, per capita rate of increase falls, leading to a decline in the rate of growth of the population. Thus we can see that per capita rate of increase is dependent on the density of the population relative to its resources; because the dependence is negative, there is a tendency for any episode of population growth to cease and for the size of the population to stabilize. The population size at which this occurs is termed the carrying capacity of its environment-that size at which the availability of resources relative to the numbers of individuals competing for them sets birth and death rates to be exactly equal (see Figure 1). Animals are still busily growing, reproducing, and dying, but the rates at which these happen balance one another and keep the size of the population constant. The logistic model can be redrawn to show the rate of growth of the population at any given population size. The growth rate is at a maximum when the population is half the size it will ultimately reach at carrying capacity.

Looked at from the perspective of the fishing industry, this logistic curve indicates that if fishing reduces the size of the population from where it was before fishing started (its virgin state when it was presumably at carrying capacity), the capacity of the population to grow will become progressively greater until the point that the population has been reduced to half its virgin size. By fishing at a rate that removes individuals quickly enough to keep the population at this size, the fishery will gain the maximum sustainable catch that the fish population is capable of providing. (This statement is correct, but doing this in a real fishery is more difficult than it might seem.)

Very similar approaches are used to maximize yields in other harvested populations. We mow pastures for hay at a frequency designed to capture the burst of rapid plant growth before the plants become large and crowded. We harvest forests on a longer cycle but follow the same principle. And we take cattle and pigs to market at an age that optimizes growth prior to sale. In all these examples, we maximize yield because of density dependence, relying on the idea that younger and less-crowded organisms grow more rapidly.

Theoretically, by doing sufficient fishing to keep a fish population at this one-half of maximum size, a well-managed fishery will be able to fish indefinitely, taking the maximum sustainable yield (MSY) of fish per year, and the population will continue to produce into the distant future. Clearly this rosy future did not befall the cod or any of a number of other species.

There are two important things to notice about this simple model of density-dependent logistic growth. First, while the capacity of individuals to grow and reproduce is highest when the population is smallest (because there are ample resources available for each individual), the overall capacity of the population to grow more abundant will decline once the population is pushed below one-half of its virgin size (because a few individuals cannot produce large numbers of offspring quickly). The desirable one-half of virgin size for the population is not a stable equilibrium-the population will tend to move away from this point unless fishing is very closely regulated, and the further it is pushed below this point, the less capacity it will have for growth and recovery. If one is interested in the long-term yield of the fishery, seeking to fish at a rate that will achieve MSY is a very risky goal that demands exquisite control of the rate of fishing.

Second, this model assumes that the production of resources and the status of all other things in the environment that impinge on the condition of the fish are unvarying. If the availability of resources varies independently of the size of the fish population, if environmental temperature changes (so that metabolic rates, and therefore rates of growth for given caloric intake, change), or if any other environmental feature changes in a way that modifies the growth, fecundity, or survivorship of the fish, then carrying capacity, rate of population growth at a given population size, and the size of the population at which maximum yield is obtained all change. Under these circumstances, maintaining the population at the magic equilibrium size can become a very difficult task indeed. Needless to say, environments are rarely unvarying.

Variability of environmental characteristics is so pervasive that we should never forget it, even if the simple logistic model of population growth assumes variability is unimportant. In fact, fish populations have been telling us for a long time just how variable their environments are. They do this by demonstrating tremendous variability in recruitment.

Recruitment is the addition of a cohort to a population. It happens to armies when raw recruits go to boot camp, and it happens to biological populations when new groups of juveniles are added to the population each breeding season or when new groups of juveniles reach adulthood. Recruitment is a measure of progress through the ranks of the population, and it can be measured at any life stage. Fishery biologists frequently measure recruitment at the time that young individuals become large enough to be caught by the particular fishing gear being used. Ecologists tend to measure recruitment to specific life stages, such as to the juvenile stage following a larval period or to the adult stage at the time of maturation.

In a fish population, recruitment-whether you measure it at the end of larval life, at sexual maturation, or at the time the fish get big enough to be caught by a gill net of a particular mesh size-is profoundly variable from year to year. Science has known about this since 1914, when the Norwegian fishery biologist Johan Hjort documented the very great variation from year to year in recruitment to populations of a number of commercial fishery species in the North Sea. In the years since, it has become abundantly clear that the production of a new cohort of fish is a very risky business that is sometimes crowned with massive success and is at other times an absolute failure. Looked at another way, while only two of a female cod's millions of offspring are likely, on average, to reach maturity, the actual breeding success of individuals varies very widely around this average, and thus the breeding success of populations varies very widely from year to year.

In species that have relatively lengthy lives, such as most fishes, the population at any particular time is composed of a number of cohorts of individuals, each the product of recruitment in a particular year. Recruitment variability means that these successive cohorts start out at very different sizes and will probably preserve these differences throughout life. Indeed, the main conclusion of Hjort's classic study was that variation in recruitment results in the formation of occasional particularly abundant cohorts, so-called strong year-classes, that tend to dominate the catch and sustain the fishery over several years.

The main reason in fishes for variation in production from year to year is that there are years when greater proportions of newly hatched eggs survive and years when smaller proportions do. Why this happens is less easy to explain but has to do with environmental variability that modifies the likelihood of survival in these early stages. Given that a cod lays several million eggs in a season, it's clear that the probability of survival is normally very low indeed (or we'd be up to our necks in cod), so very modest changes in the chance of survival will lead to very large changes in the number of fish recruiting. Among the environmental changes that may be important are weather patterns that delay the plankton blooms the newly hatched fish depend on for food, ocean current patterns that carry the larvae to places that are quite unsuitable (or, alternatively, very suitable) for their survival and development, and temperature patterns that cause them to grow more quickly or more slowly than usual and thus alter the risk of predation on or the demands for food by these tiny larvae. (A slowly growing larva is small for a longer time and runs a greater risk of getting eaten because of this.)

In a population made up of relatively long-lived individuals, the effects of good recruitment can be "stored," meaning that the reproductive capacity of strong year-classes remains for many years, buffering recruitment variability. While a population with many year-classes of animals present will receive only a modest boost in overall numbers in years when recruitment is highly successful (because each year-class is only a small component of the total population), it can survive many years with very poor recruitment (because there will still be animals maturing and reproducing). By contrast, a population of short-lived individuals containing only a handful of year-classes will exhibit far greater fluctuations in overall size as good and bad recruitments occur and will be able to tolerate only short runs of poor recruitment without going extinct.

One consequence of fishing is that, because it increases mortality and lowers average age, it tends to reduce the storage capacity of the fish population. This makes a fish population more vulnerable to a series of years of poor recruitment than it would otherwise be. When fishing pressure is relatively light, however, storage in the population makes it possible to continue having good catches despite variation in rates of replenishment of the population through reproduction and recruitment. The good year-classes sustain the fishery through years of poor recruitment.

Fishing is predation, and fishermen are efficient predators-they do not waste effort, and they are skilled or they do not survive. Fishing was a very important activity to the Melanesians and Polynesians who migrated out from Southeast Asia to populate all the scattered islands of the South Pacific, nearly one-fifth of the surface of the planet, by 1000 A.D. During this expansion into the Pacific, they developed a very broad range of fishhook styles and materials. Differences in design and manufacture of fishhooks have been used extensively by archeologists to track cultural connections, and bone or shell fishhooks of various designs are now sold as tourist curios throughout the region wherever tourists with pocket change congregate. But these fishhooks-the finely crafted tools of their trade-are really a testament to the sophisticated fishing skills of these island-dwelling people. Each hook was specifically designed to catch a particular species of fish with particular jaw structures or behaviors, in particular ways, and at a particular time and place. The level of sophistication easily rivals that of the flies tied by that Scottish trout fancier I mentioned earlier, but the Polynesians used their tools to provide food rather than for sport. Worldwide, coastal peoples still use hooks, hand spears, nets, and traps of various types, and they use their hands to pick up slower-moving species such as mollusks. They fish effectively, and fishery products form an important part of their diets and provide trade goods, including jewelry, medicines, and other materials.

Yes, fishing is predation, but it is also an economic activity: commercial fishermen fish to make money. Being an economic activity, fishing should be subject to market forces, and in many ways it is. Fishing has costs. These include the cost of the boat and the equipment and wages for the crew. Costs are linearly related to the amount of effort expended by the fishery in catching fish. Effort is a measure of the overall investment-in dollars and time-by the fishery. A fishery involving ten ships of a particular size and type costs about half per year what a fishery involving twenty ships of this type would cost. And the ten-ship fleet exerts about half the predatory pressure on the fish population that the twenty-ship fleet does. A fishery using faster, more wide-ranging vessels or vessels carrying more sophisticated gear for tracking fish costs more per year than a fishery using smaller, less elaborate vessels and exerts correspondingly greater predatory pressure on the fish population. If the value of the yield in marketable fish exceeds the cost of catching the fish, profits are made and fishing continues. If the yield does not match the cost of catching, losses are incurred, and we may anticipate some fishermen getting out of this business into something more lucrative.

Ideally, the economics of the marketplace should provide a very reliable regulator of fishing effort, because the value of the catch does not increase linearly with effort. Because fishing reduces the average age and therefore size of individuals as well as the overall abundance of the fished population, it becomes more difficult to obtain fish of high market value (larger sizes usually) as the fishing effort increases. Given that there is a certain rate of production of fish available to be caught, it follows that the value of the catch obtained will increase with effort only to a certain point. Beyond that point, increasing effort will result in a yield of lower value because few fish remain to be caught. Increasing effort still further should lead to a yield of less value than the cost of catching the fish, and fishing harder still could lead to the removal of all available fish (extinction). As Figure 2 shows, the interaction of cost, yield, and effort should lead to a stable if rather unhappy equilibrium in which effort rises to that point at which costs equal yield (and the fishery does not make a profit), but effort should not rise higher than this. This should be so even when all fishermen are thoroughly selfish and fish to obtain the maximum catch possible, so long as they do not go broke doing so.

Drawing these three factors together-storage capacity, density dependence, and the links among cost, catch, and fishing effort-theory suggests that it should be rather easy to ensure that fishing be a long-term, sustainable pattern of exploitation. The storage effect ensures that the fish population will be buffered from the natural variation in production as well as from any modest fluctuations in fishing pressure. Density dependence in demographic properties provides confidence that the fish population can compensate by increasing rates of production of new fish when fishing pressure reduces fish density. And the economic links between the cost of catching fish and the value of the yield at market should mean that fishermen, being rational beings, will never increase effort to levels that would be truly detrimental to the fish population. Would that it were all this simple.

This ideal situation is not reality. The economic extinction of the cod fishery is only the latest example in a long series of apparently well-managed fisheries that have been overexploited and have collapsed. But why is this so? Part of the problem lies in the simplicity of our model-fishing has strong ecosystem effects beyond those of simply removing some fish, and fish populations are impacted by things other than fishing. The graph in Figure 2 does not account for fishing's reduction of the storage effect, which makes the fish population less capable of weathering a series of poor years. Nor does it provide for the environmental variability (the poor years) that results in the demographic variation that makes the effort required to obtain the maximum sustainable yield (or indeed any specific yield) change from year to year. In fact, it assumes environmental variability does not exist. A more realistic Figure 2 would show two blurry clouds of points in place of the cost and yield curves that intersect so precisely at a single point. Above all, Figure 2 provides no way of showing the several other ecosystem effects of fishing that may radically alter the challenges facing fishes as they seek to survive and reproduce.

The other part of the problem lies in our tendency to think of fishery management as the management of a simple interaction between a predator and its prey. Fishermen are indeed predators-rational, intelligent predators-but they are also members of real human societies participating in an economy. Figure 2 does not include the effects of political decisions made to sustain human communities and the fishing industries they depend on for livelihood when they experience hard times-the debt relief, the unemployment benefits, and the other governmental actions intended to help families in need but that also allow people to remain fishermen when cold economic reality should be causing them to turn to other employment. Government policies intended to mitigate economic misery in the short term can have unintended negative consequences for the long-term sustainability of the fishery.

In the remaining sections of the chapter, I describe what is currently happening to fisheries worldwide. I then review the important ways in which real fishing differs from our simple theory and discuss some of the ways in which fisheries managers have been able to deal successfully with these departures from theory to make fisheries sustainable.

Current Global Trends in Fisheries

Overall global fishery production has declined slightly in recent years, despite continued growth in effort. Unless we can reverse this pattern, it probably signals the beginning of the end of our ability to extract fishery products from wild stocks in a commercially viable way. Just as long ago we learned to farm animals and plants instead of harvesting wild game and wild plant products, we will come to rely on aquaculture for our fishery products, and the eating of wild fish will become as exotic as the eating of wild game (really wild, not farmed bison or elk). The scale of modern commercial fisheries is such that this transition will represent a major shift in the ways in which we feed the world's human population, and the nature of aquaculture suggests it will not be a transition to a rosier future.

Daniel Pauly of the University of British Columbia has spent his career in efforts to improve our management of fisheries, particularly in developing countries where there was rarely an adequate management infrastructure or reservoir of expertise. Pauly and colleagues raised the alarm in a series of papers in Nature and Science at the close of the twentieth century. We had reached the point where we were no longer able to increase the tonnage of fish removed from the world's oceans, and in the process we were making substantial changes to the structure of fish populations and the ecosystems to which they belonged. Scientists from other universities who have looked at the data independently have largely supported these claims.

In fairness, Lewis (Loo) Botsford of the University of California at Davis had reported the perilous state of global fisheries in 1997, documenting the high proportion of fisheries classified by the FAO as fully or overexploited and the extent of indirect ecosystem impacts due to fishing activities; and the FAO itself has always reported dispassionately on the difficulties facing world fisheries and the relatively limited improvements in management that have occurred. The FAO predicted a limit in global marketable catch of about 80 million metric tons as early as 1971 and showed that limit as having been reached in the early 1990s. Pauly's alarm call, therefore, was based on information that had been around and publicly available for several years. Figure 3 shows the world fishery yield from 1950 to 2006.

Unfortunately, dispassionate reports by the FAO to the United Nations and its member states do not always attract media attention, and the eyes and ears of the public, in the same way that a prominent article in Science or Nature may. Fisheries management is ultimately a national responsibility, and political decisions by many nations are strongly influenced by public opinion. It's good that there are scientists such as Pauly who work in universities and nongovernmental organizations and are able to get the message out to the wider public.

Figure 3 is based on the worldwide commercial fishery data compiled by the FAO since 1950. These data are based on national statistics provided to the FAO by individual countries. The FAO cross-checks the national submissions, works with member countries to improve their fishery data, and, where necessary, makes adjustments based on other available sources of information on fisheries in each region. The FAO statistics are not perfect, but they represent the best data available on commercial fisheries per country. They include a variety of types of information beyond annual catch and include information on aquaculture yields. They provide the basis for the biennial technical reports produced by the FAO and for much of the international policy developments that have enabled fisheries management to improve to the extent it has, and they are publicly available via the FAO website for others to use. What Pauly did was to draw attention to the over-reporting of catches in the Chinese fishery, make reasonable corrections for this bias, and remove from the global total catch the widely fluctuating catch of Peruvian anchoveta in order to see the underlying global trend in total catch. The underlying trend is downward.

Over-reporting of catches is an unusual bias. (Commercial fishermen usually prefer to under-report; only recreational fishermen tend to over-report, at least when it comes to the sizes of the ones that got away.) It happens when there is political pressure on the industry to meet high catch targets. The centrally planned economy of China was routinely setting targets for all industry managers, including targets for fish to be landed, and by the early 1990s targets had grown well beyond what the fishery was able to catch. For Chinese fisheries managers, keeping one's job was dependent on meeting the set goals, whether or not the fish actually got caught. The FAO had become aware of this problem with the Chinese data at about the same time Pauly drew public attention to it and worked with the Chinese authorities to remedy the situation. Part of the remedy was a characteristically diplomatic move by the Chinese government-a public declaration of a "zero-growth" policy in 1998 that fixed catch targets (which will presumably remain in place until real catches no longer require an "upward nudge" as the data are compiled). Another part was the FAO's decision, commencing with its 2006 report, to report Chinese catches separately from those of the rest of the world.

Pauly corrected for the Chinese over-reporting, revealing a downward trend of 0.66 million metric tons per year that began in 1986. This downward trend has occurred despite the fact that the global fishing fleet is about 30 percent larger than it needs to be-the global decline in catch occurs in spite of a more-than-sufficient effort to catch fish because sufficient fish are simply no longer there.

The sizes of fish being caught have also been falling because fish are being caught at younger and younger ages. This phenomenon appears universal, and in many instances the reduction in size and age attained is profound. In 2003 Ransom Myers and Boris Worm of Dalhousie University in Halifax, Canada, documented the global extent of this phenomenon, reporting that industrialized fisheries typically reduced the community biomass (the total weight of living organisms present) by 80 percent within fifteen years of starting to fish and that the larger species are most severely impacted. They estimated that the biomass of large predatory species was on average now about 10 percent of what it had been prior to the onset of commercial fishing. This removal of 90 percent of the biomass of the larger species brings with it a marked reduction in size (and age) attained. For example, whereas Atlantic swordfish once regularly lived 20-plus years and grew to weigh more than 450 kg, the average one landed in 1995 weighed about 41 kg and was only three to four years old.

While the reductions in size are impressive, it is the reduction in average age at capture that is of most importance ecologically because of the obvious effects on the storage capacities of the populations. With fish living less long, there are fewer annual cohorts present in a species' population, and its capacity to withstand years of poor recruitment is reduced. In addition, the total lifetime production of offspring per individual is reduced because females reproduce over fewer years. Indeed, for species such as the Atlantic swordfish, the average age at capture is less than the age at maturation, meaning that the average fish is being caught before it reproduces.

The overall effect on fecundity is, of course, greater than the reduction in number of episodes of reproduction might imply. As fish grow older and larger, fecundity of the females increases exponentially because their larger body cavities can contain much greater numbers of eggs. For example, in the red snapper, Lutjanus campechanus, which commonly lives nine to eleven years and may live to twenty years, a single 61 cm (12 kg) female eight to ten years old will spawn the same number of eggs as two hundred twelve smaller females 42 cm in length (1.1 kg) and three to four years old. And that is just in the one year. While the overwhelming majority of these eggs are going to die very young, the reduction in average maximum size of the fish must greatly reduce the number of larvae being produced, making it less likely that the population will be able to produce outstandingly large cohorts of offspring in the occasional good years. This further erodes the ability of the species to capitalize on good years and thereby survive the poor ones.

Finally, in a number of fishes, including the groupers (Serranidae) and snappers (Lutjanidae) that are some of the most important fishery species in tropical coastal waters, animals normally mature as females, only to transform into males later in life. In such sequentially hermaphroditic species there is an additional potential impact of overfishing: By selectively removing the larger, older fish, the fishery selectively preys upon males, and the risk exists that the number of males may become so small as to limit the availability of sperm to fertilize the eggs. In some such species, the regulation of sex change may be strongly age based-an automatic developmental event that occurs at a specific age-but in the majority of cases it is mediated socially through behavioral interactions or pheromonal communication within the social group. With social mediation, the problem of male depletion is probably reduced-animals will simply become male at smaller sizes-but it is not eliminated.

Scarcely studied at all, but almost certainly as important as disruption of sex ratios, is the simple fact that when overfishing is sufficient to change population density and age structure, it must also disrupt social structures in those species that have social organizations more complex than a simple school. Disrupted social structures can be expected to result in reduced reproduction, regardless of whether the fish are hermaphroditic. Behavioral and ecological studies of fish on coral and rocky reefs using snorkel and scuba to permit direct observation and experimentation show us that the great majority of demersal fishes in these structurally complex habitats have complex social organizations. The idea of fish as anonymous individuals drifting haphazardly through a uniform environment and waiting to be caught makes it easy to think of fishing as a two-agent interaction involving the fisherman and his prey with nothing else being important. The reality of fish as individuals interacting differentially with other members of their population in a spatially variable environment means that fishing does have effects that depend on which of the various individual fish are caught. Catching all the older members of a population can have different consequences than catching some of the younger members; however, as yet we have paid far too little attention to this impact of overfishing.

It is not only the total marketable catch that has deteriorated. In 1998 Daniel Pauly and his coworkers used Science to alert the world to another sign of problems for world fisheries stocks. Again using the FAO database, they reported that the mean trophic level of species taken by the world's fisheries had declined between 1950 and 1994. They described the phenomenon as "fishing down the world's marine food webs."

To understand this concept, we must first appreciate the modern quantitative method for measuring an organism's trophic level. When he first introduced the concept in 1927, the Oxford ecologist Charles Elton described the trophic level of an organism quite simply: organisms exist at differing levels in a food chain, with primary producers at level 1, herbivores at level 2, carnivores that eat herbivores at level 3, and so on up as high as level 4 or 5. Omnivores that eat a mixture of plant and animal species or that eat a diet of several different types of animal from several different trophic levels were conveniently overlooked in Elton's scheme. The more quantitative approach used by Pauly requires detailed diet data for each species in the ecosystem being fished. The trophic level of a given species is determined as the average trophic level of the foods it eats plus 1. Thus a species that eats 40 percent plants (trophic level 1) and 60 percent strict herbivores (trophic level 2) sits at trophic level 2.6 (0.4 × 1 + 0.6 × 2 + 1 = 2.6).

Pauly and his coworkers computed the trophic level of each species being caught commercially; then, for each year from 1950 to 1994, they calculated the species composition of the catch for each of the major marine fishery regions of the world and for the global catch. They were then able to determine the average trophic level of the catch each year. Their results demonstrate a gradual downward trend in average trophic level of the global fishery catch, for both salt water and freshwater fishes. The global trends are each comprised of sets of trends for each major fisheries region of the world, and these are not identical. In marine regions, the majority of trends are downward, but a few regions show no significant change (e.g., Indo-west Pacific) or even a trend toward higher trophic levels (e.g., southwest-central-southeast Atlantic). These atypical trends can usually be explained by commencement of fishing in new locations or depths or on previously untapped resources-a process that is unlikely to continue much longer since all regions are now being fished.

What causes this downward trend, and what is its significance? Fisheries target species that are economically valuable. Indeed, the enormous quantities of bycatch that characterized most fisheries until recently were simply those fish that were of so little economic value that they were not worth bringing to shore. Traditionally, economically valuable species have tended to be large-bodied, usually older, and almost always piscivorous (fish eating). A decline in the average trophic level of the catch is a clear signal that fisheries have changed their targets through time. The catch has come to be comprised increasingly of fish that feed at lower trophic levels, and these tend to be smaller, younger animals.

If you have been buying fish to feed your family over the past several years, you have seen evidence of fishing down the food web, even though you may not have realized it. All those new kinds of fishery products on the supermarket shelf are there because those are the species that are now being caught. They used to be avoided or thrown back as bycatch. In the Northwest Atlantic region, the traditional target species were initially cod and, shortly thereafter, haddock. Both of these are relatively large, high-trophic-level piscivores that are of high economic value because of the relative ease of capture and the quality of the meat. By the time the trawl fishery largely collapsed in the Northwest Atlantic in the 1990s, the catch from that region was made up in about equal parts of demersal fishes, pelagic fishes, crustaceans, and mollusks. As well as cod and haddock, the catch now included hake, Atlantic redfish and American plaice (all demersal), the pelagic herring, crustaceans (shrimp, crabs, and lobster), and mollusks such as scallops. All of these additional species exist at lower trophic levels than cod and haddock. Similar changes have occurred in other regions. Large piscivores that formerly dominated catches have been replaced by smaller-bodied piscivores, planktivores, and fishery species that are not fish-the squid, clams, lobster, and shrimp that are now major parts of world fisheries. (The situation is even more obvious in the markets in developing countries, where tiny fish prevail; we in the West, snuggling in our duvets, still see lots of the high-value species because we can afford to pay for the few of them being caught.)

What is the significance of fishing down the food web? If it were just a case of changing our food preferences, this would actually be a positive development. All biological production depends ultimately on photosynthesis (except for a tiny fraction of primary production by chemotrophic bacteria that occur in such places as deep ocean vents), and, obeying the Second Law of Thermodynamics, there is loss of energy at every level of the food web. A given amount of sunlight can be used to produce many grams of plant material, which will support production of fewer grams of cow, and still fewer of farmer eating steak. By cropping a lower trophic level of organism (eating grain rather than cow, or smaller rather than larger fish species), we harvest a more plentiful supply of food resources. Unfortunately, however, the shift of fisheries to lower trophic levels has not been voluntary. It appears to have been forced due to the elimination of the higher-trophic-level species. Those large piscivores are not yet extinct (in most cases), but they have been made too rare to sustain fisheries, and in the majority of cases they are not recovering their former numbers.

That we are fishing down the food web is a clear indication of the pervasiveness of overfishing in the world's oceans. The danger is that we now are reaching the limit of fishable species, because we are already fishing organisms at an average trophic level of 3.1. Trophic level 3 in marine systems is organisms that feed on zooplankton. Our next step, if we continue this downward journey, will be to harvest krill and other tiny crustaceans. While krill do feed the giant baleen whales, grilled krill on toast will be a sad replacement for a tuna steak, quick broiled so it remains blue in the center, or many of the other quality fish products that have graced our tables.

Other Ecosystem Effects of Overfishing

So far, I have considered only direct effects of overfishing on the species being targeted. But given that fishing has routinely reduced the standing biomass of most fishery species by 80 to 90 percent, it should be no surprise that we have altered the structure of marine ecological communities. They are becoming simplified, as fewer species are present and fewer trophic connections exist. Jeremy Jackson of Scripps Institution of Oceanography in San Diego and the Smithsonian Tropical Research Institute in Panama together with several coworkers made this point cogently in 2001 in an article in Science, drawing attention to chronic long-term overfishing of marine systems and the resultant loss of larger, older, higher-trophic-level species. Jackson's focus was not on how this trend reduced our options for fishery products, but on how the losses were causing dramatic changes to the structure of the ecosystems being fished. In this article, Jackson broadened the meaning of fishing to include hunting for maritime mammals such as seals, the sea otter, and Steller's sea cow (extinct since 1768) that feed in marine environments. Using paleontological, anthropological, historical, and current ecological and fisheries management data, he documented a distressingly common and very long-term tendency to overfish larger, usually higher-trophic-level species and to cause pronounced changes to ecosystem structure as a consequence. His examples include the West Coast of North America, where the loss of Steller's sea cow and the sea otter led to a great increase in sea urchins, which reduced the capacity of kelp beds to recover from storm damage because grazing by urchins prevented the establishment of new juvenile kelp plants. As a result, the complex ecosystem of species of fish and invertebrates that depend upon the kelp to provide habitat structure and sometimes food is replaced by a simpler, less productive community occupying a largely bare, rocky habitat termed an urchin barren. He makes a similar argument for the Gulf of Maine, where overfishing of cod and other groundfish allowed urchin populations to explode. The large urchin populations led to the replacement of kelp forests by extensive urchin barrens. Paradoxically, in both of these cases, new fisheries for sea urchins (part of the fishing-down process) have allowed some recovery of kelp; however, the forests now lack the higher-trophic-level consumers that were formerly present.

Regarding coral reefs, Jackson suggests that overfishing has gone so far that in many locations major populations of herbivorous fishes have largely disappeared along with the larger piscivores. While the story is complex (see chapter 4), it appears that, at least in the Caribbean, overfishing, together with an outbreak of disease that virtually eliminated another major herbivore, the Diadema sea urchin, has been responsible for a shift from coral-dominated to algae-dominated benthic communities. The latter are notably less valuable for tourism, less diverse, and less productive of valued fishery species. Overfishing of turtles, manatees, and dugong may have made sea grass beds around the world much more susceptible to the diseases and pollution that are now prevalent causes of reduced abundance and ecological complexity, and overfishing of oysters appears to have been a primary cause of the eutrophication and consequent ecological simplification of the Chesapeake Bay. Jackson argues that overfishing has been widespread for a long time, that it predictably (thought not always) removes the larger, higher-trophic-level species, and that it causes substantial changes to the ecosystems being fished. In particular, it appears likely that in some if not all cases, overfishing changes ecosystems in ways that make them more vulnerable to other human or natural disturbances such as pollution, outbreaks of diseases or invading species, storms, and climate change. This synergism between overfishing and other forms of disturbance should be an issue of great concern because, as we shall see, our other kinds of impact on natural systems are also becoming more severe year by year. We need to recognize that by avoiding overfishing we also may be able to mitigate the effects of these other disturbances.

Overfishing also impacts species other than the ones being targeted. These impacts can be separated into bycatch issues and habitat effects, and they vary in importance depending on the fishing techniques a particular fishery uses. In 1994 the FAO reported on rates of bycatch in global fisheries during the 1980s and early 1990s. Quantities were enormous. The global bycatch was estimated to be 27 million metric tons, more than one-third of the marketed catch of 77 million metric tons. This "wasted catch" was widely recognized as undesirable, and there was widespread support to improve the situation. U.N. resolutions and the FAO's Code of Conduct for Responsible Fisheries all called for steps to reduce bycatch. An update by the FAO in 1998 reported global bycatch as 20 million metric tons, and its 2004 assessment reported bycatch at 7.3 million metric tons, or 8 percent of the global marketable catch. That's quite an improvement in ten years, and by 2006 the FAO did not bother to mention bycatch at all. As we will see, that does not mean the problem has disappeared, only that it has changed.

Bycatch is a mixture of uneconomic specimens caught unintentionally and not worth bringing to market. These consist of undersized members of the target species and individuals of undesired species. Because these organisms are not valued, they are discarded at sea, and rarely are any data collected concerning the amount or the species composition of the bycatch. Now, consider first the bycatch comprised of small specimens of the desired species, and consider a fishery being managed very close to or exceeding its MSY. The bycatch is unreported, and therefore is an excess catch of unknown extent that may not be fed into the equations used to monitor the fishery and the state of the population being harvested. Further, because these are small (therefore young) individuals, the bycatch is reducing the number of new juveniles entering the fishery in subsequent years. The result of a substantial bycatch of this type is that the population performs less well than expected under a given level of fishing, because the fishing (including bycatch) is substantially more intense than the level intended.

The bycatch that is comprised of unvalued species poses a different and more serious problem. This bycatch is an unmonitored fishery on a group of species that together provide the ecosystem that sustains the species of fishery interest. Assuming the different species making up this ecosystem are variably susceptible to being caught and variably able to sustain the level of fishing that is being imposed on them, some populations will barely be modified by the slight levels of bycatch, while other species may be severely overfished. As a consequence, over a period of time there will be definite changes in the relative abundances of the various species that make up the community. Now remember that the capacity of a fish population to grow in size depends upon a full suite of environmental factors that affect the individual's capacity to survive and reproduce-a supply of food is one of these factors. If some of the bycatch species are more important than others as prey for the commercially targeted species, and if these happen to be the ones that are overfished as bycatch, the ecosystem becomes less able to sustain populations of the economically valued fishes.

There is a lot of variation among fisheries in the extent of bycatch, based on differences in gear and in uses of the fish. Bycatch in artisanal fisheries is very low (less than 1 percent of catch) or nonexistent; people who fish to feed their families eat the bulk of what they catch or sell it to raise funds for other needs. Among commercial fisheries, trawling has a particularly bad record-both bottom trawling for a range of groundfish such as cod and midwater trawling for shrimp. A trawl catches everything that enters its mouth and is larger than the mesh-size of the cod end. On a good day, as the net fills, even organisms that might have passed through the mesh can become trapped because the mesh becomes plugged by the mass of fish, invertebrates, and algae that have already been caught. Shrimp trawlers typically discard about 62 percent of the catch as bycatch, while bottom trawlers discard 10 percent as bycatch. The reported bycatch in the long-line fishery for tuna and other open-ocean, migratory species is surprisingly high at 29 percent, but this number includes the carcasses of "finned" sharks. (Finning is an exceptionally wasteful practice in which the fins are cut off sharks for the Asian shark fin trade and the carcasses dumped back into the ocean, often still alive but unable to swim.) Bycatch also varies geographically, depending on the nature of fishery practice in different locations, from a high of 22 percent of catch, or 1.7 million metric tons, in the central Atlantic fishery (where long-lining predominates) to less than 7 percent in the Pacific and Indian Oceans. (All percentages are 2004 values.)

Reductions in bycatch have been achieved in two ways. The first has been through the refinement of gear and fishing practices to reduce the capture of unwanted fish. Most notably, the Gulf of Mexico shrimp trawl fishery has improved its (still disappointing) record by developing nets with exclusion devices of various types to prevent unwanted organisms, from small fish to large turtles or dolphins, being trapped even if they enter the mouth of the net. (It takes some skill to build a net that will sift out and retain the small creatures while rejecting the large creatures, so there are limits to what can be expected in modifying this gear.)

The second way in which bycatch has been reduced is the more important one and is paradoxically part of our fishing down the food web. Bycatch has been reduced by finding ways to make these species and small sizes marketable. Changes in regulations have eliminated minimum size limits in some fisheries so that younger fish, when caught, can still be marketed. Reduced availability of the former target species means that vessels have excess hold capacity and can afford to bring less valuable species ashore. And new ways of processing have been developed that make use of smaller specimens or different species.

For example, many fish that would formerly have been discarded as bycatch are now used in the manufacture of surimi-that amazing Japanese product that looks and tastes almost like crabmeat but does not put people with allergies to crustacea into anaphylactic shock. Madison Avenue has stepped forward and convinced the public that some other fish that would previously have been avoided are in fact highly desirable foods. One case in point is the monkfish-the name is applied to several species of the genus Lophius-a large, ugly, bottom-dwelling anglerfish that was routinely discarded as bycatch until the early 1980s. It was not considered attractive enough to be marketable. This fish has an enormous head and a much smaller body, but the flesh in the trunk and tail is delicious. Consumers rarely ever see a monkfish with its head on or learn that it is an anglerfish-marketers decreed that it should be brought to market already reduced to the tail section or a fillet, probably to disguise its appearance and avoid turning consumers away. It became a viable fishery in many regions including the United States, Europe, and Australia without most consumers in these locations knowing what it looked like. Of course, as in so many fishing tales, this one has mixed endings. Since the early 1980s, increased pressure on this slow-growing, deep-water fish has led to chronic overfishing of many populations, some fishery closures, and a few apparent recoveries under better management. So, the monkfish went from bycatch to mostly overfished in two short decades, but its use has indeed reduced bycatch.

Changes to marketing practice do not mitigate the damage caused to ecological communities by bycatch. Fisheries are still removing large numbers of young fish from populations and are removing individuals of many different species, frequently at unsustainable rates, because these are an incidental catch not being targeted (and management pays less attention to incidental species). Still, if we consume more of what we catch, perhaps our need to catch ever more fish will grow more slowly.

Some kinds of fishing also have profound effects on habitat. Again, bottom trawling is a particularly egregious example. If you think about it, trawling involves dragging a rather heavy net and a couple of heavy barn doors across the substratum in an attempt to catch those organisms that swim about just above it. To be effective, the trawl must hug the bottom so that fish can't escape underneath. As a consequence, trawling has substantial effects on the structure of the substratum, particularly when that structure is relatively delicate, made up of various sponges, bryozoans, oyster reefs, algae, and corals. Trawling rips these up while generally leveling any topography of the ocean floor. This is a little like clear-cutting a forest but using a bulldozer to do the clearing. (Actually, it may be quite a lot like clear-cutting because many of the structure-forming benthic organisms such as sponges can be quite slow-growing, long-lived creatures-five hundred years is possible for many sponges. These are removed by trawling, much as old-growth trees are removed from forests.)

Now if trawling occurs at a rate such that a trawl crosses a particular area only once every decade or so, the system is probably capable of recovering, and in any event there will be ample undisturbed area in the vicinity. But with overfishing, trawling can become so intense that the disturbance occurs repeatedly, and there is seldom time for the system to recover its former structure before it is trawled again. In 1998 Les Watling from the University of Maine and Elliott Norse of the Marine Conservation Biology Institute in Redland, Washington, examined catch and effort data from shrimp fisheries to reach an estimate for the total amount of trawling (of all types) taking place around the world. They found that trawls sweep over an area equal in size to all the world's continental shelves once every two years! Trawling is not uniformly distributed, however, and they noted that while there were shelf locations that had never been trawled, other locations may be trawled as many as four hundred times per year.

Some other forms of fishing also have undesirable effects on habitat. Chief among these are the use of dynamite and other explosives and the use of various chemicals, from household bleach to cyanide, to catch fish on coral reefs for the aquarium trade or food. The explosives or chemicals make collection in this structurally complex habitat much more effective. They also severely damage other components of the community, particularly the corals that provide the habitat on which the rest of a reef biota depends. Using dynamite and chemicals is universally condemned, and these methods are illegal in virtually every jurisdiction that has laws to manage use of coral reefs. That does not mean these methods are seldom used. Of the two, "blast fishing" with dynamite has the more serious environmental effects. A 1-kg beer bottle bomb produces a rubble crater 1-2 meters in radius and kills most of the coral in that area. While occasional damage on this scale is easily repaired by natural processes, it is the extensiveness of the practice that causes the problems. In Indonesia, blast fishing has been estimated to destroy 3.75 square meters of live coral cover per 100 square meters of reef per year-a rate substantially above the rate at which reef growth can regenerate the habitat. By contrast, cyanide fishing (and fishing using other chemicals) has more modest impacts on nontarget species. The damage is primarily the physical destruction of delicate coral growth, which occurs while extracting the catch and so is really incidental to the use of chemicals. Still, the habitat destruction can be substantial when the fishing effort is high. Since both forms of fishing are peculiar to coral reefs, I discuss them in more detail in chapter 4.

Collapse of Fishery Stocks: Do They Ever Recover?

Conventional thinking suggests that if we reduce the abundance of a fish species by overfishing it, reducing or suspending fishing will permit the population to recover. The history of overexploited fisheries does not support this expectation, however. There are now innumerable instances of fish species, such as the Atlantic cod, whose numbers have been greatly reduced and have not recovered, even though fishing has been abandoned, banned, suspended, or in other ways halted. That they tend not to recover should be a very clear message to us: Our simple notion of the natural world as one in which sizes of populations are carefully regulated by mechanisms that will tend to protect them from extinction is a flawed one. That attractive, dependable world is apparently not the world in which we and fishes live.

Why fish populations that have been severely overfished do not recover can be understood by reviewing the various aspects of overfishing enumerated in this chapter. Overfishing reduces the size of the fished population and can disrupt social structures vital to population integrity in the process. Overfishing severely depletes a population of its older individuals, dramatically reducing its capacity to weather periods of poor recruitment and its capacity to rapidly increase its numbers when conditions are favorable for recruitment. Overfishing usually also depletes many other species from the community of which the target species is a member. Some of these other species may play particularly important roles, as prey or in other ways, in facilitating the success of the target species. Frequently, overfishing leads to increases in abundance of those species that are less susceptible to being caught by the fishing gear in use, and these now more abundant species use many of the resources formerly available to the species that have now been depleted. Finally, overfishing can have substantial habitat impacts that may make the environment one that is no longer favorable for populations of the target species.

In addition to all of these factors, it is also extremely difficult for societies to reduce their fishing efforts until overfishing has become extreme. And it is equally difficult to refrain from starting to fish again before the fish population has had sufficient time to recover (assuming that the fishery is one in which some recovery of abundance does take place). Before turning to why we overfish and whether we can do anything about the sorry state of the world's fisheries, let me briefly squelch the idea that aquaculture will come to our rescue and the rescue of the world's coastal ecosystems.

The Limited Promise of Aquaculture

Aquaculture is an enormous and growing industry around the world. In practice and effect, it is very different from fishing, although many of the fish, crustaceans, and shellfish we consume today are aquaculture products, and it is often difficult to tell the difference. Extensive areas of freshwater ponds and lakes and coastal wetlands are employed to raise aquaculture species, and pen culture (also termed sea ranching) is extending aquaculture out across the continental shelves. A logical and commonly held view is that just as agriculture replaced hunting and gathering as a much more efficient way of acquiring terrestrial food products, aquaculture will eventually replace fishing of wild stocks. It is quite true that aquaculture has become important and will continue to grow in importance, and it is probably also true that our seafood diet will become predominantly based on aquaculture species over the next few years. Indeed, the only way of further increasing our global consumption of seafood is through increased aquaculture. But it would be unwise to anticipate that a shift to aquaculture will permit us to market ever-greater quantities of seafood while permitting natural marine systems to recover from the present state of overfishing.

Marine systems and terrestrial systems are very different in structure, and we enter them at very different ecological places when we seek to consume their species for food. Humans consume a broad range of plant products and a number of animal species, particularly herbivores, from the land. By contrast, plants from the ocean play a tiny role in human food products, even in Japan, where the use of algae as food has been taken farthest. Most oceanic plants, after all, are single-celled phytoplankton. Nor do we make much use of marine herbivores as food-most of these are minute zooplankton. Instead, we prefer to fish for top carnivores, the tuna, swordfish, grouper, cod, and so on that feed on smaller fish and are at levels 3.5 to 4.5 on the trophic web. There are of course some interesting exceptions to this rule. Among the marine herbivores we consume are abalone, conch, sea urchins, and parrot fishes. We also eat a number of suspension feeders (consuming phyto- and zooplankton and suspended organic matter), such as oysters, mussels, and certain sea cucumbers, and various detrital feeders, including many burrowing clams. Perhaps the most unusual herbivore we eat is the giant clam, which both suspension feeds and obtains nutrients from the symbiotic zooxanthellae (single-celled algae) that occupy the surface layers of its mantle.

Now, if you find you have not eaten very many of the animals on this list of exceptions, you need to eat more sushi and to try some of the more unusual dishes in other types of Asian restaurants. The nature of the list justifies my claim that feeding on fishes other than top carnivores is an unusual event-although it becomes ever more common as we fish down the food web.

If one plans to farm on land, it's possible to focus attention on specific species of plants, providing them with sunlight, water, and nutrients, or to focus on herbivores, supplying them with plant food. Farming the ocean is a different matter. There are no marine plants with the potential to become human staples in the way that grain crops have become, although certain suspension feeders such as oysters can be farmed in a manner analogous to that of terrestrial plants by providing them sites with a steady supply of plankton-filled water. The animals that are of sufficient economic value to be worth raising under aquaculture nearly all require foods derived largely from animal tissues. These animal-derived foods are obtained primarily by fishing wild stocks of small fish. So, far from ameliorating the need to fish, the rise of aquaculture is generating a new market for fishery products-products that used to be bycatch. Aquaculture is both energetically and economically expensive because of the food requirements of the species being raised, and it has proved difficult to develop aquaculture species that can be raised for a cost that is less than the cost of catching them in the wild. Obviously, this difficulty will be eased as all seafood becomes more expensive due to its reduced availability in the wild, and we can anticipate ever more aquaculture products on the supermarket shelves.

This increase in aquaculture, however, is going to come at a real cost. While some progress is being made in developing plant-derived foods, we will continue to need to fish wild populations to obtain much of the animal protein for the aquaculture enterprise. In addition, the enormous densities of animals living in aquaculture pens or ponds create local aquatic pollution due to their own production of waste and to the usual practice of providing surplus feed to maximize rates of growth. Then there is the problem of the introduction to the coastal marine environment of antibiotics, used to maintain the health of the crowded fish. And through inevitable escapes or releases, individuals with novel genetic makeup have been introduced to native populations-strains that have been selected for fast growth under crowded conditions, not necessarily for traits that will be adaptive in the wild. Each of these problems is real and growing as the use of aquaculture grows, but the biggest may be the continued need for animal food.

In its 2008 report on world fisheries and aquaculture, the FAO reported that global aquaculture production reached 51.7 million metric tons in 2006, having grown 8.7 percent per year since the early 1970s. The FAO stated that there was going to be a need for increased fish production, that it was unlikely that capture fisheries could provide much increase, and that it was going to be necessary for aquaculture to make up the difference. By evaluating national plans for increases in aquaculture production through the next thirty years, the FAO suggested that there was reason for cautious optimism that the world's need for fishery products in 2030 could be met by growth in aquaculture. It noted, however, that among other things the availability of fishmeal (for feed) was a "much-debated issue." This cautious report is about as close as the FAO has ever gotten to suggesting we will not be able to achieve stated goals for fishery production, and nothing was said concerning the mix of species that we may be eating in 2030. In 2006 world aquaculture was using 3.06 million metric tons of fishmeal and 0.78 million metric tons of fish oil as feed-56 percent and 87 percent of total production, respectively. While there have been impressive developments in aquaculture feeds (so that salmon diets, for example, now are only 30 percent fishmeal), I suspect we will be fishing the oceans for krill after all-to feed to aquaculture species. And after the krill, what then?

Why Do We Overfish? How Can We Stop?

We are officially Homo sapiens, the wise humans. If overfishing has been going on so long, if we have multiple stories of species that have been fished nearly to extinction and failed to recover, and if we have extensive efforts to manage fisheries, to monitor them, and to investigate what is going wrong, why are we still trying to catch more fish than are available to be caught? Surely it is in our collective best interest to do a much better job of managing these incredibly valuable resources. Are we less wise than we believe?

On a positive note, we are doing a much better job than we used to do. When the cod fishery began in the sixteenth century, it was an open access, unregulated fishery. Anyone who wanted to enter it and had a vessel and crew was free to do so. This has been the typical state of fisheries when they first start and has been the usual state of fishing enterprises since our Pleistocene ancestors speared fish and collected shellfish on their shores. Such a fishery is far from being a logical interaction between fish and fisherman acting together to achieve a long-term stable output of product. It is a scramble by a group of competing individual fishermen, each seeking to maximize his or her catch and to take fish as rapidly as possible, reasoning, "If I do not catch the fish, someone else will take them."

Garrett Hardin coined the phrase tragedy of the commons to describe the problem inherent in this type of interaction. Before they are caught, fish are a commons in that they belong to nobody and are available for all to make use of, much as the commons of the English village was a pastoral area on which any farmer was free to graze his cattle. The tragedy is that in such circumstances, it is in nobody's best interest to moderate his or her behavior to ensure that the commons will be fit for grazing next week or next year. If I don't catch those fish, somebody else will. Marine fishery resources were treated as a commons for many years, partly because most of the world's oceans were outside the territorial waters of any nation and laws governing the use of the oceans did not exist. Now that the Law of the Sea has established the right to an exclusive economic zone extending 200 nautical miles out from shore, most countries are claiming national ownership of fishery resources within this zone, and many fisheries are managed on a limited entry basis. This means that the yet-to-be-caught fish are collectively owned by the fishery and that the fishery is of a fixed size. A new fisherman can enter the fishery only by purchasing a license from someone seeking to leave it. Indeed, it is now widely recognized that limited entry is an essential part of the management of any fishery, if that fishery has any real chance of being sustainable.

In other words, we now have established law governing the use of marine resources, and there are mechanisms that can be put in place to avoid the tragedy of the commons. However, the great majority of fisheries are still not being managed effectively. Some are unmanaged, some are not managed effectively, and some are managed in ways that permit unlimited entry and the resulting growth in effort that results. Inadequate management is widespread in developing countries that either lack the resources to provide effective management and enforcement or could muster the needed resources but lack the political will to do so. Developed countries also have poorly managed fisheries. They occur under three types of circumstances: the fishery is a new one, tapping a previously unfished resource; it could be managed more effectively but there is a lack of political will to do so; or it is in international waters or on populations that straddle boundaries of different nations' territorial waters.

It takes time for a management agency to recognize the need to develop management policies for a newly targeted species. Personnel must be deployed to work on this new species, and data must be collected to determine its basic demographic characteristics. Laws governing the fishery must be introduced and implemented. Sometimes management agencies are simply less nimble than they should be. Unfortunately, in recent years, new fishable stocks have frequently been discovered in deeper, colder waters. Such fish tend to be very slow growing and long-lived. The initially bountiful catches are comprised mainly of old individuals, and these animals are removed in a short time, because the population lacks the capacity to rapidly replace itself. The result is that an initially promising fishery quickly shows declining catches of much smaller, younger animals and can become unprofitable almost before the management agency has begun to gather the information needed for sustainable management. Fisheries for monkfish and for orange roughy provide many examples, although some of these are now managed sustainably. (A roughly parallel problem can develop even in an established fishery when new technology leads to rapid changes in effective effort. Unless the management agency is alert to the innovations, the fishery can overfish even while obeying the regulations to the letter.)

Lack of political will arises because fishery management is a governmental activity that exists partly to sustain a fishery resource but primarily to ensure the continuation of an economically valuable industry that creates jobs and wealth. When fishermen have extensive investments in the vessels and fishing gear and when fishing is a primary source of employment and income in a region, governments have a way of pressuring management agencies to permit fishing effort to remain as it was or to grow, even if the data say that the population is being overfished and that effort must be reduced. Sometimes the pressure is quite indirect. Canada's particularly favorable regulations governing unemployment insurance for "seasonal workers" such as fishermen seem, on the surface, socially responsible, but they have had the effect of keeping people in the fishing industry long after economics would suggest they seek other employment. A large population that wants to earn a living by fishing and that is able to hang in through lean years because the unemployment benefits are pretty good remains a constant spur to the management agencies to provide good news in the form of renewed opportunities to fish.

In many developing countries with large coastal populations dependent on artisanal fishing for their own food as well as their livelihood, overfishing gets ignored because there are no obvious alternative sources of employment or food. How do you tell an artisanal fisherman to stop fishing when he has no other way of feeding his family? As often as not, these countries have so little invested in fisheries management that the data to confirm that the resources are being overfished are simply not available, and if they were available, lack of will and of viable alternatives to fishing would ensure that little attention would be paid to them.

Finally, there is the issue of straddling stocks (species whose distributions and fishery cross two or more different jurisdictions) and of stocks that are fished primarily in international waters. In the case of straddling stocks, differing management regulations may not be complementary or may not be equivalently enforced, and damage to the stock caused by overfishing within one jurisdiction is transferred to all regions of the fishery. In the case of open-ocean fisheries, management policies depend far more on consensus among fishermen from different nations than on enforceable regulations, and all the problems of open access and limited management effectiveness remain. That it is still possible to buy whale meat legally in Japan and that the finning of sharks continues as the primary way of harvest for the shark fin trade are testaments to the lack of effective management of high seas fisheries in the twenty-first century. Both cases are clear examples of unsustainable fishing practices, widely condemned except by the people who make money engaging in them.

So, what does the future hold for fishing? There is reason for limited optimism because of a number of improvements in both the science and the sociology of how we manage fisheries. Scientists understand that the oceans cannot provide ever-increasing quantities of fish, and we know that much of the demographic theory that led to the concept of managing for MSY was overly simple and was, in any event, asking for a much finer control of effort than would ever be possible in the real world. There is widespread acceptance among fishermen, managers, governments, and the general public that fisheries management requires adoption of the precautionary principle-that we should fish cautiously, erring on the side of taking less than the resource can sustain-to maximize the chance that fishery resources will remain available to future generations. The creation of various types of marine protected areas, particularly so-called fishery reserves, as a way of both conserving species and sustaining fisheries has been widely adopted as a useful additional tool for managing fisheries in coastal waters (discussed in chapter 4). There is also widespread appreciation that fishery species are embedded in marine ecosystems and cannot be extracted without attention to the impacts on the sustaining ecosystems. With this deeper understanding, we are in a much better position to devise effective ways of managing fisheries sustainably.

There is also much better appreciation of the linkages among government, society, management agency, and fishermen and of the complexities (contradictions, perhaps?) involved in trying to manage in a way that both sustains resources and ensures economic viability for the industry. (Coincidentally, our better understanding of fisheries management also serves to inform our management of other types of resources, such as forest products.) Considerable success has been achieved in some jurisdictions in efforts to make fishery management a cooperative, shared responsibility between the industry and the management agency, and the FAO Code of Conduct for Responsible Fisheries is being adopted widely. Best of all, perhaps, fisheries that are being managed sustainably are being marketed as "green," and the public that ultimately buys fish is beginning to differentiate and buy from the responsible fisheries instead of from those that are being managed less well.

Despite these reasons for optimism, however, I remain concerned. The human population is still growing, and coastal populations are growing more quickly than those inland. Fisheries provide 15.3 percent of the animal protein we consume, and the need for that protein is not going to disappear if fishery yield continues to decline. Too many fisheries are overfished, and there are few available stocks that have not yet been targeted. And while aquaculture might manage to fill our increasing needs, it's more likely that it will not. My fear is that, in the final analysis, it is going to become more important to put food into people's mouths today than to ensure that fishery resources remain available for use in the next decade or next century. With the loss of fishery resources will come a need for more food production from agriculture, more use of limited water supplies for farming or pond aquaculture, and the resulting stresses on terrestrial ecology that these changes could bring. In other words, the possible loss of fisheries is not a local problem or a marine problem. It has ramifications that will ripple across other parts of the world as we struggle to grow the food that fisheries formerly provided for us. I hope I am wrong, but I fear that the bountiful and nutritious food we have obtained from the oceans throughout our history is no longer going to be available to us, and that in the process of exploiting it to the very end we are going to irrevocably change the structure of marine ecosystems. Wild fish are going to become as rare as wild game or wild forests (never mind the fish that will become extinct). Ultimately, even if we can get that 15.3 percent of protein from some other source, we become poorer because we occupy an impoverished planet.