Fisheries play an important role in global food security, employment and income (Smith et al., 2010; Ye et al., 2013). Meanwhile, it is well known that fisheries tend to have tremendous impacts on marine ecosystems, possibly reducing biodiversity and impairing ecosystem functioning and even resulting in changes in their states, exemplified by overfished stocks and economic loss of fishing industry (Butchart et al., 2010; Srinivasan et al., 2012; Sumaila et al., 2012). The period following the Second World War saw significant changes of fisheries; world’s marine capture fisheries expanded continuously to a production peak of 86.4 million tonnes in 1996 but have since exhibited a general declining trend at an approximate rate of 0.4 million tonnes per year (FAO, 2014). In contrast, demand for fish is expected to grow given the increasing human population, rising incomes and growing urbanization (FAO, 2014). Fish consumption per capita in the world increased from an average of 9.9 kg in the 1960s to 17.0 kg in the 2000s, fishing pressures on fish resources continue to mount and the number of fishers has kept increasing over time (FAO, 2014). Ongoing declines in production of the world’s capture fisheries may have serious ecological and socioeconomic consequences (Balmford et al., 2002; Mora et al., 2009).

Indeed, the poor state of global fisheries might have long been masked by improved technology and large geographic expansion of fishing grounds (Pauly, 2008). The growth in the global marine fisheries was driven through a sequential expansion, fisheries exploitation has spread from coastal waters off North Atlantic and Pacific to waters in the Southern Hemisphere and into the high seas (Swartz et al., 2010). With the decline of fish stocks in shallow coastal waters, increasing demand, and introduction of new technology, fisheries were evidently expanding offshore and into deeper waters. However, the era of great expansion seems to come to an end now (Morato et al., 2006). According to recent assessments, up to one-third of the world’s fish stocks are now overexploited (Branch et al., 2011; FAO, 2014). Overcapacity in fishing fleets and overexploitation of fish resources are two major factors result in the unsatisfactory stock status of world fisheries (Ye et al., 2013). However, global fishing intensity continues to mount, and the number of fishers has kept increasing over time (Ye et al., 2013). To put fisheries on a path towards sustainability, international efforts have sought to improve management for avoiding overexploitation (Worm et al., 2009; Ye et al., 2013).

An understanding of the dynamics of marine ecosystems induced by fishing activities is important for formulating marine policy. Much effort has been spent to assess the status of many exploited stocks and fisheries globally (Worm et al., 2006; Branch et al., 2011; FAO, 2014). Nevertheless, few studies have been done to quantify the foregone catch due to overfishing of fish stocks. Srinivasan et al., (2008) estimated the foregone catch duo to overexploitation of fish stocks for low, middle and high income groups from 1961 to 2000 using catch data and maximum sustainable yield (MSY) that estimated by devising a conservative approach. The study provides new insight on understanding the global status of fisheries. However, species composition varies from area to area around the world, and exploitation history for different categories of species are significantly different (Caddy et al., 1998; Caddy and Garibaldi, 2000). For successful fisheries management, it will be necessary to further examine the exploitation state for different components of community structure in detail. Using the conservative approach described in Srinivasan et al. (2008), we examined the foregone catch trends at the level of FAO fishing areas. Contrary to early studies on the global fisheries losses due to overfishing, in which all the fisheries stocks were considered as a whole, in this study we further examined potential losses due to overfishing for different categories of resources in the marine ecosystems, which can provide more accurate estimates.

The study, which covers the period from 1950 to 2010, aims to estimate the potential catch and revenue that have been lost as a result of overexploitation at the FAO fishing area level based on the MSY estimate methodology described in Srinivasan et al. (2008). To minimize the attribution of natural fluctuations in fish populations to overfishing, we classified the fish stock as overfished only if its time-smoothed landings fell below 50% of its maximum value for at least 10 years in a row or 15 years in total following the year of maximum recorded catch in this study (stocks, defined here as species-FAO area combinations) (Kleisner et al., 2013; Srinivasan et al., 2008, 2010, 2012). We included that stock in our analyses if cumulative catch from 1950 to 2010 exceeded 1 000 tonnes and smoothed the data using three-year running average. In accordance with stock status plots (SSPs) (Kleisner et al., 2013), we excluded taxonomic groupings higher than family and pooled groups from this analysis. It should be noted that the concept of “overfished” in this study refer to species that have or had overfished during the study period, so we also included stocks which were dropped to “overexploited” and “collapsed” status previously but recovered to “rebuilding” and “exploited” status nowadays.

The estimation of corresponding maximum sustainable yield for each species already identified as overfished was according to the method developed from simple, conservative guidelines based on species’ lifespans (t_max), as well as the maximum catch (C_max) over the period and selected MSY and catch estimates from the Northeast Fisheries Science Center (NEFSC) by Srinivasan et al. (2008). In this study, we used the medium value of the lower and upper assignments of MSY by Srinivasan et al. (2008) to estimate the foregone catch. Lifespans of fish species were obtained from Fishbase (www.fishbase.org), and lifespans for genus and family were taken as the mean lifespan of constituent species. For invertebrates taxa MSY levels, the estimates were an average of all invertebrates from Srinivasan et al. (2008), at about 35% of maximum catch. In order to provide more accurate estimates, we further revised MSY assignments for fish species that had MSY estimates calculated from NEFSC data and for invertebrates that have accurate lifespans according to Srinivasan et al. (2008).

Potential total catch losses (L, tonnes) in each FAO fishing area in year t were tallied for all overfished species-fishing area pair i for year t subsequent to the year of maximum catch in which the estimated MSY was greater than the recorded catch (C_it) (Srinivasan et al., 2010). It can be defined by the equation:

Annual catch data were taken from the Sea Around Us Project (SAUP) database (www.seaaroundus.org). In order to more accurately measure fishing impact on different components of community structure in an ecosystem, all the exploited fish species/groups in each fishing area were grouped into four categories: demersal marine fishes (including demersal fish, benthic fishes, sharks and rays, and flatfishes), small-medium sized pelagics, oceanic and deep-water resources (consist of large pelagics and deep-water fishes), and invertebrates.

Price data were from the SAUP ex-vessel price database. In some cases, more than one price time series for a taxon is available; for example, when there are directed fisheries for a taxon in multiple regions. To maximize information content, we averaged across all available price records discounted to 2005 $US over time and space to construct the price index (Sethi et al., 2010). Using catch loss calculated by the above equation and price information obtained from the SAUP database, we estimated potential revenues lost due to overfishing in each FAO area. This study not only maps catch losses at fishing area-level that have not been analyzed previously, but also explores the foregone catch for different categories of resources in each fishing areas.

Overexploitation has long been recognized as a leading socioeconomic problem (The World Bank and FAO, 2009). Global landings are an indicator of benefits that human society can obtain from fisheries, and declines in landings mean loss of economic benefits (Ye et al., 2013). This study provides another aspect for understanding the exploitation history and status of global marine fisheries. This study confirms, through evaluating trends in potential catch lost and revenue lost for fishing areas in the Atlantic, Pacific and Indian Oceans, that global marine fisheries are currently underperforming due to overfishing. Based on the relatively conservative approach by Srinivasan et al. (2008, 2010, 2012), 35% of stocks in global oceans were deemed to be overfished in this study. The proportion of fished stocks that were considered overfished in this study also includes stocks that were overexploited previously but now recovered to be classified as “rebuilding” and “exploited”. Thus, this proportional value is more conservative than a recent assessment by Branch et al. (2011) in which 28%–33% of stocks were considered overexploited at the time of the study. Global catch losses caused by overfishing amounted to 322.8 million tonnes from 1950 to 2010, resulting in a corresponding value loss of US$ 298.9 billion (constant 2005 US$).

Historical data from marine ecosystems clearly suggest that intensive exploitation of fish communities often leads to substantial reductions in the abundance of target species and significantly reduce biodiversity in the global oceans (Pauly et al., 1998; Myers and Worm, 2003; Butchart et al., 2010). By the 2000s, fishery landings were three times annually more than what they were in the 1950s, global marine fisheries spread from the North Atlantic to the waters in the Southern Hemisphere (Swartz et al., 2010). Regarding the status of exploitation according to the catch-based method, up to half of fish stocks were deemed overfished in the Northwest Atlantic, Northeast Atlantic and Northwest Pacific Oceans (Table 1). Not unexpectedly, the above three areas ranked the 3rd, 5th, and 4th in the overall potential losses. Catch losses in the Northwest Atlantic Ocean were mostly from the collapse of demersal fishes such as Atlantic cod Gadus morhua, American plaice Hippoglossoides platessoides and redfishes Sebastes spp.. Marine ecosystem used to be dominated by demersal species in the 1970s and had changed to the one dominated by invertebrate and pelagic fish species in the 2000s, implying a regime shift (FAO, 2011). In addition, the increasing catch losses of invertebrates could be viewed as a warning to managers that management actions should be taken to prevent the collapse of these species (Foley, 2013). For the Northeast Atlantic Ocean, main target species have shifted from traditional species, such as Atlantic cod and haddock Melanogrammus aeglefinus, to formerly lower-valued species such as sandeels Ammodytes spp. and blue whiting Micromesistius poutassou. The sharply decreased catch of capelin Mallotus villosus together with the collapse of demersal species such as Norway pout Trisopterus esmarkii and Sebastes spp. led to a substantial increase in catch losses in the Northeast Atlantic in the 2000s (FAO, 2011). Despite recent developments in management measures, excessive fishing capacity is still one of the major issues in the Northwest Pacific Ocean, and CPUE has decreased rapidly since the late 1990s (Watson et al., 2013). Overall landings peaked in the late 1980s and have been declining ever since, making overfishing losses in landings increase greatly since the 1990s. Many important demersal fishes such as large yellow croaker Larimichthys croceus were overfished (FAO, 2011). In addition, Pacific sardine Sardinops sagax, which was once one of the largest single-species fisheries in the world, is now considered overfished, and the catch of S. sagax fishery has remained at an extremely low level since the mid-1990s, resulting in catch losses in the Northwest Pacific being sharply increased since the 1990s (FAO, 2011).

Catch-based method of stock classification for the levels of exploitation has been extensively used to assess the status of fisheries globally (Kleisner et al., 2013; Tsikliras et al., 2013, 2015). However, previous studies often emphasize the proportion of overexploited stocks, but fail to investigate the foregone catch behind it. When the catch-based method was applied in the 14 FAO areas, we found that the percentage of overfished stocks were all within the range 40%–50% in the Southeast Atlantic, Northeast Pacific, Eastern Central Pacific, and Southeast Pacific Ocean, but in 2010, these stocks lost 138% of the actual catch in Southeast Atlantic Ocean, while 15%, 10%, 29% for the latter three areas respectively. The small pelagic fisheries account for the highest proportion of the landings in the Southeast Atlantic Ocean, and the collapse of Southern African pilchard Sardinops ocellatus resulted in an increased catch loss in the 1980s, but its biological conditions had improved by a strict control of landings imposed by Namibia and Angola, causing a reverse in catch losses in the 2000s (FAO, 2011). Fisheries in the Northeast Pacific Ocean underwent a series of regulatory and market-driven reforms to reduce fishing effort to sustainable levels, which might contribute to a generally stable catch in most fisheries (FAO, 2014). The significantly increased overfishing loss in landings in the 1980s was resulted from the overexploitation of Sebastes spp.. In addition, the resources were still not recovered and required further rebuilding (FAO, 2011). Catch in the Eastern Central Pacific mostly consists of small and medium pelagic species. High fishing intensity and adverse environmental conditions might contribute to the collapse of Californian anchovy Engraulis mordax fishery since the early 1980s, resulting in rapid increases in catch loss from the 1980s to the 1990s in the Eastern Central Pacific, but the catch loss tended to be leveled off in the 2000s (FAO, 2011). According to FAO (2011), the Eastern Central Pacific had the highest proportion of non-fully exploited stocks, about 38% in 2009, which provides possible room for an increase in catch. Similarly, catch losses in the Southeast Pacific mainly resulted from the overfishing of small pelagic species. Intensive fishing effort together with natural environmental fluctuation played a significant role in large fluctuations of decadal catch losses. The collapse of the world’s largest single species fishery, anchoveta, placed Southeast Pacific the first in catch losses in the 1970s and the 1980s (Pauly et al., 2002). The sharply increased loss in catch in the 2000s was attributed to the collapse of South American pilchard Sardinops sagax, which was caused by the heavy fishing in the 1980s and the environmentally driven long-term fluctuations in stock abundance (Schwartzlose et al., 1999).

Despite the Western Central Atlantic, Eastern Central Atlantic, Southwest Atlantic, and Southwest Pacific Oceans all had about 35% of fish stock overfished, different categories of resources suffer from different degrees of overfishing losses. Catch in the Eastern Central Atlantic was dominated by small pelagic fishes, and fish stocks did not suffer from severe overfishing based on our analysis. In general, pelagic species were considered fully exploited (FAO, 2011). In contrast, the situation was more severe for the valuable invertebrate species, with a larger percentage of catch losses compared to the small pelagic species. Total landings in the Western Central Atlantic were in a state of downward trend since 1984, which was mainly caused by the diminished catches of Gulf menhaden Brevoortia patronus. However, its stock abundance was estimated to be between its target and limit reference points and, thus, not considered to be overfished nor subject to overfishing (Vaughan et al., 2007). Although the Western Central Atlantic did not seriously suffer from overfishing based on the method in this study, this area has a high incidence of stocks for which the state of exploitation is reported as uncertain. With an increased fishing pressure in this area, efforts should be concentrated to improve the quality of data and information to improve stock assessment and reduce the uncertainty. For the Southwest Atlantic and Southwest Pacific Oceans, peak landings were achieved in the 1990s. Fishery resources in the Southwest Atlantic were in a state of accelerating development since the 1980s, demersal species and squids were major contributors to the catches from this region. Although heavy catch loss did not showed in the Southwest Atlantic Ocean, excessive fishing pressure maintained main target small pelagic species such as Brazilian sardinella Sardinella brasiliensis under a state of overexploitation, and there was no sign of stock recovery (FAO, 2011). In contrast, the Southwest Pacific only included two countries, Australia and New Zealand, and they are often referred to having good practices in fisheries management (Pitcher et al., 2009). Catch loss in this region was mainly from the overfished invertebrates, however, rebuilding plans are in place in many fisheries to allow them to rebuild to target levels (FAO, 2011).

Although fishery resources in the Western Central Pacific and Indian Oceans are presently under stress, continuously upward trend of fisheries landings makes catch losses relatively low in these regions. We should note that many countries in these areas do not have well-developed management systems in place, and the capacity of data collection and stock assessment are generally poor in comparison with other regions (Mora et al., 2009). Recent studies also found that decrease in CPUE was consistent in both the Eastern and Western Indian Oceans, and monitoring programs are urgently needed (Watson et al., 2013).

Because of lack of information on the economic health of the world’s fisheries, highlighting economic objectives and the current level of global economic revenue loss in fisheries has never been so important (The World Bank and FAO, 2009; Srinivasan et al., 2010, 2012). By exploring trend of global catch losses at the level of FAO fishing areas over a critical period in the history of fishing, it is clear that unsustainable fishing caused substantial potential losses worldwide, especially in the northern hemisphere. Estimated potential losses due to overfishing for different groups of resources showed that the low-value but abundant small-medium pelagics made the largest contribution to the global catch loss, and the sustainable fishery management practices together with favorable environmental conditions can even reverse the losses to overfishing.

Although overexploitation of global fisheries cannot be attributed solely to fishing, other factors such as pollution and climate change may also play a role (Pitcher and Cheung, 2013). However, fishing is considered the greatest single cause of such depletion (The World Bank and FAO, 2009). As human populations continue to grow, the future benefits that fishery resources can provide depend largely on how well they are rebuilt and managed. Awareness and understanding of potential and actual economic benefits for avoiding overexploitation can provide incentives to take the necessary actions to sustainably manage fish stocks.

Data were obtained from the Sea Around Us Project and Fishbase for which we are grateful.