Regime Shifts DataBase

Large persistent changes in ecosystem services

High fish biomass – pre-1990s

This regime was dominated by sardines up until the late 1960's. The stock started to decline due to overfishing in the late 1960's and by 1975 the stock had collapsed (Cury and Shannon 2004). The depletion of the sardine stock gave space for the stocks of other pelagic fish to grow, such as the anchovy (Engraulis capensis), horse mackerel (Trachurus trachurus capensis), cape hake (Merluccius capensis and Merluccius paradoxus) and bearded goby (Sufflogobius bibarbatus) (Hutchings et al. 2009; de Young et al. 2004; Cury & Shannon 2004). After the collapse of the sardines, fishing pressure shifted to other pelagic fish. Fluctuations in these stocks occurred throughout the 1980s (Roux and Shannon 2004). The Cape fur seal (Arctocephalus pusillus pusillus) and seabirds such as Cape Gannet (Morus capensis), Cape cormorant (Phalacocorax capensis) and the African penguin (Spheniscus demersus) are the top predators of pelagic fish; therefore populations of these predators fluctuated in tandem with the pelagic stocks (Wickens et al.1992; Boyer & Hampton 2001).

Low fish biomass – from the 1990s

The regime shifted in the 1990's and is characterized by depleted stocks of sardine and other pelagic fish (Hutchings et al. 2009).  There has been a general increase in sea surface temperature (SST) and longer hypoxic events (Monteiro & van der Plas 2006; Cury & Shannon, 2004; Pörtner & Langenbuch 2005). These conditions combined with low fish biomass have created a beneficial niche for jellyfish (Chrysaora hysoscella and Aequorea aequorea) (Cury & Shannon 2004) to increase in abundance (Kreiner et al. 2011). A recent study estimated the biomass of jellyfish in this region to be 2.4 times the amount of three of the most important commercial fish species combined (Roux et al. 2013). Seabirds and seals declined in population due to the depletion of sardine and anchovy stocks (Boyer & Hampton 2001). There is evidence that some of these populations are recovering and have stabilized, although the state remains one of overall low fish biomass, in particular low sardine biomass (Boyer & Hampton 2001).

The combined effects of overfishing and a series of large-scale environmental perturbations caused the system to shift from a high to low biomass regime (Boyer & Hampton, 2001; Cury & Shannon, 2004). The Benguela region has been a favored fishing ground for nearly a century (due to its fertile upwelling), although it wasn't until the 1960's that resource exploitation was intensified, with the arrival of large commercial fishing fleets (Boyer & Hampton 2001). Sardines were targeted as the preferred species and intensively harvested - primarily for overseas markets (Boyer & Hampton, 2001; Cury & Shannon, 2004). The pressure on this stock effected the foodweb dynamics (Cury & Shannon, 2004). Sardine stocks declined in the 1970s after which fishing was aimed primarily at anchovy in an attempt to allow sardine stocks to recover. This, however, led to a collapse of both stocks and, in the 1980s, the system became dominated by horse mackerel, bearded goby and jellyfish (de Young et al. 2004, Cury & Shannon 2004). The top-down pressure of intense fishing, in particular on sardines, likely paved the way for a regime shift to take place by reducing ecosystem resilience to withstand perturbations (Cury & Shannon 2004).

Several environmental anomalies of the 1990s acted as important drivers of this shift. These periodic bottom-up pressures are natural to the system although they were of particularly large magnitude during these years (Cury & Shannon 2004, Boyer & Hampton 2001). In 1993/1994 a low-oxygen water event from the Angolan current caused unusually extensive hypoxia in Namibian waters (Boyer & Hampton 2001; Cury & Shannon 2004). This was followed by a Benguela Niño event in 1995 (an inter-decadal climatic event) where warm, nutrient-poor water entered the system (Gammelsrød et al. as cited in Heymans et al. 2004; Cury & Shannon, 2004). The combination of these events led to poor recruitment conditions and high pelagic fish mortality resulting in a decline of most stocks (Heymans et al. 2004; Cury & Shannon, 2004; Boyer & Hampton, 2001). A general trend of warming SST in the Northern Benguela has also been identified as a potential indirect cause behind this shift although there remains some uncertainty around this, as there does around the potential contribution of other factors attributed to climate change (Bakun et al. 2010; Cury & Shannon, 2004; Hutchings et al. 2009).

Like most upwelling systems, the Northern Benguela is dominated by a small number of species that play an important role in maintaining ecosystem structure and function (Cury & Shannon 2004). Here, sardines are the dominant species structuring the ecosystem by performing a 'wasp-waist' control on species both above and below them in the foodweb (Cury & Shannon 2004). Nutrient rich, cooler waters of the upwelling provides favorable conditions for phytoplankton growth, which is controlled by sardine populations and allows for an enriched trophic web supporting productive fisheries (de Young et al. 2004; Bakun et al. 2009).  The system is vulnerable to periodic environmental variability (such as Benguela niño events) with some years allowing better recruitment than others (Boyer & Hampton 2001).

Consistent fishing, particularly on sardines, creates a top-down pressure on this system reducing its resilience to environmental variability, to the point where a series of bottom-up pulses (93/94 and 95 events) pushes the system into a new dynamic regime of depleted fish biomass (Cury & Shannon, 2004; Boyer et al. 2001; de Young et al. 2001). A key threshold, although difficult to determine in any detail, is possibly crossed when sardine stocks get so low that they were no longer able to maintain their populations (Allée effect), with repercussions throughout the whole foodweb. Fishing pressure continues during the poor recruitment years following the environmental perturbations most likely enhancing sardine collapse (Hutchings et al. 2009; Boyer et al. 2001.)

The new state of depleted fish biomass is reinforced by continued fishing pressure. Decreased fish biomass allows jellyfish to consume a greater proportion of plankton, increasing the number of jellyfish (Bakun et al. 2009). This has creates a reinforcing feedback loop whereby fish stock recovery is impeded by jellyfish competition for food as well as a possible jellyfish predation on certain fish larvae (Cury & Shannon 2004). It has been suggested that the increase of phytoplankton in the system due to less predation creates hypoxic conditions that hamper fish spawning and recruitment (Boyer et al. 2001 as cited by Cury & Shannon, 2004). These feedback mechanisms appear to lock the system in this new state, potentially difficult to reverse (Cury & Shannon 2004). There is uncertainty about the intensity and duration of low oxygen events and how these may contribute to keeping the system in this state (Hutchings et al. 2009). Warming sea temperatures may also contribute to the continuation of this state which is favorable for jellyfish that cope better in these conditions (Richardson et al. 2002). It is unknown how the apparent decline in upwelling intensity might affect this social-ecological system as well as whether or not this is an effect of anthropogenic climate change (Bakun et al. 2009).

Provisioning services decline after the shift from high biomass to low. Fish catch are plummeted, as exemplified by sardines, decreasing from 700,000 tons caught at the height of the industry to only 2,000 in 1996 (Boyer & Hampton 2001). Seal populations also decline (Roux 1998 as cited by Cury & Shannon 2004), possibly impacting the sealing industry. Jellyfish has direct negative economic impacts on the fishing industry, for example, spoiling fish catches and busting trawl nets (Lynam et al. 2006; Roux et al. 2013). The regulating service of water purification is negatively impacted via disruption of vital ecosystem processes necessary for maintaining a healthy water state. The cultural service of recreational fishing is reduced, as low fish biomass lead to fewer catches by anglers (Kirchner 1998 as cited in Boyer & Hampton 2001).

Loss of these services has unequal negative consequences on all the user groups of the resource. As a vast majority of Namibian marine catch is exported, therefore the food security of international consumers decreases (Sowman & Cardoso 2010). However, this impact is masked by the international market (Berkes et al. 2006). Additionally, local fishers and international investors lose income from decreased fish catch (Sowman & Cardoso 2010). Recreational fishers and few subsistence fishers were impacted by the shift, via decreasing access to recreational activities and food security, respectively (Sowman & Cardoso 2010).

Prior to Namibian independence in 1990 the waters of the Northern Benguela were heavily exploited by foreign fishing fleets (Roux & Shannon 2004; Boyer & Hampton 2001). Management is based on single-stock assessments with the goal of increasing commercially important stocks, such as the sardine (Boyer & Hampton 2001). Therefore focus is not on enhancing resilience of the ecosystem, but rather increasing the stocks of the socio-economically important species (Boyer et al. 2001; Boyer & Hampton 2001). When the sardine stock initially declined in the 1960s (Boyer et al. 2001) management actions were based on observations of the Southern Benguela, where sardine domination alternates with anchovy domination. By shifting fishing pressure from the sardine to the anchovies, it was believed that sardines would recover due to decreased inter-species competition (Boyer et al. 2001; Roux & Shannon 2004; Shannon et al. 2004). Unexpectedly, through this one-stock approach, resilience was undermined leading to an initial decline in anchovy, later followed by a decline in all other pelagic fish and no recovery of sardines (Boyer et al. 2001).

After independence, the fishery shifted from international to local dominance (Boyer & Hampton 2001) and the Ministry of Fisheries and Marine Resources was established in 1991 with the mission to "[...] strengthen Namibia's position as a leading fishing nation and contribute towards the achievements of [their] economic, social and conservation goals for the benefit of all Namibians" (FAO 2002). Great effort was directed at stopping illegal fishing (Oelofsen 1999 as cited by Roux & Shannon 2004) and strict control on fishery vessels and in total allowable catches was implemented (Boyer & Hampton 2001). Additionally, the Benguela Large Marine Ecosystem research programme was launched in 2002 with the aim to develop an ecosystem-wide approach to environmental research (FAO 2002). Ecosystem-based management has been proposed as a better approach to manage complex adaptive systems (Sowman & Cardoso 2010; Roux & Shannon 2004). It takes into account trophic interactions as well as environmental variations effecting fish spawning and fish recruitment to set appropriate target exploitation rates for fisheries. Future management of exploited fisheries must also be flexible enough to deal with delayed responses to environmental perturbations transferred across scales. Successful management of marine ecosystem thus requires research traversing various disciplines as well as coordination and cooperation at national and international levels (Hofmann & Powell 1998). After decades of declining fish stocks, signs of recovery are evident in several of Namibia's marine resources but sardine biomass has yet to recover (Boyer & Hampton 2001). With regards to current management, it is still unknown what sustainable harvest levels should be in order to maintain the ecological stability and resilience of the system. (Boyer & Hampton 2001).

High fish biomass – pre-1990s

This regime was dominated by sardines up until the late 1960's. The stock started to decline due to overfishing in the late 1960's and by 1975 the stock had collapsed (Cury and Shannon 2004). The depletion of the sardine stock gave space for the stocks of other pelagic fish to grow, such as the anchovy (Engraulis capensis), horse mackerel (Trachurus trachurus capensis), cape hake (Merluccius capensis and Merluccius paradoxus) and bearded goby (Sufflogobius bibarbatus) (Hutchings et al. 2009; de Young et al. 2004; Cury & Shannon 2004). After the collapse of the sardines, fishing pressure shifted to other pelagic fish. Fluctuations in these stocks occurred throughout the 1980s (Roux and Shannon 2004). The Cape fur seal (Arctocephalus pusillus pusillus) and seabirds such as Cape Gannet (Morus capensis), Cape cormorant (Phalacocorax capensis) and the African penguin (Spheniscus demersus) are the top predators of pelagic fish; therefore populations of these predators fluctuated in tandem with the pelagic stocks (Wickens et al.1992; Boyer & Hampton 2001).

Low fish biomass – from the 1990s

The regime shifted in the 1990's and is characterized by depleted stocks of sardine and other pelagic fish (Hutchings et al. 2009).  There has been a general increase in sea surface temperature (SST) and longer hypoxic events (Monteiro & van der Plas 2006; Cury & Shannon, 2004; Pörtner & Langenbuch 2005). These conditions combined with low fish biomass have created a beneficial niche for jellyfish (Chrysaora hysoscella and Aequorea aequorea) (Cury & Shannon 2004) to increase in abundance (Kreiner et al. 2011). A recent study estimated the biomass of jellyfish in this region to be 2.4 times the amount of three of the most important commercial fish species combined (Roux et al. 2013). Seabirds and seals declined in population due to the depletion of sardine and anchovy stocks (Boyer & Hampton 2001). There is evidence that some of these populations are recovering and have stabilized, although the state remains one of overall low fish biomass, in particular low sardine biomass (Boyer & Hampton 2001).

The combined effects of overfishing and a series of large-scale environmental perturbations caused the system to shift from a high to low biomass regime (Boyer & Hampton, 2001; Cury & Shannon, 2004). The Benguela region has been a favored fishing ground for nearly a century (due to its fertile upwelling), although it wasn't until the 1960's that resource exploitation was intensified, with the arrival of large commercial fishing fleets (Boyer & Hampton 2001). Sardines were targeted as the preferred species and intensively harvested - primarily for overseas markets (Boyer & Hampton, 2001; Cury & Shannon, 2004). The pressure on this stock effected the foodweb dynamics (Cury & Shannon, 2004). Sardine stocks declined in the 1970s after which fishing was aimed primarily at anchovy in an attempt to allow sardine stocks to recover. This, however, led to a collapse of both stocks and, in the 1980s, the system became dominated by horse mackerel, bearded goby and jellyfish (de Young et al. 2004, Cury & Shannon 2004). The top-down pressure of intense fishing, in particular on sardines, likely paved the way for a regime shift to take place by reducing ecosystem resilience to withstand perturbations (Cury & Shannon 2004).

Several environmental anomalies of the 1990s acted as important drivers of this shift. These periodic bottom-up pressures are natural to the system although they were of particularly large magnitude during these years (Cury & Shannon 2004, Boyer & Hampton 2001). In 1993/1994 a low-oxygen water event from the Angolan current caused unusually extensive hypoxia in Namibian waters (Boyer & Hampton 2001; Cury & Shannon 2004). This was followed by a Benguela Niño event in 1995 (an inter-decadal climatic event) where warm, nutrient-poor water entered the system (Gammelsrød et al. as cited in Heymans et al. 2004; Cury & Shannon, 2004). The combination of these events led to poor recruitment conditions and high pelagic fish mortality resulting in a decline of most stocks (Heymans et al. 2004; Cury & Shannon, 2004; Boyer & Hampton, 2001). A general trend of warming SST in the Northern Benguela has also been identified as a potential indirect cause behind this shift although there remains some uncertainty around this, as there does around the potential contribution of other factors attributed to climate change (Bakun et al. 2010; Cury & Shannon, 2004; Hutchings et al. 2009).

Like most upwelling systems, the Northern Benguela is dominated by a small number of species that play an important role in maintaining ecosystem structure and function (Cury & Shannon 2004). Here, sardines are the dominant species structuring the ecosystem by performing a 'wasp-waist' control on species both above and below them in the foodweb (Cury & Shannon 2004). Nutrient rich, cooler waters of the upwelling provides favorable conditions for phytoplankton growth, which is controlled by sardine populations and allows for an enriched trophic web supporting productive fisheries (de Young et al. 2004; Bakun et al. 2009).  The system is vulnerable to periodic environmental variability (such as Benguela niño events) with some years allowing better recruitment than others (Boyer & Hampton 2001).

Consistent fishing, particularly on sardines, creates a top-down pressure on this system reducing its resilience to environmental variability, to the point where a series of bottom-up pulses (93/94 and 95 events) pushes the system into a new dynamic regime of depleted fish biomass (Cury & Shannon, 2004; Boyer et al. 2001; de Young et al. 2001). A key threshold, although difficult to determine in any detail, is possibly crossed when sardine stocks get so low that they were no longer able to maintain their populations (Allée effect), with repercussions throughout the whole foodweb. Fishing pressure continues during the poor recruitment years following the environmental perturbations most likely enhancing sardine collapse (Hutchings et al. 2009; Boyer et al. 2001.)

The new state of depleted fish biomass is reinforced by continued fishing pressure. Decreased fish biomass allows jellyfish to consume a greater proportion of plankton, increasing the number of jellyfish (Bakun et al. 2009). This has creates a reinforcing feedback loop whereby fish stock recovery is impeded by jellyfish competition for food as well as a possible jellyfish predation on certain fish larvae (Cury & Shannon 2004). It has been suggested that the increase of phytoplankton in the system due to less predation creates hypoxic conditions that hamper fish spawning and recruitment (Boyer et al. 2001 as cited by Cury & Shannon, 2004). These feedback mechanisms appear to lock the system in this new state, potentially difficult to reverse (Cury & Shannon 2004). There is uncertainty about the intensity and duration of low oxygen events and how these may contribute to keeping the system in this state (Hutchings et al. 2009). Warming sea temperatures may also contribute to the continuation of this state which is favorable for jellyfish that cope better in these conditions (Richardson et al. 2002). It is unknown how the apparent decline in upwelling intensity might affect this social-ecological system as well as whether or not this is an effect of anthropogenic climate change (Bakun et al. 2009).

Provisioning services decline after the shift from high biomass to low. Fish catch are plummeted, as exemplified by sardines, decreasing from 700,000 tons caught at the height of the industry to only 2,000 in 1996 (Boyer & Hampton 2001). Seal populations also decline (Roux 1998 as cited by Cury & Shannon 2004), possibly impacting the sealing industry. Jellyfish has direct negative economic impacts on the fishing industry, for example, spoiling fish catches and busting trawl nets (Lynam et al. 2006; Roux et al. 2013). The regulating service of water purification is negatively impacted via disruption of vital ecosystem processes necessary for maintaining a healthy water state. The cultural service of recreational fishing is reduced, as low fish biomass lead to fewer catches by anglers (Kirchner 1998 as cited in Boyer & Hampton 2001).

Loss of these services has unequal negative consequences on all the user groups of the resource. As a vast majority of Namibian marine catch is exported, therefore the food security of international consumers decreases (Sowman & Cardoso 2010). However, this impact is masked by the international market (Berkes et al. 2006). Additionally, local fishers and international investors lose income from decreased fish catch (Sowman & Cardoso 2010). Recreational fishers and few subsistence fishers were impacted by the shift, via decreasing access to recreational activities and food security, respectively (Sowman & Cardoso 2010).

Prior to Namibian independence in 1990 the waters of the Northern Benguela were heavily exploited by foreign fishing fleets (Roux & Shannon 2004; Boyer & Hampton 2001). Management is based on single-stock assessments with the goal of increasing commercially important stocks, such as the sardine (Boyer & Hampton 2001). Therefore focus is not on enhancing resilience of the ecosystem, but rather increasing the stocks of the socio-economically important species (Boyer et al. 2001; Boyer & Hampton 2001). When the sardine stock initially declined in the 1960s (Boyer et al. 2001) management actions were based on observations of the Southern Benguela, where sardine domination alternates with anchovy domination. By shifting fishing pressure from the sardine to the anchovies, it was believed that sardines would recover due to decreased inter-species competition (Boyer et al. 2001; Roux & Shannon 2004; Shannon et al. 2004). Unexpectedly, through this one-stock approach, resilience was undermined leading to an initial decline in anchovy, later followed by a decline in all other pelagic fish and no recovery of sardines (Boyer et al. 2001).

After independence, the fishery shifted from international to local dominance (Boyer & Hampton 2001) and the Ministry of Fisheries and Marine Resources was established in 1991 with the mission to "[...] strengthen Namibia's position as a leading fishing nation and contribute towards the achievements of [their] economic, social and conservation goals for the benefit of all Namibians" (FAO 2002). Great effort was directed at stopping illegal fishing (Oelofsen 1999 as cited by Roux & Shannon 2004) and strict control on fishery vessels and in total allowable catches was implemented (Boyer & Hampton 2001). Additionally, the Benguela Large Marine Ecosystem research programme was launched in 2002 with the aim to develop an ecosystem-wide approach to environmental research (FAO 2002). Ecosystem-based management has been proposed as a better approach to manage complex adaptive systems (Sowman & Cardoso 2010; Roux & Shannon 2004). It takes into account trophic interactions as well as environmental variations effecting fish spawning and fish recruitment to set appropriate target exploitation rates for fisheries. Future management of exploited fisheries must also be flexible enough to deal with delayed responses to environmental perturbations transferred across scales. Successful management of marine ecosystem thus requires research traversing various disciplines as well as coordination and cooperation at national and international levels (Hofmann & Powell 1998). After decades of declining fish stocks, signs of recovery are evident in several of Namibia's marine resources but sardine biomass has yet to recover (Boyer & Hampton 2001). With regards to current management, it is still unknown what sustainable harvest levels should be in order to maintain the ecological stability and resilience of the system. (Boyer & Hampton 2001).

High fish biomass – pre-1990s

This regime was dominated by sardines up until the late 1960's. The stock started to decline due to overfishing in the late 1960's and by 1975 the stock had collapsed (Cury and Shannon 2004). The depletion of the sardine stock gave space for the stocks of other pelagic fish to grow, such as the anchovy (Engraulis capensis), horse mackerel (Trachurus trachurus capensis), cape hake (Merluccius capensis and Merluccius paradoxus) and bearded goby (Sufflogobius bibarbatus) (Hutchings et al. 2009; de Young et al. 2004; Cury & Shannon 2004). After the collapse of the sardines, fishing pressure shifted to other pelagic fish. Fluctuations in these stocks occurred throughout the 1980s (Roux and Shannon 2004). The Cape fur seal (Arctocephalus pusillus pusillus) and seabirds such as Cape Gannet (Morus capensis), Cape cormorant (Phalacocorax capensis) and the African penguin (Spheniscus demersus) are the top predators of pelagic fish; therefore populations of these predators fluctuated in tandem with the pelagic stocks (Wickens et al.1992; Boyer & Hampton 2001).

Low fish biomass – from the 1990s

The regime shifted in the 1990's and is characterized by depleted stocks of sardine and other pelagic fish (Hutchings et al. 2009).  There has been a general increase in sea surface temperature (SST) and longer hypoxic events (Monteiro & van der Plas 2006; Cury & Shannon, 2004; Pörtner & Langenbuch 2005). These conditions combined with low fish biomass have created a beneficial niche for jellyfish (Chrysaora hysoscella and Aequorea aequorea) (Cury & Shannon 2004) to increase in abundance (Kreiner et al. 2011). A recent study estimated the biomass of jellyfish in this region to be 2.4 times the amount of three of the most important commercial fish species combined (Roux et al. 2013). Seabirds and seals declined in population due to the depletion of sardine and anchovy stocks (Boyer & Hampton 2001). There is evidence that some of these populations are recovering and have stabilized, although the state remains one of overall low fish biomass, in particular low sardine biomass (Boyer & Hampton 2001).

The combined effects of overfishing and a series of large-scale environmental perturbations caused the system to shift from a high to low biomass regime (Boyer & Hampton, 2001; Cury & Shannon, 2004). The Benguela region has been a favored fishing ground for nearly a century (due to its fertile upwelling), although it wasn't until the 1960's that resource exploitation was intensified, with the arrival of large commercial fishing fleets (Boyer & Hampton 2001). Sardines were targeted as the preferred species and intensively harvested - primarily for overseas markets (Boyer & Hampton, 2001; Cury & Shannon, 2004). The pressure on this stock effected the foodweb dynamics (Cury & Shannon, 2004). Sardine stocks declined in the 1970s after which fishing was aimed primarily at anchovy in an attempt to allow sardine stocks to recover. This, however, led to a collapse of both stocks and, in the 1980s, the system became dominated by horse mackerel, bearded goby and jellyfish (de Young et al. 2004, Cury & Shannon 2004). The top-down pressure of intense fishing, in particular on sardines, likely paved the way for a regime shift to take place by reducing ecosystem resilience to withstand perturbations (Cury & Shannon 2004).

Several environmental anomalies of the 1990s acted as important drivers of this shift. These periodic bottom-up pressures are natural to the system although they were of particularly large magnitude during these years (Cury & Shannon 2004, Boyer & Hampton 2001). In 1993/1994 a low-oxygen water event from the Angolan current caused unusually extensive hypoxia in Namibian waters (Boyer & Hampton 2001; Cury & Shannon 2004). This was followed by a Benguela Niño event in 1995 (an inter-decadal climatic event) where warm, nutrient-poor water entered the system (Gammelsrød et al. as cited in Heymans et al. 2004; Cury & Shannon, 2004). The combination of these events led to poor recruitment conditions and high pelagic fish mortality resulting in a decline of most stocks (Heymans et al. 2004; Cury & Shannon, 2004; Boyer & Hampton, 2001). A general trend of warming SST in the Northern Benguela has also been identified as a potential indirect cause behind this shift although there remains some uncertainty around this, as there does around the potential contribution of other factors attributed to climate change (Bakun et al. 2010; Cury & Shannon, 2004; Hutchings et al. 2009).

Like most upwelling systems, the Northern Benguela is dominated by a small number of species that play an important role in maintaining ecosystem structure and function (Cury & Shannon 2004). Here, sardines are the dominant species structuring the ecosystem by performing a 'wasp-waist' control on species both above and below them in the foodweb (Cury & Shannon 2004). Nutrient rich, cooler waters of the upwelling provides favorable conditions for phytoplankton growth, which is controlled by sardine populations and allows for an enriched trophic web supporting productive fisheries (de Young et al. 2004; Bakun et al. 2009).  The system is vulnerable to periodic environmental variability (such as Benguela niño events) with some years allowing better recruitment than others (Boyer & Hampton 2001).

Consistent fishing, particularly on sardines, creates a top-down pressure on this system reducing its resilience to environmental variability, to the point where a series of bottom-up pulses (93/94 and 95 events) pushes the system into a new dynamic regime of depleted fish biomass (Cury & Shannon, 2004; Boyer et al. 2001; de Young et al. 2001). A key threshold, although difficult to determine in any detail, is possibly crossed when sardine stocks get so low that they were no longer able to maintain their populations (Allée effect), with repercussions throughout the whole foodweb. Fishing pressure continues during the poor recruitment years following the environmental perturbations most likely enhancing sardine collapse (Hutchings et al. 2009; Boyer et al. 2001.)

The new state of depleted fish biomass is reinforced by continued fishing pressure. Decreased fish biomass allows jellyfish to consume a greater proportion of plankton, increasing the number of jellyfish (Bakun et al. 2009). This has creates a reinforcing feedback loop whereby fish stock recovery is impeded by jellyfish competition for food as well as a possible jellyfish predation on certain fish larvae (Cury & Shannon 2004). It has been suggested that the increase of phytoplankton in the system due to less predation creates hypoxic conditions that hamper fish spawning and recruitment (Boyer et al. 2001 as cited by Cury & Shannon, 2004). These feedback mechanisms appear to lock the system in this new state, potentially difficult to reverse (Cury & Shannon 2004). There is uncertainty about the intensity and duration of low oxygen events and how these may contribute to keeping the system in this state (Hutchings et al. 2009). Warming sea temperatures may also contribute to the continuation of this state which is favorable for jellyfish that cope better in these conditions (Richardson et al. 2002). It is unknown how the apparent decline in upwelling intensity might affect this social-ecological system as well as whether or not this is an effect of anthropogenic climate change (Bakun et al. 2009).

Provisioning services decline after the shift from high biomass to low. Fish catch are plummeted, as exemplified by sardines, decreasing from 700,000 tons caught at the height of the industry to only 2,000 in 1996 (Boyer & Hampton 2001). Seal populations also decline (Roux 1998 as cited by Cury & Shannon 2004), possibly impacting the sealing industry. Jellyfish has direct negative economic impacts on the fishing industry, for example, spoiling fish catches and busting trawl nets (Lynam et al. 2006; Roux et al. 2013). The regulating service of water purification is negatively impacted via disruption of vital ecosystem processes necessary for maintaining a healthy water state. The cultural service of recreational fishing is reduced, as low fish biomass lead to fewer catches by anglers (Kirchner 1998 as cited in Boyer & Hampton 2001).

Loss of these services has unequal negative consequences on all the user groups of the resource. As a vast majority of Namibian marine catch is exported, therefore the food security of international consumers decreases (Sowman & Cardoso 2010). However, this impact is masked by the international market (Berkes et al. 2006). Additionally, local fishers and international investors lose income from decreased fish catch (Sowman & Cardoso 2010). Recreational fishers and few subsistence fishers were impacted by the shift, via decreasing access to recreational activities and food security, respectively (Sowman & Cardoso 2010).

Prior to Namibian independence in 1990 the waters of the Northern Benguela were heavily exploited by foreign fishing fleets (Roux & Shannon 2004; Boyer & Hampton 2001). Management is based on single-stock assessments with the goal of increasing commercially important stocks, such as the sardine (Boyer & Hampton 2001). Therefore focus is not on enhancing resilience of the ecosystem, but rather increasing the stocks of the socio-economically important species (Boyer et al. 2001; Boyer & Hampton 2001). When the sardine stock initially declined in the 1960s (Boyer et al. 2001) management actions were based on observations of the Southern Benguela, where sardine domination alternates with anchovy domination. By shifting fishing pressure from the sardine to the anchovies, it was believed that sardines would recover due to decreased inter-species competition (Boyer et al. 2001; Roux & Shannon 2004; Shannon et al. 2004). Unexpectedly, through this one-stock approach, resilience was undermined leading to an initial decline in anchovy, later followed by a decline in all other pelagic fish and no recovery of sardines (Boyer et al. 2001).

After independence, the fishery shifted from international to local dominance (Boyer & Hampton 2001) and the Ministry of Fisheries and Marine Resources was established in 1991 with the mission to "[...] strengthen Namibia's position as a leading fishing nation and contribute towards the achievements of [their] economic, social and conservation goals for the benefit of all Namibians" (FAO 2002). Great effort was directed at stopping illegal fishing (Oelofsen 1999 as cited by Roux & Shannon 2004) and strict control on fishery vessels and in total allowable catches was implemented (Boyer & Hampton 2001). Additionally, the Benguela Large Marine Ecosystem research programme was launched in 2002 with the aim to develop an ecosystem-wide approach to environmental research (FAO 2002). Ecosystem-based management has been proposed as a better approach to manage complex adaptive systems (Sowman & Cardoso 2010; Roux & Shannon 2004). It takes into account trophic interactions as well as environmental variations effecting fish spawning and fish recruitment to set appropriate target exploitation rates for fisheries. Future management of exploited fisheries must also be flexible enough to deal with delayed responses to environmental perturbations transferred across scales. Successful management of marine ecosystem thus requires research traversing various disciplines as well as coordination and cooperation at national and international levels (Hofmann & Powell 1998). After decades of declining fish stocks, signs of recovery are evident in several of Namibia's marine resources but sardine biomass has yet to recover (Boyer & Hampton 2001). With regards to current management, it is still unknown what sustainable harvest levels should be in order to maintain the ecological stability and resilience of the system. (Boyer & Hampton 2001).

High fish biomass – pre-1990s

This regime was dominated by sardines up until the late 1960's. The stock started to decline due to overfishing in the late 1960's and by 1975 the stock had collapsed (Cury and Shannon 2004). The depletion of the sardine stock gave space for the stocks of other pelagic fish to grow, such as the anchovy (Engraulis capensis), horse mackerel (Trachurus trachurus capensis), cape hake (Merluccius capensis and Merluccius paradoxus) and bearded goby (Sufflogobius bibarbatus) (Hutchings et al. 2009; de Young et al. 2004; Cury & Shannon 2004). After the collapse of the sardines, fishing pressure shifted to other pelagic fish. Fluctuations in these stocks occurred throughout the 1980s (Roux and Shannon 2004). The Cape fur seal (Arctocephalus pusillus pusillus) and seabirds such as Cape Gannet (Morus capensis), Cape cormorant (Phalacocorax capensis) and the African penguin (Spheniscus demersus) are the top predators of pelagic fish; therefore populations of these predators fluctuated in tandem with the pelagic stocks (Wickens et al.1992; Boyer & Hampton 2001).

Low fish biomass – from the 1990s

The regime shifted in the 1990's and is characterized by depleted stocks of sardine and other pelagic fish (Hutchings et al. 2009).  There has been a general increase in sea surface temperature (SST) and longer hypoxic events (Monteiro & van der Plas 2006; Cury & Shannon, 2004; Pörtner & Langenbuch 2005). These conditions combined with low fish biomass have created a beneficial niche for jellyfish (Chrysaora hysoscella and Aequorea aequorea) (Cury & Shannon 2004) to increase in abundance (Kreiner et al. 2011). A recent study estimated the biomass of jellyfish in this region to be 2.4 times the amount of three of the most important commercial fish species combined (Roux et al. 2013). Seabirds and seals declined in population due to the depletion of sardine and anchovy stocks (Boyer & Hampton 2001). There is evidence that some of these populations are recovering and have stabilized, although the state remains one of overall low fish biomass, in particular low sardine biomass (Boyer & Hampton 2001).

The combined effects of overfishing and a series of large-scale environmental perturbations caused the system to shift from a high to low biomass regime (Boyer & Hampton, 2001; Cury & Shannon, 2004). The Benguela region has been a favored fishing ground for nearly a century (due to its fertile upwelling), although it wasn't until the 1960's that resource exploitation was intensified, with the arrival of large commercial fishing fleets (Boyer & Hampton 2001). Sardines were targeted as the preferred species and intensively harvested - primarily for overseas markets (Boyer & Hampton, 2001; Cury & Shannon, 2004). The pressure on this stock effected the foodweb dynamics (Cury & Shannon, 2004). Sardine stocks declined in the 1970s after which fishing was aimed primarily at anchovy in an attempt to allow sardine stocks to recover. This, however, led to a collapse of both stocks and, in the 1980s, the system became dominated by horse mackerel, bearded goby and jellyfish (de Young et al. 2004, Cury & Shannon 2004). The top-down pressure of intense fishing, in particular on sardines, likely paved the way for a regime shift to take place by reducing ecosystem resilience to withstand perturbations (Cury & Shannon 2004).

Several environmental anomalies of the 1990s acted as important drivers of this shift. These periodic bottom-up pressures are natural to the system although they were of particularly large magnitude during these years (Cury & Shannon 2004, Boyer & Hampton 2001). In 1993/1994 a low-oxygen water event from the Angolan current caused unusually extensive hypoxia in Namibian waters (Boyer & Hampton 2001; Cury & Shannon 2004). This was followed by a Benguela Niño event in 1995 (an inter-decadal climatic event) where warm, nutrient-poor water entered the system (Gammelsrød et al. as cited in Heymans et al. 2004; Cury & Shannon, 2004). The combination of these events led to poor recruitment conditions and high pelagic fish mortality resulting in a decline of most stocks (Heymans et al. 2004; Cury & Shannon, 2004; Boyer & Hampton, 2001). A general trend of warming SST in the Northern Benguela has also been identified as a potential indirect cause behind this shift although there remains some uncertainty around this, as there does around the potential contribution of other factors attributed to climate change (Bakun et al. 2010; Cury & Shannon, 2004; Hutchings et al. 2009).

Like most upwelling systems, the Northern Benguela is dominated by a small number of species that play an important role in maintaining ecosystem structure and function (Cury & Shannon 2004). Here, sardines are the dominant species structuring the ecosystem by performing a 'wasp-waist' control on species both above and below them in the foodweb (Cury & Shannon 2004). Nutrient rich, cooler waters of the upwelling provides favorable conditions for phytoplankton growth, which is controlled by sardine populations and allows for an enriched trophic web supporting productive fisheries (de Young et al. 2004; Bakun et al. 2009).  The system is vulnerable to periodic environmental variability (such as Benguela niño events) with some years allowing better recruitment than others (Boyer & Hampton 2001).

Consistent fishing, particularly on sardines, creates a top-down pressure on this system reducing its resilience to environmental variability, to the point where a series of bottom-up pulses (93/94 and 95 events) pushes the system into a new dynamic regime of depleted fish biomass (Cury & Shannon, 2004; Boyer et al. 2001; de Young et al. 2001). A key threshold, although difficult to determine in any detail, is possibly crossed when sardine stocks get so low that they were no longer able to maintain their populations (Allée effect), with repercussions throughout the whole foodweb. Fishing pressure continues during the poor recruitment years following the environmental perturbations most likely enhancing sardine collapse (Hutchings et al. 2009; Boyer et al. 2001.)

The new state of depleted fish biomass is reinforced by continued fishing pressure. Decreased fish biomass allows jellyfish to consume a greater proportion of plankton, increasing the number of jellyfish (Bakun et al. 2009). This has creates a reinforcing feedback loop whereby fish stock recovery is impeded by jellyfish competition for food as well as a possible jellyfish predation on certain fish larvae (Cury & Shannon 2004). It has been suggested that the increase of phytoplankton in the system due to less predation creates hypoxic conditions that hamper fish spawning and recruitment (Boyer et al. 2001 as cited by Cury & Shannon, 2004). These feedback mechanisms appear to lock the system in this new state, potentially difficult to reverse (Cury & Shannon 2004). There is uncertainty about the intensity and duration of low oxygen events and how these may contribute to keeping the system in this state (Hutchings et al. 2009). Warming sea temperatures may also contribute to the continuation of this state which is favorable for jellyfish that cope better in these conditions (Richardson et al. 2002). It is unknown how the apparent decline in upwelling intensity might affect this social-ecological system as well as whether or not this is an effect of anthropogenic climate change (Bakun et al. 2009).

Provisioning services decline after the shift from high biomass to low. Fish catch are plummeted, as exemplified by sardines, decreasing from 700,000 tons caught at the height of the industry to only 2,000 in 1996 (Boyer & Hampton 2001). Seal populations also decline (Roux 1998 as cited by Cury & Shannon 2004), possibly impacting the sealing industry. Jellyfish has direct negative economic impacts on the fishing industry, for example, spoiling fish catches and busting trawl nets (Lynam et al. 2006; Roux et al. 2013). The regulating service of water purification is negatively impacted via disruption of vital ecosystem processes necessary for maintaining a healthy water state. The cultural service of recreational fishing is reduced, as low fish biomass lead to fewer catches by anglers (Kirchner 1998 as cited in Boyer & Hampton 2001).

Loss of these services has unequal negative consequences on all the user groups of the resource. As a vast majority of Namibian marine catch is exported, therefore the food security of international consumers decreases (Sowman & Cardoso 2010). However, this impact is masked by the international market (Berkes et al. 2006). Additionally, local fishers and international investors lose income from decreased fish catch (Sowman & Cardoso 2010). Recreational fishers and few subsistence fishers were impacted by the shift, via decreasing access to recreational activities and food security, respectively (Sowman & Cardoso 2010).

Prior to Namibian independence in 1990 the waters of the Northern Benguela were heavily exploited by foreign fishing fleets (Roux & Shannon 2004; Boyer & Hampton 2001). Management is based on single-stock assessments with the goal of increasing commercially important stocks, such as the sardine (Boyer & Hampton 2001). Therefore focus is not on enhancing resilience of the ecosystem, but rather increasing the stocks of the socio-economically important species (Boyer et al. 2001; Boyer & Hampton 2001). When the sardine stock initially declined in the 1960s (Boyer et al. 2001) management actions were based on observations of the Southern Benguela, where sardine domination alternates with anchovy domination. By shifting fishing pressure from the sardine to the anchovies, it was believed that sardines would recover due to decreased inter-species competition (Boyer et al. 2001; Roux & Shannon 2004; Shannon et al. 2004). Unexpectedly, through this one-stock approach, resilience was undermined leading to an initial decline in anchovy, later followed by a decline in all other pelagic fish and no recovery of sardines (Boyer et al. 2001).

After independence, the fishery shifted from international to local dominance (Boyer & Hampton 2001) and the Ministry of Fisheries and Marine Resources was established in 1991 with the mission to "[...] strengthen Namibia's position as a leading fishing nation and contribute towards the achievements of [their] economic, social and conservation goals for the benefit of all Namibians" (FAO 2002). Great effort was directed at stopping illegal fishing (Oelofsen 1999 as cited by Roux & Shannon 2004) and strict control on fishery vessels and in total allowable catches was implemented (Boyer & Hampton 2001). Additionally, the Benguela Large Marine Ecosystem research programme was launched in 2002 with the aim to develop an ecosystem-wide approach to environmental research (FAO 2002). Ecosystem-based management has been proposed as a better approach to manage complex adaptive systems (Sowman & Cardoso 2010; Roux & Shannon 2004). It takes into account trophic interactions as well as environmental variations effecting fish spawning and fish recruitment to set appropriate target exploitation rates for fisheries. Future management of exploited fisheries must also be flexible enough to deal with delayed responses to environmental perturbations transferred across scales. Successful management of marine ecosystem thus requires research traversing various disciplines as well as coordination and cooperation at national and international levels (Hofmann & Powell 1998). After decades of declining fish stocks, signs of recovery are evident in several of Namibia's marine resources but sardine biomass has yet to recover (Boyer & Hampton 2001). With regards to current management, it is still unknown what sustainable harvest levels should be in order to maintain the ecological stability and resilience of the system. (Boyer & Hampton 2001).

High fish biomass – pre-1990s

This regime was dominated by sardines up until the late 1960's. The stock started to decline due to overfishing in the late 1960's and by 1975 the stock had collapsed (Cury and Shannon 2004). The depletion of the sardine stock gave space for the stocks of other pelagic fish to grow, such as the anchovy (Engraulis capensis), horse mackerel (Trachurus trachurus capensis), cape hake (Merluccius capensis and Merluccius paradoxus) and bearded goby (Sufflogobius bibarbatus) (Hutchings et al. 2009; de Young et al. 2004; Cury & Shannon 2004). After the collapse of the sardines, fishing pressure shifted to other pelagic fish. Fluctuations in these stocks occurred throughout the 1980s (Roux and Shannon 2004). The Cape fur seal (Arctocephalus pusillus pusillus) and seabirds such as Cape Gannet (Morus capensis), Cape cormorant (Phalacocorax capensis) and the African penguin (Spheniscus demersus) are the top predators of pelagic fish; therefore populations of these predators fluctuated in tandem with the pelagic stocks (Wickens et al.1992; Boyer & Hampton 2001).

Low fish biomass – from the 1990s

The regime shifted in the 1990's and is characterized by depleted stocks of sardine and other pelagic fish (Hutchings et al. 2009).  There has been a general increase in sea surface temperature (SST) and longer hypoxic events (Monteiro & van der Plas 2006; Cury & Shannon, 2004; Pörtner & Langenbuch 2005). These conditions combined with low fish biomass have created a beneficial niche for jellyfish (Chrysaora hysoscella and Aequorea aequorea) (Cury & Shannon 2004) to increase in abundance (Kreiner et al. 2011). A recent study estimated the biomass of jellyfish in this region to be 2.4 times the amount of three of the most important commercial fish species combined (Roux et al. 2013). Seabirds and seals declined in population due to the depletion of sardine and anchovy stocks (Boyer & Hampton 2001). There is evidence that some of these populations are recovering and have stabilized, although the state remains one of overall low fish biomass, in particular low sardine biomass (Boyer & Hampton 2001).

The combined effects of overfishing and a series of large-scale environmental perturbations caused the system to shift from a high to low biomass regime (Boyer & Hampton, 2001; Cury & Shannon, 2004). The Benguela region has been a favored fishing ground for nearly a century (due to its fertile upwelling), although it wasn't until the 1960's that resource exploitation was intensified, with the arrival of large commercial fishing fleets (Boyer & Hampton 2001). Sardines were targeted as the preferred species and intensively harvested - primarily for overseas markets (Boyer & Hampton, 2001; Cury & Shannon, 2004). The pressure on this stock effected the foodweb dynamics (Cury & Shannon, 2004). Sardine stocks declined in the 1970s after which fishing was aimed primarily at anchovy in an attempt to allow sardine stocks to recover. This, however, led to a collapse of both stocks and, in the 1980s, the system became dominated by horse mackerel, bearded goby and jellyfish (de Young et al. 2004, Cury & Shannon 2004). The top-down pressure of intense fishing, in particular on sardines, likely paved the way for a regime shift to take place by reducing ecosystem resilience to withstand perturbations (Cury & Shannon 2004).

Several environmental anomalies of the 1990s acted as important drivers of this shift. These periodic bottom-up pressures are natural to the system although they were of particularly large magnitude during these years (Cury & Shannon 2004, Boyer & Hampton 2001). In 1993/1994 a low-oxygen water event from the Angolan current caused unusually extensive hypoxia in Namibian waters (Boyer & Hampton 2001; Cury & Shannon 2004). This was followed by a Benguela Niño event in 1995 (an inter-decadal climatic event) where warm, nutrient-poor water entered the system (Gammelsrød et al. as cited in Heymans et al. 2004; Cury & Shannon, 2004). The combination of these events led to poor recruitment conditions and high pelagic fish mortality resulting in a decline of most stocks (Heymans et al. 2004; Cury & Shannon, 2004; Boyer & Hampton, 2001). A general trend of warming SST in the Northern Benguela has also been identified as a potential indirect cause behind this shift although there remains some uncertainty around this, as there does around the potential contribution of other factors attributed to climate change (Bakun et al. 2010; Cury & Shannon, 2004; Hutchings et al. 2009).

Like most upwelling systems, the Northern Benguela is dominated by a small number of species that play an important role in maintaining ecosystem structure and function (Cury & Shannon 2004). Here, sardines are the dominant species structuring the ecosystem by performing a 'wasp-waist' control on species both above and below them in the foodweb (Cury & Shannon 2004). Nutrient rich, cooler waters of the upwelling provides favorable conditions for phytoplankton growth, which is controlled by sardine populations and allows for an enriched trophic web supporting productive fisheries (de Young et al. 2004; Bakun et al. 2009).  The system is vulnerable to periodic environmental variability (such as Benguela niño events) with some years allowing better recruitment than others (Boyer & Hampton 2001).

Consistent fishing, particularly on sardines, creates a top-down pressure on this system reducing its resilience to environmental variability, to the point where a series of bottom-up pulses (93/94 and 95 events) pushes the system into a new dynamic regime of depleted fish biomass (Cury & Shannon, 2004; Boyer et al. 2001; de Young et al. 2001). A key threshold, although difficult to determine in any detail, is possibly crossed when sardine stocks get so low that they were no longer able to maintain their populations (Allée effect), with repercussions throughout the whole foodweb. Fishing pressure continues during the poor recruitment years following the environmental perturbations most likely enhancing sardine collapse (Hutchings et al. 2009; Boyer et al. 2001.)

The new state of depleted fish biomass is reinforced by continued fishing pressure. Decreased fish biomass allows jellyfish to consume a greater proportion of plankton, increasing the number of jellyfish (Bakun et al. 2009). This has creates a reinforcing feedback loop whereby fish stock recovery is impeded by jellyfish competition for food as well as a possible jellyfish predation on certain fish larvae (Cury & Shannon 2004). It has been suggested that the increase of phytoplankton in the system due to less predation creates hypoxic conditions that hamper fish spawning and recruitment (Boyer et al. 2001 as cited by Cury & Shannon, 2004). These feedback mechanisms appear to lock the system in this new state, potentially difficult to reverse (Cury & Shannon 2004). There is uncertainty about the intensity and duration of low oxygen events and how these may contribute to keeping the system in this state (Hutchings et al. 2009). Warming sea temperatures may also contribute to the continuation of this state which is favorable for jellyfish that cope better in these conditions (Richardson et al. 2002). It is unknown how the apparent decline in upwelling intensity might affect this social-ecological system as well as whether or not this is an effect of anthropogenic climate change (Bakun et al. 2009).

Provisioning services decline after the shift from high biomass to low. Fish catch are plummeted, as exemplified by sardines, decreasing from 700,000 tons caught at the height of the industry to only 2,000 in 1996 (Boyer & Hampton 2001). Seal populations also decline (Roux 1998 as cited by Cury & Shannon 2004), possibly impacting the sealing industry. Jellyfish has direct negative economic impacts on the fishing industry, for example, spoiling fish catches and busting trawl nets (Lynam et al. 2006; Roux et al. 2013). The regulating service of water purification is negatively impacted via disruption of vital ecosystem processes necessary for maintaining a healthy water state. The cultural service of recreational fishing is reduced, as low fish biomass lead to fewer catches by anglers (Kirchner 1998 as cited in Boyer & Hampton 2001).

Loss of these services has unequal negative consequences on all the user groups of the resource. As a vast majority of Namibian marine catch is exported, therefore the food security of international consumers decreases (Sowman & Cardoso 2010). However, this impact is masked by the international market (Berkes et al. 2006). Additionally, local fishers and international investors lose income from decreased fish catch (Sowman & Cardoso 2010). Recreational fishers and few subsistence fishers were impacted by the shift, via decreasing access to recreational activities and food security, respectively (Sowman & Cardoso 2010).

Prior to Namibian independence in 1990 the waters of the Northern Benguela were heavily exploited by foreign fishing fleets (Roux & Shannon 2004; Boyer & Hampton 2001). Management is based on single-stock assessments with the goal of increasing commercially important stocks, such as the sardine (Boyer & Hampton 2001). Therefore focus is not on enhancing resilience of the ecosystem, but rather increasing the stocks of the socio-economically important species (Boyer et al. 2001; Boyer & Hampton 2001). When the sardine stock initially declined in the 1960s (Boyer et al. 2001) management actions were based on observations of the Southern Benguela, where sardine domination alternates with anchovy domination. By shifting fishing pressure from the sardine to the anchovies, it was believed that sardines would recover due to decreased inter-species competition (Boyer et al. 2001; Roux & Shannon 2004; Shannon et al. 2004). Unexpectedly, through this one-stock approach, resilience was undermined leading to an initial decline in anchovy, later followed by a decline in all other pelagic fish and no recovery of sardines (Boyer et al. 2001).

After independence, the fishery shifted from international to local dominance (Boyer & Hampton 2001) and the Ministry of Fisheries and Marine Resources was established in 1991 with the mission to "[...] strengthen Namibia's position as a leading fishing nation and contribute towards the achievements of [their] economic, social and conservation goals for the benefit of all Namibians" (FAO 2002). Great effort was directed at stopping illegal fishing (Oelofsen 1999 as cited by Roux & Shannon 2004) and strict control on fishery vessels and in total allowable catches was implemented (Boyer & Hampton 2001). Additionally, the Benguela Large Marine Ecosystem research programme was launched in 2002 with the aim to develop an ecosystem-wide approach to environmental research (FAO 2002). Ecosystem-based management has been proposed as a better approach to manage complex adaptive systems (Sowman & Cardoso 2010; Roux & Shannon 2004). It takes into account trophic interactions as well as environmental variations effecting fish spawning and fish recruitment to set appropriate target exploitation rates for fisheries. Future management of exploited fisheries must also be flexible enough to deal with delayed responses to environmental perturbations transferred across scales. Successful management of marine ecosystem thus requires research traversing various disciplines as well as coordination and cooperation at national and international levels (Hofmann & Powell 1998). After decades of declining fish stocks, signs of recovery are evident in several of Namibia's marine resources but sardine biomass has yet to recover (Boyer & Hampton 2001). With regards to current management, it is still unknown what sustainable harvest levels should be in order to maintain the ecological stability and resilience of the system. (Boyer & Hampton 2001).

High fish biomass – pre-1990s

This regime was dominated by sardines up until the late 1960's. The stock started to decline due to overfishing in the late 1960's and by 1975 the stock had collapsed (Cury and Shannon 2004). The depletion of the sardine stock gave space for the stocks of other pelagic fish to grow, such as the anchovy (Engraulis capensis), horse mackerel (Trachurus trachurus capensis), cape hake (Merluccius capensis and Merluccius paradoxus) and bearded goby (Sufflogobius bibarbatus) (Hutchings et al. 2009; de Young et al. 2004; Cury & Shannon 2004). After the collapse of the sardines, fishing pressure shifted to other pelagic fish. Fluctuations in these stocks occurred throughout the 1980s (Roux and Shannon 2004). The Cape fur seal (Arctocephalus pusillus pusillus) and seabirds such as Cape Gannet (Morus capensis), Cape cormorant (Phalacocorax capensis) and the African penguin (Spheniscus demersus) are the top predators of pelagic fish; therefore populations of these predators fluctuated in tandem with the pelagic stocks (Wickens et al.1992; Boyer & Hampton 2001).

Low fish biomass – from the 1990s

The regime shifted in the 1990's and is characterized by depleted stocks of sardine and other pelagic fish (Hutchings et al. 2009).  There has been a general increase in sea surface temperature (SST) and longer hypoxic events (Monteiro & van der Plas 2006; Cury & Shannon, 2004; Pörtner & Langenbuch 2005). These conditions combined with low fish biomass have created a beneficial niche for jellyfish (Chrysaora hysoscella and Aequorea aequorea) (Cury & Shannon 2004) to increase in abundance (Kreiner et al. 2011). A recent study estimated the biomass of jellyfish in this region to be 2.4 times the amount of three of the most important commercial fish species combined (Roux et al. 2013). Seabirds and seals declined in population due to the depletion of sardine and anchovy stocks (Boyer & Hampton 2001). There is evidence that some of these populations are recovering and have stabilized, although the state remains one of overall low fish biomass, in particular low sardine biomass (Boyer & Hampton 2001).

The combined effects of overfishing and a series of large-scale environmental perturbations caused the system to shift from a high to low biomass regime (Boyer & Hampton, 2001; Cury & Shannon, 2004). The Benguela region has been a favored fishing ground for nearly a century (due to its fertile upwelling), although it wasn't until the 1960's that resource exploitation was intensified, with the arrival of large commercial fishing fleets (Boyer & Hampton 2001). Sardines were targeted as the preferred species and intensively harvested - primarily for overseas markets (Boyer & Hampton, 2001; Cury & Shannon, 2004). The pressure on this stock effected the foodweb dynamics (Cury & Shannon, 2004). Sardine stocks declined in the 1970s after which fishing was aimed primarily at anchovy in an attempt to allow sardine stocks to recover. This, however, led to a collapse of both stocks and, in the 1980s, the system became dominated by horse mackerel, bearded goby and jellyfish (de Young et al. 2004, Cury & Shannon 2004). The top-down pressure of intense fishing, in particular on sardines, likely paved the way for a regime shift to take place by reducing ecosystem resilience to withstand perturbations (Cury & Shannon 2004).

Several environmental anomalies of the 1990s acted as important drivers of this shift. These periodic bottom-up pressures are natural to the system although they were of particularly large magnitude during these years (Cury & Shannon 2004, Boyer & Hampton 2001). In 1993/1994 a low-oxygen water event from the Angolan current caused unusually extensive hypoxia in Namibian waters (Boyer & Hampton 2001; Cury & Shannon 2004). This was followed by a Benguela Niño event in 1995 (an inter-decadal climatic event) where warm, nutrient-poor water entered the system (Gammelsrød et al. as cited in Heymans et al. 2004; Cury & Shannon, 2004). The combination of these events led to poor recruitment conditions and high pelagic fish mortality resulting in a decline of most stocks (Heymans et al. 2004; Cury & Shannon, 2004; Boyer & Hampton, 2001). A general trend of warming SST in the Northern Benguela has also been identified as a potential indirect cause behind this shift although there remains some uncertainty around this, as there does around the potential contribution of other factors attributed to climate change (Bakun et al. 2010; Cury & Shannon, 2004; Hutchings et al. 2009).

Like most upwelling systems, the Northern Benguela is dominated by a small number of species that play an important role in maintaining ecosystem structure and function (Cury & Shannon 2004). Here, sardines are the dominant species structuring the ecosystem by performing a 'wasp-waist' control on species both above and below them in the foodweb (Cury & Shannon 2004). Nutrient rich, cooler waters of the upwelling provides favorable conditions for phytoplankton growth, which is controlled by sardine populations and allows for an enriched trophic web supporting productive fisheries (de Young et al. 2004; Bakun et al. 2009).  The system is vulnerable to periodic environmental variability (such as Benguela niño events) with some years allowing better recruitment than others (Boyer & Hampton 2001).

Consistent fishing, particularly on sardines, creates a top-down pressure on this system reducing its resilience to environmental variability, to the point where a series of bottom-up pulses (93/94 and 95 events) pushes the system into a new dynamic regime of depleted fish biomass (Cury & Shannon, 2004; Boyer et al. 2001; de Young et al. 2001). A key threshold, although difficult to determine in any detail, is possibly crossed when sardine stocks get so low that they were no longer able to maintain their populations (Allée effect), with repercussions throughout the whole foodweb. Fishing pressure continues during the poor recruitment years following the environmental perturbations most likely enhancing sardine collapse (Hutchings et al. 2009; Boyer et al. 2001.)

The new state of depleted fish biomass is reinforced by continued fishing pressure. Decreased fish biomass allows jellyfish to consume a greater proportion of plankton, increasing the number of jellyfish (Bakun et al. 2009). This has creates a reinforcing feedback loop whereby fish stock recovery is impeded by jellyfish competition for food as well as a possible jellyfish predation on certain fish larvae (Cury & Shannon 2004). It has been suggested that the increase of phytoplankton in the system due to less predation creates hypoxic conditions that hamper fish spawning and recruitment (Boyer et al. 2001 as cited by Cury & Shannon, 2004). These feedback mechanisms appear to lock the system in this new state, potentially difficult to reverse (Cury & Shannon 2004). There is uncertainty about the intensity and duration of low oxygen events and how these may contribute to keeping the system in this state (Hutchings et al. 2009). Warming sea temperatures may also contribute to the continuation of this state which is favorable for jellyfish that cope better in these conditions (Richardson et al. 2002). It is unknown how the apparent decline in upwelling intensity might affect this social-ecological system as well as whether or not this is an effect of anthropogenic climate change (Bakun et al. 2009).

Provisioning services decline after the shift from high biomass to low. Fish catch are plummeted, as exemplified by sardines, decreasing from 700,000 tons caught at the height of the industry to only 2,000 in 1996 (Boyer & Hampton 2001). Seal populations also decline (Roux 1998 as cited by Cury & Shannon 2004), possibly impacting the sealing industry. Jellyfish has direct negative economic impacts on the fishing industry, for example, spoiling fish catches and busting trawl nets (Lynam et al. 2006; Roux et al. 2013). The regulating service of water purification is negatively impacted via disruption of vital ecosystem processes necessary for maintaining a healthy water state. The cultural service of recreational fishing is reduced, as low fish biomass lead to fewer catches by anglers (Kirchner 1998 as cited in Boyer & Hampton 2001).

Loss of these services has unequal negative consequences on all the user groups of the resource. As a vast majority of Namibian marine catch is exported, therefore the food security of international consumers decreases (Sowman & Cardoso 2010). However, this impact is masked by the international market (Berkes et al. 2006). Additionally, local fishers and international investors lose income from decreased fish catch (Sowman & Cardoso 2010). Recreational fishers and few subsistence fishers were impacted by the shift, via decreasing access to recreational activities and food security, respectively (Sowman & Cardoso 2010).

Prior to Namibian independence in 1990 the waters of the Northern Benguela were heavily exploited by foreign fishing fleets (Roux & Shannon 2004; Boyer & Hampton 2001). Management is based on single-stock assessments with the goal of increasing commercially important stocks, such as the sardine (Boyer & Hampton 2001). Therefore focus is not on enhancing resilience of the ecosystem, but rather increasing the stocks of the socio-economically important species (Boyer et al. 2001; Boyer & Hampton 2001). When the sardine stock initially declined in the 1960s (Boyer et al. 2001) management actions were based on observations of the Southern Benguela, where sardine domination alternates with anchovy domination. By shifting fishing pressure from the sardine to the anchovies, it was believed that sardines would recover due to decreased inter-species competition (Boyer et al. 2001; Roux & Shannon 2004; Shannon et al. 2004). Unexpectedly, through this one-stock approach, resilience was undermined leading to an initial decline in anchovy, later followed by a decline in all other pelagic fish and no recovery of sardines (Boyer et al. 2001).

After independence, the fishery shifted from international to local dominance (Boyer & Hampton 2001) and the Ministry of Fisheries and Marine Resources was established in 1991 with the mission to "[...] strengthen Namibia's position as a leading fishing nation and contribute towards the achievements of [their] economic, social and conservation goals for the benefit of all Namibians" (FAO 2002). Great effort was directed at stopping illegal fishing (Oelofsen 1999 as cited by Roux & Shannon 2004) and strict control on fishery vessels and in total allowable catches was implemented (Boyer & Hampton 2001). Additionally, the Benguela Large Marine Ecosystem research programme was launched in 2002 with the aim to develop an ecosystem-wide approach to environmental research (FAO 2002). Ecosystem-based management has been proposed as a better approach to manage complex adaptive systems (Sowman & Cardoso 2010; Roux & Shannon 2004). It takes into account trophic interactions as well as environmental variations effecting fish spawning and fish recruitment to set appropriate target exploitation rates for fisheries. Future management of exploited fisheries must also be flexible enough to deal with delayed responses to environmental perturbations transferred across scales. Successful management of marine ecosystem thus requires research traversing various disciplines as well as coordination and cooperation at national and international levels (Hofmann & Powell 1998). After decades of declining fish stocks, signs of recovery are evident in several of Namibia's marine resources but sardine biomass has yet to recover (Boyer & Hampton 2001). With regards to current management, it is still unknown what sustainable harvest levels should be in order to maintain the ecological stability and resilience of the system. (Boyer & Hampton 2001).

High fish biomass – pre-1990s

This regime was dominated by sardines up until the late 1960's. The stock started to decline due to overfishing in the late 1960's and by 1975 the stock had collapsed (Cury and Shannon 2004). The depletion of the sardine stock gave space for the stocks of other pelagic fish to grow, such as the anchovy (Engraulis capensis), horse mackerel (Trachurus trachurus capensis), cape hake (Merluccius capensis and Merluccius paradoxus) and bearded goby (Sufflogobius bibarbatus) (Hutchings et al. 2009; de Young et al. 2004; Cury & Shannon 2004). After the collapse of the sardines, fishing pressure shifted to other pelagic fish. Fluctuations in these stocks occurred throughout the 1980s (Roux and Shannon 2004). The Cape fur seal (Arctocephalus pusillus pusillus) and seabirds such as Cape Gannet (Morus capensis), Cape cormorant (Phalacocorax capensis) and the African penguin (Spheniscus demersus) are the top predators of pelagic fish; therefore populations of these predators fluctuated in tandem with the pelagic stocks (Wickens et al.1992; Boyer & Hampton 2001).

Low fish biomass – from the 1990s

The regime shifted in the 1990's and is characterized by depleted stocks of sardine and other pelagic fish (Hutchings et al. 2009).  There has been a general increase in sea surface temperature (SST) and longer hypoxic events (Monteiro & van der Plas 2006; Cury & Shannon, 2004; Pörtner & Langenbuch 2005). These conditions combined with low fish biomass have created a beneficial niche for jellyfish (Chrysaora hysoscella and Aequorea aequorea) (Cury & Shannon 2004) to increase in abundance (Kreiner et al. 2011). A recent study estimated the biomass of jellyfish in this region to be 2.4 times the amount of three of the most important commercial fish species combined (Roux et al. 2013). Seabirds and seals declined in population due to the depletion of sardine and anchovy stocks (Boyer & Hampton 2001). There is evidence that some of these populations are recovering and have stabilized, although the state remains one of overall low fish biomass, in particular low sardine biomass (Boyer & Hampton 2001).

The combined effects of overfishing and a series of large-scale environmental perturbations caused the system to shift from a high to low biomass regime (Boyer & Hampton, 2001; Cury & Shannon, 2004). The Benguela region has been a favored fishing ground for nearly a century (due to its fertile upwelling), although it wasn't until the 1960's that resource exploitation was intensified, with the arrival of large commercial fishing fleets (Boyer & Hampton 2001). Sardines were targeted as the preferred species and intensively harvested - primarily for overseas markets (Boyer & Hampton, 2001; Cury & Shannon, 2004). The pressure on this stock effected the foodweb dynamics (Cury & Shannon, 2004). Sardine stocks declined in the 1970s after which fishing was aimed primarily at anchovy in an attempt to allow sardine stocks to recover. This, however, led to a collapse of both stocks and, in the 1980s, the system became dominated by horse mackerel, bearded goby and jellyfish (de Young et al. 2004, Cury & Shannon 2004). The top-down pressure of intense fishing, in particular on sardines, likely paved the way for a regime shift to take place by reducing ecosystem resilience to withstand perturbations (Cury & Shannon 2004).

Several environmental anomalies of the 1990s acted as important drivers of this shift. These periodic bottom-up pressures are natural to the system although they were of particularly large magnitude during these years (Cury & Shannon 2004, Boyer & Hampton 2001). In 1993/1994 a low-oxygen water event from the Angolan current caused unusually extensive hypoxia in Namibian waters (Boyer & Hampton 2001; Cury & Shannon 2004). This was followed by a Benguela Niño event in 1995 (an inter-decadal climatic event) where warm, nutrient-poor water entered the system (Gammelsrød et al. as cited in Heymans et al. 2004; Cury & Shannon, 2004). The combination of these events led to poor recruitment conditions and high pelagic fish mortality resulting in a decline of most stocks (Heymans et al. 2004; Cury & Shannon, 2004; Boyer & Hampton, 2001). A general trend of warming SST in the Northern Benguela has also been identified as a potential indirect cause behind this shift although there remains some uncertainty around this, as there does around the potential contribution of other factors attributed to climate change (Bakun et al. 2010; Cury & Shannon, 2004; Hutchings et al. 2009).

Like most upwelling systems, the Northern Benguela is dominated by a small number of species that play an important role in maintaining ecosystem structure and function (Cury & Shannon 2004). Here, sardines are the dominant species structuring the ecosystem by performing a 'wasp-waist' control on species both above and below them in the foodweb (Cury & Shannon 2004). Nutrient rich, cooler waters of the upwelling provides favorable conditions for phytoplankton growth, which is controlled by sardine populations and allows for an enriched trophic web supporting productive fisheries (de Young et al. 2004; Bakun et al. 2009).  The system is vulnerable to periodic environmental variability (such as Benguela niño events) with some years allowing better recruitment than others (Boyer & Hampton 2001).

Consistent fishing, particularly on sardines, creates a top-down pressure on this system reducing its resilience to environmental variability, to the point where a series of bottom-up pulses (93/94 and 95 events) pushes the system into a new dynamic regime of depleted fish biomass (Cury & Shannon, 2004; Boyer et al. 2001; de Young et al. 2001). A key threshold, although difficult to determine in any detail, is possibly crossed when sardine stocks get so low that they were no longer able to maintain their populations (Allée effect), with repercussions throughout the whole foodweb. Fishing pressure continues during the poor recruitment years following the environmental perturbations most likely enhancing sardine collapse (Hutchings et al. 2009; Boyer et al. 2001.)

The new state of depleted fish biomass is reinforced by continued fishing pressure. Decreased fish biomass allows jellyfish to consume a greater proportion of plankton, increasing the number of jellyfish (Bakun et al. 2009). This has creates a reinforcing feedback loop whereby fish stock recovery is impeded by jellyfish competition for food as well as a possible jellyfish predation on certain fish larvae (Cury & Shannon 2004). It has been suggested that the increase of phytoplankton in the system due to less predation creates hypoxic conditions that hamper fish spawning and recruitment (Boyer et al. 2001 as cited by Cury & Shannon, 2004). These feedback mechanisms appear to lock the system in this new state, potentially difficult to reverse (Cury & Shannon 2004). There is uncertainty about the intensity and duration of low oxygen events and how these may contribute to keeping the system in this state (Hutchings et al. 2009). Warming sea temperatures may also contribute to the continuation of this state which is favorable for jellyfish that cope better in these conditions (Richardson et al. 2002). It is unknown how the apparent decline in upwelling intensity might affect this social-ecological system as well as whether or not this is an effect of anthropogenic climate change (Bakun et al. 2009).

Provisioning services decline after the shift from high biomass to low. Fish catch are plummeted, as exemplified by sardines, decreasing from 700,000 tons caught at the height of the industry to only 2,000 in 1996 (Boyer & Hampton 2001). Seal populations also decline (Roux 1998 as cited by Cury & Shannon 2004), possibly impacting the sealing industry. Jellyfish has direct negative economic impacts on the fishing industry, for example, spoiling fish catches and busting trawl nets (Lynam et al. 2006; Roux et al. 2013). The regulating service of water purification is negatively impacted via disruption of vital ecosystem processes necessary for maintaining a healthy water state. The cultural service of recreational fishing is reduced, as low fish biomass lead to fewer catches by anglers (Kirchner 1998 as cited in Boyer & Hampton 2001).

Loss of these services has unequal negative consequences on all the user groups of the resource. As a vast majority of Namibian marine catch is exported, therefore the food security of international consumers decreases (Sowman & Cardoso 2010). However, this impact is masked by the international market (Berkes et al. 2006). Additionally, local fishers and international investors lose income from decreased fish catch (Sowman & Cardoso 2010). Recreational fishers and few subsistence fishers were impacted by the shift, via decreasing access to recreational activities and food security, respectively (Sowman & Cardoso 2010).

Prior to Namibian independence in 1990 the waters of the Northern Benguela were heavily exploited by foreign fishing fleets (Roux & Shannon 2004; Boyer & Hampton 2001). Management is based on single-stock assessments with the goal of increasing commercially important stocks, such as the sardine (Boyer & Hampton 2001). Therefore focus is not on enhancing resilience of the ecosystem, but rather increasing the stocks of the socio-economically important species (Boyer et al. 2001; Boyer & Hampton 2001). When the sardine stock initially declined in the 1960s (Boyer et al. 2001) management actions were based on observations of the Southern Benguela, where sardine domination alternates with anchovy domination. By shifting fishing pressure from the sardine to the anchovies, it was believed that sardines would recover due to decreased inter-species competition (Boyer et al. 2001; Roux & Shannon 2004; Shannon et al. 2004). Unexpectedly, through this one-stock approach, resilience was undermined leading to an initial decline in anchovy, later followed by a decline in all other pelagic fish and no recovery of sardines (Boyer et al. 2001).

After independence, the fishery shifted from international to local dominance (Boyer & Hampton 2001) and the Ministry of Fisheries and Marine Resources was established in 1991 with the mission to "[...] strengthen Namibia's position as a leading fishing nation and contribute towards the achievements of [their] economic, social and conservation goals for the benefit of all Namibians" (FAO 2002). Great effort was directed at stopping illegal fishing (Oelofsen 1999 as cited by Roux & Shannon 2004) and strict control on fishery vessels and in total allowable catches was implemented (Boyer & Hampton 2001). Additionally, the Benguela Large Marine Ecosystem research programme was launched in 2002 with the aim to develop an ecosystem-wide approach to environmental research (FAO 2002). Ecosystem-based management has been proposed as a better approach to manage complex adaptive systems (Sowman & Cardoso 2010; Roux & Shannon 2004). It takes into account trophic interactions as well as environmental variations effecting fish spawning and fish recruitment to set appropriate target exploitation rates for fisheries. Future management of exploited fisheries must also be flexible enough to deal with delayed responses to environmental perturbations transferred across scales. Successful management of marine ecosystem thus requires research traversing various disciplines as well as coordination and cooperation at national and international levels (Hofmann & Powell 1998). After decades of declining fish stocks, signs of recovery are evident in several of Namibia's marine resources but sardine biomass has yet to recover (Boyer & Hampton 2001). With regards to current management, it is still unknown what sustainable harvest levels should be in order to maintain the ecological stability and resilience of the system. (Boyer & Hampton 2001).

High fish biomass – pre-1990s

This regime was dominated by sardines up until the late 1960's. The stock started to decline due to overfishing in the late 1960's and by 1975 the stock had collapsed (Cury and Shannon 2004). The depletion of the sardine stock gave space for the stocks of other pelagic fish to grow, such as the anchovy (Engraulis capensis), horse mackerel (Trachurus trachurus capensis), cape hake (Merluccius capensis and Merluccius paradoxus) and bearded goby (Sufflogobius bibarbatus) (Hutchings et al. 2009; de Young et al. 2004; Cury & Shannon 2004). After the collapse of the sardines, fishing pressure shifted to other pelagic fish. Fluctuations in these stocks occurred throughout the 1980s (Roux and Shannon 2004). The Cape fur seal (Arctocephalus pusillus pusillus) and seabirds such as Cape Gannet (Morus capensis), Cape cormorant (Phalacocorax capensis) and the African penguin (Spheniscus demersus) are the top predators of pelagic fish; therefore populations of these predators fluctuated in tandem with the pelagic stocks (Wickens et al.1992; Boyer & Hampton 2001).

Low fish biomass – from the 1990s

The regime shifted in the 1990's and is characterized by depleted stocks of sardine and other pelagic fish (Hutchings et al. 2009).  There has been a general increase in sea surface temperature (SST) and longer hypoxic events (Monteiro & van der Plas 2006; Cury & Shannon, 2004; Pörtner & Langenbuch 2005). These conditions combined with low fish biomass have created a beneficial niche for jellyfish (Chrysaora hysoscella and Aequorea aequorea) (Cury & Shannon 2004) to increase in abundance (Kreiner et al. 2011). A recent study estimated the biomass of jellyfish in this region to be 2.4 times the amount of three of the most important commercial fish species combined (Roux et al. 2013). Seabirds and seals declined in population due to the depletion of sardine and anchovy stocks (Boyer & Hampton 2001). There is evidence that some of these populations are recovering and have stabilized, although the state remains one of overall low fish biomass, in particular low sardine biomass (Boyer & Hampton 2001).

The combined effects of overfishing and a series of large-scale environmental perturbations caused the system to shift from a high to low biomass regime (Boyer & Hampton, 2001; Cury & Shannon, 2004). The Benguela region has been a favored fishing ground for nearly a century (due to its fertile upwelling), although it wasn't until the 1960's that resource exploitation was intensified, with the arrival of large commercial fishing fleets (Boyer & Hampton 2001). Sardines were targeted as the preferred species and intensively harvested - primarily for overseas markets (Boyer & Hampton, 2001; Cury & Shannon, 2004). The pressure on this stock effected the foodweb dynamics (Cury & Shannon, 2004). Sardine stocks declined in the 1970s after which fishing was aimed primarily at anchovy in an attempt to allow sardine stocks to recover. This, however, led to a collapse of both stocks and, in the 1980s, the system became dominated by horse mackerel, bearded goby and jellyfish (de Young et al. 2004, Cury & Shannon 2004). The top-down pressure of intense fishing, in particular on sardines, likely paved the way for a regime shift to take place by reducing ecosystem resilience to withstand perturbations (Cury & Shannon 2004).

Several environmental anomalies of the 1990s acted as important drivers of this shift. These periodic bottom-up pressures are natural to the system although they were of particularly large magnitude during these years (Cury & Shannon 2004, Boyer & Hampton 2001). In 1993/1994 a low-oxygen water event from the Angolan current caused unusually extensive hypoxia in Namibian waters (Boyer & Hampton 2001; Cury & Shannon 2004). This was followed by a Benguela Niño event in 1995 (an inter-decadal climatic event) where warm, nutrient-poor water entered the system (Gammelsrød et al. as cited in Heymans et al. 2004; Cury & Shannon, 2004). The combination of these events led to poor recruitment conditions and high pelagic fish mortality resulting in a decline of most stocks (Heymans et al. 2004; Cury & Shannon, 2004; Boyer & Hampton, 2001). A general trend of warming SST in the Northern Benguela has also been identified as a potential indirect cause behind this shift although there remains some uncertainty around this, as there does around the potential contribution of other factors attributed to climate change (Bakun et al. 2010; Cury & Shannon, 2004; Hutchings et al. 2009).

Like most upwelling systems, the Northern Benguela is dominated by a small number of species that play an important role in maintaining ecosystem structure and function (Cury & Shannon 2004). Here, sardines are the dominant species structuring the ecosystem by performing a 'wasp-waist' control on species both above and below them in the foodweb (Cury & Shannon 2004). Nutrient rich, cooler waters of the upwelling provides favorable conditions for phytoplankton growth, which is controlled by sardine populations and allows for an enriched trophic web supporting productive fisheries (de Young et al. 2004; Bakun et al. 2009).  The system is vulnerable to periodic environmental variability (such as Benguela niño events) with some years allowing better recruitment than others (Boyer & Hampton 2001).

Consistent fishing, particularly on sardines, creates a top-down pressure on this system reducing its resilience to environmental variability, to the point where a series of bottom-up pulses (93/94 and 95 events) pushes the system into a new dynamic regime of depleted fish biomass (Cury & Shannon, 2004; Boyer et al. 2001; de Young et al. 2001). A key threshold, although difficult to determine in any detail, is possibly crossed when sardine stocks get so low that they were no longer able to maintain their populations (Allée effect), with repercussions throughout the whole foodweb. Fishing pressure continues during the poor recruitment years following the environmental perturbations most likely enhancing sardine collapse (Hutchings et al. 2009; Boyer et al. 2001.)

The new state of depleted fish biomass is reinforced by continued fishing pressure. Decreased fish biomass allows jellyfish to consume a greater proportion of plankton, increasing the number of jellyfish (Bakun et al. 2009). This has creates a reinforcing feedback loop whereby fish stock recovery is impeded by jellyfish competition for food as well as a possible jellyfish predation on certain fish larvae (Cury & Shannon 2004). It has been suggested that the increase of phytoplankton in the system due to less predation creates hypoxic conditions that hamper fish spawning and recruitment (Boyer et al. 2001 as cited by Cury & Shannon, 2004). These feedback mechanisms appear to lock the system in this new state, potentially difficult to reverse (Cury & Shannon 2004). There is uncertainty about the intensity and duration of low oxygen events and how these may contribute to keeping the system in this state (Hutchings et al. 2009). Warming sea temperatures may also contribute to the continuation of this state which is favorable for jellyfish that cope better in these conditions (Richardson et al. 2002). It is unknown how the apparent decline in upwelling intensity might affect this social-ecological system as well as whether or not this is an effect of anthropogenic climate change (Bakun et al. 2009).

Provisioning services decline after the shift from high biomass to low. Fish catch are plummeted, as exemplified by sardines, decreasing from 700,000 tons caught at the height of the industry to only 2,000 in 1996 (Boyer & Hampton 2001). Seal populations also decline (Roux 1998 as cited by Cury & Shannon 2004), possibly impacting the sealing industry. Jellyfish has direct negative economic impacts on the fishing industry, for example, spoiling fish catches and busting trawl nets (Lynam et al. 2006; Roux et al. 2013). The regulating service of water purification is negatively impacted via disruption of vital ecosystem processes necessary for maintaining a healthy water state. The cultural service of recreational fishing is reduced, as low fish biomass lead to fewer catches by anglers (Kirchner 1998 as cited in Boyer & Hampton 2001).

Loss of these services has unequal negative consequences on all the user groups of the resource. As a vast majority of Namibian marine catch is exported, therefore the food security of international consumers decreases (Sowman & Cardoso 2010). However, this impact is masked by the international market (Berkes et al. 2006). Additionally, local fishers and international investors lose income from decreased fish catch (Sowman & Cardoso 2010). Recreational fishers and few subsistence fishers were impacted by the shift, via decreasing access to recreational activities and food security, respectively (Sowman & Cardoso 2010).

Prior to Namibian independence in 1990 the waters of the Northern Benguela were heavily exploited by foreign fishing fleets (Roux & Shannon 2004; Boyer & Hampton 2001). Management is based on single-stock assessments with the goal of increasing commercially important stocks, such as the sardine (Boyer & Hampton 2001). Therefore focus is not on enhancing resilience of the ecosystem, but rather increasing the stocks of the socio-economically important species (Boyer et al. 2001; Boyer & Hampton 2001). When the sardine stock initially declined in the 1960s (Boyer et al. 2001) management actions were based on observations of the Southern Benguela, where sardine domination alternates with anchovy domination. By shifting fishing pressure from the sardine to the anchovies, it was believed that sardines would recover due to decreased inter-species competition (Boyer et al. 2001; Roux & Shannon 2004; Shannon et al. 2004). Unexpectedly, through this one-stock approach, resilience was undermined leading to an initial decline in anchovy, later followed by a decline in all other pelagic fish and no recovery of sardines (Boyer et al. 2001).

After independence, the fishery shifted from international to local dominance (Boyer & Hampton 2001) and the Ministry of Fisheries and Marine Resources was established in 1991 with the mission to "[...] strengthen Namibia's position as a leading fishing nation and contribute towards the achievements of [their] economic, social and conservation goals for the benefit of all Namibians" (FAO 2002). Great effort was directed at stopping illegal fishing (Oelofsen 1999 as cited by Roux & Shannon 2004) and strict control on fishery vessels and in total allowable catches was implemented (Boyer & Hampton 2001). Additionally, the Benguela Large Marine Ecosystem research programme was launched in 2002 with the aim to develop an ecosystem-wide approach to environmental research (FAO 2002). Ecosystem-based management has been proposed as a better approach to manage complex adaptive systems (Sowman & Cardoso 2010; Roux & Shannon 2004). It takes into account trophic interactions as well as environmental variations effecting fish spawning and fish recruitment to set appropriate target exploitation rates for fisheries. Future management of exploited fisheries must also be flexible enough to deal with delayed responses to environmental perturbations transferred across scales. Successful management of marine ecosystem thus requires research traversing various disciplines as well as coordination and cooperation at national and international levels (Hofmann & Powell 1998). After decades of declining fish stocks, signs of recovery are evident in several of Namibia's marine resources but sardine biomass has yet to recover (Boyer & Hampton 2001). With regards to current management, it is still unknown what sustainable harvest levels should be in order to maintain the ecological stability and resilience of the system. (Boyer & Hampton 2001).

High fish biomass – pre-1990s

This regime was dominated by sardines up until the late 1960's. The stock started to decline due to overfishing in the late 1960's and by 1975 the stock had collapsed (Cury and Shannon 2004). The depletion of the sardine stock gave space for the stocks of other pelagic fish to grow, such as the anchovy (Engraulis capensis), horse mackerel (Trachurus trachurus capensis), cape hake (Merluccius capensis and Merluccius paradoxus) and bearded goby (Sufflogobius bibarbatus) (Hutchings et al. 2009; de Young et al. 2004; Cury & Shannon 2004). After the collapse of the sardines, fishing pressure shifted to other pelagic fish. Fluctuations in these stocks occurred throughout the 1980s (Roux and Shannon 2004). The Cape fur seal (Arctocephalus pusillus pusillus) and seabirds such as Cape Gannet (Morus capensis), Cape cormorant (Phalacocorax capensis) and the African penguin (Spheniscus demersus) are the top predators of pelagic fish; therefore populations of these predators fluctuated in tandem with the pelagic stocks (Wickens et al.1992; Boyer & Hampton 2001).

Low fish biomass – from the 1990s

The regime shifted in the 1990's and is characterized by depleted stocks of sardine and other pelagic fish (Hutchings et al. 2009).  There has been a general increase in sea surface temperature (SST) and longer hypoxic events (Monteiro & van der Plas 2006; Cury & Shannon, 2004; Pörtner & Langenbuch 2005). These conditions combined with low fish biomass have created a beneficial niche for jellyfish (Chrysaora hysoscella and Aequorea aequorea) (Cury & Shannon 2004) to increase in abundance (Kreiner et al. 2011). A recent study estimated the biomass of jellyfish in this region to be 2.4 times the amount of three of the most important commercial fish species combined (Roux et al. 2013). Seabirds and seals declined in population due to the depletion of sardine and anchovy stocks (Boyer & Hampton 2001). There is evidence that some of these populations are recovering and have stabilized, although the state remains one of overall low fish biomass, in particular low sardine biomass (Boyer & Hampton 2001).

The combined effects of overfishing and a series of large-scale environmental perturbations caused the system to shift from a high to low biomass regime (Boyer & Hampton, 2001; Cury & Shannon, 2004). The Benguela region has been a favored fishing ground for nearly a century (due to its fertile upwelling), although it wasn't until the 1960's that resource exploitation was intensified, with the arrival of large commercial fishing fleets (Boyer & Hampton 2001). Sardines were targeted as the preferred species and intensively harvested - primarily for overseas markets (Boyer & Hampton, 2001; Cury & Shannon, 2004). The pressure on this stock effected the foodweb dynamics (Cury & Shannon, 2004). Sardine stocks declined in the 1970s after which fishing was aimed primarily at anchovy in an attempt to allow sardine stocks to recover. This, however, led to a collapse of both stocks and, in the 1980s, the system became dominated by horse mackerel, bearded goby and jellyfish (de Young et al. 2004, Cury & Shannon 2004). The top-down pressure of intense fishing, in particular on sardines, likely paved the way for a regime shift to take place by reducing ecosystem resilience to withstand perturbations (Cury & Shannon 2004).

Several environmental anomalies of the 1990s acted as important drivers of this shift. These periodic bottom-up pressures are natural to the system although they were of particularly large magnitude during these years (Cury & Shannon 2004, Boyer & Hampton 2001). In 1993/1994 a low-oxygen water event from the Angolan current caused unusually extensive hypoxia in Namibian waters (Boyer & Hampton 2001; Cury & Shannon 2004). This was followed by a Benguela Niño event in 1995 (an inter-decadal climatic event) where warm, nutrient-poor water entered the system (Gammelsrød et al. as cited in Heymans et al. 2004; Cury & Shannon, 2004). The combination of these events led to poor recruitment conditions and high pelagic fish mortality resulting in a decline of most stocks (Heymans et al. 2004; Cury & Shannon, 2004; Boyer & Hampton, 2001). A general trend of warming SST in the Northern Benguela has also been identified as a potential indirect cause behind this shift although there remains some uncertainty around this, as there does around the potential contribution of other factors attributed to climate change (Bakun et al. 2010; Cury & Shannon, 2004; Hutchings et al. 2009).

Like most upwelling systems, the Northern Benguela is dominated by a small number of species that play an important role in maintaining ecosystem structure and function (Cury & Shannon 2004). Here, sardines are the dominant species structuring the ecosystem by performing a 'wasp-waist' control on species both above and below them in the foodweb (Cury & Shannon 2004). Nutrient rich, cooler waters of the upwelling provides favorable conditions for phytoplankton growth, which is controlled by sardine populations and allows for an enriched trophic web supporting productive fisheries (de Young et al. 2004; Bakun et al. 2009).  The system is vulnerable to periodic environmental variability (such as Benguela niño events) with some years allowing better recruitment than others (Boyer & Hampton 2001).

Consistent fishing, particularly on sardines, creates a top-down pressure on this system reducing its resilience to environmental variability, to the point where a series of bottom-up pulses (93/94 and 95 events) pushes the system into a new dynamic regime of depleted fish biomass (Cury & Shannon, 2004; Boyer et al. 2001; de Young et al. 2001). A key threshold, although difficult to determine in any detail, is possibly crossed when sardine stocks get so low that they were no longer able to maintain their populations (Allée effect), with repercussions throughout the whole foodweb. Fishing pressure continues during the poor recruitment years following the environmental perturbations most likely enhancing sardine collapse (Hutchings et al. 2009; Boyer et al. 2001.)

The new state of depleted fish biomass is reinforced by continued fishing pressure. Decreased fish biomass allows jellyfish to consume a greater proportion of plankton, increasing the number of jellyfish (Bakun et al. 2009). This has creates a reinforcing feedback loop whereby fish stock recovery is impeded by jellyfish competition for food as well as a possible jellyfish predation on certain fish larvae (Cury & Shannon 2004). It has been suggested that the increase of phytoplankton in the system due to less predation creates hypoxic conditions that hamper fish spawning and recruitment (Boyer et al. 2001 as cited by Cury & Shannon, 2004). These feedback mechanisms appear to lock the system in this new state, potentially difficult to reverse (Cury & Shannon 2004). There is uncertainty about the intensity and duration of low oxygen events and how these may contribute to keeping the system in this state (Hutchings et al. 2009). Warming sea temperatures may also contribute to the continuation of this state which is favorable for jellyfish that cope better in these conditions (Richardson et al. 2002). It is unknown how the apparent decline in upwelling intensity might affect this social-ecological system as well as whether or not this is an effect of anthropogenic climate change (Bakun et al. 2009).

Provisioning services decline after the shift from high biomass to low. Fish catch are plummeted, as exemplified by sardines, decreasing from 700,000 tons caught at the height of the industry to only 2,000 in 1996 (Boyer & Hampton 2001). Seal populations also decline (Roux 1998 as cited by Cury & Shannon 2004), possibly impacting the sealing industry. Jellyfish has direct negative economic impacts on the fishing industry, for example, spoiling fish catches and busting trawl nets (Lynam et al. 2006; Roux et al. 2013). The regulating service of water purification is negatively impacted via disruption of vital ecosystem processes necessary for maintaining a healthy water state. The cultural service of recreational fishing is reduced, as low fish biomass lead to fewer catches by anglers (Kirchner 1998 as cited in Boyer & Hampton 2001).

Loss of these services has unequal negative consequences on all the user groups of the resource. As a vast majority of Namibian marine catch is exported, therefore the food security of international consumers decreases (Sowman & Cardoso 2010). However, this impact is masked by the international market (Berkes et al. 2006). Additionally, local fishers and international investors lose income from decreased fish catch (Sowman & Cardoso 2010). Recreational fishers and few subsistence fishers were impacted by the shift, via decreasing access to recreational activities and food security, respectively (Sowman & Cardoso 2010).

Prior to Namibian independence in 1990 the waters of the Northern Benguela were heavily exploited by foreign fishing fleets (Roux & Shannon 2004; Boyer & Hampton 2001). Management is based on single-stock assessments with the goal of increasing commercially important stocks, such as the sardine (Boyer & Hampton 2001). Therefore focus is not on enhancing resilience of the ecosystem, but rather increasing the stocks of the socio-economically important species (Boyer et al. 2001; Boyer & Hampton 2001). When the sardine stock initially declined in the 1960s (Boyer et al. 2001) management actions were based on observations of the Southern Benguela, where sardine domination alternates with anchovy domination. By shifting fishing pressure from the sardine to the anchovies, it was believed that sardines would recover due to decreased inter-species competition (Boyer et al. 2001; Roux & Shannon 2004; Shannon et al. 2004). Unexpectedly, through this one-stock approach, resilience was undermined leading to an initial decline in anchovy, later followed by a decline in all other pelagic fish and no recovery of sardines (Boyer et al. 2001).

After independence, the fishery shifted from international to local dominance (Boyer & Hampton 2001) and the Ministry of Fisheries and Marine Resources was established in 1991 with the mission to "[...] strengthen Namibia's position as a leading fishing nation and contribute towards the achievements of [their] economic, social and conservation goals for the benefit of all Namibians" (FAO 2002). Great effort was directed at stopping illegal fishing (Oelofsen 1999 as cited by Roux & Shannon 2004) and strict control on fishery vessels and in total allowable catches was implemented (Boyer & Hampton 2001). Additionally, the Benguela Large Marine Ecosystem research programme was launched in 2002 with the aim to develop an ecosystem-wide approach to environmental research (FAO 2002). Ecosystem-based management has been proposed as a better approach to manage complex adaptive systems (Sowman & Cardoso 2010; Roux & Shannon 2004). It takes into account trophic interactions as well as environmental variations effecting fish spawning and fish recruitment to set appropriate target exploitation rates for fisheries. Future management of exploited fisheries must also be flexible enough to deal with delayed responses to environmental perturbations transferred across scales. Successful management of marine ecosystem thus requires research traversing various disciplines as well as coordination and cooperation at national and international levels (Hofmann & Powell 1998). After decades of declining fish stocks, signs of recovery are evident in several of Namibia's marine resources but sardine biomass has yet to recover (Boyer & Hampton 2001). With regards to current management, it is still unknown what sustainable harvest levels should be in order to maintain the ecological stability and resilience of the system. (Boyer & Hampton 2001).

High fish biomass – pre-1990s

This regime was dominated by sardines up until the late 1960's. The stock started to decline due to overfishing in the late 1960's and by 1975 the stock had collapsed (Cury and Shannon 2004). The depletion of the sardine stock gave space for the stocks of other pelagic fish to grow, such as the anchovy (Engraulis capensis), horse mackerel (Trachurus trachurus capensis), cape hake (Merluccius capensis and Merluccius paradoxus) and bearded goby (Sufflogobius bibarbatus) (Hutchings et al. 2009; de Young et al. 2004; Cury & Shannon 2004). After the collapse of the sardines, fishing pressure shifted to other pelagic fish. Fluctuations in these stocks occurred throughout the 1980s (Roux and Shannon 2004). The Cape fur seal (Arctocephalus pusillus pusillus) and seabirds such as Cape Gannet (Morus capensis), Cape cormorant (Phalacocorax capensis) and the African penguin (Spheniscus demersus) are the top predators of pelagic fish; therefore populations of these predators fluctuated in tandem with the pelagic stocks (Wickens et al.1992; Boyer & Hampton 2001).

Low fish biomass – from the 1990s

The regime shifted in the 1990's and is characterized by depleted stocks of sardine and other pelagic fish (Hutchings et al. 2009).  There has been a general increase in sea surface temperature (SST) and longer hypoxic events (Monteiro & van der Plas 2006; Cury & Shannon, 2004; Pörtner & Langenbuch 2005). These conditions combined with low fish biomass have created a beneficial niche for jellyfish (Chrysaora hysoscella and Aequorea aequorea) (Cury & Shannon 2004) to increase in abundance (Kreiner et al. 2011). A recent study estimated the biomass of jellyfish in this region to be 2.4 times the amount of three of the most important commercial fish species combined (Roux et al. 2013). Seabirds and seals declined in population due to the depletion of sardine and anchovy stocks (Boyer & Hampton 2001). There is evidence that some of these populations are recovering and have stabilized, although the state remains one of overall low fish biomass, in particular low sardine biomass (Boyer & Hampton 2001).

The combined effects of overfishing and a series of large-scale environmental perturbations caused the system to shift from a high to low biomass regime (Boyer & Hampton, 2001; Cury & Shannon, 2004). The Benguela region has been a favored fishing ground for nearly a century (due to its fertile upwelling), although it wasn't until the 1960's that resource exploitation was intensified, with the arrival of large commercial fishing fleets (Boyer & Hampton 2001). Sardines were targeted as the preferred species and intensively harvested - primarily for overseas markets (Boyer & Hampton, 2001; Cury & Shannon, 2004). The pressure on this stock effected the foodweb dynamics (Cury & Shannon, 2004). Sardine stocks declined in the 1970s after which fishing was aimed primarily at anchovy in an attempt to allow sardine stocks to recover. This, however, led to a collapse of both stocks and, in the 1980s, the system became dominated by horse mackerel, bearded goby and jellyfish (de Young et al. 2004, Cury & Shannon 2004). The top-down pressure of intense fishing, in particular on sardines, likely paved the way for a regime shift to take place by reducing ecosystem resilience to withstand perturbations (Cury & Shannon 2004).

Several environmental anomalies of the 1990s acted as important drivers of this shift. These periodic bottom-up pressures are natural to the system although they were of particularly large magnitude during these years (Cury & Shannon 2004, Boyer & Hampton 2001). In 1993/1994 a low-oxygen water event from the Angolan current caused unusually extensive hypoxia in Namibian waters (Boyer & Hampton 2001; Cury & Shannon 2004). This was followed by a Benguela Niño event in 1995 (an inter-decadal climatic event) where warm, nutrient-poor water entered the system (Gammelsrød et al. as cited in Heymans et al. 2004; Cury & Shannon, 2004). The combination of these events led to poor recruitment conditions and high pelagic fish mortality resulting in a decline of most stocks (Heymans et al. 2004; Cury & Shannon, 2004; Boyer & Hampton, 2001). A general trend of warming SST in the Northern Benguela has also been identified as a potential indirect cause behind this shift although there remains some uncertainty around this, as there does around the potential contribution of other factors attributed to climate change (Bakun et al. 2010; Cury & Shannon, 2004; Hutchings et al. 2009).

Like most upwelling systems, the Northern Benguela is dominated by a small number of species that play an important role in maintaining ecosystem structure and function (Cury & Shannon 2004). Here, sardines are the dominant species structuring the ecosystem by performing a 'wasp-waist' control on species both above and below them in the foodweb (Cury & Shannon 2004). Nutrient rich, cooler waters of the upwelling provides favorable conditions for phytoplankton growth, which is controlled by sardine populations and allows for an enriched trophic web supporting productive fisheries (de Young et al. 2004; Bakun et al. 2009).  The system is vulnerable to periodic environmental variability (such as Benguela niño events) with some years allowing better recruitment than others (Boyer & Hampton 2001).

Consistent fishing, particularly on sardines, creates a top-down pressure on this system reducing its resilience to environmental variability, to the point where a series of bottom-up pulses (93/94 and 95 events) pushes the system into a new dynamic regime of depleted fish biomass (Cury & Shannon, 2004; Boyer et al. 2001; de Young et al. 2001). A key threshold, although difficult to determine in any detail, is possibly crossed when sardine stocks get so low that they were no longer able to maintain their populations (Allée effect), with repercussions throughout the whole foodweb. Fishing pressure continues during the poor recruitment years following the environmental perturbations most likely enhancing sardine collapse (Hutchings et al. 2009; Boyer et al. 2001.)

The new state of depleted fish biomass is reinforced by continued fishing pressure. Decreased fish biomass allows jellyfish to consume a greater proportion of plankton, increasing the number of jellyfish (Bakun et al. 2009). This has creates a reinforcing feedback loop whereby fish stock recovery is impeded by jellyfish competition for food as well as a possible jellyfish predation on certain fish larvae (Cury & Shannon 2004). It has been suggested that the increase of phytoplankton in the system due to less predation creates hypoxic conditions that hamper fish spawning and recruitment (Boyer et al. 2001 as cited by Cury & Shannon, 2004). These feedback mechanisms appear to lock the system in this new state, potentially difficult to reverse (Cury & Shannon 2004). There is uncertainty about the intensity and duration of low oxygen events and how these may contribute to keeping the system in this state (Hutchings et al. 2009). Warming sea temperatures may also contribute to the continuation of this state which is favorable for jellyfish that cope better in these conditions (Richardson et al. 2002). It is unknown how the apparent decline in upwelling intensity might affect this social-ecological system as well as whether or not this is an effect of anthropogenic climate change (Bakun et al. 2009).

Provisioning services decline after the shift from high biomass to low. Fish catch are plummeted, as exemplified by sardines, decreasing from 700,000 tons caught at the height of the industry to only 2,000 in 1996 (Boyer & Hampton 2001). Seal populations also decline (Roux 1998 as cited by Cury & Shannon 2004), possibly impacting the sealing industry. Jellyfish has direct negative economic impacts on the fishing industry, for example, spoiling fish catches and busting trawl nets (Lynam et al. 2006; Roux et al. 2013). The regulating service of water purification is negatively impacted via disruption of vital ecosystem processes necessary for maintaining a healthy water state. The cultural service of recreational fishing is reduced, as low fish biomass lead to fewer catches by anglers (Kirchner 1998 as cited in Boyer & Hampton 2001).

Loss of these services has unequal negative consequences on all the user groups of the resource. As a vast majority of Namibian marine catch is exported, therefore the food security of international consumers decreases (Sowman & Cardoso 2010). However, this impact is masked by the international market (Berkes et al. 2006). Additionally, local fishers and international investors lose income from decreased fish catch (Sowman & Cardoso 2010). Recreational fishers and few subsistence fishers were impacted by the shift, via decreasing access to recreational activities and food security, respectively (Sowman & Cardoso 2010).

Prior to Namibian independence in 1990 the waters of the Northern Benguela were heavily exploited by foreign fishing fleets (Roux & Shannon 2004; Boyer & Hampton 2001). Management is based on single-stock assessments with the goal of increasing commercially important stocks, such as the sardine (Boyer & Hampton 2001). Therefore focus is not on enhancing resilience of the ecosystem, but rather increasing the stocks of the socio-economically important species (Boyer et al. 2001; Boyer & Hampton 2001). When the sardine stock initially declined in the 1960s (Boyer et al. 2001) management actions were based on observations of the Southern Benguela, where sardine domination alternates with anchovy domination. By shifting fishing pressure from the sardine to the anchovies, it was believed that sardines would recover due to decreased inter-species competition (Boyer et al. 2001; Roux & Shannon 2004; Shannon et al. 2004). Unexpectedly, through this one-stock approach, resilience was undermined leading to an initial decline in anchovy, later followed by a decline in all other pelagic fish and no recovery of sardines (Boyer et al. 2001).

After independence, the fishery shifted from international to local dominance (Boyer & Hampton 2001) and the Ministry of Fisheries and Marine Resources was established in 1991 with the mission to "[...] strengthen Namibia's position as a leading fishing nation and contribute towards the achievements of [their] economic, social and conservation goals for the benefit of all Namibians" (FAO 2002). Great effort was directed at stopping illegal fishing (Oelofsen 1999 as cited by Roux & Shannon 2004) and strict control on fishery vessels and in total allowable catches was implemented (Boyer & Hampton 2001). Additionally, the Benguela Large Marine Ecosystem research programme was launched in 2002 with the aim to develop an ecosystem-wide approach to environmental research (FAO 2002). Ecosystem-based management has been proposed as a better approach to manage complex adaptive systems (Sowman & Cardoso 2010; Roux & Shannon 2004). It takes into account trophic interactions as well as environmental variations effecting fish spawning and fish recruitment to set appropriate target exploitation rates for fisheries. Future management of exploited fisheries must also be flexible enough to deal with delayed responses to environmental perturbations transferred across scales. Successful management of marine ecosystem thus requires research traversing various disciplines as well as coordination and cooperation at national and international levels (Hofmann & Powell 1998). After decades of declining fish stocks, signs of recovery are evident in several of Namibia's marine resources but sardine biomass has yet to recover (Boyer & Hampton 2001). With regards to current management, it is still unknown what sustainable harvest levels should be in order to maintain the ecological stability and resilience of the system. (Boyer & Hampton 2001).

High fish biomass – pre-1990s

This regime was dominated by sardines up until the late 1960's. The stock started to decline due to overfishing in the late 1960's and by 1975 the stock had collapsed (Cury and Shannon 2004). The depletion of the sardine stock gave space for the stocks of other pelagic fish to grow, such as the anchovy (Engraulis capensis), horse mackerel (Trachurus trachurus capensis), cape hake (Merluccius capensis and Merluccius paradoxus) and bearded goby (Sufflogobius bibarbatus) (Hutchings et al. 2009; de Young et al. 2004; Cury & Shannon 2004). After the collapse of the sardines, fishing pressure shifted to other pelagic fish. Fluctuations in these stocks occurred throughout the 1980s (Roux and Shannon 2004). The Cape fur seal (Arctocephalus pusillus pusillus) and seabirds such as Cape Gannet (Morus capensis), Cape cormorant (Phalacocorax capensis) and the African penguin (Spheniscus demersus) are the top predators of pelagic fish; therefore populations of these predators fluctuated in tandem with the pelagic stocks (Wickens et al.1992; Boyer & Hampton 2001).

Low fish biomass – from the 1990s

The regime shifted in the 1990's and is characterized by depleted stocks of sardine and other pelagic fish (Hutchings et al. 2009).  There has been a general increase in sea surface temperature (SST) and longer hypoxic events (Monteiro & van der Plas 2006; Cury & Shannon, 2004; Pörtner & Langenbuch 2005). These conditions combined with low fish biomass have created a beneficial niche for jellyfish (Chrysaora hysoscella and Aequorea aequorea) (Cury & Shannon 2004) to increase in abundance (Kreiner et al. 2011). A recent study estimated the biomass of jellyfish in this region to be 2.4 times the amount of three of the most important commercial fish species combined (Roux et al. 2013). Seabirds and seals declined in population due to the depletion of sardine and anchovy stocks (Boyer & Hampton 2001). There is evidence that some of these populations are recovering and have stabilized, although the state remains one of overall low fish biomass, in particular low sardine biomass (Boyer & Hampton 2001).

The combined effects of overfishing and a series of large-scale environmental perturbations caused the system to shift from a high to low biomass regime (Boyer & Hampton, 2001; Cury & Shannon, 2004). The Benguela region has been a favored fishing ground for nearly a century (due to its fertile upwelling), although it wasn't until the 1960's that resource exploitation was intensified, with the arrival of large commercial fishing fleets (Boyer & Hampton 2001). Sardines were targeted as the preferred species and intensively harvested - primarily for overseas markets (Boyer & Hampton, 2001; Cury & Shannon, 2004). The pressure on this stock effected the foodweb dynamics (Cury & Shannon, 2004). Sardine stocks declined in the 1970s after which fishing was aimed primarily at anchovy in an attempt to allow sardine stocks to recover. This, however, led to a collapse of both stocks and, in the 1980s, the system became dominated by horse mackerel, bearded goby and jellyfish (de Young et al. 2004, Cury & Shannon 2004). The top-down pressure of intense fishing, in particular on sardines, likely paved the way for a regime shift to take place by reducing ecosystem resilience to withstand perturbations (Cury & Shannon 2004).

Several environmental anomalies of the 1990s acted as important drivers of this shift. These periodic bottom-up pressures are natural to the system although they were of particularly large magnitude during these years (Cury & Shannon 2004, Boyer & Hampton 2001). In 1993/1994 a low-oxygen water event from the Angolan current caused unusually extensive hypoxia in Namibian waters (Boyer & Hampton 2001; Cury & Shannon 2004). This was followed by a Benguela Niño event in 1995 (an inter-decadal climatic event) where warm, nutrient-poor water entered the system (Gammelsrød et al. as cited in Heymans et al. 2004; Cury & Shannon, 2004). The combination of these events led to poor recruitment conditions and high pelagic fish mortality resulting in a decline of most stocks (Heymans et al. 2004; Cury & Shannon, 2004; Boyer & Hampton, 2001). A general trend of warming SST in the Northern Benguela has also been identified as a potential indirect cause behind this shift although there remains some uncertainty around this, as there does around the potential contribution of other factors attributed to climate change (Bakun et al. 2010; Cury & Shannon, 2004; Hutchings et al. 2009).

Like most upwelling systems, the Northern Benguela is dominated by a small number of species that play an important role in maintaining ecosystem structure and function (Cury & Shannon 2004). Here, sardines are the dominant species structuring the ecosystem by performing a 'wasp-waist' control on species both above and below them in the foodweb (Cury & Shannon 2004). Nutrient rich, cooler waters of the upwelling provides favorable conditions for phytoplankton growth, which is controlled by sardine populations and allows for an enriched trophic web supporting productive fisheries (de Young et al. 2004; Bakun et al. 2009).  The system is vulnerable to periodic environmental variability (such as Benguela niño events) with some years allowing better recruitment than others (Boyer & Hampton 2001).

Consistent fishing, particularly on sardines, creates a top-down pressure on this system reducing its resilience to environmental variability, to the point where a series of bottom-up pulses (93/94 and 95 events) pushes the system into a new dynamic regime of depleted fish biomass (Cury & Shannon, 2004; Boyer et al. 2001; de Young et al. 2001). A key threshold, although difficult to determine in any detail, is possibly crossed when sardine stocks get so low that they were no longer able to maintain their populations (Allée effect), with repercussions throughout the whole foodweb. Fishing pressure continues during the poor recruitment years following the environmental perturbations most likely enhancing sardine collapse (Hutchings et al. 2009; Boyer et al. 2001.)

The new state of depleted fish biomass is reinforced by continued fishing pressure. Decreased fish biomass allows jellyfish to consume a greater proportion of plankton, increasing the number of jellyfish (Bakun et al. 2009). This has creates a reinforcing feedback loop whereby fish stock recovery is impeded by jellyfish competition for food as well as a possible jellyfish predation on certain fish larvae (Cury & Shannon 2004). It has been suggested that the increase of phytoplankton in the system due to less predation creates hypoxic conditions that hamper fish spawning and recruitment (Boyer et al. 2001 as cited by Cury & Shannon, 2004). These feedback mechanisms appear to lock the system in this new state, potentially difficult to reverse (Cury & Shannon 2004). There is uncertainty about the intensity and duration of low oxygen events and how these may contribute to keeping the system in this state (Hutchings et al. 2009). Warming sea temperatures may also contribute to the continuation of this state which is favorable for jellyfish that cope better in these conditions (Richardson et al. 2002). It is unknown how the apparent decline in upwelling intensity might affect this social-ecological system as well as whether or not this is an effect of anthropogenic climate change (Bakun et al. 2009).

Provisioning services decline after the shift from high biomass to low. Fish catch are plummeted, as exemplified by sardines, decreasing from 700,000 tons caught at the height of the industry to only 2,000 in 1996 (Boyer & Hampton 2001). Seal populations also decline (Roux 1998 as cited by Cury & Shannon 2004), possibly impacting the sealing industry. Jellyfish has direct negative economic impacts on the fishing industry, for example, spoiling fish catches and busting trawl nets (Lynam et al. 2006; Roux et al. 2013). The regulating service of water purification is negatively impacted via disruption of vital ecosystem processes necessary for maintaining a healthy water state. The cultural service of recreational fishing is reduced, as low fish biomass lead to fewer catches by anglers (Kirchner 1998 as cited in Boyer & Hampton 2001).

Loss of these services has unequal negative consequences on all the user groups of the resource. As a vast majority of Namibian marine catch is exported, therefore the food security of international consumers decreases (Sowman & Cardoso 2010). However, this impact is masked by the international market (Berkes et al. 2006). Additionally, local fishers and international investors lose income from decreased fish catch (Sowman & Cardoso 2010). Recreational fishers and few subsistence fishers were impacted by the shift, via decreasing access to recreational activities and food security, respectively (Sowman & Cardoso 2010).

Prior to Namibian independence in 1990 the waters of the Northern Benguela were heavily exploited by foreign fishing fleets (Roux & Shannon 2004; Boyer & Hampton 2001). Management is based on single-stock assessments with the goal of increasing commercially important stocks, such as the sardine (Boyer & Hampton 2001). Therefore focus is not on enhancing resilience of the ecosystem, but rather increasing the stocks of the socio-economically important species (Boyer et al. 2001; Boyer & Hampton 2001). When the sardine stock initially declined in the 1960s (Boyer et al. 2001) management actions were based on observations of the Southern Benguela, where sardine domination alternates with anchovy domination. By shifting fishing pressure from the sardine to the anchovies, it was believed that sardines would recover due to decreased inter-species competition (Boyer et al. 2001; Roux & Shannon 2004; Shannon et al. 2004). Unexpectedly, through this one-stock approach, resilience was undermined leading to an initial decline in anchovy, later followed by a decline in all other pelagic fish and no recovery of sardines (Boyer et al. 2001).

After independence, the fishery shifted from international to local dominance (Boyer & Hampton 2001) and the Ministry of Fisheries and Marine Resources was established in 1991 with the mission to "[...] strengthen Namibia's position as a leading fishing nation and contribute towards the achievements of [their] economic, social and conservation goals for the benefit of all Namibians" (FAO 2002). Great effort was directed at stopping illegal fishing (Oelofsen 1999 as cited by Roux & Shannon 2004) and strict control on fishery vessels and in total allowable catches was implemented (Boyer & Hampton 2001). Additionally, the Benguela Large Marine Ecosystem research programme was launched in 2002 with the aim to develop an ecosystem-wide approach to environmental research (FAO 2002). Ecosystem-based management has been proposed as a better approach to manage complex adaptive systems (Sowman & Cardoso 2010; Roux & Shannon 2004). It takes into account trophic interactions as well as environmental variations effecting fish spawning and fish recruitment to set appropriate target exploitation rates for fisheries. Future management of exploited fisheries must also be flexible enough to deal with delayed responses to environmental perturbations transferred across scales. Successful management of marine ecosystem thus requires research traversing various disciplines as well as coordination and cooperation at national and international levels (Hofmann & Powell 1998). After decades of declining fish stocks, signs of recovery are evident in several of Namibia's marine resources but sardine biomass has yet to recover (Boyer & Hampton 2001). With regards to current management, it is still unknown what sustainable harvest levels should be in order to maintain the ecological stability and resilience of the system. (Boyer & Hampton 2001).

High fish biomass – pre-1990s

This regime was dominated by sardines up until the late 1960's. The stock started to decline due to overfishing in the late 1960's and by 1975 the stock had collapsed (Cury and Shannon 2004). The depletion of the sardine stock gave space for the stocks of other pelagic fish to grow, such as the anchovy (Engraulis capensis), horse mackerel (Trachurus trachurus capensis), cape hake (Merluccius capensis and Merluccius paradoxus) and bearded goby (Sufflogobius bibarbatus) (Hutchings et al. 2009; de Young et al. 2004; Cury & Shannon 2004). After the collapse of the sardines, fishing pressure shifted to other pelagic fish. Fluctuations in these stocks occurred throughout the 1980s (Roux and Shannon 2004). The Cape fur seal (Arctocephalus pusillus pusillus) and seabirds such as Cape Gannet (Morus capensis), Cape cormorant (Phalacocorax capensis) and the African penguin (Spheniscus demersus) are the top predators of pelagic fish; therefore populations of these predators fluctuated in tandem with the pelagic stocks (Wickens et al.1992; Boyer & Hampton 2001).

Low fish biomass – from the 1990s

The regime shifted in the 1990's and is characterized by depleted stocks of sardine and other pelagic fish (Hutchings et al. 2009).  There has been a general increase in sea surface temperature (SST) and longer hypoxic events (Monteiro & van der Plas 2006; Cury & Shannon, 2004; Pörtner & Langenbuch 2005). These conditions combined with low fish biomass have created a beneficial niche for jellyfish (Chrysaora hysoscella and Aequorea aequorea) (Cury & Shannon 2004) to increase in abundance (Kreiner et al. 2011). A recent study estimated the biomass of jellyfish in this region to be 2.4 times the amount of three of the most important commercial fish species combined (Roux et al. 2013). Seabirds and seals declined in population due to the depletion of sardine and anchovy stocks (Boyer & Hampton 2001). There is evidence that some of these populations are recovering and have stabilized, although the state remains one of overall low fish biomass, in particular low sardine biomass (Boyer & Hampton 2001).

The combined effects of overfishing and a series of large-scale environmental perturbations caused the system to shift from a high to low biomass regime (Boyer & Hampton, 2001; Cury & Shannon, 2004). The Benguela region has been a favored fishing ground for nearly a century (due to its fertile upwelling), although it wasn't until the 1960's that resource exploitation was intensified, with the arrival of large commercial fishing fleets (Boyer & Hampton 2001). Sardines were targeted as the preferred species and intensively harvested - primarily for overseas markets (Boyer & Hampton, 2001; Cury & Shannon, 2004). The pressure on this stock effected the foodweb dynamics (Cury & Shannon, 2004). Sardine stocks declined in the 1970s after which fishing was aimed primarily at anchovy in an attempt to allow sardine stocks to recover. This, however, led to a collapse of both stocks and, in the 1980s, the system became dominated by horse mackerel, bearded goby and jellyfish (de Young et al. 2004, Cury & Shannon 2004). The top-down pressure of intense fishing, in particular on sardines, likely paved the way for a regime shift to take place by reducing ecosystem resilience to withstand perturbations (Cury & Shannon 2004).

Several environmental anomalies of the 1990s acted as important drivers of this shift. These periodic bottom-up pressures are natural to the system although they were of particularly large magnitude during these years (Cury & Shannon 2004, Boyer & Hampton 2001). In 1993/1994 a low-oxygen water event from the Angolan current caused unusually extensive hypoxia in Namibian waters (Boyer & Hampton 2001; Cury & Shannon 2004). This was followed by a Benguela Niño event in 1995 (an inter-decadal climatic event) where warm, nutrient-poor water entered the system (Gammelsrød et al. as cited in Heymans et al. 2004; Cury & Shannon, 2004). The combination of these events led to poor recruitment conditions and high pelagic fish mortality resulting in a decline of most stocks (Heymans et al. 2004; Cury & Shannon, 2004; Boyer & Hampton, 2001). A general trend of warming SST in the Northern Benguela has also been identified as a potential indirect cause behind this shift although there remains some uncertainty around this, as there does around the potential contribution of other factors attributed to climate change (Bakun et al. 2010; Cury & Shannon, 2004; Hutchings et al. 2009).

Like most upwelling systems, the Northern Benguela is dominated by a small number of species that play an important role in maintaining ecosystem structure and function (Cury & Shannon 2004). Here, sardines are the dominant species structuring the ecosystem by performing a 'wasp-waist' control on species both above and below them in the foodweb (Cury & Shannon 2004). Nutrient rich, cooler waters of the upwelling provides favorable conditions for phytoplankton growth, which is controlled by sardine populations and allows for an enriched trophic web supporting productive fisheries (de Young et al. 2004; Bakun et al. 2009).  The system is vulnerable to periodic environmental variability (such as Benguela niño events) with some years allowing better recruitment than others (Boyer & Hampton 2001).

Consistent fishing, particularly on sardines, creates a top-down pressure on this system reducing its resilience to environmental variability, to the point where a series of bottom-up pulses (93/94 and 95 events) pushes the system into a new dynamic regime of depleted fish biomass (Cury & Shannon, 2004; Boyer et al. 2001; de Young et al. 2001). A key threshold, although difficult to determine in any detail, is possibly crossed when sardine stocks get so low that they were no longer able to maintain their populations (Allée effect), with repercussions throughout the whole foodweb. Fishing pressure continues during the poor recruitment years following the environmental perturbations most likely enhancing sardine collapse (Hutchings et al. 2009; Boyer et al. 2001.)

The new state of depleted fish biomass is reinforced by continued fishing pressure. Decreased fish biomass allows jellyfish to consume a greater proportion of plankton, increasing the number of jellyfish (Bakun et al. 2009). This has creates a reinforcing feedback loop whereby fish stock recovery is impeded by jellyfish competition for food as well as a possible jellyfish predation on certain fish larvae (Cury & Shannon 2004). It has been suggested that the increase of phytoplankton in the system due to less predation creates hypoxic conditions that hamper fish spawning and recruitment (Boyer et al. 2001 as cited by Cury & Shannon, 2004). These feedback mechanisms appear to lock the system in this new state, potentially difficult to reverse (Cury & Shannon 2004). There is uncertainty about the intensity and duration of low oxygen events and how these may contribute to keeping the system in this state (Hutchings et al. 2009). Warming sea temperatures may also contribute to the continuation of this state which is favorable for jellyfish that cope better in these conditions (Richardson et al. 2002). It is unknown how the apparent decline in upwelling intensity might affect this social-ecological system as well as whether or not this is an effect of anthropogenic climate change (Bakun et al. 2009).

Provisioning services decline after the shift from high biomass to low. Fish catch are plummeted, as exemplified by sardines, decreasing from 700,000 tons caught at the height of the industry to only 2,000 in 1996 (Boyer & Hampton 2001). Seal populations also decline (Roux 1998 as cited by Cury & Shannon 2004), possibly impacting the sealing industry. Jellyfish has direct negative economic impacts on the fishing industry, for example, spoiling fish catches and busting trawl nets (Lynam et al. 2006; Roux et al. 2013). The regulating service of water purification is negatively impacted via disruption of vital ecosystem processes necessary for maintaining a healthy water state. The cultural service of recreational fishing is reduced, as low fish biomass lead to fewer catches by anglers (Kirchner 1998 as cited in Boyer & Hampton 2001).

Loss of these services has unequal negative consequences on all the user groups of the resource. As a vast majority of Namibian marine catch is exported, therefore the food security of international consumers decreases (Sowman & Cardoso 2010). However, this impact is masked by the international market (Berkes et al. 2006). Additionally, local fishers and international investors lose income from decreased fish catch (Sowman & Cardoso 2010). Recreational fishers and few subsistence fishers were impacted by the shift, via decreasing access to recreational activities and food security, respectively (Sowman & Cardoso 2010).

Prior to Namibian independence in 1990 the waters of the Northern Benguela were heavily exploited by foreign fishing fleets (Roux & Shannon 2004; Boyer & Hampton 2001). Management is based on single-stock assessments with the goal of increasing commercially important stocks, such as the sardine (Boyer & Hampton 2001). Therefore focus is not on enhancing resilience of the ecosystem, but rather increasing the stocks of the socio-economically important species (Boyer et al. 2001; Boyer & Hampton 2001). When the sardine stock initially declined in the 1960s (Boyer et al. 2001) management actions were based on observations of the Southern Benguela, where sardine domination alternates with anchovy domination. By shifting fishing pressure from the sardine to the anchovies, it was believed that sardines would recover due to decreased inter-species competition (Boyer et al. 2001; Roux & Shannon 2004; Shannon et al. 2004). Unexpectedly, through this one-stock approach, resilience was undermined leading to an initial decline in anchovy, later followed by a decline in all other pelagic fish and no recovery of sardines (Boyer et al. 2001).

After independence, the fishery shifted from international to local dominance (Boyer & Hampton 2001) and the Ministry of Fisheries and Marine Resources was established in 1991 with the mission to "[...] strengthen Namibia's position as a leading fishing nation and contribute towards the achievements of [their] economic, social and conservation goals for the benefit of all Namibians" (FAO 2002). Great effort was directed at stopping illegal fishing (Oelofsen 1999 as cited by Roux & Shannon 2004) and strict control on fishery vessels and in total allowable catches was implemented (Boyer & Hampton 2001). Additionally, the Benguela Large Marine Ecosystem research programme was launched in 2002 with the aim to develop an ecosystem-wide approach to environmental research (FAO 2002). Ecosystem-based management has been proposed as a better approach to manage complex adaptive systems (Sowman & Cardoso 2010; Roux & Shannon 2004). It takes into account trophic interactions as well as environmental variations effecting fish spawning and fish recruitment to set appropriate target exploitation rates for fisheries. Future management of exploited fisheries must also be flexible enough to deal with delayed responses to environmental perturbations transferred across scales. Successful management of marine ecosystem thus requires research traversing various disciplines as well as coordination and cooperation at national and international levels (Hofmann & Powell 1998). After decades of declining fish stocks, signs of recovery are evident in several of Namibia's marine resources but sardine biomass has yet to recover (Boyer & Hampton 2001). With regards to current management, it is still unknown what sustainable harvest levels should be in order to maintain the ecological stability and resilience of the system. (Boyer & Hampton 2001).

High fish biomass – pre-1990s

This regime was dominated by sardines up until the late 1960's. The stock started to decline due to overfishing in the late 1960's and by 1975 the stock had collapsed (Cury and Shannon 2004). The depletion of the sardine stock gave space for the stocks of other pelagic fish to grow, such as the anchovy (Engraulis capensis), horse mackerel (Trachurus trachurus capensis), cape hake (Merluccius capensis and Merluccius paradoxus) and bearded goby (Sufflogobius bibarbatus) (Hutchings et al. 2009; de Young et al. 2004; Cury & Shannon 2004). After the collapse of the sardines, fishing pressure shifted to other pelagic fish. Fluctuations in these stocks occurred throughout the 1980s (Roux and Shannon 2004). The Cape fur seal (Arctocephalus pusillus pusillus) and seabirds such as Cape Gannet (Morus capensis), Cape cormorant (Phalacocorax capensis) and the African penguin (Spheniscus demersus) are the top predators of pelagic fish; therefore populations of these predators fluctuated in tandem with the pelagic stocks (Wickens et al.1992; Boyer & Hampton 2001).

Low fish biomass – from the 1990s

The regime shifted in the 1990's and is characterized by depleted stocks of sardine and other pelagic fish (Hutchings et al. 2009).  There has been a general increase in sea surface temperature (SST) and longer hypoxic events (Monteiro & van der Plas 2006; Cury & Shannon, 2004; Pörtner & Langenbuch 2005). These conditions combined with low fish biomass have created a beneficial niche for jellyfish (Chrysaora hysoscella and Aequorea aequorea) (Cury & Shannon 2004) to increase in abundance (Kreiner et al. 2011). A recent study estimated the biomass of jellyfish in this region to be 2.4 times the amount of three of the most important commercial fish species combined (Roux et al. 2013). Seabirds and seals declined in population due to the depletion of sardine and anchovy stocks (Boyer & Hampton 2001). There is evidence that some of these populations are recovering and have stabilized, although the state remains one of overall low fish biomass, in particular low sardine biomass (Boyer & Hampton 2001).

The combined effects of overfishing and a series of large-scale environmental perturbations caused the system to shift from a high to low biomass regime (Boyer & Hampton, 2001; Cury & Shannon, 2004). The Benguela region has been a favored fishing ground for nearly a century (due to its fertile upwelling), although it wasn't until the 1960's that resource exploitation was intensified, with the arrival of large commercial fishing fleets (Boyer & Hampton 2001). Sardines were targeted as the preferred species and intensively harvested - primarily for overseas markets (Boyer & Hampton, 2001; Cury & Shannon, 2004). The pressure on this stock effected the foodweb dynamics (Cury & Shannon, 2004). Sardine stocks declined in the 1970s after which fishing was aimed primarily at anchovy in an attempt to allow sardine stocks to recover. This, however, led to a collapse of both stocks and, in the 1980s, the system became dominated by horse mackerel, bearded goby and jellyfish (de Young et al. 2004, Cury & Shannon 2004). The top-down pressure of intense fishing, in particular on sardines, likely paved the way for a regime shift to take place by reducing ecosystem resilience to withstand perturbations (Cury & Shannon 2004).

Several environmental anomalies of the 1990s acted as important drivers of this shift. These periodic bottom-up pressures are natural to the system although they were of particularly large magnitude during these years (Cury & Shannon 2004, Boyer & Hampton 2001). In 1993/1994 a low-oxygen water event from the Angolan current caused unusually extensive hypoxia in Namibian waters (Boyer & Hampton 2001; Cury & Shannon 2004). This was followed by a Benguela Niño event in 1995 (an inter-decadal climatic event) where warm, nutrient-poor water entered the system (Gammelsrød et al. as cited in Heymans et al. 2004; Cury & Shannon, 2004). The combination of these events led to poor recruitment conditions and high pelagic fish mortality resulting in a decline of most stocks (Heymans et al. 2004; Cury & Shannon, 2004; Boyer & Hampton, 2001). A general trend of warming SST in the Northern Benguela has also been identified as a potential indirect cause behind this shift although there remains some uncertainty around this, as there does around the potential contribution of other factors attributed to climate change (Bakun et al. 2010; Cury & Shannon, 2004; Hutchings et al. 2009).

Like most upwelling systems, the Northern Benguela is dominated by a small number of species that play an important role in maintaining ecosystem structure and function (Cury & Shannon 2004). Here, sardines are the dominant species structuring the ecosystem by performing a 'wasp-waist' control on species both above and below them in the foodweb (Cury & Shannon 2004). Nutrient rich, cooler waters of the upwelling provides favorable conditions for phytoplankton growth, which is controlled by sardine populations and allows for an enriched trophic web supporting productive fisheries (de Young et al. 2004; Bakun et al. 2009).  The system is vulnerable to periodic environmental variability (such as Benguela niño events) with some years allowing better recruitment than others (Boyer & Hampton 2001).

Consistent fishing, particularly on sardines, creates a top-down pressure on this system reducing its resilience to environmental variability, to the point where a series of bottom-up pulses (93/94 and 95 events) pushes the system into a new dynamic regime of depleted fish biomass (Cury & Shannon, 2004; Boyer et al. 2001; de Young et al. 2001). A key threshold, although difficult to determine in any detail, is possibly crossed when sardine stocks get so low that they were no longer able to maintain their populations (Allée effect), with repercussions throughout the whole foodweb. Fishing pressure continues during the poor recruitment years following the environmental perturbations most likely enhancing sardine collapse (Hutchings et al. 2009; Boyer et al. 2001.)

The new state of depleted fish biomass is reinforced by continued fishing pressure. Decreased fish biomass allows jellyfish to consume a greater proportion of plankton, increasing the number of jellyfish (Bakun et al. 2009). This has creates a reinforcing feedback loop whereby fish stock recovery is impeded by jellyfish competition for food as well as a possible jellyfish predation on certain fish larvae (Cury & Shannon 2004). It has been suggested that the increase of phytoplankton in the system due to less predation creates hypoxic conditions that hamper fish spawning and recruitment (Boyer et al. 2001 as cited by Cury & Shannon, 2004). These feedback mechanisms appear to lock the system in this new state, potentially difficult to reverse (Cury & Shannon 2004). There is uncertainty about the intensity and duration of low oxygen events and how these may contribute to keeping the system in this state (Hutchings et al. 2009). Warming sea temperatures may also contribute to the continuation of this state which is favorable for jellyfish that cope better in these conditions (Richardson et al. 2002). It is unknown how the apparent decline in upwelling intensity might affect this social-ecological system as well as whether or not this is an effect of anthropogenic climate change (Bakun et al. 2009).

Provisioning services decline after the shift from high biomass to low. Fish catch are plummeted, as exemplified by sardines, decreasing from 700,000 tons caught at the height of the industry to only 2,000 in 1996 (Boyer & Hampton 2001). Seal populations also decline (Roux 1998 as cited by Cury & Shannon 2004), possibly impacting the sealing industry. Jellyfish has direct negative economic impacts on the fishing industry, for example, spoiling fish catches and busting trawl nets (Lynam et al. 2006; Roux et al. 2013). The regulating service of water purification is negatively impacted via disruption of vital ecosystem processes necessary for maintaining a healthy water state. The cultural service of recreational fishing is reduced, as low fish biomass lead to fewer catches by anglers (Kirchner 1998 as cited in Boyer & Hampton 2001).

Loss of these services has unequal negative consequences on all the user groups of the resource. As a vast majority of Namibian marine catch is exported, therefore the food security of international consumers decreases (Sowman & Cardoso 2010). However, this impact is masked by the international market (Berkes et al. 2006). Additionally, local fishers and international investors lose income from decreased fish catch (Sowman & Cardoso 2010). Recreational fishers and few subsistence fishers were impacted by the shift, via decreasing access to recreational activities and food security, respectively (Sowman & Cardoso 2010).

Prior to Namibian independence in 1990 the waters of the Northern Benguela were heavily exploited by foreign fishing fleets (Roux & Shannon 2004; Boyer & Hampton 2001). Management is based on single-stock assessments with the goal of increasing commercially important stocks, such as the sardine (Boyer & Hampton 2001). Therefore focus is not on enhancing resilience of the ecosystem, but rather increasing the stocks of the socio-economically important species (Boyer et al. 2001; Boyer & Hampton 2001). When the sardine stock initially declined in the 1960s (Boyer et al. 2001) management actions were based on observations of the Southern Benguela, where sardine domination alternates with anchovy domination. By shifting fishing pressure from the sardine to the anchovies, it was believed that sardines would recover due to decreased inter-species competition (Boyer et al. 2001; Roux & Shannon 2004; Shannon et al. 2004). Unexpectedly, through this one-stock approach, resilience was undermined leading to an initial decline in anchovy, later followed by a decline in all other pelagic fish and no recovery of sardines (Boyer et al. 2001).

After independence, the fishery shifted from international to local dominance (Boyer & Hampton 2001) and the Ministry of Fisheries and Marine Resources was established in 1991 with the mission to "[...] strengthen Namibia's position as a leading fishing nation and contribute towards the achievements of [their] economic, social and conservation goals for the benefit of all Namibians" (FAO 2002). Great effort was directed at stopping illegal fishing (Oelofsen 1999 as cited by Roux & Shannon 2004) and strict control on fishery vessels and in total allowable catches was implemented (Boyer & Hampton 2001). Additionally, the Benguela Large Marine Ecosystem research programme was launched in 2002 with the aim to develop an ecosystem-wide approach to environmental research (FAO 2002). Ecosystem-based management has been proposed as a better approach to manage complex adaptive systems (Sowman & Cardoso 2010; Roux & Shannon 2004). It takes into account trophic interactions as well as environmental variations effecting fish spawning and fish recruitment to set appropriate target exploitation rates for fisheries. Future management of exploited fisheries must also be flexible enough to deal with delayed responses to environmental perturbations transferred across scales. Successful management of marine ecosystem thus requires research traversing various disciplines as well as coordination and cooperation at national and international levels (Hofmann & Powell 1998). After decades of declining fish stocks, signs of recovery are evident in several of Namibia's marine resources but sardine biomass has yet to recover (Boyer & Hampton 2001). With regards to current management, it is still unknown what sustainable harvest levels should be in order to maintain the ecological stability and resilience of the system. (Boyer & Hampton 2001).

High fish biomass – pre-1990s

This regime was dominated by sardines up until the late 1960's. The stock started to decline due to overfishing in the late 1960's and by 1975 the stock had collapsed (Cury and Shannon 2004). The depletion of the sardine stock gave space for the stocks of other pelagic fish to grow, such as the anchovy (Engraulis capensis), horse mackerel (Trachurus trachurus capensis), cape hake (Merluccius capensis and Merluccius paradoxus) and bearded goby (Sufflogobius bibarbatus) (Hutchings et al. 2009; de Young et al. 2004; Cury & Shannon 2004). After the collapse of the sardines, fishing pressure shifted to other pelagic fish. Fluctuations in these stocks occurred throughout the 1980s (Roux and Shannon 2004). The Cape fur seal (Arctocephalus pusillus pusillus) and seabirds such as Cape Gannet (Morus capensis), Cape cormorant (Phalacocorax capensis) and the African penguin (Spheniscus demersus) are the top predators of pelagic fish; therefore populations of these predators fluctuated in tandem with the pelagic stocks (Wickens et al.1992; Boyer & Hampton 2001).

Low fish biomass – from the 1990s

The regime shifted in the 1990's and is characterized by depleted stocks of sardine and other pelagic fish (Hutchings et al. 2009).  There has been a general increase in sea surface temperature (SST) and longer hypoxic events (Monteiro & van der Plas 2006; Cury & Shannon, 2004; Pörtner & Langenbuch 2005). These conditions combined with low fish biomass have created a beneficial niche for jellyfish (Chrysaora hysoscella and Aequorea aequorea) (Cury & Shannon 2004) to increase in abundance (Kreiner et al. 2011). A recent study estimated the biomass of jellyfish in this region to be 2.4 times the amount of three of the most important commercial fish species combined (Roux et al. 2013). Seabirds and seals declined in population due to the depletion of sardine and anchovy stocks (Boyer & Hampton 2001). There is evidence that some of these populations are recovering and have stabilized, although the state remains one of overall low fish biomass, in particular low sardine biomass (Boyer & Hampton 2001).

The combined effects of overfishing and a series of large-scale environmental perturbations caused the system to shift from a high to low biomass regime (Boyer & Hampton, 2001; Cury & Shannon, 2004). The Benguela region has been a favored fishing ground for nearly a century (due to its fertile upwelling), although it wasn't until the 1960's that resource exploitation was intensified, with the arrival of large commercial fishing fleets (Boyer & Hampton 2001). Sardines were targeted as the preferred species and intensively harvested - primarily for overseas markets (Boyer & Hampton, 2001; Cury & Shannon, 2004). The pressure on this stock effected the foodweb dynamics (Cury & Shannon, 2004). Sardine stocks declined in the 1970s after which fishing was aimed primarily at anchovy in an attempt to allow sardine stocks to recover. This, however, led to a collapse of both stocks and, in the 1980s, the system became dominated by horse mackerel, bearded goby and jellyfish (de Young et al. 2004, Cury & Shannon 2004). The top-down pressure of intense fishing, in particular on sardines, likely paved the way for a regime shift to take place by reducing ecosystem resilience to withstand perturbations (Cury & Shannon 2004).

Several environmental anomalies of the 1990s acted as important drivers of this shift. These periodic bottom-up pressures are natural to the system although they were of particularly large magnitude during these years (Cury & Shannon 2004, Boyer & Hampton 2001). In 1993/1994 a low-oxygen water event from the Angolan current caused unusually extensive hypoxia in Namibian waters (Boyer & Hampton 2001; Cury & Shannon 2004). This was followed by a Benguela Niño event in 1995 (an inter-decadal climatic event) where warm, nutrient-poor water entered the system (Gammelsrød et al. as cited in Heymans et al. 2004; Cury & Shannon, 2004). The combination of these events led to poor recruitment conditions and high pelagic fish mortality resulting in a decline of most stocks (Heymans et al. 2004; Cury & Shannon, 2004; Boyer & Hampton, 2001). A general trend of warming SST in the Northern Benguela has also been identified as a potential indirect cause behind this shift although there remains some uncertainty around this, as there does around the potential contribution of other factors attributed to climate change (Bakun et al. 2010; Cury & Shannon, 2004; Hutchings et al. 2009).

Like most upwelling systems, the Northern Benguela is dominated by a small number of species that play an important role in maintaining ecosystem structure and function (Cury & Shannon 2004). Here, sardines are the dominant species structuring the ecosystem by performing a 'wasp-waist' control on species both above and below them in the foodweb (Cury & Shannon 2004). Nutrient rich, cooler waters of the upwelling provides favorable conditions for phytoplankton growth, which is controlled by sardine populations and allows for an enriched trophic web supporting productive fisheries (de Young et al. 2004; Bakun et al. 2009).  The system is vulnerable to periodic environmental variability (such as Benguela niño events) with some years allowing better recruitment than others (Boyer & Hampton 2001).

Consistent fishing, particularly on sardines, creates a top-down pressure on this system reducing its resilience to environmental variability, to the point where a series of bottom-up pulses (93/94 and 95 events) pushes the system into a new dynamic regime of depleted fish biomass (Cury & Shannon, 2004; Boyer et al. 2001; de Young et al. 2001). A key threshold, although difficult to determine in any detail, is possibly crossed when sardine stocks get so low that they were no longer able to maintain their populations (Allée effect), with repercussions throughout the whole foodweb. Fishing pressure continues during the poor recruitment years following the environmental perturbations most likely enhancing sardine collapse (Hutchings et al. 2009; Boyer et al. 2001.)

The new state of depleted fish biomass is reinforced by continued fishing pressure. Decreased fish biomass allows jellyfish to consume a greater proportion of plankton, increasing the number of jellyfish (Bakun et al. 2009). This has creates a reinforcing feedback loop whereby fish stock recovery is impeded by jellyfish competition for food as well as a possible jellyfish predation on certain fish larvae (Cury & Shannon 2004). It has been suggested that the increase of phytoplankton in the system due to less predation creates hypoxic conditions that hamper fish spawning and recruitment (Boyer et al. 2001 as cited by Cury & Shannon, 2004). These feedback mechanisms appear to lock the system in this new state, potentially difficult to reverse (Cury & Shannon 2004). There is uncertainty about the intensity and duration of low oxygen events and how these may contribute to keeping the system in this state (Hutchings et al. 2009). Warming sea temperatures may also contribute to the continuation of this state which is favorable for jellyfish that cope better in these conditions (Richardson et al. 2002). It is unknown how the apparent decline in upwelling intensity might affect this social-ecological system as well as whether or not this is an effect of anthropogenic climate change (Bakun et al. 2009).

Provisioning services decline after the shift from high biomass to low. Fish catch are plummeted, as exemplified by sardines, decreasing from 700,000 tons caught at the height of the industry to only 2,000 in 1996 (Boyer & Hampton 2001). Seal populations also decline (Roux 1998 as cited by Cury & Shannon 2004), possibly impacting the sealing industry. Jellyfish has direct negative economic impacts on the fishing industry, for example, spoiling fish catches and busting trawl nets (Lynam et al. 2006; Roux et al. 2013). The regulating service of water purification is negatively impacted via disruption of vital ecosystem processes necessary for maintaining a healthy water state. The cultural service of recreational fishing is reduced, as low fish biomass lead to fewer catches by anglers (Kirchner 1998 as cited in Boyer & Hampton 2001).

Loss of these services has unequal negative consequences on all the user groups of the resource. As a vast majority of Namibian marine catch is exported, therefore the food security of international consumers decreases (Sowman & Cardoso 2010). However, this impact is masked by the international market (Berkes et al. 2006). Additionally, local fishers and international investors lose income from decreased fish catch (Sowman & Cardoso 2010). Recreational fishers and few subsistence fishers were impacted by the shift, via decreasing access to recreational activities and food security, respectively (Sowman & Cardoso 2010).

Prior to Namibian independence in 1990 the waters of the Northern Benguela were heavily exploited by foreign fishing fleets (Roux & Shannon 2004; Boyer & Hampton 2001). Management is based on single-stock assessments with the goal of increasing commercially important stocks, such as the sardine (Boyer & Hampton 2001). Therefore focus is not on enhancing resilience of the ecosystem, but rather increasing the stocks of the socio-economically important species (Boyer et al. 2001; Boyer & Hampton 2001). When the sardine stock initially declined in the 1960s (Boyer et al. 2001) management actions were based on observations of the Southern Benguela, where sardine domination alternates with anchovy domination. By shifting fishing pressure from the sardine to the anchovies, it was believed that sardines would recover due to decreased inter-species competition (Boyer et al. 2001; Roux & Shannon 2004; Shannon et al. 2004). Unexpectedly, through this one-stock approach, resilience was undermined leading to an initial decline in anchovy, later followed by a decline in all other pelagic fish and no recovery of sardines (Boyer et al. 2001).

After independence, the fishery shifted from international to local dominance (Boyer & Hampton 2001) and the Ministry of Fisheries and Marine Resources was established in 1991 with the mission to "[...] strengthen Namibia's position as a leading fishing nation and contribute towards the achievements of [their] economic, social and conservation goals for the benefit of all Namibians" (FAO 2002). Great effort was directed at stopping illegal fishing (Oelofsen 1999 as cited by Roux & Shannon 2004) and strict control on fishery vessels and in total allowable catches was implemented (Boyer & Hampton 2001). Additionally, the Benguela Large Marine Ecosystem research programme was launched in 2002 with the aim to develop an ecosystem-wide approach to environmental research (FAO 2002). Ecosystem-based management has been proposed as a better approach to manage complex adaptive systems (Sowman & Cardoso 2010; Roux & Shannon 2004). It takes into account trophic interactions as well as environmental variations effecting fish spawning and fish recruitment to set appropriate target exploitation rates for fisheries. Future management of exploited fisheries must also be flexible enough to deal with delayed responses to environmental perturbations transferred across scales. Successful management of marine ecosystem thus requires research traversing various disciplines as well as coordination and cooperation at national and international levels (Hofmann & Powell 1998). After decades of declining fish stocks, signs of recovery are evident in several of Namibia's marine resources but sardine biomass has yet to recover (Boyer & Hampton 2001). With regards to current management, it is still unknown what sustainable harvest levels should be in order to maintain the ecological stability and resilience of the system. (Boyer & Hampton 2001).

High fish biomass – pre-1990s

This regime was dominated by sardines up until the late 1960's. The stock started to decline due to overfishing in the late 1960's and by 1975 the stock had collapsed (Cury and Shannon 2004). The depletion of the sardine stock gave space for the stocks of other pelagic fish to grow, such as the anchovy (Engraulis capensis), horse mackerel (Trachurus trachurus capensis), cape hake (Merluccius capensis and Merluccius paradoxus) and bearded goby (Sufflogobius bibarbatus) (Hutchings et al. 2009; de Young et al. 2004; Cury & Shannon 2004). After the collapse of the sardines, fishing pressure shifted to other pelagic fish. Fluctuations in these stocks occurred throughout the 1980s (Roux and Shannon 2004). The Cape fur seal (Arctocephalus pusillus pusillus) and seabirds such as Cape Gannet (Morus capensis), Cape cormorant (Phalacocorax capensis) and the African penguin (Spheniscus demersus) are the top predators of pelagic fish; therefore populations of these predators fluctuated in tandem with the pelagic stocks (Wickens et al.1992; Boyer & Hampton 2001).

Low fish biomass – from the 1990s

The regime shifted in the 1990's and is characterized by depleted stocks of sardine and other pelagic fish (Hutchings et al. 2009).  There has been a general increase in sea surface temperature (SST) and longer hypoxic events (Monteiro & van der Plas 2006; Cury & Shannon, 2004; Pörtner & Langenbuch 2005). These conditions combined with low fish biomass have created a beneficial niche for jellyfish (Chrysaora hysoscella and Aequorea aequorea) (Cury & Shannon 2004) to increase in abundance (Kreiner et al. 2011). A recent study estimated the biomass of jellyfish in this region to be 2.4 times the amount of three of the most important commercial fish species combined (Roux et al. 2013). Seabirds and seals declined in population due to the depletion of sardine and anchovy stocks (Boyer & Hampton 2001). There is evidence that some of these populations are recovering and have stabilized, although the state remains one of overall low fish biomass, in particular low sardine biomass (Boyer & Hampton 2001).

The combined effects of overfishing and a series of large-scale environmental perturbations caused the system to shift from a high to low biomass regime (Boyer & Hampton, 2001; Cury & Shannon, 2004). The Benguela region has been a favored fishing ground for nearly a century (due to its fertile upwelling), although it wasn't until the 1960's that resource exploitation was intensified, with the arrival of large commercial fishing fleets (Boyer & Hampton 2001). Sardines were targeted as the preferred species and intensively harvested - primarily for overseas markets (Boyer & Hampton, 2001; Cury & Shannon, 2004). The pressure on this stock effected the foodweb dynamics (Cury & Shannon, 2004). Sardine stocks declined in the 1970s after which fishing was aimed primarily at anchovy in an attempt to allow sardine stocks to recover. This, however, led to a collapse of both stocks and, in the 1980s, the system became dominated by horse mackerel, bearded goby and jellyfish (de Young et al. 2004, Cury & Shannon 2004). The top-down pressure of intense fishing, in particular on sardines, likely paved the way for a regime shift to take place by reducing ecosystem resilience to withstand perturbations (Cury & Shannon 2004).

Several environmental anomalies of the 1990s acted as important drivers of this shift. These periodic bottom-up pressures are natural to the system although they were of particularly large magnitude during these years (Cury & Shannon 2004, Boyer & Hampton 2001). In 1993/1994 a low-oxygen water event from the Angolan current caused unusually extensive hypoxia in Namibian waters (Boyer & Hampton 2001; Cury & Shannon 2004). This was followed by a Benguela Niño event in 1995 (an inter-decadal climatic event) where warm, nutrient-poor water entered the system (Gammelsrød et al. as cited in Heymans et al. 2004; Cury & Shannon, 2004). The combination of these events led to poor recruitment conditions and high pelagic fish mortality resulting in a decline of most stocks (Heymans et al. 2004; Cury & Shannon, 2004; Boyer & Hampton, 2001). A general trend of warming SST in the Northern Benguela has also been identified as a potential indirect cause behind this shift although there remains some uncertainty around this, as there does around the potential contribution of other factors attributed to climate change (Bakun et al. 2010; Cury & Shannon, 2004; Hutchings et al. 2009).

Like most upwelling systems, the Northern Benguela is dominated by a small number of species that play an important role in maintaining ecosystem structure and function (Cury & Shannon 2004). Here, sardines are the dominant species structuring the ecosystem by performing a 'wasp-waist' control on species both above and below them in the foodweb (Cury & Shannon 2004). Nutrient rich, cooler waters of the upwelling provides favorable conditions for phytoplankton growth, which is controlled by sardine populations and allows for an enriched trophic web supporting productive fisheries (de Young et al. 2004; Bakun et al. 2009).  The system is vulnerable to periodic environmental variability (such as Benguela niño events) with some years allowing better recruitment than others (Boyer & Hampton 2001).

Consistent fishing, particularly on sardines, creates a top-down pressure on this system reducing its resilience to environmental variability, to the point where a series of bottom-up pulses (93/94 and 95 events) pushes the system into a new dynamic regime of depleted fish biomass (Cury & Shannon, 2004; Boyer et al. 2001; de Young et al. 2001). A key threshold, although difficult to determine in any detail, is possibly crossed when sardine stocks get so low that they were no longer able to maintain their populations (Allée effect), with repercussions throughout the whole foodweb. Fishing pressure continues during the poor recruitment years following the environmental perturbations most likely enhancing sardine collapse (Hutchings et al. 2009; Boyer et al. 2001.)

The new state of depleted fish biomass is reinforced by continued fishing pressure. Decreased fish biomass allows jellyfish to consume a greater proportion of plankton, increasing the number of jellyfish (Bakun et al. 2009). This has creates a reinforcing feedback loop whereby fish stock recovery is impeded by jellyfish competition for food as well as a possible jellyfish predation on certain fish larvae (Cury & Shannon 2004). It has been suggested that the increase of phytoplankton in the system due to less predation creates hypoxic conditions that hamper fish spawning and recruitment (Boyer et al. 2001 as cited by Cury & Shannon, 2004). These feedback mechanisms appear to lock the system in this new state, potentially difficult to reverse (Cury & Shannon 2004). There is uncertainty about the intensity and duration of low oxygen events and how these may contribute to keeping the system in this state (Hutchings et al. 2009). Warming sea temperatures may also contribute to the continuation of this state which is favorable for jellyfish that cope better in these conditions (Richardson et al. 2002). It is unknown how the apparent decline in upwelling intensity might affect this social-ecological system as well as whether or not this is an effect of anthropogenic climate change (Bakun et al. 2009).

Provisioning services decline after the shift from high biomass to low. Fish catch are plummeted, as exemplified by sardines, decreasing from 700,000 tons caught at the height of the industry to only 2,000 in 1996 (Boyer & Hampton 2001). Seal populations also decline (Roux 1998 as cited by Cury & Shannon 2004), possibly impacting the sealing industry. Jellyfish has direct negative economic impacts on the fishing industry, for example, spoiling fish catches and busting trawl nets (Lynam et al. 2006; Roux et al. 2013). The regulating service of water purification is negatively impacted via disruption of vital ecosystem processes necessary for maintaining a healthy water state. The cultural service of recreational fishing is reduced, as low fish biomass lead to fewer catches by anglers (Kirchner 1998 as cited in Boyer & Hampton 2001).

Loss of these services has unequal negative consequences on all the user groups of the resource. As a vast majority of Namibian marine catch is exported, therefore the food security of international consumers decreases (Sowman & Cardoso 2010). However, this impact is masked by the international market (Berkes et al. 2006). Additionally, local fishers and international investors lose income from decreased fish catch (Sowman & Cardoso 2010). Recreational fishers and few subsistence fishers were impacted by the shift, via decreasing access to recreational activities and food security, respectively (Sowman & Cardoso 2010).

Prior to Namibian independence in 1990 the waters of the Northern Benguela were heavily exploited by foreign fishing fleets (Roux & Shannon 2004; Boyer & Hampton 2001). Management is based on single-stock assessments with the goal of increasing commercially important stocks, such as the sardine (Boyer & Hampton 2001). Therefore focus is not on enhancing resilience of the ecosystem, but rather increasing the stocks of the socio-economically important species (Boyer et al. 2001; Boyer & Hampton 2001). When the sardine stock initially declined in the 1960s (Boyer et al. 2001) management actions were based on observations of the Southern Benguela, where sardine domination alternates with anchovy domination. By shifting fishing pressure from the sardine to the anchovies, it was believed that sardines would recover due to decreased inter-species competition (Boyer et al. 2001; Roux & Shannon 2004; Shannon et al. 2004). Unexpectedly, through this one-stock approach, resilience was undermined leading to an initial decline in anchovy, later followed by a decline in all other pelagic fish and no recovery of sardines (Boyer et al. 2001).

After independence, the fishery shifted from international to local dominance (Boyer & Hampton 2001) and the Ministry of Fisheries and Marine Resources was established in 1991 with the mission to "[...] strengthen Namibia's position as a leading fishing nation and contribute towards the achievements of [their] economic, social and conservation goals for the benefit of all Namibians" (FAO 2002). Great effort was directed at stopping illegal fishing (Oelofsen 1999 as cited by Roux & Shannon 2004) and strict control on fishery vessels and in total allowable catches was implemented (Boyer & Hampton 2001). Additionally, the Benguela Large Marine Ecosystem research programme was launched in 2002 with the aim to develop an ecosystem-wide approach to environmental research (FAO 2002). Ecosystem-based management has been proposed as a better approach to manage complex adaptive systems (Sowman & Cardoso 2010; Roux & Shannon 2004). It takes into account trophic interactions as well as environmental variations effecting fish spawning and fish recruitment to set appropriate target exploitation rates for fisheries. Future management of exploited fisheries must also be flexible enough to deal with delayed responses to environmental perturbations transferred across scales. Successful management of marine ecosystem thus requires research traversing various disciplines as well as coordination and cooperation at national and international levels (Hofmann & Powell 1998). After decades of declining fish stocks, signs of recovery are evident in several of Namibia's marine resources but sardine biomass has yet to recover (Boyer & Hampton 2001). With regards to current management, it is still unknown what sustainable harvest levels should be in order to maintain the ecological stability and resilience of the system. (Boyer & Hampton 2001).