Introduction

The collapse of California’s sardine fishery in the 1950s remains one of the most consequential fisheries failures in North America. By the end of that decade, canneries had closed, and many coastal communities experienced lasting economic disruption. Similar ecological and management pressures have since re-emerged in the California Current Large Marine Ecosystem (CCLME), a highly productive coastal upwelling system extending from British Columbia to Baja California. This region supports large populations of small, schooling forage fish that occupy a central role in both fisheries and food webs (Cury et al., 2000).

Pacific Sardine (Sardinops sagax), Northern Anchovy (Engraulis mordax), and Pacific Herring (Clupea pallasii) are key components of this system. These species support commercial fisheries and serve as essential prey for salmon, seabirds, and marine mammals. Because they occupy a mid-trophic position, changes in their abundance can ripple rapidly through the ecosystem (Cury et al., 2000). 

Management concern intensified when NOAA Fisheries declared the northern subpopulation of Pacific Sardine overfished in 2019. Under the Coastal Pelagic Species Fishery Management Plan (CPS FMP), the directed sardine fishery has been closed since July 1, 2015, as estimated age-1+ biomass has remained below the threshold that permits directed harvest. At the same time, limited harvests have continued under annual specifications in specific sectors, such as live bait and incidental catch, including allowances set for the 2024-2025 fishing year. These conditions raise a broader question about how harvest levels should be set to sustain fisheries while maintaining adequate prey availability for dependent predators.

This paper examines how forage fish harvest levels in the California Current can be managed to balance fisheries production with ecosystem needs. It reviews the relevant scientific literature and management framework, synthesizes implications for predator populations, and evaluates ecosystem-based approaches that extend beyond traditional single-species reference points.

Statement of Issue to be Addressed

Setting harvest levels for forage fish in the California Current presents a central management challenge because these species are simultaneously commercially valuable and ecologically indispensable. Three species dominate this discussion: Pacific Sardine, Northern Anchovy, and Pacific Herring, although other coastal pelagic species (e.g., mackerels, market squid) contribute to predator diets and fisheries. These small pelagic fishes are characterized by rapid growth, high fecundity, and relatively short lifespans, resulting in large natural fluctuations in abundance that are strongly influenced by environmental variability (MacCall, 1990). 

The ecological importance of forage fish in upwelling systems has been recognized for decades. Early ecosystem syntheses emphasized that small pelagic fishes occupy a narrow mid-trophic “waist” that constrains energy transfer from plankton to higher trophic levels, including fishes, seabirds, and marine mammals. Because of this structural role, Cury et al. (2000) cautioned that continued expansion of pelagic fisheries in upwelling systems is unlikely to be sustainable without major ecosystem disruption. This perspective highlights an inherent management challenge: forage fish harvest decisions influence not only fishery yield, but also the stability of entire food webs.

The relevance of forage fish availability became especially clear during the 2013-2016 California Sea Lion Unusual Mortality Event (UME). During this event, NOAA Fisheries documented 8,122 stranded juvenile California Sea Lions, most of which were alive but in poor body condition, consistent with nutritional stress (National Marine Fisheries Service, 2022). Peer-reviewed analyses concluded that prey limitation contributed to pup malnutrition and to elevated strandings and mortality during this period (McClatchie et al., 2016). 

Predator responses were not limited to pinnipeds; multiple seabird species exhibited diet shifts, reduced prey quality, and increased energetic stress during the same general period, conditions that reduce breeding performance and indicate broader ecosystem stress (McClatchie et al., 2016; Warzybok et al., 2018).

These predator impacts coincided with a steep decline in Pacific Sardine biomass. For example, the 2017 stock assessment estimated age‑1+ biomass at 86,586 metric tons, below the CPS FMP cutoff of 150,000 metric tons that permits directed harvest (Hill et al., 2017). Earlier assessments indicate that sardine biomass was substantially higher in the mid‑2000s, exceeding one million metric tons in U.S. management areas, illustrating the magnitude of the subsequent decline (Hill, 2006).

The economic stakes are substantial. Koehn et al. (2017), citing PFMC (2014) and PacFIN (2014), reported that from 2004 to 2013, the Pacific Sardine fishery averaged approximately $13.7 million in annual ex-vessel revenue, while Northern Anchovy averaged about $1 million annually. However, forage fish deliver considerable indirect value by supporting predators that underpin other commercial fisheries (notably salmon) and species of conservation concern (e.g., Marbled Murrelets, Humpback Whales, and multiple seabirds). In practice, management must account for both direct fishery benefits and the ecosystem services provided by forage fish.

The Pacific Fishery Management Council (PFMC), established under the Magnuson-Stevens Fishery Conservation and Management Act (MSA) in 1976, holds primary authority for managing Coastal Pelagic Species (CPS) stocks in U.S. waters. The CPS FMP governs sardine, anchovy, mackerel, squid, and krill. Following the 2019 determination that Pacific Sardine was overfished, the PFMC and NOAA Fisheries implemented a rebuilding plan (Amendment 18) with a rebuilding target of 150,000 metric tons of age-1+ biomass (Pacific Fishery Management Council, 2023). Subsequent litigation (Oceana v. Raimondo) concluded that aspects of the rebuilding plan’s scientific support, particularly reliance on the CalCOFI temperature index in the management framework, were insufficient, prompting the development of CPS FMP Amendment 23 to revise the rebuilding plan.

Finally, Pacific Sardine distributions extend into Mexican and Canadian waters, and the stock has historically been managed without a comprehensive coastwide harvest allocation agreement across all jurisdictions. This creates coordination challenges when ocean conditions shift distribution and productivity. Effective forage fish management, therefore, requires tools that are robust to environmental variability, that incorporate predator needs, and that are resilient to multi-jurisdictional dynamics.

State of the Science

Forage fish life history and ecosystem role

Pacific Sardine, Northern Anchovy, and Pacific Herring are small pelagic schooling fishes with fast growth and high fecundity (MacCall, 1990). Their ecological importance stems from their position in “wasp-waist” ecosystems, where a small number of mid-trophic species mediate energy flow from plankton to higher trophic levels (Cury et al., 2000). These forage species feed primarily on zooplankton and are prey for predators such as salmon, groundfish, seabirds, and marine mammals protected under the Marine Mammal Protection Act (MMPA).

Natural variability, regimes, and long-term cycles

Multiple lines of evidence demonstrate that sardine and anchovy abundance are naturally variable over interannual to multidecadal timescales. Baumgartner et al. (1992) used fish scales preserved in anoxic sediments (e.g., Santa Barbara Basin) to reconstruct sardine and anchovy abundance over millennia, documenting repeated booms and busts long before modern commercial fishing. These patterns are associated with oceanographic regime shifts and illustrate that climate variability can dominate population trajectories. Chavez et al. (2003) described multidecadal alternation between anchovy- and sardine-dominated regimes in the Pacific, often with inverse abundance patterns. This observation is frequently interpreted as “compensation” (one species increases when the other declines), but more recent work cautions that such compensation is neither immediate nor guaranteed. Hinchliffe et al. (2025) documented long-term trends for Northern Anchovy and emphasized that recent ecosystem conditions can produce outcomes that do not perfectly mirror historical alternations, particularly under novel climate forcing.

Recruitment drivers and environmental controls

Environmental conditions strongly influence recruitment success for sardine and anchovy. Jacobson and MacCall (1995) linked variability in sardine recruitment to factors such as sea surface temperature, upwelling intensity, and large-scale climate variability, with recruitment generally higher under warmer oceanographic regimes. In contrast, Northern Anchovy recruitment and biomass have historically been higher during cooler regimes, although recent analyses indicate that this relationship is not stationary and that anchovy populations can decline even under low fishing pressure when environmental conditions are unfavorable (Hinchliffe et al., 2025). Consequently, the predictive power of any single environmental index is limited under changing baseline conditions and evolving ecosystem interactions.

The use of the CalCOFI temperature index in sardine management illustrates this problem. The sardine harvest control rule historically incorporated temperature to set the harvest fraction (Zwolinski & Demer, 2012). More recently, a federal court concluded that the rebuilding plan did not provide sufficient scientific support for aspects of the temperature-based approach as applied, leading to revision of the rebuilding framework through Amendment 23.

Biomass estimation and harvest control rules

NOAA’s Southwest Fisheries Science Center and PFMC stock assessment processes use multiple data sources (including acoustic-trawl surveys and modeling) to estimate sardine and anchovy biomass. Under the CPS FMP, directed sardine harvest is prohibited when estimated age‑1+ biomass falls below the cutoff of 150,000 metric tons. For the 2017-2018 management season, the estimated age‑1+ sardine biomass (86,586 mt) was below the cutoff of 150,000 mt, leading to continued directed fishery closure (Hill et al., 2017). Subsequent assessments indicated that sardine biomass declined further and later fell below the minimum stock size threshold (50,000 mt), which formed the basis for the 2019 overfished declaration.

Importantly, management closure of the directed fishery does not necessarily imply zero sardine removals. Annual specifications may authorize limited catch in specific sectors (e.g., live-bait and incidental-catch caps). For the 2024-2025 fishing year, NOAA Fisheries specified allowances (including an annual catch target) and trip/landing limits to constrain these non-directed pathways.

Predator demand and nutritional quality

A critical scientific question is how much forage biomass must remain in the ecosystem to support predator fitness and population stability. Warzybok et al. (2018) estimated that seabirds breeding in the central California Current consume greater than 60,000 metric tons of forage fish annually, with Common Murres, Brandt’s Cormorants, and Rhinoceros Auklets among the dominant consumers.

Predator impacts are influenced not only by forage quantity but also by forage quality. Synthesis of predator diet data from the California Current demonstrates strong reliance on a relatively small subset of forage taxa, including sardine, anchovy, herring, euphausiids, and market squid, with substantial spatial and temporal heterogeneity in prey use (Szoboszlai et al., 2015). Numerous studies indicate that sardine and anchovy are among the more energy-dense forage fishes available to predators, whereas alternative prey such as market squid, euphausiids, and juvenile fishes are often lower in energetic value, depending on species, size, season, and lipid content. When high-quality forage is scarce, predators may attempt prey switching; however, energetic constraints, spatial accessibility, and temporal mismatch can limit effective compensation, particularly for central-place foragers such as breeding seabirds and lactating pinnipeds that must provision offspring (McClatchie et al., 2016; Warzybok et al., 2018).

McClatchie et al. (2016) documented that both sardine and anchovy declined simultaneously in the Southern California Bight during 2009-2014, while lower-energy-density prey such as market squid and juvenile rockfish increased, indicating that prey switching occurred but involved nutritionally inferior alternatives.

Synthesis of Information

Three patterns from the literature challenge conventional single-species harvest management for forage fish in the California Current:

1) Fishery-protection thresholds do not necessarily protect predators. 

Current CPS harvest rules are designed primarily to reduce the risk of stock collapse and to promote long-term fishery yield by closing the directed fishery when age‑1+ biomass falls below 150,000 mt. However, predator impacts can occur at higher biomass levels because predator performance depends on local prey availability, prey quality, spatial overlap, and seasonal timing. The 2013-2016 Sea Lion UME occurred during a broader period of forage scarcity and altered prey availability; the resulting nutritional stress indicates that ecosystem impacts can emerge before single-species thresholds are triggered (McClatchie et al., 2016).

This pattern is consistent with ecosystem modeling results that explicitly quantify trade-offs between forage fish fisheries and their predators in the California Current. Using a multispecies food-web framework, Koehn et al. (2017) demonstrated that fishing forage species often increases direct fishery yield while simultaneously causing substantial declines in biomass of nonmarket predators, particularly seabirds and marine mammals that depend strongly on forage fish. Importantly, these predator losses frequently occurred even at harvest levels considered sustainable from a single-species perspective, reinforcing the conclusion that fishery-protection thresholds alone are insufficient to safeguard predator populations.

2) “Compensation” among forage species is not reliable at predator-relevant scales.

Sardine-anchovy alternation is well documented over long timescales (Chavez et al., 2003), yet compensation may fail in a given region or season, and both species can be simultaneously low under unfavorable conditions. Even when an alternative forage species becomes available, it may not be accessible to predators at the right time or place, and nutritional differences among prey can limit substitution. Consequently, management strategies that assume predators will seamlessly switch among forage species underestimate risk during low-biomass periods.

3) Climate variability and marine heatwaves increase management uncertainty. 

The 2014-2016 marine heatwave highlighted how prolonged warm anomalies can suppress recruitment, shift distributions, and alter predator-prey overlap. Climate change is expected to increase the frequency, duration, and intensity of marine heatwaves, amplifying instability in recruitment–environment relationships and increasing the probability of mismatches between management assumptions and ecological outcomes (Frölicher et al., 2018).

These patterns also expose a limitation of conventional stock assessment frameworks that treat natural mortality as a fixed, background parameter. For forage fish, a substantial portion of natural mortality reflects predation by seabirds, marine mammals, and other fishes, many of which are protected under the Marine Mammal Protection Act or the Endangered Species Act. When forage availability declines, predator populations can experience food limitation even when forage stocks remain above fishery reference points established under the Magnuson-Stevens Act. 

This creates management tension because fisheries may be considered compliant with the Magnuson-Stevens Act while simultaneously undermining obligations under the Marine Mammal Protection Act and the Endangered Species Act, which prohibit population-level impacts on protected predators caused by prey limitation. For forage fish, harvest decisions therefore operate at the intersection of multiple federal conservation mandates rather than within a single-species framework. As a result, management actions that satisfy single-species fishery objectives may still conflict with broader federal conservation mandates if prey limitation undermines protected predator populations.

Global syntheses reinforce the need for precaution in forage fish management. The Lenfest Forage Fish Task Force concluded that forage fish fisheries should leave substantially more biomass in the ecosystem than conventional single-species approaches would suggest, often on the order of twice as much, to protect ecosystem functions and predator populations (Pikitch et al., 2012).

Climate-driven changes also alter interactions among mid-trophic species, further complicating forage fish dynamics. Pacific Hake (Merluccius productus) is a dominant semi-pelagic predator in the California Current whose abundance, recruitment, and spatial distribution are shaped by environmental and ecological processes. Vestfals et al. (2023) demonstrated that hake recruitment and population distribution are structured by stage-specific physical and ecological drivers, including mesoscale ocean dynamics, alongshore transport, storm frequency, and prey-field interactions, rather than by temperature alone.

Importantly, Vestfals et al. (2023) found evidence of a negative relationship between Pacific Herring abundance and hake recruitment mediated through adult female preconditioning. The authors hypothesized that elevated herring biomass on hake summer feeding grounds may reduce euphausiid availability, leading to poorer adult female condition prior to spawning and, consequently, lower recruitment. Thus, forage fish abundance can suppress hake recruitment under some conditions through indirect effects on adult condition rather than through direct impacts on early life stages (Vestfals et al., 2023).

Taken together, these findings support a broader conceptual model of bidirectional interactions between mid-trophic predators and forage fishes, although such interactions are inferred from multiple lines of evidence rather than demonstrated directly in a single study. More broadly, periods of elevated hake abundance, which often coincide with favorable environmental conditions documented elsewhere in the California Current, may increase predation pressure on juvenile forage fishes. During marine heatwaves and other anomalous conditions, these interactions may intensify, adding non-fishery sources of variability and mortality to forage fish populations.

Forage availability to predators and fisheries reflects the combined effects of climate variability, fishing, and interspecific interactions rather than harvest alone. In the California Current, this implies that management focused solely on single-species reference points may be insufficient to reduce ecosystem-level risk, particularly as climate variability increases.

Recommendations for Fisheries Science and Management

1) Establish predator-informed ecosystem reference points. 

The PFMC and NOAA Fisheries should develop ecosystem reference points explicitly tied to predators' nutritional requirements and reproductive success, rather than solely to avoiding stock collapse. This can be implemented by integrating predator bioenergetics models with empirical indicators of predator performance (e.g., seabird breeding success, pinniped pup condition) into harvest advice. Management should include explicit buffers that reduce allowed removals during periods of low forage availability, reduced prey quality, or elevated environmental stress. While uncertainty in predator demand estimates is unavoidable, failing to explicitly incorporate predator needs risks systematic underprotection of dependent species during periods of low forage availability.

2) Implement multi-species, aggregate forage management triggers. 

Management should better reflect that predators depend on the total available forage, not on single-species abundance in isolation. An aggregate framework could set a combined cap on removals across key forage taxa within the CPS complex (or a defined forage subset), with automatic reductions when any primary forage stock is depressed or when predator indicators signal prey limitation. Such rules would reduce the risk of shifting fishing pressure onto alternative forage species during periods when predators are already constrained. Ecosystem modeling indicates that managing forage species independently or allowing fishing pressure to shift sequentially among species can amplify predator impacts, whereas coordinated, multi-species constraints reduce the risk of disproportionate biomass losses to seabirds and marine mammals (Koehn et al., 2017).

Empirical evidence from the California Current shows that when sardine and anchovy decline, predators do not simply disengage from the forage base. Instead, they shift toward alternative prey such as juvenile rockfish and market squid, which are generally lower in energy density and often less accessible in space or time (McClatchie et al., 2016; Warzybok et al., 2018). Allowing fishing pressure to sequentially shift across forage species therefore risks progressively eroding the prey base available to predators rather than relieving ecosystem stress. Aggregate forage management is necessary to prevent serial depletion of prey resources that predators also rely upon.

3) Use spatial and temporal protections near predator breeding and nursing sites. 

Spatially explicit management, such as seasonal buffers around major seabird colonies and pinniped rookeries during breeding and nursing, can reduce localized competition between fisheries and central-place foragers. These measures are most defensible when targeted to predictable windows and locations, and when paired with monitoring that evaluates predator response and fishery impacts. Moreover, long-standing ecological theory predicts that competition between commercial fisheries and marine mammal predators will intensify over time, most commonly through fisheries depleting localized prey resources at scales relevant to predator foraging, even when target fish stocks are not formally overfished (DeMaster et al., 2001).

Conclusion

The California Current has experienced repeated periods of sardine scarcity, including a mid‑twentieth-century collapse and a recent decline that contributed to ecosystem stress. While the CPS FMP includes precautionary harvest control rules for Pacific Sardine, the 2013-2016 Sea Lion UME and concurrent seabird breeding failures indicate that ecosystem impacts can occur before single-species reference points trigger strong management action. Three overarching conclusions follow: (1) directed-fishery closures based on stock thresholds do not guarantee predator protection; (2) compensation among forage species is unreliable at predator-relevant spatial and temporal scales; and (3) climate variability and marine heatwaves increase uncertainty and strengthen the case for precaution.

Sustaining both fisheries and predator populations will require ecosystem-based management that incorporates predator-informed reference points, multi-species safeguards, and spatial-temporal protections during critical predator life stages. These reforms are increasingly urgent as climate change amplifies environmental variability in the California Current.

References

Baumgartner, T. R., Soutar, A., & Ferreira-Bartrina, V. (1992). Reconstruction of the history of 

Pacific Sardine and Northern Anchovy populations over the past two millennia from sediments of the Santa Barbara Basin, California. CalCOFI Reports, 33, 24-40.

Chavez, F. P., Ryan, J., Lluch-Cota, S. E., & Ñiquen, M. (2003). From anchovies to sardines and 

back: Multidecadal change in the Pacific Ocean. Science, 299(5604), 217-221.

Cury, P., Bakun, A., Crawford, R. J. M., Jarre, A., Quiñones, R. A., Shannon, L. J., & Verheye, 

H. M. (2000). Small pelagics in upwelling systems: Patterns of interaction and structural changes in “wasp-waist” ecosystems. ICES Journal of Marine Science, 57(3), 603–618.

DeMaster, D. P., Fowler, C. W., Perry, S. L., & Richlen, M. F. (2001). Predation and competition: 

The impact of fisheries on marine-mammal populations over the next one hundred years. Journal of Mammalogy, 82(3), 641–651.

Frölicher, T. L., Fischer, E. M., & Gruber, N. (2018). Marine heatwaves under global warming. 

Nature, 560(7718), 360–364.

Hill, K. T. (2006). Assessment of the Pacific Sardine (Sardinops sagax caerulea) population for 

U.S. management in 2006 (NOAA Technical Memorandum NMFS-SWFSC-385). U.S. Department of Commerce.

Hill, K. T., Crone, P. R., Dorval, E., & Macewicz, B. J. (2017). Assessment of the Pacific Sardine 

resource in 2017 for U.S. management in 2017-2018 (NOAA Technical Memorandum NMFS-SWFSC-577). U.S. Department of Commerce.

Hinchliffe, C., Kuriyama, P. T., Punt, A. E., Field, J. C., Thompson, A. R., Santora, J. A., 

Muhling, B. A., Koenigstein, S., Hernvann, P.-Y., & Tommasi, D. (2025). Long-term population trend of Northern Anchovy (Engraulis mordax) in the California Current system. ICES Journal of Marine Science, 82(1).

Jacobson, L. D., & MacCall, A. D. (1995). Stock-recruitment models for Pacific Sardine 

(Sardinops sagax). Canadian Journal of Fisheries and Aquatic Sciences, 52(3), 566-577.

Koehn, L. E., Essington, T. E., Marshall, K. N., Sydeman, W. J., Szoboszlai, A. I., & Thayer, J. 

A. (2017). Trade-offs between forage fish fisheries and their predators in the California Current. ICES Journal of Marine Science, 74(9), 2448-2458.

MacCall, A. D. (1990). Dynamic geography of marine fish populations. University of 

Washington Press.

McClatchie, S., Field, J., Thompson, A., Gerrodette, T., Lowry, M., Fiedler, P., Watson, W., 

Nieto, K., & Vetter, R. (2016). Food limitation of Sea Lion pups and the decline of forage off southern California. Royal Society Open Science, 3(1), 150628.

National Marine Fisheries Service. (2022). Frequent questions: 2013-2016 California Sea Lion 

unusual mortality event. NOAA Fisheries.

Pacific Fishery Management Council. (2023). Coastal Pelagic Species Fishery Management 

Plan for U.S. West Coast Fisheries: Pacific sardine, northern anchovy, Pacific mackerel, jack mackerel, market squid, and krill. Pacific Fishery Management Council.

Pikitch, E. K., Boersma, P. D., Boyd, I. L., Conover, D. O., Cury, P., Essington, T. E., Heppell, S. 

S., Houde, E. D., Mangel, M., Pauly, D., Plagányi, É., Sainsbury, K., & Steneck, R. S. (2012). Little fish, big impact: Managing a crucial link in ocean food webs. Lenfest Ocean Program.

Szoboszlai, A. I., Thayer, J. A., Wood, S. A., Sydeman, W. J., & Koehn, L. E. (2015). Forage 

species in predator diets: Synthesis of data from the California Current. Ecological Informatics, 29, 45-56.

Vestfals, C. D., Marshall, K. N., Tolimieri, N., Hunsicker, M. E., Berger, A. M., Taylor, I. G., 

Jacox, M. G., & Turley, B. D. (2023). Stage-specific drivers of Pacific hake (Merluccius productus) recruitment in the California Current Ecosystem. Fisheries Oceanography, 32(4), 352–389.

Warzybok, P., Santora, J. A., Ainley, D. G., Bradley, R. W., Field, J. C., Capitolo, P. J., Carle, R. 

D., Elliott, M., Beck, J. N., McChesney, G. J., Hester, M. M., & Jahncke, J. (2018). Prey switching and consumption by seabirds in the central California Current upwelling ecosystem: Implications for forage fish management. Journal of Marine Systems, 185, 25-39.

Zwolinski, J. P., & Demer, D. A. (2012). A cold oceanographic regime with high exploitation 

rates in the Northeast Pacific forecasts a collapse of the sardine stock. Proceedings of the National Academy of Sciences, 109(11), 4175-4180.