End of a Dynasty: The Loss of the King Salmon

The river canyons, where the old bars were located, were romantic places previous to being disturbed and torn up by the gold-digger. The water was as clear as crystal, and above each ripple or rapid place was a long, deep pool, with water as blue as turquoise, swarming with fish. Salmon at that time ran up all the streams as far as they could get, until some perpendicular barrier which they could not leap prevented further progress.
Angel M. History of Placer County, California. 1882.

The Central Valley Chinook Salmon (Oncorhynchus tshawytscha), also known as the King Salmon, is one of the most iconic species of the Bay Area. In four seasonal runs, the Chinook salmon used to slam the rivers of the Central Valley Basin to spawn in the cold, well-oxygenated water draining from the Cascades and the Sierra Nevada. Every winter, spring, fall, and late-fall these anadramous fish would return from the Pacific Ocean, charging through the San Francisco Bay in numbers reaching the hundreds of thousands. As the largest of all salmon species, with adults often exceeding 40 pounds, the Chinook are important for commercial and recreational anglers alike, supporting the billion dollar California fishing industry. Since the era of California dam building, however, the Chinook populations have plummeted. All four runs have been listed under the Federal and State Endangered Species Acts. Even with hatchery stocks augmenting the wild population, the commercial salmon fishing season had to be closed this year for the second year in a row in order to protect the once robust fall run. The staggering decline of the Chinook spells both ecological and economic ruin for Northern California. Last year's closure of the salmon fishing season amounted to a loss of $255 million and 2,263 jobs.

While some of the salmon runs were nearly extirpated by dam installation in Central Valley Rivers, the unique life history of the fall run allowed it to persist as the backbone of the West Coast fishery for many years. The fall run spends less time in rivers than the other runs, so it is less impacted by land based activities. Also, it tends to spawn downstream of the other runs, so its passage is not blocked by the upstream dams. In recent history, the fall run population had been fairly stable, even experiencing a huge increase in 2000. Despite the large numbers of fish returning to spawn, the fall Chinook run was not as strong as it appeared. In fact, it was on the verge of an unexpected collapse. The fall run experienced record low returns in 2007, prompting the closure of the fishing season in 2008. Even when the fishery was closed, only 66,000 salmon returned to spawn.

Fisheries biologists believed that most, if not all, of the fall salmon returning to spawn came from the hatchery stock. Salmon hatcheries were originally established as part of a compromise between the fishing industry and California energy interests to sustain salmon populations that would be hindered by dam construction. However, the hatcheries have done more harm than good. The addition of a hatchery stock creates the illusion of a robust wild population, even though the wild population is no longer self-sustaining. In fact, the hatchery stocks have reduced the vitality and resilience of the wild population.

But how did the fall run, augmented by hatchery-borne fish, collapse? At the 2009 State of the Estuary Conference, Steve Lindley of NOAA explained how fisheries biologists were able to solve this mystery. Knowing the salmon’s age of sexual maturity, fish that should have been returning to spawn in 2007 and 2008 are from the broods that hatched in 2004 and 2005. Spawning of the wild population and the hatchery population was normal in those years, but survival rates between the broods’ departure from the estuary and their return to spawn were terrible. Therefore the secret to the broods’ failure lies in the environmental conditions they encountered in their three years at sea.

In 2005 and 2006 the water off the California coast was unusually warm. The spring upwelling, which is usually activated by shifting winds to deliver cold, nutrient-rich waters to the surface, had a late start in 2006. As a result, the Chinook salmons’ typical prey species struggled, while anchovies, a fish too large for the young salmon to eat, thrived. The Farallon Islands population of the Cassin’s Auklet, a seabird with the same diet as the Chinook salmon, completely failed to reproduce in 2005. The struggle of the Cassin’s Auklet population should have been a warning sign for salmon returns. Without a reliable food source, the 2004 and 2005 salmon broods suffered in the ocean, unbeknownst to biologists and fishermen until their poor returns in subsequent years.

Even though Delta water management and land-based activities impact Chinook salmon populations indirectly, the proximate cause of the fall run’s most recent collapse was simply a lack of adequate prey. Poor feeing conditions, often driven by natural climate variability, are in no way unprecedented, but the Chinook salmon was not resilient enough to overcome this challenge. Wild fish populations maintain their resilience by having a high degree of biocomplexity, described by Steve Lindley as a portfolio of environmental conditions, such as habitat and prey, which can satisfy the animals’ requirements. If the different runs, including those differentiated by spawning site as well as time, exhibit a diversity of life history patterns, the population as a whole is not intensely impacted by a shift in environmental conditions. In the case of the Central Valley Chinook, however, hatchery stocks have diluted the biocomplexity of the population, so that all the animals exhibit similar life history patterns. Whereas historically there was variability across the runs, today the reliance on hatchery production has pruned back the life history diversity of the Chinook salmon, making the population more prone to booms and busts.

Unfortunately the management strategy of the Central Valley Chinook salmon must rely on hatchery stocks to produce numbers of salmon great enough to sustain the fishing industry. When the salmon returns are poor, managers increase hatchery production, causing overall fitness of the population to decline. But even when the returns are great enough to reopen the fishing season, the apparent strength of the population is merely an illusion. When environmental conditions are favorable, the Chinook population appears to thrive, but the success of the hatchery stock makes it nearly impossible to detect inevitable declines in the wild population. When environmental conditions shift again, the homogeneous population will be vulnerable to another devastating crash.

Decades of hatchery-dependent management of the Chinook has produced a population that is unreliable for the fishing industry and deleterious to the recovery of wild stocks. Even if the fishing season is re-opened for 2010, the idea that a large number of salmon reflects the strength of the population is a dangerous misconception. To ensure long-term viability of the Central Valley Chinook as a species, something must change in order to improve wild salmon production. While there are management options, including brood stock selection and more variability in the timing of hatchery stock release, the best way to improve the fitness of the wild stock is to eliminate the hatcheries entirely. Without hatcheries, however, the wild population may never return in numbers great enough to harvest. Even though the small wild population could regain some of its biocomplexity and vitality, salmon fishing would have to end altogether, an admission of the failure of hatchery-dependent management. It is likely that Californians will have to mourn the end of their famed Chinook fishery in order to celebrate the survival of the species.

Photo: Late-fall Chinook salmon spawning, courtesy of USFWS

Lindley, Steve. 2009. The Once and Future Kings. Presented at the State of the Estuary Conference. Oakland, CA.

Yoshiyama, Ronald M., Eric R. Gerstung, Frank W. Fisher, and Peter B. Moyle. Historical and Present Distribution of Chinook Salmon in the Central Valley Drainage of California. Contributions to the Biology of Central Valley Salmonids. Fish Bulletin 179: Volume One.

The Woods and Your Health

The opportunity to hike beneath the orange and red foliage of a deciduous forest in autumn is one of the greatest perks of living in New England. I spent a fair amount of time in the wooded margins of the Colby College campus, drawn out of bed at sunrise and enduring chilly mornings on the cusp of winter to collect field data. While studying the relative distribution and abundance of small mammals, I had several opportunities to walk beneath the fall foliage, wrapped in silence except for the sounds of our footsteps and our research subjects scurrying through discarded leaves. Despite the obvious ecological transition into dormancy, the New England woodland in autumn conjures feelings of freshness and vitality rather than decay. However, beneath the duff on the forest floor and within the variegated canopy lurks an invisible threat. Secretly, this serene setting could be a hotbed of disease.

Richard Preston’s The Hot Zone, painted a thrilling picture of an exotic disease festering in a remote jungle until it aggressively emerged into a vulnerable human population through an unconfirmed, but likely encounter with a reclusive animal. A more temperate, developed setting, however, is certainly not disease-free. Extremely virulent pathogens can lurk in your own backyard in natural reservoir species as common as the white-footed mouse. Zoonotic diseases, or those that can be transferred to humans from other vertebrate animals, are so common that the majority of the diseases that infect humans today originated in other animals. According to the CDC, approximately 75% of recently emerged infectious diseases, including West Nile virus, Eastern Equine Encephalitis, and Lyme disease, are zoonotic. The natural reservoir of these diseases is an animal that acts as a long-term host for the pathogen and for which an infection is non-lethal. The natural reservoir therefore acts as a continual source of a pathogen, which would otherwise eliminate itself by killing hosts too quickly to perpetuate the infection.

The emergence of new zoonotic pathogens in human populations has risen in recent years due to increased contact between humans and wildlife. As humans push back the margins of animal habitat for settlement and exploit animal resources through hunting and husbandry, we increase our contact with animals that may harbor pathogens, thereby increasing our exposure to zoonotic disease. Zoonotic pathogens do not require humans to complete their life cycle; under natural conditions the pathogen would stay within the animal population. If exposed to a disease reservoir or vector, however, humans can become an incidental host. Habitat destruction and resource exploitation increases our risk of infection by potentially deadly pathogens simply by accident.

My research partner and I were warned about disease exposure as we emptied Sherman traps, studying the relative abundance of the white-footed mouse (Peromyscus leucopus) and the deer mouse (P. maniculatus) in the margins of the woodland surrounding Colby College. Worldwide, rodents are natural reservoirs for over 35 diseases. The pathogens that cause these diseases can be spread to humans either directly through contact with saliva, urine, and feces; or indirectly via insect vectors. Lyme disease is a well-known and well-publicized disease that is caused by bacteria harbored in a mouse reservoir but spread by ticks as vectors. When summer ends and ticks overwinter on the forest floor, humans can still fall victim to the potentially deadly hantavirus pulmonary syndrome. Hantavirus is spread to humans by direct contact with the feces and urine of an infected mouse, causing renal failure and fatal hemorrhagic fever. While the symptoms sound like something straight out of The Hot Zone, there are no monkeys, fruit bats, or protective suits involved. Hantavirus can be contracted by inhaling aerosolized waste of common and abundant rodents, and hundreds of cases have been reported in the United States, especially in the western states.

While disease was not the primary focus of our research, it had very clear epidemiological implications. As humans encroach upon animal habitat, we upset the population balance of the ecological community by eliminating cover, introducing invasive species, and reducing food availability. Some species, such as the white-footed mouse, are generalists that can thrive in highly-disturbed, marginal habitat. Their populations tend to explode when specialist species populations decline. Therefore, as we push into deciduous and mixed forests, the white-footed mouse may benefit under the altered ecological conditions and increase in numbers, but overall biodiversity collapses. Incidentally, the white-footed mouse is a reservoir for both Lyme disease and hantavirus.

Zoonotic diseases are caused by pathogens that will readily infect a variety of different host species, but for which there are very few natural reservoir species. This means that only a few species can survive and spread the infection, thereby assisting the long-term survival of the pathogen. A high density of a natural reservoir species in a community would increase the presence of the pathogen and also increase the likelihood that the disease will be passed to another host or vector species. In contrast, high biodiversity would reduce the density of the natural reservoir species and increase the number of incompetent reservoirs that cannot spread the disease. It has therefore been hypothesized that biodiversity can prevent the spread of zoonotic disease and in this way protect human health.

In a ground-breaking 2000 paper in the journal Conservation Biology, Richard Ostfeld and Felicia Keesing applied this hypothesis to the spread of Lyme disease, calling it the “dilution effect.” They predicted that high species diversity within the host community would increase the likelihood that ticks would feed on animals that do not harbor the Lyme disease bacteria. As a result the infection power of the natural reservoir, the white-footed mouse, would be diluted. With a lower prevalence of infection in ticks, the vectors of the disease, Ostfeld and Keesing hypothesized that high species diversity would also reduce transmission of Lyme disease to humans. Sure enough, through analysis of state level ecological and epidemiological data, they found that states with a greater number of small mammal species in the host communities had fewer reported cases of Lyme disease per capita.

The work of Ostfeld and Keesling demonstrates that the dilution effect can apply to indirect infection through an insect vector. A recently published study in the journal PLoS One, however, took their theory one step further. While studying the relationship between small mammal biodiversity and the prevalence of hantavirus in Panama, scientists from the Museum of Southwestern Biology and the University of New Mexico found that the dilution effect holds up for direct infections as well. They theorized that when species diversity in a host community is reduced and the density of the natural reservoir species increases, infected reservoirs will have more encounters with uninfected reservoirs and therefore more opportunities to transmit the virus. In contrast, high species diversity would reduce the encounter rate and similarly reduce the transmission rate. After setting up experimental field plots, the researchers reduced species diversity within some of the plots by removing the non-reservoir species. When the number of species was reduced, both the density of the natural reservoir population and the prevalence of hantavirus among those individuals increased. From this finding one could hypothesize that high biodiversity would consequently reduce the rate of direct infection to humans who are exposed to the reservoir species.

The idea that biodiversity can offer a service to human health is not a new concept. In fact, when people mobilize to save large swaths of highly diverse habitat from destruction, human health is one of the most argued cases for its preservation. However, people usually argue that within these areas of high biodiversity may be un-studied organisms with compounds for potential pharmaceuticals or that can act as laboratory models. While certainly compelling, this case incentivizes the preservation of certain species of potential value, not necessarily biodiversity. The ability of a diverse community to dilute the presence of a zoonotic pathogen and reduce human exposure to some of the deadliest diseases is an even stronger correlation between high biodiversity and improved human health.



Ostfeld, Richard S. and Felicia Keesing. 2000. Biodiversity and Disease Risk: the Case of Lyme Disease. Conservation Biology. 14(3):722-278

Suzan, Gerardo, Erika Marce, J. Tomasz Giermakowski, James N. Mills, Gerardo Ceballos, Richard S. Ostfeld, Blas Armien, Juan M. Pascale, Terry L. Yates. 2009. Experimental Evidence for Reduced Rodent Diversity Causing Increased Hantavirus Prevalence. PLoS One. 4(5)

Photo: Peromyscus leucopus; Encyclopedia of Life