Sierra Nature Notes, Volume 9, February 2010
Why do the Salmon Continue to Disappear from Our Mother Lode Rivers?
By Craig A. Will
WHERE HAVE ALL THE SALMON GONE?
Fewer Chinook salmon returning from the ocean to spawn in the rivers means that fishing will be closed for two successive years. What happened, and what role might Tuolumne and Calaveras counties play in restoring the salmon fishery?
In the fall of 2007, fish biologists counting the number of Chinook salmon returning from the ocean to spawn via the Sacramento River got an unwelcome surprise. Only 87,881 salmon were found in the river – way below the predicted 265,000, and below the minimum goal of 122,000 considered necessary for conserving the fishery. A year later, counts were even lower, with only 66,286 spawners found – the lowest number ever recorded. Numbers were also far less than hoped-for in the tributaries of the San Joaquin River, including the Stanislaus and Tuolumne Rivers. On the Tuolumne River, 211 were counted in 2007 – the lowest count ever, and far below the count of 18,000 in the year 2000.
These numbers sent shock waves through the California coastal fishing fleet,
from Humboldt County to San Diego. The Chinook salmon fishery that fishermen
depended upon for their livelihood had collapsed.
Federal regulators had no choice. For the first time in history, they closed both commercial and recreational ocean fishing for Chinook salmon in all of California and most of Oregon. The ban, put in place last year, will continue this year as well.
The effects of the closure have been devastating to salmon fishermen and to small fishing communities up and down the coast.
As a result of the closure, more than 2,200 fishermen and other workers in
the fishing industry have lost their jobs. Losses to fishing and related industries
have been estimated at $250 million to $300 million, with an even greater indirect
Chinook salmon – also known as “king salmon” – is perhaps the choicest salmon to eat, with a flavor described as rich and buttery. And it’s the real thing. Its deep reddish-pink color is the result of krill (a small crustacean similar to shrimp) in its diet – not food dye. And it has a high omega-3 fat content – good for your heart – unlike the vast majority of salmon now eaten in the United States, which are raised in fish farms.
At Fisherman’s Wharf in San Francisco – where in the outer lagoon, away from the tourists, there are still working fishing boats – the closure was the first blow in a one-two punch. The punch was to an industry based on two icons of the Northern California coast – the king salmon and the Dungeness crab. The second blow was the worst crab season in 40 years.
How could this icon be in such danger?
Salmon Before the White Men Came
The Chinook is the largest salmon, on average 24 pounds, although they often grow to 40 pounds and some over 100 pounds have been seen. They are known for the bright silver color of their sides, and have a blue-green back.
Historically, salmon in the Sacramento and San Joaquin Rivers and their tributaries were abundant and far ranging. Although the San Joaquin River (and its tributaries) now provides less than 10 percent of the total number of salmon that move through the Sacramento-San Joaquin Delta, that proportion was once far higher. Salmon were once so abundant in the San Joaquin that farmers would harvest the fish for hog feed.
Salmon return from the ocean to rivers and streams to “spawn”, or breed. They typically return to the precise location they were hatched at. It is believed that they use the smell of minerals and organic materials to do this. Some will breed directly in the large rivers – the San Joaquin and Sacramento. Others will go far up a tributary river and perhaps eventually up a tiny stream.
On the Stanislaus River, salmon once went up its North Fork as far up as McKay’s Point – just below what is now Big Trees State Park, east of Arnold. They went up the Middle Fork as high as what is now Beardsley Reservoir (north of Pinecrest) – and may have gone higher.
On the Tuolumne River, salmon went up as far as Preston Falls – near what
is now the boundary to Yosemite National Park.
Salmon return in characteristic “runs” that take place in a particular season, at least partially the result of genetic influence. The Central Valley Chinooks have historically had four runs – a fall run, a late-fall run, a winter run, and a spring run. The runs are named for the time that the salmon return to the breeding ground, and in most cases they will breed immediately. However, in some cases – particularly the spring run –the salmon will hang around in the river for a few months before breeding.
It’s an arduous journey. Salmon do not eat while on the trip, and some die because they haven’t stored up enough body fat to make it. Even those that do aren’t in very good shape – gaunt, sometimes described as having “grotesquely humped backs” with “large white patches of bruised skin”.
Even in the days of a pristine environment, salmon had to contend with rapids and waterfalls, and predators such as eagles, bears, and river otters. Many didn’t make it.
“Anyone who sees a spawning salmon – and who understands the countless barriers it has avoided to reach the breeding ground – can only marvel at its persistence,” said John Buckley, executive director of the Central Sierra Environmental Resource Center, in an interview.
Once back in the spawning grounds, salmon will battle each other, sometimes
tearing off fins. Females will battle other females for the best location
to make a nest, and males will battle each other over available females.
A female will be carrying a pound or more of eggs, and will seek out a place to build a nest, known as a redd. Preferred places are relatively large gravel, with little silt – which has a negative effect on egg fertility. Depth of water, temperature, and water velocity are also considerations. Using her tail and fins, a female will excavate a shallow depression in the gravel, and lay eggs. A male salmon will deposit sperm, or milt, over the eggs. The female will then cover the eggs with gravel.
A single redd may contain five thousand eggs and cover about thirty square feet. The female will then move upstream to create another redd, typically making four or five redds before exhausting her supply of eggs.
The males will die within a few days of spawning. Females will guard their
eggs for a few weeks, but then also die.
Salmon carry many important nutrients from the ocean, and some environmentalists claim that salmon spawning is not only important to keep the salmon population intact, but also to maintain other species that depend on the nutrients from salmon carcasses.
The eggs hatch after 90 to 150 days, depending on the temperature. Eggs hatch into fry, sometimes known as sac fry because they feed on the yolk sac still attached to them. They may stay for several weeks in the gravel before swimming into the stream, feeding on terrestrial and aquatic insects.
Most Chinooks spend 3-6 months developing in freshwater, then begin migrating to the sea. As they begin to enter salt water, physiological changes take place. The vertical bars and spots that they had developed as juveniles for use as camouflage are lost, and their back becomes dark and their belly light. Their gills and kidneys change to accommodate seawater.
Once in the ocean, they remain quite near the shore for several months, feeding largely on smaller fish and crustaceans, including krill. When they get larger, they feed mostly on small fish. Chinook salmon will stay for 1 to 4 years in the ocean – most stay 3 years – traveling thousands of miles, until, if they survive, they begin the long migration back to their stream of origin.
Mining, Diversions, Dams, Pollution, Invasions, and Pumps
The threat to the salmon began in the Gold Rush. Gold during that period was primarily found in the streams and rivers, and miners dug dirt and gravel out of these waters to wash it for gold. This removed spawning gravel and covered other gravel with silt when the dirt was dumped back into the streams.
Water was also needed to wash dirt that looked promising but wasn’t in a stream, and so streams were diverted to provide the water. This reduced water in streams and typically also disturbed spawning grounds. Such diversions also confused juvenile fish that might follow the new stream to the gold-washing facility.
Especially damaging was the development of hydraulic mining, in which a stream of water under high pressure was used to excavate dirt and gravel into sluice boxes, where the gold could be removed. The practice could wash away entire hillsides within a matter of days. Hydraulic mining destroyed gravel beds in streams and rivers and deposited silt in gravel beds downstream. It also caused diversions of water. The practice continued for decades until it was effectively outlawed as the result of a court case.
As the Gold Rush wound down and new emigrants turned to farming, the diversion of water for irrigation of crops began to be a significant threat. Again, the problems were the reduction of water in the streams and rivers, the destruction of gravel beds, and the luring of juvenile fish down paths that they would not survive.
Mining and water diversions were significant enough that the California Fish Commission said in 1886, “The Tuolumne, a branch of the San Joaquin, at one time was one of the best salmon streams in the state. Salmon have not ascended the stream for some years.”
Perhaps the biggest threat to the salmon, however, was the construction of dams. On the Tuolumne River, La Grange Dam was completed in 1893 and has effectively blocked any spawning above it since. Don Pedro Dam, above La Grange, was completed in 1923 and rebuilt in 1971 to raise its height to 580 feet.
On the Stanislaus River, Goodwin Dam, a small diversion dam (intended to
divert the direction of the river rather than store water), was completed
in 1913. Although Goodwin Dam initially had fish ladders, unreliable water
flow above the dam apparently prevented fish from spawning above the dam,
and the fish ladders were eventually removed. Melones Dam was completed in
1926, and the Tri-Dam project was built in the 1950s, which included dams
at Tulloch, Beardsley, and Donnell. Still another dam, New Melones Dam, was
completed in 1979, increasing the storage capacity over that of Melones Dam.
The lack of interest in fish ladders for these dams resulted from a belief that there were still large numbers of salmon spawning in the river below the lowest dam, the reality that destruction of the habitat had already prevented most spawning above the dams, and the cost and practicality of the ladders.
Juvenile fish that hatched in the rivers below the dams still had to migrate down the San Joaquin River through the delta to get to San Francisco Bay and the ocean. Beginning during the Gold Rush, however, the delta began to be transformed. The wetlands were drained and made into a series of islands where farmers could grow crops. Irrigation channels were built to provide water for the crops.
As the soil decomposed and the islands sunk – many are now ten feet or
more below sea level – levees were built to protect them from flooding.
There are now 1,100 miles of levees and about 95% of the original marsh is
As one fishery biologist told the San Francisco Chronicle: “It was a huge marsh, habitat for all of the runs. Now its been diked, levied and rip-rapped until it’s not more than a big ditch.”
Much of the water is polluted, either with pesticides and herbicides from
the farms or by ammonia from human sewage.
Invasive species have also become a problem in the delta. Largemouth bass, among others – introduced more than a century ago for recreational fishing – are predators for juvenile salmon, who now have few cattails and bulrushes to hide in – the traditional hiding places.
The delta has also become the hub for water supply in California. Although historically composed of brackish water, it has been transformed to fresh water so that it can supply a vast aqueduct system that supplies farmers in the southern Central Valley and urban areas in Southern California with water. About 25 million Californians rely on the delta for part or all of their water.
Huge pumps in the southern part of the delta – near Tracy – suck
water from the delta. Operators of the pumps look through buckets for fish
that have been sucked into the pumps and truck those still alive back to
During periods of drought, like we have today, large-scale pumping of water from the delta can reduce the water the juvenile salmon need and increase their mortality.
A UC Davis fisheries professor, Peter Moyle, recently did a study indicating that 20 of the 31 species of salmon and trout in California are at risk of extinction. His comment to the Sacramento Bee: “Maybe we should be surprised the salmon are here at all.”
Fish hatcheries – in which the spawning and rearing of juvenile salmon are done artificially – began more than a century ago in response to the various insults to salmon habitat, particularly dams.A typical hatchery is located on a tributary river or stream and has mechanisms that entrap fish and move them into holding tanks until they are ready for spawning. Eggs are taken from a female fish and mixed with milt (sperm) from a male fish, and the resulting fertilized eggs kept in an incubator.
After the eggs hatch, the fry are kept in incubators for a time and then transferred into concrete-lined pools, known as raceways. A typical size for a raceway is 100 feet long and 10 feet wide. Raceways have high chain-link fences and netting above to protect the young fish from predatory birds such as hawks and herons who might otherwise swoop down and grab them. A hatchery will typically release some of its young fish into the nearby river when they are fry so that they will eventually return and their eggs and sperm can be used. Other fish are raised to be fingerling or even yearling (one-year-old fish) size and then are moved in trucks with oxygenated and temperature-controlled tanks to be released in the delta or San Francisco Bay.
There are two hatcheries on tributaries of the San Joaquin River – one
on the Merced River and one on the Mokeulumne. There are four more hatcheries
on tributaries of the Sacramento River.
Hatcheries were initially viewed as simply a way of increasing the population
of salmon in the ocean for fishermen and as having little or no effect on
natural breeding and the ecosystem. Their goal was to produce as many young
fish as possible that would survive as adults in the ocean.
One perspective on this is provided by a Fort Bragg fisherman, who told the San Francisco Chronicle that hatcheries were supposed to allow fishing to continue even after the dams were built. “Fishermen are ecologists. We want things right, and we don’t want to catch the last fish. But we also want a chance to make a living at what we’ve done all our lives. They knew the wild fish wouldn’t survive when they put up the dams, but they promised us that we would have something to fish.”
The view of hatcheries has changed in recent years, as various concerns began to be raised. One concern is that hatchery fish who “stray” – return to a location other than the hatchery to spawn – will interbreed with the natural population. Over time, this would reduce the genetic variation in the wild population, reducing their ability to cope with the particular ecosystem that they have adapted to.
The practice of “outplanting” – releasing young salmon in the delta or San Francisco Bay, rather than the river next to the hatchery – has particularly been criticized, mainly because it results in straying. Fish who have not have the experience of swimming down the streams and rivers from where they were hatched to the ocean have no sense of where “home” is and are likely to go anywhere. Releasing young salmon in the bay may also be undesirable because these fish compete with naturally-bred fish for what might be limited food or habitat in the bay and ocean, but are unlikely to return and spawn in normal ways.
A study from University of California, Santa Cruz researchers in 2008 caused
surprise when it showed that 90% of the Chinook salmon caught by fishermen
in the ocean were from hatcheries. It had long been assumed that the majority
of salmon were spawned naturally. The UCSC study analyzed the ear bones of
salmon to determine whether the fish was from a hatchery or natural. Ear
bones, like tree rings, show patterns of growth, and can distinguish between
a naturally spawned fry that will slow its growth after the food in its yolk
sac runs out and a hatchery fry that is provided with abundant food.
This study has heightened the concern about the impact of hatcheries, since the count of 66,286 fish returning late last year to the Sacramento River means that only 6629 of them were likely non-hatchery fish.
The collapse of 2007 and 2008 was a huge shock to everyone. What happened? Newspaper reports of the collapse frequently used the phrase that has now become a cliche: it must have been a “perfect storm” of all of the factors affecting the salmon going wrong at once.
We now know that this wasn’t the case.
In a March 2009 report to the Pacific Fishery Management Council – which
regulates the fishery – fish biologists painstakingly pieced together the
available evidence. Salmon returning in 2007 – the first year of the collapse–were
mostly in “brood year 2004” – hatched sometime between November
2004 and March 2005. Conditions were favorable in the gravel natural breeding
grounds then – temperatures were “optimal” and water flows
sufficient. In the hatcheries, nothing was unusual.
There was also nothing unusual in the rivers that would prevent the growth of salmon fry into smolts – salmon that were ready for their migration to the ocean. And the fact that smolts did come out of the streams and rivers into and through the delta is confirmed by counts made at Chipps Island – a location near where the delta connects with San Francisco Bay. Counts showed a slightly above average number of smolts. As for hatchery fish, the state hatcheries released a normal number of smolts into San Francisco Bay, and there were no unusual environmental conditions in the bay.
The conclusion, then, is that a normal number of brood year 2004 salmon made it safely into the ocean. There, something caused most to not survive. Analysis of coded hatchery tags from 2-year-old fish caught by recreational fishermen showed that the number of such fish was way down – only 8 percent of the number of salmon from the relatively abundant brood year 2000. And the return for spawning of “jack” salmon – an unusual group that mature very rapidly and spawn at the age of one year – was at a record low level, far below numbers for previous years.
Most young salmon from brood year 2004 arrived in the ocean in May of 2005. And in that month winds and temperatures in the north Pacific were quite unusual. Typically northwest winds push surface water away from land, causing it to be replaced by colder water from the deep that is rich with nutrients, a process known as “upwelling”. These nutrients – including plankton and small crustaceans that young fish feed on – are critical for their survival.
In May of 2005, however, winds tended to be southwest rather than northwest,
which prevented upwelling. In addition, water temperatures at the surface
were unusually high. This causes an increased layering of waters with different
densities – and
These unusual wind and temperature conditions upset a delicate balance and
prevented the young salmon from feeding – and the vast majority of
them either starved or were made much more susceptible to predators.
The same thing happened to seabirds nesting on the Farallon Islands. Birds known as Cassin auklets abandoned nearly all of their nests, apparently because of lack of food. These birds have a diet similar to that of young salmon.
The situation for salmon in the following year – brood year 2005 – was essentially the same. Eggs hatched into fry and developed successfully into smolts, which proceeded into an ocean that had the same unfavorable conditions – again causing starvation conditions.
Why Did It Happen?
Global warming may have had much to do with the particular ocean conditions that caused the young salmon to starve. And if so, it is likely to happen again. But ultimately, the crash of the salmon fishery was caused as much by a brittle, excessively simple system for breeding and rearing as it was by ocean conditions.
“Simple” – in our complicated world – is often regarded as desirable and refreshing. In this case it’s not. The natural, historic world of salmon was a place with many different populations that were each adapted to particular conditions. Some of this is genetic, some created by the particular environment a salmon grows up in. In the particular case at hand timing was critical – the vast majority of salmon for a particular spawning year emerged in the ocean within a narrow window of time – a particularly bad time. Moreover, they were also pretty much the same size. If the salmon had been spread out wider in time rather than the narrow window, however, far more would likely have survived. And if some of the salmon had been bigger when they had emerged into the ocean, they would have had a better chance of survival.
One way in which the salmon world can be diverse is in the different runs. Historically, Central Valley Chinooks have had fall, late-fall, winter, and spring runs. Today, only the fall run has a substantial number of fish, with two of the other runs so tiny that they are listed under the Endangered Species Act. Another way in which the salmon world can be diverse is in its different habitat for rearing. However, nearly half of the nearly 1600 miles of stream that was historically available for salmon breeding and rearing has been lost. And it’s been lost in systematic ways, so that particular kinds of habitat has been lost more than others, with higher-elevation, cooler water habitat being lost almost completely. Genetics is still another way in which salmon can offer diversity – one that the heavy use of hatcheries has probably compromised.
What can be done? The quickest way to get results is probably to make some
changes to the way hatcheries operate.
One change is to rear different batches of salmon to be deliberately different, rather than the longstanding practice of standardization. This might be done, for example, by giving different batches different amounts of food – so that different batches would migrate to the ocean at different times.
Another change might be an overall reduction in the number of hatchery fish bred – so as to reduce their influence on the natural population. The number might even be adjusted to match predicted ocean conditions – breeding and releasing more juvenile salmon if survival rates were expected to be low.
An often-discussed change is the reduction or elimination of “outplanting” – releasing young salmon in the Delta or Bay – so as to reduce strays and prevent the interbreeding and excessive influence the hatchery fish have on the genetics of naturally spawning salmon.
What will happen next? We do know a little about what will happen in the
fall run of 2009. We know that “brood year 2006” salmon were raised during
drought conditions in the delta, with little water coming in but normal pumping
of water out. Counts of juveniles passing Chipps Island in 2007 were about half
of those in the two previous years. Ocean conditions were better, but the count
of the early-maturing “jacks” was still the second-lowest return
on record. Even with a ban on fishing, the number of salmon returning for spawning
in 2009 is expected to be only 122,196 – just above the minimum goal
of 122,000 for sustaining the fishery.
Bring Back the Spring Run
The idea that particular ocean conditions for a single month can kill most of the juvenile salmon for an entire season – and do it two years in a row – is a new and frightening idea. If it is linked to global warming, it is especially scary, because it may mean that as the climate heats up the extinction of the salmon is simply a matter of time.
There are many other threats to the salmon. Competition for water in the delta between the fish and the pumps supplying farms and urban areas is particularly an issue, especially given the current drought. The delta, indeed, is universally acknowledged to be a mess that everyone – environmentalists, water interests, and politicians – agrees has to be fixed. If plans to build a new large-scale water project to move water past the delta is approved, part of the money will be used to restore the delta, including fish habitat. But the diversion of water will still have a big negative impact on the fish.
With the Stimulus Bill providing large sums of money for public works projects, and the Obama Administration supporting science both intellectually and financially in a way that the previous administration did not, there will likely be considerable new work on improving salmon habitat. Recent congressional approval of the plan to restore the San Joaquin River also includes funding for such work.
However, the fact that ocean conditions – maybe related to global warming – make the salmon’s survival so sensitive to their time of arrival in the ocean is a new idea. Changes to the delta – providing more places for juvenile fish to hide, feed, and grow – can help spread out their time of arrival. But there is another approach. Historically, the spring run in the Stanislaus and Tuolumne Rivers was likely the largest run in the Central Valley salmon fishery, exceeding the fall run at its levels then and now. And that run has different timing than the fall run – indeed, studies have shown that the window of time during which salmon in the spring run enter the ocean is the largest of any run. At present, the strategy is to restore salmon habitat in the rivers and delta only below the dams – Goodwin Dam and LaGrange Dam.
But could the spring run of Chinook salmon in the foothills and mountains be reestablished? The fish exist – there are still small spring runs both in the Sacramento River and the San Joaquin. We are close to Silicon Valley, a hotbed of innovation and technology. With clever planning and design, we could perhaps do better than has been done in the past in reestablishing salmon in rivers that have lost their fish. Fish ladders are sometimes impractical or are very expensive, but computer-controlled fish elevators or devices not yet invented might work just fine and be relatively cheap. Using electronic sensors (to sense water levels, temperature, and the presence of salmon), wireless communication, and appropriate channels and valves in the river, it might be feasible to provide habitat for salmon to migrate upstream for spawning and for fry to develop and migrate downstream using relatively little water.
Could this really be done? How much would it cost? Who would pay for it? These are the big questions. But if practical, it would go far to help save – and even expand – a $250 million salmon industry, its thousands of direct jobs, and the additional jobs that depend on them. It would provide jobs in Tuolumne and Calaveras counties in helping the restoration effort. It would even attract tourists, who might like to see a salmon climb up a waterfall on the Tuolumne River. Not to mention a few locals who might enjoy the same sight.
This article first appeared in the April 24th, 2009 issue of the Sierra Mountain Times.
Further ReadingLowest delta-rivers salmon run ever in 2009
and Game Releases Flawed Report on Fish-stocking Impacts,
Fails to Propose Measures to Protect Native Fish,
Center for Biological Diversity press release
Notes is now entirely volunteer and self-supporting. Please
help us stay online and consider a tax-deductable and secure donation
Questions? Go to About Our New Site