Herring Q&A

Doug Hay

Many fisheries generate controversy – including the BC herring fishery. Concerns vary in substance, severity, alleged causes and recommended remedies – but rarely are issues articulated from a biological perspective. This article considers some biological aspects of the controversies. Here are four topics that frequently are stated or implied.

1.    Herring, like salmon, “home” to spawn in the areas where they were spawned;
2.    Herring, like salmon, consist mainly of numerous small, distinct stocks or populations;
3.    Herring stocks are depleted relative to historic levels;
4.    Herring stock depletion is due to overfishing.

Common to each topic are assumptions about the nature of herring stocks – what they are, how they persist in space and time, and their genetic composition. The subject is complex, not fully resolved in the scientific literature, and still under active investigation. Until the last decade, research in BC has been at the forefront of these topics. Genetic analysis of BC herring, started in the 1990s and published in 2008, was one of the largest investigations of its kind in the world. Subsequently, other substantial studies on Pacific herring in the US (Washington State and Alaska) have generally corroborated the BC studies. New technological developments in genetic analyses show promise of further insight, but to date, none has increased understanding of BC herring stocks. Below I pose some questions and answers on some key topics.

Four rhetorical questions and answers concerning herring stocks

1. Do herring, like salmon, ‘home’ precisely to natal spawning locations?

Natal homing must involve “imprinting” – or memorization of the location where parental fish spawned. If fish do not imprint they cannot home: “memory” is learned, not inherited. Salmon detect and memorize certain key chemical constituents of water in their natal streams, which they recognize by smell (olfaction). This is well established. The developmental stage of salmonid imprinting (embryos in pre-hatch eggs, hatched alevins, parr, smolts, etc.) is still under investigation, and this may vary among species and/or locations.

How does this apply to herring? Herring eggs and larvae are much smaller (by 10-100 times), and the duration of all their early life stages, when imprinting may occur, is much shorter. Further, nearshore coastal habitats where herring spawn are dynamic, subject to tides and winds that are continuously changing water masses. Even if the imprinting process in herring were biologically identical to that of salmon, their spawning habitats are different.

Once hatched, planktonic herring larvae may drift widely, by many kilometres in a day. Therefore, in general, marine coastal habitats would not provide the same opportunities for olfactory imprinting on the natal area that occur in freshwater habitats. This is not to say that olfactory imprinting does not occur – rather if it does occur it probably is less geographically precise than that of most salmon species. An exception to this may be herring that spawn at the heads of inlets or fjords that are fed by coastal streams. In such instances the olfactory signal from freshwater input may be transmitted well into coastal waters, perhaps for many kilometres.

Herring research in BC has conducted some of the largest herring tagging investigations in the world, beginning in the 1930s and continuing, in various forms, until the last decade. These studies show that some herring move a lot – and in some regions, some herring stay put. In short, if one were to ask: do herring home back to the Salish Sea, the answer would be an emphatic yes. Do they home back to different locations within the Salish Sea? The response to this is less emphatic and must be qualified.

The response to question 1 is yes: there is evidence of homing by herring to different regions of the BC coast. Less clear is the geographical size of the “natal” areas to which, based on tagging data, herring home. The figure below shows estimates of natal homing in herring that were tagged and recovered in subsequent years on or close to spawning areas during spawning periods. When the level of homing precision or fidelity is based on large geographic areas such as one of the five major coastal regions (Georgia Strait, West Coast Vancouver Island, Central Coast, Haida Gwaii, North Coast) the estimated level of homing fidelity is high. If the same approach is used to estimate fidelity to smaller geographical groups (Statistical Areas or Sections), the estimate of homing fidelity decreases as the size of the areas decreases, with much lower fidelity to Statistical Areas (30 on the BC coast) or smaller herring Sections (about 100 on the BC coast) or Locations (more than 1000). See figure below.

2. Do BC herring consist of numerous, small herring stocks or populations?

There is general agreement from all quarters that there are a number of different populations in BC that vary in size and genetic composition. The contentious issues concern the number, size and geographic distribution of putative populations. This much is certain: in BC, from Juan de Fuca Strait to Dixon Entrance, there are some genetically distinct, late-spawning populations. In Haida Gwaii, genetically distinct herring populations occur in Masset Inlet and another in Skidegate. There are others on the mainland side: the Douglas Channel/Gardner Canal and Burke Channel, near Bella Coola (but not all have been examined genetically). There is no argument about the distinct status of these populations, which are not subject to commercial fisheries.

A once large, but late-spawning and genetically distinct, population spawn in northern Puget Sound, at Cherry Point, Washington. The status of this population was examined by the US National Marine Fisheries Service in 2006 under the Endangered Species Act – but after a detailed review, the conclusion was that while the population was “discrete” it was not exceptionally different than other herring in other parts of the Salish Sea. The recommendation was to not list this population as endangered – i.e. that it was not at risk of extinction. The purpose of mentioning this is not to endorse or refute their decision but rather point out that this independent group of federal US scientists, who had no direct management involvement with State-managed herring, chose a course that, in many ways, is much less sensitive to issues of geographical stock structure than is the system presently used by DFO.

A more difficult biological question concerns the degree to which the migratory herring are differentiated into different populations. Migratory herring move out of an area like the Strait of Georgia and then return to spawn in the same general areas in which they were spawned. For instance, are herring that spawn in the Lambert Channel part of the same biological and “demographic” stock as those herring that spawn in other parts of the Strait? From a strict genetic perspective there is no evidence of differentiation among herring that comprise the main groups of fish that are targets of fisheries. There is increasing recognition, however, that within these large groups there may be other forms of biological variation that may be recognized as sub-populations with varying degrees of “connectivity” or exchange of individuals among the groups. This is an area of active fisheries investigation for herring and other species.

The answer to question 2 is, in part, yes: there are a number of small populations. But there also are large populations, which occur in areas like the east coast of Vancouver Island and Chatham Sound. There are some coastal herring populations that are genetically distinct and that are adapted to local environments. A vital point, however, is that no genetically distinct populations occur within any of the areas subject to roe fisheries.

3. Are present herring stocks depleted relative to historic levels?

A series of workshops and reports from the UBC Fisheries Centre in 1998 and 2004 considered some of these issues. It is certain that the Strait of Georgia has undergone significant ecological change in the last several hundred years. Some marine species that once were present have either disappeared or declined significantly. Has herring abundance declined accordingly? Clearly there will be uncertainty, but several points can be raised. One is that within the last 100 years the abundance of herring in the Strait of Georgia has fluctuated, although less so than in some other parts of the coast. In the last two decades herring abundance in the Strait has reached unprecedented high levels, but herring have declined in some local areas. The most recent reports from DFO indicate that herring spawning biomass in 2015 was close to historical high levels, although spawning biomass in other areas on the coast, especially the west coast of Vancouver Island, was low.

One of the most striking changes in Strait of Georgia herring is a change in spawning distribution. In general there has been a decline in spawning on the mainland (eastern side) and an increase in spawning on the western side. Most areas with declines have never had roe fisheries, and areas with increases in spawning have generally been the sites of the largest roe fisheries. The roe fishery catches fish on the spawning grounds so it is very unlikely that it intercepts fish destined to spawn in other locations.

The answer to question three is yes and no. In some areas, especially the west coast of Vancouver Island, herring abundance is relatively low, but in other areas, such as the east coast of Vancouver Island, the abundance is close to historical highs.

4. In areas where herring have declined, is the cause of the decline overfishing?

There clearly was serious overfishing, coast-wide, of herring during the reduction fishery (1940s to the late 1960s). The spawning biomass was reduced to a point where it could not provide sufficient offspring to sustain the population, and spawning biomass plummeted in all areas.

Is there evidence of overfishing in the roe-fishery era? In part, this question must be examined at different geographical levels.

Spawning abundance has declined in some areas where roe fisheries have occurred, especially parts of the west coast of Vancouver Island and Haida Gwaii. In contrast, spawning abundance has increased in other areas with roe fisheries, such as Chatham Sound and Lambert Channel areas. The Strait of Georgia, which has had the most intense roe fisheries, has generally had the highest estimates of spawning biomass observed since record-keeping began in late 1920s.

A different question might also be posed: have roe fisheries eliminated spawning in some local areas? Some observers feel that roe fisheries in the 1970s and 1980s resulted in cessation of spawning in locations such as Ganges Harbour or the Powell River area. Records of catches and spawn exist for nearly all areas of the BC coast and can be examined at relatively small geographical units called Sections, of which there are about 100 on the BC coast.

There are many areas where no fisheries have ever occurred, and some of these show that spawning has both started and stopped in the past. There are two sections, Section 152 in the Powell River area and Section 181 called Swanson Channel, which includes Ganges Harbour, where there were controversial roe fisheries in the 1970s or 1980s. These fisheries were followed by a decline in spawning that many commentators attributed to the removal of spawning herring by the fishery. The declines in these areas occurred several years after the fisheries, and there has been spawning in these areas, sometimes considerable, in many of the subsequent years.

There are a number of Sections where roe fisheries have occurred almost every year for several decades with no indication of any decline in abundance associated with the roe fishery. On the contrary, the spawning biomass on the east coast of Vancouver Island, where most roe fisheries have occurred, has reached record highs in the last decade.

Ideally the herring roe fishery should be conducted in a way that does not reduce the spatial distribution of herring. Within the Strait of Georgia the spatial distribution of herring has declined in areas where no fisheries have ever occurred but has increased in others (between Nanaimo and Comox) where roe fisheries have concentrated for most of the last three decades.

In summary, herring abundance has increased in some Sections and decreased in others, including some that have never had fisheries.

5. Are there other explanations for changes in herring?

The BC coastal environment is changing. Physical changes in coastal waters include changes in temperature, pH, and in timing and volume of river-discharge. In the last few decades herring size-at-age has decreased, in areas both with and without fisheries. In the last five years, this trend has reversed in all areas. Concurrently we see increasing changes associated with residential and industrial development.

Other biota has changed. Coho and chinook salmon have declined in the last few decades, as have several species of forage fish (surf smelts, fall-spawning capelin, eulachon). Concurrently harbour seals have increased substantially – to the point where there is more than one seal per 100 meters of shoreline in the Strait of Georgia. Energy-rich herring are prime prey species for seals. Seal predation may be important in explaining the decline of herring on the mainland side of Strait of Georgia. Humpback whales, which are prodigious predators of herring in Prince William Sound, have increased in recent years and may present new challenges to continuation of herring abundance in some areas. Similarly, grey whales are fond of fresh herring eggs and could impact herring in some areas. The Pacific hake and sardine, both of which are subject to natural, long- term fluctuations, may be formidable competitors of herring in offshore waters, and hake also may prey on herring.

The consuming preoccupation with alleged or presumed deleterious impacts of the fishery, even if valid, has caused all of us to overlook the impacts of climate change and especially the continuous and egregious impacts of industrial pollution, shoreline disruption and inadequate sewage treatment and discharge. It is my view that, within the Salish Sea, such impacts can explain much of the change in herring distribution and we ignore such change at our peril.


After gaining his PhD in fish ecology at UBC, Doug Hay worked at universities and the Pacific Biological Station (Nanaimo) 1977-2005 where he is now a scientist emeritus, writing research reports/books on herring and eulachon.

Related Posts

Watershed Sentinel Original Content

5 Issues/yr — $25 print; $15 digital