From the Western Cape to the Eastern Cape and Namibia, reports of dead sardines washing ashore have transformed what began as a localised mortality event into a widening concern, with scientists unable to determine how much of South Africa's sardine stock has been lost.
The department of forestry, fisheries and the environment (DFFE) has confirmed the presence of pilchard herpesvirus (PHV) genetic material in sardines sampled during the Western Cape mortality event.
PHV is a highly contagious virus affecting sardines, aka pilchards, and remains the leading explanation for the deaths. But whether the virus, which can impair fish respiration by damaging the gills, is responsible for the more recent mortalities further along the coast is being established.
After no new mass-mortality incidents were reported between 16 and 26 August, isolated sightings at Bakoven and off Tsitsikamma were followed on 27 August by widespread reports of sardines washing ashore around Gqeberha and floating offshore and inside the harbour.
Sardine strandings have also been reported along the Namibian coast, with authorities agreeing to provide samples for PHV testing. The department is considering a possible link between the Western Cape and Gqeberha events.
The department has clarified that it has not estimated that the mortality represents 1% of the sardine population, despite the figure having entered public discussion.
The 1% figure, department spokesperson Zolile Nqayi said, arose during confidential, preliminary scientific deliberations in the department's small pelagic scientific working group and "was not approved for public release or presented as a departmental estimate".
The calculation applied a surface density recorded near Dyer Island across the West Coast, producing a quantity of about 1% of biomass measured earlier this year. Nqayi said that was a crude illustration, not an estimate of total mortality. "No comprehensive seabed assessment has been conducted and DFFE has not claimed that the true mortality is only 1%. An unknown proportion of fish may have sunk, washed ashore or remained undetected."
Craig Smith, a senior technical specialist for marine at WWF South Africa, said reports of dead sardines being caught in hake trawl nets and live and dead sardines accumulating in the Koeberg cooling-water system suggested that a significant proportion of the mortality could be occurring below the surface.
"We don't know the magnitude of the impact," he said.
"The department has provided a rudimentary estimate of 1% of the biomass that has succumbed but we are mindful that that is probably a gross underestimate because that is estimating what has been washed up. We don't understand how much of that sardine has sunk to the bottom of the ocean."
The concern over PHV is rooted in two major sardine mortality events in South Australia in 1995 and 1998. A 2001 study found that spawning biomass fell by more than 75% after the 1995 event and by more than 70% after the 1998 outbreak.
The deaths were associated with PHV. The outbreaks occurred after large-scale tuna farming began in South Australia, where untreated imported frozen sardines were used as feed for caged southern bluefin tuna. Researchers proposed infected imported fish as a possible pathway, although the virus's original source was never established.
Nqayi said the department was consolidating sightings, transect counts, photographs, videos, fishing information and observer records to assess whether the impact would be low, moderate or high. Its pelagic scientific working group would meet on Friday to review the latest evidence and consider the implications for fisheries management.
Dr Alistair McInnes, the seabird conservation programme manager at BirdLife South Africa, said the mass sardine die-off could be significant for endangered African penguins.
"If the sardine stock is reduced by similar magnitudes to what transpired in the Australian sardine stock in the 1990s, the outcome will not be good for African penguins who rely heavily on this stock especially during their non-breeding season in spring and summer."
McInnes said research showed that penguin mortality rose when sardine abundance west of Cape Agulhas fell. "When the sardine abundance west of Cape Agulhas is below 25% of the long-term sardine biomass maximum, penguin mortalities increase substantially — this is a realistic scenario if 70% of the stock is lost."
The concern, he said, was compounded by record-low levels of young anchovy entering the fish population, leaving penguins with fewer alternative sources of food. Anchovy and sardine were the most important prey for the iconic seabirds.
BirdLife's automated monitoring systems have provided an early indication of differing conditions among penguin colonies. Raw data showed improved breeding adult body condition at Robben Island compared with the previous two years but poor body condition at Dyer Island.
"Since the mortality event, the condition of birds at Dyer Island has deteriorated further and birds at the adjacent Stony Point colony have also shown signs of deterioration," McInnes said. "There is no indication of a deterioration in body condition from birds at Robben Island and we are hoping that the majority of chicks fledged before the sardine mortality event at this site."
The department's October/November hydroacoustic survey is expected to provide the first reliable post-event estimate. The timetable is at the heart of a debate over whether fishing should continue while the mortality remains uncertain.
WWF-SA is calling for an immediate moratorium on small pelagic fishing, which targets sardine and anchovy, until the remaining sardine biomass can be assessed.
Smith said the total allowable catch (TAC) was based on pre-outbreak surveys. "You can't just continue fishing this stock. Remember, the total allowable catches have already been determined for this season based on surveys that were done last year. But now that latest information is not going to hold true because you've got this major viral outbreak and you need to understand that this is an additional impact on mortality to the stock that hasn't been taken into account when the TAC was first formulated."
Nqayi said the department has not ruled out additional management measures but no decision has been taken on a moratorium. He said the scientific working group would consider on Friday whether the available evidence warranted emergency or other management action.
He said the Marine Living Resources Act required a precautionary approach to fisheries management, while stressing that it did not prescribe one automatic response.
Smith said a moratorium would give sardines a chance to recover rather than adding fishing pressure to a stressed population. Sardines are also a major forage fish for predators throughout the marine ecosystem.
The small pelagic industry, however, is urging caution.
Mike Copeland, the chairperson of the South African Pelagic Fishing Industry Association (Sapfia), said the industry was extremely concerned about the mortality and its potential effects on the ecosystem.
The industry regarded the hydroacoustic survey as imperative, he said. "If the mortality is found to have significantly reduced the available sardine biomass, any significant reduction in the sardine biomass will result in a significant reduction of total allowable catch in the directed sardine sector," Copeland said.
He said that would be severe for rights holders, boat owners and crews, processors, workers, marketing and selling organisations, communities and suppliers, as well as having implications for food security.
Sapfia, however, does not support the proposed moratorium. Copeland said the proposal, covering the directed sardine fishery and sardine caught as by-catch, would effectively close the entire small pelagic sector.
Sapfia instead supported what he described as the "urgent structured and evidence based approach" being taken by the minister and department to investigate the event.
"We remain committed to participate in this process and to be part of the team that evaluates the situation and decides on the actions required."
The outbreak has also raised questions about how PHV reached South African waters. Genetic sequencing of sampled sardines produced a 99.7% match to a PHV gene sequence from Australia.
But Nqayi warned that the genetic match does not establish the virus's origin or route into South Africa.
Smith said a pathway could be imported frozen sardines, pointing to the Australian experience where contaminated frozen sardines used as feed in bluefin tuna farming were believed to have been the source.
South Africa should investigate whether similar pathways existed locally, including imported sardines used by fish-processing establishments and sardines imported as bait.
Nqayi said there was no evidence that imported frozen sardines had introduced PHV. But the department has obtained samples of imported sardines from a processor for testing and might extend the screening to imported bait products.
Finding PHV in an imported sardine would not, on its own, prove imports introduced the virus into South African waters. Nqayi said wider testing, including of sardines caught before the mortality event, would be needed to establish whether PHV was introduced or was present in local sardines.
Smith said the West Coast sardine stock was considered overexploited and warned that the virus-driven mortality could compound pressures on the small pelagic ecosystem.
"You can't be fishing a resource to such low levels that makes them vulnerable to collapse."