The Scottish Fish Farm Production Survey 2024 has dominated aquaculture headlines and for good reason. It’s the government’s official record of how the country’s salmon farms are performing, covering everything from smolt numbers and harvest volumes to employment, escapes, and survival.
But one number has outshined the rest: 61.8 percent survival.
That means nearly four out of ten fish never made it to harvest, the lowest rate in over three decades. Depending on who you ask, that figure either proves that Scottish salmon farming is thriving despite the odds or that it’s fundamentally broken. The truth is neither. It’s not about who’s right, it’s about how we’re looking at the data. Because salmon farming doesn’t run on quarterly cycles. It runs on biological timelines measured in years, not months.
So what does the long-term data actually show? Have we really gotten worse at farming? And why, after decades of progress, aren’t we seeing continuous improvement?
The Fish Farm Production Survey is one of the longest-running continuous datasets in aquaculture. Each year, it compiles data from every licensed salmon producer in Scotland. It tracks smolt transfers, ova laid, harvest volumes, employment, escapes, welfare, and survival. Because it’s been running for decades, it’s more than a snapshot. It’s the chronological record of an entire industry learning how to farm better, through new diseases, evolving regulations, and changing environments. This year’s edition focused public attention on one figure: 61.8% survival for the 2022 year-class, the lowest in 35 years. But the value of the report isn’t in that single number; it’s in the long arc behind it.
It’s not a verdict. It’s a logbook documenting how an industry learns, adapts, and corrects over time.
This SalmonBusiness article captured the divide perfectly. Tavish Scott, CEO of Salmon Scotland, described the numbers as evidence of “another successful period,” pointing to 192,000 tonnes of production, a 27 percent increase, and nearly £1 billion in exports.
“These figures reflect another successful period for our sector, demonstrating the hard work of our farmers and the growing demand for our highly nutritious fish,” he said.
Welfare groups, however, highlighted individual farms reporting 30–40 percent monthly mortality, describing it as proof of a system still under biological stress.
The irony? Both sides are right. They're just drawing from different time slices.
Salmon Scotland’s argument looks at six months of improvement; campaigners focus on the worst individual months. Both are technically true, but biologically incomplete. Salmon don’t live on six-month cycles. They live on eighteen-month ones.
Biology moves in waves. Short-term calm doesn’t erase long-term volatility.
A salmon’s seawater phase lasts roughly 540 days, about a year and a half. Mortality is cumulative. Each percentage lost each month stacks on the last. So when a report highlights “99 percent survival,” that’s only a monthly view. Multiply that out across 18 months, and it can easily add up to 18 percent mortality or more. A farm can look stable for months yet still end its cycle 25 percent below expected harvest volume. That doesn’t make the farm bad, it just means biology doesn’t follow spreadsheets.
To see what’s really happening, we need to zoom out. The chart below shows 18-month cumulative mortality rates across Scottish farms from 2018 through September 2025.
From 2018–2020, mortality stayed fairly stable, around 23–27 percent. By 2021, the curve began to rise. Losses peaked above 40 percent in 2023 and early 2024, before falling sharply back to 23 percent by mid-2025. That’s a full biological recovery cycle — a spike, a plateau, and a return to baseline. The “99 percent survival” headlines capture the tail end of that curve, not the valley that came before it.
At first glance, yes. But when you look closer, also, no.
Survival rates today are lower than in the 2010s. But farmers haven’t suddenly forgotten how to farm. The reality is that both the biology and the rules have changed. If we look back through the government’s mortality records (which stretch all the way to the 1980s) a clear pattern emerges.
In 1988–1989, average mortality hovered around 40 percent. By the mid-1990s, it dropped to around 10–12 percent — the best performance ever recorded. But by 1998, it had risen again and stabilized between 20–25 percent through the 2000s and 2010s, before climbing again in the early 2020s. That isn’t steady decline — it’s adaptation. Each decade reflects new pressures, new tools, and new definitions of success.
Each major change in performance aligns with broader shifts in farming practice:
Antibiotic use has dropped more than 95 percent since the 1990s. An environmental milestone, but one that removed a key safety net.
Sea-lice control has shifted from chemical to mechanical and biological methods — cleaner for the environment, but harder on fish.
Stocking densities and environmental limits are far stricter, leaving less flexibility during biological stress.
Welfare and reporting standards are higher than ever, capturing losses that once went unreported.
Each of these steps improved sustainability, but also made farming less forgiving. The low-mortality years of the 1990s weren’t a golden era, they were the byproduct of higher chemical use and looser regulation.
Biological improvement costs capital, and Scotland has long struggled on profitability.
Even during record-high salmon prices, Scottish EBIT per kilo hovered near break-even or negative. Higher logistics costs, tighter regulation, and frequent treatments all eat into margins. That leaves less to reinvest in welfare technology, fewer wellboats, limited delousing capacity, slower innovation cycles. The result is a feedback loop the industry knows too well:
Low profitability → limited reinvestment → slower progress → renewed welfare strain.
Ironically, the first half of 2025 delivered one of Scotland’s best biological periods in years (low mortality, stable fish health) but also one of its worst financial ones as prices dropped sharply.
The fish were healthier. The farms weren’t richer.
This report should have sparked a conversation about the long-term evolution of Scottish aquaculture, its progress, trade-offs, and resilience. Instead, the discussion collapsed into a headline war over one number: 61.8 percent. When industry insists every year is a success, credibility erodes. When critics frame every loss as failure, they erase years of progress.
Both sides lose the nuance and the nuance is where the science lives.
Aquaculture evolves through cycles, not verdicts. Every peak and dip is a signal, not a score.
The real story isn’t a discussion over the industry's survival rate figure; it’s the shape of the curve (the rise, the dip, the recovery) and what it tells us about an industry constantly balancing biology, welfare, and economics.
The Scottish Fish Farm Production Survey isn’t a verdict but a time capsule. It captures forty years of change, constraint, and progress: from antibiotics to AI, from chemical control to biological balance. If there’s one takeaway, it’s that progress in aquaculture isn’t linear, but it is measurable.
That’s why our data intelligence platform Manolin’s Watershed exists: to help farms see these long arcs clearly, connect the patterns, and make proactive decisions grounded in biological reality. Scotland’s story isn’t failure. It’s evolution written in data.