← Back

Deep dive

Two ways to
count the same fish

Norwegian salmon farmers report four categories of sea phase losses to the Directorate of Fisheries: dead fish, slaughterhouse rejects, escapees, and "others." The Veterinary Institute's annual Fiskehelserapporten calculates mortality risk using only the dead fish category. Everything else — millions of fish per year — sits outside the headline figure.

2024 total losses
70.3M
fish removed from stock
Reported mortality
59.1M
84% of total — Dødfisk
"Others" category
8.3M
12% of total — Annet
Slaughterhouse rejects
3.0M
harvested but rejected
01 — The two figures
Reported mortality and total losses are not the same number
The Fiskehelserapporten uses reported dead fish (Dødfisk) to calculate mortality risk. The total loss figure from Fiskeridirektoratet includes all four reporting categories. The chart below shows both over time. In 2024 the difference was around 11 million fish.
Total Losses vs. Reported Mortality
Thousands of fish — sea phase (all species) · 1998–2024
How the four categories are defined

The Fiskehelserapporten 2025 (Norwegian Veterinary Institute) gives the official definitions. "Død" covers fish physically removed from cages and recorded as dead. "Utkast" covers downgraded fish sorted at the slaughterhouse — harvested but unfit for human consumption. "Rømming" covers fish lost through escape events. "Annet" is used for losses that do not fit the other three categories.

The Veterinary Institute does not merge "annet" with "tellefeil" (counting errors) when calculating mortality risk — unlike Fiskeridirektoratet's biomassestatistikk, which combines them. The "Others" figures in the charts on this page, sourced from Fiskeridirektoratet, may therefore include both genuine unclassified losses and stock count corrections in the same number.

02 — The "Others" category
The "Annet" category over time
"Annet" has been present in every year of reporting. The charts below show its absolute volume and its share of total losses from 1998 to 2024.
Others — Absolute Volume
Thousands of fish — 1998–2024
Others — Share of Total Losses
% of all sea phase losses — 1998–2024
Cause-of-death data from AquaCloud

AquaCloud is an industry initiative collecting daily cause-of-death data from farm management systems. It covered 408 sea sites in 2025, representing approximately 49% of all Norwegian salmon sea sites. The three leading causes of registered mortality were: Injuries and trauma (33.5%), Infectious diseases (28.1%), and Unknown cause (27.4%).

Injuries and trauma include damage from delousing operations and handling. Infectious diseases include HSMB, CMS, and winter ulcer among others. The 27.4% in "Unknown cause" represents registered dead fish where no cause was recorded at the site. These categories apply to the "Dødfisk" column only. "Annet" losses are not included in AquaCloud's figures.

03 — What is in "Others"
No single cause — and the data systems treat it differently
"Annet" is a residual category for losses that do not fit the definitions of dead, escaped, or slaughterhouse-rejected fish. Fiskeridirektoratet's loss statistics merge "annet" with "tellefeil" (counting errors) in their biomassestatistikk, while the Veterinary Institute keeps them separate in the Fiskehelserapporten. No further breakdown of "annet" is published at national level.
Counting errors (Tellefeil)
Discrepancies between expected and actual stock counts — both positive and negative — arising from stocking, transfers, harvesting, or escape records. Fiskeridirektoratet merges these with "annet" in their published loss tables. The Veterinary Institute excludes them from mortality calculations.
Unclassified losses
Losses that do not fit the three formal categories — dead, escaped, or slaughterhouse-rejected. Could include fish removed during emergency operations, losses during transfers between sites, or mortality events where the cause was not recorded by the time of monthly reporting.
Jellyfish & algae events
Mass mortality from jellyfish or harmful algae blooms may be registered under "Dødfisk" or "Annet" depending on reporting practice. In 2025, algae blooms in PO9 contributed to an increase in that region's mortality risk according to the Fiskehelserapporten. These events are well documented and recurring in Norwegian fjords.
Unknown cause — registered dead
AquaCloud data (49% of sea sites, 2025) shows that 27.4% of registered dead fish had no cause recorded. These are counted in "Dødfisk" — not "Annet." This means the true share of unexPlained losses across all categories is likely higher than the "Others" chart alone suggests.
How the Fiskehelserapporten handles this

The Veterinary Institute calculates mortality risk using only the "Dødfisk" category — dead fish physically removed from cages. Slaughterhouse rejects, escapees, and "annet" are all excluded from the mortality risk figure. This is the calculation behind the reported annual mortality risk of 14.2% for salmon in 2025.

The total loss rate — which includes all four categories — gives a broader picture of fish that entered the production system and did not reach slaughter as intended. Both figures matter. They are measuring different things.

04 — All categories together
All four reported categories stacked
The chart below shows all four recorded categories of sea phase losses together. Slaughterhouse rejects (Utkast) are fish that reached harvest but were rejected as unfit for consumption. Escapees (Rømming) are included for completeness. Each category is defined separately by Fiskeridirektoratet.
Sea Phase Losses — Full Breakdown
Thousands of fish by category · 1998–2024 (stacked)
Reading the chart

Reported mortality (Dødfisk) has grown in absolute terms as production scales up. In 2005, 182.6 million fish were put to sea; by 2024 that was 461.9 million. Raw counts are not directly comparable across years without accounting for production volume — the rate charts in Section 05 show the same data relative to fish stocked.

"Annet" reached its highest absolute volume in 2011 (12.1M) and 2022 (8.8M). Slaughterhouse rejects (Utkast) have remained broadly stable at 3–4 million per year across the full period.

The Norwegian government's target is 5% mortality per production cycle. The Fiskehelserapporten 2025 reports a production cycle mortality risk of 14.7% for salmon cycles completed in 2025, based on Dødfisk only.

05 — Relative to fish stocked
Losses and mortality as a share of fish put to sea
Dividing by fish put to sea makes years comparable regardless of production volume. The first chart uses total losses (all four categories). The second uses Dødfisk only, matching the method in the Fiskehelserapporten.
Sea Phase Loss Rate — All categories
Total sea losses ÷ fish put to sea · % · 2001–2024
Sea Phase Mortality Rate — Dødfisk only
Reported dead fish ÷ fish put to sea · % · 2001–2024 · method used by Fiskehelserapporten
2024 in numbers

In 2024, 462.8 million fish were put into Norwegian sea cages. Total losses were 70.3 million — a loss rate of around 15%. Reported mortality (Dødfisk) was 59.1 million — a mortality rate of around 13%. The difference between the two rates reflects the other three reporting categories.