Norway's salmon lice "traffic light" system has been running since 2017. Nine years and five assessment rounds in, the industry now has enough history to ask a harder question than which production areas turned red this year. The question is whether the system's basic design, a 13-area map built to limit infection between farms, is even the right tool to answer the question it was actually built for: how much lice-driven harm farms are causing wild salmon and sea trout.
The 2026 decision, combined with a new scientific report released ten days later, suggests Norway is now answering that question directly. The map itself is under review.
Sea lice regulation in Norwegian aquaculture predates the traffic light system by close to two decades. A February 2000 regulation set the first numeric threshold still recognizable today: no more than 0.5 adult female lice per fish, with counts required at least every 14 days. A 2012 regulation, in force from January 2013, tightened the counting cadence to every 7 days whenever water temperature reached 4°C or above, while keeping the 0.5 threshold. In March 2017, the threshold itself got stricter during the spring migration window: 0.2 adult female lice per fish during the weeks when wild salmon smolt are moving through coastal waters, with 0.5 remaining the limit the rest of the year.
These are treatment-action thresholds, the point at which a farm is required to delouse. The traffic light system, launched later in 2017, operates on a different axis entirely: instead of regulating individual farm behavior, it regulates how much total biomass an entire production area is allowed to carry, based on the area's assessed impact on wild fish.
Norway's coast is divided into 13 production areas, PO1 through PO13, running north from the Swedish border to eastern Finnmark. Each area gets a color rating from HI, the Institute of Marine Research, based on lice-induced mortality risk to wild salmonids: green permits biomass growth, yellow freezes capacity, red requires a reduction.
Here's what that's looked like since 2017. It's clear over time what direction this has been going.
Green areas have shrunk from 8 in 2017 to 3 in 2026. Yellow areas have grown from 3 to 9 over the same period. PO3 has been cut in three of the last three rounds. PO4 was cut in 2020, 2022, and 2024. The system was designed with the expectation that farms could earn back growth by reducing their lice impact. Instead, the number of areas judged capable of supporting growth has fallen by more than half.
The 2026 decision, announced June 19, allowed roughly 8,300 tonnes of additional maximum permitted biomass across the three green areas, PO1, PO12, and PO13, while PO3 took its third consecutive reduction.
One area's shift stands out for a different reason. PO9, Vestfjorden and Vesterålen, had been rated low-impact in every assessment since the system began. In 2026, it stayed yellow instead of moving to green. Norway's Ministry of Trade and Fisheries pointed to monitoring data showing moderate mortality risk for salmon smolt and moderate-to-high impact on sea trout, and specifically flagged that sea temperatures in late summer 2024 and summer 2025 were unusually high, a condition that shortens lice development time and lets lice populations build faster. A production area's rating isn't fixed by its history. Warming water can move a previously low-risk area into a more cautious category even without new farms or added biomass.
Fisheries minister Marianne Sivertsen Næss was direct about what the growing yellow count means: the goal of reducing wild salmonid impact isn't being met, and the traffic light system by itself isn't sufficient to get there. She said that for the industry to grow, farms need to reduce the larval lice they're emitting into the water, not just avoid exceeding a per-fish count.
Why per-fish and per-area miss the target
The lice-per-fish threshold has a structural problem that becomes obvious once you compare two farms directly. A farm holding 1 million fish at 0.5 lice per fish is technically over the treatment threshold. A farm holding 2 million fish at 0.3 lice per fish is technically under it. But the second farm is emitting more total lice into the surrounding water than the first. Regulating the ratio instead of the total volume means a larger farm can look compliant while contributing more to the actual problem the rule exists to address.
That mismatch is now part of the active policy conversation. Norway's 2025 aquaculture white paper, the Havbruksmelding, proposes moving toward lice quotas based on total emitted lice rather than lice-per-fish ratios. HI's companion report on acceptable emission levels, also released in June 2026, argues explicitly that a quota system based on total adult female lice count would be more predictable and easier to enforce than the current ratio-based approach. Farms that exceed their quota would need to buy capacity from farms with a surplus, or cut emissions directly.
The same reasoning applies at the level of the production area, and it's the subject of HI's most consequential 2026 finding.
The 13 production areas were built to solve a different problem
On June 29, 2026, HI published a report proposing that Norway divide its 13 production areas into 28 smaller regulatory zones. The Ministry commissioned this work to answer a specific question: are the current 13 areas the right geographic unit for regulating lice emissions to wild fish, as opposed to the problem they were originally built to solve, which was limiting lice spread between farms.
HI frames the question using two criteria. Homogeneity means the effect of lice emissions on wild fish is roughly consistent no matter where inside an area the emissions occur, which matters because it lets regulators apply one rule fairly across every farm in that area. Independence means emissions inside one area don't meaningfully affect wild fish outside it, which matters because it lets regulators set limits area by area without modeling every neighboring area at the same time.
These two goals pull in opposite directions. Larger areas tend to have better independence, since they contain more of the lice spread internally, but worse homogeneity, since conditions can vary widely within a large area depending on exactly where the farms sit. Smaller areas improve homogeneity but worsen independence, since lice move more easily across a larger number of new boundaries.
HI's assessment of the current 13-area map: high independence, weak homogeneity. The areas were drawn to contain farm-to-farm infection networks, and they do that reasonably well. But they're too internally variable to fairly regulate lice impact on wild fish, because a handful of high-impact farm locations inside a production area can dominate that area's average risk score while other farms in the same area contribute comparatively little.
HI's answer is 28 zones, with each existing production area split into one to four smaller zones depending on internal variability. The methodology behind that number is a clustering analysis run on a model connecting farm locations to wild salmon and sea trout migration routes, built from lice-dispersion modeling and river-by-river migration timing across nearly a thousand farm sites. HI tested cluster counts from 8 up to 64 before settling on 28 as a working balance point, and the report is explicit that this isn't a mathematically unique answer. More zones would improve homogeneity further but would also increase cross-boundary infection, making independence worse. Fewer zones would do the reverse. HI describes 28 as one reasonable compromise among many possible ones, not a final optimized number.
The traffic light system's colors are driven primarily by projected impact on wild Atlantic salmon. But HI's own 2026 risk assessment shows a starker pattern for a different species. Under the same 2026 data, two production areas rate high risk for salmon-smolt mortality, eight rate moderate, and three rate low. For sea trout and Arctic char, the picture is considerably worse: nine production areas rate high risk, two moderate, two low.
Salmon smolt move through a production area in a matter of days to a few weeks during their spring migration to the ocean. Sea trout and Arctic char don't migrate the same way. They stay in fjords and near-shore coastal waters through the entire summer, which means a much longer exposure window to whatever lice larvae are in the water during that period. The 2024 and 2025 warm summers made this worse specifically for sea trout and Arctic char, since warmer water shortens lice development time and increases the larvae's ability to infect a host. HI describes this year's sea trout and Arctic char risk picture as unchanged from the year before, meaning it isn't a one-time spike. It's a persistent gap between a regulatory system built around salmon migration timing and where the actual biological risk has been concentrated for at least two consecutive years.
Norway's Ministry has indicated sea trout will be formally added as a traffic light indicator starting in 2028. Until then, the headline color decisions will keep being driven by an indicator that, on HI's own numbers, currently understates where the more severe impact is happening.
Millions of dollars have gone into new production systems and treatment vessels since the traffic light system began, and farm-level reporting patterns have shifted accordingly. Manolin's own aggregated treatment data shows mechanically treated cages rising from 12,341 in 2024 to 14,340 in 2025, an increase of roughly 16 percent in a single year, concentrated heavily in the July-through-October window when lice pressure peaks.
Wild salmon populations haven't followed. Norway's Scientific Advisory Committee for Salmon Management, VRL, put the 2025 wild salmon influx at roughly 456,000 fish, up from a record low of about 323,000 in 2024, but still near the bottom of the historical range. VRL's own framing: the current influx is about a third of what it was in the early 1980s. Norway's Environment Agency has separately classified wild Atlantic salmon as near-threatened on the national red list.
Investment in treatment capacity and a genuine shift in how farms report and manage lice exposure haven't yet shown up as a meaningfully different trajectory for wild salmon. HI's 28-zone proposal and the emerging lice-quota framework are both attempts to close that gap structurally, rather than by adjusting more colors.
This will take time. Each phase of this regulatory system, from the original 2000 lice thresholds to the 2017 launch of the traffic light system to the current round of structural proposals, has unfolded over roughly a decade of legal and political process. The traffic light system itself is now approaching a decade old, and its measurable effect on wild salmon populations remains, on the industry's own data, minimal. Whether a finer-grained zone map and an emissions-based quota system produce a materially different outcome in the next decade is not yet answerable. What's clear is that the current system, built on a 13-area map designed for a different problem and a single-species indicator that understates risk to at least two other wild fish, has reached the edge of what it can accomplish without a structural redesign.