Scotland’s Fish Farms Face Longer Heatwaves: Could Past Records Hide New Pressures?

The researchers used daily sea surface temperature estimates from OSTIA, a dataset combining satellite and in-water observations, to reconstruct heatwave histories around Scottish aquaculture locations.

For a salmon or mussel farm, a spell of unusually warm seawater can become a very different challenge when it stretches from days into weeks. New research around Scotland shows that marine heatwaves are becoming more frequent, with some farming locations experiencing much faster increases in how long these events last and how much extra warmth builds up.

The study, "Marine heatwave exposure is changing unevenly across Scottish aquaculture sites," published in Frontiers in Marine Science, examined temperature records from 1990 to 2025 at locations associated with 361 active farms. Led by Camila Toledo-Jaime, the research team found that places with lower historical heatwave intensity were experiencing some of the fastest changes, raising questions about how reliably past conditions can guide future farm planning.

Reading 36 Years of Warm Water Around Scotland's Farms

The researchers used daily sea surface temperature estimates from OSTIA, a dataset combining satellite and in-water observations, to reconstruct heatwave histories around Scottish aquaculture locations. The sample covered 183 finfish farms, 174 shellfish farms and four sites producing both.

Several farms occupied the same roughly five-kilometre grid cell and shared identical temperature records. The team analysed 218 distinct cells to avoid counting those shared records as independent measurements, identifying 21,099 heatwave events across the cells over 36 years.

A marine heatwave meant at least five consecutive days above a locally defined temperature threshold, calculated from the warmest 10% of conditions for that time of year using a fixed 1990–2020 reference period. This definition captures unusual warmth relative to local seasonal conditions, so an event does not necessarily mean the water reached temperatures dangerous to farmed animals.

The team checked the dataset against measurements taken about five metres below the surface at eight salmon farms between 2015 and 2025. The records closely tracked temperature changes at those farms, supporting their use for broad comparisons across locations.

Two Groups Reveal Different Heatwave Histories

Grouping locations by their heatwave characteristics revealed two broad patterns. The first group contained 173 grid cells spread around the Scottish coast, including the Northern Isles, Outer Hebrides and west coast. The second contained 45 cells concentrated along parts of the west coast, including several sea lochs, inlets and channels.

Across the full study period, the first group experienced stronger temperature departures from seasonal averages. Its mean heatwave intensity was about 0.83°C above the local seasonal average, compared with 0.47°C in the second group, and it also had higher peak and accumulated intensity.

Differences in event length and frequency were smaller: heatwaves averaged roughly 15.7 days in the first group and 16.4 days in the second, with about 95 and 102 events per cell across the record.

These groups describe overlapping patterns rather than firm boundaries between coastal environments. Their geography offers clues worth investigating, including differences in water exchange, mixing and heat retention, but the study did not establish which physical processes caused the patterns. The faster changes were occurring in the second group, despite its lower historical heatwave intensity.

Heatwaves Are Staying Longer, With Sharper Changes at Some Sites

Heatwave frequency increased at almost identical rates in both groups, with the models estimating an average annual rise of about 5.9% across the study period. The larger differences emerged in event duration, accumulated warmth and severity. Comparing 1990–2019 with 2020–2025, estimated event duration rose from 9.52 to 13.8 days in the first group, an increase of 44.6%. The second group experienced an increase of 89.9%, from 8.99 to 17.1 days.

The researchers also measured cumulative intensity, which adds up each day's extra warmth across an event. An anomaly that persists for weeks can produce a large cumulative total even without a dramatic rise in average intensity. This measure increased by 50% in the first group and approximately 127.5% in the second. Average event intensity changed much less, rising from 0.830°C to 0.855°C in the first group and from 0.452°C to 0.538°C in the second. Longer events accounted for much of the growth in accumulated warmth.

The estimated probability that a detected heatwave reached the Severe or Extreme category rose from 1.8% to 6.1% in the first group. In the second, it climbed from 2.2% to 26.2%, or roughly one in four events during 2020–2025. Those categories measure departures from local historical temperature thresholds; they do not measure animal deaths or damage.

The full-record analysis found a clear upward severity trend in the second group, but no clear equivalent trend in the first. The researchers used 2020 as a comparison boundary, without claiming it marked a sudden environmental shift, and found similar contrasts using 2017, 2018 or 2019.

What Longer Warm Spells Could Mean for Farm Life

Research discussed in the paper links sustained warmth with changes in salmon feeding, energy demands, immune responses and tolerance of low oxygen. Handling can become more stressful at high temperatures. Studies of oysters and mussels describe depleted energy reserves, altered behaviour and effects that can continue into recovery periods, making the spacing between heat events another useful question for future research.

The researchers propose combining satellite information with continuous farm measurements at relevant depths, supported by oxygen readings and stock-health observations. Forecasts showing how long warm conditions may persist could help farms prepare monitoring and review operations that add stress.

Five-kilometre surface data cannot capture every temperature difference inside a narrow loch or below the surface, and validation covered only eight salmon farms on the west and northwest coasts. Connecting these temperature histories with records of fish and shellfish health, growth and survival would show how the changing exposure translates into farm outcomes.