Atlantic salmon: what RiverNotes (ASF-FSA) says… and especially what it doesn't say

Before the first CAST

Analysis of the Atlantic Salmon Federation's RiverNotes dated August 13, 2026, with scientific context and official data from Quebec.

Good local news is good news. It does not automatically become proof of provincial recovery.

Analyse_RiverNotes_ASF_13_aout_2026_Quebec (1).pdf

Download PDF • 768KB

The ASF correctly puts heat at the center of the debate. But its own text also reveals the heart of the Quebec problem: only 4 rivers have a formal provincial regulatory mechanism, while the other 110 rely on local means [1].

Preface

This analysis is not intended to attack the Atlantic Salmon Federation (ASF), nor to downplay the good news of 2026. On the contrary: several elements of the August 13 RiverNotes are useful, responsible, and deserve to be highlighted. The ASF speaks openly about temperature, funds thermographs, works to standardize instrument calibration, supports local managers, and showcases thermal mapping work elsewhere in the salmon range [1].

But when a conservation document highlights improvements in upstream migration on a few major rivers while describing a heat protocol that formally covers only four rivers in Quebec, the two messages must be considered together. The goal here is therefore to distinguish what the RiverNotes document actually demonstrates, what it suggests, what it leaves unanswered, and what should logically follow if we want to move from data collection to genuine heat risk management.

Executive Summary

• ASF states that 2026 so far represents a significant improvement compared to the previous two years, while dedicating its Quebec section to heat, thermographs and positive signals on the Matapédia and the Grande Cascapédia [1].

• RiverNotes confirms that only four Quebec rivers are formally covered by the MELCCFP heat protocol and that the other 110 rely on local means [1].

• The Quebec regulatory threshold for closure is above 22°C in the morning when conditions are likely to persist; the government specifies that high temperatures reduce the survival of salmon released back into the water [2].

• The scientific literature does not support presenting 22°C as the point where the risk begins. A 2020 Canadian review found release mortality rates of 7 to 33% between 18 and 20°C (average 16%) and 14 to 61% between 20 and 25°C (average 35%), with high variability between studies [5].

• The 2025 report of the MELCCFP provides an essential reminder of the context: -45% for the total upstream migration and -68% for the multi-sea-winter salmon (MSW) / large salmon rivers compared to the five-year average of the same 32 rivers, despite +10% for the grilse / one-sea-winter salmon (1SW) rivers [3].

• Conclusion: Thermographs and thermal shelters are necessary, but data is not a conservation measure unless it is linked to triggers, temporary closures, reopening criteria and consistent communication at the provincial level.

  
  

Verdict: RiverNotes makes the right diagnosis - heat is the elephant in the room - but it remains too cautious on the crucial question: what do we do, concretely and uniformly, when the data shows that the risk increases?

1. What the RiverNotes actually says

The Quebec section is signed by Caroline Côté, Quebec program director. She begins by identifying the temperature as "the elephant in the room" and points out that four rivers—Malbaie, à Mars, Petit-Saguenay, and Saint-Jean (Saguenay)—are formally covered by the MELCCFP's heat protocol. The described mechanism provides for a temporary closure of at least 72 hours when the water reaches 22°C in the morning and the forecast remains unfavourable [1].

The text then adds that the other 110 salmon rivers in Quebec "govern themselves with their local resources." The ASF describes its support to the Gaspé Rivers Management Society and Sipuminu for the acquisition of thermographs, as well as its work to share a protocol in French aimed at data calibration and comparability [1].

Finally, RiverNotes highlights encouraging results: 1,751 large salmon were counted on the Matapedia River as of July 30, representing a 62% increase compared to 2025 and a 12% increase compared to the 2019-2024 average for the same date. On the Grande Cascapédia River, June saw 408 salmon counts (up 31%) and July 350 salmon, already exceeding the July 2025 count [1].

These comparisons are interesting because, for the Matapédia region, ASF compares equivalent dates. They support a positive local signal. However, they are not sufficient to describe the situation of the entire Quebec population.

2. What is solid and useful in ASF's message

There are several commendable aspects to this RiverNotes report. First, ASF does not deny the thermal issue. It explicitly places it at the heart of its Quebec section. Second, the investment in thermographs and data quality is relevant: without comparable time series, it becomes difficult to distinguish a one-off event from a lasting change in the thermal regime.

Work on calibration is particularly important. Poorly calibrated, poorly positioned, or used instruments with different methods can make comparisons unreliable. The idea of strengthening local capacity and enabling managers to understand their own river is therefore defensible and constructive [1].

ASF is also right to talk about thermal refuges. In its Research and Environment section, the organization describes significant thermal imaging work on the Margaree River in Nova Scotia, with the explicit goal of translating thermal mapping into protection decisions, restoration priorities, and mitigation measures [1]. Scientific work conducted in Canada also shows that thermal refuges are dynamic and that their availability varies depending on hydrological conditions and the type of refuge [7].

What makes RiverNotes particularly interesting is that it provides its own logical framework: acquisition -> understanding -> action. The question then becomes whether this chain truly exists in Quebec for the 110 rivers not covered by regulatory protocols.

3. The real blind spot: measuring is not managing

Perhaps the most significant figure in the Quebec section isn't the Matapédia River. It's the ratio of 4 to 110. ASF explains that four rivers are covered by the provincial regulatory protocol, while all the others rely on local measures [1]. The Quebec government confirms that in 2026, four rivers are subject to possible closures when temperatures become critical [2].

This does not mean that the other 110 rivers all have the same thermal risk or that they should all be closed at the same temperatures. Populations, flows, habitats, refuges, and fishing pressures differ from river to river. The government also states that the four selected rivers were identified as priorities based on several criteria, including temperature, population status, and vulnerability to fishing activities [2].

But this nuance doesn't solve the governance problem. When an uncovered river reaches a state of thermal stress, who decides? What threshold is used? Over what period? What data is considered authoritative? What closure is implemented? What are the reopening rules? How is the angler informed? And above all: can two neighboring rivers with the same temperatures be managed in completely different ways simply because they don't have the same local resources?

A thermograph answers the question "what is it?". A protocol answers the question "what do we do now?". These are two different tools.

4. 22°C: regulatory threshold, not a magic biological line

The Quebec government indicates that a closure can be triggered when the water temperature exceeds 22°C in the morning and that this situation is likely to persist. It also explains that warm water reduces available oxygen, increases stress, and decreases the survival rate of salmon caught and released [2].

However, extreme precision is necessary: the fact that a regulatory threshold is set at 22°C does not mean that salmon is biologically safe at 21.9°C. Management decisions transform a gradual biological reality into an operational criterion. Science clearly shows that the risk increases below 22°C.

A review published in 2020 in the Canadian Journal of Fisheries and Aquatic Sciences compiled available studies on post-release mortality. At temperatures below 12°C, mortality was generally low (<5%). Between 18 and 20°C, observed mortalities ranged from 7% to 33%, with an average of 16%. Between 20 and 25°C, they ranged from 14% to 61%, with an average of 35%. The authors emphasize the variability between studies, particularly at higher temperatures [5].

A more recent study, published in 2024, observed 4 mortalities out of 18 salmon released at temperatures >=20 °C (22.2%), compared to 4 out of 96 at <=20 °C (4.2%); one of these four mortalities below 20 °C was attributed to a hook injury. The authors conclude that the temperature at the time of capture remains a particularly useful predictor of survival after release [6].

As early as 1996, Wilkie and his colleagues had documented a more difficult physiological recovery in warm water and significant delayed mortality in a subgroup caught at 22°C [4].

5. The good news of 2026: to be celebrated, not generalized

ASF is right to highlight the results from the Matapédia and Grande Cascapédia rivers. A 62% increase in large salmon on the Matapédia compared to 2025 and a level 12% higher than the 2019-2024 average at the same date are encouraging signs [1]. They indicate that 2026 could represent a better year on some major rivers.

But the interpretation must remain rigorous. First, this is data from the current season, not a complete provincial assessment. Second, two major rivers do not automatically represent the state of all populations in Quebec. Third, an increase compared to 2024 or 2025—two very low years in several areas—may be significant in percentage terms while still being insufficient to conclude that there has been a sustainable recovery.

The improvement must therefore be formulated correctly: “several major rivers show very positive signs in 2026.” This is not the same as “Quebec populations have returned to normal.” RiverNotes does not provide, in its Quebec section, the data necessary to support this second conclusion [1].

A good year is an opportunity to maximize reproduction. It should not become a reason to reduce caution precisely when more spawners are present in the river.

6. The context that cannot be forgotten: the 2025 Assessment

To understand the significance of the good news for 2026, we need to remember where we started. The MELCCFP's 2025 report indicates that 13,133 adult salmon were observed in the 34 rivers surveyed. Of the 32 rivers with data for each of the last six years, the total run in 2025 was 45% below the five-year average, the multi-sea-winter salmon (MSW) / large salmon 68% below average, while the grilse were 10% above [3].

The same report also indicates that the small salmon sport fishery harvest was 75% below the five-year average, the large salmon harvest was 72% lower, river traffic was down 29%, and permit sales were down 22% [3]. The ministry specifies that several factors contributed to the operating data: low returns, more restrictive regulations, and measures to protect large salmon.

This context does not negate the improvement observed in 2026. It simply prevents us from too quickly transforming an annual improvement into a conclusion of recovery.

2026 should be judged in relation to a trajectory, not just in relation to the two weakest years preceding it.

7. The RiverNotes itself provides an instructive contrast

In the New Brunswick section of the same RiverNotes, ASF reports that warm water protocols have been activated or expanded on the Miramichi, Nepisiguit, and Upsalquitch rivers. The organization reminds anglers to check the current restrictions [1].

In the Research section, ASF explains that thermal mapping of the Margaree River is used to locate refuges, identify sources of thermal pollution, establish restoration priorities, and inform mitigation plans. The message is clear: the data must be used to make decisions [1].

The legal frameworks and biological realities of New Brunswick, Nova Scotia, and Quebec are not identical. It would therefore be simplistic to simply copy a rule from one province to another. But it is equally difficult to ignore the contrast: in the same bulletin, the ASF elsewhere describes active protocols and an explicit "data-to-action" chain, while in Quebec it primarily emphasizes the local capacity for measurement and interpretation.

The question is not: should we choose between local science and a provincial protocol? We need both. Local data improves decision-making; a provincial framework ensures that the decision actually exists.

8. The questions that RiverNotes leaves open

• Why do only four rivers have a regulatory heat protocol in place by 2026?

• What specific criteria would allow a fifth, tenth, or twentieth river to enter the scheme?

• What mechanism applies to uncovered rivers when temperatures reach levels associated with a documented increase in the risk of mortality after release?

• At what temperature should public communication shift from "be careful" to an operational restriction?

• Will the data collected by thermographs be publicly accessible on a daily basis, or only for research purposes?

• How can we prevent purely local management from creating different thresholds, methods, and responses for biologically comparable situations?

• How to integrate population size, the proportion of large salmon, flow rate and the availability of refuges into closure decisions?

• How will the positive results of 2026 be interpreted when the complete provincial assessment is available - and not just that of a few major rivers?

  
  

These are not accusations. These are the normal questions that a proper conservation protocol should be able to answer before the next heat wave.

9. Clear answer

The Atlantic Salmon Federation's RiverNotes of August 13, 2026, is worth reading because it finally puts into words what many salmon anglers have been saying for a long time: in August, water temperature is indeed the elephant in the room. On that point, I completely agree with the ASF. Where I disagree is when they move from stating the facts to proposing a concrete solution.

The ASF points out that only four rivers – Malbaie, à Mars, Petit-Saguenay and Saint-Jean du Saguenay – are formally covered by the MELCCFP heat protocol. For the other 110 salmon rivers in Quebec, we are essentially told that managers are working with their local resources [1].

This is precisely where the problem becomes impossible to circumvent.

I am fully in favor of thermographs. I support uniform calibration. I support the study of thermal refuges, thermal imaging, and ensuring that local agencies own and understand their data. All of this is necessary. But we must stop confusing a thermometer with a conservation policy. Measuring a river at 20, 21, or 22°C will not protect any salmon if no one has determined in advance what will happen when the temperature reaches a concerning level.

Quebec now has a regulatory trigger that can lead to a closure above 22°C in the morning when conditions are likely to persist [2]. This is a management threshold. It is not a biological boundary. Scientific literature published over the past few decades shows that the effects of fishing and catch-and-release become more concerning below 22°C.

A 2020 Canadian review reports that, between 18 and 20°C, release mortality rates range from 7% to 33%, with an average of 16%. Between 20 and 25°C, the reported average rises to 35%, with very high variability between studies [5]. In other words, we need to stop talking as if everything is fine up to 21.9°C and suddenly becomes dangerous at 22°C.

This doesn't mean that a salmon caught at 19°C will die. Nor does it mean that an automatic closure at 18°C is scientifically mandatory everywhere. That would be just as simplistic. It means that the risk increases progressively and that management should also be progressive: increased monitoring, reduced effort, mandatory or strictly regulated practices, time restrictions, and then temporary closure when conditions warrant it.

I also want to be very clear about the good news for 2026. Yes, the Matapedia River is showing a very encouraging sign. As of July 30, ASF reports 1,751 large salmon, 62% more than in 2025 and 12% above the 2019-2024 average for the same date [1]. Yes, the Grande Cascapedia River is also showing a positive trend. That's great. That's exactly what we want to see.

But a strong Matapédia and an encouraging Grande Cascapédia do not represent all of Quebec. An improvement compared to 2024 and 2025 should not erase the context in which 2026 begins. The MELCCFP's own 2025 assessment shows a total run 45% below the five-year average on the same 32 rivers and a 68% deficit among the reedbeds [3]. The good news of 2026 must therefore be protected, not used as a tranquilizer.

This is probably what bothers me most about the way we talk about salmon. As soon as one or two rivers have a good year, we breathe a sigh of relief and start talking about things returning to normal. But when a small river only has a few dozen or a few hundred salmon, its margin for error isn't the same as the Matapédia's. A serious policy must be able to recognize this difference.

The ASF itself states in the same RiverNotes that its thermal mapping of the Margaree River should be used to guide protection decisions, restoration priorities, and mitigation measures [1]. This is precisely the model that needs to be applied: from data to action. If we install thermographs on Quebec rivers, we must first know what decisions this data will trigger.

The debate is therefore not between fishermen and biologists, nor between closing all rivers and doing nothing. It is between improvising management when the heat arrives and preparing for it beforehand.

A modern provincial protocol could very well retain local flexibility. It could take into account morning temperature, flow rate, weather trends, population size, the proportion of large salmon, and the presence of thermal refuges. But the basic principles, measurement methods, risk levels, intervention criteria, and reopening rules should be known to everyone before the start of the season.

This isn't about closing the fishing season for the sake of closing it. It's exactly the opposite. A clear protocol allows for short, targeted, and predictable closures when they are truly necessary, rather than letting a river reach a point where much harsher measures might become unavoidable.

So yes, I applaud the ASF's RiverNotes. I applaud the thermographs, the comparable data, and the good news for 2026. But if temperature is truly the elephant in the room, we need to stop simply measuring it. We need to decide what we're going to do when it rises up before us.

And most importantly, it must be decided before the next heat wave - not during.

10. What a coherent answer might contain

This analysis does not claim to establish the definitive biological protocol for Quebec. However, it does allow us to identify the components that a credible protocol should contain.

• Standardized provincial measurement method: location, reading time, depth, calibration and validation of probes.

• Progressive risk levels rather than a single binary threshold: information, increased caution, operational restrictions, temporary closure.

• Integration of flow rate and weather forecasts, since the same temperature does not necessarily have the same meaning under all conditions.

• Taking into account the size and condition of the population: a small vulnerable river does not have the same cushion as a large system.

• Explicit protection of thermal refuges when they concentrate fish, with a clear rule on the rest of the river to avoid inconsistent management.

• Reopening criteria known in advance and applied transparently.

• Rapid publication of temperatures, flow rates, closures and reopenings so that fishermen can make their decisions before travelling.

• Annual evaluation of the protocol: observed mortality, thermal episodes, use of refuges, fishing effort and economic effect of closures.

• Official mechanism allowing new rivers to be added quickly to the protocol without waiting several years for administrative procedures.

  
  

The best protocol is not the one that closes the most. It's the one that protects at the right time, reopens at the right time, and gives everyone the same rules of the game.

11. Historical context – why one good year is not enough

Compilation of data from the 1984-2025 working period based on data from the MELCCFP's 2025 Balance Sheet [3]. The series is used to visualize the trajectory; it does not replace standardized river-by-river analyses.

This long series reveals cycles, recoveries, and troughs. Above all, it serves as a reminder of why comparisons must be made across multiple cohorts. A population can experience a very good year in the midst of a fragile trend, just as a single bad year is not enough to signal a collapse.

The cautious conclusion is therefore as follows: 2026 brings encouraging signs that must be monitored until the end of the season and compared river by river with consistent reference periods. These results do not exempt Quebec from immediately improving its thermal risk management.

Conclusion

The RiverNotes of August 13, 2026 is more important than it looks. In a few paragraphs, it brings together almost all the pieces of the puzzle: summer warming, heat protocols, thermographs, thermal refuges, local data, good runs on some large rivers and the need to turn knowledge into action [1].

The main weakness of Quebec's message, therefore, is not what it says, but what it doesn't fully commit to. If climate is the elephant in the room, and if 110 rivers still rely primarily on local resources, the next logical step isn't simply to install more instruments. It's to build a clear decision-making framework, consistent in its principles and adaptable in its application.

The good news of 2026 should give us room to act intelligently. It shouldn't become an excuse to wait for the next crisis. When more salmon return, the best conservation response is to maximize their chances of reaching spawning grounds in the best possible conditions.

We can celebrate 2026 and demand better at the same time. This is not contradictory: it is exactly what responsible management should do.

Numbered sources

[1] Atlantic Salmon Federation (ASF). Rivernotes August 13, 2026. Phil Monahan (ed.), with Quebec section by Caroline Côté. Published August 13, 2026. https://www.asf.ca/rivernotes-august-13-2026/

[2] Government of Quebec - MELCCFP. Regulatory management of salmon fishing based on river temperature. Updated June 5, 2026. Closure criterion: water >22°C in the morning when the situation is likely to continue; four rivers targeted in 2026.

[3] Ministry of the Environment, the Fight Against Climate Change, Wildlife and Parks (MELCCFP). Salmon Exploitation in Quebec - 2025 Assessment. Official report provided for this analysis. The report covers abundance and exploitation from 1984 to 2025.

[4] Wilkie, MP et al. (1996). Physiology and Survival of Wild Atlantic Salmon following Angling in Warm Summer Waters. Transactions of the American Fisheries Society, 125(4), 572-580. DOI: 10.1577/1548-8659(1996)125<0572>2.3.CO;2.

[5] Lennox, RJ et al. (2020). Mortality of Atlantic salmon after catch and release angle: assessment of a recreational Atlantic salmon fishery in a changing climate. Canadian Journal of Fisheries and Aquatic Sciences, 77(9), 1518-1528. DOI: 10.1139/cjfas-2019-0400.

[6] Van Leeuwen, TE et al. (2024). Water temperature at the time of the catch-and-release event is a better predictor of survival in Atlantic salmon (Salmo salar) than acute water temperature changes before and after. Journal of Fish Biology. DOI: 10.1111/jfb.15667.

[7] Dugdale, SJ et al. (2013). Temporal variability of thermal refuges and water temperature patterns in an Atlantic salmon river. Remote Sensing of Environment, 136, 358-373. DOI: 10.1016/j.rse.2013.05.018.

Methodological note

The 2026 data cited for the Matapédia and Grande Cascapédia rivers are current seasonal data reported by ASF. They are used as local indicators, not as a final provincial assessment. The 2025 data come from the official MELCCFP report. Scientific results on post-release mortality show real variability between studies; they are used here to demonstrate a risk gradient, not to propose a universal mortality rate applicable to each river.