Photo: National Geographic Photo Ark
Are California's Chinook salmon populations physiologically distinct, and does that matter for their survival under a warming climate?
Chinook salmon (Oncorhynchus tshawytscha) are one of the Pacific Coasts most ecologically and culturally important species and, unfortunately, one of the most imperilled by climate change. As rivers warm, questions about which populations can tolerate higher temperatures, and why, have become urgent management priorities.
My dissertation research tackled this question at scale, designing and conducting some of the largest metabolic performance experiments ever performed on teleost fishes. Working across eight hatchery populations spanning California, Oregon and Washington's diverse river systems, I measured aerobic scope, growth rate, and critical thermal maxima to ask whether populations differ physiologically, and whether those differences reflect local adaptation to historical thermal environments.
The answer was a clear yes. Populations differ substantially in how they perform across a temperature gradient, in ways that are consistent with adaptation to local conditions. These findings have direct implications for how we prioritize conservation and manage water for salmon under ongoing climate warming.
Lab performance and field survival tell different stories — and closing that gap is the key to protecting salmon under climate change.
This research shows that salmon populations differ physiologically in meaningful, management-relevant ways. However their impressive laboratory performance masks real ecological vulnerabilities that may only become visible when food and predators enter the picture.
The central contribution of this review is a distinction between two complementary ways of understanding how temperature affects fish:
Fundamental thermal physiology describes what an organism is physiologically capable of under controlled, idealized conditions — its thermal performance curve, aerobic scope, critical thermal limits, and optimal temperature measured in the lab with clean water, ample food and no competitors. This captures intrinsic physiological capacity and is where population-level differences in thermal adaptation are most clearly observed.
Ecological thermal physiology is the limiting of the fundamental physiology by environmental forces in the wild — where food may be limited, predators exploit thermal performance mismatches, and the costs of suboptimal temperatures compound over time. This is where physiology meets ecology, and where laboratory performance routinely fails to predict field outcomes.
The review found that most salmonid populations are considerably more thermally tolerant in the laboratory than field survival data would predict. This gap implies that ecological constraints — not physiological limits alone — are often what make high temperatures lethal for wild fish. Identifying and quantifying those constraints became the organizing focus for all subsequent work on this page.
Studies 02–04 characterize differences in fundamental thermal physiology among Chinook salmon populations. Studies 05–06 and the spinoff projects address the ecological constraints that determine realized survival in the wild.
Eight hatchery populations of juvenile Chinook salmon were acclimated to a common set of temperatures and measured for aerobic scope — the difference between maximum and resting metabolic rate — across a thermal gradient spanning their ecologically relevant temperature range.
This is one of the largest metabolic performance experiments conducted on any teleost fish, involving hundreds of respirometry trials across populations and temperatures. Tightly controlled rearing and acclimation conditions ensured that observed differences reflect true physiological divergence among populations.
Populations showed clear and statistically significant differences in their thermal performance curves. Populations from historically warmer river systems tended to have higher optimal temperatures for aerobic performance and maintained aerobic scope at higher temperatures than populations from cooler systems. This pattern is consistent with local thermal adaptation, fish are physiologically tuned to the temperatures of their native rivers
Importantly, no single population performed best across all temperatures, suggesting real trade-offs in thermal specialization.
California's Chinook salmon are divided into four seasonal "runs" — winter, spring, fall, and late-fall — that differ in their timing of ocean entry, river migration, and spawning. These runs encounter very different temperature regimes during their freshwater residency, providing a natural experiment in thermal adaptation.
This study compared thermal physiology among populations representing each seasonal run, using the same aerobic scope framework as Study 02 but with a focus on whether run-timing is associated with distinct physiological profiles.
Seasonal runs did show physiological differences consistent with the temperatures they encounter during their freshwater life stages. Winter-run Chinook — which migrate and rear in the coldest conditions and are listed as Endangered — showed the lowest thermal optima and the greatest sensitivity to warming temperatures, highlighting their particular vulnerability to climate change.
These findings provide a physiological basis for understanding why different runs respond differently to thermal stress, and why a one-size-fits-all thermal threshold for water management is insufficient.
This study moved beyond aerobic scope to examine acute thermal tolerance. Critical thermal maximum (CTmax) was measured across populations and acclimation temperatures using standardized ramping protocols. Critically, the same populations used for aerobic scope measurements were also assessed for CTmax, allowing direct examination of the relationship between these two performance traits.
At the individual level, fish exhibited a trade-off between thermal tolerance and plasticity — those with higher baseline CTmax showed less capacity to adjust that limit through acclimation or heat hardening, and vice versa. Interestingly, this trade-off did not hold at the population level, suggesting that individual variation and population-level adaptation operate through partially distinct mechanisms.
Thermal plasticity was associated with the thermal environment of origin: populations from historically warmer habitats demonstrated greater acclimation capacity and heat hardening ability than those from cooler systems. Populations from colder habitats — most notably the Trinity population — showed both lower thermal plasticity and limited capacity to buffer against warming, marking them as particularly conservation-vulnerable.
These findings underscore the importance of accounting for interpopulation differences when predicting species-wide responses to climate change. Aggregate or average thermal tolerance metrics obscure meaningful variation in which populations are most at risk and least able to adjust — precisely the information needed for targeted conservation planning.
Most populations — winter-run excepted — showed surprising thermal tolerance under controlled conditions. This gap between laboratory performance and field outcomes drove a new research focus: the ecological constraints that make high temperatures lethal even when the physiology says they shouldn't be.
The fundamental thermal physiology studies revealed that most Chinook populations are more thermally tolerant in the lab than field survival data predict. One key hypothesis for this gap: high temperatures impair burst swimming performance — the explosive anaerobic sprint used to escape predators — while simultaneously benefiting warm-adapted predators. If so, temperature could drive mortality through trophic interactions even at physiologically sublethal levels.
This study combined burst swimming performance measurements across temperatures with field-based predation trials to directly test whether temperature-induced impairment translates into increased predation vulnerability in juvenile Chinook salmon. It is one of the first studies to link laboratory thermal performance data to field predation outcomes in salmonids, and provided early field validation of the laser-timed burst tunnel methodology.
Temperature significantly impaired burst swimming, and these effects predicted predation vulnerability in the field — fish performing worse at high temperatures were more susceptible to predators. The results provide direct empirical support for the ecological thermal physiology framework: temperature reshapes predator–prey performance mismatches in ways that determine survival independent of physiological thermal limits.
In drought years, California rivers simultaneously warm and experience reduced prey availability — thermal and nutritional stress co-occur precisely when salmon are most vulnerable. The fundamental physiology studies assumed well-fed fish, but wild salmon rarely have unlimited food. If nutritional status modifies thermal tolerance, then lab-derived performance curves may be unrealistically optimistic for real-world conditions.
This project directly tests whether restricted food intake collapses the thermal performance observed in well-fed fish — extending the fundamental vs. ecological framework to include resource availability as a key constraint on realized performance.
Juvenile Chinook salmon are subjected to a range of feed ration levels — from satiation to severe restriction — then assessed for aerobic scope and CTmax using the same protocols as the earlier studies. This allows direct, controlled comparison of thermal performance under ecologically realistic nutritional conditions versus the idealized laboratory baseline.
We predict that feed restriction will reduce aerobic scope and lower CTmax, and that these effects will interact with temperature to compound risk — bringing predicted performance closer to field conditions than fundamental physiology data alone would suggest. Two manuscripts are in preparation.
Controlled laboratory experiments are essential for establishing mechanisms, but individual fish trials cannot directly answer the management question that matters most: across an entire watershed, under realistic thermal regimes, how many salmon are physiologically at risk — and where? Bridging this gap requires embedding population-specific physiology into spatial and temporal frameworks that operate at the scales where conservation decisions are made.
This collaborative work integrates thermal performance data from the Chinook salmon physiology program with river temperature models and individual-based population frameworks. By coupling empirical performance curves to landscape-scale temperature scenarios, it becomes possible to forecast how physiological stress is distributed across populations, life stages, and river reaches — and how those distributions shift under climate projections.
The approach treats physiology not as a fixed endpoint but as a spatially and temporally dynamic variable — asking not just whether a population can tolerate warming, but when, where, and for how long individuals within that population will be operating above critical thresholds.
Individual-level thermal tolerance data, no matter how precise, has limited management utility unless it can be scaled to population outcomes. This line of work represents a broader interest in applying mechanistic physiology at the scales relevant to conservation planning — translating what we measure in the lab into predictions that inform habitat protection, flow management, and species recovery.
Peer-reviewed papers arising from this research program
Zillig, K. W., Bell, H. N., FitzGerald A.M., Fangue, N.A. Patterns of Interpopulation Variation and Physiological Trade-offs in the Acute Thermal Tolerance of Juvenile Chinook Salmon (Oncorhynchus tshawytscha).
Frontiers in Fish Biology
Zillig, K. W., FitzGerald, A. M., Lusardi, R. L., Cocherell D. E., Fangue, N. A. Intraspecific variation among Chinook salmon populations indicates physiological adaptation to local environmental conditions.
Conservation Physiology, 11(1)
Zillig, K. W., Lusardi, R. L., Cocherell D. E., Fangue, N. A. Interpopulation variation in thermal physiology among seasonal runs of Chinook salmon.
Canadian Journal of Fisheries and Aquatic Sciences
McInturf, A. G., Zillig, K. W., Cook, K.*, Fukumoto, J. A.*, Jones, A.A.*, Patterson, E.*, Cocherell, D. E., Michel, C., Caillaud, D., Fangue, N. A. In hot water? Assessing the link between fundamental thermal physiology and predation risk of juvenile Chinook salmon.
Ecosphere, 13(11)
Zillig, K.W., Lusardi, R.A., Moyle, P., Fangue, N.A. One-size does not fit all: variation in thermal eco-physiology among Pacific salmonids.
Reviews in Fish Biology and Fisheries, 31(1)
An interactive tool for exploring thermal performance curves across Chinook salmon populations — compare aerobic scope, optimal temperatures, and CTmax across river systems.
Interactive data explorer coming soon
Replace this with your ShinyApps.io iframe when ready