Biology Climate Change Ecology

Fatal attraction: Under climate change, fish swimming towards predators, not away from them

Paper: Welch, M. J., Watson, S.-A., Welsh, J. Q., McCormick, M. I., & Munday, P. L. (2014). Effects of elevated CO2 on fish behaviour undiminished by transgenerational acclimation. Nature Climate Change, 2(October), 2–5. doi:10.1038/nclimate2400


Many studies have already been done showing that more carbon dioxide (CO2) in the water (a condition called hypercapnia) can affect the behavior of fish. It causes them to lose their sense of direction, behave strangely, process smells differently, and even increases their stress levels. All of those aspects may seem inconsequential, but when you’re a fish, whether you live or die depends on your ability to get away from predators.

One of the ways that fish communicate in the water to each other is through chemical cues, similar to smells. The chemical molecules dissolve in the water immediately around the fish and spread outward from that point, much like an open bottle of perfume will soon be smelled throughout a room. Some of these chemical signals attract mates, called pheromones, and others communicate danger, called alarm cues. Fish innately know to swim towards the pheromones and swim away from the alarm cues, much like you innately know to go towards the smell of cooking and away from the smell of rotting food.

Studies have already shown that fish raised under high CO2 conditions don’t process chemical signals correctly. Instead of swimming away from these chemical alarm cues, they won’t notice them at all, or worse, they’ll swim towards them. This study is different in that it wanted to see if these fish could adapt to high CO2 conditions. They raised parents in control and higher CO2 environments, then saw how their offspring were responding to these chemical alarm cues.


This study was conducted using a common fish found on coral reefs, the spiny chromis (Acanthochromis polyacanthus), as the model species.


Spiny Chromis
Spiny Chromis

The researchers tested their hypothesis, that these fish would be (hopefully!) be able to adapt to higher CO2 conditions, using a purely factorial design illustrated in the diagram below. The reason they chose to do this was to test out all possible combinations of parent and offspring conditions to get the most correct picture of what was happening to these fish.

Experimental design, testing all the alternatives.
Experimental design, testing all the alternatives.

They raised parents in control conditions (representing CO2 levels right now), mid-CO2 conditions (what’s expected in 50 years) and high CO2 conditions (what’s expected by 2100). They then took the eggs and raised them in one of the three CO2 levels, then tested their sense of smell.

To do that, the researchers used a flume tank, illustrated below. The researchers pumped one side with regular, untreated seawater, and the other side with seawater treated with the chemical alarm cue. The fish is free to swim back and forth between the sides, and the researchers can calculate how much time the fish spends in either side of the tank to determine its preference. Each trial involved 2 minutes of the fish in this tank, then a rest period, then 2 more minutes of the fish in the tank with the treatments switched to eliminate any preference the fish had for the side of the tank.

Visual explanation of how the flume tank works.
Visual explanation of how the flume tank works.


The offspring showed the same trends regardless of the CO2 level in which the parents were raised. The offspring that were raised in a high CO2 environment sent an average of 75% of their time in the tank swimming in the side with the alarm cue. That suggests that they’re not recognizing the cue as something they should be afraid of, and instead something they should be attracted to. The offspring raised in the mid-CO2 environment were moderately attracted to the alarm cue and spent about 55% of their time on that side of the flume tank.

The offspring raised in the control environment showed a more variable result. If the parents were also reared in that control environment, the offspring showed the “correct” response to the chemical alarm cue and stayed away from it (only 10% of their time was spent on that side of the tank). However, if the parents were raised in a higher CO2 environment (either mid or high CO2), the offspring spent more time in the alarm cue (30% of the time). That suggests that even if the oceans go back to normal CO2 levels, those offspring are already likely to spend more time not responding correctly to the alarm cues just because their parents came from a high CO2 environment.

Results of the chemical alarm cue experiment.
Results of the chemical alarm cue experiment.


These results illustrate that there is very little to suggest that these reef fish will be able to adapt to a high CO2 ocean. Even when the parents also came from a high CO2 environment, the offspring were at a significant disadvantage for dealing with that high CO2 and went towards the cue that normally indicates the presence of a predator. They showed a small decrease in time spent in the alarm cue (75% with control parents to 70% with high CO2 parents) but overall, the capacity for acclimatization doesn’t look promising. As the ocean gets more and more CO2, these fish will continue to move towards alarm cues instead of away from them and will get eaten more often.

One of the best hopes we have for dealing with climate change in general is that animals will adapt to the new conditions, just like they have to different conditions in the past. However, we are changing the conditions too rapidly for these animals to gradually adapt, and instead we’ll see more results like this that could result in changing food web dynamics and the loss of biodiversity.

Leave a Reply

Your email address will not be published.