I fall in love easy. Must be my Latino upbringing. Whatever it is, I have no choice on the matter. So for five years and counting, I have been passionate about lingcod (Ophiodon elongatus) and rockfish (Sebastes spp.), upper- and mid-level predatory fishes on rocky reefs of the Northeast Pacific.
Lingcod are beautiful and fierce. Rockfish are cosmic. Both taste mighty good and—surprise, surprise—have been overfished to smithereens throughout much of their range. Howe Sound, my field site near Vancouver, British Columbia, is no exception, although new protective legislation might be starting to give them some slack.
Our dive surveys1 and earlier studies, in combination, have pieced together a story of ecosystem change. In the Howe Sound of today, lingcod rarely exceed body lengths of 80 cm. But up to 30 years ago, when overfishing had yet to inflict the full extent of its current damage, lingcod with lengths of 90 to 100 cm had been common in the area. There is nothing unique about this; most fisheries target the biggest individuals, ultimately reducing maximum body size within each species of predatory fish.
As predators shrink, the vibrant tension of predation risk slips away. The mechanism of change has a lot to do with mouth size. Predatory fishes swallow prey whole, usually head or tail first, so it is impossible for them to eat prey bigger than the width and height of their open jaws. And bigger fishes have bigger jaws, which makes them capable not only of consuming larger prey, but also of scaring bigger prey into using antipredator behaviours, such as hiding in rocky crevices. As predators shrink, big prey enter a size refuge and only small prey remain at risk, which can alter trophic cascades and other indirect species interactions.
Variants of this phenomenon—in which smaller mesopredators become the new top predators of over-exploited communities—keep sprouting everywhere on land and in the water2. Mulling over this state of affairs, it dawned on me one day that understanding how antipredator behaviour varies between different species of mesopredators while top predators are still around might help ecologists predict shifts in exploited food webs. This idea emerged not from careful thought but rather from the dregs of failure.
I was running preliminary trials for a field experiment, presenting rockfish with a food reward—live shrimps of the genus Pandalus tethered to a chain—eventually to test how the presence of a large lingcod, in the form of a fibreglass model, might affect their behaviour. My expectation was that rockfishes would feed on shrimps without the model lingcod—which was still being made, meticulously and sloooowly, by a taxidermist—but once in place, the model lingcod would scare rockfish away from shrimps: a slam dunk into experimental evidence showing that large lingcod affect the feeding rates of rockfish on invertebrate prey, which could potentially underlie a trophic cascade3. Simple. And, as it turned out, horribly rockfish-centric.
A different mesopredator was quick to remind me that reality can derail well-dreamt experiments. During the preliminary trials, kelp greenling (Hexagrammos decagrammus)—who I had barely acknowledged in prior thoughts—kept nailing shrimps ahead of everyone else. Rockfish were abundant on the reefs and did attack the shrimps, but at a much slower rate than kelp greenling. And it was not a mere matter of kelp greenling being quicker at finding shrimps; video from fixed video cameras showed that rockfish often were first to inspect shrimps from a distance. Perhaps rockfish were being more cautious than kelp greenling in assessing perceived risks associated with the chains and plastic ties to which shrimps were tethered. After all, these were novel structures with which reef fishes had not evolved, so species differences in willingness to attack shrimps could turn out to be interesting. Still, I was not enjoying what appeared to be a waste of my funds and get-wet-and-cold currency. So I would go home after a field day, hang my dripping dry suit from the porch rafters, crank up some angry rock music and review video footage from fixed cameras while trying to summon loving thoughts about kelp greenling. Eventually those thoughts crystallised into two words: life history.
Thanks to kelp greenling, I began to see that the reef mesopredators of Howe Sound—including subadult lingcod at risk of cannibalism from adult lingcod—encompassed a spectrum of life history speeds. Copper and quillback rockfish were at the slow and slowest end of the spectrum. Copper rockfish live to 50 years and most individuals do not reproduce until 6 to 7 years old. Quillback rockfish live to 95 years and most individuals do not start reproducing until 11 to 20 years old. To top it all, rockfish mothers produce more abundant and stronger young as they age. For rockfishes, producing the most offspring during their lives might require living cautiously and avoiding predators—even at the cost of missing out on meals—especially for quillbacks that live longer and begin reproduction later in life. In contrast to rockfishes, lingcod and kelp greenling have much faster life histories. They live only 20 and 12 years, respectively, and begin reproducing earlier in life than rockfishes. For them, living dangerously to secure food that enhances short-term reproduction might produce the most offspring during their lives, particularly so for the shorter-lived kelp greenling.
So there you have it: a hypotheses emerging from the scraps of a failed experiment and a little help from theory on state-dependent behaviour4. And a testable one to boot. All four mesopredators have adult lingcod as a common enemy and eat the same kinds of shrimps.
The fibreglass lingcod model eventually arrived. Cast from a 125 cm-long female caught in Alaska, it was scary-looking. From then on colleagues and I spent many dives tethering live shrimp to chains, setting up the model lingcod and the video cameras that would record the action, getting the hell out for three to four hours, and returning to retrieve our junk and survey fish and invertebrates.
The results were way cool5. Kelp greenling, which live the fastest life style, took the highest risks and were the only species to attack shrimps adjacent to the model lingcod. Subadult lingcod attacked shrimps only when these were far from the model lingcod, and copper rockfish attacked shrimps only when the model lingcod was absent from the reef. Quillback rockfish, which live the slowest life style, took the fewest risks, never leaving the vicinity of rocky shelter to attack the shrimps.
The lesson from all this is that different species of reef mesopredators, despite occupying similar positions in the food web, do not necessarily play the same ecological role. Some respond more strongly to predation risk than others, and these differences seem to be predictable from life history characteristics. This sort of information can be used to predict how ecosystems might change in response to the loss of top predators. The study also made me realise that we should keep a tighter eye on kelp greenling. With their greater gusto for risk-taking and lower appeal to fishers—who would rather eat lingcod and rockfish—kelp greenling could well be rising to become the top dog of overfished reefs: a scenario similar to exploited terrestrial communities where former top predators like wolves are replaced by mesopredators with faster life histories like coyotes.
In fact, the potential for analogues in terrestrial and freshwater ecosystems is my main motivation for writing this post. Would your study system be amenable for testing whether life history characteristics explain differences in risk-taking by different mesopredators? Would that line of inquiry improve predictions on the ecological costs of losing top predators? I’d love to hear.
- Frid A, Connors B, Cooper AB, Marliave J (2013) Size-structured abundance relationships between upper- and mid-trophic level predators on temperate rocky reefs. Ethology Ecology & Evolution doi:10.1080/03949370.2013.798350
- For one of many reviews see: Prugh LR, Stoner CJ, Epps CW, Bean WT, Ripple WJ, Laliberte AS, Brashares JS (2009) The rise of the mesopredator. BioScience 59: 779-791. doi:10.1525/bio.2009.59.9.9
- Frid A, Marliave J (2010) Predatory fishes affect trophic cascades and apparent competition on temperate reefs. Biology Letters 6: 533-536. doi:10.1098/rsbl.2010.0034
- Clark CW (1994) Antipredator behavior and the asset-protection principle. Behavioral Ecology 5 159-170. doi:10.1093/beheco/5.2.159 (see also other papers and books by Marc Mangel and Colin Clark, and by Alasdair Houston and John McNamara)
- Frid A, Marliave J, Heithaus MR (2012) Interspecific variation in life history relates to antipredator decisions by marine mesopredators on temperate reefs.PLoS ONE 7(6): e40083. doi:10.1371/journal.pone.0040083