How Evolutionary Patterns Helped Find the Gnarliest Bird You’ve Never Heard Of

Here is how to make a model of a Kittlitz’s murrelet: take a nerf football, stretch it out lengthwise until its about 10 inches long, give it paintjob more or less like a dirty ashtray, throw on some stubby wings capable of sustained flight over 100 mph (which are also excellent for “flying” underwater), add a pair of webbed duck feet, and voila, you have a passable model for one of the strangest members of the seabird family Alcidae (Photo: J. Lawonn/USFWS). If you extend the model further, place your model atop a single giant egg on a steep slope of rocks on the flanks of a tall mountain, all preferably within sight of a huge Alaskan glacier. Really. In breeding ecology, behavior and plumage, the Kittlitz’s murrelet (KIMU) is one of a very small clade of supreme oddballs, the Brachyramphini, within the Alcidae, a family that otherwise shows remarkable similarity among species. Like the KIMU, all alcids share a similar body plan and forage exclusively at sea, and all have wings that are optimized for both fast flight and swimming efficiency. But that is where the similarities end. All alcids besides the Brachyramphini have a penguin-like black and white “countershaded” plumage pattern, nest colonially (and conspicuously) immediately adjacent to the sea, and breed almost exclusively on islands or cliffs that are inaccessible to terrestrial predators.

I began studying KIMU breeding biology in Kodiak, Alaska, in 2008 as part of a conservation effort to determine the causes of a dramatic population-wide decline that had been detected by boat-based surveys. Through the course of my research I have used the extreme phenotypic differences and the evolutionary relationships apparent in the Alcidae, as well as inferences made from patterns in convergently evolved species, to make predictions about emerging aspects of the breeding ecology of KIMU, and make hypotheses about this intriguing species.

Kittlitz’s murrelet has been the enigma nonpareil among the alcids, and a large part of this owes to the extreme difficulty researchers have had in finding their nests. They breed non-colonially in rugged alpine areas, nest up to 75 km inland (where predators are often abundant), are cryptically colored, and they are extremely secretive near their nest sites (Day, Kuletz, & Nigro, 1999). Its habits are so mysterious and its Beringian alpine breeding habitats so remote that only 25 nests had been described prior to 2005. I used the smattering of anecdotal data from these previous nests, along with information about its relatively well-described kin, the marbled murrelet (MAMU), to design an expedition that would yield a large number of nests. My team knew about the close evolutionary relationship between the KIMU and MAMU from phylogenetic relationships determined from DNA and morphological analysis (Cracraft et al., 2004; Friesen, Baker, & Piatt, 1996; Moum, Johansen, Erikstad, & Piatt, 1994; Strauch Jr, 1985) (Figure 1, Friesen et. al, 1996). Since the two species appeared to have diverged only about 2 mya, I inferred that the breeding biology of the two species would be much more similar than a comparison between KIMU and other clades of the Alcidae, since many of these clades radiated from a common ancestor about 18 mya, and would be expected to have independently evolved many derived traits that would not be shared with the KIMU. Using MAMU as a proxy species, I made the predictions that: 1) KIMUs nest in a very dispersed manner on the landscape to avoid predation; 2) KIMUs exhibit nest area fidelity, meaning that pairs usually return to the same place every year to breed. I also made predictions based on the commonalities among the dispersion patterns of other cryptic species such as grassland birds and peppered moths to infer that these groups convergently evolved their dispersion strategies (i.e. these species are well-spaced throughout their potential habitat) to avoid predation.

So, following these predictions, I prepared an expedition that could move great distances and cover as large an area as possible to account for high dispersion of nests, and that would be reproducible in following years to find birds that were nesting in the same area annually. The effort turned out to be a success. By spending up to three months at a time backpacking among likely alpine habitats, my group discovered 53 active KIMU nests during 2008-2011, generating a wealth of new information about the breeding ecology of the species. And, as is usually the case in science, we also came up with a lot of new questions.

The first and most obvious question we had is: why are these birds doing such a poor job of reproducing? Only 9 of 53 (17%) nests produced a fledged chick, a very low figure that has recently been corroborated by other studies in different areas of Alaska. Again, looking at phylogenetic relationships we saw that most of the Alcidae experiences much higher rates of nest success. We have no idea whether our small window of data on KIMU is representative, but we do know that the rarity of the species, and the relatively small area of breeding habitat available to it, that it may never have been abundant in recent history. “Recent history” is the operative phrase. The combination of KIMUs affinity for glaciated habitats, and for foraging in glacial outflow areas, and it’s apparent emergence from a common ancestor with the MAMU 2 million years ago, suggests that KIMU may in fact be an ice age relic species, adapted to a eking out its nesting period on bare mountaintops surrounded by glacial ice, as well as in the scree and talus fields created by retreating glaciers. Such habitats would be much less common today than they were during the major glacial retreats that occurred about 10,000 years ago, when large areas of Alaska were melting out from a blanket of ice several kilometers thick. Certainly this may only be circumstantial evidence, but it’s tempting to think that the shrinking glaciers of today may be leading to the shrinking population of KIMUs as well. This hypothesis could be tested, though at great expense, from a large scale study of nest sites and their relationships with glacier dynamics.

But how long do KIMUs live? If their life spans are long enough, even a 17% nest success rate would be sufficient to sustain the population. As you might expect, however, we have no clue about the average adult mortality of KIMU, so we go back again to the phylogenetic drawing board to determine how long its closest relatives typically live. The results are a little sobering: a 15 to 20 year average life span is what is predicted for MAMU, a species that also happens to be in decline through most of its range. Though models of MAMU survivorship and nest success can’t be expected to fully explain KIMU population dynamics, they do provide our best guess as to how long KIMU adults would need to live on average for population stability, and our best models suggest that 15-20 years may not be long enough.

It’s also tempting to compare the breeding strategy and plumage of the KIMU with its alcid kin. Since most of the Alcidae are colonial, nest very close to the sea, and exhibit countershaded coloration, we could presume that these traits are somehow adaptive. KIMU does not share these traits, presumably because of trade-offs resulting from nesting in predator dense areas: it would be a liability to have contrasting coloration and dense colonies where foxes, for example, may be abundant. As a result of these trade-offs, KIMUs may need to spend more time and energy foraging and provisioning fish than other alcid species, meaning that changes in food supply, water temperature, and other factors may have a disproportionate impact on KIMU, relative to other alcids. Again, this is a testable hypothesis, one that I am exploring by studying the diet composition and growth rates of KIMU chicks.

The patterns of evolution have been fundamental for making predictions and hypothesis for a seabird that whose breeding habits were formerly almost completely unknown. Not only have I used phylogenetic relationships to make informed hypotheses relating to dispersion and breeding ecology of KIMUs, but managers will use inferences informed by phylogeny to make educated guesses about life history parameters such as adult mortality for which data does not yet exist. Far more than a merely academic pursuit, principles of evolution may help us to ensure that this cagey little seabird will continue to be found for a long time in the Alaskan alpine gnar. If you look hard enough, that is.

Cracraft, J., Barker, F. K., Braun, M., Harshman, J., Dyke, G. J., Feinstein, J., Stanley, S., et al. (2004). Phylogenetic relationships among modern birds (Neornithes). Assembling the tree of life, 468–489.

Day, R. H., Kuletz, D. J., & Nigro, D. A. (1999). Kittlitz’s Murrelet (Brachyramphus brevirostris). In A. Poole & F. Gill (Eds.), The Birds of North America. Philadelphia, Pennsylvania: The Birds of North America, Inc.

Friesen, V., Baker, A., & Piatt, J. (1996). Phylogenetic relationships within the Alcidae (Charadriiformes: Aves) inferred from total molecular evidence. Molecular Biology and Evolution, 13(2), 359.

Moum, T., Johansen, S., Erikstad, K. E., & Piatt, J. F. (1994). Phylogeny and evolution of the auks (subfamily Alcinae) based on mitochondrial DNA sequences. Proceedings of the National Academy of Sciences, 91(17), 7912.

Strauch Jr, J. (1985). The phylogeny of the Alcidae. The Auk, 520–539.