The Krill Surplus Hypothesis and the Power of Data

This post was chosen as an Editor's Selection for ResearchBlogging.orgAlmost a year ago, we discussed briefly the Krill Surplus Hypothesis. In this model, the removal of large baleen whales created a competitive release for Minke whales, Balaenoptera bonaerensis, exponentially increasing their food supply and and allowing their population to boom. By removing all other krill eating whale from the Antarctic, Minke whales were allowed to thrive, gorging on an endless supply of krill. The flipside to this hypothesis is that now Minke whales have become competitive excluders of other baleen whales, preventing their re-population post-whaling. Minke whale may be preventing the recovery of other whale species.

This is Japan’s primary scientific and ethical justification for the continued removal of Minke whales from the Antarctic. The earliest reports put forth by Japan’s Institue for Cetacean Research put the Minke population at 8 times greater than it was pre-whaling (Ohsumi 1979), a number that was recently revised to 3-fold (Mori and Butterworth 2006). Provided that number was accurate, the minke whale population could very well have earned its title as “the cockroaches of the sea”. If the Krill Surplus Hypothesis is valid, then a culling of minke whales may actually be the best management strategy to promote the recovery of large baleen whales.

So how do you test that?

Three surveys conducted over the last three decades have estimated the current Minke whale population to be between 338,000 and 766,000 (Branch and Butterworth 2001; Branch 2006). There are no historical surveys of Minke whales pre-whaling. In order to determine what the historical population of Minke whales was, we have to look at genetics.

There is a fundamental misconception when people think about populations over time. Populations do not look like this:

An idealized 'stable' population

Rather, all populations are in a state of flux. Under ideal conditions, they are still subject to stochastic variation each generation. Real populations look more like this:

A more accurate view of a 'stable' population

Even healthy populations look like this, undergoing periods of booms and busts. A robust population can endure the lean generations and recover. These cycles of booms and busts leave a signature in the genetic code. It is easier to lose genetic diversity than to gain it. The increase from generation 2 to generation 4 will marginally increase the diversity of the population, while the rapid decrease from generation 4 to 5 will remove huge amounts of diversity in the gene pool. So any estimate of historic population size is going to be skewed towards the lower end of the spectrum. There are other measures of population size, including mean population and harmonic population size, both of which more accurately reflect the boom to varying degrees of accuracy. Both of which are also more appropriate for estimating contemporary population size.

Back to the Minke whales. Dr. Steve Palumbi and his team sampled Minke whale meat collected from seafood markets in Japan. They amplified several regions of DNA in order to get an estimate of historic population diversity and calibrated that against mutation rates and generation times to produce an estimate of historic population size. They determined that the long-term estimate of population size for the Minke whale was 671,000 (with a 95% confidence interval between 374,000 and 1,150,000 individuals), well within the range for estimates of contemporary population size. In short, the hypothesis that contemporary minke whale populations are larger than historic populations was rejected.

But beyond simply rejecting the null hypothesis, they rejected it with the most conservative possible parameters. They made every effort to confirm the Krill Surplus Hypothesis but could not. It would have been simple to use a different mutation model, to calculate a harmonic population size that would have put the estimate higher, increasing the likelihood that their results would confirm their bias (I can only assume that Palumbi et al are opposed to Japanese whaling). But by collecting and analyzing the data in the most rigorous way, they’ve provided the strongest possible rejection of the Krill Surplus Hypothesis.

This is what I was talking about in “The Data Speak” good data don’t need spin, they need to be publicly available and presented to stakeholders. Steve Palumbi is an advocate, a strong passionate advocate, but his data stand on their own, independent of the scientist’s ideals.

The paper is available online at the link below. All the sequences presented are in GenBank, a publicly available database. The main programs used to analyze the data are all free online. Anyone who wants to can access the data and do the analyses themselves.

As I often say, scientists have a duty to present their findings to the broadest possible audience. Palumbi did just that. Enjoy.

And head over here to see the rest of these short documentaries.

~Southern Fried Scientist

RUEGG, K., ANDERSON, E., SCOTT BAKER, C., VANT, M., JACKSON, J., & PALUMBI, S. (2010). Are Antarctic minke whales unusually abundant because of 20th century whaling? Molecular Ecology, 19 (2), 281-291 DOI: 10.1111/j.1365-294X.2009.04447.x

Comments are closed.