Important Biodiversity

In the Nov 3 Journal Science, Boris Worm and others predict that all the ocean's populations of fish and invertebrates will have decreased to pitiful numbers by the year 2048. They link current and recent declines in biodiversity with exponential declines in marine food resources, water quality, and, worst, the potential for future ecosystem recovery.

Exponentials are potent. "Exponential" does not simply mean "very fast," or "a lot," it specifies a kind of decay (or growth) with a constant halving time. For example, if an exponentially decaying population had 200 members yesterday and 100 today; it would have only 50 members tomorrow. By next week it would be down to 1 individual. So with exponentials, big numbers fall fast and bigger numbers fall faster.

It's easy to imagine that biodiversity, fish populations, water quality, and ecosystem robustness would vary together. Perhaps some kind of pollution would hinder all these things. Worm and the others suggest a causal role for biodiversity itself.

This also makes sense considering that a diverse group of primary producing species might exploit more environmental niches than a homogeneous group, and that herbivore and predator species may have different food needs at different life stages. A similar "portfolio" effect would be especially important in the face of aggressive, selective fishing practices. If catches of an overfished species declined, fishers would be able to switch to another species and let the first recover.

Worm and the others present a survey of many previous studies on the small, medium, and large scales, and the important role of biodiversity emerges.

In 32 small, controlled experiments, higher diversity of species like algae, seagrasses, and planktons led to improvements in ecosystem functions like population growth, nutrient cycling, and stability in the face of perturbations. Since these experiments had controls, the causal role of biodiversity seems certain.

12 Medium-sized coastal ecosystems lost similar functions as they lost biodiversity: fishery catches declined, nursery habitats shrank, and natural filtering mechanisms, such as the cleaning effect that wetlands can have on runoff, failed. These real-world studies lack deliberate controls, so it is not certain that biodiversity was a causal player. Presumably, biodiversity played a similar causal role as it did in the smaller studies.

The most startling effect was visible among the world's 64 largest marine ecosystems. For these, Worm and the others define a population collapse as membership falling below 10% of the historic maximum. They show that of 65% of all known fish and invertebrate taxa (the genus Homo and the species sapiens are two examples of taxa) have collapsed since 1950! This is the accelerating trend that points to 2048.

New England Lobsters

The predictions of Worm and the others rely on two assumptions: that the causality of biodiversity on ecosystem functions is consistent on all scales, and that all of the studies surveyed have good data about their taxa populations and catches. One study of New England lobsters (not included in the Worm survey) shows that estimating population size from fishing catches can be misleading.

In order to estimate population size from catch size, the two quantities must be strongly dependent: if there are more lobsters (or fish) out there, you should be able to catch more. This is a straightforward assumption, but not necessarily valid.

First of all, check out the lobsters ignoring and escaping from this trap used in a study conducted by Steven Jury and others off the coast of New Hampshire:

Watch the video of lobster escaping from kitchen.

The top half of the lobster trap is called the kitchen. It is where lobsters are supposed to enter and eat some bait. Then they are supposed to move into the parlor and, if they are too big for an escape vent, stay trapped. But that plan fails. The large lobster in the kitchen finds the way out through the kitchen door.

Many other lobsters approach the trap and don't enter. Even if several of the peripheral lobsters in the video are the same individual, they clearly are not bound to enter the trap. These observations question the first assumption of how lobster traps work.

There is also a larger dynamic that obscures trapping estimates of lobster population. The phenomenon is called "saturation," but it could also be called bullying:

Watch the video of larger lobster chasing away smaller lobsters that are entering the trap and chasing a smaller lobster out of the trap.

It takes some staring to see: the bully lobster in question "runs" down the left side of the image and around to the right side. It patrols the right side for a while, gets some bait from the right side of the kitchen, then runs back around and into the left side of the kitchen to evict the smaller lobster feeding there. This observation suggests that lobster traps saturate with bullies to the exclusion of additional lobsters.

If only a few lobsters in the area enter the trap, and only some of those stay around to get caught, then traps are a lousy way to estimate lobster population. Consider the following:



The graph shows imaginary lobster catches from traps that saturate (in green) and from hypothetical non-saturating traps (in blue), each as they might vary given small and large wild populations.

There are also two backwards "F" shapes that represent population estimates from trapping. The red "F" is consistent with our initial straightforward assumption: there aren't many lobsters out there so we don't catch that many. The black "F" is not consistent: if I tell you there was a small catch of lobsters, you won't know which "F" leads you to the size of the wild population. In this sense, the two quantities are not strongly dependent.

Trap saturation doesn't mean that lobster populations can't be estimated; only that it is better estimated at the level of a whole fishery, not trap-by-trap. This is one kind of caveat which suggests taking the conclusions of Worm and the others with caution. If, perhaps, several of the studies in the Worm survey suffered from similar kinds of saturation bias, then the prediction of total fishery collapse by 2048 would be too pessimistic.

Since the many studies in the Worm survey were mutually consistent, it would seem unlikely that their prediction is too far off. If there are looming biases, further studies like this by Jury and others should illuminate and help correct them.

Further population research, and conservation of ecosystems on all global scales will cost money, but that should be no obstacle here. Worm and the others point specifically to economic benefits like larger, easier fishing catches, fisheries that are more resilient to perturbations and stresses, and increased tourism revenue. In this case, money should speak for itself very well.

« Previous Next »