Fish and Bugs

This post originally appeared on the Software Carpentry website.

The May/June 2012 issue of Washington Monthly has an article by Alison Fairbrother titled "A Fish Story". Near the top, it says, "In 2009, a routine methodological upgrade at NOAA—and the subsequent discovery of a few lines of faulty computer code—forced the start of a profound shift in the ASMFC's estimates of menhaden stocks." A few pages later, we get more details:

In 2009, the Menhaden Technical Committee updated its methodology for estimating the menhaden population—something it does every five years—and then ran the menhaden catch data through a new computer model. The results weren't much different: although the numbers of menhaden were declining, the estimated number of eggs produced by spawning female menhaden was at the target level, so according to the reference point, menhaden weren't being overfished.

Shortly thereafter, a colleague of Jim Uphoff's, a biologist named Alexei Sharov, got hold of the computer model that had been updated by NOAA scientists. Going through the code line by line, Sharov, one of Maryland's representatives on the Technical Committee, found a fundamental miscalculation buried inside the model. Uphoff, meanwhile, studied the methodology of the code and discovered that NOAA had both underestimated the amount of fish killed by the industry and overestimated the spawning potential. Sharov brought these two mistakes to his peers on the committee, and it was agreed that corrections needed to be made.

Several months later, after the model had finished running a second time, the science finally caught up with what Jim Price and the anglers had been saying for decades: even using the lax reference points developed by the ASMFC, menhaden had been subject to overfishing in thirty-two of the past fifty-four years. When the assessment was then peer reviewed by a group of international scientists, the reviewers deemed that the reference point currently in use for menhaden—8 percent of maximum spawning potential—was not sufficiently safe or precautionary.

Furthermore, the number of menhaden swimming in the Atlantic had declined by 88 percent since 1983—to a level so low that it caused George Lapointe, former commissioner of Maine's Department of Marine Resources, to have what he called an "oh shit moment."

If anyone knows more about the "fundamental miscalculation", I'd be grateful for a summary.