Thursday, January 13, 2011

How Computers Help Us to Understand Mark Selective Fishing

Mark selective fishing is a strategy to protect endangered wild salmon and steelhead while allowing fishers to harvest abundant hatchery fish. The idea is to release unmarked wild fish and keep only hatchery fish marked with an adipose fin clip. Releasing unmarked wild fish is an easy concept to understand but estimating the impacts to the unmarked fish is a lot more complicated. There are infinite combinations of unmarked to marked ratios, size of mark selective fishery relative to non-selective fishery, and harvest rates, all of which affect the outcomes.

For problems that are too complicated to solve with a pencil and paper, we turn to computer models, just like forecasting the weather or the stock market, designing airplanes and bridges, you name it. For “what if” type of questions, a common technique is called Monte Carlo simulation because probabilities to a computer are not much different than a person throwing a dice. For our mark-selective fishing question, we add another technique where we track the fate of each and every fish in the fishery, all two hundred and fifty thousand of them.

Tell the computer to pick a fish at random. Is it marked or unmarked? Pick a fishery at random. If the fishery is non-selective, then flag the fish as caught. If the fishery is mark selective and the fish is marked, then flag the fish as caught. But if the fish is unmarked, then add one to the number of times the fish has been released. Not all fish survive being released because survival depends on where they were hooked and how they were handled. Pick another random number. If it is better than the odds, then the fish is still alive to be caught again. Otherwise, flag the fish as dead. Pick another fish at random until the catch quota has been achieved and add up the number of dead unmarked fish.

You have to repeat the procedure many times, until the average of all trials is stable. The result is an example of how we use computers to help us understand complex problems that are difficult to observe in the field or, in this case, under water.

Submitted by Henry Yuen

No comments:

Post a Comment