Sampling Strategy

This section will be most helpful to those observers that sample onboard. Depending on the RFMO, the catch can be sampled either onboard by observers or by port samplers when the vessel returns to port.

The ideal goal of catch sampling is to determine the age of every fish caught. This information forms the basis of the common stock assessment tools used today. When combined with other stock abundance estimates, such as catch per unit effort (CPUE) and research vessel surveys, various aspects of the population dynamics and life histories of the species can be determined. This enables biologists to assess the size of the fish stocks and make recommendations on how the resource may be harvested without endangering the stocks.

But, of course, there is no readily available technique to make such a direct determination of age, and there is certainly not the time to do so for every fish landed. Instead, the observer must use fish length as a proxy for age (length is well correlated to age and weight of a fish), and collect that data from a random sample of the fish that are brought on board.

One technique for random sampling is to begin with the total number of brails the crew estimates is in the net. A good goal is to sample 10 percent of that total. For example, if there are an estimated 30 brails of fish in the net, the observer would want to sample three brails. Next, select a random number (or have someone else pick a number for you) between 1 and 10. That tells you which brail to begin with. If the random number was 4, for example, you would sample the 4th, 14th, and 24th brails to come on board. If 30 brails was an underestimate and there were really 35 in total, then you would sample the 34th brail as well. If your random number had been 7, then you would sample only the 7th, 17th, and 27th brailer.

For any selected brailer, it is not possible to identify and count all fish. Each brailer can be 3 or 5 tons in capacity, which means 500 fish or more, depending on their sizes. You may need to take a subsample in order keep up with the fish coming on board and ensure an even sampling through the haul. Subsamples can be done via “grab” or “spill” sampling.

Grab vs. Spill Sampling. “Grab sampling” is when an observer pulls by hand fish from a pile. Because of the subconscious human tendency to select larger individuals, many programs are moving away from using grab sampling to “spill sampling” in their sampling strategy. Spill sampling is when the fish are sampled from a receptacle that has been filled directly from a single brail. If this technique is used, the observer is often asked to also note the brail number sampled and the total number of fish measured. Be sure to follow your regional observer training protocols on how to obtain a grab or spill sample.