The point I was trying to make is that pinniped harvest from nets is significant now and is not considered in harvest allocation models. (Release mortality is a different subject.)

In my observation of gill nets there is often a problem with fish dropping out as the net leaves the water and rises into the boat. Some fishermen even hold a landing net under these fish for insurance. I assumed that this was what the 2-3% drop out referred to as it would be a reasonable estimate of those losses.

My understanding of the official run size is that it is the summation of harvested fish including NDO plus release mortality plus spawner counts. Net drop out was not accounted for in the models prior to the 2014 model year. In 2014 NDR was added as 2% for coho and 3% for chinook in both Grays Harbor and Willapa. This increased the harvest mortality and caused all the historical run sizes to increase. These values have been used for the past three years as well as a 5% drop off for sport catches. I doubt if any of these include pinniped mortality. When I look at the model for tribal harvest I cannot find any calculation for net drop out. It appears to me that only non-tribal harvests are charged with this loss reducing their allocation.

If I were a modeler trying to model harvest mortality (but not subject to politics) I would consider including pinniped losses to be critical for accurate run management. My first guess at numbers would be 40% of the fish that impact a manned set net are lost to pinnipeds. Unmanned set nets would be much higher. I would consider drift nets to lose 30% of the fish that engage the net and then request a research project be initiated to get better data.