Discussion in 'Steelhead' started by Jeremy Floyd, Jan 23, 2007.
Re: Do you care enough about wild steelhead to stop fishing for them?
I’m not sure if Ricker invented it, and there is another name important in global fisheries management that was once drilled into my brain that I’ve managed to forget after all these years. Nonetheless, anyone can conclude an MSY management strategy by using the spawner:recruit relationships described in Ricker’s book(s). While the model was built on fishery data, MSY as a strategy is a management theory based on the model that was based on the data. That’s my best off-the-cuff explanation. Since MSY hasn’t sustained a single fishery where it’s been used, managers ought to reject the theory. Instead, it appears to me that they argue that the strategy was improperly implemented. Sounds sorta’ like fishery politics, doesn’t it?
Regarding the remainder of your post, see Smalma’s explanation, as it’s more detailed and accurate than what I wrote.
WDFW has access to statistical expertise at every college degree level. I can only say that if fishery statistical modeling were as simple as you seem to assume, this would have been done 10 times over by now. Unfortunately, it hasn’t worked this way with fish populations, at least up til now.
I differ with you. You wanted the model fine tuned enough to accurately predict required steelhead escapement goals for each river system. My position, and I cannot prove it by anything other than the body of evidence, it that the variation within systems equals or exceeds the variability between systems such that fine tuning the model for each river individually, or fine tuning the model using the original 12 index systems, is no more precise than the general model developed about 1984. In theory, if one did annual sampling in individual river basins over a long enough period of time, a more precise estimate for that basin would be the result. I reject the theory because experience and observation have revealed to me that enough key environmental parameters change enough over a relatively few years, that a model would never “keep up” with the moving target of variation.
It isn’t for lack of trying. Many scientists and statisticians from numerous agencies have worked this type of problem for more than the last 20 years. The problem seems to come down to there being huge natural variation and too many uncontrollable variables. In my estimation, lack of statistical precision is a fact of life in fisheries management in the PNW. And to me, that’s really not the issue. I think the issue is whether the models being used are precise enough so that the management objectives of resource conservation and harvest prediction can be met. The answer is yes to the first and usually to the second. Regarding the second (and shifting topics somewhat), harvest predictions are over-estimated half the time, but significantly over-estimated at a much smaller frequency. And that can be offset with an error buffer if the managers choose to do so. As you might guess, some managers prefer a buffer while others do not, and that usually correlates to what interests they serve.
Your bone with WDFW is based on your point of view. If you read the applicable laws, you will find that WDFW serves multiple masters. If it seems that management is more favorable to harvest than to conservation, then you will also quickly note that WDFW operates in a pluralistic political system. Management decisions are a reflection of the prevailing biological, social (legal), and economic (political) conclusions. State law does not direct WDFW to conserve fish (particularly, which is why wildlife are more conservatively managed) at great expense to harvest. Harvest enjoys very significant legal status, and that is why harvest opinions carry so much clout. Lowly steelhead anglers are a pretty small speck of dust in the realm of statewide fishery management. When it comes to the law, you don’t get it your way. Many a manager and Commissioner would prefer more conservation of salmon and steelhead, but they are constrained by law. If you don’t like that, I suggest you join a group that seeks to modify the laws.
With respect to your comments to Smalma:
1. The Methow wild steelhead were made functionally extinct by having to run 9 mainstem dams, downstream and up. Hatchery steelhead help maintain natural production where it otherwise would either disappear entirely or functionally disappear.
2. At present, most river systems, whether they support weak runs or strong, are stocked with hatchery smolts. Would managers like to experiment with some to see what could be achieved? Yes. Is that alternative politically viable? Not yet, due to social and political repercussions.
3. Don’t know about zero methodology (whatever that means), but there is some genetic evidence (described earlier in this thread) that temporal separation has been effective in maintaining intact wild steelhead gene pools.
I must be an extremely poor writer - you seemed to have missed much of what I was trying to get to.
regarding the need to continue to stock fish in specific drainages - see Cascadekiller's and Salmo g.'s posts.
Regarding the separation of hatchery and wild steelhead spawning - I don't think I said anything about the timing of the planting of the smolts. Rather I was saying that selection of the earliest maturing hatchery fish allowed for the development of hatchery fish that would spawn at a different time than the wild fish thus eliminating the potential of genetic interactions between the two.
Regarding that stocking must continue - Please not I was referrring to the conditions 25 years ago. believe it or not things and management do change and it drives me nuts that folks refuse to recognize as new information is developed management does change. In fact I can thing of several systems where planting has ended - the Wenatchee, Entiat, Nisqually, Cedar.
I remain a skeptic when it comes to the use of integrated hatchery programs. In the middle of the last century there wasn't much in the literature about the dangers of hatchery fish; in fact as I pointed out the thinking was they were more than adequate replacement of wild fish.
However what we do know is that a successful integrated hatchery program requires several factors which include that the brood stock used is representative of the wild population with which they are going to integrate with and that the longer a fish is in the hatchery environment the less likely the resulting product will be the same as the natural produced fish. Those requirements indicate that steelhead are of the anadromous salmonids are the most difficult to develop a program that will be well integrated with the wild population.
The complex life histories, run timing, spawn timing, diversity of stocks (most large western Washington system contain 3 or more stocks), and extend hatchery rearing timed (In the wild it takes approx. 2 years to grow a smolt) all conspire against the development of hatchery program that would be well integrated with the wild population. In some cases such as the Methow such an approach may well be the best alternative in a bad situation. In other cases a poorly integrated hatchery program may well create more problems than it would solve.
thanks for your responses smalma and salmo g. i will continue to disagree with some of your positions.
your clarifications are appreciated and understood.
statistical modelling is used for all sorts of complex problems. i am sorry, but your explanation of the issues involved for fisheries management simply points out that building your model is not a priority. i will not delve into the detail of various models of behavior with which i am famaliar other than to say our present telephone switching system was derived by modelling human behavior, as one small example. obviously there was zero chance of understanding in any precise fashion what was going on in someones mind. but there you are, you pick up, dial and you are connected to the other side of the world. of course there are unknowns, i already said that. that is precisely why you turn to a statistical model. running the data for decades back, from whatever watershed, should provide some determination of variance which can be put to good use in tightening projections. what may be needed here, is some fresh thinking regarding an attack strategy. simply doing what you have been doing is obviously not producing any better predictive value.
you do have the perfect opportunity to try something different with the elwha, hope you lobby for no stocking of steelhead and make that stick. the columbia r. dams are a large problem for fishes but that simply does not apply to undamed rivers on the westend, as an example.
sure politics plays a role, but science can only trump that card when science gets bold enough to take the high road. if WDFW is not the major voice for conservation as opposed to consumption, the remainder of our native fishes will be extinct, just like those on the methow.
all the best to yah both.