Discussion in 'Steelhead' started by 808steelheader, Jan 31, 2013.
Try this one.
Thanks for posting this. I agree and am also concerned. As I try to understand the cause for the decline in native steelhead I seem to come across many potential causes including deforestation, netting, farming and agriculture, unregulated ocean fishing, etc. I think what you've posted is interesting but really represents a correlation (e.g., more hatchery fish is related to decreases in native fish). The problem is that this doesn't really say much about the actual mechanism involved in the loss of native fish. For example, we can't really say what percentage of loss in native runs is directly attributed to hatchery fish. I think it is hard to narrow this problem down to one or two variables. The problem seems quite complex and involves multiple problems, including hatcheries.
Not to take away from the tremendous amount of work that is presented in these papers and by this group of researchers. Thanks for posting!
Hi Curt, Do you have any other examples of rivers that have been cut off from hatchery plants, and later steelhead populations declined?
I agree. Eliminate the variables we have proof on and then figure out what the "mystery" is after we have done everything we know works.
Regarding other systems where the steelhead populations fell after the ending of the steelhead planting.
The obvious example of course theVAncouver streams where the populations fell some years after the ending of wild brood stock programs.
In Puget Sound the Cedar is an interesting example. During the early 1980s the wild escapements were in the 1,500 range (high year 2,200). The population fell pretty steadily through the rest of the 1980s and early 1990s due to sea lion predation (which no doubt was aggravated by having early returning hatchery fish in the system). By the time the problem sea lions were dealt with the hatchery plantshad been discontinued. Within a steelhead generation the escapements increased to the 600 range and then began falling once again. In recent years the steelhead returns could be counted on a single hand; except for the resident population and its contribution one could consider the anadromous population extinct.
One variable rarely discussed is the unnatural concentrations of fish created by hatchery releases. Predators get used to where their food will be and when. If there are wild fish in the vicinity they are at greater risk.
This may be more meaningful/problematic in the Columbia basin but predators are predators regardless. If we concentrate a food source it makes it easier and more attractive for the predators...any wild fish (alevin, fry, parr or smolt) that happen to be in the vicinity is at much greater risk, and hatchery smolt are predators.
Each spring on the Klickitat the hatchery semi's pull through town accompanied by flocks of seagulls. The seagulls and fish eating ducks follow the mass of hatchery smolts down the Klickitat and into the Columbia where they continue to follow, and of course the pikeminnows are waiting also. Any wild fish in the vicinity is at greater risk. I'd bet >50% of the hatchery releases are consumed well before they get to the Pacific and if there are wild fish in the mix they get consumed at a similar rate...it's PURE economics for the predators.
There is a report available from 2001 that indicates that males have a very wide spawning time and spawn multiple times with multiple females. Females spawn only once and leave. There is apparently a common steelhead management assumption in Washington that only considers female spawning time as the distinguishing trait that separates wild from hatchery interactions. Male hatchery and wild steelhead apparently have significant overlaps with female hatchery and wild spawning times. It is indicated that males are the primary vectors for hatchery/wild interactions. You can access this paper from the Native Fish Society at: http://nativefishsociety.org/conser...ry/special_contributions/strays/Steelhead.pdf
It is also my understanding that hatchery smolts do not all go out to the ocean but stay in the river. Considerable numbers of these are apparently little males that are already mature and ready to spawn immediately on release in spring. They have been observed to be part of the spawning population with wild steelhead in an Olympic Peninsula study. These little hatchery males are there and ready to spawn right when wild females are prevalently spawning in spring along with male wild resident rainbow in the work by John McMillan. There are very good explanations, discussions, and underwater photos in the recent book by the McMillans, May the Rivers Never Sleep.
I wonder about the Cedar often. I wonder what role residualization has on the steelhead population. It would appear that there is little need for the anadromous life history, given that the adfluvial/residents can clearly find adequate nutrients and reach steelheadish proportions. That coupled with the fact that smolts are much less likley to outmigrate once they reach +/- 210 mm. However, it does beg the question 'why would this only happen recently?', especially since the adfluvial life history has been available since they rerouted the river. Maybe the residents could out-compete the steelhead once they fell below a threshold. Maybe it's just the ocean...
I just read through the whole of McMillan's 'report' from 2001 that you provided the link to. Instead of a report, I read an essay with a lot of qualifiers (appears, could be, might be, etc.) in it that purports to show why hatcheries are bad for steelhead. Granted, McMillan states right up front in the beginning that it isn't a scientific paper, which as he states is why he used a forward instead of a summary of his hypothesis and findings.
Unfortunately, it lacks real data that backs his hypothesis that hatcheries are bad for steelhead due to hatchery males spawning with wild females. Granted, he makes that supposition and purports to show why it is happening, but the 'proof' or 'data' in support of his conclusion that his hypothesis is true is not provided. Instead, there is a lot of supposition with no data to back the suppositions. Thus, he offers nothing to back his hypothesis as being true, as opposed to the only other conclusion testing a hypothesis can provide; namely that is is not true.
To McMillan's credit, he very carefully states in the beginning of this essay (or article) that he is only presenting a hypothesis for discussion and for consideration of research. Unfortunately, he then goes off in the body of it attempting to make the case without data to back it that his hypothesis is true; thus implying that steelhead hatchery plants have harmed wild steelhead and should cease.
In other words, this is not a research paper, eventhough it cites some research (not all of the works in his bibliography are scientific hypothesis tested papers or works) that appear to back his hypothesis. The unfortunately result of essays like this that use science language is folks who don't read it carefully or don't know what to look for to see if his conclusion really is based upon testing his hypothesis make the incorrect assumption and conclusion that the hypothesis is correct despite the complete and utter lack of evidence that he tested his hypothesis.
In other words, it is a nice opinion piece/editorial essay that is designed to get people to think differently and to sway opinion. It is not a report of results from him testing his hypothesis. Therefore, I submit it is dangerous to base steelhead management on his unproven and untested hypothesis.
Could one of our scientists (SMalma, Curt, FT) please share with us the data and conclusive reports that have scientifically proven that using Chambers Creek hatchery steelhead has strengthened and improved the genetics and survival rate of our wild steelhead?
Dear FT, not sure if you read the same paper as I did:
It is true, it is a hypothesis to be further tested for conclusive findings, but it does provide numerous examples from which the male vector hypothesis is based that begs for further research (all science begins with such a hypothesis, and he has gone well into providing the needed blocks for proof throughout). And some of his observations have since been well proven in peer reviewed work. His personal sightings are quite conclusive that there is an interactive range for spawning overlap found on the Washougal River on pages 17-18:
Washougal River: Observed range wild females to potentially spawn, January 9 to July 12 (6 months); observed range for wild males to spawn, November 29-July 22 (8 months). Observed range for hatchery females to potentially spawn, December 20-March 19 (3 months); observed range for hatchery male adults, November 29-March 25 (4 months); if hatchery residual male smolt at sexual maturity included found as a kelt on April 16, the range is 4.5 months.
The examples from the steelhead spawning research on the Siuslaw River using tributary creek weirs to document steelhead spawning entry and exit in Oregon provide further data from which to base the hypothesis on pages 20-23:
Siuslaw River: Range for wild female entry to spawn was December 28 into June, 5.5-6 months; range for wild male entry to spawn was December 28th into June. Range for hatchery female entry to spawn was December 19 into May (4.5-5 months); range for hatchery male entry to spawn was December 19 into June (5.5-6 months).
The Siuslaw weir data provide clear indicators for the differing potentials for hatchery and wild fish to spawn together with differences by sex: Hatchery females were only 8% of the total spawning population after March 15th, while 52% of hatchery males spawned after March 15th. Those hatchery males after March 15th had a female population to spawn with that was 92% wild with choice for mates heavily dominated by wild females. 33% of the wild females spawned prior to March 15th with a total male population composed of 65% hatchery males in that time period. Obviously, the hatchery males greatly extended the potential interactive spawning period between wild and hatchery fish. The Oregon researcher further indicated that even though most hatchery females spawned on or prior to March 15th, that there were significant opportunity for spawning overlaps between wild and hatchery fish and it could be very high some years. The data indicated that some years there was relatively equal potential for wild and hatchery fish to spawn together. These specific weir data from Siuslaw tributary spawning are high informative as explanations for wild/hatchery interactions and well detailed in these pages.
Page 23-24 provides the Cowlitz River hatchery data for steelhead: radio tagged fish hatchery males were found to remain in the river and spawn over a period of 2-3 weeks (compared, of course, to females that spawn and are done). In the hatchery males were used for repeated spawnings for up to 2 months. Page 25 provides a figure for the periods of time male and female, hatchery and wild steelhead were found to have overlapping times to spawn together beyond the typical time of separation for wild and hatchery spawning used in Washington for management purposes and often in Oregon of March 15th. Pages 26-36 well depict the overlap of wild and hatchery steelhead spawning times found in varied studies in Washington and Oregon.
The continuing pages through page 53 provide continuous examples of hatchery wild spawning overlaps, or potential for overlaps, as found in science literature or through his personal field observations that include his many spawning surveys of the Washougal River in graphic form. His observations of resident and anadromous spawning interactions are among the few early ones prior to eventual proofs now found in more recent studies on the Olympic Peninsula, Hood River, and other areas of the West Coast.
Could all of the above "possibly" explain the presence of hatchery markers in the DNA of sampled steelhead in the study provided in the other thread started by Andrew Lawrence?
Thanks again for sending me Bill's escapement graphs.
One of the important aspects in analyses like these is the selection of samples. You make a reasonable case for the rivers you compare assessing the hatchery fish variable and ignoring all other variables. As Smalma points out, that conclusion is not supported by including a N. Van. Is. river in the sample. Going the other direction, south, the Nisqually River is the southern most Puget Sound river hosting wild steelhead. This run was healthy through the 1980s and declined in the early 1990s, same as all other PS rivers. An important difference is that the Nisqually has never had a regular hatchery steelhead stocking program, only irregular plants over the years, and certainly none in the last 15 years. Yet it shows that wild steelhead health in PS is positively correlated to the north and negatively to the south. At any rate, if it and the Van. Is. streams were included in the group you presented in Bill's paper, the conclusion that hatchery stocking negatively affects wild steelhead production is no longer supported.
Smalma and I are not trying to be disagreeable. We're just looking as many places as we can at as many variables as we can to try and determine the proximate cause for low PS wild steelhead abundance. While hatchery stocking may be adverse, it's not near the top of the list. The best indicators point to early marine survival, and have for a number of years, yet we still do not know what (or which) early marine factor(s) are responsible, making it impossible to even try to address.
Your question/request was -
" Could one of our scientists (SMalma, Curt, FT) please share with us the data and conclusive reports that have scientifically proven that using Chambers Creek hatchery steelhead has strengthened and improved the genetics and survival rate of our wild steelhead?"
As you probable expect neither I or to my knowledge anyone else can provide such data that Chambers Creel hatchery steelhead has been good for wild steelhead in general and wild Skagit in particular.
Just as none of us can provide any data that shows catch and release fishing is good for the wild fish.
While neither is particular good for Skagit wild steelhead after review all the available data, informationa and my own observations neither is particular bad for wild steelhead.
I can't help but notice that you used the qualifier 'possibly' in your final question (which really is a "begging the question" statement, and not a true question) re: "Could all of the above 'possibly' explain the hatchery markers in the DNA sampled steelhead in the study provided in the other thread started by Andrew Lawrence." The answer of course based upon what you posted and what McMillan put into his essay is that this is unknown because it hasn't been tested to see if it is true or not true.
Also we have to keep in mind that just because some hatchery males are present when some wild fish are spawning doesn't mean the hatchery fish are actually spawning with the wild females. To imply they are, as McMillan does, is rigtfully known as "begging the question" and hoping the reader will jump to the conclusion that this is in fact happening to a large enough degree that it is detrimental to the wild fish population. The bottom line is this: it is unkown at present if or how much spawning interaction is occuring between hatchry males and wild females.
Lest we forget, McMillan did not test his hypothesis at all, he merely wrote an essay designed to promulgate his hypothesis and wrote in such a manner as to sway opinion without any testing whatsoever. However, the lack of testing his hypothesis did not keep him from leading readers to think and conclude that his hypothesis was in fact happening.
This is why I said he provided no evidence only an opinion piece/editorial essay designed to get folks to accept his untested hypothesis as being true, when it was not tested.
I'm familar with McMillian's 2001 paper and read it from cover to cover when it first came out and had a hard copy in my library.
I have to basically agree with FT's assessment with the addiitional comment that some of his obsrevations were dated and likely did not reflect changes in management of hatchery stocks.
That is not say that his observations and resulting concerns were not worth considering. In fact he was hardly the first to make those kinds of observations and have those concerns. The first that I recall were where they were expressed was Lloyd Royal's report in early 1970s. Those concerns help form the foundation for a number of management actions ranging from hatchery/wild interaction studies, manipulation of hatehery brood stocks, stocking protocol, etc.
I think it is important clear up a couple assumptions that you/McMillian have put forward. One it is not true that a female steelhead only spawns once. In fac typically will deposit 500 or so eggs at a time and over a day or two repeat that process as many as 8 to 12 times potentially spawning wth several different males. While it is certainly true that an individual wild male will stay on the spawning grounds for days or even weeks after reaching sexually maturity and first spawning and an individually female while vacate the area reasonably quickly following completion of her spawning. That however does not mean that individual males will remain on the spawning ground much passed the time the last female in the population spawns; after all with whom would those late males spawn with?
On the Skagit one can find adult males through out the summer. I have personally seen such fish as late as October however by mid-August those fish have clearly been post spawners (kelts). While the peak spawning (redd construction) of wild females is typically mid-May some females have been seen actively spawning (digging redds) as late as late July (latest date such activity noted was 7/25). In short the "ripe" males can be found only a week or two past the spawning of the last female in the population. Why would not that be true with the hatchery fish?
Currently in the Skagit basin the latest spawning hatchery females is temporally separated by approiximate a month from the earliest spawning wild females. Even if on the off chance that a ripe hatchery male managed to stay the spawning grounds for that month he would be nearly spend/tired fish that would have to compete (fight) with a fresh/vigorous wild male for spawning rights. While such interactions would be possible one would have to expect it to be rare occurence. While we can debate the significance of such interactions to the longer viability or risk to the population my personal opinion is it is not very significant risk in comparison to the other threats facing the populations.
To your questin to FT -
"Could all of the above "possibly" explain the presence of hatchery markers in the DNA of sampled steelhead in the study provided in the other thread started by Andrew Lawrence?"
Those same "hatchery markers" in sampled wild steelhead populations could easily be the result of the Chambers Creek and say the Skagit fish being from the same geographic area and as part of the same major ancestoral linage (MAL) share some of the same genetic markers. I have asked at least a 1/2 dozen steelhed geneticists whether any of the genetic results would preclud the likelyhood that the two populations would share some genetic background. they all had the same answer -either they did not know or that such sharing would be possible. In fact such "sharing" is a fundamental aspect of stocks within a MAL as well within the ESA concept of a Distinct Popuation Segment (DPS).
Keep in mind that the so-call Chambers Creek hatchery steelhead population was not a PS wild stock but rather a composite stock developed for the hatchery program using as many as 7 or 8 donor stocks. I think most would be surprised if the various PS wild stocks and the Chambers creek stocks did not share some of the same DNA markers.