Results 1 to 18 of 18

Thread: Didson Update

  1. #1

    Default Didson Update

    Update:

    It appears that the Funny River final count will be around 750 Kings. The Killey River weir is on track and will be approximately 1300-1500 kings. With the Funny and Killey Rivers alone, the current Didson count of 2038 has been exceeded. As stated in the USFW Report Abundance and Run Timing of Adult Chinook Salmon in the Killey River 2012, "Information collected during 2010 and 2011 indicates similar results with the majority of the early run returning to the Killey River(48% to 58%), main-stem Kenai River (24% to 37%), and the Funny River (11% to 12%) (Adam Reimer, Alaska Department of Fish and Game, personal communinication)." The current Didson numbers are not accounting for the main-stem Kenai River (24%-37%).

    The Kenai ER Chinook Didson was off by at least 100%

    As of July 10th the Didson sonar estimate is 2,709 and the Cumulative Net
    Apportioned Sonar estimate is 5,703. Why is there such a large discrepencacy
    between the two indices?

    The department used ER data to make a LR forecast and management decisions. Have they changed their forecast to account for this ER error?

    From Emergency Order number 2S-09-13 The department states as of July 11, approx 25% of the late run of Kenai River king salmon based on historical run-timing. All indicies point to the run being late like last year. Why are they using historical average run timing?

  2. #2
    Member fishNphysician's Avatar
    Join Date
    Jun 2006
    Location
    Aberdeen WA
    Posts
    4,516

    Default

    I think you have very legitimate concerns, tb.

    Correlating the sonar estimate to actual weir counts is crucial to validating the accuracy of the technology. For as long as we've had the sonar in the 80's, this is something that's never really been done.

    But there are some assumptions being made in your analysis that may no longer hold water.

    First is the spawning distribution of ER fish (defined in this case as anything passing the sonar prior to July 1).

    1) That may no longer look like the original Bendock distribution because mainstem spawners have been subject to disproportionate exploitation over the past 24 years.
    2) The effects of designating trib-mouth sanctuaries for the entire fishing season may well be allowing a greater proportion of fish to make it into the tribs.

    The combined effect is an increase in the relative trib proportion and a reduction in the mainstem proportion of ER spawners.

    As the chinook team finishes up its current radio-transitter spawning distribution study, we'll have a better handle on how that proportion has changed over time.

    To say the DIDSON is off by a factor of 100% is probably unfair and is overstating things. Lets use the USFWS data you cited. Assuming the mainstem component was still intact and still comprised about 30% of the ER, the combined trib component is about 70% of the ER. Now extrapolate the weir count of 2200 by the appropriate proportion (1/0.7) would give a total run of about 3100 and change. That is NOT a factor of 100%.... it's closer to 50%.

    Now, if the mainstem component is something lower than 30%, the magnitude of undercounting becomes less and less. In the Bendock study, his group only found about 20% of the radio-tagged sample spawned in the mainstem.

    At 20% mainstem, the extrapolated run size is 2750.

    At 10% mainstem, the extrapolated run size is 2444.

    Bottom line, the magnitude of undercounting cannot be reliably determined on weir counts alone. We must have a better handle on the mainstem proportion in order to arrive at the appropriate expansion factor. And even then, that number is subject to year to year fluctuations. My contention is that the mainstem piece has been disproportionately exploited by the in-river fishery and no longer constitutes even 20% of the ER. I think Nerka would echo those concerns.
    "Let every angler who loves to fish think what it would mean to him to find the fish were gone." Zane Grey
    http://www.piscatorialpursuits.com/uploads/UP12710.jpg
    The KeenEye MD

  3. #3

    Default

    So, to put it in the simplest form,

    If I recall, 2011 Funny River weir count was 16% of the ER estimate (going from memory here) on a weir count of around 600? 16% is right in the middle of the historical average of what the Funny River weir comprises. (I realize I quoted the USFS study that showed 11-12% for 2010-2011 - don't know what to think there.)

    If the Funny Weir ends up at 750, that's just over 20% better escapement than last year (similar to many other UCI Chinook runs this year)

    750/.16= 4,687 fish
    750/.12=6,250 fish

    If I've got bad info or have errored somewhere, someone please point it out. We all benefit from accurate counting.

    Thanks for the discussion Doc. Much more than anyone's gotten from the department thus far.

    If the mainstem spawning component of the early run is no longer a component of that run, I would like the department to tell the public that, as well as any theories (I'm believing yours!) as to why that may be. It would be very helpful in explaining how inriver issues affect King abundance every bit as much as saltwater issues, and it may raise some valid concerns that could be addressed before the BOF this year. As the BOF process takes a long time and we are watching both the Kenai and the Kasilof go off the charts on sockeye counts, that info is needed sooner rather than later.

  4. #4
    Member fishNphysician's Avatar
    Join Date
    Jun 2006
    Location
    Aberdeen WA
    Posts
    4,516

    Default

    Again, I think the assumption of 16% Funny or even 12% Funny will lead you to gross hyperinflation of the run size.

    The truth is we don't know what proportion of the current ER is Funny. I believe the historic distribution has changed... count on it. How much? Who knows. And as I said earlier, there will be year -to-year fluctuations within that overall proportionality.

    Moreover, the more you hang your hat on just a small sample of the tribs (ie as in one trib) the greater the likelihood of an expansion error. Much better to run that expansion on the full suite of tribs, not just Funny. You end up with a very different result.

    Look, even the currently observed proportion of 750 Funny vs 1400-1500 Killey does not fit the USFWS data you cite. That data shows Killey dominating 4:1 or perhaps even 5:1 over Funny. So either the USFWS data is suspect, or Funny is having one helluva good year. Make sense?

    I will pull the old Bendock tables to give you the proportion they found back in 1990-91.
    "Let every angler who loves to fish think what it would mean to him to find the fish were gone." Zane Grey
    http://www.piscatorialpursuits.com/uploads/UP12710.jpg
    The KeenEye MD

  5. #5

    Default

    Ok, I screwed up. 2012 Funny River weir was 879. It was 16% of the ER estimate.

    At this point, I don't know what to think. Looks like me and ADFG have something in common!

  6. #6
    Member fishNphysician's Avatar
    Join Date
    Jun 2006
    Location
    Aberdeen WA
    Posts
    4,516

    Default

    Dammitt.... just lost my entire post so had to re-type it. GRRRRRRRR>>>>

    Here's the Bendock distribution in nice round numbers.

    Killey: ~40%
    Funny: ~20%
    Mainstem : ~20%
    Benjamin: ~15%
    All others: ~ 5%

    And just so no one thinks I made this crap up out of thin air....



    Consider that Benjamin is part of the Killey, so the entire Killey system counts for 55%

    That makes the proportion of Funny:Killey come out to 20:55 or about 1:2.75. Now pretend you did not have the Killey weir number for 2013. If you were to expand the 2013 Funny number by the Bendock proportions to arrive at the Killey number, you would get 750 x 2.75 = 2063 Killey kings. The actual observed 2013 value is only ~1400 measured at the weir. Clearly the relative proportion has changed for 2013.

    Now let's expand the Funny number for total escapement... 750/0.2.... and we get an escapement of 3750.

    If we try the same exercise with the observed 2013 Killey weir count... 1400/0.55.... we get 2545.

    This illustrates the perils of expanding a limited data set.... you get very different answers. Yes, both answers clearly point out that the DIDSON undercounted ER this year, but if you look at Funny alone, DIDSON looks horribly flawed. If you look at Killey alone, well it's not so bad.

    So let's try it knowing what we know in 2013 for combined Funny+Killey weir counts.... again this should much more reliable than either count alone. For round numbers lets say the combined weir count is 2200 kings.... expand 2200/0.75... and we get 2933.

    So which of these is correct?

    I'd hedge my bets on the 2933. But again this is making the assumption that F+K = 75% and mainstem = 20%. That's simply not something we can hang our hats on.

    What if mainstem is now 15% and F+K = 80%. The expansion comes out to an escapement of... 2200/0.8.... 2750.

    What if mainstem is now only 10% and F+K = 85%. The expansion comes out to.... 2200/0.85.... 2588.

    ....

    In each of these scenarios, we FAIL to meet not just the OEG 5300-9000, but also the BEG 4000-9000
    "Let every angler who loves to fish think what it would mean to him to find the fish were gone." Zane Grey
    http://www.piscatorialpursuits.com/uploads/UP12710.jpg
    The KeenEye MD

  7. #7

    Default

    Thanks Doc, although I remember reading in one study that they think the the Funny river proportion seems to be dropping. As I said, last year it was 16%. USFS said 12%. Is that Bendock study from the 90's?

    One thing is for sure about all of the scenarios you laid out - ER is a mess and the methods to count it are a mess. The ER estimate is probably not the best predictor of LR strength at this point.

    We need to be killing Sockeye right now.

  8. #8
    Member fishNphysician's Avatar
    Join Date
    Jun 2006
    Location
    Aberdeen WA
    Posts
    4,516

    Default

    Quote Originally Posted by smithtb View Post

    We need to be killing Sockeye right now.
    You may well get your wish in a big way tomorrow morning.

    Have you been to the beach this afternoon to see if that SW wind is starting to blow any rollers ashore?
    "Let every angler who loves to fish think what it would mean to him to find the fish were gone." Zane Grey
    http://www.piscatorialpursuits.com/uploads/UP12710.jpg
    The KeenEye MD

  9. #9
    Member
    Join Date
    Apr 2010
    Location
    Anchorage
    Posts
    499

    Default

    FNP and Smith help me out here. If I recall correctly from the 2012 early run reconstruction report. About half of the killey river tagged fish spawned below the killey river weir. OF course these percentages can change year to year as your figure illustrates FNP. But if last years percentage is close to this year then that means more like 3,000 fish could actually go up the killey, if the weir counts only about half of the fish that go up there. Well have to see the tagged fish data here in another couple weeks to know what the percentage for above/below the weir was this year. At least that's the way I read the early-run king report data on the Killey.

  10. #10
    Member
    Join Date
    Apr 2010
    Location
    Anchorage
    Posts
    499

    Default

    Don't forget that we've pretty much lost the Slikok contribution to the early run. With only 27 kings entering last year and probably not much more than that this year. I have friends that live along Slikok that when they were young used to see kings up there all the time. Now though? Hardly anything at all.

    So if we lost the Slikok portion, which should easily produce a couple hundred kings a year, and we've lost some of our early mainstemers, then it doesn't take a NASA scientist to figure out that the early run fish count as a whole is going to take a hit.

  11. #11
    Member fishNphysician's Avatar
    Join Date
    Jun 2006
    Location
    Aberdeen WA
    Posts
    4,516

    Default

    Quote Originally Posted by 33outdoorsman View Post
    FNP and Smith help me out here. If I recall correctly from the 2012 early run reconstruction report. About half of the killey river tagged fish spawned below the killey river weir. OF course these percentages can change year to year as your figure illustrates FNP. But if last years percentage is close to this year then that means more like 3,000 fish could actually go up the killey, if the weir counts only about half of the fish that go up there. Well have to see the tagged fish data here in another couple weeks to know what the percentage for above/below the weir was this year. At least that's the way I read the early-run king report data on the Killey.
    Not sure how far upriver the Killey weir is located, but depending on flows, sometimes an obstruction, even a partial one like a weir, can change spawning behavior causing some fish to do their deed below it rather than above, had it not been there in the first place.
    "Let every angler who loves to fish think what it would mean to him to find the fish were gone." Zane Grey
    http://www.piscatorialpursuits.com/uploads/UP12710.jpg
    The KeenEye MD

  12. #12
    Member
    Join Date
    Jul 2006
    Posts
    5,526

    Default

    Doc, take a look at the early run escapement report to get the in-river estimates used to establish the goal. Then take a peek at the 2012 USFWS report and you will see that the Funny is about 12%. The numbers are slightly off for some reason (probably post season adjustments) but you get the idea. If the contribution of Funny River is going down then that means the estimate must be higher and the counter is farther off not closer to the weir counts - your examples do not match with the data as you are missing some tributaries that contribute to the total return and the assumption about main stem not being there is not valid - more on that below. Here is the section from the USFWS report. Below it I hope to post the tagging data.

    The contribution of the Funny River to the total Kenai River early-run Chinook salmon
    escapement appears to be decreasing over time. An ad-hoc estimate of the total proportional
    contribution of early-run Chinook salmon destined for the Funny River can be projected for
    every year since 2006 by dividing the passage estimates at the Funny River weir (Gates and
    Palmer 2007, 2008; Gates and Boersma 2009a, 2009b, 2011) by the remainder of the passage
    estimates obtained with sonar in the lower Kenai River (Miller et al. 2011; Jeff Perschbacher,
    Alaska Department of Fish and Game, personal communication; Appendix 5) and the harvest
    estimates for early-run Chinook salmon (Jennings et al. 2010a, 2010b; McKinley and Fleischman
    2010; Jennings et al. 2011a, 2011b; Jeff Perschbacher, Alaska Department of Fish and Game,
    personal communication; Appendix 5). These estimates indicate a downward trend in the
    contribution of Chinook salmon from the Funny River to the overall Kenai River early-run
    escapement (0.20 of overall run in 2006, 0.25 in 2007, 0.16 in 2008, 0.14 in 2009, 0.10 in 2010,
    and 0.12 in 2011). However, the scale of the contribution and the slope of change could be
    markedly different than estimates using this method. Potential sources of bias could include
    imprecise estimation of the Kenai River early-run Chinook salmon escapement and imprecise
    estimation of the Kenai River early-run Chinook salmon harvest.

    The tagging data for 2010, 2011, and 2012 indicates for the early run that 19.6, 25.6, and 11.6 of the early run tagged at the lower tagging site was mainstem spawners. The portion that was Funny River was 12.5, 10.8, and 11.6 percent. So your recalculation assuming no main stem spawners is off. However, having said that who knows what is early run or late run. It may be the main stem spawners are just one run that starts at the end of June and goes through August. That does not mean they are not being harvested at a differential rate but it would make more biological sense that a July 1 cut off date.

    Maybe we need to rethink this whole thing in terms of a tributary goal and a main stem goal with equal harvest rates over the whole return. In that case the early portion of the main stem spawners would have to be protected from over-harvest in July above the bridge.

    The tagging data in 2012 shows the following. Funny 11.6 percent, Killey 69.7, Other systems 7, and main stem 11.6 if I am reading the spread sheet correctly. It looks like B. Creek is separate from the Killey R main stem so I added them together. Not sure if the weir is above or below these tags so need to ask. However, bottom line is that the Funny has been on the low end not the high end.

    Only time will tell when the Killey comes in but no matter there are serious questions that need to be answered for everyone to feel comfortable with the estimates of escapement.

  13. #13
    Member fishNphysician's Avatar
    Join Date
    Jun 2006
    Location
    Aberdeen WA
    Posts
    4,516

    Default

    Quote Originally Posted by Nerka View Post

    ... but no matter there are serious questions that need to be answered for everyone to feel comfortable with the estimates of escapement.
    All the mathematic masturbation can be overwhelming, but I stand by the rationale for all the expansions in post #6. But I'll concede using the 1400 weir-count as the Killey escapement number gives the wrong answer. Based on 33o's post, that is clearly is NOT the whole Killey escapement because 1400 fails to capture spawners below the weir.

    Re: not capturing all the tribs, that was never my intention. The whole point of the exercise was to show that the answer for estimated escapement changes as we include more tribs, as opposed to just extrapolating strictly from the Funny proportion alone.

    None of my examples presumed that mainstem was absent... I simply ran scenarios for 30%, 20%, and 10% mainstem, reapportioning the change in mainstem back to the F+K component. Who knows what the real proportion of mainstem actually is? All I know is that it changes the mathematical answer for estimated ER escapement significantly.
    "Let every angler who loves to fish think what it would mean to him to find the fish were gone." Zane Grey
    http://www.piscatorialpursuits.com/uploads/UP12710.jpg
    The KeenEye MD

  14. #14
    Member fishNphysician's Avatar
    Join Date
    Jun 2006
    Location
    Aberdeen WA
    Posts
    4,516

    Default

    Quote Originally Posted by Nerka View Post
    The tagging data in 2012 shows the following. Funny 11.6 percent, Killey 69.7, Other systems 7, and main stem 11.6 if I am reading the spread sheet correctly. It looks like B. Creek is separate from the Killey R main stem so I added them together. Not sure if the weir is above or below these tags so need to ask. However, bottom line is that the Funny has been on the low end not the high end.
    So compared to the 1990 data which was Funny 20, Killey 55, mainstem 20, and others 5, what we're seeing in 2012 is that "other" is pretty close to the same, mainstem is down by 40%, and combined K+F proportion picks up the bulk of the difference.

    So now, instead of 75% F+K, we're at 81%. Mainstem was 20 and now down to 12. The difference between 5 and 7 for other is probably NOT significant due to the small sample size.

    We'll never know what the real mainstem number is... can only calculate it indirectly.

    The easiest numbers to wrangle with are the actual kings observed from Funny and Killey. Once the Killey count is finalized, we'd have a pretty good feeling for escapement by taking the sum of F+K (actual observed) and dividing by 0.81 to get a pretty reasonable escapement estimate.

    If the DIDSON is anywhere near accurate, the number should match up pretty closely, esp since ER harvest was next to nothing this year.

    I don't think anyone should hold their breath for a match.

    Just wondering if the accuracy of the new chinook DIDSON site further upriver fared any better
    "Let every angler who loves to fish think what it would mean to him to find the fish were gone." Zane Grey
    http://www.piscatorialpursuits.com/uploads/UP12710.jpg
    The KeenEye MD

  15. #15
    New member
    Join Date
    Feb 2004
    Location
    Soldotna
    Posts
    5,639

    Cool Bingo . . . !

    Quote Originally Posted by fishNphysician View Post
    All the mathematic masturbation can be overwhelming . . .

    You got that right . . . . . ad nauseam . . . . +1

  16. #16

    Default

    Quote Originally Posted by fishNphysician View Post
    So now, instead of 75% F+K, we're at 81%. Mainstem was 20 and now down to 12. The difference between 5 and 7 for other is probably NOT significant due to the small sample size.
    Wait... I thought it was Funny that was 20 and now down to 12. Crap. I'm confused. Oh man, lets just all go Sockeye fishing!

    As to your earlier question about fish on the beach - yes, good wind. Very fishy out there...

  17. #17
    Member fishNphysician's Avatar
    Join Date
    Jun 2006
    Location
    Aberdeen WA
    Posts
    4,516

    Default

    NOAA says wind has shifted S... dang, might throw things off a smidge... only an hour or two til the ebb. We'll see if they come ashore.
    "Let every angler who loves to fish think what it would mean to him to find the fish were gone." Zane Grey
    http://www.piscatorialpursuits.com/uploads/UP12710.jpg
    The KeenEye MD

  18. #18

    Default

    Supposed to be SW20 on Tuesday. Could be an optimal day to surgically target Sockeye... too bad Tuesday is a mandatory closure day for gillnets. Mother nature doesn't seem to care about the calendar. This optimist is thinking that Monday will show good numbers for both Sockeye and Kings. Reports are that Sockeye are thick near the Kenai.

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •