***************************************************************************
I-1a. COHO RETURNS AS % OF RELEASE AT NEHALEM. If the number of smolts released is increased, then it would be expected that the number of jacks or adults returning would also increase. The number of smolts released was quite variable and averaged about 125,000 fewer smolts during the nonhazing years of 1975-1984 than during hazing years (Appendix III). Accordingly, differences in returns between hazing and nonhazing years could reflect differences in the numbers of smolts released and not just the occurrence of hazing, per se.
One way to analyze returns so that the number of smolts released is not a significant factor is to calculate the percent return by dividing the number of returns by the number of smolts released. However, fingerling coho were released into the Nehalem basin through 1990 and in 1992 (Table 3); they may have migrated out as smolts the following year, but their return rate is unknown and would be expected to be less than for fish released as smolts. Thus, it is only possible to accurately determine the percent return in 1992 and 1994-1997, when returns would only be from smolts (Table 3).
The percent return of jacks or adults does not seem very high at the Nehalem hatchery in some of the years with hazing, and the adult return rate in three of four hazing years was actually less than from smolts released in the nonhazing year of 1995 (Fig. 12).
When return rates at the Nehalem are compared to those at the Salmon River (where hazing did not occur and fingerlings were not released in these years), Nehalem return rates averaged 0.04% more for jacks and 0.12% less for adults (Fig. 13) and were not statistically different for jacks (paired-sample t=0.63, df=4, one-tailed P>0.25) or adults (paired- sample t=-0.91, df=3, one-tailed P>0.10).
Differences for the percentage return between the Nehalem and Siletz or Alsea hatcheries were not compared because fingerlings were sometimes released into the Siletz and Alsea and their smolt releases were not always at the hatchery, so their returns may have been affected by the release site.
---------------------------------------------------------------------------FIGURE 12. Percentage of coho smolts returning as jacks or adults to the North Fork of the Nehalem River hatchery. Releases prior to 1992 and in 1993 are not included because fingerling releases into the Nehalem or smolt releases elsewhere in the Nehalem than at the hatchery in these years may have augmented returns to the hatchery (see Table 3). These data were calculated from data provided by Rick Klumph, John Leppink, and Tracy Cabe of the ODFW. Jacks returned in the same year as the Year of Smolt Release, and adults returned the following year.
H=affected by hazing O=not affected by hazing
% of Coho Smolts Returning to Nehalem Hatchery as
P Jacks Adults
e 1.0-| 1.0-|
r -| -|
c 0.8-| 0.8-|
e -| -|
n 0.6-| 0.6-| H O
t -|------ ------------------ 0.5% -------|---H-- ---H---O--------
a 0.4-| H 0.4-| H H O H H
g -| H H -| H H O H H
e 0.2-| H H 0.2-| H H O H H
-| H H H -| H H O H H
0.01-0.04-| H H O H H H 0.01-0.04-| H H O H H
0-| H H O H H H 0-| H H O H H
| | | | | | | | | | |
92 94 95 96 97 98 92 94 95 96 97
Year of Smolt Release
---------------------------------------------------------------------------
FIGURE 13. Differences in the percentage of coho smolts returning as jacks or adults between the North Fork of the Nehalem River and Salmon River hatcheries. The number of smolts released at the Salmon River were provided by John Leppink and Tracy Cabe of the ODFW; all other data are calculated from data provided by Rick Klumph of the ODFW. Jacks returned in the same year as the Year of Smolt Release, and adults returned the following year.
Releases in 1985-1991 and 1993 are not included because fingerling releases or smolt releases elsewhere into the Nehalem may have augmented returns to the Nehalem Hatchery (see Table 3). An undetermined portion of these fish would be expected to return to the Nehalem Hatchery; however, their return rate would be expected to be much less than for smolts released at the hatchery.
H=affected by hazing O=not affected by hazing
Nehalem-Salmon River % of Coho Smolts Returning as
Jacks Adults
0.5-| 0.5-|
-| Hazing Mean= 0.04% -| Hazing Mean= -0.12%
0.4-| 0.4-|
-| -|
0.3-| H 0.3-|
-| H -|
P 0.2-| H 0.2-|
e -| H -|
r 0.1-| H 0.1-| H H
c -| H H -| H H
e 0-|==H== ==H===O===H===H===H 0-|===H== ==H===O===H===H
n -| H O H -| H O H
t -0.1-| -0.1-| H O H
a -| -| H O H
g -0.2-| -0.2-| H O H
e -| -| H O H
-0.3-| -0.3-| H H
-| -| H
-0.4-| -0.4-| H
-| -|
|_____ ___________________ |______ _______________
| | | | | | | | | | |
92 94 95 96 97 98 92 94 95 96 97
Year of Smolt Release
---------------------------------------------------------------------------
I-1b. NUMBER OF RETURNING COHO. The numbers of jacks returning to the Nehalem and Trask hatcheries were variable; the average was slightly higher during hazing at the Nehalem but during nonhazing at the Trask (Fig. 14). Returns with hazing are not significantly greater (Table 13).
The number of returning adults was also variable and averaged twice as great with hazing at the Nehalem but only slightly greater at the Trask hatchery (Fig. 14). Hazing-affected returns were not significantly greater for the Trask hatchery but were so for the Nehalem hatchery (Table 13). However, there was an extraordinarily high adult return for smolts released in 1990 at the Nehalem that would have returned as adults in 1991 (Fig. 14); this anomaly may not be related to hazing (section K-2b). Without the 1991 anomaly, hazing results for adults at the Nehalem were not significantly greater (Table 13: footnote b). For both hatcheries, returns during some years with hazing were less than in some years without it (Fig. 14).
If hazing is as beneficial as has been claimed for coho hatchery returns (Erickson 1995a, Monroe 1995b, 1996a; Nokes 1995), one would expect differences with hazing to be greater than they are. But even if hazing improved hatchery returns of adults, this is not a goal of salmon management because increased returns are not very useful. As it is, there are often more coho returning than are needed to coastal hatcheries (Orcutt 1992) and at the North Fork of the Nehalem hatchery, generally about 400-600 female coho were spawned each year, regardless of the number returning (John Leppink and Tracy Cabe, ODFW, pers. comm.). The number of females that were not spawned (i.e., surplus) at the Nehalem does not appear to be related to hazing; for example, there were 1,743 surplus females in the nonhazing-affected return year of 1985 but only 79 surplus females in the hazing-affected return year of 1990 (John Leppink and Tracy Cabe, ODFW, pers. comm.). Surplus fish are generally unfit for human consumption and are sold for fish food (Orcutt 1992).
---------------------------------------------------------------------------TABLE 13. Statistical tests of whether the number of returning coho jacks or adults to the Nehalem or Trask hatcheries is significantly greater with hazing. Hazing-affected years of smolt releases were 1988-1994 and 1996-1998 for the Nehalem hatchery and 1988 and 1996-1998 for the Trask hatchery in the Tillamook Basin. See Fig. 14 for details and yearly graphs. The larger mean is underlined. U=Mann-Whitney statistic for the hypothesis that the number of returns at the Nehalem or Trask hatcheries is more abundant with hazing than without it, N=number of years, P=probability, NS=not significant (one-tailed P>0.05).
---------------------------------------------------------------------------
Number of Returning Coho to Hatchery...... Hazing>
Hazed............... Nonhazed............ Nonhazing
Comparison N Mean Range N Mean Range U P
---------------------------------------------------------------------------
Coho Jacks
Nehalem Hatchery 10 1,564 80-4,570 11 1,433 45-7,499 62 NS
Trask Hatchery 4 405 216-591 20 608 48-2,803 34 NSa
Coho Adults
Nehalem Hatchery 9 5,185 1,553-14,732 11 2,587 71-7,988 74 <0.05b
Trask Hatchery 3 4,138 789-10,174 20 3,833 930-9,326 26 NSa
a The number of years with known hazing is less than five; statistical
tests would be more robust with larger sample sizes (section E-4e).
b Without the anomalous 1991 return of adults from smolts released in
1990 (section K-2b), returns with hazing are not significantly
greater (U=63, N1=8, N2=11, one-tailed P>0.05).
---------------------------------------------------------------------------
---------------------------------------------------------------------------
FIGURE 14. Counts of coho jacks or adults returning to the North Fork of the Nehalem River or Trask hatcheries for each Year of Smolt Release during 1975-1998. These data were provided by Rick Klumph (ODFW, pers. comm.), except for the 1988 number of jacks returning to the Nehalem hatchery, for which I use the 4,570 jacks given by Leslie Schaeffer (ODFW, pers. comm.) and John Leppink (ODFW, pers. comm.) rather than the 424 jacks given by Klumph. Jacks return in the same year as the Year of Smolt Release, and adults return during the following year.
H=affected by hazing ?=unknown if affected by hazing
o=not affected by hazing
------------------ Number of Returning Coho Jacks --------------
N 8,000-| Nehalem Hatchery -| Trask Hatchery
u | o Nonhazing Mean= 1,433 | Nonhazing Mean= 608
m 7,000-| o Hazing Mean= 1,564 -| Hazing Mean= 405
b | o |
e 6,000-| o -|
r | o ? |
5,000-| o ? -|
o | o ?H H |
f 4,000-| o ?H H -|
| o ?H H |
J 3,000-| o o ?H H -| o
a | o o ?H H H | o
c 2,000-| oo o ?HHH H -| o
k | oo o? ?HHH H | o o
s 1,000-| oo o o o? ?HHHHH -| oo o o oo
| oo oo oo o???HHHHH oH H | ooo ooo oHoooo o H
1-249-| oooooooooo???HHHHHHHoHHH -| oooooooooooooHoooooooHHH
0-| oooooooooo???HHHHHHHoHHH -| oooooooooooooHoooooooHHH
|''''|''''|''''|''''|''' |''''|''''|''''|''''|'''
75 80 85 90 95 75 80 85 90 95
Year of Smolt Release
------------------- Number of Returning Coho Adults -------------------
Nehalem Trask
Nonhazing Mean= 2,587 Nonhazing Mean= 3,833
Hazing Mean= 5,185 Hazing Mean= 4,138
14,732-H
|
10,000-| H -| H
| H -| o H
9,000-| H -| o H
| H -| o H
N 8,000-| o ? H -| o H
u -| o ? H -| o H
m 7,000-| o ? H -| o o H o
b -| o ?H H -| o o o H o
e 6,000-| o ?H H -| o o o oHo o
r -| o ?H H H -| o o o oHoo o
5,000-| o o ?H H HH -| o o o oHoo oo
o -| o o ?H H HHHo -| o ooo oHoo oo
f 4,000-| o o o ?H H HHHo -| o ooo oHoo oo
-| o o oo ?H HHHHHo -| o o ooo oHoo oo
A 3,000-| o o o? ?H HHHHHo -| o o ooo oHoooooo
d -| o oo o? ?H HHHHHoHH -| o oo ooo oHoooooo
u 2,000-| o oo o? ?H HHHHHoHH -| o oo ooo oHoooooo
l -| o oo oo? ?HHHHHHHoHH -| ooooo oooooHoooooo H
t 1,000-| oo oo o oo???HHHHHHHoHH -| oooooooooooooHoooooooHH
s -| oo oo oooo???HHHHHHHoHH -| oooooooooooooHoooooooHH
0-| oooooooooo???HHHHHHHoHH -| oooooooooooooHoooooooHH
|''''|''''|''''|''''|'' |''''|''''|''''|''''|''
75 80 85 90 95 75 80 85 90 95
Year of Smolt Release
---------------------------------------------------------------------------
---------------------------------------------------------------------------
I-2a. INTRODUCTION. Hatchery winter steelhead are released into the Nehalem, Tillamook, and Nestucca Basins, and summer steelhead are released into the Tillamook and Nestucca Basins (Appendix II-2a). Unfortunately, there are difficulties in determining if hazing has influenced hatchery returns. For example, the number of winter steelhead smolts released has varied at the Nehalem (Appendix III), and it would be expected that more fish would return as a consequence of larger releases. To reduce this factor, it would be ideal to calculate the percentage of smolts that were released that return to the hatchery; however, this is not appropriate for Nehalem hatchery returns for two reasons. First, most adults return as 2- or 3-salts and the proportion returning as each may vary yearly (Appendix II-2d). Second, many winter steelhead smolts reared at the North Fork of the Nehalem hatchery were released elsewhere in the Nehalem or nearby Necanicum basins (Appendix III), and a significant proportion of those would be expected to stray back to the Nehalem hatchery, since 16-24% of steelhead caught in 1993 at the Trask and Nestucca Rivers were strays from releases into other streams (Lindsey et al. 1994:9).
Another challenge in analyzing winter steelhead hatchery returns to the Nehalem hatchery is that the number of steelhead returning depends upon how long they are counted. For example, many can return in February or March, but counting ceased in January or early February in some years (Table 4) when enough fish had returned for hatchery purposes, not because fish had ceased returning (Gary Yeager, ODFW Nehalem Hatchery manager, pers. comm.).
Because of these challenges, returns must be examined cautiously to determine if hazing is correlated with increased returns. Below, returns to the Nehalem and Cedar Creek (Nestucca Basin) hatcheries are examined. There were too few returns (i.e., usually less than 100 adults and 5 jacks) to the Trask and Trask Pond (Tillamook Basin) hatcheries (John Leppink, ODFW, pers. comm.) to be useful in analyses.
I-2b. WINTER STEELHEAD. The average number of jacks returning to the Nehalem hatchery is much greater with hazing (Fig. 15), and returns are significantly greater with hazing (Mann-Whitney U=76, N1=9, N2=11, one-tailed P<0.05). However, Nehalem returns are not always greater with than without hazing, and the average number of winter steelhead jacks returning to the Cedar Creek hatchery for two years of hazing is less than the average for 13 nonhazing years (Fig. 15).
Annual adult returns to Nehalem with hazing are not consistently higher (Fig. 15). On average, slightly more adults returned, but annual returns with hazing are not significantly greater (Mann-Whitney U=41, N1=6, N2=10, one-tailed P>0.10). Years in which only 2- or 3-salts were possibly affected by hazing are not included in analyses, and returns in these years are greater than in some years when all ages were affected (Fig. 15).
Because hazing commenced in the spring of 1996 at the Nestucca, the first return year with both 2- and 3-salts would be 1998-1999. Consequently, there are not yet enough years of adult returns at the Cedar Creek Hatchery to see if hazing is correlated with increased returns.
I-2c. SUMMER STEELHEAD. Summer steelhead were not released at the Nehalem (Appendix II-2a). It is inappropriate to examine summer steelhead returns to the Trask hatchery (Tillamook Basin) because too few (i.e., less than 30 adults and 0-1 jacks) return there annually (John Leppink, ODFW, pers. comm.).
Only the returns of summer steelhead of Stock 47 to the Cedar Creek hatchery are numerous enough to analyze. However, since hazing commenced at the Nestucca in the spring of 1996 and adults may return as 2- or 3- salts as do winter steelhead (Appendix II-2d), there are no data that may entirely represent adults affected by hazing. Even for jacks (1-salts), the only two years of hazing-affected returns that are available are for 1997 and 1998; 0-1 jacks returned in these years, compared to an average of 1.1 jacks (range 0-5) during the nonhazing-affected 1984-1992 and 1994-1996 (John Leppink, ODFW, pers. comm.). So hazing is not correlated with an increase in jack returns.
I-2d. STEELHEAD DISCUSSION. The assertion that steelhead hatchery returns have increased with hazing (Erickson 1989a,b,d; Monroe 1996a) is supported only for Nehalem winter steelhead jacks, not for Cedar Creek winter or summer steelhead jacks or Nehalem winter steelhead adults. Similarly, Berry (1995) testified to the Oregon Legislature that steelhead returns to the Nehalem hatchery for 1993 releases were nearly identical to those not affected by hazing that returned to the Tillamook Basin hatchery. Further, since it is adults, not jacks, that contribute materially to fisheries and hatchery brood stock, the increased jack returns to the Nehalem with hazing are not very beneficial.
---------------------------------------------------------------------------
FIGURE 15. Number of winter steelhead jacks or adults returning to the North Fork of the Nehalem River and Cedar Creek hatcheries. The Run-Year is when the steelhead returned to the hatchery. For smolts released during the spring of 1988, jacks (1-salts) would return during 1988-1989 Run-Year, and most adults would return during the 1989-1990 Run-Year as 2-salts, some would return during the 1990-1991 Run-Year as 3-salts, and a few would return in later years as repeat spawners (Appendix II-2d).
Nehalem data for the 1997-1998 Run-Year were provided by John Leppink and Tracy Cabe of the ODFW; data for previous years were from Rick Klumph (ODFW). The adult return in the 1975-1976 Run-Year is excluded because only 2-salts returned then to the Nehalem hatchery; the first Run-Year with both 2- and 3-salts was 1976-1977.
Cedar Creek data were provided by John Leppink; data prior to the 1983-1984 Run Year were not necessary because there were enough years without hazing. Only jacks are graphed because the first year with hazing-affected returns of both 2- and 3-salts would be 1998-1999.
H=affected by hazing ?=unknown if affected by hazing
o=not affected by hazing #=wild steelhead released
2=only 2-salts affected by hazing 3=only 3-salts affected by hazing
-------------------- Returning Winter Steelhead Jacks --------------------
Nehalem Hatchery Cedar Creek Hatchery
Nonhazing Mean= 5 Nonhazing Mean= 7
Hazing Mean= 19 Hazing Mean= 4
80-| 80-|
70-| H 70-|
60-| H 60-|
J 50-| H H 50-|
a 40-| H H 40-| o
c 30-| H oH 30-| o
k 20-| ? H HH HoH 20-| o
s 10-|--------oo?--H-HH--HoH- 10 10-|-o-----ooo--o-- 10
1-4-| oo? H HHHHHoH 1-4-| oooo ooo ooHH
0-|oooooooooo???HHHHHHHoHH 0-|oooooooooooooHH
|________________####### |________#######
|''''|''''|''''|''''|'' ''|''''|''''|''
75- 80- 85- 90- 95- 85- 90- 95-
76 81 86 91 96 86 91 96
Run-Year
Winter Steelhead Adults Returning to Nehalem
-| Nonhazing 1976-77 through 1985-86 Mean= 631
2,500-| Hazing 1990-91 through 1995-96 Mean= 737
-| o
2,000-| o
-| o o
A 1,500-| o o
d -| o o H 3
u 1,000-|-----------------o-o-o-----------------H-H-3--- 1,000
l -| o o o ? 2 H H H 3 2
t 500-| o o o o ? ? ? 2 H H H H H 3 2
s -| o o o o o o o o ? ? ? 2 H H H H H H 3 2
1-124-| o o o o o o o o o o ? ? ? 2 H H H H H H 3 2
0-| o o o o o o o o o o ? ? ? 2 H H H H H H 3 2
|_________________________________#_#_#_#_#_#_#
| ' ' ' ' | ' ' ' ' | ' ' ' ' | ' ' ' ' | ' '
75- 80- 85- 90- 95-
76 81 86 91 96
Run-Year
---------------------------------------------------------------------------
Hatchery fall and spring chinook are released into the Tillamook and Nestucca Basins, but no chinook are released into the Nehalem Basin (Appendix II-3a).
The number of fall chinook jacks returning to the Trask hatchery did not show an increase during the two years with hazing compared to nonhazing years (Fig. 16). The fall chinook hatchery program at the Cedar Creek hatchery appears to be ceasing as eggs were last taken in 1992 and jack returns have greatly declined in following years (John Leppink, ODFW, pers. comm.). Accordingly, it would be inappropriate to test Cedar Creek jack returns for the effects of hazing.
Spring chinook jack returns to the Trask and Cedar Creek hatcheries for the two years affected by hazing were less than or about equal to the years without hazing; there clearly was not an increase with hazing (Fig. 16).
For 1986-1997, the only year with any returning winter chinook jacks to the Trask hatchery was 1993 when there were two (John Leppink, ODFW, pers. comm.). So no more returned in the hazing affected years of 1989 and 1997 than in other years, but winter chinook jacks are rare in any year.
In summary, the number of returning chinook jacks did not appear to be correlated with hazing.
---------------------------------------------------------------------------FIGURE 16. Number of fall and spring chinook jacks returning to the Trask hatchery (Tillamook Basin) and spring chinook jacks returning to the Cedar Creek Hatchery (Nestucca). Data were provided by John Leppink (ODFW, pers. comm.); I did not request 1998 Trask data, so they are not shown.
H=affected by hazing o=not affected by hazing
------------------ Number of Returning Chinook Jacks ----------------------
Trask Hatchery Trask Hatchery Cedar Creek Hatchery
Fall Chinook Jacks Spring Chinook Jacks Spring Chinook Jacks
N 100-| o 100-| 100-|
u 90-| o 90-| 90-| o
m 80-| o 80-| 80-| o
b 70-| o 70-| 70-| o o
e 60-| o 60-| 60-| o oo
r 50-| o o o 50-| 50-| o oo
40-|oo o o 40-| 40-| oo oo
o 30-|oooo o 30-| 30-| ooooo
f 20-|oooo Ho 20-| 20-| ooooo
10-|oooooHo o o 10-| oo o oo 10-| oooooooo o HH
J 1-4-|oooooHo o oo 1-4-|ooooo o o oo H 1-4-| ooooooooooooHH
a 0-|oooooooooooooH 0-|oooooHoooooooH 0-|oooooooooooooHH
c '|''''|''''|'' '|''''|''''|'' '|''''|''''|'''
k 85 90 95 85 90 95 85 90 95
s
Year of Return
***************************************************************************
J-1a. INTRODUCTION. The coho fishery is important, although recovery of wild coastal coho takes precedence (OCSRI 1997a,b; NMFS 1998, Kitzhaber 1999). The challenge in using estuarine and stream sports catches to determine if hazing has been successful is that at least four other factors also affect catches. First, estuarine and freshwater catches are expected to be greater with hazing because ocean exploitation rates were less during hazing years than without it (Appendix V-8), so relatively more coho should escape to freshwater, where they could be caught.
Second, the number of coho caught can depend upon the number of smolts released, which has averaged greater since hazing began at the Nehalem (Appendix III) and has been variable at the Salmon and Alsea Basins (ODFW 1997a:61, 1997b:45). Standardizing catches by the number of hatchery smolts released at the Nehalem is not possible because fingerlings were released into the Nehalem Basin (e.g., Table 3), and their survival is probably much lower than for smolts, although they could still contribute to catches. Additionally, much of the available catch data prior to 1994 includes wild coho, so using the number of hatchery smolts released to standardize catches would not be appropriate.
A third factor that can influence sport catches is straying, whereby coho caught in one basin may have been released elsewhere (Appendix V-9). This factor is important along the Oregon central coast because Oregon Aqua-Foods (OAF)(a now defunct private aquaculture firm at Yaquina Estuary) released an annual average of 10.9 million coho smolts (range 3.9-20.6 million) during the nonhazing years of 1978-1984 that would have been caught during the 1979-1985 fisheries and an average of 2.9 million smolts (range 0-8.0 million) during the years of known hazing at the Nehalem in 1988-1992 that would have been available to fisheries in 1989- 1993 (ODFW 1991:37; Bob Buckman, ODFW, pers. comm.); OAF ceased releases in 1993. A high percentage of OAF smolts did not return to the Yaquina but strayed elsewhere, particularly to streams between the Yachats and Salmon River Basins (Jacobs 1988:17-18, ODFW 1997a:62, 1997b:43, 1997c:57). In 1985, 78% of the adult coho captured at the Salmon River hatchery and 54% of the coho at Salmon River spawning areas were released by OAF (Jacobs 1988:17-18). Consequently, if catches are relatively less at the Alsea, Siletz, and Salmon River Basins during years with hazing at the Nehalem, the reason may be the lack of OAF strays to these basins in recent years rather than increased returns to the Nehalem from hazing.
Another factor that affects stream sports catches is fishing effort. For example, coho fishing effort at the Salmon River has ranged from 7,470 to 78,663 Angler Hours annually during 1976-1996 (ODFW 1997b:46) and could have also varied considerably at other coastal streams. Thus, a larger catch in one year than another may reflect greater fishing effort that year, not necessarily a greater number of available fish.
To sum up, the estuarine and stream catch reflects the effects of many factors (also see Appendix V), so it may not be possible to distinguish how much the catch depends upon any one factor, such as hazing. Nonetheless, fisheries are an important aspect of salmon management, so it is important to examine catch data cautiously to determine if hazing may have improved fisheries.
J-1b. INCREASED COHO POST-HAZING CATCHES. Catches are significantly greater at the Nehalem since hazing began (Table 14, Fig. 17). The huge catch in 1991 at Nehalem (Fig. 17) is probably not a result of hazing because 1991 was an anomalous year for adult returns (section K-2b), and an unusually large catch was also reported then at the nonhazing-affected Tillamook Basin and composite of other coastal streams (Fig. 17). At Tillamook Bay, this extraordinary catch in 1991 was attributed to large numbers of non-Tillamook Bay coho being caught just inside the mouth of the Bay during late summer (R. Klumph in Ellis 1998:3-8). Other evidence that the 1991 Nehalem catch may not reflect the benefits of hazing is that this catch is comparatively less than at the nonhazed Umpqua (Fig. 18).
The differences in catches between the Nehalem and the nonhazed Salmon, Siletz, and Alsea Basins are significantly greater since hazing began at the Nehalem (Table 14, Fig. 18). These nonhazed basins did not share in the extraordinarily good year of 1991 (ODFW 1998b), but even without that year, Nehalem catches were relatively greater with hazing (Fig. 18). However, it is unclear if these differences result from hazing at the Nehalem or from the decline in OAF releases during the years of hazing, so that fewer OAF coho strayed to the Salmon, Siletz, and Alsea Basins where they contributed to fisheries (section J-1a).
J-1c. UNCHANGED OR DECREASED COHO POST-HAZING CATCHES. Not all evidence suggests that catches have increased at the Nehalem with hazing. For example, the difference in catches between the Nehalem and the Tillamook Basin or the composite of all Oregon coastal streams other than the Nehalem and Tillamook averaged greater before known hazing than after hazing at the Nehalem, although these differences are not significant (Table 14). Further, differences in catches between the Nehalem Basin and the nonhazed Umpqua Basin are significantly greater without than with hazing (Table 14: footnote b).
Finally, the 1989 catch at Tillamook Basin that would have been affected by the 1988 cormorant hazing at Tillamook is much less than for the nonhazing-affected years of 1986 and 1991 (Fig. 17).
J-1d. COHO FISHERIES DISCUSSION. Evidence that hazing resulted in increased coho catches is muddled because not all tests indicate a significant increase with hazing and because several factors other than hazing have also changed that could have contributed to relatively greater catches during the years with hazing (Appendix V).
---------------------------------------------------------------------------TABLE 14. Statistical tests of whether sport catches of coho at the Nehalem Basin or differences between the Nehalem and other basins are significantly greater with hazing. See Fig. 17 for annual catches at Nehalem Basin, Tillamook Basin, and other coastal streams and bays and Fig. 18 for differences in catches between the Nehalem and Salmon River, Siletz, Alsea, and Umpqua Basins; see the legends of these figures for streams included within each of these basins. Starting in 1994, coho sport fisheries were limited to a few streams (Appendix V-8c), so comparisons are not possible for these years.
The larger mean is underlined. Other=sum of coastal stream catches minus catches at the Nehalem and Tillamook Basins; the Columbia River and its tributaries are not included with coastal stream data. N=number of years, U=Mann-Whitney statistic for the hypothesis that Nehalem or Nehalem-(basin[s]) catches are relatively greater with hazing at the Nehalem than without it, P=probability, NS=not significant (one-tailed P>0.05).
---------------------------------------------------------------------------
Number of Coho Caught in Sports Fishery........... Hazing>
Hazed (1989-1993)........ Nonhazed (1975-1985)... Nonhazing
Comparison N Mean Range N Mean Range U P
----------------------------------------------------------------------------
Nehalem Basin 5 3071 1280-8318 11 804 154-2534 51 <0.01
Nehalem-Tilla. 4a -1810 (-9521)-1999 11 -376 (-1545)-404 26 NSb,c
Nehalem-Other 5 -15637 (-40945)-(-4418) 11 -8430 (-12760)-(-646) 20 NSb
Nehalem-Salmon 5 2822 1130-7955 11 465 (-97)-1971 52 <0.01
Nehalem-Siletz 5 2724 795-7861 11 -124 (-1101)-504 55 <0.01
Nehalem-Alsea 5 1149 (-1410)-6426 11 -1578 (-4833)-5 44 <0.05
Nehalem-Umpqua 5 -4937 (-14841)-107 11 -515 (-3197)-1293 10 NSb
a 1989 is not included for Nehalem-Tillamook because there was hazing at
Tillamook in 1988 that may have affected the 1989 sports catch
there.
b When the hypothesis that nonhazing catches are greater than those with
hazing is tested, the results are nonsignificant for
Nehalem-Tillamook and Nehalem-Other but are significant for
Nehalem-Umpqua catches (U=45, one-tailed P<0.05).
c The number of years with known hazing is less than five; statistical
tests would be more robust with larger sample sizes (section E-4e).
---------------------------------------------------------------------------
---------------------------------------------------------------------------
FIGURE 17. Sport catches of hatchery and wild coho at Oregon coastal streams and bays other than the Columbia River during 1975-1993. 1994-1996 data are not included because there was not a sports fishery at most streams in these years, although there was a fishery in part of the Nehalem Basin at the North Fork of the Nehalem River (Appendix V-8c). Data are from ODFW (1987, 1998b) punchcard/tag returns, which are explicitly for adults, although a few jacks may be mistakenly recorded (Eric Schindler, ODFW, pers. comm.). Adults return after two summers in the ocean, so smolts entering the ocean during hazing in the spring of 1988 at the Nehalem and Tillamook would be caught as adults during the 1989 Run-Year.
Other Coastal Streams & Bays=all coastal streams and bays except for Nehalem and Tillamook Basins. The Nehalem Basin includes Nehalem Bay, Nehalem River above and below Elsie, North Fork of Nehalem River, Cook Creek, Rock Creek, and Salmonberry River. The Tillamook Basin includes Tillamook Bay and the following rivers: Kilchis, Miami, Tillamook, Trask, North and South Forks of the Trask, and Wilson.
H=affected by hazing ?=unknown if affected by hazing
o=not affected by hazing
*=anomalously high catches in 1991; only Nehalem catch affected by hazing
----------------------- Coho Sport Catches ----------------------------
Other Coastal Nehalem
Streams & Bays Basin
Nonhazing Mean= 804
Hazing Mean= 3,071
50,000-| *
N -| *
u 45,000-| * 9,000-|
m -| * -| *
b 40,000-| * 8,000-| *
e -| * -| *
r 35,000-| * 7,000-| *
-| * -| *
o 30,000-| * 6,000-| *
f -| * -| *
25,000-| * 5,000-| *
C -| o * -| *
o 20,000-| o * 4,000-| *
h -| o *o -| *
o 15,000-| o o o *o 3,000-| ? *
-|oo o ooo o *o -| o ? * H
10,000-|ooo oo ooo oo *oo 2,000-| o ? ? H* H
-|ooo oooooooooo*oo -|-o---------?-?HH*HH
5,000-|ooo ooooooooooo*oo 1,000-|ooo oo? ?HH*HH
-|oooo ooooooooooo*oo -|ooo oo o oo???HH*HH
1 -1,249-|oooooooooooooooo*oo 1-249-|ooooooooooo???HH*HH
0-|oooooooooooooooo*oo 0-|ooooooooooo???HH*HH
|''''|''''|''''|*'' |''''|''''|''''|*''
75 80 85 90 75 80 85 90
Run-Year
Tillamook
Basin
17,839-*
N |
u 9,000-| *
m -| *
b 8,000-| *
e -| *
r 7,000-| *
-| *
o 6,000-| o *
f -| o *
5,000-| o *
C -| o *
o 4,000-| o *
h -| o *
o 3,000-| o *
-| o o *
2,000-| oo o o o oH *
-| oo o ooo oH *o
1,000-|ooo o ooo oHo*o
-|oooooooo oooooHo*oo
1-249-|ooooooooooooooHo*oo
0-|ooooooooooooooHo*oo
|''''|''''|''''|*''
75 80 85 90
Run-Year
---------------------------------------------------------------------------
---------------------------------------------------------------------------
FIGURE 18. Differences in sports catches of coho between the hazing-affected Nehalem Basin and the nonhazing-affected Salmon River, Siletz, Alsea, and Umpqua Basins. The streams included in Nehalem Basin are in the legend of Fig. 17, which graphs Nehalem catches. The Siletz Basin includes Siletz Bay, Siletz River, North and South Forks of the Siletz River, Drift Creek, Rock Creek, Little Rock Creek, and Schooner Creek. The Alsea Basin includes Alsea Bay, Alsea River, North and South Forks of the Alsea River, Drift Creek, Fall Creek, Five Rivers, and Lobster Creek. The Umpqua Basin includes Umpqua River & (Winchester) Bay, North and South Umpqua River, Smith River, and North Fork of the Smith River. These data are from punchcard/tag data in ODFW (1987, 1998b).
H=Nehalem affected by hazing
?=unknown if Nehalem affected by hazing
o=not affected by hazing
*=1991 was an anomalously high catch year for the hazing-affected Nehalem
but not for all nonhazing-affected basins (also see Fig. 17)
------------------------ Coho Sport Catches -------------------------
Nehalem-Salmon River Nehalem-Siletz
Nonhazing Mean= 465 Nonhazing Mean= -124
Hazing Mean=2,822 Hazing Mean= 2,724
*-7,955 *-7,861
| |
N 3,000-| * -| *
u | * | *
m 2,500-| * -| *
b | ? * | *
e 2,000-| o ? * H -| ? * H
r | o ? ? H* H | ? ? H* H
1,500-| o ? ? H* H -| ? ? H* H
o | o ? ?HH*HH | ? ? H* H
f 1,000-| o ? ?HH*HH -| ? ?HH* H
| o o? ?HH*HH | ? ?HH*HH
C 500-|ooo o o? ?HH*HH -| o? ?HH*HH
o |ooo oo o oo???HH*HH | o oo o???HH*HH
h 0-|ooooooooooo???HH*HH -|ooooooooooo???HH*HH
o | |o o o oo
-500-| -| o
| | o
-1,000-| -| o
| |
|___________________ |___________________
|''''|''''|''''|*'' |''''|''''|''''|*''
75 80 85 90 75 80 85 90
Run-Year
Nehalem-Alsea Nehalem-Umpqua
Nonhazing Mean= -1,578 Nonhazing Mean= -515
Hazing Mean= 1,149 Hazing Mean= -4,937
*-6,426
|
5,000-| * -|
| * |
4,500-| * -|
| * |
4,000-| * -|
| * |
3,500-| * -|
| * |
3,000-| * -|
| * |
2,500-| * -|
| * |
2,000-| * -|
N | * |
u 1,500-| * -|
m | H* | o
b 1,000-| H* -| o
e | H* | o
r 500-| H* -| o
| ? H* H | o oo
o 0-|ooooooooooo???HH*HH -|ooooooooooo???HH*HH
f |oooo oooooo ? H H |o oo oooo ???H *HH
-500-|oooo oooooo H H -|o oo oooo ? H *HH
C |oo ooooo H H |o ooo ? H *HH
o -1,000-|oo ooooo H H -|o ooo H *HH
h |oo ooooo H | o H *HH
o -1,500-|oo oo H -| o H *HH
|oo oo | o *HH
-2,000-|oo oo -| o *H
| o oo | o *H
-2,500-| o o -| o *H
| o o | o *H
-3,000-| o o -| o *H
| o o | o *H
-3,500-| o -| *H
| o | *H
-4,000-| o -| *H
| o | *H
-4,500-| o -| *H
| o | *H
-5,000-| -| *H
| | ||
| | (-14,841)-*H-(-6,619)
|___________________ |___________________
|''''|''''|''''|*'' |''''|''''|''''|*''
75 80 85 90 75 80 85 90
Run-Year
---------------------------------------------------------------------------
J-2a. INTRODUCTION. There is no ocean fishery for steelhead (Pauley et al. 1986, ODFW 1998b), and many factors, including the number of smolts released (see Appendix III), can affect the number of fish caught (Appendix V). For example, starting in 1992, only hatchery-reared steelhead could be caught and kept (Chilcote 1998:31). Additionally, more steelhead would be expected to be caught with increased fishing effort, and effort can be quite variable, since yearly steelhead fishing effort at the Alsea during 1975-1987 ranged from 40,000 to 153,000 hours (ODFW 1997a:79).
Standardizing catches as a percentage of the number of steelhead smolts released is not appropriate because of straying of Nehalem-reared fish released into the Necanicum (see Appendix III) back into the Nehalem, because steelhead catches could include both 2- and 3-salts and the proportion returning in each age-class can vary yearly (Appendix II-2d), and because catches prior to 1992 included wild fish.
Usually less than 50 summer steelhead were caught at the Nehalem (ODFW 1998b), which are too few to correlate with hazing. Summer steelhead catch data are not available at the Tillamook and Nestucca Basins to see if they have changed with hazing during the springs of 1996-1998. Consequently, catch data are only analyzed for winter steelhead.
J-2b. WINTER STEELHEAD CATCHES. Graphical results indicate that winter steelhead catches have not increased with hazing at the Nehalem (Fig. 19), the average Nehalem catch with hazing is less than half of the nonhazing catch, and nonhazing-affected catches are significantly greater (Table 15: footnote a). Since hazing occurred only in 1988 and 1996-1998 at the Tillamook, there are no years of catch data yet available that include all age-classes of adult steelhead that could have been affected by hazing (Fig. 19).
Annual catch differences between the Nehalem and Siletz or Alsea Basins are generally larger (Fig. 20), average more, and are significantly greater with hazing (Table 15). On the other hand, annual catch differences between the Nehalem and Necanicum and Salmon River Basins are usually less (Fig. 20), average less (Table 15), and are significantly less with hazing for the Necanicum (Table 15: footnote a).
---------------------------------------------------------------------------TABLE 15. Statistical tests of whether sport catches of winter steelhead at the Nehalem Basin or differences between the Nehalem and other basins are significantly greater with hazing. See Fig. 19 for annual catches at the Nehalem Basin and Fig. 20 for annual differences in catches between the Nehalem and the Necanicum, Salmon, Siletz, and Alsea Basins; see the legends of these figures for streams included within each of these basins.
Although the catch of 2-salts in 1989-1990 at Nehalem was affected by hazing, it is not included because the proportion of the 1989-1990 catch that was 2-salts is unknown, and the proportion of the catch that year that returned as 3- or 4- salts or repeat spawners was not affected by hazing.
The larger mean is underlined. Necan.=Necanicum River. N=number of years, U=Mann-Whitney statistic for the hypothesis that Nehalem or Nehalem-(basin) catches are relatively greater for hazing than nonhazing, P=probability, NS=not significant (one-tailed P>0.05).
---------------------------------------------------------------------------
Number of Winter Steelhead Caught...............
Hazed (1990-91 Nonhazed (1975-76 Hazing>
through 1995-96)..... through 1985-86)....... Nonhazing
Comparison N Mean Range N Mean Range U P
---------------------------------------------------------------------------
Nehalem Basin 6 2107 1064-3997 11 4479 2090-10162 10 NSa
Nehalem-Necan. 6 1109 426-2312 11 2652 885-7227 15 NSa
Nehalem-Salmon 6 1653 809-3560 11 2570 314-6844 25 NSa
Nehalem-Siletz 6 549 (-126)-2292 11 -1266 (-3621)-2665 55 <0.01
Nehalem-Alsea 6 662 (-75)-2212 11 -2595 (-5232)-514 62 <0.01
a When the hypothesis that nonhazing catches are greater than hazing is
tested, the results are not significant for Nehalem-Salmon,
significant for catches at the Nehalem (U=56, one-tailed P<0.01),
and significant for Nehalem-Necanicum (U=51, one-tailed P<0.05).
---------------------------------------------------------------------------
---------------------------------------------------------------------------
FIGURE 19. Sport catches of hatchery and wild winter steelhead at Oregon coastal streams and bays other than the Columbia River. Data are from ODFW (1987, 1998b) punchcard/tag returns, which are explicitly for adults, although a few jacks may be mistakenly recorded on them (Eric Schindler, ODFW, pers. comm.). For smolts released during the spring of 1988, most adults would return during the spring of 1990 (1989-1990 Run-Year) as 2-salts, some would return during the spring of 1991 (1990-1991 Run-Year) as 3-salts, and a few would return in later years as repeat spawners. Kenaston (1989:16) gives the percentage of winter steelhead that were caught that were of hatchery origin.
Starting in 1992, only fin-clipped (hatchery) steelhead could be kept (Chilcote 1998:31-33), so this regulation could have reduced catches.
Other Coastal Stream & Bays=all coastal streams except for Nehalem and Tillamook Basins. The Nehalem Basin includes Nehalem Bay, Nehalem River above and below Elsie, North Fork of Nehalem River, Cook Creek, Rock Creek, and Salmonberry River; during 1980-81 through 1984-85, 59-84% of winter steelhead caught here were of hatchery origin (Kenaston 1989:16). The Tillamook Basin includes Tillamook Bay and the following rivers: Kilchis, Miami, Tillamook, Trask, North and South Forks of the Trask, Wilson, and Devils Lake and Little North Forks of the Wilson.
H=affected by hazing ?=unknown if affected by hazing
o=not affected by hazing #=wild steelhead released
2=only 2-salts affected by hazing 3=only 3-salts affected by hazing
@=only 4-salts and repeat spawners affected by hazing
---------------- Winter Steelhead Sport Catches (Thousands) ----------
Other Coastal Nehalem
Streams & Bays Basin
Nonhazing Mean= 4,479
Hazing Mean= 2,107
100-| 16-|
-| o -|
90-| o o 14-|
-| o o o -|
80-| o o o 12-|
T -| o o o -|
h 70-| o o o 10-| o
o -|o o oo o -| o
u 60-|o oooo o 8-| oo
s -|o ooooo ooooo -| oo
a 40-|ooooooo ooooo o 6-| o oo ?
n -|ooooooo ooooooo -| o oo ?
d 30-|ooooooo ooooooooo 4-| o oo oo ?? 2 H
s -|oooooooooooooooooo o -|oooooo ooo???2 H
20-|ooooooooooooooooooooo 2-|ooooooooooo???2HHH
-|ooooooooooooooooooooo -|ooooooooooo???2HHHHHH
0-|ooooooooooooooooooooo 0-|ooooooooooo???2HHHHHH
|_________________#### |_________________####
|''''|''''|''''|''''| |''''|''''|''''|''''|
75- 80- 85- 90- 95- 75- 80- 85- 90- 95-
76 81 86 91 96 76 81 86 91 96
Run-Year
Tillamook
Basin
16-|
-| o
14-| o
-| o o
12-| o o
T -| ooo
h 10-| ooo
o -| oooo o
u 8-|oooooo o o
s -|ooooooo oo oo
a 6-|ooooooo oo oo
n -|oooooooooooooo2
d 4-|oooooooooooooo2 @
s -|oooooooooooooo23@@
2-|oooooooooooooo23@@@
-|oooooooooooooo23@@@@@
0-|oooooooooooooo23@@@@@
|_________________####
|''''|''''|''''|''''|
75- 80- 85- 90- 95-
76 81 86 91 96
Run-Year
---------------------------------------------------------------------------
---------------------------------------------------------------------------
FIGURE 20. Differences in sports catches of winter steelhead between the Nehalem Basin and the Necanicum River, Salmon River, Siletz Basin, and Alsea Basins. The streams included in Nehalem Basin are listed in Fig. 19, which graphs Nehalem catches. The Siletz Basin includes Siletz Bay, Siletz River, Drift Creek, Rock Creek, Little Rock Creek, Schooner Creek, and the North and South Forks of the Siletz River. The Alsea Basin includes Alsea River and Bay, Drift Creek, Fall Creek, Five Rivers, Lobster Creek, and North and South Forks of the Alsea River. Data are from ODFW (1987, 1998b) punchcards/tags.
H=Nehalem affected by hazing ?=unknown if Nehalem affected by hazing
o=not affected by hazing #=wild steelhead released
2=only 2-salts affected by hazing
------------------------ Winter Steelhead Sport Catches -------------------
Nehalem-Necanicum R. Nehalem-Salmon R.
Nonhazing Mean= 2,652 Nonhazing Mean= 2,570
N 8,000-| Hazing Mean= 1,109 -| Hazing Mean= 1,653
u | o |
m 7,000-| oo -| o
b | oo | oo
e 6,000-| oo -| oo
r | oo | oo
5,000-| oo -| oo
o | oo | oo ?
f 4,000-| oo -| oo ?
| oo ? | o oo ? H
S 3,000-| o oo ? -| o oo ? 2 H
t |o o oo ? 2 H | oo oo ? 2 H
e 2,000-|oo oo ooo? 2 H -| oo ooo???2 H
e |oo oo oooo? ?2H H |oo ooo oooo???2HHH
l 1,000-|ooooooooooo???2HHH H -|oooooo oooo???2HHHHHH
h |ooooooooooo???2HHHHHH |ooooooooooo???2HHHHHH
e 0-|ooooooooooo???2HHHHHH -|ooooooooooo???2HHHHHH
a |_________________#### |_________________####
d |''''|''''|''''|''''| |''''|''''|''''|''''|
75- 80- 85- 90- 95- 75- 80- 85- 90- 95-
76 81 86 91 96 76 81 86 91 96
Nehalem-Siletz Basin Nehalem-Alsea Basin
Nonhazing Mean= -1,266 Nonhazing Mean= -2,595
N 3,000-| Hazing Mean= 549 -| Hazing Mean= 662
u | o H |
m 2,000-| o ? H -| H
b | oo ? 2 H | H
e 1,000-| oo ? ?2 H -| 2H H
r | oo ? ?2 HH | o ? 2H H H
0-|ooooooooooo???2HHHHHH -|ooooooooooo???2HHHHHH
o |oooooooo o ? |oooooooo oo ?
f -1,000-| oooooo ? -|o ooooo oo ?
| oooooo ? |o ooooo o ?
S -2,000-| ooooo -|o o ooo o
t | ooooo |o o ooo o
e -3,000-| ooo o -|o o ooo o
e | oo |o o o o
l -4,000-| -|o o o o
h | |o o o o
e -5,000-| -| o o
a | | o
d |_________________#### |_________________####
|''''|''''|''''|''''| |''''|''''|''''|''''|
75- 80- 85- 90- 95- 75- 80- 85- 90- 95-
76 81 86- 91 96 76 81 86 91- 96
Run-Year
---------------------------------------------------------------------------
J-3a. NEHALEM CHINOOK CATCHES. The classification of the type of chinook at Nehalem as fall, fall/summer, summer, and/or spring (Appendix II-3a) confuses data analyses because chinook catches at the Nehalem are divided into fall and spring chinook catches with no spring chinook catches prior to 1989 (ODFW 1998b). Because Nicholas and Hankin's (1989) report may have affected classification of Nehalem chinook catch data, I suspect that "spring chinook" catches prior to 1989 were lumped with fall chinook catches.
Adult chinook can return as 3-6 yr olds (Appendix II-3c), 2-5 yrs after migrating to the ocean, so the 1993 Run-Year would be the first when all adults may have been affected by hazing. Since complete data are only available through 1996 (ODFW 1998b), there are only three years for fall chinook and four years for spring chinook that could be totally ascribed as affected by hazing (Fig. 21).
Fall chinook catches at the Nehalem have clearly increased during the time of hazing (Fig. 21), and catches in hazing years (1993-1995) are significantly greater than in nonhazing years (1975-1986)(Mann-Whitney U=36, N1=3, N2=12, one-tailed P<0.01). However, it is unclear if hazing is the cause for this increase or if changes in other factors such as fishing effort are involved. Indeed, the Nehalem run is classed by Nicholas and Hankin (1989:204) as wild, and the increase in catches since 1990 (Fig. 21) is also correlated with a decrease in the number of wild fall chinook at spawning grounds (section G-6). Consequently, the increased catches may be more a result of increased fishery exploitation in recent years, rather than hazing. A concern about overfishing was raised in 1997, when some local residents tried to reduce the daily bag limit from two to one fall chinook in Tillamook County (Monroe 1997).
Spring chinook catches have also increased in recent years, but there are no catch data for nonhazing-affected years (Fig. 21).
J-3b. NEHALEM TILLAMOOK AND NESTUCCA CATCHES. Because adult chinook return at various ages up to six years old (Appendix II-3c), hazing only consistently began at the Tillamook and Nestucca in the spring of 1996, and complete fishery data are only available through the 1995-1996 Run Year (ODFW 1998b), there are not enough data to examine chinook fishery catches at the Tillamook and Nestucca Basins to see if they may have been affected by hazing.
---------------------------------------------------------------------------FIGURE 21. Sport catches of fall and spring chinook salmon at Nehalem. Nehalem includes Nehalem Bay, Nehalem River above and below Elsie, and North Fork of Nehalem River; Cook Creek, Rock Creek, and Salmonberry River are also tributaries of the Nehalem but are excluded because of missing data in some years. Data are from ODFW (1987, 1998b) punchcard/tag returns, which are explicitly for adults, although a few jacks may be mistakenly recorded on them (Eric Schindler, ODFW, pers. comm.). Adult chinook return as 3-6 yr olds (Appendix II-3c), 2-5 years after migrating to sea, so juvenile chinook entering the ocean during hazing in the spring of 1988 at the Nehalem would be caught as adults during the 1990-1991 through 1993-1994 Run-Years, and 1993-1994 would be the first Run-Year in which all adult chinook may have been affected by hazing. It is unclear if hazing occurred in 1985-1987.
Spring chinook at the Nehalem were first included in catch data in 1989 (ODFW 1987, 1998b). Since the fall, fall/summer, summer, and/or spring classification of stock(s) at the Nehalem varies among references, it is possible that "spring" chinook catches were pooled with "fall" chinook catches at the Nehalem prior to 1989 (section J-3a).
H=all affected by hazing ?=unknown if affected by hazing
o=not affected by hazing P=part of catch affected by hazing
--------------------------- Nehalem Catches -------------------------------
"Fall" Chinook
Nonhazing Mean= 672.4
Hazing Mean= 3493.7 "Spring" Chinook
4,000-| H
-| H
3,500-| H H
C -| H H
h 3,000-| HHH
i -| PHHH
n 2,500-| PHHH -| H
o -| ? PHHH 1,000-| H
o 2,000-| ?? PPHHH -| H
k -| ?? PPHHH 800-| HH
1,500-| o???PPPHHH -| HHH
-| o o???PPPHHH 600-| HHH
1,000-| o o???PPPHHH -| P HHH
-| o o ooo???PPPHHH 400-| PPHHHH
500-| oooooooooo???PPPHHH -| PPHHHH
-| oooooooooo???PPPHHH 200-| PPPHHHH
1-124-|oooooooooooo???PPPHHH -| ?PPPHHHH
0-|oooooooooooo???PPPHHH 0-|<=No Data====>?PPPHHHH
|''''|''''|''''|''''| |''''|''''|''''|''''|'
75 80 85 90 95 75 80 85 90 95
Run-Year
***************************************************************************
The goals of salmon management are the recovery of wild salmonids and, secondarily, maintaining or improving fisheries (OCSRI 1997a,b; NMFS 1998, Kitzhaber 1999).
Improving smolt survival by hazing seems to make common sense, which is why it has been so appealing. The logic used by proponents is that cormorants eat smolts, and hazing stops cormorants from eating smolts; therefore, hazing improves smolt survival. There is no dispute that cormorants can eat smolts, although the amount and kind (wild or hatchery) can be debated (e.g., Bayer 1989:25-29, Roby et al. 1998). But the most important issue is whether hazing helps attain the goals of salmon management, and this does not appear to be the case.
K-2a. INTRODUCTION. Correlation does not mean causation--this is an axiom in statistics. But it is also important in evaluating hazing because there are several variables that can affect salmonid returns other than hazing (Appendix V). Thus, any increases or decreases in salmonid returns that are correlated with the occurrence of hazing may not have been caused by hazing.
K-2b. 1991 COHO ANOMALY. An example of where correlating returns with hazing may be misleading occurred in 1991 for returning coho. At Nehalem, the return of adults in 1991 was much higher than normal based on CWT survival rates (section H), returns to the hatchery (section I-1b), and fishery catches (section J-1b). Further, spawning ground counts of adults were also slightly higher than usual in 1991 (Fig. 4). Whatever affected the high return rate of adults in 1991 was not reflected in greater returns of jacks in 1991 to the spawning grounds (Fig. 2) or hatchery (Fig. 14).
One could assume that this extraordinary return may have been caused by hazing during 1990; however, increased CWT recoveries (section H) and sports catches (section J-1b) in 1991 also occurred at some nonhazed basins. In addition, if hazing was the cause of the greater adult returns to Nehalem in 1991, one would expect that the number of jacks returning in 1990 would also be much higher than normal. But the number of jacks returning in 1990 to spawning grounds was less than usual (Fig. 2), and the number returning to hatcheries in 1990 was not as anomalously high as the 1991 adult return (Fig. 14), so the increased return of adults in 1991 appears to be a consequence of some factor that occurred after the hazing in the spring of 1990 and the return of jacks in the fall of 1990.
Hazing may affect the survival of juvenile salmonids during their outmigration to the ocean, so an immediate measure of their survival to the ocean would be informative (also see section E-2). However, delayed measures (i.e., returns of jacks and adults) are essential in evaluating hazing for three reasons. First, proponents have claimed that hazing has improved salmon and steelhead returns (Erickson 1989a,b,d; 1992, 1993, 1995a,b; Monroe 1995b, 1996a [see section E-3c]; Nokes 1995). For instance, Erickson (1993) wrote:
"Many people are asking why the Nehalem River has rebounded to
record numbers of coho and steelhead. Those of us who have worked on a
smolt protection plan for the last six years feel very strongly that
our efforts have helped considerably in the increased fish runs."
Second, the goal of recovery plans and fisheries is improving returns of adults, and increasing the number of smolts reaching the ocean may not help. For example, increasing the number of coho smolts released in Oregon has not resulted in greater adult returns (Pearcy 1992:48). Third, using returns is appropriate in assessing hazing because smolts saved from cormorants in estuaries may die anyway from disease, parasites, oceanic predators, and/or unfavorable ocean conditions (Appendix V).
The numbers of wild adult coho (section G-3), winter steelhead (section G-4), and fall chinook (section G-6) returning to spawning grounds have averaged less with than without hazing at hazed basins, so wild adult abundance has not increased significantly with hazing (Table 16). Similarly, the abundance of jack coho (section G-2) and chinook (section G-5) also was usually less with hazing at hazed basins, although the relative difference between Nehalem and nonhazed basins for coho jacks was sometimes significantly greater with hazing (Table 16).
Thus, hazing does not appear to be useful in attaining the goal of recovery of wild salmonids in the Oregon Plan, in removing coastal wild coho from a federal ESA listing, or in helping coastal steelhead from becoming listed under the federal ESA.
---------------------------------------------------------------------------TABLE 16. Summary of statistical tests comparing results of hazing with nonhazing. A significant difference does not establish that hazing is the cause of a difference, only that hazing is correlated with the difference. Hazed Basins=Nehalem, Tillamook, and Nestucca Basins. Statistical tests for Hazing>Nonhazing are expressed as the number of tests with a significant difference divided by the total number of tests.
---------------------------------------------------------------------------
Stat.
Tests
Haz>
Taxon Site(s) Type of Count Non Section
------------------------------------------------------------------------
Wild Fish (jacks)
coho Hazed Basins spawn grounds 0/3 G-2
coho Nehalem-Nonhazed Basins spawn grounds 2/6 G-2
fall chinook Hazed Basins spawn grounds 0/3 G-5
fall chinook Nehalem-Nonhazed Basins spawn grounds 0/3 G-5
Wild Fish (adults)
coho Hazed Basins spawn grounds 0/3 G-3
coho Nehalem-Nonhazed Basins spawn grounds 0/6 G-3
winter steel Nehalem spawn grounds 0/1 G-4
fall chinook Nehalem spawn grounds 0/1 G-6
fall chinook Nehalem-Nonhazed Basins spawn grounds 0/3 G-6
Hatchery Fish (jacks)
coho Nehalem-Nonhazed Basin % return 0/1 I-1a
coho Hazed Basins returns 0/2 I-1b
winter steel Hazed Basins returns 1/2 I-2b
summer steel Nestucca returns * I-2c
chinook Hazed Basins returns * I-3
Hatchery Fish (adults)
coho Nehalem-Nonhazed Basin % return 0/1 I-1a
coho Hazed Basins returns 1/1 I-1b
winter steel Nehalem returns 0/1 I-2b
Hatchery Fish (jacks + adults)
coho Nehalem Coded Wire Tag 0/1 H
coho Nehalem-Nonhazed Basins Coded Wire Tag 2/3 H
Fisheries (adults)
coho Nehalem catches 1/1 J-1
coho Nehalem-Nonhazed Basins catches 3/6 J-1
winter steel Nehalem catches 0/1 J-2
winter steel Nehalem-Nonhazed Basins catches 2/4 J-2
fall chinook Nehalem catches 1/1 J-3
* No statistical test because of too few years of hazing-affected
returns, but there was no apparent increase with hazing.
---------------------------------------------------------------------------
Hatchery return data do not show a consistent trend of significantly increased returns for jacks, adults, or CWT marked fish with hazing (Table 16). As discussed in each of the sections cited in Table 16, even in the few cases when increased returns are correlated with hazing, other factors may have been responsible. Further, even if hazing increased the number of fish returning to hatcheries, this is not a goal of salmon management because coastal hatcheries have a surplus of fish to use for broodstock and surplus fish are usually unfit for human consumption (Orcutt 1992; section I-1b).
A secondary goal of salmon management is to maintain or improve fisheries, which have been curtailed in recent years because of concerns about wild salmonids. Increased coho and fall chinook catches (but not winter steelhead) are correlated with hazing at the Nehalem as are increased relative differences in catches between the Nehalem and half of the nonhazed basins for coho and winter steelhead (Table 16). However, the increased catches of coho may be more a result of factors other than hazing (section J-1, Appendix V). Additionally, greater fall chinook catches may have arisen from increased fishing effort, not hazing, and to have resulted in significantly decreased wild fall chinook escapement to spawning grounds (section G-6), which, potentially, could lead to a re-examination of listing coastal fall chinook under the federal ESA. Thus, evidence that hazing has improved fisheries is mixed.
Evidently the cormorant predation issue is not as simple as it may first seem or returns of wild and hatchery fish would have improved much more significantly with hazing than shown in Table 16. Similarly, Draulans (1987:221-223) noted that other fish-eating bird control programs have not demonstrated an increase in fish abundance.
There are three reasons why returns may not have greatly improved with hazing. First, smolts saved with hazing may have died anyway from factors such as disease, parasites, and/or predators in the ocean (Appendix V). Second, an increase in returns as a result of hazing may have been masked by factors such as unfavorable ocean conditions and changing fishery regulations and effort (Appendix V).
Finally, hazing may not have improved returns because hazing did not significantly reduce cormorant predation. During 1996-1998, counts of cormorants indicate that they were present after hazing commenced (section F-7), some smolts may have already migrated before hazing began each year (section C-4), hazing did not occur throughout the day at each estuary (sections C-5 and F-6), and one hazer might not have been able to adequately cover each estuary. In addition, scarecrows were used somewhat to keep cormorants away when hazers were absent (section C-2), but scarecrows are usually ineffective after a short period of time (Draulans 1987, Stickley et al. 1995). In 1999, after SPP hazing began on April 1 (Stahl et al. 2000:7), it was noted that hazed cormorants were interrupted in their feeding but did not leave an estuary (Stahl et al. 2000:34), and substantial numbers of potential predators were still counted (Stahl et al. 2000:50-51, 53, 67, 69). Further, 55% of 20 radiotagged hatchery coho smolts released at Nehalem during 1999 hazing were thought to have been caught by predators before they reached the ocean (Stahl et al. 2000:7, 28, 44), so it appears that appreciable predation occurred in spite of hazing. However, radiotagging may increase smolt vulnerability to mortality such as predation (e.g., Adams et al. 1998, Hockersmith et al. 1999), although Stahl et al. (2000:29) do not think this is true in their study. In any case, if hazing reduced predation, it may not have done so significantly.
At best, cormorant hazing is correlated with mixed results for fisheries and hatchery returns, but hazing is not correlated with increased returns of wild salmonids, which are the focus of the Oregon Plan and federal ESA listings. Consequently, it is unclear if the benefits of hazing equal the $100,000 spent on it during 1996-1999 by the Oregon Legislature and its biological and social costs (e.g., the harassment of wildlife other than cormorants, see section F-3; Bayer 1989:40-46).
***************************************************************************
brant, black Branta bernicla nigricans
cormorant spp. Phalacrocorax spp.
cormorant, Brandt's Phalacrocorax penicillatus
cormorant, double-crested Phalacrocorax auritus
cormorant, pelagic Phalacrocorax pelagicus
crow spp. Corvus spp.
dogfish, spiny Squalus acanthias
eagle, bald Haliaeetus leucocephalus
gull spp. Larus spp.
heron, green Butorides virescens
heron, great blue Ardea herodias
lion, sea Zalophus californianus or Eumatopias jubatus
loon spp. Gavia spp.
mackerel, Pacific Scomber japonicus
merganser, hooded Lophodytes cucullatus
mink Mustela vison
murre, common Uria aalge
osprey Pandion haliaetus
otter, northern river Lutra canadensis
pelican, brown Pelecanus occidentalis
"perch, pink-tailed" ? (probably one of surfperches in Family
Embiotocidae)
"pogey" ? (Cottidae ?)
raccoon, common Procyon lotor
rockfish spp. Sebastes spp.
rockfish, black Sebastes melanops
salmon, Atlantic Salmo salar
salmon, chinook Oncorhynchus tshawytscha
salmon, coho Oncorhynchus kisutch
sandlance, Pacific Ammodytes hexapterus
sculpin ? (probably in Family Cottidae)
seal, harbor Phoca vitulina
shearwater spp. Puffinus spp.
"shiner" ? (probably one of surfperches in Family
Embiotocidae)
steelhead Oncorhynchus mykiss
tern spp. Sterna spp.
tern, Caspian Sterna caspia
trout, cutthroat Oncorhynchus clarki
***************************************************************************
In Oregon, most wild juvenile coho appear to remain in freshwater one year before migrating (K. W. Myers 1980:121-122), a few may be present less than a year (Reimers 1971 as cited in K. W. Myers 1980:121), and others reside two years (Chapman 1961:35, Moring and Lantz 1975:23).
Peak emigration of wild coho smolts through Oregon estuaries is in May, but some also migrate through estuaries in April and June (Bottom and Forsberg 1978:55, K. W. Myers 1980:38, 120; J. A. Johnson et al. 1986:6, 8; O. W. Johnson et al. 1991:6). Most hatchery smolts were released in late March or April at the Nehalem (Table 3). Coho smolts only appear to remain a few days to a few weeks in estuaries (Emmett et al. 1991:138), and, at Yaquina Estuary, 50% of hatchery smolts appeared to have left within 2-9 days, and 90% or more left within a month of release (K. W. Myers 1980:56-68, 122, 133-139; K. W. Myers and Horton 1982). Thus, hazing may have the greatest effect on hatchery fish if it is done within a month of a release.
Precocious males (jacks) or females (jills or jennies) return to freshwater in the fall of the same calendar year, and adults return to spawn in freshwater during the fall of the next calendar year (Laufle et al. 1986, Kostow 1995:65, Weitkamp et al. 1995:25).
II-2a. STEELHEAD TYPES. Steelhead returning to freshwater from the ocean in May-October are classed as summer steelhead, and those returning in November-April as winter steelhead (Pauley et al. 1986:4-5, Kostow 1995:99). Except for the timing of their return, it is unclear how life histories of summer and winter steelhead differ.
Kostow (1995:111-112) writes that wild steelhead along the north and mid-coast of Oregon are winter steelhead except for a severely depressed summer steelhead run at the Siletz River. Hatchery winter steelhead have been released into the Nehalem, Tillamook, and Nestucca Basins (Kostow 1995:112, A2-A4; Chilcote 1998:33-35, Ellis 1998:3-12). However, many hatchery winter steelhead release programs were terminated or reduced in 1995 in portions of the Nehalem, Tillamook, and Nestucca Basins, although releases of 63-83 thousand winter steelhead smolts continue into the North Fork of the Nehalem (Chilcote 1998:33-35; Appendix III). Hatchery summer steelhead are released into the hazed Tillamook and Nestucca Basins but not the Nehalem Basin (Kostow 1995:112, A2; Chilcote 1998:34-35, Ellis 1998: p. 3-12, John Leppink, ODFW, pers. comm.). Generally fewer than 50 summer steelhead are caught annually in the Nehalem Basin (ODFW 1998b); these may be strays from hatchery programs in other basins.
II-2b. FRESHWATER RESIDENCE. Hatchery-reared juvenile steelhead are often considered to remain in freshwater only one year (Pauley et al. 1986:6, Busby et al. 1996:15), but, at the Alsea in the 1950's, 32% were in freshwater two years (Chapman 1958:125). Wild juveniles usually spend two and sometimes three years in freshwater with a range of 1-4 years (Chapman 1958:125, Wagner et al. 1963, Lindsey et al. 1991:19, 1992:21, 1993:28, 1994:32, 1995:29; Busby et al. 1996:25).
II-2c. TIMING OF SMOLT EMIGRATION. Most wild winter steelhead passed through the North Fork of Nehalem River in April through mid-May of 1998-1999 (ODFW 1999), which is characteristic of other Oregon coast streams, although appreciable numbers can also be migrating in March and June (Chapman 1958:132, Wagner et al. 1963:205, ODFW 1999). At the Nehalem, hatchery smolts were generally released in April (Table 3). Accordingly, wild and hatchery steelhead smolts could have been present during cormorant hazing. Steelhead smolts spend little time in estuaries while migrating to the ocean (Emmett et al. 1991:148).
II-2d. AGE AT RETURN. Jacks return after one summer in the ocean and are consequently categorized as 1-salts; about 1-10% of returning winter steelhead were jacks (Chapman 1958:125, Weber and Knispel 1977:41- 43, Lindsay et al. 1991:18-19, 1992:20-21, 1993:27-28, 1994:31-32, 1995:28-29). On average, about 75-79% of returning hatchery and 66-80% of wild winter steelhead spent two summers in the ocean (2-salts), 12% of hatchery and 8-26% of wild steelhead were in the ocean three summers (3-salts), 0-3% of steelhead were in the ocean four summers (4-salts), and 6-15% were repeat spawners, although the percentage returning in each age class often varied annually and hatchery steelhead tended to return earlier than wild steelhead (Chapman 1958:125, Weber and Knispel 1977:41- 43, Lindsay et al. 1991-1995; Busby et al. 1996:29). Because many return as 3-salts, hazing must occur for at least three consecutive years before the vast majority of those that return could be assumed to have been affected by hazing.
II-3a. CHINOOK TYPES. The life history of chinook salmon is complex (Nicholas and Hankin 1989, Emmett et al. 1991:160-168, Kostow 1995:18), and the classification of run type sometimes differs among references, probably because the timing of runs sometimes does not fit into simple categories. Chinook return to the Nehalem from July through October, fall chinook return to the Tillamook Basin from September through February, fall chinook return to the Nestucca Basin from August through January, and spring chinook return to the Tillamook and Nestucca Basins from April through July (Nicholas and Hankin 1989:23, Kostow 1995:23, 25). Runs at the Nehalem have been classed as fall (Nicholas and Hankin 1989:23, 27; Jacobs and Cooney 1997), fall and summer (Nicholas and Hankin 1989:91, Kostow 1995:19, 23, 25), fall and spring (ODFW 1998b), or summer (J. M. Myers et al. 1998:43). In the Tillamook Basin, runs have been categorized as fall and spring (Nicholas and Hankin 1989:23, 134; Kostow 1995:19, 23, 27; Lewis 1997:12, J. M. Myers et al. 1998:44, ODFW 1998b), with a winter run at the Trask River (Lewis 1997:12, John Leppink, ODFW, hatchery return data). At the Nestucca Basin, runs have been described as fall and spring (Nicholas and Hankin 1989:23, 27; Kostow 1995:19, J. M. Myers et al. 1998:44, ODFW 1998b) or as fall, spring, and summer (Nicholas and Hankin 1989:94).
The Nehalem hatchery currently does not release chinook into the Nehalem system (R. Klumph, ODFW, pers. comm.), and hatchery releases of chinook in the past have been very limited (Nicholas and Hankin 1989:204, 253; Kostow 1995:26, A2-A4). However, hatchery releases of fall and spring chinook have been extensive in the Tillamook and Nestucca Basins (Nicholas and Hankin 1989: 257, 283, 289; Kostow 1995: A2-A5).
In this paper, I list the run type as the one given in a reference.
II-3b. ESTUARINE OCCURRENCE OF JUVENILES. Juvenile chinook in Oregon predominately enter the ocean before they are a year old (Nicholas and Hankin 1989:16, Kostow 1995:18, J. M. Myers et al. 1998:48). Juveniles are known to be present in the lower Nehalem estuary from late spring through at least mid-September (Nicholas and Hankin 1989:91). In Tillamook Bay, they were reported as present from January through at least November (Nicholas and Hankin 1989:134) or June through November with a peak in July (Forsberg et al. 1977:17, 34; Bottom and Forsberg 1978:44). In Nestucca Bay, juveniles occur from late spring through at least October (Nicholas and Hankin 1989:94), and they were also observed during hazing in early June 1996 (SPP 1996). Elsewhere along the Oregon coast, juvenile chinook can be present in estuaries during April through summer or fall with one or more peaks in May-August (Reimers et al. 1978:40-42, 1979:38-42; Mullen 1979:29-33, K. W. Myers 1980:38, 124-128; K. W. Myers and Horton 1982:385-388, Reimers and Downey 1982:11-13, J. A. Johnson et al. 1986:6, 8; Nicholas and Hankin 1989:11, 14-15, 18; Fisher and Pearcy 1990). There is some annual variation in the timing of peak abundance of juveniles in Oregon coast estuaries (Mullen 1979:29-33, Reimers and Downey 1982:11-13), and peaks occurred earlier in the upper than in the lower portion of an estuary (Reimers et al. 1978:18-21, 1979:38-42; Mullen 1979:29-33, K. W. Myers 1980:38, 124-128; Fisher and Pearcy 1990).
The length of residence of juvenile chinook in estuaries is variable (Nicholas and Hankin 1989:11, Emmett et al. 1991:162, Kostow 1995:18). Hatchery juvenile chinook were estimated to remain a few weeks to several months after a release in Yaquina Estuary (K. W. Myers 1980:69-70,139- 144), about 10 days for spring chinook and a month for fall chinook at Coos Bay (Fisher and Pearcy 1990), but more than half appeared to have left in a week after a release at the Salmon River (Mullen 1979:29). In contrast to Coos Bay, juvenile spring chinook at Yaquina Bay seemed to reside longer than some juvenile fall chinook (K. W. Myers 1980:70, 141).
Based on their seasonal occurrence, some or many juvenile hatchery or wild chinook may have been present during April-June hazing at Nehalem, Tillamook, and Nestucca Bays.
II-3c. AGE AT RETURN. Chinook males are 2-6 years and females 3-7 years old when they return; however, most males usually return at four years and most females at five years (Nicholas and Hankin 1989:27, 93, 97, 100, 119, 123, 138, 141, 156, 162). Jacks (which may be mature, Kostow 1995:18) return the calendar year following the migration of juvenile salmon into the ocean at two years of age (Nicholas and Hankin 1989:197). Part of the considerable yearly variation in age at return (Reimers and Downey 1982:1, 8-9; Nicholas and Hankin 1989) may reflect the bias of different methods of collection. Because of the variation between years and between sexes, hazing must occur for six consecutive years before it can be fairly certain that all returning adults may have been affected by hazing.
***************************************************************************
---------------------------------------------------------------------------TABLE 17. Number of coho and winter steelhead smolts (in thousands) reared and released at the North Fork of the Nehalem hatchery. 1975-1998 coho data were provided by Rick Klumph (ODFW, pers. comm.). In 1985- 1987, 15,000-25,000 coho smolts were also released elsewhere in the Nehalem River or its tributaries, and some may have also been done so prior to 1985 (John Leppink and Tracy Cabe, ODFW, pers. comm.). Steelhead data for 1984-1998 were provided by John Leppink and Tracy Cabe of the ODFW. See Table 3 for release dates into the Nehalem Basin. Occurrence of hazing at the Nehalem is from Table 2. -=data not requested. *=not enough data to calculate.
---------------------------------------------------------------------------
Thousands Thousands of
of Coho Winter
Smolts Steelhead
Released Smolts Released at...........
Smolt at North North Rest of
Release Known Fork Fork Nehalem Necanicum
Year Hazing Nehalem Nehalem Basin River
---------------------------------------------------------------
1975 No 230 - - -
1976 No 1288 - - -
1977 No 189 - - -
1978 No 954 - - -
1979 No 871 - - -
1980 No 53 - - -
1981 No 602 - - -
1982 No 540 - - -
1983 No 85 - - -
1984 No 722 67 128 43
1985 ? 439 54 113 36
1986 ? 568 51 107 0
1987 ? 568 52 116 44
1988 Yes 741 67 90 40
1989 Yes 805 64 90 40
1990 Yes 831 72 85 40
1991 Yes 736 76 63 40
1992 Yes 832 79 65 40
1993 Yes 760 77 48 31
1994 Yes 840 94 51 30
1995 No 790 80 7 40
1996 Yes 637 83 7 32
1997 Yes 629 78 15 35
1998 Yes 193 63 15 40
Nonhazing Mean 575 * * *
Range 53-1288 * * *
Hazing Mean 700 75 53 37
Range 193-840 63-94 7-90 30-40
---------------------------------------------------------------------------
***************************************************************************
It is not possible to directly test whether hazing is successful because there are no control groups. When a change in hatchery practices is tested to determine if it affects survival, there is a treated group and an untreated (control) group; these groups are reared as similarly as possible and released at about the same time and location, so that the only difference between them is the treatment. Such tests are not possible with cormorant hazing because the treatment (hazing) either occurs at an estuary or it does not.
One way to design treated and untreated groups is to assume that a smolt group released into a basin before hazing commences is a control group and a smolt group released during hazing is a treated group. This assumption is not valid for two reasons. First, smolts released prior to hazing may linger in an estuary until hazing occurs (Appendix II), so that they, as well as a treated group, are affected by hazing. Second, the seasonal timing of a release can greatly affect survival (e.g., see data in Appendices of Lewis 1994, 1997), and there has been a search for the optimal time to release smolts (e.g., Bilton et al. 1982, Parker and Stohr 1983, Martin and Wertheimer 1987, Gowan 1988, Irvine and Ward 1989, Mathews and Ishida 1989). For example, the seasonal variation in the survival of coho smolts released at the North Fork of the Nehalem hatchery prior to known hazing is shown in Table 18. Prior to hazing, there was an appreciable difference of 0.5% or more between early and late releases in 4 of 6 yr, and the difference was 1.0% or more in two of six yrs (Table 18). These differences were not just a result of larger smolts surviving better because in some years they did not (Table 18). Thus, differences in survival between smolt groups released before or during hazing may reflect seasonal factors unrelated to hazing.
The converse way of creating treated and untreated groups at a basin is to assume that smolts released early in a year during hazing are a treatment group and smolts released after hazing has ceased are a control group. This is also not valid--smolts released during hazing may linger until after hazing has ceased and any differences in survival between the two groups may reflect seasonal variation in survival that is not associated with hazing.
Control and treatment groups can be formed by assigning years without hazing as control years and years with hazing at the same basin as treatment years. Then, the assumption is that the only variable affecting survival between control and treatment years is the occurrence of hazing. However, this assumption is not robust because there is substantial annual variation in catches, returns, or survival of salmon and steelhead in a basin (e.g., ODFW 1987, 1998b; Jacobs and Cooney 1997, Lewis 1997, Chilcote 1998), whether hazing has occurred or not. This variation may be a consequence of factors such as variable ocean conditions or fisheries regulations (see Appendix V).
Another way of creating control and treated groups is to designate basins without hazing as control groups and basins with hazing as treatment groups. Here, the assumption is that the only difference between basins is the occurrence of hazing. This assumption is not robust because there can be substantial differences among basins in trends of returns (Jacobs and Cooney 1997, Chilcote 1998:25-26, 43, 45), survival (Lewis 1997:5-6), and catches (ODFW 1987, 1998b), whether hazing occurs or not.
--------------------------------------------------------------------------TABLE 18. Within-year variation in survival rates of coho smolts released at the North Fork of the Nehalem hatchery prior to known hazing. These data are from CWT data in Lewis (1994:19-21, 33-34) and do not include coho smolts reared at this hatchery and released into Fishhawk Creek. Two stocks are reared at this hatchery: 32=Nehalem stock reared at the hatchery for years, 99=stock from Fish Hawk Lake on Nehalem River (John Leppink, ODFW, pers. comm.). Only smolts from the same stock that were released at different times during a year are included, so that a possible difference in survival as a result of stock origin is not a factor. Weight is in grams/fish; Surv. Rate=survival rate as a percentage of smolts released that are recovered.
---------------------------------------------------------------------------
Av. Surv.
Brood Release Weight Rate
Year Stock Date (g) (%)
---------------------------------------
1977 32 3/15/79 23 1.2
" 32 5/1/79 24 0.5
1978 32 3/15/80 30 1.0
" 32 5/1/80 32 0.2
1979 99 3/15/81 30 0.6
" 99 5/1/81 32 0.7
1980 99 3/15/82 31 0.6
" 99 5/1/82 32 1.8
1981 99 3/15/83 47 1.6
" 99 5/1/83 41 2.9
1982 99 3/15/84 41 1.1
" 99 5/1/84 42 1.3
---------------------------------------------------------------------------
***************************************************************************
Proponents have claimed that hazing has improved salmonid returns (Erickson 1989a,b,d; 1992, 1993, 1995a,b; Monroe 1995b, 1996a; Nokes 1995) and the Oregon Legislature has subsequently acted to permit and fund hazing (section C-1). However, other factors influence salmonid returns besides hazing. Acknowledging these factors is important for two reasons. First, separating out their effects from those of hazing is generally not possible. Second, these other factors may result in the mortality of those smolts that are saved by hazing, so that the end result of hazing may not be as beneficial as it might first seem. Some of these factors are examined below.
Ocean conditions in recent years have been unfavorable for juvenile salmon growth or survival, and some researchers regard ocean conditions as the most important factor in reduced salmon abundance (Nickelson 1986, Pearcy 1992, 1997; Botkin et al. 1995:122-128, National Research Council 1996:39-45, Beamish et al. 1997, Emmett and Schiewe 1997, Kaczynski 1998).
One would expect more fish to return if more smolts emigrate to estuaries or the ocean, although this has not been true for hatchery- reared coho salmon (e.g., Pearcy 1992:48). The number of hatchery coho and steelhead smolts has varied at the Nehalem (Appendix III), and the number of wild salmonids produced probably also varies considerably. For example, there has been a decline in the number of spawning adult coho, steelhead, and chinook (sections G-3, G-4, and G-6), so one would expect that the total number of resulting smolts would also decline because fewer eggs would have been produced.
Because of the variation in smolt production, it would be best if the number of smolts was known, so that returns could be measured in terms of percent smolt survival. However, such data are not available for wild salmonids.
Some studies have found a positive correlation between fisheries and stream flows during freshwater rearing of wild salmonids, but other studies have not (Knight 1980:60-63, Hall and Knight 1981:5-9, Scarnecchia 1981). It is clearer that flooding during rearing can have a negative impact on survival (Weber and Knispel 1977:33, National Research Council 1996:186-188); for example, the 1996 flood along the Oregon Coast washed eggs away (B. Buckman in Gallob 1999).
Both hatchery and wild fish may be affected by flows or turbidity at the time of outmigration. Weber and Knight (1977:33-34, 84) found that hatchery steelhead returns at the Nehalem were better when there were high flows at the time of release, perhaps because high flows reduced susceptibility to Ceratomyxa shasta infection (see below). Also at the Nehalem, Stahl et al. (2000:31) noted that hatchery coho smolt migration rates were greater with increased flows at the time of release. Similarly, Hvidsten and Hansen (1988) observed increased returns with high flows during release of Atlantic salmon smolts, and Gregory and Levings (1998) found that high turbidity (which can be associated with high flows) during smolt migration also reduced predation of juvenile salmon by piscivorous fish.
Hatchery fish can be impaired after release for several reasons, and thus they may be more vulnerable to mortality (Mesa et al. 1994). First, they have been stressed, disoriented, and forced into an unfamiliar environment; these stressors can be compounded by the additional stress of trying to adapt to seawater (Wedemeyer 1980, Schreck 1990:31-32). Second, hatchery smolts are easily detectable to predators because some behave inappropriately in their new environment. Since they feed on pellets spread on the water surface, many hatchery smolts come to the surface to feed shortly after release and in so doing are easily seen by potential predators; they also often jump out of the water or roll, exposing their highly conspicuous silver sides, which makes it very easy for predators to find them (Bayer 1986). Third, hatchery smolts are vulnerable to predation because they are not wary of predators (e.g., Bayer 1989:61-63; Suboski and Templeton 1989, Olla et al. 1998).
Post-release mortality of juvenile salmonids can result directly from diseases or parasites (e.g., Noga 1996). In the Nehalem Basin, both Ceratomyxa shasta and Nanophyetus salmincola occur and can cause mortality in salmonids (Weber and Knispel 1977, Wade 1986:12-13). In Washington, it has been suggested that Nanophyetus infection plus other stressors resulted in the death of coho smolts (Schroder and Fresh 1992: 207-208, 262-263). Further, chemical contaminants can make juvenile salmonids more susceptible to disease (Arkoosh et al. 1998). Together, or separately, these factors may also increase mortality by impairing smolts, so that they are more vulnerable to predation (Mesa et al. 1994).
V-7a. PREDATORS OF JUVENILE SALMONIDS. Although cormorants are the predator of most interest in Tillamook County, Caspian terns are of more concern in the Columbia River (Roby et al. 1998), and common murres have been suggested to have had a major impact at Yaquina and Coos Bays (Varoujean in Ward 1983, Matthews 1983, Bayer 1986, Emmett 1997:152). In the ocean near the mouth of the Nestucca, hazers reported that "birds" and seals were feeding on smolts (section F-10), and, in the Pacific Northwest, several species of birds can prey on juvenile salmonids in the ocean (Bayer 1989:36-38, Fresh 1997).
Fish can also be significant predators. Limited studies in Oregon nearshore areas indicate that various subadult or adult salmonids (Angstrom and Reimers 1964, Fresh et al. 1981:17, 27, 35; Stuart and Buckman 1985, Brodeur et al. 1987:9) and black rockfish (Brodeur et al. 1987:9) prey on juvenile salmonids. Elsewhere in the Pacific Northwest, significant predation of juvenile salmonids has been reported by spiny dogfish, Pacific mackerel, Pacific whiting, and lamprey (Beamish et al. 1992, Beamish and Neville 1995, National Research Council 1996:40, Pearcy 1997:341).
Predation in the ocean is important to acknowledge in assessing hazing because juvenile salmonids saved by hazing in estuaries may be taken by nearshore predators (Bayer 1989:36).
V-7b. PREDATORS OF JACK OR ADULT SALMONIDS. Marine mammals are predators of adult salmonids, although the significance of this predation is controversial (Kaczynski and Palmisano 1993, Botkin et al. 1995:132-141, Emmett 1997, Fresh 1997). Bald eagles and osprey can also prey on adult salmonids in estuaries (Emmett 1997), but there are so few of them along the Oregon Coast that they probably have little impact. This predation would reduce numbers of returning salmonids, and thus complicate analyzes of hazing on adult returns.
V-8a. INTRODUCTION. Fishing regulations alter the length of season and number and type of fish (wild or hatchery) caught; they have become more restrictive as wild coho and steelhead populations have become of increasing concern. The goal of these regulations is to increase the number of wild fish returning to spawning areas, but they could also increase the number of wild or hatchery fish returning to hatcheries and the number caught in freshwater fisheries.
V-8b. OCEAN FISHERIES. In an attempt to improve the escapement of wild coho to freshwater, regulations reducing ocean commercial and sports catches of both coho and chinook (e.g., PFMC 1999:I-18, I-19) have decreased the ocean exploitation rate of coho (Fig. 22). In 1994, ocean coho fisheries were limited to incidental catches during the ocean chinook fishery (Kostow 1995:84, 97). For the nonhazing years of 1975- 1985, the average ocean exploitation rate of coho was generally 70% or more (Fig. 22) and averaged 68% (range 27-88%), but the average for the hazing-affected years at the Nehalem of 1989-1997 was 30% (range 2- 62%)(calculations from PFMC 1999:I-20).
The number of fish caught depends upon how much effort is spent trying to catch them. Ocean fishing effort has declined greatly in Oregon; for example, ocean salmon commercial troll effort has declined from about 25,000-40,000 days fished annually in 1979-1990 to 4,000-9,000 in 1992-1998, and the ocean salmon recreational effort has declined from about 225,000 angler trips yearly in 1981-1990 to 25,000-45,000 in 1994- 1998 (PFMC 1999:I-8, I-10). Most of this decline can be attributed to the increasingly restrictive fishing regulations, but ocean fishing effort can also depend upon weather and, for commercial fishers, the price of salmon and the profitability of the salmon fishery.
---------------------------------------------------------------------------FIGURE 22. Coho Oregon Production Index (OPI) Ocean Exploitation Rate Index. The OPI extends from Leadbetter Point, Washington to the U.S./Mexican border. Data are from PFMC (1999:I-20); 1998 data are preliminary.
Coho: OPI Ocean Exploitation Rate Index (%)
P 100-|
e -| X X
r 80-| X X X X X X X X X X
c -|---X-X-X-X-X-X-X-X-X-X-X---X------------------------------- 70%
e 60-| X X X X X X X X X X X X X X X X
n -|-X-X-X-X-X-X-X-X-X-X-X-X-X-X-------X-X-X-X----------------- 50%
t 40-| X X X X X X X X X X X X X X X X X X X X X
a -|-X-X-X-X-X-X-X-X-X-X-X-X-X-X-X-X-X-X-X-X-X-X-X-X----------- 30%
g 20-| X X X X X X X X X X X X X X X X X X X X X X X X X
e -|-X-X-X-X-X-X-X-X-X-X-X-X-X-X-X-X-X-X-X-X-X-X-X-X---X-X-X-X- 10%
1-4-| X X X X X X X X X X X X X X X X X X X X X X X X X X X X X
0-| X X X X X X X X X X X X X X X X X X X X X X X X X X X X X
|__________________________________________________________
| | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| | | | | |
70 75 80 85 90 95
Year
---------------------------------------------------------------------------
V-8c. ESTUARINE AND FRESHWATER FISHERIES. These fisheries have also been restricted in recent years to increase the escapement of wild salmon and steelhead to spawning areas. In 1994-1996, estuary and freshwater sports coho fisheries were discontinued in most coastal streams; however, fisheries continued in the North Fork of the Nehalem River (but not the rest of the Nehalem Basin), North Umpqua River, Umpqua River and Bay (1995 only), Coos River and Bay, Tenmile Creek and Lakes (1994 only), and Rogue River (ODFW 1998b). Starting in 1992, recreational fishery regulations along the Oregon coast required the release of all wild steelhead (Chilcote 1998:31).
Catches also depend on fishing effort, and effort has been variable in estuarine and freshwater fisheries (sections J-1a and J-2a).
Another challenge in testing returns for the effects of hazing is straying, whereby salmon or steelhead reared or released in one basin return as adults to a different basin. Although some straying is a natural process, the number of hatchery-reared coho (Jacobs 1988, ODFW 1997a:62, 1997b:43, 1997c:57; Quinn 1997) or hatchery-reared winter steelhead (Lindsey et al. 1995, Chilcote 1998:31-36) that stray can be significant. For instance, strays from Oregon Aqua-Foods coho releases increased the number of coho in streams from the Salmon River to Yachats (section J-1a). If there is appreciable straying, then it cannot be robustly assumed that all salmon or steelhead returning to a basin with hazing were released there and affected by hazing.
***************************************************************************
Adams, N. S., D. W. Rondorf, S. D. Evans, J. E. Kelly, and R. W. Perry.
1998. Effects of surgically and gastrically implanted radio
transmitters on swimming performance and predator avoidance of
juvenile chinook (Oncorhynchus tshawytscha). Can. J. Fish. Aquat.
Sci. 55:781-787.
Angstrom, R. L. and P. E. Reimers. 1964. Occurrence of juvenile salmon in
stomachs of adult coho salmon. Oregon Fish Commission, Research
Briefs 10(1):69.
Arkoosh, M. R., E. Casillas, E. Clemons, A. N. Kagley, R. Olson, P. Reno,
and J. E. Stein. 1998. Effect of pollution on fish diseases:
potential impacts on salmonid populations. J. Aquatic Animal Health
10:182-190.
Bayer, R. D. 1986. Seabirds near an Oregon estuarine salmon hatchery in
1982 and during the 1983 El Nino. Fishery Bulletin 84:279-286.
Bayer, R. D. 1989. The cormorant/fisherman conflict in Tillamook County,
Oregon. Studies in Oregon Ornithology No. 6. (It is planned to be
available at: http://www.orednet.org/~rbayer/salmon/s6_cnt.htm [all
lower case letters]).
Bayer, R. D. 1996. Censuses of black brant at Yaquina Estuary, Lincoln
County, Oregon. Journal of Oregon Ornithology 6:723-780. (See
http://www.orednet.org/~rbayer/j/j723.htm [all lower case
letters]).
Beamish, R. J. and C. M. Neville. 1995. Pacific salmon and Pacific
herring mortalities in the Fraser River plume caused by river lamprey
(Lampetra ayresi). Can. J. Fish. Aquat. Sci. 52:644-650.
Beamish, R. J., B. L. Thomson, and G. A. McFarlane. 1992. Spiny dogfish
predation on chinook and coho salmon and the potential effects on
hatchery-produced salmon. Trans. Am. Fish. Soc. 121:444-455.
Beamish, R. J., C. Mahnken, and C. M. Neville. 1997. Hatchery and wild
production of Pacific salmon in relation to large-scale, natural
shifts in the productivity of the marine environment. ICES Journal of
Marine Science 54:1200-1215.
Berry, R. 1995. [27 March 1995 oral testimony by the ODFW about Senate
Bill 707 to Oregon Senate Committee on Agriculture, Natural Resource
and Environment.] This is available on audiotape 59A at the Oregon
State Archives, 800 Summer St. NE, Salem, Oregon 97310.
Bilton, H. T., D. F. Alderdice, and J. T. Schnute. 1982. Influence of
time and size at release of juvenile coho salmon (Oncorhynchus
kisutch) on returns at maturity. Can. J. Fish. Aquat. Sci.
39:426-447.
Botkin, D., K. Cummins, T. Dunne, H. Regier, M. Sobel, L. Talbot, and
L. Simpson. 1995. Status and future of salmon of western Oregon and
northern California: findings and options. Center for the Study of
the Environment, Santa Barbara, California. Report No. 8. (This
study was financed in part by Oregon Senate Bill 1125 [1991]).
Bottom, D. and B. Forsberg. 1978. The fishes of Tillamook Bay. Oregon
Dept. of Fish and Wildlife, Research and Development Section, Federal
Aid Progress Reports Fisheries, F-100-R.
Bowmaker, A. P. 1963. Cormorant predation on two central African lakes.
Ostrich 34:2-26.
Brodeur, R. D., H. V. Lorz, and W. G. Pearcy. 1987. Food habits and
dietary variability of pelagic nekton off Oregon and Washington, 1979-
1984. U.S. Dept. of Commerce, National Oceanic and Atmospheric
Administration, National Marine Fisheries Service, NOAA Tech. Report
NMFS 57.
Busby, P. J., T. C. Wainwright, G. J. Bryant, L. J. Lierheimer, R. S.
Waples, F. W. Waknitz, and I. V. Lagomarsino. 1996. Status review of
West Coast steelhead from Washington, Idaho, Oregon, and California.
U. S. Dept. Commerce, National Oceanic and Atmospheric Administration,
National Marine Fisheries Service, NOAA Tech. Memo. NMFS-NWFSC-27.
(This is at http://www.nwfsc.noaa.gov/pubs/tm/tm27/tm27.htm)
Chapman, D. W. 1958. Studies on the life history of Alsea River
steelhead. J. Wildl. Mgmt. 22:123-134.
Chapman, D. W. 1961. Factors determining production of coho salmon,
Oncorhynchus kisutch, in three Oregon streams. Ph.D. Thesis. Oregon
State Univ., Corvallis.
Chilcote, M. W. 1998. Conservation status of steelhead in Oregon. Oregon
Dept. of Fish and Wildlife, Fish Division, Information Report 98-3.
Draulans, D. 1987. The effectiveness of attempts to reduce predation by
fish-eating birds: a review. Biological Conservation 41:219-232.
Ellis, R. 1998. Biological resources. Chap. 3 in Tillamook Bay National
Estuary Project "Tillamook Bay Environmental characterization: a
scientific and technical summary." Prepared under Cooperative
Agreement with the U.S. Environmental Protection Agency.
Emmett, R. L. 1997. Estuarine survival of salmonids: the importance of
interspecific and intraspecific predation and competition. P. 147-158
in R. L. Emmett and M. H. Schiewe (eds.), Estuarine and ocean survival
of Northeastern Pacific salmon: proceedings of the workshop. U. S.
Dept. Commerce, National Oceanic and Atmospheric Administration,
National Marine Fisheries Service, NOAA Tech. Memo. NMFS-NWFSC-29.
(This is at http://www.nwfsc.noaa.gov/pubs/tm/tm29/Papers/Emmett.htm)
Emmett, R. L., S. L. Stone, S. A. Hinton, and M. E. Monaco. 1991.
Distribution and abundance of fishes and invertebrates in west coast
estuaries, Vol. II: species life history summaries. U.S. Dept.
Commerce, National Oceanic and Atmospheric Administration, ELMR Rep.
No. 8, NOAA/NOS Strategic Environmental Assessments Division.
Emmett, R. L and M. H. Schiewe (eds.). 1997. Estuarine and ocean survival
of Northeastern Pacific salmon: proceedings of the workshop. U. S.
Dept. Commerce, National Oceanic and Atmospheric Administration,
National Marine Fisheries Service, NOAA Tech. Memo. NMFS-NWFSC-29.
(This is at http://www.nwfsc.noaa.gov/pubs/tm/tm29/index.html).
Erickson, J. 1988. Letter to Editor: More than $30,000 a day. P. A-4 in
May 11 Headlight-Herald (newspaper). Tillamook, Oregon.
Erickson, J. 1989a. Letter to Editor: borderline insanity. P. A-5 in
March 1 Headlight-Herald (newspaper). Tillamook, Oregon.
Erickson, J. 1989b. Letter to Editor: cormorant harassment. In March 6
Daily Astorian (newspaper). Astoria, Oregon.
Erickson, J. 1989c. [13 April 1989 oral testimony about House Bill 3185
to Oregon House Agriculture, Forestry and Natural Resources
Committee.] This is available on audiotapes 73B and 74B at the Oregon
State Archives, 800 Summer St. NE, Salem, Oregon 97310. (Also see
Bayer 1989:24, section 2-E-1 for a quotation from Erickson's
testimony.)
Erickson, J. 1989d. [13 April 1989 written testimony about House Bill
3185 to Oregon House Agriculture, Forestry and Natural Resources
Committee.] Copies are available at the Oregon State Archives,
800 Summer St. NE, Salem, Oregon 97310. (This is reprinted in Bayer
1989:82-83.)
Erickson, J. 1989e. [24 April 1989 letter petitioning cormorant
harassment permits from the Oregon Fish and Wildlife Commission.]
(This is reprinted in Bayer 1989:81.)
Erickson, J. 1992. Letters and Issues: What makes a river successful?
Northwest Steelheader 7(4):10-11.
Erickson, J. 1993. Cormorant control: chasing the birds away on the
Nehalem Estuary. Northwest Steelheader 8(1):18-19.
Erickson, J. 1995a. Letter to Editor: Emotional. P. A-4 in May 10,
Headlight-Herald (newspaper). Tillamook, Oregon.
Erickson, J. 1995b. [27 March 1995 oral testimony about Senate Bill 707
to Oregon Senate Committee on Agriculture, Natural Resource and
Environment.] This is available on audiotapes 58A and 59A at the
Oregon State Archives, 800 Summer St. NE, Salem, Oregon 97310.
Fisher, J. P. and W. G. Pearcy. 1990. Distribution and residence times of
juvenile fall and spring chinook salmon in Coos Bay, Oregon. Fishery
Bull. 88:51-58.
Forsberg, B. O., J. A. Johnson, and S. M. Klug. 1977. Identification,
distribution, and notes on food habits of fish and shellfish in
Tillamook Bay, Oregon. Oregon Dept. of Fish and Wildlife, Research
Section, Federal Aid Progress Reports Fisheries.
Fresh, K. L. 1997. The role of competition and predation in the decline
of Pacific salmon and steelhead. P. 245-275 in D. J. Stouder,
P. A. Bisson, and R. J. Naiman (eds.), Pacific salmon & their
ecosystems. Chapman & Hall, New York.
Fresh, K. L., R. D. Cardwell, and R. R. Koons. 1981. Food habits of
Pacific salmon baitfish, and their potential competitors and predators
in the marine waters of Washington, Au