260 
Fishery Bulletin 117(3) 
22 
21 
_ 20 
9, 19 
CD 18 
I 17- 
S. 16 - 
I 15 - 
1“ 
14 -I 
13 
12 
Sockeye 
Chinook 
>. c 
CT3 ^ 
§ 1 
O) O) O) CL O. Q. ft 
D “J D £)) ti) 0 fX 
^ ' •» - ■ ' <<<C/)COC/)V 
CO CO co ' 
<r0C0 ^ T-WWWWl-T-Y-^ 
T_ 1- CM 1- CM 
r 3.5% 
- 3.0% 
2.5% 
2 . 0 % 
1.5% 
1 . 0 % 
0.5% 
0 . 0 % 
o R 
O O 
CNJ CO 
120 
100 
o 
2.0 80 
0 CM 
60 
A 
cd 
3 
2 
<55 
E 40 
CD 
20 
B 
O CD 
,o O 
OOO 
O o ° 
o 
° o° 
cP 
o 
o 
o 
o 
ePo 
%n° 
o O 
° o cP 
o o 
0 4—63“ 
o 
c 
o 
■c 
o 
a. 
o 
1960 1970 1980 1990 2000 2010 2020 
Figure 2 
(A) Migration timing of sockeye (Oncorhynclius nerka ) and Chinook 
(O. tshawytscha) salmon at the locks in the Lake Washington water¬ 
shed in Washington (black lines) and mean lock temperature (gray 
line) during May-October in 2005 and 2006. The timing of migration 
(black lines) is presented as the proportion of the time of the run fish 
spent at various temperatures. (B) Number of days, by year, between 
the first and last days when the epilimnetic temperatures exceeded 
20°C in Lake Washington. Data were provided by C. DeGasperi, 
King County Department of Natural Resources and Parks, Seattle, 
Washington. 
locks to the lake) with a mean depth of 9 m and includes 
shallow Lake Union (mean depth: 10.5 m; maximum: 
16 m). Lake Washington has a surface area of 87.6 km 2 , a 
mean depth of 32.9 m (maximum: 65.2 m), and a summer 
epilimnion to about 10 m. Lake Washington’s 2 largest 
tributaries are the Cedar River and the Sammamish 
River, which drains Lake Sammamish (surface area: 
21 km 2 ; mean depth: 18 m). 
The locks include a tidally influenced fish ladder, small 
(8 m deep) and large (16 m deep) lock chambers, a spill¬ 
way bay (5 m deep), and a drain above the large lock to 
return salt water to Puget Sound. The large lock is used 
by migrating salmon and is included in the area defined 
as the lower estuary. The locks physically separate the 
freshwater system, without any tidal elevation changes, 
from the marine system where tidal amplitudes reach 4 m. 
The exchange of fresh and salt water occurs pre¬ 
dominantly in localized pulses during locking 
events, resulting in unusual circulation patterns 
within the estuarine portion of the ship canal. 
Salt water enters when the locks lift vessels from 
Puget Sound up and into the ship canal; there¬ 
fore, temperatures in the large lock chamber can 
change up to 9°C in less than 1 h. Salt water 
intrusion in spring and early summer is usually 
restricted to the lower 1.5 km of the ship canal, 
but by late summer it reaches Lake Union (4-6 
km upstream), although at lower concentrations 
(salinity: 0-2) than near the locks (salinity up to 
15). Vertical stratification in temperature and 
salinity occurs in Puget Sound, below the locks, 
within the large lock, in the saltwater wedge in 
Salmon Bay, in the ship channel (salinity only), 
and in Lake Union, and temperature stratifica¬ 
tion occurs in Lake Washington and Lake Sam¬ 
mamish. Low DO levels (<5 mg/L) occur near the 
bottom from the upper estuary starting in August 
until the end of September. 
Fish capture and tagging 
In 2005 and 2006, the Washington Department 
of Fish and Wildlife (WDFW) and R2 Resource 
Consultants 1 (R2, Redmond, WA) captured 368 
Chinook salmon by using a purse seine in the 
lower half of the large lock from early August 
through early September, tagged, and immedi¬ 
ately released them below the locks in the lower 
estuary. This net was designed for sampling and 
safely releasing juvenile salmonids (Durkin and 
Park, 1967; Ledgerwood et al., 1991), and only 
fish in good condition were used for tagging, in an 
effort to minimize the effects of handling on fish 
survival and behavior. Catch and release, from 
commercial, scientific, and recreational fisher¬ 
ies, can affect survival, physiology, and behavior 
of salmon, but the effects are context dependent 
(Candy et al., 1996; Donaldson et al., 2011; Raby 
et al., 2015). Given the nature of our study sys¬ 
tem, there was no way to know how salmon would have 
behaved had we not caught and tagged them, nor did it 
allow us to compare survival rates among salmon with dif¬ 
ferent types of tags. However, our goal was not to assess 
survival, it was to examine thermoregulatory behavior, for 
which our approach was standard and appropriate. 
The Chinook salmon had both archival temperature data 
loggers, implanted by using tagging techniques similar to 
those in Newell and Quinn (2005), and passive integrated 
transponder (PIT) tags; approximately 25% also received 
acoustic tags (Table 1). These types of tags yield different 
but complementary kinds of information. The fish ladder 
1 Mention of trade names or commercial companies is for identifi¬ 
cation purposes only and does not imply endorsement from the 
National Marine Fisheries Service, NOAA. 
