HomeMy WebLinkAboutSWP273046(5)MAR-04-03 TUE 04:33 PM USACE SEATTLE DISTRICT FAX NO, 206 764 4470 P. 01
ILI I I M.11 I I
US Army Carps
N
of Engineers
Soattle District
l'u. rjax 371.1�i
Sn;At11A.WA 99771.3755
47US East Murgir el Wa) Sam{
Snottle,WA %ij.1-230i
016) 7G.1-3652
r-wo: pool 1,girLim, juh f)
rrmy roll
I lamo! (20r) 70 1.11391
Fax: t206) 7C 1-4170
MAR-04-03 TUE 04:34 PM USACE SEATTLE DISTRICT FAX NO, 206 764 4470 P, 02
C;'E"Nws-E?C-I)TI-CS (1110-2-1150a)
Mf.?I`1()I:1�1�Tnt7M rC)R Chief, CENWS-PM-PL
ATTN: N. Gil.brough
26 February 2003
:'Ui3JECT, Ceci,-L River. Side C:hanne'l, Renton, Washington
1. A field grip to the above referenced project was conducted
c?r) 18 P'ebcuary 2o03 to investigate the subsurface soils. Three
I-ackhoe test, pi -Ls (03-I:H--01, 02, and 03) were dug to depths of
(;.5 to '1.5 feet: (see enc.l.. 1, sheets 1 and 2) using a Case 560
backhoe owned and operated by City of Renton. The materials
oncourltured in trig upstream most test pit (03 -EH- O1) consist of
,c��proximatel.y 6 i.nclies of silly sand with organic content
ovca-r.'I ying :iandy gravels with cobbles to 6 inch diameter. The
l.i����czr four f'ect of this material had some roots and organic
debris (de(:rlyiclq logs) interspersed in the formation. Static
grroundw,atcr was encountered at a depth of six feet and the pit
wi.,tt; terminated at seven feet due to caving. The materials
E:>•rc,ountcrc:d in r.he middle Lest Pit (03-BH-02) were similar to
Llie first. tcc.t; pit; except that the organic debris was
ezGc.;()j.ant_cre,d only in the upper 2 feet of exploration. Static
.r.-ou.ndwatc;r was again encountered at a depth of 6 feet and the
pit. w,.,s t..prininated at %.5 feet due to severe caving and an
:inability t,,) make much progress with depth. The downstream most
piL (03-BIi-o3) also had about the same formation as the
,,E.c.oI1cl. pit with the exception that a lense of what appeared to
br_ vE?ry clayey gravel was cricountered at about 3.S to
Lc;et. ()ncc this ].ease was penetrated, groundwater started
t;o Percolate into the excavation indicating Borne degree of
c:onf infinent, by Lhe clayey soil. The water level was measured at
U depth of f011r. feet after reaching a static state and the pit
okra S terminarc:d at. 6.5 feet; in severely caving soils. The soils
}�ra�rzesgivcly less stable due to lack of finer. grained
materl.r�.1 zis exploration progressed downstream.
Ga•.*ouridwatr r and river levels were surveyed in at the
loci 1t..i.o-Is of these test pits and the groundwater levels were
fc)uria to be consistently about '4. foot lower than the adjacent
z .i. ver level .
;j. .t3aried can the findings of this field investigation, it can be
asr.;ume(I t.hc t a channel placed at this site will be influenced by
))nth the river level and groundwater migration from up,Lope.
rllw, proposc!d channel will also rapidly run dry when both
c.I.roundwat Pr and river levels drop below the channel invert.
C:()nstzuctic.m of this side channel will require very flat
3icl:�^1o�>e;5 at least 3 horizontal (H) on 1 vertical M to
MAR-04-03 TUE 04:34 PM USACE SEATTLE DISTRICT FAX NO. 206 764 4470 P. 03
maintain stable side slopes and channel alignment. We
«cicli?:Y'Eli: �rld that some sloughitlg and channel migration will be
�ar.c�pt_ablt: fox, this project and therefore recommend that the
"lc,pes be no stceper than 2 I -I on 1 V. It is recommended that
of t'.he channel where it reconnects with the Cedar
be, t-_ransitioned Lo 31I on 1V.
4. excavated for the proposed channel. are
;;sti•til ��blc by clef i nition for structural fill if they are re -worked
by mixing the: c:l.eaner and coarser downstream materials with the
i CA'c 1111ifornily graded upstream soils and the surfieial silti.er
oJ1s The cleaner downstream gravels can be used in confined
slicii rFis trenches or. depressions for fill and for drains.
'lilEs majoriti, of material encountered consists of predominately
clea-n sandy gravel. tilat. is highly permeable and will not lend
I:o a 11.ig1i de>.greo of compaction, therefore, heavy loading
and t•c-)ad corlsttLiction are not recommended on these soils unless
t;llcy are rc-worked with suitable fine grade soils.
5. If YOU lhtrvc-' ally questions, please call M. Kaiser at ext .
6194.
lll. S iFISC ER, P. '.
Chief., Civi.]. Soils Section
CF.
M-!,NW;z,-PM--PL (N. Gi•lbrough)
C1J"sNWS-I.C-DI3-CS (Kaiser)
MAR-04-03 TUE 04:35 PM USACE SEATTLE DISTRICT FAX NO, 206 764 4470 P, 04
�,�,�'� •` .'-�.,`. `.\ ^tip.
F�1.7Q9,950 �•�
•. - ``_``
\` X
o s- at-j-oz
tj
t�1�7I O.:SO
V
C(CAYITE 7
srz r--V/'Z 7E;5r
------------
E 1.7t o,loo
Z 16L , l
.5,477. 1
MAR-04-03 TUE 04:35 PM
•1
ij
USACE SEATTLE DISTRICT FAX NO, 206 764 4470
")-;- r 4- c)3
e-4
P. 05
Al
7>�j
I j
406
_��77C 69, �Ulll L.,!5),115L
w V
F—� e— reAE' 7P�
6441JAL-1--
W
N
3<
El ❑ ❑
CA
Cn
Oz CD
D D z
_
< m 0
O
cn
(A
CD
3 Cn
(n
(D
(D
:3 3
�.
:3
(D
CL
W
o L
C W
N
N p
N (n
0.
CD -O
W
O
S
fv ?
5
CD
S CD 0)o
7 (D
=' Cn
O
�cn = cn
EF
O3
�' ((DD CD
�.
o 3
O O
0 X C2 CD
cn
(DO
COl1 O
m<
CD 7 CD.
.-' C j
O<
(D
sZ
O N 0
7
0 C7
C
== CL
N O
7
C)
CD
-
- CD
N N
00 (n
^' CD
'S -«
0 ITl c
o
v
�'a
•)F)� w
O Q
-.
O _r
3
N N CD
Cu
< to cQ
(D S N
(n'
_
N -i
(D = -
s? 0 (n
=r =
(�D
O
N !Z
O
CD
OO
N
CD
�• (D =
�• p=
0
O
1 (n
a)
� Cn n
<-
C c
3 3
O
oo�
3vsv
(Tom
``�cn
v�
G�� -
_C.-0(D
0
—
El El
(D CD
vi
v
<' (D �
n
Cr
^
moo
�(DCD
LO
�_�
v
O
9 o
• (D
co-p
=
::3
CD
O
T� (a=
(D
(n Nam_
3 hn :'
CD
N
9
m
El El El
O
fn
�-' 3
n.
CL
C)
3 (D
OL
N O
0
,(On
N �
0'
=
0)
Cn
-n -v C)
(D 7
3
(D
<
O N
_. (D
.<
S C:
3 (T
N
O
-a
ty'
/�
VJ
o
3 (D
C =
x
CD
= (n
O (n —
cn 0
cn
Q
O
Cn
N O
O 3
CD
Cn
O CD
(3 C)
=
c!a
a
(n n D
Z3<
O
N (p
=
O
-n
O
n 3 c
o
v
Q
lD
m (n (n
�sv o
O
o
0
C
m a)D
CCD
O
S
(D
(D
�,
-,
(D.:;
CD
O
D
O O m
3 3
v
3 3
CD CD CD
(D
_
O
O
X a D -n3
o
m
'a(3D
5' :E a m
z
C)
6
m
<
O w CZg-0 w wow Jozr
3 5N v n
(D (D v
za)
O
3 5C -0
Q O
z
O
c
(3
O COED sZ (D n a) n Z) (D (D (n
O O In c 0 _ < 0
3
((DD
v N 3
N (D 0 _ (n - j
sZ
CL
(n CD � cfl � (�
3
Q
O
(�
C L N p� c = fD N CD - -
O -� X CD O -6 <n �« = (D
C) a (n Cp O '< (D
CD Cz = (D (3 —
—
(v (a S
(n (n 7' Cv
�
(D
.. p
C7 m
O CD 3 cn (v ((DD �' Q �•
O
in N O cn<
(D
S g O Q CL
<
O
O
Cr
�n_ cn3(D
_
cc=
—
E0
CD @
�ZCD3o
�
�3-N
�o
x O
_a)
(n CD Q �n
t= (D =,, Cr (D N
S
n' (D (a
(n O o-
a ,
°
1lv(D
3
3
(D — n < CD C) (D
(D
0
:3 (D
CD
(fl (D
C)
CD m
< N N N cr
(D (D
O
Ci =
CD
(D (n
CxD Cr (D
= (n
In
D
`C
(D
� w
(D
D n 77
n �
O
(D
00
cp
-0 Cl)
-a C/)
C/)
CD
cn
(n
(Q
(°CD (D
(fl CD
0
(n N 0
CD N
(n 0
(D 0
(�
CDp
CD n
m 0
m n
m
m p� O
O
O
CD
m
m =3
m 3
=3
m W
.. 7
�-D W
.+
w
CD W
0 W
0 W
0 W
N W
0-0 N
0=T N
5�
0 S fV
j
o�
W
W
W
W
N
m N
m N
m
O
51
rn
M.
v v m 3< p
(n c� < O m
c =3CL-0-°-
(n O
n
c O
-o m
3=
O o
cn
�
�,
��
° °-
. 2 z
CD lD
C) m c
O � 0
N 6
0
(Q
m
(n 2. in
m N (A (D
m n 7
m
_
_Y
v (� N 3
7 -� (D 7 (D
7<
m e
N
� m w CD
< m
O N CD �
=� m
'w
N
w
Q OZ3 ° �.
O O
cr
Q z o 7-
0
O
,< 6 D -� SEY
(D
0
-
cfl
C cl 0- m
O
Q 0 �. �
3
� (D
C O O— C
(n
O Q
3 O
�, O O
n. cn
C
O
0 (Q Q
CT
v
O 3
Q cn
'< -
(fl 3 O (n (n cD
-
r
O
v m (o 0 (D
c� c -
m
3 �_
O
�
O
*
o
— o to to m
Q O
O
0 Q 0 ° (D Q
-0
<
(n 00
c O Q o
(D 7
Q
0
O
3 o 0 m O
CD
a
CD n (D
00 53' v (D 3 Q
—
O v
(� O m 7 (D
=. - m
(D
=—
CAD
0 Q 0 ,-+
O<
cn
�o � ='
�o
O
(n � CDCD u
�
o
o
0 * 3 o v
=o
° O m umi
`< Q-
CD
< cn OL
,��.
v_,...
o
� to c Q 07
O—
0 3
CD
J !D
O
J (D
�.
C p' O
m O
O
s
0
0 3
v (D O
3
(D O
=
O
((nn
SD
(n cn ((nn O
cn Q
-0
° m
CD
�. N (p
CD 0
(D 7
OmmCD
fl
C
r«
CD
O
=< ?�
(D fl-
o3CD
<
OQ(oQ-0-
(�
00
O cn Cr
CD
CD N _
m
(D (D
_0
O C
N
cn
v
-6
m (n
= (D 7
O
C: -- " _•
CD
O
cn " CD (D
O
(D
(n
(D
3. v 7
Q N'
m
(n
m m ° C'
3Qw
Z3 -00 : =
cn
N
Q
�' — N —
3 O 3
0 v
•J -s m<
�<
�'
(n ,.'(�
0 m (D
c 3 -'. Q-
cc
(D
O
Q
0
•J
(n
O N n
cfl CD
v v
3
0 M 3.
v c o
—(a
0
Q
° �� m
m (n
c
$
O
:3
� crw <
m O o
m �. �
o
-a
D
(D
o'm
7 (D v
N 0=
vcn
Q (D (n
-a omo'
v cn
�og30
?
n
�•
o
�cmn�
0 m
(D m
o
3
° « c �
m- cD
v
<<
o
3 cn (n O
m 0 v.
= �
s
0
N•
m
3 m
_ zr
CD
-
0
M m
cCDD m o CD : m y 3
gc—� o wm 3
o <oCD
0
0
0
°'
wwwwSlnmm°3(nm(nmm
(nD0 0
N ° Q cn O � (� � m (p O
00m
(n X 7 � (p
owl<-�('3D
-a (D (D
cn�oa
CD
vCD
m
Q
m
Q
(D
Q
Z-�
: CL v o' � m —v 0 m Q
O D m o m —0
to (D
v (ncD p
m
0
m
En CD � Q D O < O 0 m -
v v ao m o (D cn -o m O m- O v m
yr Q m
a m nm 9
v
m O
a� fl m
�
m m'
a<
CD O
M Q vCD s � m 5' m� O (o (a
fl-v (n C m 3 n)-0(o— v m C m
N O° M.
� c 3 3 v
C7 o (Q
(D °=
�n)°c�
0
o•(0n
mQ
Z3ou�o°c(n°�-„ca.��Q°'�CD
= Q N (_n o' m (D
:3 W(n n.
cQ cn (° 0 <
CD 3 M
M(n
° ai
G
�-0
l�—D (n
C a)
• j �, 0 7 (D
v
O m z3 (n m
7
(n
w
CL :3Z) m2 �57 (n CL
0_ m -0 , m :E O M m v
��, nco
C7 , -0
°�v
(D (D
m3
a m
7 M
-0 0 CDm Q- (B
O 0 m
z m O-
(n m
0 7
in
Q �
o
N
A
N
W
N
N
N
—
N
O
(O
OD
�J
(3)
U1
(n
in
(n
(Q (D
(D
(D
CD
(D
(D
0
(D
n
(D
0
m
(D �
(D
(D �
(D
(D
-0,
= O
(D =
O,
33
O,
O,
3
,
O
=
•
O
:3
O,
=
O O.
M_
3 O.
_
3 O.
CD_
N (Jt
Ul
(11
(T
A
A
W
0 3-0
m
W
CW
C �. C C
M
O N
(D p
N
= p �
CD
CD Q� 3
3
3
CD
3
w N
3
3 m
O
C S 3
�.
`. N O
o
J
c l<D (3D (&D
Q
O
m
n
?m
(CD
=
< °
'
(n
O
n Q O
(n =
'Zl (D
<
=-
(�
mCD
(<D o C 3
W-
3
n
N
C (D
(D
�'
zm
O v C
o
(D O 3
3
J (�
n
� In
j 0
°
(D O
-
N
O
(D (D
O
cD cn cn _,
Q (Q v
(D
"� O Q
(D
N
v
(D
3
7
p= (Q (D
O
In
p O
z
Q
n <
=�
J O (D N
m
w m
*(n O
N
N
o
m
3
`2 M (D
-a
(D
_
O
(p
m �'
(n 3
cn
J
N
= C-)
LD.`G
N
Q
3
(n
Q-
O m 0
O
(D
3
Q
,71 Dcn
m g ��
CD
��
:c
CD
E
o
v
CD ��
Qmp
=
c�3
zT
m
°
°
3
Q
M.
=
CD
Q N `<
Cr
m CD
m
m
y
O 0�
(D 'J �
cn
(D
<
•J
p =-
(n N O
m
o
(D 3
-
CD
Q
CD
—
cn-:
3 M.
cn
(Q, C) w
Q
J
cn
cr
CD
cn
(D
oo
m
0
CDQ O
QO"a
Q.
cn3 _
0
(D
CD-((nn
O
(CD
m
Q-
N
-p =r
m=
0 (n (
n
0
o N
N
(D
0
Cr
(n
N
(D 3
(D
In
Cr (n
N
W
(Q
a p 0
C
0
(D - 0
�
TI
v m= L
-
-
C
o
_' N cfl
3. O .J
0
Q�(Q(n
•J = 0
j
v
O
m
'�
(Q�
=
��0
3
a
-Ci
m
Q 3
.-�
N
(D
,x.�
.-
O
Q n
S
N C7 v
O.
3
m
-
�2
m O O
Ocr
73
(Q 3
Q-
Q
O
CD m =
= Q
Q
(n
cn
o 53 9c w -3 �
o m v m a m O
Z
O
Z
O
Z
O
Z
O
Z=
O
= Q= v D
m n= CD n-°
aL �3cn o ao < o p
N m C (CD O (D v (D
Q V,
ma Q-(n;U
n (D C m
Cr
-• o m
.__•
Q m
a CCD'-
(D
Q
(D
a
m
Q
(D
Q
(D
Q
(On m m v 5- r
Q p �
(� _�
C zr� ' Q (D N C (� O 3
m 3,
6
m O0 v m
m N
Q cn
° (<D N= �. o
(n 0
(D
Q
3 w (D - 3 G (=p
3 0
N°
C O
N 6
CD
In v (�
n O N O
Q
O=
J 3 3 N (p Q m (mp (Qp O �'.
O
c3i j "a
N-
N(D Q
7
CD m (D (D (D w
O
m
< N (Q n (D -p D (DN
m(D
m
O N
� m 3
p<� m (D v (D
m m Cr O' 3
Q Q m N p
f�D 3_ 0 m w = lD _ -+
N' O O (Q m = m O
?_
(D .G . 0
cn
Q O (Q
N
_ m< n
r2 (n N N
N Q
CD CD m lD m O Q to
<
CD
m �
O
o
m O= =° 7 -3(n
< CD lD
Q m ? N
° O N' CCD O D cn(D
n
S
cf.)
O
M
m (=D N (D cn
cQ
Q �.
3
OZ
E n
-
(D (O
W
co
W
N
W
W
O
N
(0
N
00
N
N
O)
N
al
CD
n
n
n
n
n
n
n
n
O
O
O
O
O
O
O
O
O
7
7
7
7
7
7
7
7
Cn
cn
m
Ul
U'i
(TT
U1
(n
U1
W
W
W
W
W
W
W
W
W
W
w
W
N
N
N
W
G)
(n 6 CD
CD
Q O =,
n CD r
n n (D W �' p
-� co
cn (n O � 7' 0 cn 3 �
N CD m N W y
-I
v
(n Q
p (D y
cn
(n
0�7
0 C j
< W 3 CD
(n
m
(D CD
- W
7
cn
7
t
m�
Q —
7 p
J
7 7 N °6 -i
"6 CD cD (D W
-�
(D
3 7 n
~ 7. N .-. (D (D
0
n 7
< (D
Q m
c
C
(D
(D
7
E � .-.
`< Q
5 (n
(n
CD D O 7- 7
J
7
co 7C W n W W
.J ,--. O° 7 7•
m 7
(Q
(D
0)
CD
o m =r
�0
o m °
m
v cn m� o CD W_
� CD
10 <�
�CD
Nv v
m�
7W
7'
Q5'
70�<<�O om
nm_
�. 7
7-
O
7' (D CD (n
O W -p C
(n (D
(D O
(D 7' cn O Q
°(a 3 7
7 'O
O W
7 C O W N (D
(D Q C 7 (n 7' (n
7 (D
Q
`� zr
n m
(D (D
7
C
Q
C cn _
ZY W
Q (n
7
- (D
•J W W n
Z
_� -
ca
cQ O cn (D =r .-«
m 7 -6 N O a7,
�
cn
W
n
CD CD
Q cn
O (n
� O. 7- CD
<' (D
Q—° (CD
(D W a 3
(p (n (n (p m 3
C v
(D —
O
C
N umi cc Q
m Q
-
n
n m
(n
Q 7
cn
O
W
O (n (a
m 7 n (p C
7
CD �
0—
(p
N Q W
7
W CDD
O
CL Q
°
p W W
7
Q7
`<_
. F
C O 7 C
p C 0,
(nm
m W m (D
n O =3 O °'
O
6 m W 7 (n W (�
(a CD n CL cn � W `< 7
(D W
O
W Q
(n W 7
(D
m W
O
CD
W Cf CL
�. (D
C
W Q- <
(n
(D
O -:? 7< O= �.
C< W
Q
O
N
cn (D m
7= 3 7
(n CL
cn _
CD < (D
m o Q Q 71
00
-,
n n W v (D (D CD
C W— (D (n (n O (D
-
`t
0 W
7—
m=
0 cn
W cn
N
7
?
N
(m—D m
m n Q �'
n m
_
m m
T. O O W
-
n (n N
p (mn
n W n CD CD A `<
(D
p
7
W N
7
Q m W 7
0
m
0
CD
0 0 (D
C
3
W
O n m W W fl
� C °- 7 0� W
3
C v
(D
3 0
3 (n
W
O. _
Q
cQ
���
7
O
—
CD
Zc 7
j D
(n
3-„D��O
O 0 O O
�� n
(D Q= �_ N
(D
W
v
<ZY
(D �.
(mp
J
7 O (c <
cD 7
(n
3
3
m � (D 07
CD
W
7= °to : (p =
-gyp
CT
3 m
° ('D
y <r C
< mn
(D O
(n J
v
Cl..
n
in-' 3 (D �. W
cn �
O
-
-
_. Q
7 n
-°
-
(mj cn (CD
N
� O cn
,�
7 j (n
ZT
(D
° W
W
��
7. (D O Q
C
.+ (Q (D N (n O
W
W W
=
�_
cn 0
(D
� cQ r: p
3
m m W cn 7
a sZY (D
c
OL
0 CD
O
O
N
7 Q E (D
W
n
p
cn J
C
Z 0 CD
m
7
(n (n 7
°
7
n
CD
(n
7 W
(�
(n
TI
lD
0 0 _E W
(D ZrO Q (n W
uO r
W W n (� <
n 6n 3"° ;:V --1
o (D O T C
TIC
W (Q
Z7� (n
W< 0°
z
0
Z
0
Z
0
n a'(n
0 n Q
07
CD O D W (D
cn cn
C C x� p
_?
°O 7 (D N W < (n
(D O
W
lD
Q
(D
Cl
(D
a
c W m
°
.n3i sin? m Q Cn
U)` iC j (r
v° g
o o lD w
c cl) 0 c �v
o 3 0 7 3 n(n
m
3
o n>>
N (n(p
(n Zr
(D W
Q
(D
n p (n O
W M ° W
5r
O (D p) n_ j
7°
N
<
W 7 (n
(D .+
—^
°
SD
W O (Q °
3
M S 0-
cn
m
W° p m o N
7
O Cl
(D CD
(D
r7.
7 N n (D - *.
7 W
C7
o m (D
7 rn* ; Q
O m C (�D
(n W- 3 - (D
m
fl' cn a)
<
cn O
7
a
CD
(D 'D C CD
Q C? O
(D _
to 00 W (n n 7
O (D - C <
7 (D W< tn.
(D (�D
<
CO W (D
(o —
O p
zr
O (D
° 7 (D (n �
7
n (� O O
O- 7
7 7 Q n—
7 (D
(Q (fl (n C O cr- Q
O p
(n
� (D
m
Oi -a
CD °n
(n
O _ C
(D m 3 O (n
* m
(D (D (D
(n
m (
7 O
0 CD Q Q
z
'17
C
Cl)
N
O
cD
W
,J
..Q).
CIi
-P,
(D
(D
(D
(D
(D
(D
(D
(D
m
(D
O
O
O
o
O
O
O
O
O
O
C
C
7
C
C
D
C
do
W�o
v
v
v
v
P
9)
m�-IDmmo6�woM�o�a-2.�p
v ° m °
O =
<cZmm3'-(n,ww(pogvorsom��rn
Q
!i C N
O
v
((DD
(n.
Q _a Q
w o r Cl
m
(/.
_° ((DD
w C Q
C„ 'Oc cQ (Nn w
w (D c (D �_
v O (D 3 (n
-, 3
(n c 6
Q
o w
o
c
�_
C cn 3 -• ()
(n m (� (n o o m '' c
o
o C w
6 O
w w m N w
o n — 3
3 O p
O w =3 w
C
m m C
(n cn—
(,
(n C_
(n—
(�
om
m(D
Q (n C7
O < (D (n
O
m
(� (D m v O O
(D O
p m O A�
a w D
O_0
-� �
(D
(o
C
0- p�
Q- (D O
.
o Q- Q Q N m cn —(n
m Q
Q- (D E O
73 o x
n N0
C F -n
w 3 3
v -� m o c�
C. (Q
C C ,
p(a w
07(D m
m m 3
(D
3 3
o
(D C)
ffl
�'
(D
- < _ ZT'
(D w ' m - cQ
Q Q Q O
w
(D o
,
N
m (D
(n
Q * C —U p
O < O
(o �(n v
�o -p
w w cn zr
3 m '
cn C Z
C.
in w '
C
C
<(Q
o
C o
�oQ� cn
v °
m<
3 oc
(�
w o
Q
cno cQ(nc�
3 v �`�' 3
C o
(�' op
c
oo
��
�'
3 o
o (D
;:I-
-
�_ .o n.w
O O?(D
o
Q `2
v
v a OC
c
J 'C
D
m
w
�
C O (D —
Q m
O ''
(� m (D
p —
w
— C
s
X
^D
w
T r-I (n `< '� (`G/7
(D
C- C.,
Q w
- Q- ((DD
m
Ill Q (D -- (D O C
cn w cn O Q Q
(Q cn
(Q O o
O<
(D
(D
m
w w (n cn 5' 0 (D
.cn -0 v o (D 6 v cn
om-
N l<
(On
(D cn r (n w
(D (D M 'ts
`n.3
p 03
m (D
3 (D
o
3
(D
(D D :.- m m
m
�c w
sm Q
o <cn �3
N
o c�i``
D cn (n
�o Q-m
c (n
g.
p
Q n j Q ciUn
(D
50.
a
O C C C n
(D
m p r
cn fl
'J
Q(n
v v (no� o 0
3 m � -6 n
G o v
n 6
o� o�
3 3
a� 53.v v Q=
(D m o D
3 a0 _Q
� v �
°�
v
c
C
(D
�'
C
(D C o=
cn w �. w< n' ff� (D
'
C
D ccnn O
(D Q
w C C
Q C O, G �+ Q
p O C 3
n �.
c w
v D C
C
�2
0
�G — Q N o N' Q �
-0 � w =
C v
w m
� (D
w -
cD n Lll W (D " v (D
3 Q. cn. -'
-+, c ?
cD
Q =
c� (n (D
v
cn
Q
Q
(D N (� C 73 m N Q- (D (n
o 7 3 w (� < to p
-'
p' (� (n
C w-S2
Q a m (n (n (D o
ZY
-s `G
_0< cn
(� o
C
C
O w (D
v
m
0-
6
W m
n N ° p n. Q
p�
(n O
(n
C (D
o m Q o Q
(�'
N O< �. O
CD cD
—
�� v
(ten
-
=.. (n CCi n
O
N
-
(D 0
..
Ci cD m W OC = O
cn E l<
3
C w w
m (Q n
C
-a
<
(D
w -* , cn
c.) O w= G w w
. o
'�3
c
�+
tl► :41 N c C — 3
o
m m O zT
-ow
(D C :'
6 o (D
w
<
w�cn(n 0
(D cn w O O Q
O
u,o
m e
m'(n3�-Q�w
M •-,- 0 n m *—
'm O
(D
0(D D Q•
w77
�
� w
D (n w
m r
CC
Q m
'' O (D w
x O O C
!-
O m
D�
Q -� 6
C O
O C C
a
(D -' m n m Q- (�D (D
<
O w
e < 0 C Q Q m Q
g O
Q? m
n (D Q
N
�•
C
—
= (n O m D (CD v
w o C -+ (D (Q
o
O (D
O� C, w O (n � w _„
3 Q 0- ZT o
(n O O
?
Q Q-
Q -6
O
Q C 'W - r w m
(D
3 cn
-0
p C)
C _ - m (D Q-
Fn
C O ... Qi-...,.,
C m
U � 5,
(D
O (n
(n
w cn -6 O cn cn `< m�
m -- v m N QQ
O (D (D
c
c w
� O o
m(D C O
w 1 a1
m 3
" (0
m cn
C C (D
m
m 3( �' 0��
Q
a r
Q o
(Dm
Q
0 cn
m 5
c m
„acQ
� o
3 �'O
u,
o
< w w`<
i o Q -
O
Q
Ln �
� w o
En w,
N
- Q
c<D
a F)'
(D
C: CD
cn x
(/�
(D
w umc W
o m X Q
—u3
3 C m
cn �mc
n m x
—07 �7
3 C (D
U) D
(Q
D
(Q
D
(Q
m(?'-
Q
-z
m
m
OZ3-a�v
C Q o O�
om<05-0
G
C Q p O 0
r om<
<
o
C (D
(D
m
o wv--
-
!�- w O
�
7 wQw
��
p�
n (D
p
m
m
-�p�
Z3Q
ca (n0-a-_
QQ
�v7w(
m (D
Ca
Q
O m o 0
O o n n
wwv
o (D o 0
�w
O
o c
w
(n
0-w c
m
n0
c� c c
(T w c
v
0 c
(Dm
O Q
Qc
rm
<
D Q
vcnc°QQ(n<
O C
'
o,
0cQ
O
Q
���
zTQ
Q
<o
.
on
�<o
.m
m
c
(n
^
(D w
c
cnQ
(D w
3
�'Q
`�Q
C)
'T7
"'
-j
rn
�
o
((o
coo
m
m
cn
Cl)
Cl)
cn
Cl)
cn
(D
0
(D
0
(D
0
(D
0
(D
0
(D
0
0
0'
o
0
0
0'
o'
o'
00
:3
D
D
v
v
�o
W
00
00
00
0:)
W
W
Ui
A
W
D -n
Z
(�
(n'
0 (D O 2
U O (D S �
�
(D N a n cn
O rt
o
(D3
()
C a
(nv
n N p�
0- 0
(D
3
Q (D < O
-
(D (D
270
3
c��
°' v
C-
3 Sv °' 3
Z- me
(D CD !?
�Z3
v
(0v(
N
-Do
o'�
O
,vQ�(D�
ill
CD
-�� (D
>_ (o
�° (D
(D o
� (� CD(
3
= <
(D (D _
O
(D
_
cn
0
� �
O' v �
(D (D
0 O
� - C
M. (D �
N v 5' (D �
�
(D
� C � (D
(D tll
.
(D
73U)a)
No
<ocr (D
`nCD3(DO
3�07T
(D
Cn
(D
Q
0+
(D
�M"
C
M w O O
Q
N r
_
00 (D
cn T
0
c (D
Q < _
((DD (D Z' N
Q
O
(n
x- -O
o N
0 v fl (D
0'
� _� (D cn
O
J (D
o -*
N (o
0
!v
3
ZT
O
O
(D p v
cn73 (D
N
(D v 3 0
-
T<
Q
,y
-
fy CD
(D O
cn
J
N Q- (D
0
�_ (D
(D
= 'O O
Q
!1l
Q
(D
N w
n
T(D
0 0
n-
(D
(n
!v
73 c Q
O n'
0 ZT 0 0
:3 (D (D (D
n
Q
(D
(D
zj
z
p o
p O � (n
O-0 (D (n
0
a c (D (n
(D to (D
3
(D
(D
(D (n
C) p
(D
0
0 (� N
0 (D —
� � �
Q 0 O O
�
(fl
n
(�D
-0
O
Q
(D
3
0 . O c
w
O v= 3
O O
�
- 0.
n (D Q
- (D O tD
x
<
p
(D
Q
O�
(n
((DD O 00 (D
(D
(0
O a3' O 0
(D cn
N
cn
Q
N
cD
(D 0 Co
-
O (D
n
O cD ;*
(D
�_
(n
_
(D
J
Q
_ v
(D O
v
,
Q (D Q
O
0
(n (LZ (n
0. 0 N Q
0
� O��
'
D
Q
O
�<
0-(n
c 0 <n
N
� w r o-_�
o
0'.
0
In
(D
to
N 0-
c D
m (n c)
O (n
� O
_
O
Q
0--
O
N
(D
Q (D
0 C W 0
— 73 c
p)
-
(D (D O v
cn
X
5' O
=•
O
�G O
"�
(L1 Q Q(D
(D
Q
0 (D
�
Z
p
(n 3
O
O" C) (D O
0 Cl) Z
N C (D Z
II (n <
3 v O< < O Z
—
0-
0
7T
(D
O (D
O O in (n
v x <
� 0 (D
O () 0
3 0
(n cn O
cn (D (D
Q CD (D
(D O -o
O (D D Sv N 0
N O O (D (D O C (�
,
< 0 —
Q
n Q D
y�y
m X Q o
a Q
� D Q
x �' :
(D O
m vi cn cn m 3 0
v ((n
cQ (n
c X ao -
:
o- -�
(�
p-0(D 0-
v D O (D �n -0 3
"
0 o
(D
(n N (D -n
cn p
Q Q ?
C)
(D (D
� C) (sa n c p
0 a
(�
N (D W T
Q W* <
:
(D (D
X II ((D
O cfl Q O tn' 6
O
!v
(D Q � :
11l
N (D
N
(D Q Sv
N (D (D
0 o N C (i3 (D 0
(D --'. n v �
(0 (D
l< Q
(D
0
p� o
0
(p o< 0
0
O
- 0
O
(D
(n (D
O 0 (D
(D
:3� 0)
N (� fy
n (o T. 5 (D D Q
v
O
N
cn
'
(n
n (�
(D
<
(n
(a cn
-
n- 0)
C�' (D N = n v Q O
(n N
0
Q
!v
0 O
77 n-
(�
Q
w Q
N
; O
(D
N * CD E Q (D 0
(i
_
(D -n
v7 v Q
CQ
(n n
(n
0
❑ ❑ ❑ ❑
rn
cn
.A
coN)
3
D D z
_
n
Cn
cn
(D
C
(D
��
o
Q
�Q
�
=
Z
(D
�
�N
(D
�o
cZ)
W
o 1
(D
(D W
7
N
o
(D
SU''
—
W
m
�.
cn
W
(� _
O
Q
C-i7 0
o r < °'
Cn
Cn
v cn
(n D
5 Cn
0 Cn
O 3 (D
(D
(D
m
o ZT
o o
(D 7-
T• Q (D
(D =•
O
(D =r
tn', O
D CD (Dc�
CD 7
(0 (D
O C 0
�
*
(n c
Q
-9 a) O o
(D o M O
3 �
O Q
�` S
Q
n. m m
o ..
o � cD
�
- cD
� � _(n
0
� m
� -•.
(D
m
rn -
o
< v
O CO
v
cn N
T. Q
-
J N c N
v
o Q
3 _?
N y
< to (Q
(D o
�'
(D � �
Q= D
3
o (n
-p
O
3
(D
Q ZT,
(D
(
(D
O
po
Q
6)
O(DD-
p (n
cn
in
rl ,
(D
3 3
O(D
O
3Z
o
D
_ (D
J
El ❑
❑�
(D 4)
a)
o
(D
Q
CD
^)
Soo
o
mcD
Z-F'
v
�
O
7
p
(�(p
(D
CD�
(�
�
Q
a
U) (D
o
(n
(D (D D
o
v
O -,
.
-
O
Jam.
(D
QL)
o ((D
cn CD
(D
(D
fn
(n
El ❑ El
(D
N O
3 (n
O
v
3
Q
`<
o 77
�.
(D -o
C
�
N
(n
.0
-I
r^
5.
m 0
v
(D
7
(D
7 G
6
C)
v
VJ
Z
::3
(D
o U
t3
Ul
O
Do
v .0(D
70 (D 0)
Q
❑ El El El7
,;
(0
C)
� o D
cD <
°)
_ �,
°' c�
o
-
c o
= 7 cn
= o
<
Q
rn (n cn
(D
m °� D
((DD
° o
D
cDi
o n m
3 3 <
:3 3 3 M
(D (D (D
(D
=
D n
O
o
(D ((DD Tzl
o
0
x Q m
v (D �Q
:E CL m
(D
:3
D z
o
�
m
ovQ6-0Ov0NQ-�0Q3�(n
-*o�Q
C)
zv3�Q�
z
C
D�==(D
Q m� Qv(D
���n�ic°)i
=(D Q
v(n o D
o
(D
� �co�
QOp o
o
D
_0
��mv'v���'om�
0
(n�C:-(D'
o
(D (n
. CDQ'
o `n�((DD<
(0- n s .
(D
m
oo(D
D�
o (D p (n (D �. Q
v(D �
p D O (n (D N
D (D (D
< Q Q
(n
O C)
m(D0<a)��((D Z3
c)'Cc@���
aE; (D
�-
z
�° �.3 o 9 Q�
�_0 3
m n�' D
X o O
_�
v,cD7v- nn) -(D
o—
z�v(D
vCD(DQo
ino0—
(D
m
FT�m�.� v�'
< * Q
v
D
_0ch
(D
3 n
(D 'mom o -
7 Q
(D (D (D (D
(Q
n �
v
Q m
� o (D
73 (D ::3 in
D
z 07 o co
m
D o
n �
n
O
if
m
JZ
rr ^^1
VJ
Q O
D L-
� m
O �J
� � �
D �-
0
D
�(n <
cD
Cn
:7 Q.
(Q
O Q-
O
a (n
�< D7
(D z3
Ln. (D
(Q
((DD
;u _0
(D
- n
O (D
(D i
r+
v
FAI
..
O
(fl
00
-4
Cn
(n
-0 (n
cf)
m
(D
(D((DD
0
(D
CDCCDD
(D
0
CD
(n 0
W (D
U) C-)
.
(n .OF
O
(D .0
(D
(D :+
CD
(D N O
j (Q O
j CQ O
O
CD O
3 O'
CD 3
3 O
CD 3
O
m
m
-} 3
_0 ()
W W
W W
W W
W W
W O N
:3
O = N
:3 W=3W
O N
o
—W
—W
—W
—W
—�-
(D-
CD- -
N
CD
E0
W W m 3`
N O< O m
c =3Q-0-0-0
(n o
(_ "" O 0 CD
�_
0
�
0 3� 0 -i
O N
p j ' 0
^
(fl
(n3
N �. (p• N N
o CD
CC
O
"'
j (D j m
W
m
m O �.
3 3
� Q
((DD N CS O� CD
N
m O m S
.< 3
cr
CD
CD
Q 3 °0 D. cCD
m
3
M.
_•
_
CD
00
cCD
M.< O iU 3
3
3 (D
O O
-s W co (n Q
(0 Cr 3" (n
CD
Q) CD
-
:
W
a_
M. Q (n
(� � � �
Q
<
(Q p (n (D
O Y Q—
0
7\ '6
<
CrCD
m (Q o (D
o .-+ c—
W
3
O
�
0 �, �• 3 (D
0 (n (n
Q O
C n -O m N
(n O
O
m
Q
OO
3 Q W m O
m
�6
— CD C)
°o a m 3 Q
o v
W m m
.+ O —
o Z.3 CD
m
mQ
(D
���
n
� o<
�'-0
Qo = c
moo'
o
n m am
° W
o
o
o
O
=o
O Q m vOi
(D Q
O O(n Q
cn -�
0
to W Q CT
O
CD
=,.
n
O
00
(D m
CD
3
n W 0
3
O
J
7c
((nn
c p'
(0 O
3
v m j- O
7'
(D
j- 3 —
=
p
((nn (n
3- = Q
Q
(D W
^� a) W (D
O 3 3 70
(D
Q
— (D
o (Q CD 3 N
Q 3"_
7'
(Q
3
O
(On
0) <
=<?
•J O
m
3 (On 3— Q
o CD m
J
Q W m N j
=
m
_
Q
(D fn m
23
m v
CD
cQ 0-
(TQ Q
—
o
(n
o a
* j
(0 CD QD �. —
(�D
m -0 fD
m
(n
v3)
= m 3 to
O cn
0. j. r
l<
to
O� ("D cZ) m
CD
c
(p
�'
—
3 �,
Q N'
JCD
m O 3 (n
SCDN < 3
OQ
W c 0 CD
3 0 y 3.o
Cn
(Q
0
CD
Q
'-' .. COD —
O 3
cQ
CD, J m<
W (D CD
3. Q
CD
O
Q
C0
v
0 (n
O w
° Q
< p CAD o 3
0 Q O a
3
cQ
cQ CD = W W
CD W -' - cn
3
3
W
W
c O
(n (o
Q°
W
< CD3 m
3 N
p
O
N �O O
N O c
O
_0
3
Cp
3 m (D W
v (0
Q° m (n
c
O O
W
CD .� 3
,
o .-� c
o °
N
o
3- to (n 0
m C W *
C
O �
N
3 CD CD
O M.
CD
3
CD
5'
=r CD
Q
0
QCn3o(n0Uoo;poWw0—o7
m (D
cnD��cn
W F. O m O D (� 3 D_ W (�
mZT
0)0p
,� � O CC) O (D
n*Q-Z
O m O
(nc/)
m e
m
Z
O
Z
O
Z
O
O
O C" Q (n c 0 3 j 3� CD W 3
W �
m X? � 3 ch
(n CD 3 (Q
_ 3 CD CD
Q
m
CD
Q
CD
Q
(D
Q
O
3' 0
W W 3 cA m O " (<D 3 � W CD
3 Q• W W m O m cD W m '-' Q-
— o
° Z m O 0
(n .0
-0
W
(D m
W (n
m
(D
N Q CQ 3 3 CD 3 CD D O < Cn (n CD 3- -
Q0 o m CD O (D o v m
n
(nn Q (n (D m 3
�� 3
cWn p
3 O O
Qc Q�
3.
<
W ��
3 Q W < m �. (D Q 00 3
-0
nfD
CD W 00 o
C7 (0 � (Q
CD
0 (n
CD O
CD o
Q-v 0 �m 3 v (a w m 3� m5-
W 0 o c cn W Q o Q-W 2 CD
=
�'.� �-0 3W
3 W cn 0-
:3 �m
�'m 0 3
W° 0
3
o'N
3 to
3°
(n G
o°cr-03-00- oWIT mm
3 W 3 m o W 0 0 0
(D 3
CO
m W m<
CD 3 (�
m
o (n
0 �•
c m 3 . 3 m
CD W —' (Q � O CD I Q cc (n3 to Q° Q
O 3 m
_ Q° O
c
3 3
n W'
(n
m 3
�' w o CD m n Q m Q
0 "0(D
cD
c m
-0
0?
in
o
Q Q
O _
3
4�.
W
N
O
(O
OD
--1
O
(r
w
wCn
w
0
Cn
N
(D
UD (D
CD
CD
n
n
n
n
n
n
CD 0
(D 0
(D 0
_
Z3 p
(D =3
O
7
O
7
O
7
O
7
O
7
O
7
: 0
(M m
3 0
M 7
C O
(7
(n
Cn
Cn
Cn
?
A
W
�_ W
W W
W w
IV
N
N
W
N
W
W
W
co
CO 3-0(D
(D (gyp
O W
W
CD p
~'
O W (D
(7
CD W
to
0
�+ CD
(D O C
O
(D
'' (n
(�
7
(n
O
W
p O— 0
CL _
`<
CD S
O Q In W
W _
w
W
a ci
�
C
W CD
7
(fin
W
O W
n
pOj W
J
<CD ���'
���
0E=�
0"
cam•
�03
CD CD
v
3
M �' 0
O
p
O
CD
W
in
CD
p
� N. W
a N W
° (D
a Q-
v
COD
p CO
O
0,
Q
CD
n
(n
J O CCD W
W
w
0.
CD
(D
cD m CD
z
c
(D
cQ
f0/1
� �c cQ
O
(D
<
-
= (7
-a
W
O -
CD
En
Q
p W n
C (D n
(nJ
O
7
T. ° .Cl =
�.�
m
! to
�`
to
0
(n 0
mdO
3M
0
O
Q
(D D�
cn
(n
6
CD
03
c
W
'•
o
7 p
CL O
C� (D
°
CD
(D
(n
3
O
CD
n
0�oo
�,v�
�
(D W
"
O
CD
(2
�Qv
(D _«
Z cn
(D
W
-+
�'
=t
(n 0
7 CD W
p
c
a
"
(n v
°
-
07
�
W
�
-p
�' _
r: cn (0
a CD
— CD
(D
o
O.
Q _
�Q
< CD
m
3 r-
pCz�
°
�
r2 n W O
(D
CD
N
C]
(25• C) W
Q
(n
(D
N N ="
W
W
(D
CD
W
(n m•
N N
m O Q
C
O
Cn
cnCC O"a
p
n
v033
n CD
<
j
O
`< 3
((DD QO
a5U
O
Q
'.m(n.«
_
CD
N O
CD
g�
(D
W
��C
O 3 .-.
(n (CD �D
0
() N=
W =°
N
C,
W O 0)
I n
-0
cn
J
W
6 In W
�
0
w.
(�
In
(o
CD
Q N O
(D 7
cD x O
_n
(D Q
<
-. (Q
•v
O
Q - (o (n
p
a)
=
CD
7
(Q
�.
W 3 O
Q
. J 7 C)
.--F
O
O
�
�'
W
7
Z
C
= W
+ W N
CD
.-.
O C
Q CD
°
W
=
CD
x
O
� -
< 0 O
v
N
O
C
�
Q
�.
o EF3 EF W 53
0 (D W (D �. °' (D O �
Z
O
Z
O
Z
O
z
O
zc;-
O
Q- W DQ-•-'��(n
CD (D �. CD 7 —
o QC) < 00
O (D N C: CD CD O CD ZY (D
C,
m-0 OL
X CD C (D
X
O CD
W fl- Q CD
CD — CD
,--�
CD
Q
�«
(D
Q
(D
Q
(D
Q
m
Q
CD n— cr 0 r
(n (D (D W
Q p*
0 0 v 7 < C (D � O W�
C C- T • Q (D lD 0 M .
(D 7 (°
Q
cr
C) �� (D
CD O v
cn
N �
W
° (<D (D(D(DQ W (O—D -,
(D O Q CD
W j Q
3 ? 3 (D 3_—
�� O (ND � n �' Q �- !D a (D o;
W°
Q' Q
CD N C (D
W (s—D
N lD
Q C
CD
W WCD CL CD CD CD n C_
J 'C d (n O N
i°
O
CD C) Q W
W CD Q
(D
N (D
6 (D (n 3 W :
O T• N
< x. Q (D � (D N (D N �.
N (n
OC ,�
N U) C.
O
7 w N < 0 W C
(
.-. cn W In
(D (D (n N (SD (DCD (n Q (n
<
(D (D
�O
�. O
O
O fD � a)) (D
CD W
0 O W' N O D( Q�
(D
n
3
N
�p lD O
O Q�
COED
Z
N
(D
(°
o
.
x
O
-n
CI
W
W
Cl)
N
(D(O
00
-l1
0)
(P
Cl)
Cn
Cl)
Cl)
Cn
Cn
CA
Cn
Cn
CD
CD
n
n
n
CD
0
CD
0
CD
0
CD
0
0
0
0
O
O
O
O
O
O
7
7
7
7
7
7
7
7
7
CSI
CJI
C n
(P
Cn
C n
(P
(TI
U1
W
W
W
W
W
W
W
W
W
W
W
W
N
N
N
6CY)
(D
(A U rC
(D
Q 0 _.
(D (D
O O (D N C)
C S 7 "O O p
-1 W
N
(A O 0 3 C• 0 Cn 3
-a < p 3 -p S2
O
�I
N
(n Q
� (D '^
(n
C
0 `
0 � C7 (D
(n �. Cr
(Q N <
< N "O 3 (D
(D j v (D cn
O
(D (D
(D N (D N (D
� (� Q
cn
W CD
p
C
N
((DD
(U
J
o
(<D
T�
O CD
N 0
G
`0 (D
v cn
�_ C
O D(D 73 Cr_
J(D
0
(Q 7 v 00 3 N
< 7
(D
cr
(D y• (a
CD 0-)
_
0 <
(%1 'O N-0 (D
O 0
- CD
(n G C 7 v
O O
3
O
CD
CD
(D Cn
N O v
(n Q "�
O C
z3 v O
3 r: m'
Q
N� (D (D Q
O O < 7 O O
(D 3
(D <
o CD
6
�. :
�o(D(Q
S.
(n CD
m��(noQ
Q
z3ts
0
p
�Cosli(� ��CD
cD O
�m
OL
�;��FD
(D
C
Q
C SQ
Q
7
(n
CD
N v C0
_� fli
Z
=^N
(Q (� CD <_
=.;U�(D -,`< ?
Cl-Q-
OHO
0(n
=(Q C
7
(D (0D
O
O (n (0
�
< (D
Q (D
(Op cn O �' 3
v
�
(D Q O
O -0 (D
< O O (0 C N Q
3 rn cQ
CD �
(n =h (D
O N
(n
O (QD O fD
`< �
Q -• (fl n
. 0
-0
O (n - O� � �. cn
�, 3
v v �
0 O
J�
C
O
O (D
O
p < O N 7 0 0
-6
(D O
(D 7 O
C
(D
r(n Q O
j
O CD
cn 0 3
Q- Q 0 — (D (D
O
y�_- C N N
-0 p (n Q N
n�
(D Q
Q—
O Q_ `<
Q
,
C
Q
(n CD
0 3 0 Q
(Q C0 m CD Q N Q N << Q
p
p N 7
CD
O N
0
CD
D) Q Q
O lD
—O(n
N. CD
0 (DC
Q' N `< Cl-0 <
CD
(Q
(D
O Q N O C N
0 O (D O
C Q N
O
Q
p
-�
N
(D
m
(D 3
(n Q
(n
CD (Q CD Q-
N cD Q ClQ O
00
N "• 0
C N= lD N O O (D
cr Q .+ +
0
� —
v v
O (D
<
v (n
rt O CD
C
C (D �, (Q OL)ti,
0
N N � lD
cD
(n
(n 0 C S 3 •-�
7 In O
C
(Q Q•
5• Q
=
(D
p -O (D(D(D
((DD
Q
0 CD
CD' =
I
cn. c n N Q
V'
$
cn O (D
N (gyp 3 ,C
a
CD (D
O-0
7
O' O
(D
Q
0
Q(D o'
C0 ai O
m
C7
v, v p oo
3 C Q
�� < CD ,Yv <
° C 0 O N
3
3c
(D �
3 O
p
CD
O
7
O
D
(D Q I(n
O o
O v (n 3 O �• �
(D O
°
=
(QD 3
(D
O
Q .�'
Q (Q <
(n Cn
O O 3 O CD O
Q CT
C
,
CD
N N (Q
(D Q 3 v
cn @<
�_
cD
J
p r Q c
(D 3
(D �� CD
Q
7= a (� n= CD DO
-0
3 O
O D
O
m �� N
Q-
Jo
0
-
Q
O
0 CD6
6
v
ZT
O
�O o- (``n
O Q O CD O
a
0)
.O. �
m a
Qy O
3
0 N N-
C N
(n (U
(D O Q
C
OQ (a
< (Q CD Zr (n 0
—
v (ll
= v
Q- r
(p
co 0
3
(D _O j
Q ca
cD
to
C
O
O rr
DO
Q
Q O CD
N O Q
p�j
0
?
cn
0
O
C
0 CD
Q 0)
p 7•
(D
cn
T13--10
O =^C O --{
(D . O 0 Q
�_
O'p Cr
C
0 6C0 3"0 ;:p
Q.
1l�
v � �
.Up Cn
03
Z
Z
Z
O 07cn
=3 0
('D
C
(n—_Z7�(Q(Q
3 O
cOn �_
C �
N p (0 CA < �
- CD
Q 3 m 3
(D D
O CCDD (0 O
E 7
N
Q
N
Q
(�D
Q
CD
C O
Tl
r> 7 Q
fD (J�
cn X
� <n
O
N 7
(n
CD (00
C (D 0
-. N 3
(D
Cn
O C)
cn
C cn
(0 (D N
o
0-0
(D Q C7 `� '"
(n O
O O CD O—
0) N
CrC w C
O < p n Q p
au0i $pm
0
N C
< N
tli 7 (n
Q m
5'0 (n D� o
(n 3Q-�
o (D
o n
o(D s
v
��ovvQ��n�
�Q
cnv-3o�(D
—
o<>
>�
Q
(D O a C
(D _
O (D 6<
(D CD
C N (D
O O
A
o 00 (D con v
o 00 OC OQ
ca (fl C -- o Q
° 0
Q Q
(DD O
cn
(D N (D 3 O vCi
� C - CD
(Q
CD ° ((DD ((DD
N
(on
Z) (On
O6 CDo (D
N :
- — ?
CT
Q Q
Q
m
z
d
"rJ
R
W
N
m
ro
m
m
m
(D
m
m
(D
0
D
0
Q:3
0
O
Q
O
O
O
O
O
Q
o
M
00
00
00
D7
0)
�
�
�
N
-•
-�
N
N
N
(D
X'
D
�
(D (D p "O W O Q
Q °
v O
< 7 Z
°D
(D(n W (Wn
° Q
(�W(n
WQ
o
v
O (
W�OUI
m
Lc<
�Oc�f l
6 nro
zT �
v
°
_NQ
(D
7 r CL
5
O
nQ
(D c(D
:3 W—p-
3 O-a
v=vc-
N S
(D (DN(n<(nD
QW
°Cc�r
(D
0o
-mC -
�
moo
va
%
=3 CDd
(D
0— -W
M�0-
(D
c Q(nlvoC
,< cn
°
Q
Q
Q
_0
O
pO v
(r
ro
.Q
°
ZT O
O
Q
W -0 '
(D
nWW(D
(
=
�(D p
�
(OQ` _0
°
(D 3
ID
3 3
(/
nO
(D
(D (D (p
p 0- 0-0 o
W
O
(DIn
O_
(
v(C-73=
O
(Z
On
WC
Cc
'
(D
O
v
-
c
0
p<
O
C(
O
C:
°
3 n
O
°
_
E
� -0
:3 w
D
O
m
�On
N
O
(D
(D
(X
_XW
- (nl< (n
n
(D
< O
W Q_
(Q ._(Q
Q'Oo
O nsD
(D
<
m
(D
(i
(n
WW
Qa
D (n r w
(n
(Q £2
(D (D -(D
10
(D
(DO
W Q
o
=
(D _� (D (
c
77
=(n Q(N
:3
_
<
°
°
(D
Qo m?QWro W
��cn v
c,
-o
_c
O c T. w0 c 0 cc
(n �(D
(D -a r-
C_
(D
cn
(D
(D W (D ° W
N
(D F.
°
o(W AO�
W _
(/
n 3
W(n 0�
0'ro
m
O
ro
_
.W
. � Os N
(n W �((D
0(D Q
:3
0 vQ
O
DD O
n
n
pW(D <
(n
QQ
.
O O00
n Wp
N
Q0-
c
(D c (D W O- (D �
Q 3 O<W(npW
o o
W
Q(D D3p
-0 - (
Q<
- < (n
ro°
WO
(Dm
(D
O
(D (D
rop
�o D
(
�oC
a
W W
0N.
((nQ�
0)
�v° - N
Wm
°QE= U) (O(Dm
o
C �<
�c
W U
m a 0
N
m m p
Nv0O
OO (D
0
_
(D W Q
n �n W
o
0-
CD
n(D
n�(a)o
�
zy (D (n
QX_ 6.
�o
m
0<
_� D
o(D
Qa
(D
°=(no �
Qn
D
O (
NiQ
O m 0 (n (n =
°5•
O p o o
�(n(D Q
(n Q
cQ
v om m W n
(D (D a
o_
D
m
°
W o roQ
m (D
�
oin
717 D
3
�(Dro
v�o
cn(
n (D (n
0
m
0 (n
cm
°
3a
7
o
0
Q 0
Q
c om
(
—:3
-
(n
(D
W CTmc cn—C
o�WCmmo�
-0=;,
�o(n
o<
07Mc U)
—073 �u
Cmm
o�
(n D
o(n
(Do
cD
fl
Q A
(Do7
<�.cQOc,
0=.
l<�
�(D
m
m
°Q
W
°
o
�
(n
W Q
Q-
(n
v
Q W W
cQ
Q
_Wo
o
�
CC
W
v
O 0
°
(D
O M 0
o 0 o0
Wc�
0-O
v
M
Mv
cw0
v
(D Qc
'
o`n
QWp'mQo
<
-
0
(n <
(D
vn
(D
o�O
p� D
o
—.cc
o�o
o
(D a-
Q
.
Q
0Q
O
N
cn
(D W
(n
(D W
3
�.
Q
�.
v,
-�
Cn
O
4�1
co
4�-
00
rn
-r�
cn
X�-
m
m
cn
cn
cn
Cl)
cn
Cl)
0
o
0
0
0
0
0
0
00
00
00
OD
00
00
0^0
OD
00
O
O
(D
�`
g
r2
(A
@ @
T.
v
@
O n
N
v (D v C
n N@ C..G.
m � a
UD @
v
—
N cn
Q m
'
@
0 —,y
p
@ —3
v Q� (D m
Iv
m
.�
(D
C (Q
cD°
�cD�
��Oo�o
�l
�D.m��m
3
�0(n
cn 3
(Dv
7 U) D
N O
00 ((DD
Cl
to (D 3
3 O O
�
00
O v
(D
pC
(�
Q@ 0 O
Q < _
Q
@@ Q
CD
(D(D
can
(D (D
0-
zy
Q cfl
lD
0" �
-
N 0 O
(�
0
� O
U) w 3
�
Q�,Qm
°)
@v30
0"
�<
m 0
(D (D
N
-
- 0 (D
v
()
�—
O m 0
(D
w
0-
Q 0
m m
(�
J=
<
v Cl- m
0 •
O_ M m
(D
(n (n Q @ Q
N� -o O
N
(D
O w p)
y cD
�'
O O
(n
_�
N
v
QO Q
n -.
O N (n n
7 @
(
(n
y,
Q n
LZ
4
-.
(n0
O 00
O -0 @ O
(7
CD3 cn
(D
0 (OD
O (OD w
Q 0 o 0
�
� C� 3_
co
0
0
Q
@
3 Un
c _. W e
v
0 �' U� 3
@
(U
<
O O
0
Q
_0 0
—g
n @ -, Q-
< 3 �
�'
_0 @ (n
c�
X
Q
m
Q- O
(D
0= O
@ Q
m O O (D
Or
O@�0
(Q
0 m 0 0
W --�, Sc @
v
w
v
(n
cn
Q�.
m
(n
J
Q0v
!v
m O
C�
v
(Domes
Q m Q-
O
0
a) ^
�
cn (p �
O. D N Q-
�
� O Q (Q
0
Q
O
Q
zT
v
(0 cn
v
p- 0(D
!y T
W
�
0-00
�
(D
_0
((n
W
@ 07
(
@ Q
m 0 0
O (n o
-^ -o
<n m
: @ O_
-0
O
c=
N
m
Q @
O c w 0
— U)
(D@ O W
N
0
m m
z'
-Q
a
O (D
Zo
O_
(n 3 <
O
0m0
v X<
0Cnz
O 0 O_
vcmzII(n<p3w
U! cn O_
Q m@@—
0<<oz
O n Q v !v O
Q-�
N'
@
m@
O O V7 (n
Q 0 @
3 O
(n (D m
_.
(D 0 -p
O O m C) � (�
T.
-0@
Q
0 N Q
(D Q
C
-0(D
X O
cn m 3 O
a
v O
@
m
U) (Q O
v m -„o
c CO -0
(n o
m Q
QQ�
0 D- I
��(D
(�
0 -0 cn s
@
w c o cn 0 3
0 �'D0 0cn c 0 �O
(� (�
@p
�v°
�0
(D �(Q
N
a)�n
Q
ACT
cn
@
D(n
I I
000oQ �330-
m CQ O 6_ �'
m07
<0
m
(�
v
N' (D 11) T•
m Q
( <
N@
? 0
(�'
m Q@
X m
N(D CD
Q o
0 O 0 N O (Q @ O
(n
(0 @
(D
@ (� @
-N p
O
U) (D
(p
w n Z3 N 7
G Q
O�
OL)v O
77 zr (D&
Q Q �'
N
(OD 0 m v Q
A
Q-
O O
0
@
Q
@ (D
�
O
cl
(nN (D 0 (D O
=_ v (D
7
—
N
N C
(Q
U) (n
Ul
CENWS-EC-TB-HH
MFR
21-Feb-03
SUBJECT: CEDAR SPAWNING CHANNEL DESIGN REVISIONS DESIGN
DOCUMENTATION, AND CONSTRUCTION CONCERNS
This memo documents the design modifications and addresses additional analyses
performed since completion of the draft Hydrology and Hydraulics Design
Documentation Report (August 20, 2002) and receipt of Technical Review comments
from NHC (February 2003)(See Appendix B).
Design Modifications and Additional Design Documentation
1) Despite surveys of water levels there is still uncertainty about how the water level
compares between the proposed channel and the river. The February 18th subsurface
exploration showed that river levels were about 0.5 ft higher than the groundwater table
(GWT) along the proposed channel. The reason for this is uncertain. It may be due to a
lag in the response of the GWT to river levels, or improper measuring, although the later
appears unlikely given the consistent trends in measurement (water slope in holes and
river were both in the downstream direction, water level in river was uniformly higher).
The most significant impact of the investigations is that the side -slopes have been laid
back to 2 on 1 from 1.5 on 1.
The water levels in the river adjacent to the holes were input into the HEC-RAS
hydraulic model of the river and spawning channel to see how closely the model
represents river levels (Figure 1). The model does a god job of representing water levels
in the river at the inlet (+/- 0.2 ft) but is at least 0.8 ft to as much as 1.6 ft lower than the
measured water level downstream along the channel. The reason for this difference is
that the river cross sections in this reach are interpolated from cross sections several
hundred yards away. At low flow the interpolated cross sections don't reproduce the
water level very well. Because the previous analysis used highwater marks during a
moderate flood, the model was assumed to be acceptable and a channel profile was
developed to ensure that water did not percolate into the GWT. Since the river was
assumed to be lower during low flows, the design profile also appears lower. I have
revised the design profile and submitted it to Civil Design. It is not important that it be
incorporated into the planset immediately because it is still dependent on remaining
uncertainties. Given the uncertainties about river levels I strongly recommend that we
measure river levels at locations shown in Appendix A when flows approach 250 cfs or
below. This will allow final verification of the inlet and outlet elevations, and stream
slope before construction. The surveys show the GWT about 4-6 ft below ground. Our
centerline excavation depths exceed 10 ft in places. It may not be possible to excavate to
the profile depths shown in the plans due to bank caving in the wet.
2) Due to concerns over riverbed aggradation or downcutting addressed in previous
meetings, the inlet culvert was enlarged to 4 ft x 4 ft and the invert lowered 6" below the
streambed to El 55.70. Since normal low flow depths average 2 ft in this location, the
Cedar Spawning Channel
Final Design Documentation
river thalweg can scour as much as 2 ft before the culvert would need to be re -set. If
aggradation exceeds 1.5 ft, raising of the inlet culvert or dredging should be considered.
3) Figures 2-3 show typical depths and velocities during peak spawning conditions.
Depths average between 0.75 and 1.0 ft for most of the channel until the channel meets
the river. Here river backwater increases depths up to 2.5 ft. Velocities range from 0.4
ft/s at the mouth to 2.5 ft/s at the riffle just upstream of the mouth. Average channel
velocities are between 1.2 and 1.6 ft/s for most of the channel length. Because only flow
is increasing, Figures 5 and 6 from the draft report can be used to show that the increased
flow from the larger culvert does not adversely impact stability of the channel (assuming
uniform, non -turbulent conditions). The above figures are for the average roughness
condition and don't reflect the impact of large amounts of channel irregularity or woody
debris.
4) With the gate fully open, the flow to the channel is increased over the 3 ft x 3 ft
culvert, as shown by the following table, but only when flows exceed 200 cfs. Flows are
increased 30-40% during the peak spawning period by increasing the size of the inlet
culvert. Figures 4-5 illustrate the changes in depth and velocity during low flow
conditions.
Spawning Period
Flow in River
Flow in
Flow in
% Change
(cfs)
Channel (4 ft x
Channel (3 ft x
4 ft Box
3 ft Box
Culvert)
Culvert)
Summer Lowest
100
6
6
0
Required Normal
150
7
7
0
Minimum
(September 16-30)
90% Exceedence based
250
14
10
+40%
on 57 years of record
(Sept 1-March 1)
Required Normal Low
440
19
14
+28%
(Oct 8-Dec 30`h Average)
Required Normal High
470
20
14
+36%
(Oct 8-Dec 30`h Average)
5) The above table does not reflect the impact of large amounts of channel irregularity or
woody debris. A check was made to see what the impact of one debris installation on the
hydraulics. If large amounts of wood are placed in -channel, it is anticipated that LWD
will disrupt flow and backwater the inlet system, reducing the available flow to the
channel and promote siltation.
6) The following table shows how flows in the channel change in response to moderate
flood events up to 5000-cfs, depending on culvert type. Obviously the larger opening
increases flow. Similarly, the culvert discharge peaks out at flows nearing 3,000 cfs due
to backwater.
Z. Corum 7/1/2004 2
Cedar Spawning Channel
Final Design Documentation
Recurrence Interval Flood Flow Max Flow in Max Flow in % Change
(yr) (cfs) Channel (4 ft Channel (3 ft x 3 ft
x 4 ft Box Box Culvert, full
Culvert, full open
50% EXCEED 1,000 27 23 (14)** 17%
—1-YR 2,000 48 39 (18) 23%
—2-YR 3,000 62 46 (21) 35%
4,000 65 46 (21) 41 %
5-YR * 5,000 66 46 (21) 43%
* Channel overtopped by river above this discharge. < ---
** Figure in parentheses is flow with 6" sluice gate opening
6) Large wood installations were analyzed for stability using buoyancy and drag
computations with a factor of safety of 1.5 during the 100-year event. Logs are stabilized
by embedding anchor members into the slopes. Fill heights on the trench were assumed
to be 10 ft, with 2:1 side slopes. Granular fill was used as ballast.
7) The number of dendrites was reduced from two to one because the influence of
backwater during high flows can defeat the dendrite/spillway concept at the lower end of
the channel when flows exceed 2,000 cfs. A dendrite/spillway was added to the HEC-
RAS model to determine the amount of flow that can be added to the spawning channel
for flows ranging from 3,000 cfs up to the 5,000-cfs (5-year recurrence interval) flood.
Overflows will reach 225 cfs at a flow of 5,000 cfs in the river. The maximum increase in
flow depth in the spawning channel at the confluence of the dendrite is 0.5 ft during the
5-year event, but velocities are increased from 0.6 ft/s to 2.75 ft/s. Thus the dendrite will
function to pre -flood the channel and increase velocities to limit siltation when flows
exceed 3,000 cfs. For flows below 3,000 cfs some silt will temporarily deposit, however
the flow rates during normal conditions will be sufficient to remove this silt. Beyond
5,000 cfs the river starts to access the floodplain and it becomes exceedingly difficult to
estimate the amount of overflow to the channel with certainty. The dendrite and the
bank of the channel opposite of dendrite will require armoring with riprap sufficient to
prevent the river from overtaking the spawning channel during a major flood. I
recommend locating an engineered logjam at the confluence to collect woody debris and
to dissipate the energy of the flood. The buried riprap design in the plans should be
sufficient to protect this reach. If the expense and work required to construct this
dendrite are excessive, I still recommend constructing a limited riprap overflow weir to
provide a controlled failure point.
8) Trees placed in alcoves should be anchored by placing the tree across the channel and
bank such that they rest adjacent to live trees. The force of flowing water will push the
Z. Corum 7/1/2004
Cedar Spawning Channel
Final Design Documentation
tree downstream into the live tree and the rootball will dig into the stream bank. This
anchoring system is not analyzed and is based on judgment.
Construction Info
1) The profile and bottom width of the channel in the plans represent the top of finished
streambed (excluding pool sections). Consider over excavating up to 18 inches to place
gravel if in -situ material doesn't meet WDFW specifications.
2) Consider the impact of the GWT on construction operations. Building the channel to
the design depth will not be possible when the river is high. How and where will water
be pumped? If GW is a problem at low river levels, then the channel may have sufficient
GW to operate without a surface connection to the river.
3) The inlet headwall was modified to limit concrete pouring near the river. The
headwall and wingwalls were replaced with a metal trash rack that flares out to catch the
adjacent side slopes. 24" minus riprap is used to protect the inlet and limit turbulence (see
plans). Two geogrid layers will be constructed on top of the culvert/riprap and vegetated.
The trash rack is shop or field assembled and bolted to the flush end of the culvert.
Consider non -corrosive materials.
4) Several suppliers offer pre -cast reinforced concrete box culverts. Copies of the
drawings should be made available to ensure that the culvert will fit into the vault.
5) The sluice gate, Armtec Limited type "50-10" can be obtained locally from Beaver
Inc, Bellevue WA, 425 398 6678 (Ask for Nick). Estimated price for delivery of all
components is $5000. Check to see if they will field install. Waterman series 3000
sluice gate can be used as well.
6) The 96" diameter vault can be obtained from Shope Concrete products of Puyallup
WA 253 848 1551 (Ask for Gary Patee). Estimated price for delivery of all components
is $6,150. The vault needs to be watertight once assembled and pipes are fit. A truck and
crane capable of lifting 10 tons is required for offload and placement of heaviest sections.
7) The 48" CMP was replaced with 48" CPEP to eliminate corrosion risk.
8) All drainage elements will have to have sharp protrusions sanded down or filled.
9) Using a layer of riprap on top of the placed logs will increase the factor of safety and
reduce the need for logs along the toe for erosion protection. Rebar should be used to
hold some members to anchor members (see plans). Duckbill anchors are capable
resisting a force of 3000 lbs. These should be used to hold wood in place where noted in
the plans. Pre -excavate scour holes 1.5-2.5 ft below structure to form pools.
Z. Corum 7/l/2004 4
Cedar Spawning Channel
Final Design Documentation
11) Drive rebar through the hoop end of the duckbill anchor to secure the log. The
commercially available anchors go for around $25/ea.
12) Alcoves should be excavated to place toppled trees with rootball attached (see detail
in plans). It is likely that these trees will have to be limbed or topped to ensure safety to
workers and to prevent unwanted damage to nearby areas.
10) Work closely with the hydraulic engineer and geotech engineer if there are questions
about any of the above in the field. It is clear that the plans are a starting point and "field
fitting" will be required. The plans are prepared with consideration for a wide range of
flow conditions, however failure to install the drainage system properly may jeopardize
the success of the project.
Contingency Planning
Construction planning should consider foregoing the inlet system if sustained GW
pumping/flow rates during excavation exceed 3 cfs at or above the design streambed
elevation. If this contingency is considered it may be more appropriate to excavate the
channel before installing the inlet works. The inlet works trench can be partially
excavated and backfilled with riprap to capture more groundwater. In this event someone
will have to supply estimates of how much water the alternate system may be able to
supply. This should only be considered in the event that GW is plentiful enough to
remove fines that may deposit at the downstream end of the channel and all parties agree
that it is an appropriate action.
REFERENCED FIGURES ON FOLLOWING PAGES
ZPC
Z. Corum 7/ 1 /2004 5
Cedar Spawning Channel
Final Design Documentation
Cedar River - PL99 - Based on 2000 FIS Plan: 1x 48 RCB high flow full open 2/21/2003
�~-a
cea etbel3"'vinn ch i
iiwer %F n - -
c
--
a
Legend
_
�-
,�_ �i
�.. -
Feb lttth flow
__ __ ___ _ _
_ _./_, _
R
-__
Ground
___ _ _ ............
__ :_ _ ....__ __ `--_ __.
— - LOB —
__ _ t_
r
Ur,��
----
___
______
-X �__ ___ _ _____,_
U
Right Levee
______ _ _ __ __ ___ _ _ ___
___ _ /
P
_____
OWS Feb lath lbw
--
e
___
�lllll
,
_ __ _ _ �_ _ ___ ___ _ __ __•�_a_
___ ___ ___---------- __�.u___ ____ _,__
d
L _i , _ i
i
S_
,
,
.$e
0 260 400 600 860 1000 1200
Ma Channel Distance (R)
Figure 1: Feb 181h Water level vs. Modeled River and Channel Water Level
Cedar River- PL99 - Based on 2000 FIS Plan: I 48 RCB high flow full open 2/21/2003
SpewMna Ch In
Lpend
VN ChN Summer Critica�
VN ChM 90%exceedence
VN& AVG CRITICAL._
VW CM AVG LOW NORMAL
-------------- --- VN CMI AVG HIGH NOR
-{-�--�-- --- --- -- --- --- -- - - ---- --- - - -- -- --- -- -- --- ------
-Ili---f-- --- --- -- --- --- -- --- ------ --- --- -- -- --- -- --- --- -- --- --- -- --- -------
-------------------- r
--_'
,
,
-----------•----- - -- --- --
,
------,- - --'-------,---r -- ---,-- -----=------ - ------------ -
,
260 4og _ 600 800 1t108 1200
Main Channel Ol nce (n)
Figure 2: Velocities in Spawning Channel: Low -Flow to Normal Spawning Conditions
4x4 box, full open
Z. Corum 7/1/2004 6
Cedar Spawning Channel
Final Design Documentation
Cedar River - PL99 - Based on 2000 FIS Plan: 1x 48 RCS high flow full open 2/212003
sp.ae.rp cn In
Max CN Dplh AVG HIGH NORMAL
Max CN Dpth AVG LOW NORMAL
i
Max CH Dplh 90%excaadanca
VI
-
Max CN Opih AVG CRITICAL
r Max CN OPTh Summer CrWal
i
7
g
`u
iS
{- ------- ------------ - - -- -------- --- --- ------------
os t'
I,
0.0
0 200 40o eo0 e0o 1000 1200
Main Channel Distance (h)
Figure 3: Depths in Spawning Channel: Low -Flow to Normal Spawning Conditions, 4x4
box, full open
Cedar River - PL99 - Based on 2000 FIS Plan: 1) 1X36 56.15ES 9/4/2002 2) early 0.5 2/6/2003
Spawning Cn In
__,._,___,__I_._,_,_. -. __,. Legend
________________•___�___•__.___�__•__�___•__,______ ,______._ _ .__•___•__.__. Vsl Chnl 1-70ct Norma1-1X3856.75E5
Vel Chn11-7 Oct Normal - sany 0.5
- - - - X3
al Chnl 76 30 Sept Norm-1X3856.i5ES
i
Vol hM 1 -30 Sept o -My 0.5
, ,
4
l
260 am 060 800 --- _.. 1000 1200
Main CW" DlMrwa 0a
�,.,. 2 f�
�f
Figure 4: Compares spawning velocities at low flow for 4x4 RCB (full open) and 3x3
RCB (0.5 ft open). Note no change in channel hydraulics.
Z. Corum 7/l/2004 7
Cedar Spawning Channel
Final Design Documentation
Cedar River - PL99 - Based on 2000 FIS Plan: 1) 1%36 56.15ES 9/42002 2) early 0.5 2/62003
----------- ___________„___.__,-__._. .-_.�___.__
Oe
___ Ye CN N 1-7 Od Nenrel -aaal� O.a
___.__,___-I__.__ �__.__._._._..._-.__,__....
__.__ - . __._ _ we CM 0pla 1.7 Od Nand. 11l305a.15Ea
i
- - Me CN OON fa J08ao1 Norm -ae!/ O.a
3.0 ____________________1___.__.___.
we woga is ao 9.0 NMm•1 los 5a15E8
------------'"'------------------------- --- - - -- --_---- - -- - ----- ----- ------
ao — --- ---
-- - ----- - -- -- -- - --
. -- -" ---- ----- ----- -- -- --
is
------------------
GO
o no 40o eao aao i000?2(10
Mwn Lhannal One— (a)
Figure 5: Compares spawning depths at low flow for 4x4 RCB (full open) and 30 RCB
(0.5 ft open). Note no change in channel hydraulics.
Z. Corum 7/l/2004 8
Cedar Spawning Channel
Final Design Documentation
w
N
N
Z_
Q
C7
W
J
m
Q
F
FIGURE 6: BANK STABLE GRAIN SIZE VS FLOW AND ROUGHNESS
8.00n-
G�5
O
OO
N
0
e+Gooa\
P�
FLOW OF INTEREST
Z. Corum 7/1/2004 9
Cedar Spawning Channel
Final Design Documentation
FIGURE 7: CHANNEL STABLE GRAIN SIZE VS DISCHARGE AND ROUGHNESS
Z. Corum 7/l/2004 10
Cedar Spawning Channel
Final Design Documentation
Appendix A
RATING CURVES FOR RECOMMENDED MONITORING LOCATIONS
1) Use rod/level survey based on existing ground control network
2) Get USGS hourly flow data at the City of Renton from internet (data should be for 3
hours after survey)
3) Identify flow on rating curve
4) Plot water level on rating curve and note date
Cedar River- PL99 - Based on 2000 FIS Plan: ix 48 RCB high flow full open 2 i2003
Copy CX 14—COE OVERBANK XS N
54.5 - -- -- -- -- -- -- -- -- -- - - - -- -- -- -- Legend
54.0 -- -- -- -- -- -- — --
53.5---------------- -- - - -- ----- - -- - -- - -- _
> r -- T r - -
w 53.0 - -- -- - -- --- - -- -- - - - -- -- -
52.5 -_ -_ -- - -- -_ - ----- - - -- - - -- -- -- -- -- - --
52.0 _ - - - - -- -- - - - -- ......
.i -- -- ;- '-- T_
0 100 200 300 400 560
Q Total (cfs)
Rating curve 50 ft upstream of Outlet of Spawning Channel (120 ft E/NE from SC 2)
(approx Sta. 0+80)
Z. Corum 7/1/2004 11
Cedar Spawning Channel
Final Design Documentation
Cedar River - PL99 - Based on 2000 FIS Plan: ix 48 RCB high flow full open 2/21/2003
Copy CX 14-COE OVERBANK XS L
55.0 -- ....... - -- --- -- ---- - Legend
_
54.5 _ __ __ __ __ __ __ __ __ __ _ _ W.S. Elev
,.
54.0 =_ __ ____ __ ......
- -- -- -- - ----- __ -- --- -----
,
_ . . . . . . ---- -- ------ - -- -- -
> 53.5 ____ __ __ __ _ _ __ __ _ __ ____
---- --- -
w =_ -- - ---__ ___________--- __
053.0 - ---- ---- -- -- --- -- -- --- --- - - - --
52.5
52.0 __ ________ __------_--__-______ _ _
51.5
- - -,--- ;-- - - - - -- -- --- - ---
0 100 200 300 400 500
Q Total (cfs)
Rating curve 40 ft d/s from SC 3 and Sta 2+20 (Lower bore hole)
Cedar River- PL99 - Based on 2000 FIS Plan: 1x 48 RCB high flow full open 2/21/2003
Copy CX 15-COE OVERBANK XS H
55.5 - -- -- - - _ __ __ ___ _____ __ __ __ __ Legend
55.0 - -- - - -- - ---- ---- -- -- - --- __ ------ __ __ _- W.S. Elev
54.5 ........ - - -- - -- ----------- ------ -- -- -- -- -----
> -- - -- ---- - - -
w 54.0 - -- -- -- . -- --
-- --
5......... -- - --
3.5
53.0 - - - - --------- -- -- - _
52.5
0 100 200 300 400 500
Q Total (cfs)
Rating curve near Sta 5+20, 20 ft d/s from SC 5 and (Middle bore hole)
Cedar River- PL99 - Based on 2000 FIS Plan: 1x 48 RCB high flow full open 2/21/2003
COPY CX 17-COE OVERBANK XS A
Legend
57.0 -- - --- - - - - ...............
W.S. Elev
56.8-------- - - - - -
56.6 - - - -
w 56.4 . . . - ---
532 - -
56.0 -- ------------------ ------ - --
--------------------------
55.8 _
_ -
55.6- --
Z. C 100 200 300/1i a�4 00
12
Q Total (cfs)
Cedar Spawning Channel
Final Design Documentation
Rating curve at SC 7, Sta. 9+75, and 30 ft d/s from staff gage, just upstream of head of
spawning channel
Cedar River- PL99 - Based on 2000 FIS Plan: ix 48 RCB high flow full open 2212003
copy of 357.08
-- - - - -- -- -- -- - Legend
- -- -- -
58.0 W.S. Elev
57.5 _ - -
w. . - - --
56.5 - -- -
------------
56.0
0 100 200 300 400 500
Q Total (cfs)
Rating Curve at Inlet Culvert of Spawning Channel 70 ft u/s from SC 8
Z. Corum 7/1/2004 13
;1,
M
O
O
N
r
CM
O
>+
f0
O
O
M I
O
O
N
O
LLI
m
N
U v cr)
Q o
2
W
LL
Q
O
N
C�
M
O
O
N
� a
C a
0,.oa)�Y
0 C
J (Dcu
m
--- ---
- - - - -
- -t
-t
--- ----
--- ----
-- t -
i
I
---- ---------
-----•----•----`----
t
I
I
I
0
i
I
i
�
J
I
i
--- --- - -- - --
-
O 6) W
D
c
0
o U)
O
v
0
O
O
TO
O> co J 3 J 0 oA .9 S LO
C
(u) UOIIBA913
Cedar River - PL99 - Based on 2000 FIS Plan: 1x 48 RCB high flow full open 2/21/2003
8djr Rt�r �I inlet C -
n
t - C e l ar iveb p nin Ch--I -- - Legend
1 Feb 18th
N a uV flc
Ground
-- � � �- f - ' Gr nd
Modeled river - , - - - -
!; i _y LOB I
water level for Feb
18`h flow --
s Right Levee
65 - , --
.
r
Field measured
water level at
Theoretical approximate river G
y ,
spawning channel .......... ; station
d
water level for ; a
this flow -
------ - - - --- --
o i
LU
y i
-
Proposed ------ ,-
Invertof '---=----t--�---�--- �----'----�-- �---�---=---=------ - --- -------------- --- �---;---�--- �---
spawning `
- -------- --
channel _
S9--- '----; --- -- T ------ - -i- -- i i
This is point where the spawning
- -- --- -
channel and river model split Since
Cedar they b n profile, and
th both are shown i fil
si river ; ; , channel lengths are a little
the
thalweg different, they don't line up exactly
...
si ,
0 200 400 600 800 1000 1200
Main Channel Distance (ft)
Elevation (ft, NAVD '88)
� 1 1
N C.0 CI) A 4�- Cn Cn O __ O �I --ACO00 U) CO CD CD-'
Cn O Cn O Cn CD Cn O =(n O Cn O Cn O M O Cn O
A
O
O
Cn
Cn
O
O
O
Cn
O
O
N
O
Cn
0
0
rn
W N
O Cn
O
N
N
M
O
O
N
C.J
Cn
O
O
N
Cn
Cn
O
O
N
CA
Cn
O
O
N
J
Cn
O
O
N
OD
Cn
0
O
0
W
W
C_
CD
T\
W
r
W
W
z
W
O
-o
n
O
N
m
m
CD
W
Q-
W
D
a
n m
o<.
m
0Q
C
Q i
O
� !v
G
O
(D n
CD
�
w
w
W
�
M
�
�.
C)
S
G
Riverview Park bridge
Cedar River Trail
(old railroad bridge)
I
I I
I I I
I
I
U) U) W U) M
�,
I (D o 0 0 o z
U) 0
CD
0
r
0
0
7
Drporate Boundary
Highway 169 and downstream
pedestrian bridge
—� Corporate Boundary
Golf Course
pedestrian bridge
Location of Flow Split
Through Golf Course
\
\
\\ Corporate
\ Boundary
1 \
1 I
11
\
149th Avenue S.E.
CD
1'
i
0
-
,
• } _ L _ _ 1 _ _ _
1r'
1 _ _ _ _ 1
...
-- -- -- - - -
_ 1 _ _
.........
1
1
1 I I I
t 1
1 I
I I I
1 I
t
_ _ - _
_ _ _ - 1 _ _ _ _ L _ _ _ _ _ _ _ _ _ _ _ _ f _ _ _ - _ _ - _ _
I
-
_
1 r
- L - - -
L - - -
1
i i
1 I
I I
1 I I
L_ _ _ _ L _ .. _ _ i _ _ _ _ _ _ L _ - _ - 1
I I I
1
I I 1 1
I
--�
1
1 t I 1
- _ _ _ 1 - _ _ -
I 1
1 I 1
1 1
1 I 1 I
I
1 1 1 1
i
I I 1
1 L
_ _ _ _
_ _ _ _ 1
I 1 I I
I 1 t 1
1 I
- _ _ _ _
_ _ _ _ 1 _ _ _ _ 1 - _ _ _ L _ _ _ _ L _ _ _ _
I 1 1 I
I r
I I
1 I 1
I
t
I
I I 1
, 1
I
1
1
_ _ _
_ _ _ _ L _ _ _ _ 1 - -
_ _ _ _ L - _ _ _ L _ _ _ _ - _ _ _ _ _ _
1 i
1 1
1 1 1 1
1
I i 1
I
1
1
I I I
1 I 1
1 1 I I I
1
I t
I 1
1 I 1
F."
olleA913
O 4ip S �
O
O
O r�
to
c
0
V /
FE
MM10
ol,
Chapter 7
Rehabilitating Off -channel Habitat
D. Brent Lister, Rheal J. Finnigan'
INTRODUCTION
Development and rehabilitation of off -channel areas is receiving increased attention as a
practical means of restoring salmonid fish habitat. This approach is particularly attractive for high-
energy coastal streams where flow extremes and channel instability often make it impractical to
attempt rehabilitation of the main channel. Off -channel habitat rehabilitation can also be a
worthwhile option in the interior, where winter conditions may be severe in the main channel.
However, only certain species and life stages of salmonids utilize off -channel areas. The benefits of
rehabilitating off -channel habitat will therefore depend on the presence of an appropriate physical
environment and the fish species and life stages adapted to that environment.
The Nature of Off -channel Habitat
Off -channel habitats, including overflow, groundwater, and wall -base channels (Fig. 7-1), are
created by long-term processes of alluvial deposition, channel migration, and changes in stream bed
elevation (Kellerhals and Church 1989). While overflow channels will carry flow only during flood
events, groundwater channels can occur in a range of floodplain situations that to varying degrees
are removed from the active channel. Wall -base channels, whether groundwater or surface fed,
occupy higher portions of the floodplain or a terrace outside the influence of active channel
processes (Peterson and Reid 1984). Each of these situations generally develops in a meander
channel that has been abandoned by the main stream as it migrates across the valley. Groundwater
and wall -base environments can be expected to provide relatively stable flows, a moderated
temperature regime (Fig. 7-2) and, in some cases, a complex of channel and pond habitats. Another
common feature of these habitats is their modification by beaver (Castor canadensis) activity,
described in detail in Chapter 15 of this publication.
Salmonid Use of Off -channel Habitats
. Among the salmon species, chum and coho are most commonly associated with off -channel habitats.
These species are apparently attracted to sites fed largely by groundwater. Late -run chum stocks,
throughout their range, have been noted to spawn in groundwater -fed channels or seepage areas (Salo
1991). Coho spawn in groundwater channels to some extent (Sheng et al. 1990), but most coho spawning
occurs in relatively small surface -fed streams (Sandercock 1991). Coho juveniles, on the other hand, make
widespread use of off -channel habitats, often gaining access to small stream and pond environments that
are either inaccessible to adult coho or unsuitable for spawning (Peterson 1982a; Brown and Hartman
1988). In coastal streams, juvenile coho move into off -channel areas as post -emergent fry during spring
and early summer, or during fall, in advance of the larger mainstem freshet events (Fig. 7-3). While coastal
coho juveniles appear to use off -channel habitat mainly for overwintering, studies suggest that interior
coho characteristically move to off -channel ponds after fry emergence in spring, and may remain
D.B. Lister & Associates Ltd.. Box 2139, Sardis Stn Main, Chilliwack, BC V2R lA5
British Columbia Conservation Foundation, Suite 206 - 17564 - 56A Avenue, Surrey, BC V3S 1G3
Rehabilitating Off -channel Habitat
a
4
i
0
7-1
there for the entire I or 2 years of freshwater rearing (Bustard 1986: Swales and Levings 1989). A fall
movement of coho juveniles from mainstem to off -channel sites for overwintering has also been observed
in the interior population at Coldwater River (Beniston et al. 1988). Juvenile coho are also known to
migrate considerable distances downstream from summer rearing habitat to off -channel sites for
overwintering. Coho marked in headwater rearing areas have been recaptured at overwintering sites up to
33 km downstream in the Clearwater River, Washington (Peterson 1982a) and 52 km downstream in the
Chilliwack River, British Columbia (Fedorenko and Cook 1982).
Figure 7-1. Types of off -channel habitat (adapted from Peterson and Reid 1984).
14
U 12
`. 10
m $L
a�
6
4
2
0
Beaver Pond
i
River
I,
J F M A M J J A S O N D
Figure 7-2. Water temperature regime of an off -channel beaver pond and river mainstem,
Coldwater River, B.C. (Swales and Levings 1989).
7-2 Rehabilitating Off -channel Habitat
Figure 7-3. Seasonal pattern of juvenile coho salmon recruitment to off -channel ponds
(Peterson and Reid 1984) and relationship of coho recruitment to daily discharge (Peterson
1982a) Clearwater River, Washington.
Other salmon species have not been observed to utilize off -channel habitat to a significant
extent in British Columbia. Pink salmon are not reported to utilize off -channel areas (Heard 1991).
Chinook salmon do not spawn in off -channel habitat, but interior stocks make some use of off -
channel ponds and side channels, often associated with tributaries, for juvenile rearing and
overwintering (Anon. 1987; Swales and Levings 1989). Sockeye salmon have been reported to
spawn in off -channel spring areas in Kamchatka (Burgner 1991) and Alaska (Mathisen 1962), and
on groundwater -fed lake beaches (Burgner 1991). Though off -channel habitat use by sockeye is
apparently not common in British Columbia, groundwater channels developed at Adams River and
Kitsumkalum Lake in the interior have been used extensively by sockeye for spawning and, in the
former case, for juvenile rearing over summer (M. Foy and M. Sheng, Department of Fisheries and
Oceans, Vancouver, pers. comm.). As the capacity of nursery lakes to rear juveniles frequently
limits sockeye production (Burgner 1991), off -channel habitat rehabilitation for sockeye will be
restricted to situations where spawning habitat is considered the limiting factor.
Kehabilitating Off -channel Habitat 7-3
Of the trout species. coastal cutthroat (O. clarki clarki) are most likely to be found in off-
channel environments. Adult and juvenile coastal cutthroat can be expected to cohabit many off -
channel sites with juvenile coho (Cederholm and Scarlett 1982; Hartman and Brown 1987).
Seasonal movements between main -channel and off -channel sites enable cutthroat to utilize small
tributaries for spawning, juvenile rearing or ovenvintering. In these situations, cutthroat may be
partially segregated from coho juveniles by their preference for more permanently wetted
environments with higher flow (Hartman and Brown 1987). Interior stocks of west slope cutthroat
trout (O. clarki leivisi) have generally not been documented in off -channel habitat, but in the Elk
River and its Fording River tributary in southeastern British Columbia, yearling and older cutthroat
do overwinter in groundwater -fed off -channel ponds (G. Oliver. Interior Reforestation, Cranbrook;
D.B. Lister & Associates 1980. pers. comm.).
Steelhead trout do not commonly spawn in off -channel streams, and juvenile steelhead apparently
use such habitats to a much smaller extent than coastal cutthroat. Steelhead are not abundant in off -
channel ponds (Cederholm and Scarlett 1982; Swales and Levings 1989). In coastal streams steelhead
underyearlings and parr prefer small surface -fed tributaries to groundwater environments for rearing
and overwintering (Cederholm and Scarlett 1982). Some coastal groundwater channels do, however,
overwinter significant numbers of parr and pre -smolt steelhead (King and Young 1986) and a
groundwater channel at Deadman River, in the British Columbia interior, attracted significant
numbers of underyearling steelhead for rearing and ovenvintering (Sheng et al. 1990). Adequate
velocity and habitat diversity were likely requisites for juvenile steelhead use of these sites. Though
off -channel habitat use by resident rainbow trout has not been reported, underyearling rainbow are
known to rear in seepage -fed ponds and side channels when these habitats are available (B. Chan,
Ministry of Environment. Lands and Parks, Kamloops, pers. comm.).
The stream -dwelling species of char, Dolly Varden and bull trout, have not been commonly
observed in off -channel habitats. Studies of coastal streams in southern British Columbia and
Washington indicate little use of off -channel areas by Dolly Varden. This species may, however, be
more abundant in off -channel habitats of northern streams. In the Skeena River system, juvenile
Dolly Varden utilize off -channel ponds in the presence of juvenile coho and stickleback
(Gasterosteus acidearus) (David Bustard and Associates 1993), and have been observed to
overwinter in a groundwater channel at Kitwanga River (M. Foy, Department of Fisheries and
Oceans, Vancouver. pers. comm.). Dolly Varden are also relatively abundant in southeast Alaska
streams, where thev make considerable use of beaver ponds (Bryant 1984). Though bull trout are
suspected to spawn to some extent in groundwater -fed areas, studies to date have not revealed the
use of off -channel habitat by this species (Baxter and McPhail 1996).
Use of off -channel habitat by Arctic grayling has not been well documented. Grayling appear
likely to utilize off -channel habitats, as they have been noted to frequent marginal low velocity areas,
side channels and backwaten, as underyearlings. and to distribute into a variety of habitats, including
spring -fed areas, for feeding_ and overwintering in the juvenile to adult stages (Northcote 1993).
Evolution of Off -channel Habitat Rehabilitation
Early development of .eft -channel habitat for anadromous salmonids on the Pacific coast of
North America involved construction of surface -fed side channels, primarily for spawning
sockeye, pink and chum salmon. during the 1950's and 1960's. The first artificial spawning
channel was constructed in 10553 at Jones Creek on the lower Fraser River (Hourston and
MacKinnon 1956). Follo in_, that project. 14 artificial spawning channels were built in British
7.4 Rehabilitating Off -channel Habitat
Columbia, and another 7 channels went into operation in the northwest United States. Certain
spawning channels produced significant numbers of adult salmon to the commercial fishery (Fraser
et al. 1983; West and Mason 1987), but the relatively high capital and maintenance costs,
particularly sediment problems in non -lake -fed situations, tended to make spawning channels
Generally less attractive than other salmonid enhancement options.
In the mid-1970's, the Department of Fisheries and Oceans (DFO) began to examine the
potential for increasing chum salmon production in British Columbia by restoring groundwater -fed
spawning areas (Marshall 1986). Groundwater spawning channels were relatively inexpensive to
build and essentially free of the maintenance problems associated with many surface -fed spawning
channels. Also at this time, research on seasonal habitat and movements of juvenile coho was
revealing the importance of off -channel streams and ponds for rearing and overwintering (Bustard
and Narver 1975; Peterson 1982a). Development of off -channel habitat for chum salmon, and to a
lesser extent for coho, continued in British Columbia, Washington and Alaska during the 1980's
(Bachen 1984; Bonnell 1991; Cowan 1991). Evaluation of the groundwater chum spawning
channels indicated that these sites also provided spawning and juvenile nursery habitat for coho
(Sheng et al. 1990). The quarry rock (riprap) used to protect the banks of these channels from
erosion by chum spawners inadvertently provided excellent cover for coho juveniles, which also
benefited from stable flows and temperatures as well as an abundance of food consisting of aquatic
insects supplemented in winter by chum salmon carcasses, alevins and emergent fry. Recent
research suggests that salmon carcasses are also a significant source of nutrients to the food chain
supporting stream -rearing species such as coho, cutthroat and steelhead trout (Bilby et al. 1996). In
the 1990's, off -channel habitat projects have therefore tended to emphasize a multi -species
approach that includes, for example, creation of spawning areas for species such as chum salmon
with little or no stream -rearing requirement, as well as spawning and rearing habitat for coho,
cutthroat or steelhead, which have more complex freshwater habitat needs.
ASSESSING THE NEED AND THE OPTIONS
Effective off -channel habitat rehabilitation is best achieved from a thorough assessment of
conditions in the watershed of interest. An overview fish habitat assessment, including review of
existing information on fish resources and watershed conditions, will normally be required. It
should examine evidence for declines in fish stocks and the linkage between fish population
changes and habitat changes. If existing information is not adequate to assess the need for a project,
it may be necessary to obtain detailed site -specific habitat information through field assessment.
Guidelines for planning watershed restoration work, and fish habitat rehabilitation in particular, are
presented in Watershed Restoration (WR) Technical Circular No. 1 (Johnston and Moore 1995).
Procedures for overview and detailed field assessments are described in WR Technical Circular No.
8 (Johnston and Slaney 1996).
If initial assessment confirms the need for fish habitat rehabilitation measures, project planners
should decide on the appropriate strategy and whether or not it will involve main channel or off -
channel habitat. Off -channel habitat rehabilitation will be favoured on relatively large streams, and
in situations where the main channel is too unstable for in -channel habitat rehabilitation measures.
Selection of the off -channel option also depends, of course, on the presence of salmonid species
that will take advantage of habitat created outside the main channel. A process for selecting
between main channel and off -channel habitat options is presented by a flow chart in Chapter 8 of
this publication.
Rehabilitating Off -channel Habitat . .r 7-5
PROJECT PLANNING AND SELECTION
The process for planning. selecting and investigating suitable off -channel habitat projects is
described in the follo%ving sections and summarized in Figure 7-4.
Target Species and `eater Sources
It is important to confirm at the outset the fish species and life stages involved, and the type of
water source, surface runoff. lake or groundwater, that will be employed. One should
understand ho%% salmonid populations utilize the drainage for spawning, juvenile rearing, and
overwintering, as well as the role played by main channel, tributary and off -channel habitats. Do
any stocks spa%\ n,in lake -fed or groundwater -fed situations that offer more benign winter conditions
(moderate temperature regime. stable flows. and sediment -free water) than surface runoff streams?
The ans«ers to these questions will lead to decisions on the species, water source and habitats,
which will be the tocuc of the project.
7-6
Target:pi
lies
life se.
V
Water quagi 'I
temperature n e Watel-1,
it source',
Available", - J
flow
site .
Site
acces selectio' rM,
Land us
constraints Swta a land
od
Construction
mat6rials
Water qua
quality m niton g
Channel l po„ d Detailed s e
alignment; InVeSt1g8f1"
" fll
Topographic `.,....
survev T� Excaiation
Figure 7-4. Principal elements of project planning, selection and investigation.
Rehabilitating Off -channel Habitat
Identifying Prospective Sites
The nature of the primary water source will guide site identification. If the project is to utilize
surface water, it will need to be located on a bench or in the upper floodplain. Abandoned or cutoff
river channels offer natural depressions, which can sometimes be used to minimize excavation
quantities. A surface water source must be reliable. It should provide a year-round flow that does
not freeze in winter, and it should be relatively free of tine sands and silt, bedload (sand and
(Yravel), and wood debris. In colder climates, lakes are an attractive water source because their
moderating effect on winter temperature tends to ensure ice -free operation. To avoid sediment
bedload, intakes should be located on the outside of a stream bend. Relatively stable sections of
stream, such as a lake outlet reach, are the most favourable situations.
Groundwater quantities required for off -channel habitat development are most likely to be
available on a relatively large stream, over 300 km' in watershed area, with a wide valley and
extensive alluvial floodplain. Relic meander channels and areas of standing groundwater are
commonly candidates for detailed investigation. Identification of prospective sites will be aided by
inspection of topographic maps and recent, larger -scale aerial photographs. The latter are especially
useful for identifying abandoned river channels, which are revealed by the presence of bands of
deciduous vegetation, often less mature than surrounding vegetation because of more recent
exposure to flood flows (Fig. 7-5). A helicopter survey can be of considerable assistance. Under
winter conditions, groundwater sources can often be identified from the air by the absence of snow
and ice cover due to elevated ground temperature. Promising sites should also be inspected on the
ground to confirm their suitability for development.
Exposure to Flooding
It is important to minimize intrusion of off -channel habitat rehabilitation projects onto the
active river floodplain. Adoption of such a philosophy will avoid impact on currently functioning
stream habitat.
Most projects will be located in and around abandoned flood channels that may become active
during high flow events. This raises the question: should projects be designed to withstand extreme
floods or should they be located and designed to accommodate overtopping by floods in order to
minimize their impact on the river regime? Experience suggests that it is better to design for
overtopping by a moderate flood event with, for example, a 1 in 30-year recurrence interval. Off -
channel projects located and designed with this safe fail philosophy should require only minor
maintenance and repairs after an extreme flood. This approach tends to reduce floodplain impact
and flood protection costs.
One consideration in assessing exposure to flooding is the benefit of providing modest site
protection with a berm of granular material excavated in the course of developing a channel or
pond. The excavated material is normally used to create a low berm between a side channel and the
main river channel (see Project Design Concepts section).
It is not uncommon for a major road, railway, or flood protection dyke to cut off an active flood
channel or a channel that has been abandoned by the main river. Restoration of such habitat, using
the flood protection attributes of the existing facility, is a good strategy (see Fig. 7-19). It will be a
sound approach wherever the road or dyke is an integral part of the infrastructure and will therefore
be adequately maintained.
Rehabilitating Off -channel Habitat 7-7
Figure 7-5. Abandoned channels on the upper Pitt River floodplain.
As flooding risk varies greatly from site to site, that risk should be assessed by review of aerial
photographs and stream discharge data, by consultation with a fluvial geomorphologist, and by
contacting individuals who have experience in the project area.
Site Access
Road access will be needed for most off -channel projects. The access must be adequate for heavy
construction equipment and periodic maintenance or monitoring activities. Location of an access road
should minimize impact on the f1codplain, using the flood protection berm where possible. Cost and
feasibility of creating new access or upgrading existing road access are important to project feasibility.
7-8
Rehabilitating Off -channel Habitat
Land Tenure
Land tenure and possible restrictions on land use must also be considered. As stream habitat
rehabilitation projects will generally involve Crown forest land, it will likely be expedient to contact
the appropriate regional or district office of the Ministry of Forests. The B.C. Lands Branch can
also provide information on status, responsible management agency, and land use restrictions
pertaining to a given parcel. Through its six regional offices, the Lands Branch operates the
computerized Crown Land Registry for the province. In some instances, access to Crown land may
have to be across private land. Information on private land ownership can be obtained from one of
six regional land title offices. To initiate a title search requires specific identifying information, i.e.,
legal description or parcel identifier number, which must be obtained from the appropriate
municipal or district office, or the B.C. Assessment Authority. Information on land use restrictions
pertaining to private land may be gained from the local district or municipal office.
Construction Equipment and Materials
Availability of suitable construction equipment, particularly excavators, and off -site
construction materials are also factors to consider. For stream channel projects, rock bank
protection material, either angular quarry rock (riprap) or natural boulders, is normally employed
for lining banks to prevent erosion. Other common requirements are suitable gravel for an access
road or large wood debris for instream cover.
Operational Considerations
Some off -channel projects, particularly those with surface intakes, must be regularly inspected
and maintained to ensure reliability of the water supply. The primary task will be to remove
accumulated debris from facilities such as trash racks, water control structures, culverts and fish
enumeration fences and traps. Periodic maintenance work, involving heavy machinery, will also be
required. Availability of organizations and personnel to perform these operational functions, such
as stewardship groups or a local contractor, must therefore be taken into account.
Detailed Site Investigation
Sites that hold promise for off -channel habitat development should receive additional
investigation to determine (1) the nature and suitability of alluvial materials, (2) the quantity and
quality of surface or groundwater sources, (3) excavation quantities, (4) amount and type of material,
if any, to be moved to or away from the site, (5) the best stream channel alignment and/or pond limits,
and (6) flood protection requirements. A site topographic survey will be required to provide the
physical information and enable the channel or pond layout to be optimized. This topographic survey,
including establishment of benchmarks for reference, should provide ground elevations relative to
adjacent stream elevations and high water marks. In addition to the survey, a number of test holes
should be excavated to determine the nature and layering of substrate materials in the area to be
excavated. Another practical way of testing depth of overburden above the gravel layer, an important
consideration for a groundwater channel, is to probe the ground with a 6-12 mm diameter steel rod.
For a groundwater channel, standpipes should be installed in the test hole pits to enable
monitoring of water table level as well as groundwater temperature and dissolved oxygen.
Excessive levels of chemical constituents such as iron and hydrogen sulfide should also be
identified by analysis of groundwater samples. It is desirable to monitor groundwater levels and
quality at about monthly intervals over an extended period, ideally covering one annual cycle.
Rehabilitating Off -channel Habitat 7.9
The topographic survey should identify standpipe locations and should enable the investigator to
compare elevations of the water table and the adjacent stream.
Regulatory Agency Approval
It will be essential to obtain project approval from appropriate government agencies such as the
Ministry of Environment, Lands and Parks and the Department of Fisheries and Oceans (Johnston
and Moore 1995). For this purpose. the project proponent will have to prepare at least a conceptual
plan, describe the project's purpose, define its water requirements, and describe how the project
would affect existing fish and wildlife habitat values. The regulatory agency submission should
occur as soon as the proponent has adequate information and has decided that the project is likely
to be technically and economically viable. The regulatory agency approval process is described in
detail in Chapter I of this publication.
PROJECT DESIGN CONCEPTS
Site investigations provide the basis for an initial project layout and conceptual design, as well as
information needed to confirm feasibility. From this preliminary stage, the design should proceed to
development of detailed design drawings. Design drawings for an off -channel habitat project should
include a site plan showing ground elevation contours at 0.5 m intervals and locations of significant
trees to be avoided. The drawing should also include cross -sections that indicate elevations and the
extent of excavation at appropriate intervals along the project site. Plotting of channel or pond cross -
sections relative to the existing ground surface will enable the estimation of excavation quantities and
development of a plan for disposal of excavated material, which is commonly used for the flood
protection dyke and the access road. For spawning and rearing channels, it is also desirable to plot a
centerline profile showing the drop in bed elevation over the channel, and the bed elevation in the
adjacent river channel. Also needed are drawings of associated facilities such as channel intakes,
culverts (see Chapter 5) and beaver control features (see Chapter 15). Regulatory agency staff
typically require drawings such as a plan view, cross -sections and profile to facilitate project approval.
The following sections describe three principal concepts for off -channel habitat development: 0 )
the groundwater -fed spawning and rearing/overwintering channel; (2) the surface -fed spawning and
rearing/overwintering channel; and (3) the rearing and overwintering pond. A single project may
incorporate only one or all of these concepts, depending on available water supplies, • terrain
conditions, and the target fish species. It is desirable, however, to provide some habitat diversity within
each project to accommodate the requirements of several species and life stages. A multi -species
approach is appropriate in light of the fact that all species in a given watershed are likely to have been
affected by habitat impairment, and improvements made for one species may benefit another.
Groundwater -fed Channels
Suitable groundwater channel sites have a gravel substrate, relatively free of silts and tine
sands, with the water table near the ground surface. Such sites typically occur on streamside
benches with an overall gradient of 0.2 - 0.5% parallel to the main stream (Bonnell 1991). Lower
gradients tend to be associated with sandy materials and higher gradients with cobble -boulder
material in the alluvium. Water table elevation should be quite stable from season to season. The
groundwater should have a minimum dissolved oxygen concentration of 5 mg•L-' (Sowden and
Power 1985), and temperature and other water quality parameters should be within acceptable
ranges for the species and life stages of interest (Sigma Resource Consultants 1979; Piper et al.
7-10 Rehabilitating Off -channel Habitat
1982). Bjornn and Reiser( 1991) provide a useful review of the literature on water temperature and
dissolved oxygen requirements of salmonids.
Excavation of a groundwater channel causes drawdown of the water table in the vicinity of the
channel (Fig. 7-6). To ensure year-round flow, groundwater channels constructed by DFO in British
Columbia have generally been excavated to 0.9 - 1.2 m below the lowest level of the water table in
summer, based on water table monitoring prior to construction (Shen- et al. 1990). The bed of a
groundwater channel, which usually parallels the adjacent river. is typically about 1.5 m below the
bed of the river channel. Excavation requirements can be minimized by routing the new channel
along an abandoned meander channel to the extent possible. Depths of excavation for groundwater
channels are typically 1-2 m, but may be up to 3 m at the upper end.
In cross-section, groundwater channels are trapezoidal with bank slopes of 1.5 horizontal to I
vertical (Fig. 7-6). Bottom width and gradient are selected to achieve a water depth of 25-50 cm
(Bonnell 1991). Channel gradient is generally flat or very low, but it may range up to 0.5% in some
cases (Bonnell 1991; Cowan 1991). Bottom widths of excavated groundwater channels are
commonly 5-6 m. Where salmon spawning is expected, channel sides are armoured with a 50 cm
thick blanket of 20-50 cm diameter riprap or boulders (Fig. 7-7). Experience has shown that rock
armouring is essential to prevent serious bank erosion by salmon spawners. If the rock is installed
in a rough or non -uniform fashion at a thickness of at least two layers, the interstices in the rock
also provide good summer and winter cover for juvenile salmonids (Sheng et al. 1990). Wood
debris is generally added to the channel to provide high quality cover for salmonid juveniles and
for mature adults during spawning (see Chapter 8). Alcove -type ponds, connected to the channel,
are also being used to augment juvenile rearing and overwintering capability of groundwater
channels. Off -channel pond habitats are described later in this section.
The present custom is to use the native in -situ material for the substrate of groundwater
channels. Comparison of chum salmon survival in channels with substrates of either native gravel,
or artificially graded gravel, with smaller size fractions (<10 mm diameter) removed, has indicated
that Graded Gravel offers no advantages in terms of egg -to -fry survival or density of fry production
(Bonnell 1991). Large voids in the artificially graded gravel are thought to trap fine organics, which
could result in increased biological oxygen demand (Bonnell 1991). It has been shown
experimentally that coarse sand in the native gravel/sand matrix tends to filter out finer sands and
silts in the upper layers of the stream bed (Beschta and Jackson 1979). Coarse sands probably also
reduce the penetration of fine organic material into the spawning bed.
Excavated groundwater channels usually produce small discharge volumes (0.08-0.20 m3• sec-')
and low water velocities of 5-15 cm•s-' (Sheng et al. 1990; Cowan 1991). The minimum desirable
water depth for salmon spawning (25 cm) is maintained by the use of rock weirs installed at
intervals dictated by channel slope.
Protecting a groundwater channel from flood damage involves (1) adequate setback from the
active river channel, and (2) construction of a protective dyke. The upstream end of the channel
should be set back at least 30 m from the active channel to maintain an adequate buffer of protective
vegetation. A band of undisturbed vegetation should also be left between the channel and main river
alone the length of the channel. The flood protection dyke, constructed by sidecasting the
excavated alluvial material, is usually located between the groundwater channel and the main
channel (Fig. 7-6). These granular dykes cannot be expected to withstand the erosive forces of
extreme flows, but they can provide a measure of flood protection at moderately high flows. Both
the channel and dyke need to be located so as to minimize encroachment on the floodplain.
Rehabilitating Off -channel Habitat 7-11
A
Main channel
Rock weirs
Flood protection dyke using granular
material from channel excavation
Bottom of valley wall
i
Wood debris cover
throughout channel
Pool with large wood debris
(minimum 0.5 m water depth)
ni nni \iiE'\ni
Groundwater level before Channel water surface
channel excavation after construction Top of dyke
Original ground surface
------------
Gravel bed (native material) _� Rock weirs as required to maintain
minimum water depth of 25 cm
CHANNEL PROFILE
Undisturbed native vegetation
between dyke and main channel
Bank protected to water Topsoil dressing
surface with boulders on dyke slopes
or riprap (20 - 50 cm)
Native topsoil Dyke Wood debris cover Main
1�r_
1.5 channel
Organic debris removed
along dyke footprint
Gravel bed Water depth 25 - 50 cm
(native material)
TYPICAL CROSS-SECTION
Figure 7-6. Concept of a groundwater -fed spawning and rearing channel.
Rehabilitating Off -channel Habitat
7-12
Consideration should also be given to creating a small bench at least 1.5 m wide between the
channel and the dyke to provide a path for channel inspection, and to catch sediment that may be
eroded immediately after construction.
Figure 7-7. A broundwater channel just after construction, Usher's Channel,
Chilliwack River.
Surface -fed Channels
Side channels fed by stream surface water are being employed to an increasing extent for
rehabilitation of off -channel spawning and rearing habitat. This trend is occurring because suitable
sites for groundwater channels are being exhausted in some areas, and certain salmonid species, e.g.,
chinook and pink salmon and steelhead trout, are adapted to surface temperature regimes and
relatively high water velocities. The advantages of a surface water source lie in the availability of
large water volumes and flexibility in project siting. The disadvantage of a surface water source is
the possibility of significant sediment introduction and the need to expend considerable effort on
maintaining substrate quality. A surface -fed channel also has the cost of a river intake, which is not
a requirement for a groundwater channel.
Because sediment is such a critical concern, surface -fed channels should not be located on
systems, such as glacial streams, with high suspended sediment loads. Other siting prerequisites are
a bench of land with adequate flood protection, and a suitable water intake site. The bench on which
the channel is located should have a reasonably high overall slope of at least 0.5% parallel to the
main stream. Relatively high channel gradient is required to provide an adequate range of water
velocities and habitat types in the channel.
Intake Location and Design
Intake structures for projects that utilize surface water require special consideration. A minor
shift in the main channel during a flood event could have serious implications for the continued
operation of a surface intake structure. Hence, permanent intakes should be located on stable
reaches, which are unlikely to be affected by flood flows. In some locations it may be necessary to
armour the bank in the immediate vicinity of the intake to ensure bank stability under extreme
Rehabilitating Off -channel Habitat 7-13
flows. In addition, fill material around the intake structure must be adequately armoured with heavy
riprap or boulders to withstand high local velocities.
The intake for a surface -fed channel should be located on the outside of a bend in the main
stream, along the downstream half of the bend (Fig. 7-8). Secondary or lateral circulation in that
area tends to be toward the outside bank on the stream surface, and away from the outside bank
along the stream bottom. Such an intake location will minimize amounts of sands and gravels
entering the channel because sediment bedload tends to accumulate on the inside bend. To avoid
debris accumulation, it is also important to locate the intake where there is a positive downstream
flow along the bank. Localized backeddies should be avoided.
PLAN VIEW
Bedload deposition Surface debris near
outside bank
on inside bank
Spiralling flow pattern
SECTION A -A
Figure 7-8. Suitable channel intake location on outside of river bend.
Cross-section shows secondary circulation pattern.
The appropriate type of intake will depend on site conditions. Where the bank is stable on the
outside bend, a loa curtain wall can be employed to deflect floating log debris (Figs. 7-9 and 7-10).
The flow control valve, with a trash rack for smaller debris, is set back from the stream bank. A
settling pond, between the curtain wall and the trash rack, is an essential feature. The pond can be
j expected to settle out fine to coarse sand fractions, but silts and clays are likely to pass through to
the channel. The settling pond should be as large as site conditions permit. A smaller pond will have
to be cleaned out more frequently. It is also important to provide road access for heavy machinery
required to clean the pond.
7-14 Rehabilitatin; Of Habitat
Figure 7-9. Log curtain wall under construction, Shovelnose Creek, Squamish River.
Figure 7-10. Log curtain wall above intake, Anderson Creek Pond, Chilliwack River.
There are several intake designs that can be employed to control flow volume at the head of the
channel. The simple wood frame structure shown in Figure 7-11 will be adequate if upstream fish
passage, past the regulating valve, is not needed. If fish passage is required, then a larger slide -gate
control structure should be employed (Fig. 7-12). The latter structure will be most suited to
channels with relatively large flow volumes that might attract considerable numbers of migrating
salmon destined for spawning areas upstream of the channel.
Rehabilitating Off-eltamrel Habitat 7-15
AL
Water s
Trash rack
Native soil
Trash rack
Removable cover
CROSS - SECTION r
Valve chamber
(cover not shown)
,—Valve chamber. Laminated timber
R construction (treated wood)
Granular fill
r Pipeline
Control mechanism
for regulating valve
r Sub floor--7 ---A - frame support
ISOMETRIC VIEW
{ i
j
Figure 7-11. Wood frame intake structure, Or Creek project, Coquitlam River, B.C.
Figure 7-12. Concrete intake with box culverts and manually operated slide
gates, Centennial Channel, Chilliwack River.
Where sand and gravel bedload is a concern, it may be appropriate to locate the intake off the
stream bottom. A manifold -type intake is suited to this purpose, and has the added advantage of
7-16
Rehabilitating Off -channel Habitat
requiring little stream bank disturbance. It consists of a horizontal, heavy wall steel pipe with an
intake slot and trash bars facing upstream or downstream, parallel to the direction of flow (Fig. 7-
13). This type of intake must be located on a stable reach and at a location with a positive
downstream flow to carry floating debris past the intake. Backeddies should be avoided. To
minimize wood debris and bedload problems, water depth at low flow needs to be at least 60 cm
plus the pipe diameter.
Top of bank
i =1 I
Water supply for
fisheries project
Large rock or
concrete block
connection
flanged
e tion on
A
i Trash bars
End plate
Stream
E-7
flow
Vortex
plate
B
Openings between trash
End plate U bars not to exceed 10 cm
PL AN VIEW FROM ABOVE
' See Table 7-1 for dimensions 'a' and 'b'
30 cm from low water surface
Vortex plate Stiffener plates
Flow path
Stiffener
CROSS -SECTION A
Tapered slot in manifold
d6 .r
Centre trash bar
End plate (sides only)
END VIEW B
of stream bank
30 cm above stream bed Large rock or concrete
block to su000rt nine
VIEW C LOOKING UPSTREAM
Figure 7-13. Concept of a manifold -type intake, bolted to steel pipe buried in stream bank.
Rehabilitating Off -channel Habitat 7-17
i
Water is drawn into a manifold intake through a tapered slot in the pipe, with the wide end of
the slot at the offshore end (Fig. 7-13). Cross -sectional area of the tapered slot should approximate
the pipe cross-section. The narrow end of the slot should exceed 5 cm in width; the wide end
dimension is dictated by the required open area. Design parameters for a manifold intake are given
in Table 7-1.
Table 7-1. Design parameters for selected flow capacities of a manifold -type intake.
Design Pipe Pipe Width of Trash Rack
Flow Diameter Vortex Plate' Dimension"
(m3's'1) (cm) (cm) (cm)
0.05
20
30
90
0.12
30
40
120
0.20
40
45
150
0.35
50
50
180
0.55
60
60
275
Vortex plate width as indicated by dimension "a" in Figure 7-13.
n Trash rack dimension on one side, as indicated by dimension "b" in Figure 7-13. Trash rack
length and width should be equal.
Channel Design
A surface -fed channel should provide a range of habitat conditions to accommodate spawning,
rearing and overwintering, as well as the habitat needs of individual species. Rearing habitat
requirements, for example, can differ among species (Hartman 1965; Bisson et al. 1988), among
life stages of a species (Everest and Chapman 1972), and among seasons (Bustard and Narver
1975; Nickelson et al. 1992a). The channel should have pools interspersed between run or riffle
sections (Fig. 7-14). In a meandering stream with a natural flow regime, pools occur at intervals
of roughly 5-7 channel widths (Leopold and Langbein 1966). In a controlled flow channel,
however, pool spacing and size are not similarly constrained. The proportion of a surface -fed
channel devoted to pool habitat, relative to riffles and runs, can therefore be governed largely by
the habitat requirements of the fish species and life stages involved. Research on salmonid
behavior suggests that a larger number of smaller habitat units will promote greater fish utilization
per unit area than fewer but larger units. The head of a pool, for example, is known to be a focal
point for rearing salmonids because of its proximity to the supply of insect drift food from the
upstream riffle (Mundie 1974). and cover provided by the higher velocities and surface turbulence
at that location (Lewis 1969). Salmonids establish social hierarchies with larger, dominant
individuals usually located at the head of the pool, and smaller, sub -dominant individuals
distributed over downstream areas of the pool where the dissipation of velocity results in less
favourable feeding stations and cover (Mason and Chapman 1965; Griffith 1972; Fausch 1984).
The velocity entering pools, and the size and frequency of pools, are therefore important
considerations in channel design.
Velocities in pools, and water depth and velocity in the run and riffle sections, can be
manipulated by varying channel slope and width. Addition of boulder clusters can enhance cover
and habitat complexity in runs (see Chapter 10), and can increase channel roughness to create
suitable velocities for spawning in steep riffles (Fig. 7-15). Water depth, velocity and space
7-18 Rehabilitating Off -channel Habitat
requirements of various salmonid species and life stages are reviewed in Bjornn and Reiser (1991 )
and Keeley and Slaney (1996).
1-+ I,- _+ , + _ I on mirtnin wall to
Culvert
Valve
vac cock or riprap
PLAN VIEW OF INTAKE FACILITIES
Spawning area Pool
below pool Inlet pool
!D.
c f,:
0o Pool �
00
00
Wood debris cover
along outside bend
PLAN VIEW OF CHANNEL
Spawning section with gravel
Riffle section sloped at 0.4 - 0.7 substrate at 0.2 - 0.4 % slope
and 0.3 - 0.6 m diam. rock in clusters Spawning area
below pool
Figure 7-14. Concept of a surface -fed channel for spawning and rearing.
Rehabilitating Off -channel Habitat 7-19
Figure 7-15. Riffle section of channel with boulders added for habitat
complexity, Centennial Channel, Chilliwack River.
The cross-section of run and riffle sections in a surface -fed channel should be similar in concept
to a Groundwater channel (Fig. 7-6). To prevent bank erosion by spawners, and provide stability until
riparian vegetation establishes, channel banks should be constructed with a slope no steeper than 1.5:1
and lined with boulders or riprap (20-50 cm diameter) to a level above the design water surface. The
channel alignment may be adjusted to take advantage of large trees or stumps to provide cover features
such as undercut banks (Fig. 7-16). The bank along the outside bend of a pool can be protected from
erosion by placement of large boulders, riprap, or cabled wood debris. While both rock and wood
debris provide cover and habitat complexity for juvenile salmonids, wood debris is generally preferred
for that purpose. Where both materials are available, the best solution may be to armour the outside
bank with riprap for hydraulic protection, and employ wood debris along the bank to improve habitat
complexity and fish escape cover (Fig. 7-17). If riprap is used for bank armouring in pools, its
attractiveness to juvenile salmonids will be enhanced if rock size is large (> 50 cm mean diameter),
the bank is steep and irregular, and velocities along the bank are relatively high (Lister et al. 1995).
As discussed for groundwater channels, native gravels should be used for the channel bed
wherever possible. If gravel has to be imported for the project, its size composition should be
comparable to natural spawning beds, and it should include fines down to 1-2 mm diameter.
Guidelines on spawning gravel sizes for various salmonids are summarized in literature reviews
(Bjomn and Reiser 1991, Keeley and Slaney 1996). Designers should also check the hydraulic
stability of selected gravel and rock sizes with the tractive force equation given in Chapter 12.
Design width of a surface -fed channel should be in keeping with the available water supply,
and should take into account the site's physical constraints and fish production objectives.
Channel design width and slope can be varied, along with the planned operating discharge, to
produce desired water depths and velocities, and to provide variation between channel sections.
Once habitat requirements are defined, an experienced fluvial geomorphologist or hydraulic
engineer should be consulted to determine appropriate channel discharge, slope and width.
7-20 Rehabilitating Off -channel Habitat
Desioners should avoid the temptation to maximize channel length to gain the largest possible channel
area within the project site. Such a strategy may result in a marginally acceptable channel slope, and
increase the risk of sediment accumulation. It is therefore prudent to provide for adequate, or possibly
excessive (1-2%), channel slope. Excess slope can always be accommodated by adding weirs or by
increasing channel rouahness with boulders in steep riffle sections (Fig. 7-15).
Figure 7-16. Undercut stump habitat in pool, Grant's Tomb Channel,
Coquitlam River.
Figure 7-17. Pool with riprap bank armour and wood debris cover, Centennial
Channel, Chilliwack River.
Rehabilitating Off -channel Habitat 7-21
Rearing and Overwintering Ponds
Off -channel pond environments play a significant role in juvenile salmonid rearing and
overwintering, particularly for coho salmon (Peterson and Reid 1984: Swales and Levings 1989).
Many natural ponds result from beaver dams (Bryant 1984). Creation of pond environments for
juvenile salmonids may be the main focus of an off -channel habitat project, or it may be an adjunct
to groundwater or surface -fed channel development (Fig. 7-18).
Figure 7-18. Alcove -type pond connected to Centennial Channel, Chilliwack River.
For pond sites that are independent of spawning channels, it is advantageous to incorporate some
stream spawning habitat into the project, to provide greater assurance of juvenile recruitment and to
accommodate stream -dwelling fish species. If suitable spawning habitat cannot be provided, the outlet
channel must be designed to facilitate upstream juvenile passage from the main stream. Because
juvenile salmonids have relatively high swimming capabilities, up to 10 or more body lengths per
second (Webb 1975), they are able to surmount short, high -velocity stream sections. Though a steep
outlet channel may be required in some cases, the channel slope should probably not exceed 5CI.
Outlet channels over 2% slope should be stepped down to the main stream over a series of low drops,
not more than 10 cm high and separated by pools or runs that dissipate the hydraulic energy and
provide resting areas for the fish. Rough and irregular channel margins, with alternating sections of
high velocity and back eddy, are also essential features for juvenile passage. Conditions at the entrance
to the main river will be difficult to control and should not be a concern. Juvenile salmonids appear
capable of locating and entering off -channel sites under a variety of physical conditions.
Rearing and overwintering ponds can be created by either: (1) flooding an existing site or
increasing water depth over the site, possibly an existing wetland or abandoned river channel, through
construction of a dam or dyke: or (2) excavating to achieve adequate water depths within or adjacent
to an existing watercourse. Some projects combine these approaches. Off -channel pond projects can
vary greatly in size, depending on physical attributes of the selected site. Though pond areas of
projects in British Columbia have generally been less than 0.5 ha, ponds of up to�1.5 ha have been
constructed. It is worth noting that larger ponds will generally not support as many fish per unit area
as smaller ponds (Keeley and Slaney 1996). Examples of off -channel rearing and overwintering pond
projects are discussed in the folloNvins paragraphs. y
7-22 Rehabilitating Off -channel Habitat
The Anderson Creek project on the Chilliwack River is an example of creative use of site
characteristics to restore salmonid spawning and rearing habitat (Fig. 7-19). Since construction of the
Chilliwack Lake Road, upstream passage of fish into upperAnderson Creek had been blocked at the road
culvert, and an old meander channel of the Chilliwack River had been cut off from the main river. The
restoration effort involved diverting a portion of the Anderson Creek flow just upstream of the road
culvert, and installing a new road culvert 300 m east of Anderson Creek to flood the old meander channel
and to provide fish passage under Chilliwack Lake Road. The project has created a 1.5 ha pond (Fig. 7-
20), 2W m of inlet and outlet spawning and rearing channels, and has re-established fish access to upper
Anderson Creek. A 2 m deep channel was excavated along the dyke on the north side of the pond to
increase water depth for juvenile salmonid overwintering, and to make it difficult for beavers to construct
a dam at the pond outlet. A beaver -proof intake box (see Chapter 15) prevents blockage of the outlet
culvert while allowing adult and juvenile salmonids to move upstream into the pond. The outlet culvert
was oversized to provide low velocities for upstream passage of juveniles. The inlet and outlet streams
are now being used for spawning by coho and chum salmon, and the pond supports juvenile coho and
steelhead rearing and overwintering.
Studies of juvenile coho utilization of off -channel ponds for overwintering have indicated that while
shallows, less than 0.75 m deep, may be beneficial to coho in terms of benthic insect food production.
the presence of deeper areas (to 3.5 m) tends to maximize survival for smolt emigration (Peterson 1982b;
Cederholm et al. 1988). Oft -channel ponds that have both shallow areas or shoals for food production
and deep areas for overwinter security (Fig. 7-21) are most likely to produce good numbers of large,
viable smolts.
Alcove -type pools or ponds, which are essentially a backwater environment connected to the main
stream channel, are also being created to provide off -channel rearing and overwintering habitat,
primarily for juvenile coho. This type of habitat, similar in concept to Cook Creek pond in Figure 7-21,
fosters limited water exchange, low velocity, and development of pond -like conditions. Alcoves are
usually quite accessible to salmonid juveniles and coho utilization of alcoves is high compared to other
habitats in winter (Nickelson et al. 1992b). Alcoves have therefore been included in several off -channel
habitat developments in British Columbia. As a note of caution however, the point of connection between
the alcove and the main stream can be blocked as a result of shifts in the bed of the main channel under
high flows (Nickelson et al. 1992b). The alcove concept therefore appears best suited to hydraulically
stable environments, such as a groundwater channel or a surface -fed side channel with controlled flow.
Addition of wood debris cover to streams can improve salmonid rearing and overwintering
capability (see Chapter 8). There is limited evidence, however, of similar benefits to salmonid
production in ponds. Comparison of juvenile coho populations in a series of artificial off -channel ponds
at Coldwater River, with and without wood debris added, revealed an average of 2.5 times higher
densities in ponds with debris cover (Beniston et al. 1988). The advantage of added wood debris
appeared to be greatest in winter and smallest in summer.
Experience with performance of constructed off -channel pond environments is not extensive.
Research to date would suggest incorporation of the following design principles:
• limit the area of individual ponds to 0.1 - 0.3 ha:
• provide diverse water depths for fish growth and overwinter security;
• limit water exchange rates to foster development of a pond environment, but recognize the need
to maintain suitable water temperatures and dissolved oxygen levels;
• incorporate, where feasible, some spawning habitat upstream of the pond, to reduce reliance on
upstream movement of juveniles for recruitment, and
• add wood debris cover, to improve rearing and overwintering capability.
Rehabilitating Off -channel Habitat 7-23
>
3 0
(D
_0
RINak,
—
'r- 0
ID
CL.0
0
M-0
ca
(D
>
0
E
cn 3:
>
Na a)
.0
cc
0
(D
0
c
-0
0)
-0
a)
>
0)
_0
0
0
02
U)
a) 0)
0
(n Z
r_
70 CO
U)
LL
-0
a) CD
2 C/)
co
-0-0
.2
0
U)
u0
psi
1I
CL
I N
ti
rn
0 co
0,
3:
U — E
a)
a)
(D —C
2 co
2
_0
E
0) a)
CO >
LO
i I ca
-01
0 a)
_0
00 1 U)
ca C3)J
CL
1
r-
<
0
0
(D
,4
- 0
75
F-;
LLJ
CL
ca
0
ICI
r,
CY,
0 m
CO
LU
>
-C
Z
CL
.C: 0
-0 )
a (D
II
> 0 -0
(D
0 c_0
(D a) 0
E o-O
NEo
0 CLt CO.(D
CL
o)-o
-q 7.' 2 3:.Si -ca
.2 —:.Z5 cii CD
CO C
2
co
�-- o c 2
00 CO (D
Xro 0 r- =
-§ -0
0 C: C:
E
co
z "
0)
U) c: a)
a) L)
U) — --=> -=
C:
c:
co
-E
ca-.1,
>
Co > Zb
>
C) C:
of
a0iE .2
0 co >
ca
.C—: 0 -FD
C:
-00
ca 0
I l x
ca
�E
3:
ca -0 L)
CL m
0,
CL
(n
c7) -6 2
2
0 0
c
LLJ
C) a)
co 3�
0
0
Figure 7-20. Rearing and overwintering pond at Anderson Creek, Chilliwack River.
Cook Creek
2
1
50 m
Paradise Pond
Blasted Holes
40 m " s
NY NY
Dam and Fishway
-'Ar
Figure 7-21. Alcove pond at Cook Creek, North Thompson River (Lister and
Dunford 1989) and beaded channel at Paradise Pond, Clearwater River,
Washington (Cederholm et al. 1988).
Rehabilitating Off -channel Habitat 7-2:)
Project Costs
Costs of off -channel habitat projects vary with scale and type of habitat, as well as the particular
site conditions (Table 7-2). Rearing and overwintering ponds of various sizes can be created at
modest cost ($2.50 - 4.00 per m2) in relation to excavated groundwater -fed channels with armoured
banks ($14.00 - 18.00 per m2 ).
CONSTRUCTION PROCEDURES
Construction of off -channel habitat, as with any fish habitat project, is largely about dealing
with water. This influences construction timing and procedures, and it also poses challenges in
controlling sediment to avoid impact on downstream fish habitat.
To prepare for construction the limits of clearing and excavation as well as the channel
centerline, should be flagged. Flagging may reveal the potential loss of exceptional trees that could
be avoided by a modest shift in project alignment or by confining the excavation limits. Clearing
of vegetation should also be undertaken in a manner that avoids unnecessary disturbance and loss
of trees. Conifers that are suitable for providing habitat complexity in the finished project should
be stockpiled (Fig. 7-22).
Figure 7-22. Excavation of a surface -fed channel showing stock piles of conifer
trees and boulders, Centennial Channel, Chilliwack River.
Excavation and handling of topsoils should be undertaken during the dry summer season. A
groundwater channel should ideally be excavated in late summer when the water table is low.
Instream work windows, established by regulatory agencies to avoid fish habitat impacts. are also
an important constraint on the construction schedule. Projects should be completed early enough to
provide time for germination and growth of grasses to stabilize soils before the fall rains. Seeding
of grasses needs to be conducted before temperatures decline in fall.
7-26 Rehabilitating Off -channel Habitat
L
C
_
DC
�n
C
—
_
J
it
�.i
74,
J
=
c
7>_
c
L
—
zo
J �
t �
f „
w
L
C �
w
•
'D
tom/]
—
� T
=L 6 ,J. .>.
�
_
•� c
� c
� ? �
� -
� � � _
_ o N c
c �
" u 3
c � G �
-
^—
G
J
L
— J
to C
cp C
Y .=
G _. Jl
vl
m
u �
v
:J
+L+ CC
—
y
.L.
CE
to
In the case of groundwater channels, and some pond habitat projects, site excavation will penetrate
the water table and cause seepage into the excavated area. An earthen plug can be left at the
downstream end to isolate the site from the adjacent stream. Keeping initial excavation above the water
table, and lowering the bed of the channel or pond in stages, can also assist in dealing with seepage. A
staged excavation strategy could involve removal of overburden, then excavation to the water table, and
lastly, commencement of ditching at the lower end of the channel to pull down the water table. Water
that can't be retained on the site should be pumped or diverted to a constructed sediment settling basin
or natural depression for treatment. Clean water flowing into the construction area should be diverted
or pumped around the site to assist construction and minimize water volumes to be treated for sediment
control. It is advisable to develop a sediment mitigation plan and to obtain agreement from habitat
protection staff of the regulatory agencies in advance of construction.
Excavated overburden and gravels will normally be sidecast to form a flood protection dyke on the
main river side of the project. Where practical, topsoil should be stockpiled so that it can be later used
on disturbed areas such as channel and dyke slopes, to improve the growth medium for revegetation.
Rootwads and large boulders that have to be removed should also be stored on -site for subsequent
placement as instream cover or bank protection. Trees and shrubs, which can be salvaged alive with
intact roots, should be covered with soil and stored for site revegetation.
Following site excavation, channel side slopes should be armoured and rock weirs installed.
Vegetative techniques for slope stabilization may also be appropriate (see Chapter 6). Wood debris
and/or large rocks should be added to enhance habitat complexity.
A surface intake on a stream should be constructed during the low water period and within the
approved instream construction window. If necessary, the intake can be isolated from the stream by
retaining an earth plug or by sand bags, to avoid downstream sedimentation and facilitate the
construction.
The final stage involves stabilization of banks and other disturbed areas to control surface erosion
during the fall -winter period. Available top soil should be spread over disturbed soils, and slopes should
be lightly scarified to improve conditions for establishing grasses and shrubs. The site should be seeded
with an appropriate mix of grasses and clover. A light dressing of hay or straw may then be spread over
the seeded areas to provide some erosion protection until grasses germinate and begin to grow.
OPERATION AND MAINTENANCE
Operation and maintenance requirements of off -channel habitat projects vary considerably
depending on the type and complexity of the project. Projects that utilize a groundwater source
normally require little operational effort and also have modest maintenance needs. Surface -fed
channels or ponds, on the other hand, will often require an active operation and maintenance program.
In this respect, it is prudent to engage a local entity, such as a fish and game club or salmonid restoration
society, to carry out the routine operation and maintenance.
For a groundwater project, the operational concern is generally related to fish access, which
may be restricted by a beaver dam or inadequate flows, particularly in the early fall. The gravel
substrate and rock armour of a groundwater channel will accumulate fine organic material, and
may experience some encroachment by vegetation or rooted aquatic plants. These problems are
generally addressed by scarifying the gravel and turning over the rock armour. Reconstruction of
rock weirs. regrading of spawning _ravel, and addition of bank armouring may also be required
7-28 Rehabilitating Off -channel Habitat
from time to time. Groundwater -fed channel and pond projects should be inspected periodically during
the spawning migration period to ensure fish access and at least annually to monitor channel substrate
quality and other features.
A surface -fed channel or pond will generally require greater operational attention than a
groundwater project, because of relatively high flow volume and potential for intake blockage. The
intake and flow control gate will need to be regularly inspected and kept free of debris. During high
flow periods, inspections may be required weekly or more frequently, while at other times the
inspection frequency can likely be reduced. Once the best discharge for fish spawning is
established, the flow control gate or valve should be left at the required setting throughout the
spawning period. It may be desirable, however, to reduce channel discharge during the incubation
period in order to minimize introduction of suspended sediment to the channel.
Surface -fed channels need to be inspected annually to assess the maintenance requirements.
Features to be examined include intake condition, sediment deposition in the settling pond,
condition of weirs and bank armour, and possible accumulation of fine silts and organics in the
gravel substrate. Maintenance could involve sediment excavation from the settling pond, weir repair
or removal, bank armour repairs, and gravel cleaning by either scarification or, in an extreme case,
by removal and replacement with new or screened material. The latter activity should be only an
occasional requirement.
Though a pond environment is unlikely to require maintenance, the stream connecting it to the
main stream should be inspected periodically for potential obstructions to fish passage, particularly
beaver dams. It is also prudent to periodically monitor water quality, especially dissolved oxygen,
to identify any trend toward deteriorating water quality.
Rehabilitating Off -channel Habitat I -Z9
T#
SPAWNING CHANNEL REPLACEMENT PROJECT
STORMWATER POLLUTION PREVENTION PLAN (SWPPP)
Element 1: Mark Clearing Limits
Prior to beginning land disturbing activities, including clearing and grading, the US Army
Corps of Engineers and the City of Renton will delineate the project and clearing limits
with silt fencing and/or surveyor's tape. Trees to be cut down for construction of the
spawning channel and maintenance access road will also be marked accordingly.
Element 2: Establish Construction Access
The spawning channel maintenance access road will be constructed prior to work on the
channel and will be used as a construction access. The access road will be constructed of
two-inch minus with quarry spalls at the entrance over a compacted natural base to
reduce local erosion and tracking of mud and sediment by construction vehicles.
Easement agreements with public and private parties require the City to maintain
cleanliness of easements during construction. These will be inspected on a daily basis
and swept as necessary.
Element 3: Control Flow Rates
Construction of new impervious surfaces associated with this project are minimal.
Stormwater runoff flows and volumes from the project site will not increase, therefore no
flow control is required.
Element 4: Install Sediment Controls
Natural vegetation will be retained to the maximum extent possible. Silt fencing, riprap,
straw bales, and other temporary erosion and sedimentation control best management
practices (TESC BMPs) will be utilized as required to prevent silt -laden runoff from
exiting the site.
Element 5: Stabilize Soils
All exposed and unworked soils will be revegetated with natural vegetation and mulching
or hydroseeding of disturbed areas prior to project completion. The project is expected to
be completed by September 2003. Excavated soils stockpiled off -site will be covered
with plastic and/or surrounded by silt fencing.
Element 6: Protect Slopes
Slopes excavated and cut to create the channel will be armored with riprap or planted
with natural vegetation to prevent erosion.
H:\File Sys\SWP - Surface Water Projects\SWP-27 - Surface Water Projects (CIP)\27-3046
Spawning Channel Replacement Project\1400 - Permits, Plan Review\Doe\SWPPP.doc
Z
Element 7: Protect Drain Inlets
No drain inlets are associated with this project.
Element 8: Stabilize Channels and Outlets
No temporary channels are associated with this project. The spawning channel itself will
be armored with riprap and/or planted with natural vegetation to stabilize soils and
prevent erosion.
Element 9: Control Pollutants
No pollutants, waste materials, or demolition debris is anticipated on -site as part of this
project. Construction equipment will be inspected on a daily basis to determine if leaks
of fuels, hydraulic fluids, etc. are occurring. If leaks are found, drip pans, plastic, etc.
will be used to retain the fluids until repairs are made.
Element 10: Control De -watering
No de -watering is anticipated as part of this project. In case it becomes necessary, silt
laden runoff will be infiltrated in the channel itself.
Element 11: Maintain BMPs
All TESC BMPs will be inspected on a regular basis and after storm events to ensure
adequate protection and performance. All TESC BMPs will be removed after permanent
erosion control measures are in place upon project completion.
Element 12: Manage the Project
The USACE and the City of Renton will be responsible for ensuring TESC BMPs are
adequately installed, maintained, and removed upon project completion. Regular
inspections will be made to ensure protection from construction runoff exiting the site.
H:\File Sys\SWP - Surface Water Projects\SWP-27 - Surface Water Projects (CIP)\27-3046
Spawning Channel Replacement Project\1400 - Permits, Plan Review\Doe\SWPPP.doc
Cedar River Spawning Channel Rehabilitation
Project Management Plan
Overview: This PMP is intended to specify the roles, responsibilities, and protocols of
the U.S. Army Corps of Engineers and the City of Renton during the construction of the
PL 84-99 Cedar River Spawning Channel Rehabilitation. It will also identify construction
methods, constraints, and issues that will be utilized for this project.
Project Name: Cedar River Spawning Channel Rehabilitation
Location: Left Bank of the Cedar River, see attached map.
Sponsor:
Gary Schmick, (425) 7205
City of Renton
1055 South Grady Way
Renton, Washington
98055
Project Team:
Noel Gilbrough
Project Manager
Doug Weber
Program Manager
Rustin Director
Environmental Coordinator
Wanda Gentry
RE
Zach Corum
H&H
Monte Kaiser
Civil Design
Eric Winters
Construction Lead
Charles Ifft
ICW Program Manager
Description:
The Cedar River Spawning Channel was constructed as part of a Section 205 Flood
Damage Reduction project in 1998. The purpose of the channel was to provide spawning
habitat for Sockeye salmon to mitigate for potential impacts as a result of construction of
the flood control portion of the project. The spawning channel was subsequently
irreparably damaged in the February 28,2001 Nisqually Earthquake. After much debate,
replacement of the spawning channel was authorized and funded under the PL 84-99
rehabilitation program. The project was approved because the spawning channel was
required by permitting agencies to allow maintenance dredging of the flood control
project, which was anticipated to be required approximately every three years. Under that
condition, the spawning channel was recognized to be a vital part of the flood control
system.
Roles and Responsibilities:
To be added later
HAFile Sys\SWP - Surface Water Projects\SWP-27 - Surface Water Projects (CIP)\27-3046 Spawning Channel Replacement
Project\1100 - Design and Planning\1101 - Design\Plan - Construction Management.doc
Budget:
Project Information Report $20k
Construction
Federal
Construction
$ ????
E + D
$ ????
TOTALS
$ ????
Local
Total
$ 0
$ ????
Quality Objectives:
Project is to be restored to pre -flood condition or an alternate project to restore pre -flood
production.
Project must be in compliance with all pertinent environmental laws, regulations, and
policies.
Project shall be completed in 2003 or be transferred to a planning authority.
Schedule:
Completed P&S
January 2003
Permits Requirements Satisfied
April 2003
Real Estate Certified
April 2003
Contract Advertisement
April 15 2003
Contract Bid Opening
June 13 2003
Contract NTP
June 23 2003
Construction Start
July 012003
Construction Complete
August 29 2003
Construction Final Inspection
October 312003
Construction Methodologies and Issues:
1) General
a. Construction is scheduled to begin on July 1, 2003 and in water work will be
complete by August 31, 2003.
b. The Construction Supervisor is the lead Corps official on the construction site
and is therefore shall be held accountable for the safety, envirommental
compliance, and prompt progress of the project construction.
c. The Environmental Coordinator is responsible for environmental monitoring
and onsite agency coordination of the project.
d. The Project Manager is in charge of the overall project, answering to the
primary customers, City of Renton and Emergency Management.
2) Construction Sequencing:
• Site cleared
• Access road built
• Excavation complete with upstream and downstream plugs
• Channel features built (spawning gravels, woody debris, etc.)
• Intake structure built
• Plugs removed
• Site clean up
HAFile Sys\SWP - Surface Water Projects\SWP-27 - Surface Water Projects (CIP)\27-3046 Spawning Channel Replacement
Project\1100 - Design and Planning\1101 - Design\Plan - Construction Management.doc
• Planting
3) City of Seattle Easement Requirements:
a) The Construction Supervisor will ensure that all of the easement conditions are
adhered to during construction.
b) A gate will be upgraded prior to construction to prohibit vehicle access. The
gate will be locked anytime that the access is not being used for construction. During
construction activities, the construction supervisor will maintain access control to limit
access to authorized personnel.
c) All runoff and siltation will be controlled during construction using the
sediment control plan.
d) Load restrictions will be implemented as necessary to avoid damaging any
utilities in the easements. Areas where the trucks will travel over the pipeline will be
spanned with metal plates to distribute the load.
e) Scour protection will be constructed as shown in the design to protect utilities
in the easement.
f) At the end of the construction the easement will maintained to pre -construction
condition including asphalt paving of the road.
4) The following Best Management Practices will be utilized during construction:
a. Equipment that will be used near the water will be cleaned prior to
construction.
b. Work will be conducted during a period of low flow.
c. Biodegradable hydraulic fluids will be used for machinery at the site.
d. Refueling will occur away from the river bank.
e. Construction equipment will be regularly checked for drips or leaks.
f. At least one fuel spill kit with absorbent pads will be onsite at all times.
5) Sediment Control.
a. The Construction Supervisor is responsible for implementing whatever
sediment control measures are need to ensure the sediment mixing zone does not
exceed 300 feet downstream.
b. Visual inspections will be made by the Construction Supervisor to ensure
compliance. Upon request of the Construction Supervisor, the onsite Biologist
can provide turbidity sampling.
c. If the mixing zone exceeds 300 feet, the Construction Supervisor will modify
construction methodologies to ensure 300 feet is maintained.
d. If 300 feet cannot be maintained, the Construction Supervisor will halt the
construction activities causing the discharge until an alternative plan or waiver
can be developed in conjunction with the appropriate Corps staff and/or agencies.
e. The use of silt fencing or other sediment control devices will be implemented
as required by the design. As site conditions change (i.e. rain) implementation of
other devices may be necessary.
6) Biological Monitoring
HAFile Sys\SWP - Surface Water Projects\SWP-27 - Surface Water Projects (CIP)\27-3046 Spawning Channel Replacement
Project\1100 - Design and Planning\l 101 - Design\Plan - Construction Management.doc
a. The Environmental Coordinator will be onsite for the initiation of construction
to determine any presence of species of concern and any potential impacts.
b. The Environmental Coordinator will document any presence or non -presence
of concerned species.
c. The Corps biologist will bring any issues of concern and recommendations to
the Construction Supervisor, whereupon the Construction Supervisor and
Environmental Coordinator will determine the best course of action.
d. After initiation of construction, the Environmental Coordinator will make a
determination regarding how often appropriate environmental staff will need to be
present at the construction site.
7) On site Coordination
a. Any resource agency coordination will be done through the Project Manager or
the Environmental Coordinator, unless the coordination involves a visit to the
construction site, whereupon the Construction Supervisor will be primarily
responsible.
b. Anyone visiting the Construction site (Corps, Sponsor, Agency, Tribal, Public)
will need to coordinate on site visits with the Construction Supervisor prior to the
visit to ensure safety and the prompt progress of construction activities.
c. The Construction Supervisor may limit access to the construction site as
necessary to ensure personal safety and the prompt progress of construction
activities.
d. If excessive visitation to the site by others is interfering with construction
progress, the Construction Supervisor should coordinate with the PM and to have
the PM facilitate visitation procedures.
e. If the Environmental Coordinator is not on site and resource agencies or others
concerned about biological issues visit the site, the Construction Supervisor
should inform the Environmental Coordinator of their visit as soon as possible
such that the Environmental Coordinator can follow up with the visiting official
H:\File Sys\SWP - Surface Water Projects\SWP-27 - Surface Water Projects (CIP)\27-3046 Spawning Channel Replacement
Project\1100 - Design and Planning\1101 - Design\Plan - Construction Management.doc
DRAFT 12/21/99
Memorandum
To: LARRY BASICH, FEMA
From: David Hartley, Senior Watershed Hydrologist, King County DNR-WLRD
Date: Revised March 13, 2000
Re: FLOOD FREQUENCY CURVE FOR YEAR 2000 FLOODPLAIN
MAPPING ON THE CEDAR RIVER
cc: Ron Straka, Gary Schimek, Dave Clark, Nancy Faegenburg, Jeanne Stypula,
Andy Lesveque
Executive Summary
A flood frequency analysis was performed for the lower Cedar River between USGS gage
12117500 near Landsburg at RM 21.6 and USGS gage 12119000 at Renton at RM 1.6.
The 100-year flow at Renton is estimated to be 12,000 cfs based on the HEC-FFA
program that fits a Log -Pearson III distribution using peak flow data. The 12,000 cfs
value is consistent with the USA-COE's estimate used for design of a 1998 flood
reduction project at Renton. It is approximately 36% higher than the previous 100-year
flow assumed by the existing FIRM (-8,800 cfs), mainly because a flood of record
occurred on the river in 1990 after the existing FIRM was developed.
Introduction
The purpose of this technical memo is to propose and document the rationale for a flood
frequency curve that will be used to re -map the floodplain of the Cedar River in
accordance with FEMA methods and standards. In water year 1991, a record flood
occurred near the mouth of the river at Renton. It exceeded the previous record of 1976
by a large margin. The existing FIRM is based on a flood frequency curve that pre -dates
DRAFT 12/21 /99
the water year 1991 event. While the occurrence of a record flood often justifies re-
analysis of flood frequency and upward adjustment, the need for adjustment in this case is
greatly magnified because the 1991 flood required the very first emergency operation of
PMF-capable gates at Masonry Dam. The 1991 flood experience suggests that the
magnitude of 100-year and greater floods are probably much higher than had been
indicated by the previous record which was devoid of events that necessitated emergency
gate operations at the dam.
The flood frequency information presented herein applies to the lower 21.6 miles of the
river from the Landsburg Diversion Dam to the mouth of the river at Lake Washington in
the city of Renton. King County's floodplain re -mapping project lies entirely within this
reach of the river.
This technical memo describes the following steps in the development of a proposed
flood frequency curve for the re -mapping study:
1. A brief discussion of flow data available for flood frequency analysis
2. Selection and development of a data set suitable for Log -Pearson III analysis using
the HEC-FFA program.
3. Presentation of HEC-FFA results
4. Definition of reaches and corresponding flows for HEC-RAS analysis.
Sources of Data
The primary sources of data to support the flood frequency determination for the lower
Cedar River are
USGS 12117500, Cedar River Near Landsburg, RM 23.4, Drainage Area 121 sq. mi.,
Period of Record 1896-present.
USGS 12119000, Cedar River at Renton, RM 1.6, Drainage Area 184 sq. mi., Period of
Record 1946-present.
Both gages are downstream of Masonry Dam (RM 35.6), a facility with significant
storage capacity devoted primarily to water conservation for municipal supply.
Secondary uses of the facility include meeting instream flow objectives during the low
flow season, hydropower generation, and control of winter peak flows.
Although flood season operation and some physical characteristics of the dam outlets
have evolved over the years, the basic storage capacity provided by the facility has been
approximately consistent since water year 1920. This post-dates dam construction and
initial filling of Masonry Pool which began in water year 1915. Initially, the facility was
plagued by seepage problems that eventually culminated in the Boxley Burst of
DRAFT 12/21 /99
December 23, 1918- a catastrophic landslide and debris flow in the neighboring south
fork Snoqualmie basin. (See Bliton,1989)
The Landsburg Diversion dam at RM 21.6 is located between the two USGS gage sites,
however, there is virtually zero storage at this facility and it consequently provides
negligible attenuation of flood peaks.
Selection of Data for Log -Pearson 111 Analysis
Although the upstream USGS gage site (12117500, Cedar River Near Landsburg)
includes peak annual flow data going back to 1895, only data from WY 1920 to present
have been used in recognition of the effect of Masonry Dam's operation on downstream
peak flows. This provides a substantial record of approximately 80 peak annual flows
that can be used in flood frequency analysis.
The record at the downstream USGS gage site (12119000, Cedar River at Renton) dates
from 1946. However, regression analysis of contemporaneous peak annual values at the
two sites provides a method for filling the Renton record. As shown in Figure 1, Renton
annual peaks in cfs can be related to Landsburg annual peaks in cfs by the quadratic
relationship: Renton =-4E-05Landsburg2 + 1.3864Landsburg, W = 0.9483.
12000.0
10000.0
80000
60000
4000.0
2000.0
0.0
Figure 1. Regression of Renton on Landsburg Peaks
0.0 2000.0 4000.0 6000.0 8000.0 10000.0 12000.0
Landsburg Annual Peaks
DRAFT 12/21/99
With this relationship, peak flows at Renton can be generated for the period from 1920 to
1945. Subsequently, HEC-FFA, Log -Pearson III fits can be made for both sites using
equal record lengths of 80 years spanning water years 1920 - 1999.
Presentation of HEC-FFA Results
The full output files from the HEC-FFA runs are provided in Appendix I. A summary of
the results using the computed quantiles is presented in Table I.
TABLE I.
HEC-FFA FREQUENCY CURVES
BASED ON ANNUAL PEAKS 1920-1999
USGS 12117500- CEDAR R.
NEAR LANDSBURG
USGS 12119000- CEDAR R. AT
RENTON
%CHANCE
COMPUTED
COMPUTED
ANNUAL
EXCEED.
QUANTILE
0.05 CI
0.95 CI
QUANTILE
0.05 CI
0.95 CI
0.2
16100
21700
12800
18400
24400
14800
0.5
12500
16300
10200
14500
18600
11900
1
10300
13000
8540
12000
15000
10100
2
8340
10200
7080
9860
12000
8450
5
6220
7360
5430
7470
8760
6570
10
4880
5610
4340
5940
6780
5320
20
3720
4170
3370
4600
5120
4190
50
2370
2600
2160
3000
3270
2750
80
1640
1810
1460
2110
2330
1890
90
1390
15601
12201
1810
2010
1600
95
1230
13901
1070
16201
1810
1410
991
1020
1170
863�_l
13501
15401
1150
Explanation of Skew Coefficients Employed in HEC-FFA Analysis
Based on technical comments from FEMA (personal communication, Wilbert Thomas,
3/3/00), station skew value (0.699) calculated from the longer, historical Landsburg
record, is utilized instead of a the traditional weighted mean value that combines station
skew with a regional value. The regional value is based on data from many gaging
stations in this area with flows that are either unregulated or regulated in a different
manner than the Cedar River. The Landsburg skew value is also used in the HEC-FFA
computation of the Renton frequency curve rather than its own station skew. This assures
that flow exceedance levels increase in the downstream direction across the entire
spectrum of return periods.
Results from the HEC-FFA analysis using a single skew equal to the Landsburg station
skew of 0.699 are consistent with flow quantiles proposed by other studies. These
DRAFT 12/21/99
include flow analysis used in the recent channel improvement project done cooperatively
by he city of Renton and U.S. Army Corps of Engineers (COE). The COE has used a
100-year flow level of 12,000 cfs at Renton in the design of this project (personal
communication, Ron Straka, city of Renton). The COE based its 100-year estimate on an
earlier feasibility study (COE, 1990) using the SSARR model. The model represented
current dam operations and utilized both historical storms and design storms representing
extreme events.
Another recent study of Masonry Dam operations jointly sponsored by Renton, King
County, and Seattle (NHC, 1998) estimated 100- and 500-year flows at Landsburg and
Renton that are considerably higher that those resulting from the HEC-FFA reported in
this memo (see Table II). The higher flow quantile values reported by NHC can be
attributed to several factors:
1. The annual peak values used to compute the frequency curve are based on modeled
rather than historical annual peaks.
2. The modeled November, 1990 peak of record is 43% higher than the gaged peak of
record at Renton
3. Modeled peaks for the event of record depended questionable estimates of reservoir
inflow because of incomplete data
4. The model assumed zero peak attenuation between Masonry Dam and Renton
5. Frequency curves were estimated by eye -fit to modeled peak data using the Weibull
plotting position. The Log -Pearson III distribution was not utilized.
DRAFT 12/21 /99
TABLE II.
100- and 500-yr EXCEEDANCE LEVELS FROM NHC STUDY
RETURN PERIOD
LANDSBURG
RENTON
YEARS
EXCEEDANCE LEVEL
(CFS)
EXCEEDANCE LEVEL
(CFS)
100
11,900
13,100
500
21,400
21,200 11
Recommended Flood Quantiles for HEC-RAS Analysis And Floodplain
Mapping
Given the existence of a long record of USGS flows at Landsburg, it is proposed to use
the results of the HEC-FFA fit of the Log -Pearson III as the basis for hydraulic analysis
and 2000 year re -mapping of the lower Cedar River floodplain. The HEC-FFA results
are consistent with the recent Renton-COE channel improvement project. The higher
flow quantiles suggested by the NHC study result from the use of modeled rather than
gaged peak annual flows as well as the data and assumptions underlying the model.
Reach and Flow Assignments for HEC-RAS Modeling
For purposes of HEC-RAS analysis, the river can be broken up into reaches based on the
confluence points of tributaries entering the river between the two gage sites.
Exceedance flow levels at intermediate points are interpolated based on average
cumulative drainage area within the reach. The reach definitions and corresponding
flows are presented in Table III.
TABLE III.
RIVER REACHES AND FLOOD QUANTILES FOR HEC-RAS
LOWER
UPPER
DISCHARGE WITHIN REACH
NOTES
RIV MI.
RIV MI
10-YR
50-YR
100-YR
500-YR
(cfs)
(cfs)
(cfs)
(cfs)
0.0
2.3
5940
9860
12000
18400
MOUTH TO TRIB 0300, USGS 12119000
2.3
5.0
5834
9708
11830
18170
TRIB 0305 ENTERS @5.0, 0304
ENTERS @3.9, 0302 ENTERS @3.2
5.0
6.1
5728
9556
11660
17940
TRIB 0311 @6.1, TRIB 0307@5.3
6.1
12.0
5569
9328
11405
17595
TRIB 0317 ENTERS @12.0,
0316 ENTERS @11.2, 0308@7.1
12.0
14.0
5463
9176
11235
17365
TRIB 0328 ENTERS @14.0,
0239 ENTERS@13.0
14.01
18.0
5251
8872
10895
16905
TRIB 0338 ENTERS @18.0
18.01
19.51
5029
8553
10538
16422
TRIB 0341 ENTERS @19.5
DRAFT 12/21/99
21.6
4880
8340
10300
16100
LANDSBURG DIVERSION @21.6,
[19.5�
USGS 12117500 11
References:
Bliton, William S. 1989. Cedar River Project. In: Engineering Geology in Washington,
Volume I, Washington Division of Geology and Earth Resources Bulletin 78, pp. 225-
232.
Northwest Hydraulics Consultants. 1998. Masonry Dam Flood Operations Study. Report
prepared for Seattle Public Utilities, King County Water and Land Resources Division,
and City of Renton.
R.W. Beck and Associates, 1988. Operations and Maintenance Handbook- Cedar Falls
Headworks, Masonry Dam and RCC Overflow Dike: Prepared for Seattle City Light and
the Seattle Water Department.
U.S. Army Corps of Engineers. 1990. Summary and Status of Section 205 Cedar River
Flood Damage Reduction Feasibility Study.
DRAFT 12/21/99
Appendix 1. Output from HEC-FFA Runs
************************************
*************************************
* FFA
*
* FLOOD FREQUENCY ANALYSIS
ENGINEERS
* PROGRAM DATE: MAY 1992
CENTER *
* VERSION: 3.0
*
* RUN DATE AND TIME:
95616
* 07 MAR 00 10:44:11
*
* *
*
************************************
*************************************
INPUT FILE NAME: LANSTSK.DAT
OUTPUT FILE NAME: LANSTSK.OUT
DSS FILE NAME: LANSTSK.DSS
*
* U.S. ARMY CORPS OF
* THE HYDROLOGIC ENGINEERING
* 609 SECOND STREET
* DAVIS, CALIFORNIA
* (916) 756-1104
*
----- DSS --- ZOPEN: New File Opened, File: LANSTSK.DSS
Unit: 71; DSS Version: 6-GX
**TITLE RECORDS)**
TT FLOOD FLOW FREQUENCY ANALYSIS PROGRAM
TT FITTING THE LOG-PEARSON TYPE III SKEW FORCED TO STATION SKEW
TT NEAR LANDSBURG, 1920-1999 ANNUAL PEAKS, 80 YEARS
**STATION IDENTIFICATION**
ID LANDSBURG USGS 12117500
**SYSTEMATIC EVENTS**
80 EVENTS TO BE ANALYZED
**END OF INPUT DATA**
ED+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
************************************************************
CAUTION FROM SUBROUTINE WTSKEW
***** NO GENERALIZED SKEW PROVIDED
ADOPTED SKEW SET TO COMPUTED SKEW
AAAAAAAAAAAAAAAAAAAAAA FINAL RESULTS AAAAAAAAAAAAAAAAAAAAAA
-PLOTTING POSITIONS- LANDSBURG USGS 12117500
DRAFT 12/21/99
tffffffffffffffffffffffffffNfffffffffffffffffffffffffffffffffff»
°
EVENTS ANALYZED
3
ORDERED
EVENTS
°
°
FLOW
3
WATER
FLOW
WEIBULL °
° MON
DAY
YEAR
CFS
3
RANK
YEAR
CFS
PLOT POS °
0 0
0
1920
1860.
3
1
1991
10800.
1.23 0
0 0
0
1921
1920.
3
2
1976
7930.
2.47 0
0 0
0
1922
5960.
3
3
1934
7520.
3.70 0
0 0
0
1923
4160.
3
4
1996
6560.
4.94 0
0
0
0
1924
3100.
3
5
1951
6200.
6.17 0
0 0
0
1925
2740.
3
6
1922
5960.
7.41 0
0
0
0
1926
1720.
3
7
1978
5900.
8.64 0
0
0
0
1927
1820.
3
8
1928
4860.
9.88 0
0 0
0
1928
4860.
3
9
1932
4860.
11.11 0
0 0
0
1929
1180.
3
10
1960
4840.
12.35 0
0 0
0
1930
1350.
3
11
1965
4640.
13.58 0
0 0
0
1931
1200.
3
12
1933
4300.
14.81 0
0 0
0
1932
4860.
3
13
1947
4190.
16.05 0
0 0
0
1933
4300.
3
14
1935
4160.
17.28 0
0 0
0
1934
7520.
3
15
1923
4160.
18.52 0
0 0
0
1935
4160.
3
16
1987
4090.
19.75 0
0 0
0
1936
1900.
3
17
1984
3990.
20.99 0
0 0
0
1937
1800.
3
18
1982
3980.
22.22 0
0 0
0
1938
2360.
3
19
1972
3840.
23.46 0
0 0
0
1939
1500.
3
20
1990
3520.
24.69 0
0 0
0
1940
1880.
3
21
1959
3460.
25.93 0
0
0
0
1941
1050.
3
22
1953
3370.
27.16 0
0 0
0
1942
1830.
3
23
1956
3280.
28.40 0
0 0
0
1943
2140.
3
24
1957
3240.
29.63 0
0 0
0
1944
1380.
3
25
1924
3100.
30.86 0
0 0
0
1945
1970.
3
26
1950
3050.
32.10 0
0 0
0
1946
2040.
3
27
1954
2770.
33.33 0
0 0
0
1947
4190.
3
28
1974
2770.
34.57 0
0 0
0
1948
1940.
3
29
1925
2740.
35.80 0
0 0
0
1949
1750.
3
30
1955
2720.
37.04 0
0 0
0
1950
3050.
3
31
1999
2660.
38.27 0
0 0
0
1951
6200.
3
32
1969
2440.
39.51 0
0 0
0
1952
1740.
3
33
1980
2400.
40.74 0
0 0
0
1953
3370.
3
34
1986
2390.
41.98 0
0 0
0
1954
2770.
3
35
1995
2390.
43.21 0
0
0
0
1955
2720.
3
36
1938
2360.
44.44 0
0
0
0
1956
3280.
3
37
1961
2350.
45.68 0
0 0
0
1957
3240.
3
38
1964
2340.
46.91 0
0 0
0
1958
1570.
3
39
1975
2320.
48.15 0
0 0
0
1959
3460.
3
40
1973
2310.
49.38 0
0 0
0
1960
4840.
3
41
1981
2280.
50.62 0
0
0
0
1961
2350.
3
42
1983
2260.
51.85 0
0 0
0
1962
1960.
3
43
1968
2240.
53.09 0
0 0
0
1963
1930.
3
44
1971
2240.
54.32 0
0 0
0
1964
2340.
3
45
1967
2170.
55.56 0
0
0
0
1965
4640.
3
46
1943
2140.
56.79 0
0 0
0
1966
1380.
3
47
1997
2090.
58.02 0
0 0
0
1967
2170.
3
48
1946
2040.
59.26 0
0 0
0
1968
2240.
3
49
1945
1970.
60.49 0
° 0
0
1969
2440.
3
50
1962
1960.
61.73 0
° 0
0
1970
1620.
3
51
1948
1940.
62.96 0
0 0
0
1971
2240.
3
52
1963
1930.
64.20 0
0 0
0
1972
3840.
3
53
1921
1920.
65.43 0
0 0
0
1973
2310.
3
54
1936
1900.
66.67 0
0 0
0
1974
2770.
3
55
1940
1880.
67.90 0
0 0
0
1975
2320.
3
56
1920
1860.
69.14 0
DRAFT 12/21/99
0 0
0
1976
7930.
3
57
1942
1830.
70.37 0
0 0
0
1977
1250.
3
58
1927
1820.
71.60 0
0 0
0
1978
5900.
3
59
1937
1800.
72.84 0
0 0
0
1979
1590.
3
60
1949
1750.
74.07 0
0 0
0
1980
2400.
3
61
1952
1740.
75.31 0
0 0
0
1981
2280.
3
62
1926
1720.
76.54 0
0 0
0
1982
3980.
3
63
1989
1690.
77.78 0
0 0
0
1983
2260.
3
64
1988
1650.
79.01 0
0 0
0
1984
3990.
3
65
1992
1630.
80.25 0
0 0
0
1985
1600.
3
66
1970
1620.
81.48 0
0 0
0
1986
2390.
3
67
1998
1620.
82.72 0
0 0
0
1987
4090.
3
68
1985
1600.
83.95 0
0 0
0
1988
1650.
3
69
1979
1590.
85.19 0
0 0
0
1989
1690.
3
70
1958
1570.
86.42 0
0 0
0
1990
3520.
3
71
1939
1500.
87.65 0
0 0
0
1991
10800.
3
72
1993
1490.
88.89 0
0 0
0
1992
1630.
3
73
1966
1380.
90.12 0
0 0
0
1993
1490.
3
74
1944
1380.
91.36 0
0 0
0
1994
1170.
3
75
1930
1350.
92.59 0
0 0
0
1995
2390.
3
76
1977
1250.
93.83 0
0 0
0
1996
6560.
3
77
1931
1200.
95.06 0
0 0
0
1997
2090.
3
78
1929
1180.
96.30 0
0 0
0
1998
1620.
3
79
1994
1170.
97.53 0
0 0
0
1999
2660.
3
80
1941
1050.
98.77 0
Efffffffffffffffffffffffffflfffffffffffffffffffffffffffffffffff�-4
-OUTLIER TESTS -
HIGH OUTLIER TEST
AAAAAAAAAAAAAAAAA
BASED ON 80 EVENTS, 10 PERCENT OUTLIER TEST VALUE K(N) = 2.940
0 HIGH OUTLIERS) IDENTIFIED ABOVE TEST VALUE OF 10878.
LOW OUTLIER TEST
AAAAAAAAAAAAAAAAA
BASED ON 80 EVENTS, 10 PERCENT OUTLIER TEST VALUE K(N) = 2.940
0 LOW OUTLIERS) IDENTIFIED BELOW TEST VALUE OF 579.7
AAAAP,A�AA�A,A,P,AAAAAAP,A�.�,AIIP,AAA,P,1�bAk��b}�,�b,�AAAP,A�P,�,1a�,�,AP,AAAAAAA,P,AA�j
-SKEW WEIGHTING -
BASED ON 80 EVENTS, MEAN -SQUARE ERROR OF STATION SKEW=-99.000
DEFAULT OR INPUT MEAN -SQUARE ERROR OF GENERALIZED SKEW = .302
FINAL RESULTS
-FREQUENCY CURVE- LANDSBURG USGS 12117500
$ffffffffffffffffffffffffNfffffffffffff&fffffffffffffffffffffff»
DRAFT 12/21/99
°
COMPUTED
EXPECTED
3 PERCENT
3 CONFIDENCE
LIMITS
°
°
CURVE
PROBABILITY
3 CHANCE
3 .05
.95
°
°
FLOW
IN CFS
3 EXCEEDANCE
3 FLOW IN
CFS
°
Q
11
0
16100.
17700.
3 .2
3 21700.
12800.
0
0
12500.
13400.
3 .5
3 16300.
10200.
0
0
10300.
10800.
3 1.0
3 13000.
8540.
0
0
8340.
8650.
3 2.0
3 10200.
7080.
0
0
6220.
6350.
3 5.0
3 7360.
5430.
0
0
4880.
4940.
3 10.0
3 5610.
4340.
0
0
3720.
3750.
3 20.0
3 4170.
3370.
0
0
2370.
2370.
3 50.0
3 2600.
2160.
0
0
1640.
1630.
3 80.0
3 1810.
1460.
0
0
1390.
1380.
3 90.0
3 1560.
1220.
0
0
1230.
1220.
3 95.0
3 1390.
1070.
0
0
1020.
1010.
3 99.0
3 1170.
863.
0
Ifffffffffffffffffffffffflfffffffffffffllffffffffffffffffffffffl
°
SYSTEMATIC STATISTICS
°
°
LOG TRANSFORM: FLOW,
CFS 3
NUMBER OF EVENTS
°
°
MEAN
3.3999 3
HISTORIC EVENTS
0
°
°
STANDARD
DEV
.2166 3
HIGH OUTLIERS
0
°
°
COMPUTED
SKEW
.6985 3
LOW OUTLIERS
0
°
°
REGIONAL
SKEW
-99.0000 3
ZERO OR MISSING
0
°
°
ADOPTED
SKEW
.7000 3
SYSTEMATIC EVENTS
80
°
Effffffffffffffffffffffffffffffff�ffffffffffffffiffffffffffffff'�
+++++++++++++++++++++++++
+ END OF RUN +
+ NORMAL STOP IN FFA +
+++++++++++++++++++++++++
************************************
*************************************
* FFA
*
* FLOOD FREQUENCY ANALYSIS
*
*
U.S. ARMY CORPS OF
ENGINEERS
* PROGRAM DATE: MAY 1992
*
*
THE HYDROLOGIC ENGINEERING
CENTER *
* VERSION: 3.0
*
*
609 SECOND STREET
*
* RUN DATE AND TIME:
*
*
DAVIS, CALIFORNIA
95616
* 07 MAR 00 11:18:08
*
*
(916) 756-1104
*
*
*
*
*
************************************
*************************************
DRAFT 12/21/99
INPUT FILE NAME: RENSTSK.DAT
OUTPUT FILE NAME: RENSTSK.OUT
DSS FILE NAME: RENSTSK.DSS
----- DSS --- ZOPEN: Existing File Opened, File: RENSTSK.DSS
Unit: 71; DSS Version: 6-GX
**TITLE RECORDS)**
TT FLOOD FLOW FREQUENCY ANALYSIS PROGRAM
TT FITTING THE LOG-PEARSON TYPE III DIST SKEW FORCED TO 7500 STA SKEW
TT 80 YEARS, 1920-1945, 1951,1997-99 FILLED BY REGRESSION WITH
12117500
**STATION IDENTIFICATION**
ID RENTON USGS 12119000
**JOB RECORDS)**
IPPC ISKFX IPROUT IFMT IWYR IUNIT ISMRY IPNCH
IREG
J1 0 3 0 0 0 0 0 0
0
**GENERALIZED SKEW**
ISTN GGMSE SKEW
GS 9000 .000 .70
**SYSTEMATIC EVENTS**
80 EVENTS TO BE ANALYZED
**END OF INPUT DATA**
ED+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
FINAL RESULTS
-PLOTTING POSITIONS- RENTON USGS 12119000
ElfffffffffffffffffffffffffNfffffffffffffffffffffffffffffffffff»
°
EVENTS ANALYZED
3
ORDERED
EVENTS
°
°
FLOW
3
WATER
FLOW
WEIBULL °
° MON
DAY
YEAR
CFS
3
RANK
YEAR
CFS
PLOT POS °
0 0
0
1920
2440.
3
1
1991
10600.
1.23 0
0 0
0
1921
2514.
3
2
1976
8800.
2.47 0
0 0
0
1922
6842.
3
3
1934
8164.
3.70 0
0 0
0
1923
5075.
3
4
1996
7650.
4.94 0
0 0
0
1924
3913.
3
5
1951
7058.
6.17 0
0 0
0
1925
3498.
3
6
1922
6842.
7.41 0
0 0
0
1926
2266.
3
7
1972
6210.
8.64 0
0 0
0
1927
2391.
3
8
1960
5860.
9.88 0
0 0
0
1928
5793.
3
9
1928
5793.
11.11 0
0 0
0
1929
1580.
3
10
1932
5793.
12.35 0
0 0
0
1930
1799.
3
11
1978
5670.
13.58 0
0 0
0
1931
1606.
3
12
1984
5540.
14.81 0
0 0
0
1932
5793.
3
13
1947
5510.
16.05 0
0 0
0
1933
5222.
3
14
1982
5320.
17.28 0
0 0
0
1934
8164.
3
15
1965
5300.
18.52 0
DRAFT 12/21 /99
0 0
0
1935
5075.
3
16
1990
5240.
19.75 0
0 0
0
1936
2490.
3
17
1933
5222.
20.99 0
0 0
0
1937
2366.
3
18
1935
5075.
22.22 0
0 0
0
1938
3049.
3
19
1923
5075.
23.46 0
0 0
0
1939
1990.
3
20
1987
5070.
24.69 0
0 0
0
1940
2465.
3
21
1950
4160.
25.93 0
0 0
0
1941
1412.
3
22
1953
4110.
27.16 0
0 0
0
1942
2403.
3
23
1924
3913.
28.40 0
0 0
0
1943
2784.
3
24
1969
3720.
29.63 0
0 0
0
1944
1837.
3
25
1956
3640.
30.86 0
0 0
0
1945
2576.
3
26
1959
3520.
32.10 0
0 0
0
1946
2860.
3
27
1975
3520.
33.33 0
0 0
0
1947
5510.
3
28
1925
3498.
34.57 0
0 0
0
1948
2750.
3
29
1955
3480.
35.80 0
0 0
0
1949
2780.
3
30
1957
3460.
37.04 0
0 0
0
1950
4160.
3
31
1999
3405.
38.27 0
0 0
0
1951
7058.
3
32
1954
3250.
39.51 0
0 0
0
1952
2190.
3
33
1983
3250.
40.74 0
0 0
0
1953
4110.
3
34
1974
3190.
41.98 0
0 0
0
1954
3250.
3
35
1961
3180.
43.21 0
0 0
0
1955
3480.
3
36
1995
3170.
44.44 0
0 0
0
1956
3640.
3
37
1973
3090.
45.68 0
0 0
0
1957
3460.
3
38
1980
3080.
46.91 0
0 0
0
1958
2160.
3
39
1938
3049.
48.15 0
0 0
0
1959
3520.
3
40
1981
3020.
49.38 0
0 0
0
1960
5860.
3
41
1964
2960.
50.62 0
0 0
0
1961
3180.
3
42
1967
2960.
51.85 0
0 0
0
1962
2570.
3
43
1968
2910.
53.09 0
0 0
0
1963
2340.
3
44
1946
2860.
54.32 0
0 0
0
1964
2960.
3
45
1943
2784.
55.56 0
0 0
0
1965
5300.
3
46
1949
2780.
56.79 0
0 0
0
1966
1570.
3
47
1997
2773.
58.02 0
0 0
0
1967
2960.
3
48
1948
2750.
59.26 0
0 0
0
1968
2910.
3
49
1971
2730.
60.49 0
0 0
0
1969
3720.
3
50
1998
2660.
61.73 0
0 0
0
1970
2290.
3
51
1945
2576.
62.96 0
0 0
0
1971
2730.
3
52
1962
2570.
64.20 0
0 0
0
1972
6210.
3
53
1921
2514.
65.43 0
0 0
0
1973
3090.
3
54
1936
2490.
66.67 0
0 0
0
1974
3190.
3
55
1986
2480.
67.90 0
0 0
0
1975
3520.
3
56
1940
2465.
69.14 0
0 0
0
1976
8800.
3
57
1920
2440.
70.37 0
0 0
0
1977
1340.
3
58
1942
2403.
71.60 0
0 0
0
1978
5670.
3
59
1927
2391.
72.84 0
0 0
0
1979
1840.
3
60
1937
2366.
74.07 0
0 0
0
1980
3080.
3
61
1963
2340.
75.31 0
0 0
0
1981
3020.
3
62
1970
2290.
76.54 0
0 0
0
1982
5320.
3
63
1992
2280.
77.78 0
0 0
0
1983
3250.
3
64
1926
2266.
79.01 0
0 0
0
1984
5540.
3
65
1952
2190.
80.25 0
0 0
0
1985
1660.
3
66
1958
2160.
81.48 0
0 0
0
1986
2480.
3
67
1989
2000.
82.72 0
0 0
0
1987
5070.
3
68
1939
1990.
83.95 0
0 0
0
1988
1800.
3
69
1993
1850.
85.19 0
0 0
0
1989
2000.
3
70
1979
1840.
86.42 0
0 0
0
1990
5240.
3
71
1944
1837.
87.65 0
0 0
0
1991
10600.
3
72
1988
1800.
88.89 0
0 0
0
1992
2280.
3
73
1930
1799.
90.12 0
0 0
0
1993
1850.
3
74
1985
1660.
91.36 0
0 0
0
1994
1170.
3
75
1931
1606.
92.59 0
0 0
0
1995
3170.
3
76
1929
1580.
93.83 0
DRAFT 12/21 /99
0 0
0
1996
7650. 3
77
1966
1570.
95.06 0
0 0
0
1997
2773. 3
78
1941
1412.
96.30 0
0 0
0
1998
2660. 3
79
1977
1340.
97.53 0
0 0
0
1999
3405. 3
80
1994
1170.
98.77 0
Effffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff�-4
-OUTLIER TESTS -
LOW OUTLIER TEST
BASED ON 80 EVENTS, 10 PERCENT OUTLIER TEST VALUE K(N) = 2.940
0 LOW OUTLIERS) IDENTIFIED BELOW TEST VALUE OF 791.2
HIGH OUTLIER TEST
BASED ON 80 EVENTS, 10 PERCENT OUTLIER TEST VALUE K(N) = 2.940
0 HIGH OUTLIER(S) IDENTIFIED ABOVE TEST VALUE OF 12688.
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA,Atip.p,AAAA,A,AAAAziAA,AAA,A1atiA�,
-SKEW WEIGHTING -
A_p,AA.AAAXIA,A,p,Ap,p,A,bAAAAAAAAp,AAAla p.p.p,AAp,AAAp,p,p,A,Aiik1AAAAAAAAp,p.AAAAp,p,AAlj
BASED ON 80 EVENTS, MEAN -SQUARE ERROR OF STATION SKEW=-99.000
DEFAULT OR INPUT MEAN -SQUARE ERROR OF GENERALIZED SKEW = .302
AAAAAAAAAAAp,p,AAAAA,p,AAAAAAAAAAAAp,p.AAA,A,p.AAAAAAA,AAA,p,p.A,AAAAA,Ap,p,AAAAA
FINAL RESULTS
-FREQUENCY CURVE- RENTON USGS 12119000
EffffffffffffffffffffffffNfffffffffffff&fffffffflffffffffffffff»
° COMPUTED
EXPECTED
3
PERCENT
3
CONFIDENCE LIMITS °
° CURVE
PROBABILITY
3
CHANCE
3
.05
.95 °
° FLOW
IN CFS
3
EXCEEDANCE
3
FLOW
IN CFS °
C,AAAAAAAAAAAAAAAAAAAAAAAA�AAAAAAAAAAAAppAApAAAAAAAAziAAAAAAAAAA,91
0 18400.
20100.
3
.2
3
24400.
14800. 0
0 14500.
15400.
3
.5
3
18600.
11900. 0
0 12000.
12600.
3
1.0
3
15000.
10100. 0
0 9860.
10200.
3
2.0
3
12000.
8450. 0
0 7470.
7620.
3
5.0
3
8760.
6570. 0
0 5940.
6010.
3
10.0
3
6780.
5320. 0
0 4600.
4630.
3
20.0
3
5120.
4190. 0
0 3000.
3000.
3
50.0
3
3270.
2750. 0
0 2110.
2110.
3
80.0
3
2330.
1890. 0
0 1810.
1800.
3
90.0
3
2010.
1600. 0
0 1620.
1600.
3
95.0
3
1810.
1410. 0
° 1350.
1330.
3
99.0
3
1540.
1150. 0
Ifffffffffffffffffffffffflfffffffffffffifffffffffffffffffffffffl
°
SYSTEMATIC
STATISTICS
°
QAAApAAAAAAAAApAppAAAAAAAAAAAAAAAfaAAAAAAAAAAAAApAApAAAAAAppAAAAl
° LOG TRANSFORM: FLOW,
CFS
3
NUMBER OF
EVENTS 0
DRAFT 12/21 /99
° MEAN
3.5008 3
HISTORIC EVENTS
0 °
° STANDARD DEV
.2050 3
HIGH OUTLIERS
0 °
° COMPUTED SKEW
.3540 3
LOW OUTLIERS
0 °
° REGIONAL SKEW
.6990 3
ZERO OR MISSING
0 °
° ADOPTED SKEW
.6990 3
SYSTEMATIC EVENTS
80 °
EffffffffffffffffffffflfffffffffflffffffffffffffffffffffffffffI'�
+++++++++++++++++++++++++
+ END OF RUN +
+ NORMAL STOP IN FFA +
+++++++++++++++++++++++++
Cedar River Spawning Channel
Hydraulic Design Consultation Report
1. Background. Development of a spawning channel on the left bank of the Cedar
River near RM * * * is being proposed to replace a previously constructed spawning
channel upstream of the proposed site that was destroyed by a landslide in early 2001.
The proposed spawning channel is anticipated to be used primarily by sockeye salmon,
and to a lesser degree by chinook and possibly coho salmon. This document provides
design guidance to the Corps of Engineers per the Scope of Work to Contract * * * * * *.
2. Design Criteria. The following hydraulic design criteria was used for design of
the spawning channel and is based on information contained in the publication "Fisheries
Handbook of Engineering Requirements and Biological Criteria" prepared by Milo C.
Bell, and information contained in the Seattle District Corps of Engineers "Cedar River
Additional Mitigation Project, Hydrologic and Hydraulic Analysis" dated 28 March
2000.
• Cedar River design discharge
250-1,000 cfs
• Spawning channel discharge
2.25-3.0 cfs/ft width
• Spawning channel depth
1.5-2.0 ft
• Spawning channel velocity
1.5-2.0 fps
• Spawning channel width
12-40 ft
3. General Design Confi urg, ation. The spawning channel will include an intake
structure to provide year-round consistent water supply from the Cedar River, a
trapezoidal shaped spawning channel, and an outlet structure to attract upstream
migrating fish to the artificial spawning channel. Features will be included as required to
protect the spawning channel from Cedar River discharges up to the * *% peak annual
recurrence event (i.e., the **-year frequency discharge). Bank protection and intake
structure stabilization as required will be provided to ensure spawning channel integrity
at Cedar River discharges up to that size event. Flood events larger than that may cause
damage to the facility.
4. Cedar River Hydraulic Data. Cedar River water surface elevations at the
spawning channel entrance and exit were based on data developed by the Seattle District
Corps of Engineers through the use of the HEC-RAS water surface profile modeling
software. Model boundary conditions were established from survey data collected by the
Seattle District, and on hydrologic data found in ******. Spawning channel entrance
water surface elevations for the design discharge range of 250 and 1,000 cfs are *** ft
and *** ft (msl), respectively. Channel exit water surface elevations for the design
discharge range of 250 and 1,000 cfs are * * * ft and * * * ft, respectively. The **-year
discharge frequency water surface elevations at the spawning channel entrance and e-
are * * * ft and * * * ft, respectively.
4. Channel Design. The spawning channel will approximate a trapezoidal cross
section having a bottom width of 20-ft and 1 vertical on 2 horizontal side slopes. With
this geometry the channel discharge will be on the order of 50 — 60 cfs. Assuming a
channel roughness (Manning `n') of 0.03 to 0.04 and a hydraulic gradient of 0.001,
normal depth and velocity with discharges of 50 cfs — 60 cfs ranges from 1.3 ft — 1.7 ft
and 1.4 — 1.8 fps, respectively. These values approximate those in the hydraulic criteria.
The channel entrance and exit structures will be designed to maintain a hydraulic gradient
of about 0.001
5. Entrance Structure. The channel entrance structure will be designed to supply a
discharge of 50 — 60 cfs to the spawning channel with a range of Cedar River flows of
250 — 1,000 cfs. Three different entrance designs were developed; (1) a concrete
overflow weir structure, (2) a rock overflow weir structure and (3) a headwall with pipes
structure. Debris control provisions are included with each design to prevent floating
material from entering the spawning channel. DETAILS TO BE DEVELOPED
AFTER CORPS PROVIDES MAPPING AND CEDAR RIVER RATING CURVE
INFO FOR THE SITE.
Incipient Motion grain Size, in
O O O O O N
O IN -N O-) 00 O N J m O
O O O O O O O O O O O
O `°
0
UG)-o
/
Z i X
\
ZmCo,
G) �t a
O r- �
O.
�
Z Z D
Z m w z 7
o
�
O
0
v�
u)
n
TI N
CD
(�
O
-w
-
(D
0
CD
^
O CD
C p.p
O
m D
ZMW
Z m -T,
O
G)
N O
O o
mD�
Z
Z m
G)
v o
G)
Cn
Z��
N.
Z m w
CDT (
�
C
v
3
v
3
CQ
�
(Q
�
X
X
O
O
O
O
N
O
CD
1:
i
Shear, Ib/ft2
0
(D
(T
CT
N
Cl)
G)
N
+
7
Q
7
EtL] E
CD
G7
G7
<ci:
ill
CD
-
CD
CD
O
C1
O
n
O•
�
•
CCr
f
_�
CD
O
rt
�
n
S
G1
CD
v
CD
i
CD
p
♦
3 O
p
i
CD
Q72.
E7
r. to
O
ED
N O
CD
i
O O
fA
Q
3.
r
•
m O
7
A
Q
j,
j
f
r
N
E
i
r
O
j
O
E
i
I
cQ
cQ
c=
n
�
�
Z3
�
�
x
0 000 CO
U) Ct) cn W J
c
a
0
0
U-
rn
rn
J
d
c
co
roll
0
0
0
N
c
0
V)
co
00
rn
rn
J
m
co
a�
U
()J) u01len913
cc
V SX NNVMJ3n0 300--L4 XO AdOO M
9 SX NNV9b3A0 300--L4 XO AdoO
O SX ANVEIH3n0 300--Ll XO Adoo
a SX NNV9b3n0 300--L1 XO AdoO j
3 SX )1NVMJEAO 300--SL Xo AdoO
0 SX NNV9�i3n0 300--56 XO AdoO
H SX ANV9213n0 300--5L XO ldoO
E
U� c
cn
I SX MV92i3AO 3o0--9 L XO Adoo T
C
f SX NNVSH3n0 300--51 XO LdoO m
L
U
c
X SX NNVOHi IAO 300--56 XO AdoO
I SX MV9HIAO 300-14 XO AdoO
W SX NNV9Hi IAO 3001--ti4 Xo AdoO
N SX ANV92i3A0 300--46 Xo AdoO v
ri
aX01 cCn
a
I
9
n
w
0
0
0
0
m V 00 cD O
O O O O O
C)
C/)
Cl)
N
C/)
co
r
m
ro
o0
0
0
o
cn
Ln
o
r
CD
O
CD
a
n
T
T
n
T
N
CDy
N
CO
O
T
m
m
n
Mom
D
O
D
m
y
y
n
(D
Q
2
rnfo-W-0»
O
m
m
r
r
S
m'
Oi c
N
O
O
C>
O
fnD
m
<
`D
co >
n
o
o
0
m
m
n
m
C_�
�'
m
(DI
CD
y
>Zm
-->
O
no
ZZ
T
N
m
O�
0
0)
7
7
m
cn
a
r:
-1
0
0)
19
ro
U)
X
m
n
D
C
D
m
c)
z
2
0
m
m
C)
Z�
C)
2
m
D
D
(/) O
O
O
O
(T
N
0
C)
2
D
m
V
cn
A
w
00
W
V
O
Ou
On
00
W
ao
A C%
O
0o
V
A
W
W m
z z z z 0=
m
O
a
G m
�J
O
m
N
-•
O
A N
O
(O
V
M
Z
G)
2
D
D
D
D n
2
W N
Z7
m
al
D
<
cn
m
r
O
n
V N
m
1I m
� �
_
Z
u
D
z
z
C�
2
D
z
z
m
r
m
O
cn
O
z
D
r
c
W
`m
n
O
X
m
cn
O
Z
c
U)
D
n
m
1n
m
D
r
m
N
n
O
x
c
K
V
D
0
N
0
w d CX 13
w
w
Copy CX 14--COE OVERBANK XS N
Copy CX 14--1COE OVERBANK XS M
Copy CX 14--COE OVERBANK XS L
m
n
s
v
Copy CX 15--COE OVERBANK XS K
co
0
N
Copy CX 15--COE OVERBANK XS J
0
m
Copy CX 15--COE OVERBANK XS 1
Copy CX 15--COE OVERBANK XS H
Copy CX 15—COE OVERBANK XS G
Copy CX 15--COE OVERBANK XS F
Copy CX 17--COE OVERBANK XS D
Copy CX 17--COE OVERBANK XS C
Copy CX 17--COE OVERBANK XS B
COPY CX 17--COE OVERBANK XS A
w
0)
Elevation (ft)
U)
�
ai
�
'',
ro
0
W '.; W 7
N
O
O
O
iT
O
CD
Z7
;u
'�
a
.Z�1
0
co
a
v
1T
_N
CD
1
'0
r
co
cD
07
m
cn
m
Q
O
N
0
0
0
U)
Cedar River Spawning Channel Replacement Hydraulics & Hydrology Design Report --Draft
TABLE OF CONTENTS
1.0 Purpose......................................................................................................
2.0 Project description.........................................................................................
3.0 Background.................................................................................................
3.1 Restoration Goals and Concepts................................................................
3.2 Site Selection......................................................................................
3.3 Design Criteria.....................................................................................
4.0 Hydrologic Analysis.....................................................................................
4.1 Data................................................................................................
4.2 Methodology......................................................................................
4.3 Results..............................................................................................
5.0 Hydraulic Analysis.........................................................................................
5.1 Channel Inlet Selection...........................................................................
5.2 Hydraulic Modeling...............................................................................
5.3 Results.............................................................................................
5.3.1 Channel Inlet Selection...............................................................
5.3.2 Hydraulic Analysis....................................................................
5.3.3 Sensitivity Check.....................................................................
6.0 Geomorphic Analysis.....................................................................................
6.1 Data and Methodology...........................................................................
6.2 Results.............................................................................................
7.0 Stability Analysis........................................................................................ 1
7.1 Data and Methodology..........................................................................
11
12
7.2 Results............................................................................................
15 8.0 Hydraulic Design........................................................................................
15
8.1 Inlet...............................................................................................
15
8.2 Channel...........................................................................................
15
8.3 Riprap..............................................................................................
8.4 Large Woody Debris............................................................................ 5
8.5 Dendrites.........................................................................................
16
16
8.6 Bank Stabilization...............................................................................
8.7 Other Design Considerations..................................................................16
8.7.1 Right of way and Utilities...........................................................16
8.7.2 Geotechnical.........................................................................16
8.7.3 Site/Civil..............................................................................17
8.7.4 Public Access........................................................................17
8.7.5 Monitoring............................................................................17
LIST OF FIGURES
Figure 1: Spawning Channel Inlet Low -Moderate Flow Rating Curve for Varying Roughness and
GateOpenings................................................................................................
Figure 2: Spawning Channel Inlet Flood Flow Rating Curve for Varying Roughness...............7
Figure 3: Bank Stable Grain Size vs Flow and Roughness.............................................13
Figure 4: Channel Stable Grain Size vs Flow and Roughness..........................................14
LIST OF TABLES
Table 1: Channel Spawning -Flow Hydraulic Parameters................................................8
Table 2: Channel Flood -Flow Hydraulic Parameters.....................................................9
Cedar River Spawning Channel Replacement Hydraulics & Hydrology Design Report --Draft
1.0 Purpose
This document reports the results of the hydrologic and hydraulic analysis for the Cedar River
PL99 project to replace the Overdredge Mitigation Channel destroyed in the February 2001
earthquake. This work item is outlined in the Hydrologic Engineering Management Plan
(HEMP) dated 10 December 2 00 1.
2.0 Project description
The project is located in the floodplain of the Cedar River between River Mile (RM) 3.3 and 3.6,
within the city limit of Renton WA. The project is a proposed replacement of the spawning and
rearing habitat destroyed during the February 2001 earthquake. That channel was constructed as
mitigation for the Cedar 205 dredging project. The project involves constructing a culvert
diversion and a channel on a high -flow bench that is protected from annual flooding. The intake
and channel are designed to maximize spawning area, while minimizing the site footprint.
Generally the channel is designed to be stable over the range of expected flows. Where risk
allows, portions of the channel will be allowed to naturally conform to the flow regime. Large
Woody Debris (LWD) will be incorporated to increase bank stability and to provide habitat
complexity. The intake system is designed to take a varying quantity of flow from the Cedar
over the entire range of expected flows. Site constraints include a steep valley wall with
episodic geotechnical instability, buried utility lines at the downstream end, public and private
right-of-ways, and local public resistance to the proposed project. Site amenities include a
morphologically stable reach, native gravel -cobble substrate, flood protection, readily available
large wood, and restricted public access.
3.0 Background
3.1 Restoration Goals and Concepts
The goal of the project is to provide spawning and rearing habitat for Sockeye salmon,
and secondarily to improve spawning and rearing habitat for Coho, Chinook and other
salmonids. A pool -riffle side channel with a river fed intake was selected as the
preferred concept. The riffles will encourage sockeye spawning, the pools, Chinook.
Based on resource agency comments the ratio of pools to riffles should not exceed 1:4.
Because the Elliot channel produced favorable spawning conditions, it was used as a
partial design template. Visits to other Cedar side channels indicate that over -steepened
side slopes are a source for shade, gravel, and woody debris. And with constructed side
channels, it was found that a predictable, controlled water source was used with success
in similar river basins in British Columbia (Canada Department of Fisheries and
Oceans). This led to pursuit of a structural intake system, but a softer channel. It was a
project goal to include over steepened banks into the project design where feasible, and
to accept that the constructed channel will evolve in time in response to regular fish use
and flood flows. The large volume of available woody debris will limit migration of the
channel in response to flood flows. It was also recognized that critical facilities must be
protected from flooding and erosion.
Seattle District USACE 8/26/2002
Cedar River Spawning Channel Replacement Hydraulics & Hydrology Design Report --Draft
3.2 Site Selection
Two sites were considered for the channel replacement project. The Rolling Hills Site,
on the left bank at RM 3.3-3.6 was the primary site, the existing Elliot Channel at the
Maplewild golf course was the alternate site. The site constraints and benefits governed
the scope and scale of the project at each site. The Rolling Hills site and the Elliot
channel were compared with respect to improvements in riparian habitat, cost,
maintenance, flood damage risk, and social -economic impacts.
The Rolling hills site was selected as superior, largely because the river is
geomorphically more stable, and the flood risk is lower. It is believed that the damage to
the Rolling Hills site in the long run will be less severe, and thus the fry survival rate
will be greater. Citizen input added the soft constraint of limiting the clearing of the
cottonwood canopy forest and locating the maintenance path to the landward side of the
channel.
3.3 Design Criteria
Project team members and resource agency specialists worked together to establish the
design criteria for the project. The design criteria were used to size an inlet and channel
to provide average and minimum spawning conditions comparable to the Elliot Channel
(average stream bed slope 0.3%. depth between 0.5-1 ft, velocities 0.7-3.3 ft/s).
Additionally it was recommended that the pool riffle ratio not exceed 1:4, that channel
dendrites be added to safely flood the channel during large floods, that large woody
debris be incorporated into the design without anchoring. Rock and other structural
features should be limited to where necessary. It was desired to evaluate the performance
of the design over a range of flows that represented the rearing and spawning windows.
4.0 Hydrologic Analysis
The concern for this project is to provide adequate flow in the spawning and rearing channel
during the time period when the channel is in use. The operational time frame for this channel is
from mid -October through May, although the project is intended to operate year-round. The
design also considered use by early spawners (late summer to mid -October). Hydrologic
investigations completed for previous Cedar projects and the 2000 Cedar River Habitat
Conservation Plan were used to determine the low flow, design flow and flood flow conditions.
4.1 Data
The Instream Flow Agreement (IFA) for the Cedar River Watershed Habitat
Conservation Plan (HCP) has been finalized. The IFA set minimum flow levels in the
Cedar River measured below Landsburg Dam (USGS 12117500). These were adjusted
to reflect the expected minimum flow levels at Renton. As part of the 1999 Additional
Mitigation Channel Design ("Elliot Channel"), simulated weekly streamflow values at
Renton (USGS 12119000) using data from water years 1929 to 1992, and incorporating
the constraints proposed in the IFA, were used to define a low flow duration curve. The
duration curve values for incremental flows between Landsburg and Renton were taken
Seattle District USACE 8/26/2002
Cedar River Spawning Channel Replacement Hydraulics & Hydrology Design Report --Draft
from the HCP technical appendices. Flood flow data was obtained from previous
USACE work on the Cedar and from the 2000 FEMA FIS for the City of Renton.
4.2 Methodology
It was assumed that there was a negligible incremental inflow to the Cedar River between
the site of the proposed spawning channel and the gage at Renton. The minimum flow
requirements are valid for the flow just below Landsburg. It was necessary to determine
the flow at Renton that corresponds to the minimum flow levels imposed at Landsburg.
A partial -duration frequency analysis for low flows, consistent with ER-1110-2-1450 and
EM 1110-2-1415, was conducted to develop the weekly low -flow duration curve for
Cedar River at Renton.
Once the weekly low flow duration curve was established for Renton, the incremental
duration curve for flows between Landsburg and Renton was subtracted from the low -
flow duration curve at Renton. This is an estimate of the low flow duration curve at
Landsburg. For the Elliot Channel the design low flow was taken as the flow at Renton
that had the same percent of time exceeded as the normal minimum instream flow
requirement at Landsburg. With the HCP in place, the weekly high normal, low normal,
and critical low flows are defined. Although the Elliot results are applicable, it was felt
that they captured too broad a time scale to define early spawning and other low flows of
interest. Thus for the Rolling Hills site, the actual weekly minimum HCP flows at
Renton were used directly to establish the operating range at low flow. In highwater
years, the project will generally experience flows higher than the low flows used for
design.
4.3 Results
From the Elliot Channel Analysis:
For the period 15 October to 02 June, the normal minimum instream flow requirement at
Landsburg is 250 cfs with a 98.8 percent of time exceeded. This corresponds to a flow at
Renton of approximately 300 cfs. This value was considered the lower limit of design
flows for the hydraulic analysis at Elliot Levee.
From HCP (2000) (used for study):
Early Sockeye Spawning (9 Sept-7 Oct): 138-273 cfs, weighted average 152 cfs, critical
annual minimum flow — 100 cfs
Peak Sockeye Spawning (8 Oct- 30 Dec): 365-540 cfs, average "low normal" flow
440cfs, average "high normal" flow 468 cfs, critical minimum flow 243 cfs.
From FEMA FIS (2000) (used for study)
10 yr: 5,834 cfs, 50 yr: 9,708 cfs, 100-yr: 11,450 cfs, 500-yr: 15,830 cfs
From USACE (1995) (2, 5-yr used for study only):
2-yr: 3,200 cfs, 5-yr: 4,900 cfs, 10-yr: 6,175 cfs, 20-yr: 7,900 cfs, 50 yr: 10,100 cfs:
12,000 cfs, 500-yr: 22,000 cfs.
Seattle District USACE 8/26/2002
Cedar River Spawning Channel Replacement Hydraulics & Hydrology Design Report --Draft
5.0 Hydraulic Analysis
The hydraulic analysis section of this report includes discussions of the channel inlet
system selection, hydraulic modeling, geomorphic analysis, and stability analysis.
Methods, Data and Results are presented.
5.1 Channel Inlet Selection
NHC Inc. was retained to provide preliminary design alternatives for the inlet system,
based on their expertise with similar projects. Three alternatives were presented: A rock
berm, a logjam, and a culvert intake. Project team members evaluated the alternatives
and selected the culvert as the most appropriate system because head losses and ground
disturbance needed to be minimized, and specific hydraulic conditions were required that
could be reliably met by a culvert. This system was refined as needed to better match the
topography and other project constraints. Guidance was supplied by Canada Department
of Fisheries and Oceans experts who have used similar structures successfully. As stated
previously it was preferred to have a reliable and controlled source of flow to the channel
to meet spawning goals. It was determined that the culvert option, although more
complex to design and construct, offered, the most reliable source of flow, with the least
head loss, and least ground disturbance. Additionally the culvert gate allows for
dewatering during maintenance and flushing of fines.
5.2 Hydraulic Modeling
A backwater analysis was required to design the inlet system to create the desired
hydraulic conditions, and to analyze the stability of in -channel features (substrate on bed
and banks, LWD). A HEC-RAS model developed for the 2000 FEMA FIS was modified
for this study. This model was calibrated to a moderately high flow. Additional
calibration data were gathered and incorporated into the model as part of this study. The
HEC-RAS model included the main river channel, the spawning channel, and the inlet
and outlet works. Flooding impacts were analyzed.
Manning's n values typically increase for a given reach for lower flows due to the larger
influence of the channel's roughness elements. The channel roughness coefficients were
increased to, on average, 0.055 in the channel and 0.120 on the overbank. This is about a
20 percent increase in roughness over the original calibration done for the FIS. The
roughness values used in the recalibrated file are consistent with documented values for
stony channels at lower stages.
The previously calibrated model was checked for accuracy by comparing the computed
water surface elevations to the surveyed water elevations between RM 3.3 and 3.6 during
discharges of 2000 cfs in April 2002. The model reproduced the surveyed elevations to
within 4 tenths of a foot at the inlet during this relatively high discharge, although some
sections downstream of the inlet did not match the surveyed water surface as closely.
Seattle District USACE 8/26/2002
Cedar River Spawning Channel Replacement Hydraulics & Hydrology Design Report --Draft
All elevations in this document are reported using NAVD 88 vertical control.
New model cross sections were added to the main -stem between RM 3.3 and 3.6. These
cross sections were a combination of interpolated channel data for the main -stem Cedar,
and recent ground survey of the floodplain where the proposed channel will be located.
No new river cross -sections were taken. The channel is very stable, with little observable
deposition or scour (Perkins Geosciences, 2002). Thus it is unlikely that the FEMA
cross-section geometry has changed significantly.
Given the uncertainties inherent in every computer model, two staff gages were installed
at the proposed inlet locations to monitor conditions during low flow to verify the
hydraulic model and design before construction begins.
5.3 Results
5.3.1 Channel Inlet Selection
Careful consideration was given to the inlet selection. The options were a culvert, log
jam, and rock berm inlet. The rock berm was ruled out as environmentally and
hydraulically infeasible. Although the logjam inlet reflects successful natural analogs, the
required ground disturbance, limited life span, and extra risk were viewed as negatives.
A project at this site demands a reliable source of flow and protection against erosion of
the steep hillside. It is difficult to reliably quantify the hydraulics of the log inlet,
although there is substantial evidence of their success in nature. The need for meeting
spawning goals makes a reliable and flexible water source attractive. The concern over
hillside erosion caused by the river overtaking the channel (unlikely but possible) makes
the culvert inlet attractive because there are over a hundred feet of land separating the
inlet from the channel. A benefit of the culvert in this stable reach is that most of the
materials will last for decades baring maintenance, although metal components will
require corrosion control or occasional replacement. A "natural" inlet will degrade with
time, requiring eventual replacement of the rotting trees. Limitations of the culvert are
construction effort and cost, and use of a non-native materials. The worst -case scenario
for the culvert is if the river severely degrades or aggrades (unlikely but possible). In this
case the river would have to be re -graded (i.e. boulder riffle), or the culverts and headwall
would have to be reset.
The culvert intake design consists of a 3 ft by 3 ft concrete box culvert with a concrete
headwall and wing -walls, and a metal trash rack. A control gate would be mounted on
the downstream end where the box culvert enters a vault. The vault is located
approximately 170 ft upstream of the head of the spawning channel. A 4 ft diameter
corrugated metal (outlet) or HDPE culvert would connect the vault to the spawning
channel. The inlet culvert, manhole, and headwall will be precast, the wingwalls, cast -
in -place. Alternately two 24-inch parallel metal pipe culverts could be used in lieu of the
box culvert, but they offer less of an operating range and are more difficult to maintain.
Seattle District USACE 8/26/2002
Cedar River Spawning Channel Replacement Hydraulics & Hydrology Design Report --Draft
(Note: the culvert headwall is tentatively being redesigned. Riprap and quarry spalls will
be used in lieu of a concrete headwall and wingwalls. Woody debris will be incorporated
if feasible.)
5.3.2 Hydraulic Analysis
Low flow and high flow rating curves were developed for the site and are illustrated in
Figures 1 and 2. Roughness values were varied from 0.04 to 0.12 to capture a reasonable
range of expected hydraulic conditions. In Figure 1, only two gate settings are shown,
the maximum opening (3 ft) and the minimum opening (0.5 ft). This gives the operating
envelope for the channel for flows not exceeding 5,000 cfs in the river. It is important to
point out that the gate opening to gate discharge relationship is not linear, that is, a ten
percent increase in the gate opening does not result in a ten percent increase in flow. It is
expected that the City will field test the channel and gate to develop as -built rating
curves. When main river discharges exceed 5,000 cfs (5-10-year frequency), the high -
flow bench where the channel would be constructed begins to overtop. Discharge would
increase easily by a factor of ten if the channel were flooded. Thus the gate setting is
irrelevant because the proportion of flow from the inlet is small relative the proportion
from the river overtopping the channel banks. This is reflected in Figure 2, where the
rating curve at high flow depends only on the amount of roughness in the spawning
channel. The large difference in spawning channel discharge for different roughness
values occurs because discharge is inversely proportional to roughness, and because the
flooded channel area is large. Armored swales will be constructed to intercept overbank
flow and flood the channel before banks are eroded by overbank flows. Additionally, the
control structure for the inlet system would be located out of the 100-year floodplain.
Flood elevations during the 100-year event could be reduced by as much as 1.0 ft,
attributable to the conveyance added by the new channel. If large amounts of LWD are
added to the channel, the conveyance benefit will be negligible.
Seattle District USACE 8/26/2002
Cedar River Spawning Channel Replacement Hydraulics & Hydrology Design Report --Draft
Figure 1
Spawning Channel Inlet
60
Low -Moderate Row Rating Curve
Gate opening
for Varying Roughness and Gate Openings
0.5ft, Chn=
50
0.04
Gate opening
a;
3.0 ft, Ch n =
T 40
0.04
L
Channel Overtops,
y
Culvert Flow Insignificant
Gate opening
o
30
\
\i
0.5 ft, Ch n
�
=0.12
L)
-- `
_ .-._. Gate opening
- 20
3.0 ft, Ch n =
0.12cc
I
a
J
N 10
0'
0 1000 2000 3000 4000 5000 6000
Main River Discharge, cfs
Figure 2
Spawning Channel Inlet
1800 Flood Flow Rating Curve
for Varying Roughness
1600
42 1400
d
t 1200
N
0
°rn 1000
A
N
1
Q
a+ 600
C
C
A
L
U
o, 600
AT 400 ---"" Gate opening 3.0 ft, Ch n = 0.04
200 —Gate opening 3.0 ft, Ch n = 0.12
0
5000 6000 7000 8000 9000 10000 11000 12000
Main River Discharge, eft
Seattle District USACE 8/26/2002 7
Cedar River Spawning Channel Replacement Hydraulics & Hydrology Design Report --Draft
For all analyses it is assumed that no flow is lost in the channel, that the gate on the inlet
culvert is fully open, and that debris does not obstruct flow upstream or downstream of
the inlet. The gate can be closed to optimize the hydraulic conditions in the spawning
channel over a range of flows. The figure below compares the effect of a fully open and
partially open (6-in) gate on discharge to the spawning channel. The figures and tables
below describe hydraulic conditions expected throughout the channel over a range of
spawning and flood flows, assuming the channel roughness and inlet configuration
described above.
Table: 1 CHANNEL SPAWNING -FLOW HYDRAULIC PARAMETERS
Spawning Period
Minimum Flow in
Max Flow in
Range of Depth
Range of Velocity
River based on
Channel
(Average)
(Average)
Cedar HCP
(Flow at 6" gate
opening)
Summer Lowest
97
6 (5)
0.3-1.7 (0.6)
0.3-1.6 (1.0)
Q
16-22 SeptL'�
150
7 (6)
0.4-1.7 (0.7)
0.3-1.5 (1.1)
1-7 Octj-'
270
10 (8)
0.6-2.0 (0.9)
0.3-1.5 (1.2)
"90%Exceed"
300
11 (8)
0.7-2.1 (0.9)
0.3-1.5 (1.2)
Peak Critical
240
10 (7)
0.6-1.9 (0.8)
0.3-1.6 (1.2)
Peak Low Normal
440
14 (9)
0.8-2.5 (1.1)
0.3-1.6 (1.3)
Peak High
470
14 (10)
0.9-2.6 (1.1)
0.3-1.6 (1.3)
Normal
I/Normal minimum during early spawning period
From the Table 1, it is clear that the proposed channel will meet all the design criteria
during normal low flow conditions. Compared to the existing condition of the Elliot
channel, flow depths and velocities should be somewhat greater, due to the greater flow
capacity of the inlet. The bracketed discharges in the Max Flow column represent the
discharge to the channel if the culvert gate was closed to 6 inches. The corresponding
hydraulic parameters can be estimated by comparing with similar full -open discharges.
Despite the 80% reduction in the culvert opening, flows are cut only 30% during peak
spawning conditions. During high flows, as shown in the following table, the discharge
reduction is more significant. However, during high flow events that don't overtop the
channel, backwater will significantly increase depths and reduce velocities in the channel
resulting in deposition of suspended fines. As the backwater recedes, shear stress on the
bed should clean most if not all of the deposited fines. However, the more the culvert
gate is open, the more effective the silt transport and removal. If desired, field-testing
could be conducted to determine the best gate setting to balance fish hydraulics, habitat,
and sediment management.
Seattle District USACE 8/26/2002
Cedar River Spawning Channel Replacement Hydraulics & Hydrology Design Report --Draft
Table 2: CHANNEL FLOOD -FLOW HYDRAULIC PARAMETERS
Spawning Period
Minimum Flow
ax Flow in Channel
Range of Depth (Average)
Range of Velocity
in River
Flow at 6" gate
(Average)
pening)
50%EXCEED
1,000
22.5 (14)
1.1-4.0 (1.8)
0.2-2.3 (1.3)
—1-YR
2,000
39 (18)
2.0-5.9 (3.2)
0.2-2.9 (1.1)
—2-YR
3,000
46 (21)
3.3-7.4 (4.7)
0.1-2.4 (0.7)
4,000
46 (21)
4.6-8.9 (6.0)
0.1-1.9 (0.5)
5-YR *
5,000
46 (21)
5.7-9.8 (7.0)
0.1-1.6 (0.4)
10-YR *
5,800
790 (NA)
5.5-6.2 (5.9)
2.7-6.2 (4.1)
50-YR
9,700
1500 (NA)
9.0-11.2 (9.9)
3.0-7.0 (4.5)
100-YR
11,450
1790 (NA)
10.3-12.3 (11.2)
3.1-6.3 (4.4)
* Cedar River overtops into spawning
channel between 5 and 10-year events
5.3.3 Sensitivity Check
Because the above parameters were based on a single assumed channel n-value, a
sensitivity test was performed on the 90% exceedence flow and the 50-year flood flow.
Low flow n-values were adjusted +/- 50% (.034-.068) and the model was re -run. The
average channel depth ranged from 0.8 ft to 1.1 ft (- +/- 15%) and the average velocity
ranged from 1.0-1.5 ft/s (+/- 20%). Even within this range of error, the hydraulic design
criteria are met, showing that the channel and inlet configuration are robust. The high
flow channel n-values were adjusted upwards by 67% (0.06 to 0.10) to capture energy
losses from flow resisting LWD blockages. With the increased roughness, the average
channel depth ranged from 11 ft to 12 ft (- + 15% of average) and the average velocity
ranged from 2 - 4 ft/s (— - 38% of average). Thus woody debris that increases hydraulic
roughness should dissipate some of the erosive energy. This analysis does not
incorporate the pool riffle sequences and LWD structures that will be constructed. This
means that the as -built depths and velocities will be more variable than those shown here.
6.0 Geomorphic Analysis
6.1 Data and Methodology
Sue Perkins of Perkins GeoSciences was retained to provide an assessment of both the
Elliot channel and the Rolling Hills site. Her assessment of the Rolling Hills Site is
contained below. USACE staff inspected the site on several occasions, during high and
low flow. USACE observations follow Ms. Perkins' results.
6.2 Results
The following are excerpted from Sue Perkins Geomorphic assessment for the Rolling
Hills Site:
Seattle District USACE 8/26/2002
Cedar River Spawning Channel Replacement
Hydraulics & Hydrology Design Report --Draft
• The spawning channel should be kept far from the valley wall to minimize the risk
of burial in the event of a landslide, and to reduce the very low risk of river migration
to a position next to the valley wall where it could destabilize it.
• Drainage from roads, storm drains, and houses should be routed away from the
potentially unstable slope above the project site.
• The inlet design should account for potential future degradation, perhaps with a
permeable inlet such as a log jam that would allow flow to enter over a range of
depths. More information could be gained about recent rates of bed degradation by
comparing survey data from the new King County flood study with the 1980s FEMA
flood study, if any of the cross-section pairs are located close enough for comparison.
If the only water source is at the upstream end, sand is likely to drop out in the
lower half of the channel due to the lower gradients and backwater flooding
conditions. Unless permeability is such that groundwater would add a significant
amount of flow, a second inlet halfway down the channel may be needed to provide
additional flow to keep fine sediments moving through. There is a good example of
this upstream at the Maplewood spawning channel, where a natural side channel
enters partway downstream. During our site visit this month, the constructed
spawning channel's bed was covered with sand upstream from the side channel, but
was clean gravel downstream of the added water source.
As with any channel with a controlled inlet, there may not be enough flow to
scour the gravels and keep them clean over the years. If there were enough flow to
scour the gravels, then there would be no way for new gravel to move into the
channel to replace the gravel moved downstream. In addition, future downcutting of
the river could potentially reduce flow in the channel. The need for repeated
maintenance, design modifications and repairs should be anticipated throughout the
life of the project.
Were it not for the increased risk of landslides that would occur if the river were
to undercut the valley wall, I would advocate an uncontrolled channel opening and
allowing the river to shape the new channel as it wished. Over time, this would
provide more numerous maintenance -free side channels that provide a variety of
habitat functions in addition to spawning gravels. This still may be the best option
given the potential long-term maintenance and sediment problems alluded to above.
With this approach, the money spent on inlet construction would instead be spent to
reinforce the outer boundary of the desired channel migration zone. Other than
excavating a proto-channel and seeding it with LWD, minimal engineering of the side
channel would be needed since the river would do the engineering for you.
Given the low risk of channel migration at this site, the use of hard bank armoring
can probably be limited to the approaches to the inlet, a controlled overflow area
partway down the channel, and the downstream outlet (where it is essential to end the
spawning channel before it encounters Seattle's water pipeline). I concur with many
of Cygnia Freeland's suggestions (1/18/02 memo, Dept of Ecology) to construct a
channel with as many natural characteristics as possible. These could include an
engineered log jam at the inlet through which water flows, steeper bank angles similar
to those occurring along the river, an excavated floodplain along one bank, and
Seattle District USACE 8/26/2002 10
Cedar River Spawning Channel Replacement Hydraulics & Hydrology Design Report --Draft
clusters of logs with rootwads that are large enough to stay in the channel without
anchoring.
Site visits show that the preferred inlet location is about 150-ft upstream of a riffle in a
straight, slower, deeper reach. The cobble bank is steep and armored by small alder trees.
The inlet is proposed just downstream of a vegetated boulder and submerged woody
debris. The velocity shadow and scour hole of this feature results in the settling of some
sand, however the bed is visibly dominated by gravel and cobble. This bank feature
should be retained to ensure the scouring function. A gravel bar appears to be located
towards the middle of the river, just outside the proposed inlet. It is unknown if this
would limit flow to the inlet during minimum discharge conditions. If necessary, woody
debris or boulders could be placed to scour deposited gravels away from the inlet, or to
raise the water surface if degradation occurs.
An alternate location downstream is just upstream of the aforementioned riffle. Flow is
swift and shallower. The bed is course, dominated by cobbles with few visible fines. At
low flow it is uncertain if depths will be sufficient to sustain flow to the channel.
Stability of the riffle is critical for the success of this alternate location. Erosion of the
riffle could cut off flow to the inlet. Without in -river grade control work, the inlet invert
would have to be re -set to match the new grade.
At this time two staff gages have been installed, one at each of the two inlet locations
described above. If the results of the monitoring indicate that the alternate location is
superior, the inlet design will be modified to match that location.
7.0 Stability Analysis:
7.1 Data and Methodology
The stability of placed gravel, in -situ soil, and large woody debris were analyzed in
several ways. The hydraulic models were used to compute variables necessary for the
stability calculations. Design charts were used to compare the stability of bed and bank
soil and erosion control materials with respect to the shear stress and velocity. A critical
shear bank stability analysis was used to determine the stable natural bank side -slope. A
formal incipient motion analysis was undertaken to determine the grain size that would
first be transported given specific hydraulic forces. The above were combined to identify
where erosion protection is necessary. The stability of placed wood was not rigorously
analyzed. Buoyant forces were computed to determine required ballast for bank
revetments incorporating logs. Hydraulic forces acting on logs subject to flow were not
rigorously determined because in -channel logs will be embedded in banks, and sized and
grouped together where possible to resist hydraulic forces en masse. In -field hydraulic
calculations can be performed to check these designs if required. As logs decay, some
installations will become unstable. Bank vegetation should be well established by then,
and in -water logs should still provide cover and structure.
Seattle District USACE 8/26/2002 11
Cedar River Spawning Channel Replacement
Hydraulics & Hydrology Design Report --Draft
The maintenance commitment will be largely determined by the stability of the placed
gravel. Some bank erosion will occur during major floods and in reaches of the channel
that are designed to conform to the flow regime and recruit gravel. Gravel nourishment
stations should be located during construction to allow for convenient and effective
gravel placement.
7.2 Results
The results of the critical shear analyses are shown in Figure 2-4. For a given roughness
value, the average condition is shown, along with the maximum. The average condition
represents the conditions to expect in the majority of the channel. The maximum is
generally much greater than the average, and typically occurs at one of two locations —
during high flows, at the culvert outlet, during low flows, where the channel rejoins the
river. The maximum should be considered an upper range for typical conditions,
however larger grain sizes will be mobilized at channel bends, riffles, around LWD
during high flows, and by spawning. The results also compare the effect of an inlet
culvert gate opening of 3-ft (max. open) versus 6-inches (min. open).
The results show that the large majority of the design gravel specification is stable during
spawning flows and over a wide range of higher flows and channel roughness conditions.
Also, increasing roughness tends to decrease discharge, increase the friction slope, and
thus increase the size of particle that can just be moved by the flow. Effective roughening
of the channel will occur at low flows, especially in riffle sections, because the influence
of the grain size on roughness increases inversely with depth of flow. At higher flows
and depths, roughness is supplied by in -channel LWD and overhanging vegetation.
Closure of the gate reduces discharge and increases the stability of the bed and bank
materials.
The banks are generally less stable than the bed due to the additional gravitational force
exerted on the bank material. As shown in Figure 3, the computed stable bank size
ranges from 0.1-in to 6.8-in, with an average of about 0.8-in. As shown in Figure 4, the
stable grain sizes on the bed range from 0.1-in during the 5-year flood backwater
condition to 4.5-in at the outlet of the channel during the 10-year event, with an average
of about 0.5-in. Overtopping floods (10-year and greater) are predicted to be much
more erosive. Gravel and cobbles finer than 2.5 inches on the bed and 4.0-in on the
banks may be eroded. This will result in armoring of some portions of the streambed,
and bank scour where not sufficiently protected. Although the model predicts that
increases in roughness will increase the stable grain size, this is only generally true.
Around roughness features such as logjams, sediment movement will be restricted and
finer particles will deposit.
Periodic gravel nourishment will be required to replenish gravel mobilized from high
flows and spawning activity, especially following an overtopping flood. Normally stable
vegetated side -slopes should withstand erosive forces during large floods that fill the
channel in a controlled manner. Otherwise, uncontrolled overtopping will severely erode
Seattle District USACE 8/26/2002 12
Cedar River Spawning Channel Replacement Hydraulics & Hydrology Design Report --Draft
the channel banks. Over- steepened sections will erode over time and in response to
flows to a more stable configuration. This erosion will supply gravel to downstream
reaches. Imported streambed material should be well graded, and include rounded
cobbles up to 5-inches. If more than 30% of the material is finer than 0.5 inch, frequent
gravel nourishment may be necessary.
Seattle District USACE 8/26/2002 13
Cedar River Spawning Channel Replacement Hydraulics & Hydrology Design Report --Draft
FIGURE4: CHANNEL STABLE GRAIN SIZEVS DISCHARGE AND ROUGHNESS
5.00
M r
s
_
f
4.50
4.00
T %
On
3.50WMIM
2 - f iF}}++3 s`" -d
OW
V
"
3.00
. „,
N,
* A w 0.04-AVG
-:
Vi
3
0.04-M4,X
Z
�� 6
W
2.50
0.12 AVG
cr�
0.12-MAX
, . . , , .
��
f � rrss,4Ml
1.50
SUPN
( ff+ FAR` `5 � 'F,Yi >g ✓F � x✓ �/ 0Xf � 9 �'
1.00
0.50 .
"�"�
4
0.00
90% exceed Peak 1000 cfs 2-yr
5-yr 10-YR 50-YR 100-YR
low flow, 3- Spawning
ft gate Flow 3-ft
opening gate opening
FLOW OF INTEREST
Seattle District USACE 8/26/2002 14
Cedar River Spawning Channel Replacement Hydraulics & Hydrology Design Report --Draft
8.0 Hydraulic Design
8.1 Inlet
To provide the necessary flow to the spawning channel at the design low flow, the invert
of the 3 ft x 3 ft concrete box inlet culvert should be at elevation 56.17 ft (NAVD88) at
the inlet and 56.00 ft at the outlet. The invert of the 4 ft diameter outlet culvert should
be 56.0 ft at the inlet and 55.50 ft at the outlet. Invert of the vault should be 54.0. The
invert of the headwall apron should be 55.65. The wingwalls should be 45 degrees from
the culvert centerline. The protruding end of the outlet culvert should be mitered to the
vault wall. (Note: The inlet headwall has been tentatively redesigned as follows: The
inlet will be protected by a riprap headwall. The box culvert end will be cut off to a 1.5
to 1 side -slope to match the riprap headwall slope. A gravel filter will be placed around
the culvert end protected by the riprap headwall to reduce the void -space and prevent
the piping of fines. The culvert end will be underlain by 2.0 ft of riprap for scour
protection.).
8.2 Channel
The channel bottom at the upstream end should be 55.0 ft elevation at the downstream
confluence the invert would match the invert of the river channel (52.5 ft)
The excavated channel depth varies from 14 ft at the upstream end to 4 ft or less at the
downstream end. The channel geometry that met the design criteria was a trapezoidal
channel, approximately 900 feet in length, with a 10.0 ft bottom width, 1.0-1.5H: 1V
side slopes, and an average bed slope of 0.003 ft/ft (0.3%). The design spawning
channel roughness coefficient was 0.045, corresponding to a typical gravel bed channel
at low stages. At high flow, the LWD will begin to block flow and increase hydraulic
roughness (assume at least a factor of two greater during full effect).
8.3 Riprap
In order to protect critical infrastructure during large floods, a 3-ft thick, Class 3-4
riprap blanket should be placed at slopes no steeper than 1.5 H:1 V. See plans for
placement locations and details.
8.4 Large Woody Debris
LWD is a desired restoration project component because it provides both habitat and
hydraulic function. Analogs for use of LWD at this site are newly formed side channels
where erosion topples trees into the channel and the rootball and trunk armor the cut
bank, and the tree spans the channel, and alternately, lays downstream parallel to flow.
Wood anchored as shown in the plans should withstand a moderate to severe
overtopping flood. Severe floods will both transport some wood not securely anchored
from the site and undermine and recruit new trees on site. Until waterlogged, the large
tree sizes at the site may require ballast to resist the buoyant and drag forces exerted by
the flow. By leaving large rootballs intact, the additional frictional resistance will
counteract the hydraulic forces on the log. If the cumulative backwater effects of LWD
Seattle District USACE 8/26/2002 15
Cedar River Spawning Channel Replacement Hydraulics & Hydrology Design Report --Draft
is significant, the hydraulic forces on the channel, banks, and other LWD pieces will
decrease. If necessary, ballast can be provided by burying the majority (2/3 of length)
of a log in the bank, anchoring the log to other logs, placing large boulders on top of the
logs, and anchoring the log to the earthen slope with a dumbbell driven through the log
into the bank. LWD may be difficult to place in some portions of the channel, due to
the large size of the members, excavation depths exceeding 10 ft, and limited access.
Most cottonwood not permanently inundated will begin to rot after a 7-10 year period.
This interval should allow the native bank vegetation and side -slopes to mature. LWD in
the channel and pools could be replaced as needed.
8.5 Dendrites
Dendritic channels should be constructed at two locations shown in the plans to provide
additional rearing and spawning area and to provide a predictable overtopping
mechanism for the channel during flood events. The dendrite will consist of a swale
armored with a 2-ft thick riprap blanket that extends from the river's edge to the
spawning channel. The dendrite will be backfilled with spawning gravel and LWD will
be located at the confluence with the channel. During high flows, water will
preferentially flow down the dendrite to the spawning channel, supplying gravel, and
flooding it before overtopping flows can erode the rim of the channel.
8.6 Bank Stabilization
Side -slopes excavated to 1.5 H:1 V are computed to be stable under high flow
conditions, however, experience shows that unless the slopes are re -vegetated or
armored, the slopes will erode where flow impinges on the bank. Fortunately vegetative
growth is robust on site. Native topsoil and vegetation will be conserved and replaced
to aid re -vegetation. LWD is readily available to provide temporary stabilization of
excavated slopes, especially at the toe.
8.7 Other Design Considerations
There are several important considerations that relate to the H&H aspects of the project.
These are briefly addressed below.
8.7.1 Right of way and Utilities
Access to the site for maintenance needs to be secured by the City. Buried utilities need
protection should the river overtake the constructed channel. See the plans for buried
riprap details.
8.7.2 Geotechnical
The steep hillside south of the project must be protected from becoming unstable during
construction and from river migration following construction.. Additional investigations
are required to ensure that the project will not de -stabilize the hillside and jeopardize the
channel and properties above. The contribution of groundwater is assumed nil. If a
reliable groundwater source is found, the inlet may be scaled back or eliminated. The
Seattle District USACE 8/26/2002 16
Cedar River Spawning Channel Replacement Hydraulics & Hydrology Design Report --Draft
underlying soil appears to consist of gravels and cobbles. Thus importing spawning
gravel may not be necessary. The likelihood of seepage into the groundwater table has
not been determined, nor the availability of groundwater to supplement the surface
intake.
8.7.3 Site/Civil
Wasting excess material close to the project site will create a significant cost and time
savings. This will allow for more effort to construct the complicated features of the
restoration project such as geogrid walls and LWD structures. The extra care in these
areas should improve the long-term success of the project.
8.7.4 Public Access
A more natural ("messy") channel configuration with steeper side -slopes, higher
excavation depths, and large amounts of in -stream woody debris may discourage
poaching and vandalism by limiting access to and restricting movement within the
channel. Tamper proof trash -racks and manhole covers are necessary to protect public
safety and prevent vandalism of the control works.
8.7.5 Monitoring
It is recommended that the City periodically monitor channel cross sections to record the
changes in the channel geometry. This will allow a means of tracking the gravel
transport rates, effectiveness of gravel nourishment, success of re -vegetation, assessing
general channel stability, impacts of inlet operations, impacts of LWD, and impacts of
overtopping floods.
Seattle District USACE 8/26/2002 17
0
Upper Chilliwcck Tour
Resource, Fl\`estoration sites
Upper Chilliwack Tour, February 2002
Resource Restoration Sites
Site 1: Yukalup Side Channel.
Built in the summer of 1997, this project was funded by the British Columbia Watershed Restoration
Program and designed and constructed by Fisheries and Oceans Canada for approximately $100,000.
The project is fed by a 2ft-diameter instream intake that supplies approximately 20cfs to 2000 m2 of
spawning habitat and 4400 m2 of rearing habitat. Primary production for this project will be 2200 coho
smolts, 500,000 pink and chum fty annually.
Site 2: Centre Creek C.C., Camp Channel.
The Centre Creek Camp Channel Project, constructed in the summer of 2000 is a British Columbia
Watershed Restoration Program funded project built in partnership with Cattermole Timber and Fisheries
and Oceans Canada for approximately $60,000. This project, fed by two 16" pipelines that supply
approximately 20cfs, has created 1100 m2 of spawning habitat and 4,300 m2 of rearing habitat. The habitat
is designed to primarily benefit pink and coho salmon populations found in this part of the Chilliwack
River. Using standard estimates of production this area could produce up to 2200 coho smolts and 500,000
pink and chum fry annually.
Site 3: Angelwing Channel.
Built in the summer of 1996 as a partnership project between the British Columbia Watershed Restoration
Program and Fisheries and Oceans Canada for approximately $100,000. This project, fed a 2' diameter
concrete pipe, that supplies approximately 20cfs, has created 105,OOOm^2 of off channel habitat. The
habitat is designed to primarily benefit pink and coho salmon populations found in that part of the
Chilliwack River watershed. Using standard estimates of production this area could produce up to 50,000
coho smolts and 10,000 chum and pink fry annually.
Site 4: Borden Creek.
The Borden Creek Project, constructed in the summer of 1997 as a partnership project between the British
Columbia Watershed Restoration Program and Fisheries and Oceans Canada for approximately $245,000.
This project has created approximately 1100 m2 of spawning habitat and 25,000 m2 of rearing habitat
and is designed to primarily benefit steelhead and coho salmon populations found in that part of the
Chilliwack River. Using standard estimates of production this area could produce up to 12,500 coho smolts
annually.
MAR-J3-02 12:46 AM EFH CONSULTING LTD 6045217143 P-01
Rcochne Fisheries Habitat
Consulfing LiA.
s c
0 tK N N) Ck
71
E P_
Ir
200 — 4170 Still Creek Drive, Burnaby, BC V5C 6C6
Tel: (604) 918-5097 Fax: (604) 299-4511 WA Tel: (206) 371-7608
E-mail: Infonaecocline.com Internet: www.ecocline.com
MAR—,13-02 12:47 AM EFH CONSULTING LTD 6045217143 P-02
N
-71
Fate of Coho Salmon (Oncorhynchus kisutch) Carcasses in
anis
Spawning Stre
C. J, C ederho I ni
wa;hington Department Of Naturil Re>ourcvs, f,ptc,t Lind 'AWA!"t 11:0111 Oivisiofl, Olyrnl,ia, V%!A 98504 USA
D. B. Houston and D. L. Colc'
U.S. National P-14 A," tItA 96162 (.15A
W. ). Scarlett
Wjihinolo.n Department of Nartiral Resourr1% BOX 13175 Forks, M 98331 USA
L
Cederholm, C. J., D. B. Houston, D. L. Cole, and W, 1, Scadeit. pj8c Fite - of coho salmon (Oncorhynchus
).
kisutch) carcasses in spawning streams. Can. 1. Fish. AqUat. Sci.46.1347-1355.
J chU5 kisuich) carcasses released
We eximined the levels of retention and utilization of 945 Crho sa
lmon almon (Oncorhyn
Jy11)picPeninSt,1l-1 �%Ijjjington. Most carcasses were Telained
Kz atly into Seven spawning strf.,,irns on the 0
experiment, debris caught sqdM1d`ma6y
In acent foTtAs,,few were flushed beyond 600 rn, organtc de
2 species ofma-linials and birds. The distancq%thai;.. i0l
th mass w9ronsurned by 2
h occurrence of freshets and inversely to debris lood,
dire y tot e occ omen
ain carcasses has probably been redu e#Y 01
ngin the capacity of many streams and rivers to rel,
activities. 'lie importance of cohn carcasses to PuFA11,11 tons of carnivores anti to the dynamics of lotic food webs
nicrils additional study
-imin6s les taux de retention etd%itil;si&on C109,5 czrcas<vs de saurron quinnat (Oncorhyrichus
Les auteurs ont ex. kisutch) placdes pour 1'exp6riencp daps Sept fraytres naturolles do li p&insule Olympic clans Veat du Was-
hington. La majoritd des carcasses ont dt6 retenues dins lcs tours d'ua,j el dans !es fort,,s adjacentes, peu ont
616 trinsport6e.0 plus Cie 600 ni du point o6 elles oill 616 organiques ont retenu de nornbreuses
es )\c(,
carcasses. La maleure panic: de la chair des poissons a 66 cunsorrim,"o 22 Par sdernamnAres et d'oiseaux.
La distance de transport des carcasses semble titre diret; tot;Wnt !;6, i, la prt4ence de trues nivales et inversement
a par I ca ipacit6 (if., retention des carcasses de
rnivores La c,
-bris et .
lite h 1, :hirge de debris i la consummation p,
beaucoup dei cours (I'vau a probiblerymni Ot`! r6li.1110 SOOS I44("t dk,, icflvitor> humaines. L'importance des
niclue Cie lit chiline alimentaire lotique
- s de (11.linnaI5 l3ollulations de (ilrilivote.> %:,. pour ki clynal
carcasse pour lt-
devrait etre otudi6e plus it fond.
�u le 17 octobre 1986
Rc
Receiyod October 17, 1988 Accopid le 6 avril 7989
Acr(qirt-d April 6, 1989 # Nit
(J9913)
C,
0
rivate Harry
Fisher, forest
14th Infantry U.S. Anny, emerged
ftom dense rain onto the Jloo(lplain of the Queets
; IRiver on 23 September 1890, arter spending nearly 3 mo exploring Washington's Olympic Mountains. He was impressed
by1he abundance of salmon, noting that -_great salmon
thms1lied in the water all night ion;, in their efforts to ascend
the stream. Wild animals which 1 could not see snapped the
bathes in all directions, traveling tip and down in search off ish.
At every few yards was to be seen the remains of a fish where
cougar, coon, otter, or eagle had made a Meal" (Fisher 1890).
Th 1 9. abundance of adult salmon spawning in the waters of the
01 . ympic peniniula his declined dramatically in the intervcnirl.g
near-cen(ury because of extensive uverl'i%l1ing, habitat k""Tit_
due to ioggirig, and dam construction (Ellis 1977. Brown
*J,;1PFlv1C 1982( Houston and Contor 1984). The decline of
llielllkiry of Don Colo \010,41L!,l before tbi,!,study
teas C0111pleted.
'Deccascd
adub salmon raises important questions relating to strean, pro-
ductivity, food web complexity, and carnivore abundance
beci,us^ salmon carcas."s, if retained in,*' U ,could repfescril.
an important input of nutrients to the 'latent llplt`o�
Ian communities, 6hts contained
Accumulating evidence suggestsocktpo
in salmon carcasses affect aquati,
salmon (oncarhynchus nerka), ino piked k0wi
nec form. contributed nitrates and' is.
downstrealln from spawning ar the` U6$R
(Donaldson 1967; Niathisen 1972; rok i n 1975), And'
in-..1cp.4c1 ilitirient concentrations and primary production Of
spawnitio �treillll'i in California (Richey et all. 1975). Dccoin
C�
position of pink salmon (0. gorbascha) carcasses increased the
nitrogen concentrations of a spawning Stream and the adjacent
csitutry in Alit4a (Brickell and Goering 1970). Carcasses also
influonced the nature of the st.,diments deposited in the. estuary.
pink jilli chuni salmon (0. keta) added substantially to the sea-
1347
Can. J. Fith. Aqk,d S(L. Vol, 45, 19sy
MAR-t13-02 12:48 AM EFH CjONS�}ULTING LTD 6045217143
P.03
Strg1r0f
LA PUSH
EIWVANGT0N
ANADA
Juan de
Fuca
PORT
ANGELES
8ockrnn`r So/ Due R1Ve
r
J 'F
P FORKS
13o0chiel River
V
y\ .
H o h Ri v e r
f��7 rer
I�, t
�e
Quests tL`
N
Not to Scale j
Fio. 1. The westem Olympic Peninsula, Waeliingion. showing locatinns of the seven study streams.
U.S. Highway 101 shown as bruken line.
"TreLa 1, Channel characteristics, debris loads, and winter lowflows
for.fi0t1-m retches of the seven study streams.
Channel LOD'
Winter - --
lowflow X width
Stream (m'ly) (nt) Sinuutiny" No, Density
AinkLake` 6.52 14.61.2i(1M1�._ -1.2,-
dst Twin 0.25 18.8 1.2 206 1.8
Bear 0.24 19.2 2.3 114 1.0
'`go�kman 0,06 12.3 1.4 97 1.3
,Dickey 0.10 14.8 1.3 134 1.5
?;tiMinter 0.14 17.5 1.4 189 1 8
"Hulow 0.34 24.3 1.2 181 1.2
"Large organic dehris. Density LOD pieces - 100sil.
°Cocrficient or thalwes icilgth divided by down valley length.
`A lake occurs at the headwaters.
sonal concentrations of nutrients in Snieaton Bay, southeast
' t.A]aska (Sugai and Burrell 1984).
Comparatively little is known about the role of coho salmon
:.(0. kisuich) carcasses in stream ecology. Coho are ubiquitous
.o� the Olympid Peninsula, spawning in small streamr.�. spring
cieeks, and side channels of large rivers. Fish live about 9 d
N from the onset of spawning in this region (van den Berphc and,
dross 1986). a � �ttidies suggest that carcass retention
. }tc,p��itiv with the amount of instream debris
crholm an Peterson 985). Understandittig the rate of car-
ikko& e
sasses may he of practic;tl importance to managers responsible
for producing and harvesting wilt! salmon. Moreover, the head
watery of most rivers sin the Olympic Peninsula are now within
Olympic National Park, a 3800 km2 natural area established in
1918, Knowledge of po.sible effects from the reduction of.
salmon on stream ecology and on terrestrial carnivores is
important to park mangers charged with maintaining the area
as a dynamic "vignette of primitive America." Here we report
the results from a comparative mensurative experiment (sensu
Hurlbcrt 1984) conducted from 1984-86, designed to deter-
inine the levels of retention and utilization of coho carcasses in
spawning streams.
Study Area
Seven coho spawning streams were chosen for study on the
western Olympic Peninsula (Fig. 1, Table 1). Three streams are
within Olympic National Park (Mink Lake, West Twin, and
Harlow), tw'o are on state lands administered by the Washington
Department of Natural Resources (Minter, Bockman), and two
are on private land (Dickey, Bear). Winter low flows on these
relatively low gradient, third -order streams ranged from 0.06-
0,52 m,i,.
Ilic streams drain a variety of Quaternary glacial deposits
and Eocene sandstones (Tabor and Cady 1978). Vegetation is
mostly coniferous forest dominated by western hemlock (Tsuga
heleropltY11(t), western redcedar (Thuja plicara). Sitka spruce
(Pirca sirc•hcnsis). and Douglas -fir (Pseudolsuga menziesit)
(Fonda and Bliss 1969; Franklin and Dymess 1973; Fonda
1974). Corridors of riparian vegetation contain big leaf maple
(Accr mucrophyllum), vine maple (A. circinarcrnr), black cot-
tonwtx)d (Pnpalus trichocarpa), and red alder (Alrats rubra).
Those study streams within the park are near pristine; those
114k
Con. J, Fish. Arynat'. St I., 41(4. 46. 1989
MAR-13-02 12:48 AM EFH CONSULTING LTD
p
6045217143 P.04
6utsoe the park flow through logged furests ranging in ag:c
frore recent clearcuts to remnant old growth.
'"`+e area receives the greatest precipitation in the contcrmi-
United States (NOAA 1978). Annual precipitation aver-
U ebbut 300 cm and falls mostly as rain. Mcan annual tem-
`"tttre at Forks is 10°C, with January as the coldest month at
i4°Cand July —August the warmest at 16°C (Phillips and Don
gtdson 1972).
The combination of moderate temperatures, steep slopes,
fdott 'drainages, and great annual precipitation nieans that
;mitts are characterized by very rapid rises and falls in dis-
*6iie and by frequent high flows, Fallen trees are abundant in
i':#i$dne stream channels, and this debris modifies stream mor-
r,0.40gy and water velocity by forming pools, brcudcd channels,
;ptev (Swanson and Licnkaemper 1978).
''A 600-m section of coho spawning habitat on each stream
was trapped and flagged at 25-m intervals, with the 0-m point
..located at the upstream end of each section. The positions of
ton greater than 10 cm in diameter and 3 to long (large organic
de 3, LOD) were plotted. Temporary staff gauges and an
;`el*ttonic current meter (Marsh McBimey) were used to meas-
,umwater levels and flows. Continuous recording thermographs
(Ryan Model "J") were used to monitor water temperature. A
`study section represented the length of stream and riparian cor-
ddor that could be searched thoroughly in about one-half day.
Underwater searchgs were aided by glass-hoitomcd "viewing
tubes," —I in long, constructed of 10-cm diiuneter PVC plastic
with a waterproof flashlight attached. Each stream sup-
d low density populations of spawning coho, except where
iVere blocked by a road culvert 100 m below the West T« in
:section.
le hundred and forty-five freshly spawncd carcasses of
coho salmon (470 males, 475 females) were measured
length, in ce►ttimetres), weighed (kilograms), and tagged
gh the back below the dorsal fin with numbered 2.5-cat
.ter Peterson discs, Additiomilly, small hales, 6 men in
titer, were punched through the caudal fin, dorsal fin, and
ula of each carcass to help distinguish planted from wild
ridto code dates, Small 40 MHz radio transmitters (Smith -
models P-40-10001, and P-40-500L)'were inserted into
tophagi of 174 coho, and their jaws stitched closed to retain
ansmitters. Transmitters were distributed in proportion by
ss size, sex, and release point. All carcasses were obtained
fish hatcheries after they had been spawned artificially.
Eery returns were so poor in 1985 that 2.68 carcasses were
!n for up to 2 wk before being distributed. Otherwise, car-
s were fresh, and were usually distributed within 1 d of
Issing. The consumption of previously frozen carcasses
tallied separately and comp;ircd with fresh c4rctrsses.
Mo different field experiments were conduc:tcd: (1) autumn
ors to determine the fate of carcasses during extended
4s under varying streamtlaws, (2) a high flow study to
ment carcass movement during a flood,
"Imn Studies
k*.*4inc hundred and twenty carcasses wcrc dividcd into eight
experimental releases In seven stream, with 13ockrnan Creck
used during, both 1984 and 19li5 Crablc 2). 'three lots of car-
casses, representing 0.25, 0.50, and 0.25 or the totals, wcrc
Con. !. Fiat, ly r,;r. Sri„ Vol. 46,
distributed at 10±4 d intervals to approximate the timing and
relative abundance of wild spawners. Carcasses were distrib-
uted in small numbers at 25-m intervals starting from the 0-m
mark; total numbers/stream varied between years (Table 2).
Carcass density averaged approximately 1.2±0.50 carcasses/
100 m1 between the ordinary high water levels on all the study
streams. Carcasses were released into the thalwegs (i.e. the line
of maximum water depth) of the streams.
c;ire mses of fall run coho were used in all the streams except
Mink Lake Creek, where an unusual summer run coho stock
was useJ. Fall run coho typically spawn in late November —
December; summer coho spawn from October —mid November
(Houston 1993).
The stream sections and adjacent 30-m of forest along each
bank were searched by teams of four to six persons at intervals
of approximately 3 to 4 d initially; on most streams the interval
was lengthened to —10 d after 4 wk, as carcasses were con-
surrted or decomposed. Harlow Creek, in the backcountry of
the pork, was searched less frequently. Streams were surveyed
an average of about 10 - 3,6 times for up to 118 d, (Table 2).
All streams, except Mink Lake Creek, had about the same num-
ber of freshets during the study. Five of the freshet days at Mink
Lake occurred near the end of the survey period, after many
carcasses had been consumed.
Information recorded for each carcass found included: stream
meter location, instream deposition site (riffle, pool, eddy,
gravel, nr sandbar), instream retention clement (free-floating,
buried, snagged on large or small organic debris (SOD), or on
ruts), tissues remaining, and, if on land, the distance (in
metres) from strcarribank to the remains. Radio -instrumented
carcasses were located on cacti survey with a Smith -Root model
SR-40 receivcr. Field dataloggers (Smith -Root model FDL-
10ER) wcrc installed at the 600-m marks to monitor passage
of radio,d fish.
Eighteen additional carcasses were dissected in the labora-
tory to determine the proportion of body weight contributed by
various orans, and this information was used to estimate the
amounts of tissue consumed in the field studies. Radio-instru-
rncntcd cvcirsses were used to prorate the noninstmmented fish
that Aerc tint rcobscrved, on the assumption that both types
behaved similarly. We believe this is reasonable; the small
trtnsmitiers (22 or 45 g) represented only about 0.8-1.6% of
the mean carcass weight.
Birds wcrc identified and counted on the study sections dur-
ing each survey. Mammal tracks and sign were recorded in snow
or mud. Up to six track plates (Barrett 1983) and two to three
Polaroid cameras rigged for multiple automatic exposure (Goetz
1981) wcrc used per stream in 1994, and night -vision goggles
(I'PT l;lectro-Optical, model AN/PUS-SA) were briefly tried in
1985. Two to three snaptraps (bloodstream Corp. Museum
Special) were set at sites where carcasses were being consumed
by small mammals. Finally, we simply observed stream sec-
tioriS from blinds. The cameras and plates performed poorly in
the cold weather and high humidity; most information was
gained from the other techniques. Species were tallied as con-
%umint•, citrcnr se only when observed feeding, when tracks left
nu d� ubt, or wlicn tnipped on site.
High Flow Study
Twcrity-five additional radio -instrumented carcasses were
released into West Twin Creck on 17 January 1986 at the begin-
nin tit' ;t mujnr storm that raised the measured discharge of the
1319
'A
::fit;
MAR-13-02 12:50 AM EFH CONSULTING LTD 6045217143 P_05
TABLE 2. Distribution and survey
c1lort ror 920 coho carcaacs comprising, eight releases into seven
Spawning streams'.
_
No. surveys
16�Stream
Total No.
(X survey interval
in days)
Freshet
days'
fi
carcasses released Survey period (days)
_
Mink Lake 160
38 (31 Oct.— 7 Dec.)
12 (3.2)
10
,::'
West Twin 100
62 (26 Nov.-27 Jan.)
10 (6.2)
S
Bear 100
63 (25 Nov.-27 Jan.)
4 (7.0)
7
f
Bockman 1984 160
118 (21 Nov.-19 Mar.l
10 (7.4)
7
Bockman 1985 1W
56 (27 Nov.-22 Jan.)
8 (7.0)
6
Dickey 100
56 (27 Nov.-22 Jan.)
9 (6.2)
9
6
7
Minter 100
59 (26 Nov.-24 Jan.)
(6.6)
Harlow 100
52 (27 Nov.-19 Jan.)
5 (10.4)
5
'Work was carried out in Mink Lake and Bockmun Creek during
1984, Hockman Creek was used
again in 1995 along with five other streams.
'Number of days stream -flow was bank full or in flood, using Calawah River gauging station.
TABLE 3. instream movement and retention of 920 coho salmon
carcasses.
1
Stream
No.
released`
Final location of carcasscs2
Stream Bunk Flushed Unknown
Mink Lake
160(53)
10(ll)
79(89)
0(0)
11
Jest Twin
100(26)
45(63)
25(36)
i(l)
29
car
100(29)
42(44)
51(53)
3(31
4
ockman 1984
160(34)
73(76)
19(20)
3(4)
5
""`Bockman 1985
t , :..
100
49
36
0
15
.` hickey
100
34
53
0
13
Minter
100
6.3
30
0
7
Marlow
too
30
26
0
4;
`(Number radio -instrumented carcasses)
f .0 ' 'Percent by category (Adjusted percent by prorating unknowns from
;..`radio -instrumented fish).
.$.
�;aiijacent Hoh River to 16 times the average low flow of the
evious 6 wk, and produced flows in the creek estimated at
mils, 25 times the low flow. Carcasses were monitored
r!.; .�Iour times through 3 February as the stream returned to lo%;.,
flow. Survey techniques were similar to those used during the
`atitumn studies.
'Results
S
Autumn Studies
;t..,
arcass retention
The number of carcasses retained within the study sections
f was generally high, few were flushed beyond the 600-m points
f
_4,Y'g-Table 3). The final distribution of carcasses on the four streams
5 � Qonmining radio -instrumented fish suggested that 4% or fewer
:carcasses were flushed downstream. The unknown category
.
,:(tagged carcasses never reabserved) was prorated on the basis :. the distribution of instrumented carcasses, Apparently, few
' -`"'carcasses on the remaining streams. i.e. those without radio-
J. vostrumented fish, were flushed; high numbers were accounted
}, or by survey, with unknowns 515% on three streams. Further,
the spatial distribution of found carcasses (see below) suggested
`t .t few drifted beyond the measured sections. Harlow Crcck.
here 44% were in the unknown category, was a possible
exception. but we suspect this high proportion of unknowns
resulted more from infrequent surveys combined with hifh
consumption than from differences in retention,
The proportion of carcasses located on stream banks by the
end of the surveys varied among streams. The proportions of
radio -instrumented fish located on basks, instream, or flushed
did not differ from the distribution of noninstrumented fish ;
found on any stream (G-tests, P>0.05, Sokal and Rohlf 1981).
Unconsumed carcasses had decomposed considerably by the
end of the surveys, at water temperatures of from 0 to 7.5°C;
but decomposition rates were not measured quantitatively. Three
hundred and forty-four carcasses observed over time were
coated with fungus by 14±9.4 d after release.
Carcass movement
The di i inces drifted by the carcasses yielded frequency dis-
tributions that were often sharply skewed, so median distances
were used to characterize movements (Fig. 2). The overall
median distance drifted was 49.5 m for 773 carcasses (84% of
the 920 relensed) retained within the study sections over all.
}tuning: only 19eio of the retained carcasses traveled more than
200 m; only 951.. more than 300 m. Median distances drifted
differed among streams (multisample median test P<0.05);
distances were greatest in Bear and Bockman Creeks and were
associated with the lowest LOD and carnivore feeding recorded
(Fig, 3).
Carcass retention and deposition
Over all streams, large+:ttr�` bds s
the most. important instreQr'
(Table 4). Carcasses were.0 .
The occurrence of buried a sf'likly°`
titnatcd overall. Most (52%) of the carcasses observed in the
process of being buried were in Bockman Creek, which seemed
to have a particularly unstable streambed (perhaps due to exten-
sive clearcut logging activities upstream),
Pools represented the most important instream deposition site .
for carcasses (Table 5). Many of the' pools that retained car-
casses (either free on the streambcd, pr retained by LOD) were
formed twcause of the effects of L.00 on streamflow.
Ceircasc consumption
individual carcasses were consu*d in varying amounts;
from less than 1 %, when eyes only were removed by birds, to
complete consumption. The percentage of the total fish mass
eaten was often high (Table 6), and was greatest on Mink Lake
Creek wal ul Olympic National park, due primarily to black
bears. Ilowcver, high consumption also occurred outside 0lym-
p is Nark. particularly on Dickey (71 %) and Bear (68%) Creeks.
11$0 Can. J. ViA. Agaor. ,Sc•L, Vol, 46. 1989
MAR713-02 12:51 AM EFH CONSULTING LTD 6045217143 P-06
60' MINK LAKE
WE5T TWIN
90
20
10
BOCKMAN 85
t
I
DICKEY
t
MINTER
t
t
HARLOW
t
DISTANCE DRIFTED (m)
Ftc. 2. Distance drifted by coho salmon c:ucnsses rclaincd in ipawning streams• Bars represent 20-m
intervals, broken lines show medians.
tamptionlevels atHarlow Creek were not determined accu-
Observations of bear activity early in the survey and the
I complete consumption of carcasses on banks suggested
"jarge numbers of carcasses were removed, but that traces
not found by the infrequent surveys.
ivores moved most carcasses only short distances. Of
tarcasses removed to banks and at least partially eaten,
were found within 10 m and 88% within 15 nt of the
.
s. Again, Harlow Creek differed; SO% of 26 carcass
ruins were found from 15 to 90 m from streantbanks.
Trhe number of carcasses consumed may have been under-
eo't*nated on all streams because of our inability to find remains.
ndent searches for tagged remains were conducted JDY
teams of four persons each on 14 November 1984 along
the banks of Mink I.akc Crcck. p rizntblc �urvcy c,tim:tte
(Magnusson et al. 1978) suggested that 76 curcasscs were an
ttiee banks to be found. The tennis actually found 75 carcasses,
Cdn. J FjOl• ,hiuut, Sci., Vol, 46, 109
for a survey efficiency of over 98%. This high efficiency prob-
ably diet nat apply over extended periods on any stream as the
taped reinain� occasionally disappeared entirely. Indeed, even
on Mink Lakc Creek bears returned several tags in scat at later
dates. I-luwe\•cr, this test, plus the frequency of survey on most
streams and the proximity of remains to streambanks all sug-
o,ested that a hick pc�portion of the cotisvmed carcasses was
f()und. Harlow Creek was an exception. because survey fre-
quency was lower. Fable; 6:
(Presumably. much of the fisE >
demaining Instream or on banks•• .
by bacteria and fungi, and
Initial freeing of some carcasses prior to their release appar-
ently made little difference to their consurdption; the proportion
of pr,aviously frni.en carcasses showing consumption did not
dirfer from fresh carcases on any stream (G-tests. P>0.05).
1351
MAR-13-02 12:52 AM EFH CONSULTING LTD
6045217143
P.07
180r— 4 T,\nt^P.5. Final instrrain de
G 120 lkpusition site
t
N 7
s 2
i'V 60
:U 0 �� — 50 J�Q
1 2 0 Go�S
J LOD DENSITY
tv laic. 3. Relationship of the median distances Qrifted by salmon car-
casses to large organic debris (LOD) load and to the percentage of
t' . tmasses known to be wholly or partially consumed, LOD density is
rttunberof pieces, 100 m-' for the uprper 340 m of study stream bacaw�c
few carcasses traveled bcyonJ this distance, Streams arc.: (1) Mink
Lake; (2) West Twin; (3) Bear; (4) Bockman 19a4,. (5) Bockman 1985.,
(6) Dickey; (7) Minter; (8) Harlow.
TABU 4. lnstream retention element for 605 coho salmon carcasses'
.� Cilrca ssc 5
Retention element No. Percent
` LOD/Sol) 338 �+. 061
Live roots and brunches' 30 (5)
Free on strcambed 204 (3 1)
Buried 33 (5)
Total W5
'Last lnstream bighting, 25l earcasws subaalucntly moved to
s • stmambenks by canivores. No(c, the difference between the 773 car
a.'- .':sasses known to be retained and the 605 tallied here represent 168
*k• carcasses removed to hanks by carnivores before they wcrc sighwd
Z tnstrearn. Also, deposition sites were not recorelcd tier 37 carc:r.scs
S`:'�during several early surveys', thus the difference in totals bemcen
•, x.' Tables 4 and 5.
'Riparian trees and shrubs.
. ` Consumers
Forty-three taxa of mammals and birds acre accorded on the
study streams, and 51 % of these were found to consume sal men
carcasses (Table 7). The number of consumer species/stream
averaged 6.1 t 2.59 and ranged from three on Harlow Creel to
1 I on Mink Lake Creek.
r . The 14 species of mammalian consumers ranged in size from
;;, shrews to black bears, with raccoons and otters feeding at most
-streams. The pattern of initial carcaKs consumption somet m--s
permitted identification of the dominant consumer: typically,
`,i bears left only the lower jaw, gills, gut, and caudal tin; raccoons
�...,.
{ ,,..:.and otters peeled skin inside out from the muscle mays, much
like removing a stocking. Using these characteristics and the:
occurrence of tracks, we estimate that the minimum number of
.'. .carcasses eaten by bears on Mink Lake Crcek was 116, or 911%r
of those known to be wholly or partially consumed. Similarly,
raccoons and otters combined fed on 59, and 377 of the car-
: ., casses consumed at Bear and Dickey Crees, respe.tiscly. A
sequence corn monly observed was for raccoons. otters, or hear~
to retrieve carcasses from the streams (bears removed fish from
pools more than 1.5 m deep), to feed, and to tlten move tin.
'rhe scattered remains were next ekrnsuntcd saver several weeks
by small mantrttals and birds, until ordy scattered bones and
pyloric caeca remained. The frequent occurrence of'snutll textth
site for 568 coho salmon cats' -
Carcasses i..
No. Pettw..
Pool —........_ _..__�
257 (43).�
Riftic
62 (11)
Gravel bar
147 (26)
rddy
102 (18)
Total 568 x
'Um insiream�sighting, 224 carcasses subsequently moved toM
streamhank by carnivores. See footnote Table 4 for explanation of`
cracass
T.,.nt.F 6 Ric of coho salmon carcasses.
Fish mass
Percent'
Stream
Initial(kg)
Eaten
Stream
Bank
Undeterrnined;
nk 1.akc �..._
434
79
8
3
10
West Twin
288
36
30
7
27
Bear
272
68
21
5
6:
Bockman '94
527
30
56
7
7
Ruckman '85
281
43
37
7
13
nlckcy
295
71
12
5
12 .
Minter
301
47
42
4
7
Harlow
284
36
21
s
43,
'Percent of initial mats known to be eaten or remaining imtreamar,
on the hanks; undetermined includes flushed and unknown carcasses.;
marks on bones plus skinned fin rays'and lower jaws, suggested.;
that shrews and mull rodents were more important in this.
"cleanup" than indicated by the trapping; in part because sinall-
mammals captured in the snap traps were often removed add`
eaten by larger carnivores.
Although bears were recorded on four streams, they were,
most active on Mink Lake Creek during October —November:
1984. Many bears had probably already entered winter dens
about the time carcasses were released on the other streams
(Lindzey and Meslow 1976). Several elusive species may We'
been 1110rc important consumers than indicated, e.g. weasels
and mink. The presence of deer tracks near carcasses and the
documentation elsewhere of fish -eating -deer (Olson 1932; Shea
1973, Case and McCullough 1987) indicated that even deer may
haws consumed parts of some carcasses,
Tltc eight avian consumer species ranged in size from winter
wrens to bald eagles. As with small mammals, the frequency
of pecked carcasses, the occurrence of droppings, and theaeea-
signal unidentified tracks suggested that the importance of bird
seascnbing wag underestimated by direct observation. Feeding
by dippers, gray jays, stcllcrs jays, crows, and ravens probably
occurred throughout the study area. The carcasses did not attract
lades aggregations of scavenging birds, as reported for large
rivers (sta!master and Gessaman 1984; Hanscn et al. 1984): In
general, mammals were more important consumers thanblr&,
particularly along the streams in depse old -growth forests. -
High Flow Study
Twenty-one of 25 carcasses (84%) released into West Twin
C'rcck at'the beg.imring of a flood remained within the 600-ra'
1352 Crrn, J. Fish. Aquar. ,Sal., Vol. 46, 19d9:
MAR713-02 12:53 AM EFH CONSULTING LTD
6045217143
P-08
I
TAOLG 7, Mammals and hinds present (P) and knuss_n to contiumc salnion carcases N.
----------- ___._ Stream' ,
Species
1 --
2
3 4_
5
6
7
8
Mammals
Water shrew
Sores palustris
X
X
Masked shrew
Sores cinereus
X
Wandering shrew
Sorex vagrans
P
P
P
Mote
Deer mouse
Scapanus spp.
Perwnyscus municuluurs
X
X
P
X
X
X
X
Douglas squirrel
Tamiasciuris dnuglasii
X
P
P
P
Flying squirrel
Glauco►nys sabrinus
X
P
P
Beaver
C4,vor canadensls
P
X?
X?
Mountain beaver
Aplodowin rufa
P
P
X
Coyote
Canis Lutran•s
P
X
1'
X
I �
Black bear
Ursus arnericcrnus
X
X
X X
X
X
X
X
X Raccoon
Procyon lotor
Weasel'
Afnsrela spp.
X
P
Mink`
)(Skunk"
M. Vison
Mephitis rnephiris
X
P
X
P
X
P
P
X
Cougar
Felis concolor
P
X
P
X
P
X
P
Bobcat
Otter
Lynx rufus
LIUra canudPnsi,c
X
P X
X
X
X
Elk
Corvirs claphus
P
P
P
P P
P
P
P
P
P
P
P
Blacktail deer
0docoileus hentiortus
P
B irds
Great blue heron
Arden herudius
P
P
P
Ruffed grouse
Borursa urrthellus
P
Merlin
Falco columbarius
P
X
P
X7
P
Red-tailed hawk
Buteo jam aicensis
P
P
X
P
X
Bald eagle
Haliaeetus leueocepltalus
P
Pygmyy owl
Gluuridium •gnoma
P
P
-
Kinbftshcr
Megaceryle ulcynn
P
1,
P
l'
P
Ilairy woodpecker
Picuides viler„us
P
P
P
P
Sapsucker
Sphyrapic'us ruriuc
P
X
P
P XP
P
P
P
Dipper
Cirtc•lus mericanicc
X
Gray jay
Perisurrus camulensis
X
X
P
X
P
P
P
P-
Stellers jay
Cyartnciva stelleri
P
X
Crow•
Carus spp.
l'
f
X
P
1'
Raven
C. c orrtz
P
P
P
Hormit thrush
Carhurus gtuturus
P
Varied thrush
/xoreu.c noes' as
P
P
P
P
P
Chickadecst
Parrs spp.
P
P
P
P
Nuthatches`
Winter wren
Siva spp.
Troglodytes troglodyres
P
P
P
i' P
P
P
X
P
Kinglets"
Regulus spp.
P
P
P P
P
P
P
P
Pine siskin
Carduelis pinus
P
P
Fox sparrow
Passerella Itiaru
,
P
P
P
Song sparrow
Afelospi:a rnelodin
'Mink Lake (I), W. Twin (2), Harlow (3), Hear (4). E. Fork Dicke) (5), Minter (6), BocKman tytsg
(7), and 1985 (8).
'Probably M.frenata.
`Possibly M. americ•ana on stream No. 1.
'Possibly Spilogale xracilis.
`Corvus brachyrinnchas and C. marinas.
Torus arricapilhrs and P. rufescens,
°Siva cartarlensis and unknown.
'Regulus sairapu and R. calendulu.
section and moved a median distance of 66 m (range
't9 In)- twice the distance recorded in the earlier release.
W ''lsc tareasses were deposited well athove tits low ilow lcv> ls
on crevei bars, stream banks, or in IoFjams; resslts :+imilar to
11L� y�we observed during high flows on Mink Lakc and Bock-
riOCreeks in 1984. These observations suggest that high pro-
pvrtis�ns ttl, carc•a,res nt:ty he retained in the spawning arc as
even following iloods, particularly where 1,01) IuaJ, are hich.
1t summary, our results showed that high proportions nr the
Cap. J. ri+h Xplar. 'sur. VuL 10, 1989
coho salmon carcasses were retained in the streams and sub-
stsontial numbers Were consumed by tin array of mammals and
hird�, 1-hc rli�t +aces that C;IrcaSscS drifted appeared to be related
dirci lly to the occurrence of freshets and inversely to debris
load and carnivore scavenging, although the number of streams
way t<x) small to define these relationships quantitatively.
Discussion
Thc relatis'cly short distance traveled by most cttl10 carcasses
itt our small streams (also suggested in a preliminary study by
1353
MAR-13-02 12:54 AM EFH CONSULTING LTD 6045217143
P - 0 9
on 1985) eontra.tcd with the drift of churl
, clarhulm and Peter,
Jalmon carcasses in Washington's large Skagit River (Glock et
�d. 1980), the only other study to our knowledge where carcass
r ement and retention were assessed. Although in the Skagit
udy about one-half of 103 fish (76 postspawners, 27 car-
scs) drifted 10.4 km or less, the other half drifted as far as
�km, during an unusually severe flood where hourly flows
a . ached 650 m'/s (Glock et at. 1980). Distances traveled aside,
th this study and ours showed that MUny eareas;es were
tained in the general spawning areas.
a .(,The capacity of most streams and rivers to retain carcasses
fta9 probably been greatly reduced by human activities, espe
,• lrrajly outside national parks. Oregon's Willamette River, for
? lsample, bus been changed from having multiple channels,
hes, and sloughs on an extensive forested floodplain to
�'.� ely a single confined channel on a contracted and greatly
t,.'.., aCorested floodplain (Scdcll and Froggatt 1984). Beginning in
#70, large numbers of fallen trees and snags were removed
om the river to improve navigation, (The Skagit River,
dis-cussed above, also had many log jams removed prior to 1900
(Sedell and Luchessa 1981)). Early logging activities altered
�;.. „�trtany small streams by using "splash dams" to transport lugs
d`dell and Luchessa 1981). Debris torrents following extcn-
ire log&g additionally scoured channels (Swanson and Lien-
6;;rr' per 1978). Streams were often further straightened and
�, + 1)Sed and hnd logs, (1ebris, and racks rernoved as adjacent land`
kwr,! ate developed (13ilby 1984; Grette 1985).
`;+^ iNWe do not know either the current importance of salmon
*asses to the array of carnivores on the study areas, or the
"'I, "bohsequences to carnivore populations on the Olympic Pcnin-
Ip from the decline in salmon (and carcass) abundance dtx-
v nted historically. Assessing the importance of carcasses to
ivores would requite long-term experimental manipulation
carcass numbers with concurrent monitoring of carnivore
1.. lsity — efforts beyond the scope of this study. rlscwh:,re in
tsshington and in southeast Alaska salmon carcasses are
,M1rlr wed as essential food for wintering and breeding bald eagles
,'Stalmaster and Gessaman 1984; Hansen 1987).
t 1p�'Peninsula, appreciable numbers
Ms,n would have been available as
mo annually (Houston 1983).
tl0vr, igbest ensitles d carcasses would have occurred
m about mid -October through January, supplying abundant
w „ for carnivores when alternatives may have been scarce.
Ccounts of substantial utilization of adult salmon historically
,isher 1890), coupled with the known reduced abundance of
sf'<< i on and the heavy consumption of carcasses documented in
>fis study, suggest circumstantially, a potential for p,-)pulation
P ? Meets upon certain carnivore species, resulting from reduction
I# their food supply.
retention demonstrate a potential
,contribute nutrients to spawning
" s is unclear. At first giancei the
nGw a eonstimplion y ieirestrial carnivores would seemingly
rgcludc meaningful additions to lotic foul webs. But these
a ; } evels of consumption might he misleading. Historically. coho
spawning levels may have exceeded 125-300 adults -km` of
earn, 10 times the densities now ohserved in many areas
A. Wood, Washington Department of Fisheries, unpubl.
data), Athigh carcass levels, canlivoic fond demands ma) hasc
been satttrclled. fivcn though higher numbers cif carcusscs could
jhave been eaten overall, the proportion eaten on a sincle strcam
1354
could have bcCn much lower, leaving many carcasses. i
decompose.
,Even at high carcass density e q
tively insignificant compared
forests of 3�,4000 kg•ha -' yr..; "
wan ct al. 1984). However, auto ' •,_ , M r y s.
�y
rence of carcasses, and careassi•
position of the litter, with conseq r4 u
detrital food chains, as reported a
pseuduharengrts) (Durbin et al. 1979). The role of carcasses
the dynamics of forest streams may be subtle, and certtiit
warrants study, particularly where spawning salmon still ota
at nigh densities.
N ensurative experiments that attempt to mimic natural V
ditions must always be viewed with caution. Considerations
Ingistics, cost, and sample size dictated that we use carcais
rather than capture, tag, and release live spawners. (In ret
specs this was a fortunate decision because coho escaperrie
were very low during the study,) How closely does the reterit
and utilization of introduced carcasses parallel that of m
salmon? Our experimental carcasses were released into the It
wogs of the streams as the most conservative placement. Obs
vattons of wild postspawners suggested that some dying`f
moved into shallows and backwaters where their cartes
might initially have been more accessible to carnivores `.d
those we released. Thus our measures of utilization could
conservative,
Acknoµledgments
w,
Funds were. provided by the U.S. National Park Service and Waah
ington Daparhnent of Natural Resources, Washington Departmont 6
Fisheries personnel at the Sol Duc and Hurnptulips hatcheries supplied`,
coho carcasses. Our able field crew of A: Brastad, M. Gracz, A,
Krvrnc, K. Nelson, J. Oelfke, P. Rentschler, E. Seaman, D. Sharp,'
R. Vanderhorst, and C. Wilson was characterized by an hypertrophied
sense of humor and an atrophied sense of olfaction. D. Gouin and6,:;;s
Schreiner aided data analysis. L, Conquest and K. Loman provided;
advice on statistical analyses. We thank K. W. Cummins, G. L, Lmq` .;.
can, J. `tauliews, 1, li. Meyer, E. 0. Salo, artd M. V. Stalmaster for, M1,
resic�tis of the paper. V. Rose, L, Perini, and J. Walker typed thei
manuscript.
References ,
BARR-7T, R. 11. 1933, Smokcd aluminum beef><plptsfor determining furbra}nt,
Yy
1'
diuribuuon sod relativo abundance. Calif. Flsh and Game 69: 188-190.
Bu.bv, R. B. 1934 Removal of woody debris nay affect strcam channel sta•
hil;ty. 1, Forcary 82: 609413.
BRtc9:ri.r., 1) C.. J J. GOumc. 1970. Chemicaleffects of salmon decompo
�;1
wion on aquatic ccosystcrns, p, 125-138, /n Proc. Symp. on Water Pol•
} k
tution Control in Cold Climates. R. S. Murphy led.] U.S. Govt. Prindn`
:Y
Orrice Washington, DC.
t,
IiRow,N, B. 19 2. Nlo ni3l
clouds — a search for the wild salmon.Simon
Yi pthe
7 e?
CAsi,• D. J , km) D. R. NICCtn.l.ouGH. 1987. White -tilled deer forage on alb
wj%cs. 3. Mammal, t gB 195-197.
CF.ni.,w H �r, C. I. A.%D N. P. PVT-RSON, 1985. The retention of coho lalmo0
debris In Can.
00, :
`
(Onrnrh„rchus klsr;r�h) carcasses by organic small stroarny.
3 1'1,11 Aquas Sci..12: 1222-1225,
}
Uo �tnsnv, J R 1967. Thc phosphorus budge( of Diamna Lake, Alaska, es
? ,r
reiatcd to the cyclic abundance of sockeye falmon. Ph.D. thesis. Univ.,
t�'a:h;n�ton, Seattic, tVA, 141 p.
r '
S. W. Nixusi, Ary C. A. Ovt,ATT.1979. Effects of the spwning
mieraiion or the alewife. Alnsu pseudoharengus, on freshwuter eeosys-
ttma.
tar;N, 1) V, ltn.l 1977. ilAor,c w1monmanagoncnt Curlkoptc. Westemaeo•
t,
gi 1 `ur ai Szrics. V,d 1?. tlniv, Victoria, Victoria, B.C. 320 P.
Cara. J. Fish. iquot, S'ri., Vul. 46. J989
Im
fm
N
MAR-13-02 12:56 AM EFH CONSULTING LTD 6045217143
7
Journoof mocitriloratlancifthe Olyinnic Mountains. Univ.
9t M Seattle, WA. 128 P.
974
� . Forescaucemlon in reLaknjo river terrace development
r 'rok National Park, WA. Ecology 0! 927-942.
L. C. Buss. 1969. Forest vegetation of the montane and
Vones, Olympic Mountains, WAI Ecol. Monogr. 39; 27 140 1.
AND C. T. DY11403- 1973. Natural vegetation of Oregon and
USDA. Forest Service. Teeh.Rcp PNW-fl. 417 p.
!�AND I TuRNER. 1974. Litter production by red alder in western
Forest Scl, 20: 325-330.
G. HA;rrmAm, ANDL. CONquiv. 198.0. Skagit River churn
Titan drift study. Seattle City Light Dept., Scatt1c, WA. 4S p:s
.41911. A photographic systen fof mui4le automatic exposures
conditions. J. Wild 1. Manage. 45: 273-27 6.
1985. The role of large organic debris in juvenile salmonid
'Sabilas in small streams. M.Sc, thesis. Univ. Washington, Seattle.
1987. Regulation or bald eagle reproductive ratc-S in southeast
tatogy: 68-. 1387-1392,
E. L. BoEuR. J. 1. HODOES, AND, D. R CUse, 1984. Bald
tKithe Chilkstwliey, Alaska! ecology, behavior, and management.
4 Audubon Society. 27 p.
B. 1993. Anadramous fish in Olympic National Park' a status
,V* Nall. M Service, Seattle, WA. 12 p.
=B AND R. J. CONTOlt. 1994..Anadromous fish in Olympic
Park-, status and management considerations, P. 97-111. to, J M.
And D. B. Houston (ed.] proc. Olympic Wild Fish Conference,
publ, peninsula College. Potj Angeles. WA. field
seudoreplicalion and the design of ecological r
H. 1984, P
C 157-21 t.
u. Ecol. Monogr. 5
M
t. 1975. Transport of nutrienu by salmon migrating from the
01 systems.
"A' 4107 lakes, p. 153-456. In The coupling of land and watcr sy -
�. ]�;, Hater (ed.] Springer-Verlag, NY. -
isy E. C. MEsww. 1976, Winter dormancy in black beam
in =ashlngton. 3. Wildl. Manage. 40: 40" 15.
yUgSdtoi, W; E., G. J. CAU0111Y, AND P. C. GRI00- 1978. A double -
al alienate of populutinn aiae from Incomplete counts. J.Wildl. Man-
1972, Biogenic enrichment ofsockeye salivion I.Acs and stock
Iv. Veit. Int, Ver. Limincl. IS, 1089-10)S.
Vol 46, ONQ
NOAA (NAIKINAL 04rVANIC. A,4V A'rMOSPOCKIC ADMINISTRA110m). 1978, Cli-
CiiFilLiography of the United States No. 60. National
mate orWashington.
Oc,:1Lnjc and Atmospheric Administration, WA.
OLSOs, S, F. 1932. Fish eating deer. I Mammal. 13: 80-81
Pr�JC (PACIPC Fic.NERIV-S MANAGFMFNT COUNCIL). 1982. Perspective on man-
salmon fisheries within the fishery
artinclit of ocean chinook and cohn
corl<arvafllnn zone off California, Oregon and WashingtOn. PFMC Pubt.
Ponland, DR. 29 p.
Piuu.yvs, C. L Aso W. R DONALDSON. 1972. Washington climate. Cooper-
State University, Pullman, WA. EM
ative Extension Service, Washington
3708. 89 p,
Rxr)WAN. M. A., C. A. HARRINOTON. AND J- M. KXAFT' 1984. Litterfall and
in Washington. plant Soil 79-.
41
'W
nutrient returns in red alder stands western
341�-351.
Ricin:Y, I. E.. M. A. NkxrNs, AND C. R. GOLDMAN, 1975. Effects of koRanee
stream. J. Fish. Res.
salmon decomposition on the ecology of a subalpine
Board Can. 32: 817-1420, of streAmside forests to
SEnu.L. J. R., AND J. L. FRocnATf. 1984, Importance
Willamette Rivet, Oregon, U.S:A., from
large rivers: the isolation of the
its floodplain by snagging and sueatriside forest removal. Vtrh. lot. Vcr.
Limnol, 22: Jg28_183A.
K. 1. LucifwsSA. 1981. Using the historical record as an aid
QI
Srj)1-;ij., J. R.. AND
to salmon habitat enhancement, p. 210-233. in N. B. Armantrout led-]
1'be acquisition and utilization of aquatic habitat inventory information.
Synip. Am. Fish. Soc.. Western Div. Portland, OR"':
White-tailed deer eating salmon. Mul-telit 54: 23.
StmA, D. S. 1973.
SOKAL. R. R., AND F. J. Rolu. 1981. Biometry. W. H. Fpcoman and Co. San
Frunclico, CA. 859 p.
STALMAMR. K V.. A.4D J. A. GESSAMAN, 1994. Ecological energetics and
for3ging hehaviOr of overwintering bald eagles. Ec6l. Monogr. 54: 407-
428.
SL(',At, S. R. ANr) D. C. BuRREIX, 1934. Transport of dissOlved organic car-
Wn. nutrients and trace metals rrivin aic Wilson *M Blossom Rivers to
Alaika, Can. 1. Fish. Aquat, Sci. 41: 18t1-190-
Smeaton Day. Southeast
SWANSON, F. J., AND C. W. LTF_,4rALMFER. 1978. physical consequences of
large organic debris in Pacific Northwest Streams. USDA Forest Serv.
Tech. Rep. P,1;%%1_60. )2 p. lc map or the Olympic Pen-
TATIMR, R. W,, AND W. M. CADY, 1978. Gcolol;
il)s%113. WA. disc, Invisligalionger U.S. Gco), Survey Publ.
M. R. Cmoss. 1986. Length bf breeding life of
Vkv nL.4 Brxnim, E. P_ A.,4D
coho sallnilm (oncorhynchus khmich). Can. L Zool, 64. 1482-1486.
MAR713-02 12:57 AM EFH CONSULTING LTD 6045217143 P.11
nr L`
Ecouline Fisheries Habit
Consulf ing L
.
4° 1` i' F
y.^ e . • �<� 1. ».�b.'J , r�
..
C v • ,y�
i ro
ry
Y i
. ':far ,; ' F:.�;;°.•• l���;,._
.4,
P
t. xalr
200 - 4170 Still Creek Drive, Burnaby, BC V5C 6C6
Tel: (604) 918-5097 Fax: (604) 299-4511 WA Tel: (206) 371-7608
E-mail: Info@ecocline.com Internet; www-ecocline.com
MAR713-02 12-58 AM EFH CONSULTING LTD 6045217143 P-12
nadromous Fish as Keystone Species
Vertebrate Communities
F. WILLSON AND KARLC. HALUPKA'
Il!q C try Sciences Laboratory, 2770 Sherwood Lane, Juneau, AK 99601, U.S.A.
Many taildlffe specfea feed on anal mmous ft~sbes of several life-bistory stages There is evidence for
K U47dl jje species tbat the apallability ojanadrvmous fish is critically important for aurvival or repro-
f lJort In some regiOru anadrdmous flsbes in fresh ueter appear to be keystone food resources for vertebrate
alo►s and rtauengrnt, jorgin$ an ecologically signOr-ant link beniven aquatic and terrestrial eeorysterrts
"ttal distributionof aaadrOmOur flcb in fresb water, including the occurrence of runs in �'r3'sy 11
+ x bar important eonsequ�nces for udldlife biology (social interacrlon4 distribution, activity patterns
pg ably sumlvorsbip) and coruervation of blodivenity.
artddromos como espeeles clavcs en ass eomunidzdes de vcrtebr2dos
eat Afuchas especies so alimentan de varios de Jos estadios de desarro110 de peces or drorrtos ixiste
R (¢t>Rcia que lradlca quePara algunas espectes silvestres la disponibtlidad de peces anddromos es de ruma
da Para su supery vencla o su reproducciorL En algunas regr'ones, Jos peces anAdromos en aguas
u< P rrwn ser reeYusos alir>lenticlos traces pars los predadores varlebrados y los carroileros. carulitu-
irn erl4b6n ecoldgicame"te si8niftcativo entre los sistemas acudticos y termstres. La dlstribucidn
i41 dt los peces anddrornow en aguas dulcet incluyendo to pre encla de-corrldds" en arroyos »icy,
of lien* conseo encias Importanies pars la biologfa de to wda siltvstre (interacciones soci4le4
ueldn, patrones de activida4 posibllidades de suprevivencla) y !a conservaci6n de la blodiversidad
Atic and terrestrial ecosystems are usually studied
rely, by different sets of researchers using diffcr-
icthods and, often, different approaches. In some
1, of the world however, It is cleat that ecological
' Ctloris between the m,o ecosystems are central to
i v ecology. One such region, is the north paclfc
North America, where anadromous fish rc.
to spawn, often in huge numbers, and where they
Cu>tanI addrcm Blology Department, Albertson Collegq 2212
Viand Bouln.arg Caldwel; 1D 83(05, U.SA
y 24pe,+5a4bm1t1cd Ha)'70. 1994; revised manu1cnp1aecep1ed5ep1em•
u► 9, 1994
fall prey to numerous species of terrestrial wildlife.
Anadromy is common throughout the northern cool.
temperature and subarctic region, and it a15o occurs —
less commonly —in south -temperature regions (Mc-
Dowall 1987), suggesting that the potential for
important interactions between anadromous fishes and
wildlife predators is widespread. Although any field bi-
ologist could readily note chat herons and mink cat fish,
thus making the aquatic -terrestrial link, the magnitude
Of the Interaction in some regions warrants special cx-
unlnation and calls attention to the pervasive occur-
rencc of important aquatic -terrestrial linkages in many
other areas. The loss or severe depletion of arlsdromous
fish stocks could have major effects on the population
biology of many species of wildlife consumers and, thus,
489
Gonscnauon BloloM%Pages 4B9-497
Vulumr 0 N., w 1.. Inc
MAR713-02 12:59 AM EFH CONSULTING LTD
A6fldrpmous FW is li*Ie spedef
,! n terrestrial animal cbrdmunitia, but these possibiii-
{ have not been ex:it*cd and indeed have seldom
�A addressed at all ,
7be licit bctw-een fish populations and their verie-
fi"`- Y Mite corsumers has reccved some attention, usually of
4 One-s[ded nature, Alva st all of the existing studies
64Y
1 m from fisheries blolol y, in which a central concern
..how many fish remain r humans to haracst Harvest
. , huttsatu and other co timers has so profoundly af-
•peted the perception of almon populations that adult
t�don size is customarily referred to as "escapc-
�tnt"--dose individuals that escape from their enc-
i grslCL Our approach here is to reverse the perspective
to focus on the importance of fish populations for
3 'pure 1s precedent for such a change in per-
''"ittclvc- Some years ago ;Gould ( 1981 ) addressed the
1 •:
' Z ;*singly silly question oSwhether a zebra is basically a
animal with black s�dpes, or perhaps a black an-
` iao>tl wills a a-Wtc overlay. The question is,, in fact, any-
ig but silly for sdentlsis interested in development
y itttd evolution, because po�ing the alternatives permlts a
tktOre perceptive analysis of some basic biological pro -
Initial effect df our shift in perspective on
t A h-wildlife interactions tt�y be to complicate the study
t Qf ecological systems by 'insisting on linkages among
l 'set as, but In the long nin the change of focus should
*pow us to comprehend ti"gs that could not be under-
,, "r 00od otherwise.
_We focus on direct inter�cdons between anadromous
�ttV" t' and their vertebrate cionsumers In and near fresh-
j+, ator, but anadromous ish also have Important —
less apparent —in ect interactions with both
is and wildlife through nutrient dynamics. A grow-
"` body of evidence indicates that chemical nutrients
vercd by spawned out carcasses can play a critical
43
a t r�c in sustaining the productivity of riparian and lacus-
x t ,, a ecosystems, perhaps' including the next genera.
�, I{xf� . +ll�►s of Juvenile salmon (sCc Richcy et al, 1975; Kline ct
1990, 1993; R. Bilby et s�l, unpublished manuscript).
c cuff Cnt use of the tam "anadromous" refers spc-
>r rally to populations !n w Wch the adults migrate from
*fir• ; sa to fresh water for $pawning (McDowell 1987).
derivation of the word itself, however, refers more
Idly to a habit of runnlrig upstream (to spawn), and
" 7 s dote in a broad sense can be applied to any popu.
IS on that moves frOr11 a 14ge body of water (ocean or
;� c) to I stream environmint for breeding, We use the
ower sense of the word here, but because the eco-
y4;' cal relationships between the migrating fish and
,1r wildlife consumers similar, whether the fish
the to a stream from the sea or a lake, w-c briefly
i iscuss the relationships for inland populations re-
y *Uicted to fresh water.
We have chosen to use �c term "keystone species"
iEor anadromous fish in these interactive systcros, despite
2
Coa+avx�cn Biology
Volume 9. no ], June 1"5
6045217143 P.13
R'illson d� Nalupla
some possible difficulties (Mills et al.1993)• We find the
term useful to emphasizc our point about the likely im-
portancc of anadromy for the many predators that use
anadromous fish. We submit that lntet2ctions between
anadromous fish and terrestrial wildlife arc important
components of regional biodiversity, and that they de-
serve a far greater consideration in land -management
schemes, fishery management practices, ecosystem-
management plans, and ecological studies of ecosystems
than they have received in the past.
Background
Research on predator -prey interactions In which anad-
romous fish are the prey has strongly emphasized the
effects of predation on the fish populations. For exam-
ple, a literature search of Wlldllfs send Fisheries Review,
1971-1993 (using keywords salmon, predator, preda-
tion, and scavenger), revealed 80 papers on Interactions
in fresh water. Only 10 of these papers contained some
mention of We consequences of the interactions for the
consumers. Further Inspection of the literature reveals
the general lack of studies on the relationship of anad-
romous fish to the communities of vertebrate consum-
ers. The utility of seabird data (especially rates of energy
acquisition) as indicators of the population size of occ-
anic fish has been suggested (Calms 1992)• The focus is,
usually still on the economically important fish, how-
ever, although the potential impact of overharvesting of
fish on seabird populations may be serious (references.
in Cairns 1992). One exception to thi4 trend is it poster'
produced by the U.S. Fish and Wildlife-Scrr.lee, entitled
"Alaska's Salmon Resource: Importance to the Ecosys•
tem", ;;,hich illustrates the variety of wildlife that use
salmon as a food resource. In addition, the implicatlons
of fisheries management for fish -feeding birds in inland
waters have been emphasized by Dombeck et al.
(1984).
Many existing studies emphasize fish -hunting wildlife
species as competitors of human harvesters and often
address the question of how to reduce the effect of such
predators on the numbers of anadromous fish. The an-
swers to the question range from adjusting the timing of
hatchery releases (Mace 1983; Wood 1985q Bayer
1986) to more draconian programs of wildlife slaughter
(Shuman 1950;Anderson 1986, cited in Mills 1989). We
present two examples to illustrate some of the drastic
measures used against wildlife consumers and the attl-
rudcs that have led to these measures.
Some programs of wildlife "control" have been con-
ducted despite their dubious efficacy and uncertain
ramifications. For example, Elson (1962) reported that
merganser (Mergus merganser) populations would
need to be reduced from the estimated natural average
of 5-10 birds per 15 miles (24 km) of stream to about
MAR-13-02 01.00 AM EFH CONSULTING LTD 6045217143 P.14
cif 11ai11pa� -
bird per 15 miles, in ci dcr to increase significantlya7 ; pt�oduttlon of Adantic �almon (Sa/rno solar) smolt
`
Cu ternanad8r destroyed Ca. Elson's o ed an avcrag c
p
adult mergamers per tar for six years in a 10-milt
` tch of river and achles d an estimated increase of
,000 smolts. Stocking ratn also increased during this
however, so the es t rttated gain in smolt abun
e cannot be attributed solely and clearly to the
.: 0N91 of mergansers. Fu thcrmorc, Elson's estimates
t based on relatively Ngh densities of hatchery-
juver llcs, and hat0cry stocks are often more
tiblP to predation tha4n wild stocks (Beamish et al.
2; sec also Patten 1977i Wood & Hand 1985; Bayer
). Where the mergans r population is more abun-
' when mergansers recruit to waters with increased
4 dL hits of juvenile sal onids (Elson 1962; Wood
$q 1985b), and a-hen,the regional supply of mer-
" is able to rccolonizjt an area of merganser con -
(Mace 1983), the toll In wildlife death would be
hightr. merganser c6trol programs In eastern
t des have continued in orc recent times, although
correlation between m rganscr control and returns
adult salmon appears to be uncertain (Hunter 1959;
er 1968; Andersonet 1985; Anderson 1986 in
yr ,
w 1989; Wood 1987b). urthcrmorc. the decline of
tic salmon abundance In recent decades has been
art"t directly by excessive exploitation by humans,
t by merganser predation so the killing of mergansers
*rease populations of salmon is scientifically dubi-
(Anderson et al. 1985)'
second example may beven more telling. Between
° °47 and 1953, a bounty, ranging from 50 cents to two
k a bird, was placed on Bald Eagles (Haliaeetus
cephalus) In Alaska �Imlcr & Kalmbach 1955;
bards & King 1966). Regards n`cre kept on the num
111., '1A of eagles killed during'stils time, except for 1941—
" A, 45. when the bounty was �n effect but not funded, and
aq a 5-1949, when It was 'briefly repealed. Bem,cen
,000 and 128,000 codes were recorded In the
"unty, books during the years for which data are avail-
�ti• ' IC, and an unknown nu er w`cre killed and not re-
u d or wounded and led later (Robards & King
About 80% of these birds eamc from southeast
(Robards & King 19�6). Casual observations in-
, tcd t.`tat this mayhem w is associated with a reduced
"lr{; ber of eagles in cons Alaska (Imlcr & Kalmbach
r S), buS no assessments f the effects of reduced ea-
r'= �„numbers on prey popul floras were made (M. Jacob
W. G. Meehan, ptcso communication), although
t)iajor 1nitlal motivation for the bounty was the reduc-
on of predation by eaglcst During this time, the nui-
1�cnce value of eagles as poss blc competitors was clearly
considered to be more imp rtant than their ecological
+slue, so much so that the ei;;'ect of "nuisance" reduction
aas, not even worth meas\itint',.
dtudromous llsh as Fhstnne Specks 491
Wildlife and Anadromous Fish on the Northwest
Coast of North America
The Prey Resource In Fresh Water
Berween southern California and the Arctic Occan there
are ovcr 15 native species of dramatically anadromous
fishes (see Scott & Crossman 1973). These Include
three species of lamprey (Petrom)zontidat), several
species of smelt (Osmcridac), at least one species of
charr (Soluellnus), and seven species of Oncorhyncbus
(Salmonldae). The diversity of anadromous species is
especially high between northern California and south-
wcsicrn Alaska. Some species, especially of Oncorhyn-
cbu.; have sea -run populations that spawn in inland wa-
tors located hundreds of kilometers from the sea.
Spawners of several species may use the same strauns,
at either similar or different times. Tbesefore, any given
stream at any one time of year may harbor adults of one
or several species. The total number of strums support-
ing stocks of anadromous Ash In this region 1s enormous.
For example, in southeast Alaska alone, coho (0.
kisutch), chum (0, keta), and pint: (0. gorbuscha)
salmon each occur In over 2000 stream systems, sock-
eye (O. nerka) utilize over 200, aqd Chinook (0,
tsebawytacha) about 100 (Alaska Department of Fish
and Game, Anadromous Waters Catalog, unpublished).
The time of spawning varies considerably, both
among species and among populations in dilerent loca-
tions (Scott & Crossman 1973, references in Groot &
Margolis 1991). The lampreys enter fresh water in the
late summer and fall but do not spawn until the follow-
ing spring, Some of the smelt spawn in spring, but others
spawn in WI and winter. The charr are typically fall (and
occasionally spring) spawners. In general, the five spe-
cies of Pacific salmon spawn in summer and fall, but
spawning of coho, chum, and sockeye salmon ma)' ex-
tend into winter In some places. Both steclhcad (O.
myklss) and cutthroats (O. clarkf) are generally spring
spawners. For cortspeclfic populadons, significant differ•
ences are found among locations In the time of spawn-
Ing and the time of adult entry Into fresh water, perhaps
especially In steelhcad and chinooks. But adult In -
migrations and spawning times and smolt out -migration
times for any particular species in any one river system
are usually less variable from year to year than among
locations.
Titus, the diversity and abundance. ;of anadromous.
adults in a stream display enormous Ydriation In both s
time and space. In addition, the size' of,the populations
(arid the body size of the 1nd1v1dua11?sb) varies in re-
sponse to many factors, including hum4,n harvest. High
varlability of prey densiry Is reflected in the opportunis-
tic foraging of many wildlife cotuumcrs--but "opportu-
nistic" is not a synonym for blologlcally unimportant- In
many freshwater drainages, the presence of anadromous
Gonxr`•aclon D�o1a0•
MAR-13-02 01:02 AM EFH CONSULTING LTD 6045217143
P_IS
low 1�
AnadMMOtts risk Al A
1
fish is sWiciently reliabl
.,'11fespecies seems to be
? r' ,Wildlife Predation on Am
;a coastal Streams
As sit c:ampic of the
Spedes
that the biology of some Vvild-
cared to their exploitation.
thous Fish In Alaskan
m ldpie linkages that occur in' a
i tttativcty restricted rcSic n, we first summarize the wild-
„ fife species of coastal sou cast Alaska that arc known or
a tapceted to feed on samom We focus on southeast
' ;'`Alaska because this is One of the few places in North
<rtter(ca that contain abundant stocks of anadromous
fishes and a relatively unl nodifiicd terrestrial vertebrate
:tuna.
The regional list Ofwildllfc consumers of anadromous
Ish is impressive (Table � ). Not surprisingly, all carniv-
orous mammals in the re$lon take advant2gc of anadro-
q1 a Mous adult fish, as do a number of large earrllvorous
iat �iblyds. In some cases, pr datlon occurs on live adults
` espccially by bears and eagles). The spawned -out, mor-
.': `t r:dbund kilts and the carcasses of dead adults (many sear
Jes of anadromous fishes are semelparous and die
x Shortly after sDawrllna) ere cravrr, ,a i K_ .,...,..�
VIU5on g )Wupka
Perhaps surprisingly, carcasses are also eaten by species
such as squirrels and deer, which art typically consid-
ered to be herbivorous (sec also ICederholm et al.
1989). Juveniles fall prey to otters and many species of
birds, as well as to predatory fishes. Xggs arc eaten by
many birds, and by Dolly Vardcn chart, juvenile salmo-
nlds, and sculpins (see Armstrong 1965; Moyle 1977; ,F.
H. Everest, personal comsnunicatlon). Many of the eggs
consumed by thc.-c animals are drifting eggs that have
been displaced from redds by subsequent spawning ac-
tiviry or other disturbances, although sculpins may take
eggs from the gravel as well (Moyle 1966; Reed 1967;
Armsuong 1970). Gulls sometimes tread over redds in
shallow water to stir up eggs (Moyle 1966) or prod the
belly and vent area of gravid funalc salmon to force
them to release eggs before spawning (our observation;
R_ H. Armstrong, G. Strcvcler, personal communication).
Two lacunas in Table 1 deserve some explonGon:
there is a dearth of mammalian egg -eaters and of piscine
scavengers. We know that mammals will cat salmon
eggs bccausc bears often forage preferentially on roc
(and brains) from captured females (Frame 1974; our
observation). Therefore, It is not likely that eggs are
Wit 1. a'lldllfe consumers Of salmon In or near f;rsh wafers of sotamm ,Uaska_'
Salmon Lifc-Aisro
z consumers Eggs
luvcnilcs
f ' 1►lammais river otter (Luira)
mink (Muslela)
J'
;3
Ik 1lirds
tit i
"y ash is
a
r
rCompilcd from unpubushod e
i lal8r80l,a (1991) rbal also ocndr
So utilize salmon carcasses In o
futxnite jrs0.
I
Mallard (Anal)
Canada Gposc (Branta)
goldeneyrrs (Bucephala spp.)
gulls (>4,Larus spp•)
American'Dipper (Cinclus)
American I Robin (Turdus)
Dolly Van en (Salvelinus)
sculpins (Co. -us spp.)
coho saLyn (0ncorbrtchuS)
suckers( tostomus)
grayling ( Jrymallus)
loons (Gavia)
mergansers (Merges spp.)
Great Elue Heron (Ardea)
scaup (Ayrt w spp.)
€uLls
Arctic Tern (Srerna)
Belted Kingfisher (Megacer}'le)
crow
Black -billed Mamie
Dolly Vardcn
sculpins
coho, chinook salmon
rainbow trout/steclhead
cutthroat trout
%;-Zcye pollock (Theragra)
Pacific herring (Clupea)
adulc
bears (Ursus spp•)
minis weasels (Muslela spp.)
wolvecfne (Gulo)
wolf, coyots (Cants spp.)
red fox (Vaipes)
sins (Pboc))
sea lions (lumetopiat)
deer mouse (Pmmyscus)
shrew (Sores)
red squirrel (Tarniascfurus)
flying squlrtrl (Glaucont}'s)
black•taUcd deer (Odocoileus)
Bald Eagle (Haltacetus)
Red-tailed Hawk (Buteo)
Northern Hurier (Circus)
guUs
Black -billed Magpie (Pica)
crow, ravesn (Cortmi spp.)
Steller's Jay (Cyanoclna)
Winter Wren (Troglodytes)
American Dipper
+ ation; persona! communication; and from species lured in Cederholm of at (]989) and Croot and
Ibis ►rglon Tijo list undoubtedly omits some species of consumers, for Instance many other birds aria likely
^r regions the Variery of consumers is stilt greater because same amphibians and rrpifles pnry on eggs or
Conscn-adon Blolop
t'Olumc 9• IJ0 3, Junc I9,iA
MAR713-02 01:03 AM EFH CONSULTING LTD
, �• tiaiUplY
A.#A3ultzblc food for m mals. Probably most of the
fitunmals capable of enic Ing the water are so big rcla-
t;e to the size of the a that cgg-foraging, especially
`lror single eggs, Is not pro table. The absence of piscine
f ; t3i•engets may be attribu cd principally to the limited
es diversi and fo
�j,►ed q• ( - g habits) of fishes in these
AWecz , It is also possibl that mammalian use of eggs
VW piscine use of carcasses arc under-recorded:
Av
r Although bears are wid y recognized as salmon prtd-
j ttl, the consequences of salmon predation for bear
dlogy are virtually unknown. The large size of coastal
;b%T1 beats (Ursus areto ), compared to the conspe-
,' c $dunes in the interior, is sometimes attributed to
�X. x great availabillty of nadromous fish (Nowak &
disc 1983). Salmon consumption probably contr►b-
z .. to the fat deposition Tequired for hibernation and
the reproductive succes
of female bears (references
Willson 1993) but the r lative importance of salmon
4; pared to other food resources has not been as-
> , d. Mink (Afustela vis�n) may adjust the phenol -
of reproductlon to math the seasonal availability of
t
+ tming salmon (M. ben avid, personal communica
dA). Salmon carcasses ma be critical to the ovcrwin-
�S ` survival of Bald Eagles rd, In some areas, to their
i roducthr success (Hartrscn 1987), In addition, the
essibiliry of carcasses l�robably also facilitates the
1-21 of fledgling eagles, bvhich arc Just learning how
1<fange for themselves at bout the time when salmon
`v aT es are readily avails lc. tact runs of salmon and
Z runs of eulachon (T�aleicbtb)fs pac jicus) draw
�h r a from all over the re 'on and contribute to the
�.
tenance of regional p%mstrong,
ufations (our observation;
Msen et al. 1984; R. H. unpublished rc-
). Consumption of high }energy eggs may be critical
the survival of juvenile,sa mordds reared in fresh wa-
Eric tttgn latitudes, where the growing season is very
-'' Ott (E H. Everest, person! communication).
Pea crillc salmon probably Influence the breeding bl.
of several Predators.
P The phcnology of migration
!`reproduction of mergansers may be coordinated
f)t fry emergence and ssolt out -migration (Wood
Marquiss & Duncan; 1993). For mammals, the
�ttlon period is encrgct{c ly costly, and the mobility
talcs with young in the en may be limited. During
time1 Y, juvenile salmon a provide an important
resource for river of ers (Dolloff 1993)• Out -
at il lons of juvenile salm n often attract large num.
: of birds, including immature individuals (Mace
Wood 1985q 1985b) and the nesting density of
i fgansers has been correi ed with the abundance of
�'CrWe salmon (wood 198t ).
40me marine mammals pua�sue their anadromous prey
t farup freshwater rivers: fo example, beluga whales
(Oelpbfnapterus leucus) ha a been found hundreds of
kilometers up the Yukon rive , follonving the salmon run
6045217143
An2&VM0u5 fish is Keystone SFrdes
493
(Juneau Empire, September 14, 1993), and salmon.
hunting seals and sea lions move more than 100 tan
upstream in large rivers in Oregon and Washington (F.
H. Everest, personal communication; Juneau Empire
April 26, 1994). As adult anadromous fish approach the
coast, and as the Juveniles leave freshwater (or at high
tides), their congregatiorts arc often subject to high lev.
els of predation by several species of marine birds and
mammals, including seals, sea lions, and small whales
(Fiscus 1980; juneau Empire, December 6, 1993), and
some saltwater fishes (for example, walleye Pollock
[Tberagra ebalcogrammus}— see Armstrong (1968)-
and Pacific herring (Clupeus pallasi)).
Discussion
The few Systems about which we have found published
lrtfotmation all suggest that wildlife species capitalize on
available concentratlons of anadromous fish and may
change their distribution and even breeding biology in
response to the abundance of these Ash- Most reports of
wildlife responses to anadromous fish have emphasized
trout and salmon as the prcy; perhaps because of their
considerable commercial and recreational interest. The
relationships between anadromous fish and wildlife on
the northwest coast of North America can also be found,
with regional variations, on the north Pacific coast of
Asia and on both sides of the north Adandc.
Wildlife in inland areas also make extensive use of
migratory fish resources, Both black and gtizzly bears in
Yellowstone National Park prey on spawning cutthroats;
the level of bear activity was often correlated with the
density of fish in the stream (Reinhart & Mattson 1989).
Almost all the studied streams with spawning runs were
visited by bears, and there was evidence that bears often
tended to avoid each other by using different streams
(Reinhart & Mattson 1989).
Eagles utilize the spring runs of,suekers (Catostomi-
dac) in midwcstern streams (M. F. Willson observation).
Inland eagles whose nests arc close to spawning streams
have higher nesting success than those whose nests are
more distant (Gerrard et al. 1975). Eagles in the non -
breeding season often congregate near good fishing
sites, and the number of eagles is often correlated with
the availability of fish (see Fitzner & Hanson 1979; Spen-
cer et al. 1991, Hunt ct al. 1992; McClelland et 21. 1994).
In sc%,cral cases, the pre}, fish were not native to the wa-
tors in the region, but the eagles readily Incorporated
the new food source into their foraging patterns. Exotic
rainbow trout populations In the highly modified
aquatic ecosystem of the Grand Canyon probably ben-
cfit the eagle population but may further damage the
community of native fishes (Brown 1993)• In a pro.
foundly modified systern in W,cstcrn Montana, the num-
MAR-r13-02 01:04 AM EFH CONSULTING LTD 6045217143
4 Aakdromous fish ail'9} one spedes
„1 �
7
of Introduced kolance"on declined sharply after
$6. The numbers of cs and other wildIJe species
t gathered to harvest okancc detllned in parallel
_( those of their prey( enter ct al. 1991). Possible
�emative food sources fo eagles, such as carcasses of
icon or elk, arc no longe available.
;.$omaitnes the activiEl of one major consumer ird-
e a chain of Interactio for other wildlife species.
the Kamchatka peninsu a, for example, sea -run sock-
c salmon In Kuril Lake d its tributaries are extraor-
y numerous and spa n over a nine -month period
waters that common! remain ice -free (Ladigin
994)• This sockeye population supports a diverse ag-
tion of wildlife specs similar to that in southeast
Ica, and resident eagles ,usuaily gain weight over the
ter. Perhaps the most 'conspicuous of the wildlife
des is Steller's Sea-eagl� (H. pelagicus), a large rcl-
vc of the North American Bald Eagle. The massive bill
Steilar's Sea -eagles enab es them to open cfhiciently
e tough skin of salmon .carcasses that smaller birds
of pierce. The presen a of two smaller species of
c at Kuril lake In wint r has been attributed to the
w bility of salmon carcasses opened by Steller's Sea-
"` es (Ladigin 1994). v�hen exceptionally severe
ther covers the spawn�g grounds with ice, the in.
ecific congregation o eagles disbands; the con-c-
t tnees of being forced t� forego this rich winter rc-
urce arc unknown.
1e biological relationships between anadromous fish
a diverse array of wildlife consumers seems to stand
t from many terrestrial predator -prey interactions,
resources for almost kinds of animals are earl-
1c in space and time, but,thc anadromous fish system
An extreme case in whi h prey is temporarily very
undant, spatially constrained, relatively easy to cap -
and more or less predictable. It is different from
for -prey interactions 'with Irruptive species (lo-
lemmings) because o its interannual predictabil-
. It is similar in some r peas to tropical frugivore-
t Interactions in which me fruits serve as keystone
resources at certain 4cs (Tcrborgh 1986), but it
oot cleat whcn or where anadromous fish fill a season -
gap in other resources. S� me parallels exist with ant
i` ms and ant -following Irds In tropical forests, but
e ants themselves are not he prey and the swarm may
somewhat less predictable than the fish, Better com-
ns can perhaps be foe id with migratory ungulates
c prey species. Long-r ge mo�-ements of the Amer -
bison, were surely use by carnivorous mammals,
wLures, cowbirds, and humans, and the migrations of
V ' cic caribou are still utilf2ed by an array of carnivores.
tit the diversity of vertebr to predators and scavengers
�unported by bison and cars ou may be less than that for
:Inadromous fish. Perhaps ;the closest parallel comes
from the hordes of mlgradory antelope in Africa, on
ck•hich many species of prjt,dators and scavengers dc-
Conscn•oGon Dlolos
Volume 9. No. S, )unc 1995
tt'Yltsoo A aupkm
pond. None of these sets of interactf ons, however, is
based on a link between ecological communities that
are usually treated separately.
Not all el%cts of anadromous fish on their predators
are necessarily beneficial to Individual consumers. For
instance, when bears congregate along salmon streams,
intcnsc social strife sometimes leads to competitive dis.
placement of subordinate individuals or to the death of
cubs (S. Morello, K 'Titus, personal communlcatlon).
Dcnse aggregations in salmon -foraging season might in.
crease the transmission of pathogens within or even
among species. Furthermore, a disease kncmm as salmon
poisoning afflicts some species of carnivore in certain
parts of the west coast of North America (Schwabe ct al.
1977). The disease is caused by a rickettsia, which is
transmitted by a trematode vector whose We cycle
passes through one species of freshwater snail as the
first lntermcdlate host, then through certain fishes (in-
cluding salmonlds) as second intermediate hosts, and
finally into fish -citing mammals and birds. Of the final
hosts, only caNds and raccoons become ill, but the In-
fection can be fatal to canlds.
We have been concerned, so far, chiefly with the cf-
fects of anadromous fish on their wildlife consumers.
But predators can have effects on their prey in addition
to potentially reducing their abundance. Predation, in
general, is seldom random, and predation on anadro-
mous fish is no exception. Trade-offs between foraging
benefits and prcdation risks are thought to contribute to
patterns of habitat use by Arctic eharr and other species
(sec Huntingford et al. 1988; Magriha$cn 1988; L'Abbe-
Lund ct al. 1993) and arc probably relevant to many
anadromous fishes. Size -selective predation on Juvenile
salmonids is known to occur, but the direction of selec-
tion appears to vary with circumstances (Parker 1971;
Wood 8: Hand 1985; Wood 1987a). The juveniles of
some species may be more susceptible than others to
predation by certain wildlife species In some experi-
mental situations (Patten 1975, Hoar 1976; Hargreaves
LtBrasscur 1985); whether this is true when each
prey species occupies its own natural Habitat is appar-
cntly not known. Black bears favored chum over pink
salmon, perhaps because of their greater size (Frame
1974). Furthermore, predation can be sex -specific, but
the sex most at risk varies. Glaucous -winged Gulls
(Lanus giaucescens) have been reported to prey or
scavenge differentlally on female salmon (Mo55m2n
1958; Moylc 1966). Bcars and otters prey selectively on
male salmon in some cases (Gard 1971; Carss et al.
1990; Burgncr 1991) but on females in others (Frame
1974). Reports of selective predation Are too few and
scattered to allow us to examine the causes and consc-
qucnces of the variation. Nevertheless, it Is clear Mat
potentially important effects of selective predation may
be exerted on the life history of anadromous fishes.
The distribution of anadromous'fish on the landscape
P_ 17
MAR-13-02 01:05 AM EFH CONSULTING
d Hdvpki I I
F•
1 scvcral important impU ations for wildlife conser-
dlion. Temporary local d clines of anadromous fish
pulations must be relatively common in nature, be-
lit3�
many small-scale disturbances and stochastic
is can make a stream Im ccessible or inhospitable to
14romoµs fish. Some predators, such as bears, may be
able of consuming the a tirt run of salmon In very
as , streams (Shuman 1950), leading to depressed fu-
` returns. If a stream loss its anadromous fish pop
p �A tiflon, the spatial distribudpn of wildlife consumers or
s'. #r nutritional status and,;ultimately, their reproduc-
q" vt success are a-cly to a altered, The severiry of
l 0c effects depends in lame part on how long the fish
ulation is depressed and) on the availability of altcr-
dve resources.
1,1"Cornmerclal fisheries have great potential for persis-
y altering the spatial < istrlbution of anadromous
r•,. resources, in particularby eliminating small stocks.
Asti _ r
cries for anadromous saJ mon are generally managed
?' S the scale of regulatory districts encompassing multi-
iaatersheds, and might ry salmon arc ohcn Inter•
" ed in multiple districts efore reaching their spawn -
Fishing rea ions for each district are
i ically based on the esUrn tcd ability of the largest or
s t productive local stoc to sustain harvest, with an
�ltipnal allowance for int rccpted fish. This manage-
la�; .
saatcgy is economi y expedient, but it tacitly
is that smaller or less productive stocks, or those
,.' it� migratory paths that "pose them to a gauntlet of
•cst may be overexploitor catirpated. 1f the pop -
ohs of anadromous fish in small screams are se-
^.; ly reduced, this manag ment policy is potentially
gotmental to wildlife for evcral reasons. For many
estrial wildlife species, fish are typically easier to
itch In small, shallow strs (or at rifles and rapids in
; tr ones) than In lakes o large rivers (see Shuman
t�rr <' 5d. Reinhart & Mattson 1989). Moreover, the pres-
t of anadromous fish in umerous tributary streams
is intraspcciGc spacin of at least some predators,
\' d as begs. Female bears ith cubs, and young bears
tti►Iy independent of their mothers, all avoid mature
es, and family groups rr>lay also tend to avoid one
,;,1, plher (Reinhart & Maas n 1989). Thus, "escape-
"ts" sufficient to main a fishery in a management
tict may be inadequate om a wildlife perspective.
fish populations of sm streams are not entirely
?; ' " sticutable resources for any species of wildlife, a] -
ugh thcy may he for corr mercial Gshcrs.
;9p large spatial scales, wl en regional spawning ruts
elapse (as from chronic verfishin widespread
g or P
%3bitat modification), the o niorts for wildlife consum-
t;Xs art more limited. Lon -distance emigrations, im
Paired reproductive suecc and increased mortality
become more probable. Sp tW scale also affects the
abilin• of wildlife consumers 4o recover from declines or
Cxtispations of their prey. Fc r instance, the current pat -
LTD 6045217143 P-18
Anadromous fist+ as Ke)stoae Species 495
tern of salmon stock declines in southeast Alaska is hap-
hazard, is limited to less than 10% of the stocks, and
occurs primarily in small drainages (K C. Halupka). This
pauem resembles historical patterns of decline in the
Pacific Northwest and British Columbia (Frissell 1993).
if salmon runs are restored In these scattered systems,
the probability of rapid recovery of wildlife populations
or recolonization Is high because source populations
remain in nearby areas. In contrast, current salmon de.
clines in the Pacific Northwest are more regional in
scale (Frissell 1993), possibly compromising the viabil-
ity of wildlife populations over broad areas. Further•
more, recovery times are likely to be protracted when
immigration distances are large and source populations
arc small.
Scale is also important in recognizing the relevance of
Indirect disrurbanee events in a system characterized by
many interconnections. For instance, destruction of
headwater spawning grounds by landslides, earth-
qual cs, road -building, logging, minin& or agriculture
has consequences for the foraging ecology not only of
the a-lldllfe species that utilize these small streams but
also of those that concentrate their activities far down.
stream. Furthermore, the subpopulatlons in lower -order
streams may contribute to the genetic diversity of fish
stocks, and extirpation of the subpopulatlons might re-
duce the long-term viability of the stock even when the
numbers of fish remain temporarily unaffected (North -
cote 1992). Likewise, downstream events obviously can
have major effects on upstream ecology: Earthquakes
and dams alter strearriflow necessary for both upstream
and downstream migration, water quality (especially
tempi rature) that affects egg and juvenile survival, and
the hydraulics of spawning areas (Roys 1971; Thorsteln•
son et al. 1971). Megaharvests (such as the capture of
more than 80% of a sockeye' run in one day-, Rogers
1987) seriously deplete the number of spawners and
can change their spatial distribution. Changing climates
and oceanic cycles will have yet -to -be -determined ef-
fects on the availability of anadromous fish for both hu-
man and wildlife consumers.
Both harvesting practices and the prevalence of
hatchery stocks in some areas may also affect the tem-
poral availabiliry of anadromous fish in fresh water by
selecting for particular timing and duration of spawning
runs. It is possible that temporal changes in prey avail-
ability have important effects on wildlife biolop,, but
this possibility seems to be as neglected as the effect of
spatial changes.
Many people will no doubt say that the importance of
anadronious fish for wildlife is common knowledge,
which is true at some level, Tourists and pl.otographers
flock to particular locations in coastal Alaska to watch
bears capture salmon. Fish growers and harvesters are
obviously aware of possible competition from wlldli.fe.
Sport fishers for ;nadromous fish in fresh water know
conscn.ation D,ulog).
MAR713-02 01:06 AM EFH CONSULTING LTD
�96 .ta:dtt'xnoua' fitli u � tone Species
• roust behave circum ecdy In bear country, where
L1e peedators vie a•!th umans for prey. vPe submit,
�cer, that a change o perspcccivc-to actively in-
Ogde the wildlife panic pants in the interaction -is
$ overdue. Variation In anadromous fish populations
#n have major ceffccts one the productivity, phcnology,
Od mctapopulation d)-4ics of wildlife and hence on
isdodal biodiverslry, Whit is needed now, in terms of
iiCili is some quantifiatlon of the interaction. in-
j iAcdonS among species I, are a central component of
,, M*stem function and, once, of the maintenance of
vcisiry In ecological •smms (Willson 1995). No -
Were Is'this more evident than in the fish -wildlife in-
xior4 we have discussed here. Recognition of the
"one nature of anadromous fish populations should
K Incorporated into ec •stem -based plans for land
taaaagement, fishery hary st, and conscn•atlon,
j
i
/*" owledgments
�g are grateful for the hcl provided by our librarians,
,Bedvsnd L Pctershoar , for provocative discussions
04 numerous references' from lZ H. Armstrong that
d Initiate this essay, d for moral support from F.
Everest E. 5hoehat's p allcl concerns provided use-
�(input. We thank: K H. strong, F. H. Everest, and B.
gtston for conswctivc c mments on the manuscript.
erasure Cited
W4010n J. M. 1986. Merganser rcdation and iu impart on Atlantic
pi'wmon slow In the Resdgou a River system 1982-1985. Special
n:4�1ihUCstipn, Series no, 13, Atlantic Salmon Federation, St, Andrews,
�""Jgcw Brunswick Canada.
Orion, J. X. 1t Schicfer, and r i. Brucau Cztricr. 1985. A study of
YMerganser predation and its ' tpact on Atlantic salmon stocks In
Rn'; tie Restitouche River system 984. Atlantic Salmon Federation, SL
x: ndre�ss Nca' BrunwIck Cx ads
g R H. 1965, Some feedkts habits of the anadromous Dolly
vardenSalivitnos otahrto (W baum) in southeastern Alaska. 1n-
� YbMatto rial lestAct 51. Alaska pcpar=cnt of FLsh and Game, Ju-
h ', mau, Alaska.
lstronZ A H. 1968. An IneideIpce of walleye pollock feeding on
ipimon yAung, Transactions of the American Fisheries Society 97:
�,�'1112-203. I
ong, R. H. 1970. Age, foo t, and migration of Dotty Vardcn
is In, Southeastern Alaska,l Journal of the Fisheries Research
�d of Canada 27-991-1004
R. D. 1986. Seabirds vicar ar� Oregon esruiiinc salmon hatchery
1982 apd during the 1963 EI,Nino. Fishery Bulletin 94:279-266.
"'tilt li J, B. L Thomson, and . A. 1,1cFLrIzne 1992. Spiny dogfish
rtdatlory` on ehinook and kph salmon and the potential effects of
tehety produced salmon. T actions of the American Fisheries
"Society 21:444-455.
own, B. T. 1993. Winter fon&Ln ccoiog)' of Bald eagles in Arizona
Condor 95a132-138.
4tigner, R L 1991. Life history f 5ockcye salmon (Oncorhy nrhus
nerka), Pages 1-117 to C. Gr i and L Margolis, editors, Pacific
salmon life histories. University of British Columbia Press, Varncou•
ver. British Columbia, Canada.
I
6045217143 P_19
ts9flson & NafupA2
C lnu, D. K 1992. Bridging the gap between dr'tt},thology and f sheties
science: Use of seabird data in stock assessioent models. Condor
9d,8 U-S24.
Car'ss, D. N„ H. ivuuk and J. W. K Conroy. 1990, Predation on adult
Atlantic salmon, Saimo salor L, b)• othera,lrjtra lu&v (L), within
the River Dec system, Aberdeenshire, Scotland. Journal of Fiab
Biology 37 935-944,
Cedcrholm, C.J., D. B, Houston, D. B• Cole, and VP, J. Scanlon 1989.
Fate of coho salmon (Oncorb)richur k1sulcb) eu'caasa in spawn-
ing streams. Canadian Jourrtal of Fisheries and Aquatic Sclenccr
46:1347-1355.
Dollolr C.A 1993. Predation by river otter (lutra aanadertur) on
Juvenile coho salmon (Oncorhycbtu kkutcb) and Doily Vwden
(Saltvilnus malma) in souUtcast Alaska_ C niLdian Journal of Fish-
eries and AquatlC Sciences 50:312-315.
Dombcck M., J. Hammill, and Vi•. Buller. 1984. Fisheries management
and flsh dependent buds, Fisheries M-4.
Etson, P. F. 1962. Predator -prey relationships bean m fish -eating birds
and Atlantic salmon. Bulletin of the Ksherics Research board of
C uuda 133:1-82-
Fiscus, C 1980. hiarinc manmaJ•salmoold interactions: a review.
Pages 121-132 in W, J, McNeil and' D, C. Htruswot$ editors,
Salmonid ecosystems of the North Pacific, Oregon State Universiry
Press, Corvallis.
Fituter, A E., and W, C. Hanson. 1979. A congregation of wintering
BaJd Eagles. Condor 81:311-313.
Frame, G. W. 1974, Black bear predation on salmon .st Olscn Credo,
AlukL Zeltschrift fUr Tier'psycholooc 35:2}-A
PrlsseU, CA. 1993. Topology of extinction sold endangerment of na,
tive fishes In the Pacific Northdcst and Catifornla (USA). Con.
servation Biology 7:342-354.
Gard, R. 1971' Brown bear predation on sockeye salmon at 1:arluk
Lake, Alaska. Journal of Wildlife Manalcmcht 35093-204.
Gerrard, J. M., P. Gerard, NN'. J. Maher, and D. W. A. Whitfield. 1975,
Factors tnfluencing nest site selection of Bald Eagles In northern
Saskatchc•tvan and Manitoba. Blue Jay 33:1694176,
Gould, S, J. 1981, What color is a zebras Natural History 90(e):16-23.
Groot, C., and L Margolis, editors. 1991. Pacific salmon life histories.
Uniscrsiry of British Columbia Press, Vancouver, Canada.
Hansen, A.]. 1967. Regulation of Bald Eagle reproductive rates in
southeast Aluka Ecology 68:1387-1392.
Hansen, A. J., L L Docker, J,1. }lodges, and D. It. Clint, 1984, Bald
cagJes of the Chilkat valley, AIwka: ecology, behavior, and man-
agemenL Final report of the ChUJtat River Cooperative Ud Eagle
Study, National Audubon SWcty and U.S. Fish and Wilcilde Service.
National Audubon Society, New York.
Hargreaves, N. B., and R.J. LcBrasseur. 1985. Species selective prcda-
Lion on juvenile pink (Oncorb)rtebus gor6usrba) and chum
salmon (0, keta) by coho salmon (O, kisutcb), CarssdiaiJournal of
Fisheries and Aquatic Sciences 42,<i59-G68
Hoar, W. S. 1976. Smolt transformation: evolution. behavior, and phys-
iology, Journal of the Fisheries Research Board of Canada 33:
1234-1252.
Hunt, W. G., B. S. Johnson, and R E. Jackman 107- Cartytng capacity
for bald eagles-intering along a northwestern river. Journal of
Raptor Research 26:4"0.
Huntcr, ). G. 1959. Survl%-al and production of pink and chum salmon
In a coastal stream Journal of the Fisheries Resc2rch Board of
Canada 16:835-885
Huntingford, F..A., N. B. Metcalfe, and J. E. Thorpe. 1988, Choice of
feeding station in Atlantic salmon, Solrno talar, pan: t8etu of
predation risk; season and life history stntegy. Journal of Fish 81-
ology 33-917-924.
Imicr, R. H., and E. R. Kalmbach. 1955. The bald eagle and Its eco-
nomic status. Circulu 30. U.S, Fish and Wildlife Service. WUhing-
ton, D.C.
i
i
(oruer 'atlon Biology
1'elume 9, No, 3. Junc I995
MAF 13-02 01:08 AM
EFH CONSULTING LTD 6045217143 P.20
I
b & HaiuDl�
"ySy, •`. ' i', C.,,�r, J.J. Goering, O.A. athisen, P. H. Poe, and P. L Parker.
"• : � . Rocycling of elements red upstream by runs of Pa-
9"N and VICpidcnce in Sashln Creck sss outhet-
'u1 sari Alaska Canadian Journal Flthuies and Aquatic Science 47:
136-144, i
T. C.,Jr,J.J. Goering, O. A. �athlsen, P. H. Poe, P. L Parkcr, and
Scanlan 1993. Recycling 4 elements transported upstream by
1a is
igns d Psleilic salmon: IL 8 and S C evidence In the Kvichak
rNjjver watershed, Bristol Bay, uthwestern Alaska. Canualan Jour-
"gg''qp of Fisheries and Aquatic S�'Icnccs 50:2350`2365.
1 iLod, J. H., A. Lengcland, B, Jonsson, and O. Ugedal, 1993.
Al nisi
�Satlon b age , size In Arctic chary: a tnde.off bc-
� Y 6 and ,
kidtng possibility and Fisk o(prcdatiorL Journal of An mal
62,1 CO- 168.
w
A 1094. Fire, ice, and a es, Natural History 103(2):27-32.
P. M. 1983. Bird prcdatlo on juvcndc salmortids In the Big
cum estuary. Vancouver Island. Canadian Technical Rcpom
4 1 es brad Aquatic Sciences' 1176, Department of Fisheries and
ripctarts, Vancouver. B.C.
Eat, C 1988, Predation risMlc and foraging in juvcrWc pink (On-
r i n 'rlrbxu gorbascba) and chum salmon (0, kera). Canadian
r1 fA 5l of Fisheries and Aqua Sciences 45-592-596.
M. and K Duncan 1 3. Variation in the abundance of
JW-brcasted Mergansers M us serraror on a Scotlrah river in
w` ;r`�Nitit?n to season, year, rivers hydrography, salmon dcnslry and
y ting culling. Ibb 11533--41.�
ell". B- IL. L S. Young P.'1+. McClelland, J. G. Crenshaw, H. L
�y `fk Lll►l{ert D.S. Shea. 1994. Migration ecology of Bald Eagles from au.
� e concentrations in Glael�t National Part, Montana[ Vildiife
It. M. 1987. The oc"CnCe and distribution of diadromy
ngtlsites. American Pishc Society Symposium 111-13.
i 0. 1989, Ecology and manag mcnt of Atlantic salmon. Chapman
?1 atld liall, London.
L 5. M. F. 50ti1f, and D, F. ak 1993. The keystone-bpccles
'lkb Dncept In ecology and consent,anion. BloScicnce 43:219-224,
A. S. 19% Scicctrvt prdation of Glaucous-aw-Inged Gulls
�r r�
ItiA.
pOn malt red salmon. Eco 39:482-486.
t {C, P. 1966. Feeding behavior the Glaucous-w nged Gull on an
ansalmon stream. Wilson Bulletin 78A75-190.
r�lt t: P. B. 1977. In defense of seulpins. Fisheries 2:20-23.
+ m cote, T.L. 1997- Migration ar}d residency In stream salmonids-
s Mc t,*Vcal consideration; and evolutionary consequences.
jaw qlt relic Journal of Freshwater )[arch 67:5-17.
y, 9F, R. M.,and j. L Paradiso. 19 3. Walker's mammals of the world,
",rs VOL 2. 4th edition. Johns Hopkins University Press, Balumorc,
L R ,968. Marine mortality schedules or pink salmon of the
Coola Rivcr,'central British Culumbik Jourzul of the Fisheries
t u �J ? ►f xuch Board of Canada 25: 7-794,
err, R R 1971. Size selective predation among juvenile azimonld
ff In a aritish Columbia W r Joumil of the FLsherScs Research
air zr d of �.aruda 28:1503451 ,
a$ B. G. 1975. Comp2rativ5 s cnbllity of fr)' of PacUlc salmon
d stcdl cad trout to pnedar on by torrent seulpin In stream
'y,x jt uu'ia Fishery Bulletin 73:93'-934.
' i4t B. G. 1977. Body size and le4mcd avoldance a factors adccting
+ .? tedauon on coho salmon, On orh)cbw klsurcb, fry by torrent
I..�ulptn Corrur rhothe" FishernBulletin 75:457-459,
#: R. J. 1967. Observation of ,fishy associated with spawning
F k4ilmor. Transactions of the Am, ricun Flshcrics Socict7 96:62-67.
y, 11 C D. F., and D.J. Mattson. 989. Bear use of cutthroat trout
I
wwt:�:':'!►�'ivld+.WYV fy.,. C ,;,,_+.... • F:S' ti_'$tdf,q�i .
Ana&vmous W yes X4*bm SperJes 497
spa-viing sucsrns in Yellowstone National Pa& Intcnutional Con.
fcrence of Bear Research and Managea'tent St54y-35o.
Richey, J. E, M. A. Perkins, and C K Goldman 1975, Effeeu Of kokanee
salmon (Or3rbor6j-ncbus nerka) dceompcsltioo on the ecology of
a subaipirre stream. Journal of the Fisheries Research Board of
Canada 32:817-820.
Robards, F. C., and J. G. King. 1966. Nesting and productisity of Bald
Eagia, Southeast Aluk�-1966. US. Depznt ncnt of lntuior, Bu'
rcau of Spon F"hcria and Wildlife, Juneau, Alaska
Rogers, D. E. 1987. The regulation of age at maturity in Wood Rivet
socicyc salmon (Oncorfrynchus nerka} Pages 7"9 in H. D.
Smith, L Margolis, and C C. Wood, editors, Sockeye salmon (On-
corfrynchus nerka) population biology and future management.
5pcclal publication 96. Canadian Fisheries and Aquatic Sciences,
Reis, A S. 1971. Fdcct of tectonic deformatlons on pink salmon rums
in Prince William Sound. Pages 220-237 In The great Alaska earth-
quake of 1964. National Academy of Science'& Waishington, D.C.
Schwabe, C. W, H. P. Ricrnma. C. B. frantl, 1977. Epidemiology In
veterinary practice. Lea and Febiger. Pttiladelphla.
scot[, W. B., and E.J. Crossman- 1973. Freshwater fisllts of Carudi
Bulletin 184. Fisheries Research Board of C;ataad- OtuwtL
Shuman, R. F. 1950. Bcu depredations on w4cyt sp-MIAg popular
[ions In the Karluk Rlvcr system, 1947. Joutt d of Wildlife Man-
agement 14:1•-9.
Spencer, C, N., B. R. McClcUand, and). A. Stanford 1991. Shrimpstock-
Int. salmon collapse, and eagle displaeemen4 BloSeicnce 41:14-
21.
Tcrborgh J. 1986. Community aspects of hugJvory In tropical forests
Pages 371-384 in A. Estrada and T. H- Fleming, editors. Frugivores
and seed dlspersal. JuA Dordrecht, The Netherlands,
Thorsicinson, F. V., J.H. Helle. and D. G. Blrkhota 1971, Salmon sur-
Nl%a1 in intcriidal zones of Prince William Sound sucsurss In uplihcd
and subsided arras. Pages 194-219 In The great Alaslu earthquake
of 1964. National Academy of Selencea, Washington, D.C.
Willson, M, E 1993, Mammals as sccdd1spers4 mun-11As in North
America Oikos 67d%5 -176.
Willson, M. F. 1995. Biodivemiry and ccologleal ptCA t. . in R. S=ro,
editor. Biodivcrsity In managed landscs4xy. Oxford University
Press, New York
Wood, C C. 1985a Aggrcgattvc response of Common Mergansers
(Mergus meganser): predicting flock size and abundance on Van-
couver Island salmon streams. Canadian Jour%W of Fisheries and
Aquatic Sciences 42:1259-1271,
Wood, CC 1985b. Food•scarching bchavlour of the Common Mer.
priscr (Mergus merganser) 11: Choice of forsging location. Cana-
dian Journal of Zoology 63:1271-1279.
Wood, C C. 1986. Dispcmion of Common Mcrganttr (Afergur mer-
ganser) breeding pairs in reiauon to the avallsbiliry of juverwe
Pacific salmon in Vancouver Island streams, Canadian journal of
Zoology 64:756-765.
Wood, C. C. 1987a Predation of juvenile Pacific salmon by the Com-
mon Merprucr (Mergus merganser) on east= Vancouver island
I: Predation during the scawud migration. Canadian Journal of
Fisheries and Aquatic Sciences 44:941-N9.
Wood, C. C. 1957b, Predation of juvenile Pacific $Llnaon by the Gom•
men Mcrgart5cr (Afergw merganser) on castCM Vancouver Island
11: Predation of strcun•resident juvenile salmon by Merganser
broods Canadian Journal of Fisheries and Aquatic Sciences 441
950-9 59,
Wood, C, C. and C. M. Hand 1985. Food -searching behaviour of the
Common Mcrganccr (Magus merganser) 1: Functional responses
to prey end predator densiry. Carudlan Journal of Zoology 63:
1260-12 7C.
JUN-04-02 TUE 01:06 PM USACE SEATTLE DISTRICT FAX NO, 206 764 4470 P. 02/02
CENWS-PM-PL-ER 29 Mar 2002
MEMORANDUM FOR: File
SUBJECT: Wetland Delineation for Cedar River spawning channel site
1. Site Visit: Zach Corum, Rustin Director, Noel Gilbrough, and I visited the site on the
morning of March 28, 2002, accompanied by Gary Schimek from the City of Renton. I focused
my effort on delineating wetlands in the project area. I walked the site and took data points at
suspect depressional areas. I used methods and criteria defined the 1987 Corps of Engineers
Wetlands Delineation manual to determine if an area was wetland or upland. The eight attached
data sheets document observations of the vegetation, hydrology, and soils across the site.
2. Site Description: The site lies along the southern shore of the Cedar River and occupies a
low bench below a steep valley slope. The riverbank is abrupt and appears stable. One swale
runs immediately at the base of the valley slope. Another occupies a linear depression at the
base of a minor terrace escarpment midway between the valley slope and the river. The
proposed spawning channel would be occupy the second swale, which becomes more defined
toward the downstream end of the site. At its highest, the terrace rises about 10 feet above the
left side (looking downstream) of the swale.
3. Vegetation: The most prevalent community that occurs throughout the site is a
cottonwood/alder forest with an understory of snowberry, salmonberry, and sword fern. In
places, vine maple, blackberry, Indian plum, knotweed, bleeding heart, giant horsetail, and
Pacific waterleaf occur. New growth of buttercup and nettle was just becoming evident.
4. Hydrology: Except within Wetland A, water or soil saturation was not observed within 12
inches of the ground surface. The data points were purposely located in low areas of a narrow
wall -based swale where water would have been observed if wetland hydrology was present
(particularly considering the early spring date of the site visit).
5. Soils: Soils in Wetland A had prominent hydric features, including low chroma and gleying,
abundant mottles, and oxidized pores and root channels. All other soils had chromas ranging
from 2 to 3 with no evidence of redoximorphic features. Dominant soil texture in the upper
layers was sandy loam or fine sand. A layer of gravel (2 to 6 inch minus) was encountered in
most holes at depths ranging from 12 to 23 inches. Except in Wetland A, soils appeared to be
well -drained.
6. Wetland Description: Wetland A occupies a long, narrow low spot in the central swale. The
remainder of the site appears well drained, likely resulting from the prevalence of sands and
gravels close to the ground surface. The steep riverbank precludes any sort of wetland fringe
associated with the shoreline.
Evan Lewis, Environmental Resources Section