S. Gregory L. Ashkenas
D. Oetter P. Minear K. Wildman
Longitudinal Patterns - Channel
Human activities in the Willamette River and most of the larger rivers
of the world tend to simplify channels and minimize flooding. Dynamic
channel patterns and regular flooding are cycles of river ecosystems that
enhance biodiversity and aquatic productivity. Beginning in the 1860s, the
federal government invested in the clearing of large wood and other obstructions from the Willamette River and eliminating of side channels. Most of
these river alterations in the 1800s were intended to improve navigability of
the Willamette River, because riverboats were a primary means of transportation of people and goods, particularly in the wet winter months when valley
roads were flooded and impassable for extended periods. As landowners and
communities began to build closer to the river, floods became a major risk
and the natural processes of erosion and deposition were a hindrance to
property owners along the river. Revetments and bank armoring with riprap
were used by the federal government, local communities, and private citizens
to stop bank erosion and close off side channels. Eleven federal flood control
reservoirs were built in the basin from 1948 to 1964 (pp. 30-31), and these
reservoirs reduce peak flows of small to moderate floods by approximately
30-50%. All of these control measures over the last 150 years have simplified the channels and islands of the Willamette River, converting much of a
previously complex network of mainstem, side channels, alcoves, and islands
into a riverine thread with far less channel complexity (pp. 15-35).
The Data
As described in Historic Willamette River Channel pp. 18-25, maps of
the river from 1850, 1895, 1932, and 1995 provide a scientifically robust
basis for determining the extent and location of river channels in the
Willamette River and its floodplain. The extents of flooding were combined
for these major floods and depicted in a map of the total extent of known
floodplain inundation since EuroAmerican settlement. The floodplain axis
provides the most constant and quantifiable context for tracking changes in
the river channels, because the position of the channel changes but the
overall area inundated by past floods is relatively constant (pp. 28-29). We
mapped 1-km slices of the Willamette River floodplain at right angles to the
center axis of the floodplain (see floodplain slice map at top of facing page).
Within each 1-km slice, we measured the length and area of main channel,
side channel, alcoves, tributaries, and islands. The longitudinal display of the
length or area of all channel types combined within the floodplain creates a
linear illustration of channel complexity, a chart of the conditions of the river
and its floodplain.
Patterns
To determine the potential for restoration we have compared the
longitudinal patterns of channel complexity along the Willamette River in
1850 and 1995 (Figs. 174-78). Prior to settlement and river modification, the
Willamette River exhibited three major geomorphic reach types. The upper
(southern) river from Eugene to Albany contained the most complex river
channels, with more than 100 – 400 ha of river channel within 1 km of
floodplain. Floodplains within this reach of the river contained as much as 11
km of channel length within a 1-km floodplain distance, and most of the
reach included 4-8 km of channel per km of floodplain (Fig. 177). The reach
from Albany to Newberg flows through several volcanic mountain ranges
that extend across the valley floor. The river “bounces” between these
resistant lateral landforms, creating a floodplain that varies greatly in width.
As a result, the section from Albany to Salem was relatively simple in 1850,
and the section downstream of Salem near Mission Bottoms, one of the early
settlements in the Willamette Valley, is more extensive and complex. From
Newberg to Portland, the Willamette River flows within a narrow basaltic
trench that developed early in the geologic formation of the basin. The
floodplain in this reach is narrow and relatively unchanging.
By 1995, human attempts to control the river had eliminated
large amounts of river habitat and simplified the river. Area of river channel
in the upper reach was reduced by more than half compared to 1850. The
length of river decreased even more through the elimination of side channels
and the armoring of its banks. By 1995, the upper river remained the most
complex overall, but the lengths of channels within 1 km of floodplain have
been reduced to 20-30% of the length present before river modification (Fig.
174). The middle section of the river also changed, but not to the extent
observed in the upper river. Area and length of river channel in the
geomorphically simpler lower river remain similar to the historical patterns
of channel complexity (Figs. 24-27, p. 25).
Potential for Restoration
The potential response of the river to efforts to restore channel complexity and area of riverine habitat differ along the Willamette River in
relation to its geologic and hydrologic characteristics. Historical patterns of
channel complexity (Figs. 175, 177) provide a context for evaluating potential gains in river complexity through restoration actions. The difference
between the graphs of channel complexity in 1850 and 1995 provides a
longitudinal measure of the relative loss or gain of channel complexity (Fig.
174). This graph depicts the areas of loss and gain and illustrates the analyses
of the different river reaches described above. Clearly, the upper river section
has the greatest potential for future recovery of channel complexity lost
through past river modification. The area downstream of Salem also has been
simplified and ecological function could respond with increased aquatic area
and channel complexity. Some of this land is managed by the Oregon Parks
and Recreation Department, and park managers are actively restoring some
of the side channels and floodplain features that have been modified over the
last 150 years.
Slice Number
0
10
20
30
40
50
60
70
80
90
100 110 120 130 140 150 160 170 180 190 200 210 220
2
0
Change in Channel Length (km)
Channel Complexity
-2
-4
-6
-8
-10
Floodplain Distance
Figure 174. Net increase or decrease in channel length per 1 km of floodplain distance, between 1850 and 1995. Channel length provides a measure
of river habitat complexity.
The maps and longitudinal illustrations of channel complexity can also
be used to project future restoration potential through alternative futures. We
examined aerial photographs for remnant channels and river features and
identified historical channels that could be reconnected to the river in the
future. In the mainstem Willamette River from Eugene to Portland, approximately 200 km of river channel could be reconnected. These reconnected channels are incorporated into the Conservation 2050 alternative future (pp. 90-91, Fig. 173 p.133). If we continue to simplify the
channel at the same rate of loss that has occurred since 1932, we will lose
additional riverine habitat, an alternative depicted in Development 2050 (pp.
88-89, Fig 173 p.133).
Figures 175-178. (Facing page). Longitudinal patterns of channel complexity along the
mainstem of the Willamette River.
134
PNW Ecosystem Research Consortium
RIVER RESTORATION
River Channel
Scale: 1:1000000
Portland Metro
0 mi
1995 Urban Growth Boundaries (UGBs)
10 mi
20 mi
Slices which intersect UGBs
0 km
10 km
20 km
S
N
Salem/Keizer
30 km
Eugene/
Springfield
Albany
Corvallis
N
S
400
hectares
300
0.0 - 19.9
20.0 - 24.9
25.0 - 33.9
34.0 - 43.5
43.6 - 68.9
69.0 - 111.9
112.0 - 178.9
179.0 - 430.2
200
100
225
220
215
210
205
200
195
190
185
180
175
170
165
160
155
150
145
140
135
130
125
120
115
110
105
95
100
90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
15
5
10
0
Figure 175. Area of
river channel ca.
1850.
0
400
hectares
0.0 - 18.3
18.4 - 21.3
21.4 - 24.9
25.0 - 29.3
29.4 - 34.4
34.5 - 46.9
47.0 - 66.7
66.8 - 383.7
300
200
Figure 176. Area of
river channel 1995.
225
220
215
210
205
200
195
190
185
180
175
170
165
160
155
150
145
140
135
130
125
120
115
110
105
100
95
90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
15
10
5
0
100
0
12
kilometers
10
0.0 - 1.01
1.02 - 1.06
1.07 - 1.30
1.31 - 1.80
1.81 - 2.54
2.55 - 3.57
2.58 - 5.69
5.70 - 10.92
8
6
4
Figure 177. Length of
river channel ca. 1850.
225
220
210
215
205
200
195
190
185
180
170
175
165
160
155
150
145
135
140
130
125
120
115
110
100
105
95
90
85
80
75
65
70
60
55
50
45
40
35
25
30
20
15
10
5
0
2
0
12
kilometers
10
0.0 - 1.01
1.02 - 1.07
1.08 - 1.28
1.29 - 1.66
1.67 - 2.06
2.07 - 2.49
2.50 - 3.15
3.16 - 6.91
8
6
4
Figure 178. Length of
river channel 1995.
2
Portland Metro
Salem/Keizer
Albany Corvallis
225
220
215
210
205
200
195
190
185
180
175
170
165
155
160
150
145
140
135
130
125
120
115
110
105
100
95
90
85
75
80
70
65
60
55
50
45
40
35
30
25
20
15
10
5
0
0
Eugene/Springfield
Note: 1 hectare equals 2.47 acres, 1 kilometer equals .62 mile
Willamette River Basin Atlas
2nd Edition
135