Willamette Valley Ecoregion Summary
By Tamara S. Wilson and
Daniel G. Sorenson 1
Click to see available downloads for this ecoregion
Figure 1. Willamette Valley ecoregion and surrounding ecoregions. Information shown includes
land-use/land-cover data from the 1992 National Land Cover Dataset (Vogelmann
and others, 2001) and the 32 randomly selected 100 km2 sample blocks
used to create estimates of change for the entire ecoregion. Click to enlarge.
Ecoregion Description
The
Willamette Valley ecoregion (as defined by Omernik, 1987; Environmental
Protection Agency, 1999) covers approximately 14,400 km², making it one of the
smallest ecoregions in the conterminous United States (fig. 1). The alluvial valley
extends 193 km (120 mi) north to south over, ranging from 32 to 64 km in width
(20 to 40 mi), and is wedged between the sedimentary-basalt Coast Range to the
west and the basaltic Cascade Mountains to the east. The Willamette River watershed is the largest in Oregon, with headwaters in the Cascades draining northward into
the Columbia River near the ecoregion’s northern boundary in Washington State. Interstate Highway 5 runs the length of the Willamette Valley to its southern
boundary with the Klamath Mountains. Topography here is relatively flat with
elevations ranging from sea level to 122 meters (400 feet). The Willamette Valley is considered the most important agricultural region in Oregon given its even
terrain, mild wet winters, warm dry summers, and nutrient-rich soil (Sinclair, 2005, Oregon Department of Fish and Wildlife, 2006). More
than 2,300,000 people called Willamette Valley home in 2000 (Oregon Department
of Fish and Wildlife, 2006), and population centers are concentrated along the
valley floor (fig. 2). Portland is the
largest city with 529,121 residents (U.S. Census Bureau, 2000). Other sizable
cities in the ecoregion include Eugene, Salem (Oregon State capitol), and Vancouver, W ashington.
High-technology,
manufacturing, construction, and services industries dominate the urban economy
in the Willamette Valley, yet agriculture, forestry, and forest products are major
drivers of the region’s economy as well (figs. 3 and 4). The valley is a major
producer of grass seed, ornamental plants, fruits, nuts, vegetables, and grains,
as well as poultry, beef, and dairy products. The forestry and logging
industries are primary employers of the valley’s rural residents (Oregon
Department of Employment, 2006). Most of the forest cover is located along the
foothills lining the ecoregion’s perimeter (fig. 1). The combined impact of
timber harvesting (mechanical disturbance) and agricultural activities has altered
the Willamette Valley watershed, with forestry contributing to river
sedimentation and agricultural runoff impacting water quality in the Willamette River and its tributary streams (Oregon Department of Fish and Wildlife, 2006).
Recent years
have seen a marked decline in forest health related to the increased frequency
of multi-year droughts (ODF, 2008). Insect damage and
other diseases are also present, however, drought-related water stress is the
primary factor in coniferous tree mortality in the Willamette Valley. Trees most at risk include Douglas-fir (Pseudotsuga menziesii), grand fir (Abies
grandis), and western red cedar (Thuja plicata). Overstocking by
timber companies and planting on sites with poor soil conditions increases tree
susceptibility, with mortality occuring most often at low-elevation sites (Oregon
Department of Forestry, 2008) such as those sampled in our study.
Contemporary Land-Cover Change (1973 to 2000)
Between
1973 and 2000, the areal extent of land-use/land-cover (LU/LC) change in the Willamette Valley was 14.5 percent, or approximately 2,035.9 km² of area changed (table 1).
The footprint of change can be interpreted as the area that experienced LU/LC
change during at least one of the four multi-year periods comprising the 27-year
study period. Overall, an estimated 1,240 km² of land-cover experienced change
in at least one of the time periods, 594 km² changed during two time periods,
195 km² changed during three periods, and less than 7 km² of sampled land area
changed during all four time periods.
The
average annual rate of LU/LC change in Willamette Valley from 1973 to 2000 was 0.8
percent (113.6 km²) of change each year in the 27-year study period (table 2).
This measurement is a cumulative average of the annual average-change values
for each time period reported. A closer look at successive time intervals reveals
a steady increase in change estimates during the study period (table 2). Between
1973 and 1980, the annual rate of change was 0.4 percent (62.5 km²), increasing
to 0.7 percent (97.9 km²) from 1980 to 1986. The annual rate of change rises to
1.0 percent (139.6 km²) from 1986 to 1992 and to 1.1 percent (154.2 km²) from 1992
to 2000 (table 2).
Our results illustrate a dominance of four of the ten LU/LC classes
in Willamette Valley in 2000: agriculture (45.1 percent), forest/woodland (33.5
percent), developed/urban (12.6 percent), and mechanically disturbed (4.0 percent; table 3). These estimates from our sampled
area are similar to LU/LC percentages reported for the entire ecoregion by
the Oregon Department of Fish and Wildlife’s “Ecoregions: Willamette Valley
Ecoregion” report (2006). The remaining six LU/LC
classes account for the final 4.8 percent of the classified landscape in 2000.
Each of these classes alone represents less than two percent of the sampled area
(table 3). Between 1973 and 2000, the LU/LC classes experiencing statistically
significant (Wilcoxon statistical test for linear and quadratic trends, P<
0.05) net change in relation to the total ecoregion
area include net losses of forest cover (- 4.1 percent) and agricultural land
(- 2.2 percent) and net gains in developed area (+ 3.1 percent) and
mechanically disturbed land (+ 2.8 percent; table 3).
A
closer look at per period net change estimates reveals within-class variability
for those LU/LC classes experiencing spatio-temporal fluctuations during the
study period (fig. 5). Analysis of net change (area gained and lost) in
individual LU/LC classes by period shows class fluctuations throughout the 27-year
study period. Classes may experience gains and losses in area within and between
time intervals (fig.5). For example, mechanically disturbed land experienced a
net increase of 2.8 percent from 1973 to 2000, but variable rates of forest
cutting and other disturbances throughout the study period show a gross change (sum
of positive and negative change) of 3.3 percent (table 3). This equates to a
net change in mechanically disturbed area of 404.7 km² compared to a gross
change of 476.3 km² during the entire study period.
The
"from-to" information afforded by a post classification comparison allowed us
to identify LU/LC class conversions and to rank these conversions according to
their magnitude. The most frequent conversions from 1973 to 2000 are shown in
table 4. An estimated 80.6 percent of land-cover class conversions were related
to timber harvesting, forest loss, and successional regrowth (fig. 6). Of these
changes, 41.2 percent represent the mechanical disturbance of forests with 14.7
percent recovering directly back to forest and 11.6 percent transitioning to
grass/shrubland. Overall, the cumulative impact of forest clearing represents
1,254 km2 of disturbed landscape (table 4). The majority of LU/LC
changes in the Willamette Valley occurred in the low-elevation forested
foothills along the ecoregion periphery (fig.7). As a result, a percentage of
the forest loss in our year 2000 classifications may potentially be attributed to
hydrologic and climatologic changes affecting forest health in the ecoregion,
rather than to timber harvesting alone.
Our
analysis shows forestry intensifying during the study period - a trend not seen
in any other western U.S. ecoregion analyses. Because 96 percent of forested
lands in the Willamette Valley are privately owned (Oregon Department of Fish
and Wildlife, 2006), logging restrictions imposed on public lands by Federal
laws, such as the Endangered Species Act, are not enforceable. When the
northern spotted owl was listed as endangered in 1990, sizeable tracts of
federally-owned forest in the Pacific Northwest were taken out of production
and preserved as habitat (Daniels, 2005). Consequently, between 1988 and 1996,
timber harvesting rates in the Pacific Northwest fell 87 percent in national
forests alone and 38 percent overall (Warren 1992, 1999). However, dramatic
increases in timber harvesting occurred in the Willamette Valley at this time, in
contrast to regional trends. Our estimates show that between 1992 and 2000, an
additional 206 km2 (127.7 mi2) of timberland was
harvested or cleared as compared to the previous time period (1986 to 1992; table
4). These estimates suggest that increased logging restrictions on national
forest land outside of the Willamette Valley led to increased logging on
private land within the valley (Sinclair, 2005) to meet national and
international timber market demands (Daniels, 2005). Within the ecoregion, logging
restrictions on private land are only applicable where plots are adjacent to
fish bearing streams, as regulated by the Oregon Forest Practices Act (1971;
Sinclair, 2005). Where logging does occur in privately-held forests, the goal
is maximization of wood yield with more frequent cutting of younger trees (Sinclair,
2005).
Another
important LU/LC conversion is the loss of agricultural lands to development
(table 4). In the first change period (1973 to 1980), only 45.0 km² of
agricultural lands were converted to development, however, this number nearly
tripled between 1980 and 1986 to 132.0 km². This development trend drops off
somewhat in the third change period (1986 to 1992) to 77.0 km² and rises again
to 93.0 km² in the last change period (1992 to 2000). Individual sample block
results show these trends concentrated in 4 of the 32 sample blocks within the greater
Portland, OR. - Vancouver, WA. metropolitan area (63.7 percent between 1973
and 2000; sample blocks 17, 34, 40, 42; fig. 2). Of the four blocks, sample
block 17 alone accounted for 34.3 percent of farmland to development conversions
during the study period, and it is the only block located in Washington State. This trend may be a function of differential State land-use planning. Whereas
Oregon enacted the Land Conservation Development Plan back in 1973, Washington
land use planning law did not take effect until after passage of the Growth
Management Act in 1990 (Kline and Alig, 1997). The Oregon law mandated all cities
and counties in the state prepare urban growth plans and set growth boundaries;
this was completed in 1986. Oregon experienced some of its greatest urban growth
rates during this time (1980-1986), according to our change estimates. This
growth occurred just prior to completion and submission of the comprehensive
growth management plans.
References
Daniels, J.M., 2005, The
rise and fall of the Pacific Northwest log export market: U.S.
Department
of Agriculture, Forest Service, Pacific Northwest Research Station General Technical
Report PNW-GTR-624, 88 p. [http://www.fs.fed.us/pnw/pubs/pnw_gtr624.pdf,
last accessed Januraly 15, 2008]
EPA, 1999, Level III
Ecoregions of the Continental United States: U.S. Environmental Protection
Agency,
National Health and Environmental Effects Research Laboratory, Corvallis, Oregon, 1 map, scale 1:7,500,000.
Kline, Jeffrey, and Alig, R.J.,
1997, The impact of Oregon’s Land Use Planning Program on
forest
and agricultural land retention: Proceedings of the 31st Annual Pacific
Northwest
Regional
Economic Conference, Spokane, Wash., April 24-26, 1997.
Omernik, J.M., 1987, Ecoregions
of the conterminous United States: Annals of the Association of
American
Geographers, v. 77, p.118-125.
Oregon Department of
Employment, 2006, Oregon’s forestry and logging industry: Particularly
important
to rural areas: [http://www.qualityinfo.org/olmisj/ArticleReader?itemid=00004977,
last
accessed May 12, 2008]
Oregon
Department of Forestry, 2008,
Drought and conifer mortality in the Willamette Valley:
last
accessed April 16, 2008].
Oregon Department of Fish
and Wildlife, 2006, The Oregon Conservation Strategy:
Willamette Valley
Ecoregion: [http://www.dfw.state.or.us/conservationstrategy/document_pdf/b-eco_wv.pdf,
p. 234-235, last accessed May 23, 2008].
Sinclair, M., 2005, Willamette River Basin: Challenge of change: The Willamette Partnership, Oregon
Explorer,
p. 44.
U.S. Census Bureau, 2008, [http://www.census.gov/prod/www/abs/decennial/index.htm,
last
accessed
April 28, 2008].
Vogelmann, J.E., Howard,
S.M., Yang, L., Larson, C.R., Wylie, B.K., and van Driel, N., 2001,
Completion
of the 1990s National Land Cover Data Set for the conterminous United States
from
Landsat Thematic Mapper data and ancillary data sources: Photogrammetric
Engineering
&
Remote Sensing, v. 67, p. 650-662.
Warren, D.D., 1992,
Production, prices, employment, and trade in Northwest forest industries, fourth
quarter
1991: U.S. Department of Agriculture, Forest Service, Pacific Northwest
Research
Station,
Resource Bulletin PNW-RB-192, 112 p.
Warren, D.D., 1999,
Production, prices, employment, and trade in Northwest forest industries,
fourth
quarter
1997: U.S. Department of Agriculture, Forest Service, Pacific Northwest
Research
Station,
Resource Bulletin PNW-RB-230, 130 p.
Figures and Tables
Figure 1. Willamette Valley ecoregion and surrounding ecoregions. Information shown includes
land-use/land-cover data from the 1992 National Land Cover Dataset (Vogelmann
and others, 2001) and the 32 randomly selected 100 km2 sample blocks
used to create estimates of change for the entire ecoregion.
Figure 2:
Figure 2. Willamette Valley ecoregion and major metropolitan areas. 32 - 100
km² sample blocks were randomly selected. Green shaded areas represent
metropolitan areas.
[1]
U.S. Geological Survey, Western Geographic Science Center, Menlo Park, CA 94025
U.S. Department of the Interior |
U.S. Geological Survey
