Hierarchical Subdivisions of the Columbia Plateau
and Blue Mountain Ecoregions, Oregon and
Washington
by
Sharon E. Clarke
Sandra A. Bryce
United States Environmental Protection Agency
National Health and Ecological Effects Research Laboratory
Western Ecology Division-Corvallis, Oregon
-------
Hierarchical Subdivisions of the Columbia Plateau
and Blue Mountain Ecoregions, Oregon and
Washington
by
Sharon E. Clarke
Sandra A. Bryce
united States Environmental Protection Agency
National Health and Ecological Effects Research Laboratory
Western Ecology Division-Corvallis, Oregon
-------
Clarke and Bryce-2
Hierarchical Subdivisions of the Columbia Plateau and Blue Mountains Ecoregions,
Oregon and Washington
Sharon E. Clarke
Sandra A. Bryce
.P'
-AUTHORS
SHARON E. CLARKE is a faculty research assistant, Oregon State University, Corvallis,
OR 97331; SANDRA A. BRYCE is a staff scientist, Dynamac International, Inc.,
National Health and Environmental Effects Research Laboratory, Western Ecology
Division, Corvallis, OR 97333.
-------
Clarke and Bryce--3
ABSTRACT
Clarke, Sharon E; Bryce, Sandra A. 199X. Hierarchical subdivisions of the
Columbia Plateau and Blue Mountains ecoregtons, Oregon and Washington. Gen.
Tech. Rep. PNW-GTR-XXX. Portland, OR: U.S. Department of Agriculture, Forest
Service, Pacific Northwest Research Station. XX p
This document presents two spatial scales of a hierarchical, eooregional framework
and provides a connection to both larger and smaller scale ecological classifications.
The two spatial scales are subregions (1:250,1300} and landscape-level ecoregtons
(1:100,000), or Level IV and Level V ecoregions. Level IV ecoregions were developed
by the Environmental Protection Agency when it became apparent that the resolution
of national scale ecoregions provided insufficient detail to meet the needs of State
agencies for establishing biocriteria, reference sites, and attainability goats for water-
quaiity regulation. For this project, two ecoregions-the Columbia Plateau and the
Blue Mountains—were subdivided into more detailed Level IV ecoregions. Similarly,
the finer scale landscape-level ecoregions (Level V) were developed to address local
land-management issues. The landscape-level ecoregions for northeast Oregon and
southeast Washington were created specifically to address the issue of anadromous
fish habitat. However, their delineation employed landscape information similar to that
used in other levels of the ecoregion hierarchy, indicating the potential for general
application of these regions to both terrestrial and aquatic research questions. The
study area for the landscape-level ecoregions was defined by contiguous watersheds
-------
Clarke and Bryce--4
within the Columbia Plateau and Blue Mountain ecoregions to merge the ecoregional
information with units corresponding to fish distribution.
Keywords: Ecoregions, anadromous fish habitat, fish habitat, watershed classification,
landscape ecology, water quality, environmental mapping, classification.
-------
Clarke and Bryce-5
CONTENTS
Introduction
Background
Ecoregion Methodology
Applications of the Ecoregion Framework to Resource Management
Scope of Work
Literature Cited
Section 1 — Level IV Ecoregions of the Columbia Plateau Ecoregion of Oregon,
Washington, and Idaho
Columbia Plateau Ecoregion/Subregion Project
Water Resource Issues of the Columbia Plateau Ecoregion
Description of the Columbia Plateau Ecoregion
Materials
Ecoregion Boundary Decisions
Description of the Columbia Plateau Subregions
Literature Cited
Section 2—Level IV Ecoregions of the Blue Mountain Ecoregion of Oregon,
Washington, and Idaho
Climate
Geology—The Birth of the Blues
Blue Mountain Soils
Blue Mountain Vegetation
-------
Clarke and Bryce--6
Descriptions of the Blue Mountains Subregions
Subregions in the Cascade Rainshadow
Maritime-Influenced Zone
Melange Subregion
Wallowas/Seven Devils Subregion
Canyons and Dissected Highlands
Snake and Salmon River Canyons
Continental Zone Highlands
Continental Zone Foothills
Batholith Contact Zone
Blue Mountain Basins
Literature Cited
Section 3-Landscape-Level Ecoregions for Seven Contiguous Watersheds, Northeast
Oregon and Southeast Washington
Introduction
Objectives of the Project
Connection to Other Approaches
Rationale for Delineating Landscape-Level Ecoregions
Delineation of Landscape-Level Ecoregions
Conclusion
Literature Cited
Acknowledgments
-------
Clarke and Bryce-7
INTRODUCTION
Much of our disagreement about the proper role and function of regions, models and
other generalizations probably should be attributed to basic and inherent differences
between humans and their need for order, in which case it cannot be resolved by
rational discussion. Most of us are probably somewhere on a continuum between two
polar extremes. At one extreme are those tidy-minded souls whose instincts tell them
that the world should move with all the precision of a finely tuned watch,. . .that our
inability to detect regularity and order reflects our own weak analytical skills rather than
the possibility that it may not exist. At the other extreme are those rambunctious types
who perceive the world as a massive stochastic process, who glory in its disorder,
chaos, and complexity, and who revel in the thought that every leaf on every tree is
different. This battered old planet has quite enough evidence to keep both extremes
happily convinced that they are right.
—John Fraser Hart, 1982
Background
The condition of the Columbia Plateau and Blue Mountain ecosystems is presently the
focus of several Federal and State agencies. One hundred and fifty years of intensive
land use-logging, road building, mining, grazing, and fire suppression-have
-------
Clarke and Bryce--8
transformed forests and grasslands, threatening the existence and long-term
productivity of native ecosystems (Johnson and others 1994). Plans for restoration
include more holistic management of forest, range, and water resources.
The concept of ecosystem management has gained acceptance among land
management agencies, but implementing it can be a daunting prospect. It is difficult
for resource managers to apply the results of many individual studies to broad areas.
Different agencies have conflicting missions, overlapping jurisdictions, and
administrative boundaries that restrict an ecosystem focus over all. Canadian
resource managers have experienced the same difficulties; they have stressed the
importance of an ecosystem perspective, arguing that the bulk of environmental
research is concentrated on single-issue and single-medium subjects (Government of
Canada 1991, Omernik 1995).
How do we move spatially and conceptually from single issue research to consider
cumulative stresses and ecosystem response? To coordinate ecosystem
management in environmental research, Federal land-management agencies formed
an interagency task force to draft a Memorandum of Agreement that promotes the use
of a common, hierarchical ecosystem framework. Existing ecoregion frameworks
(Bailey 1983, 1994; Omernik 1987, 1995) provide the template for a common
framework, by delineating areas of similar landscape characteristics that reflect the
similarities in ecosystem type. By using a scheme that divides the landscape
ecologically rather than administratively, agencies will be better able to develop
-------
Clarke and Bryce--9
strategies that are focused, ecologically significant, and thus, more cost-efficient.
Federal agencies already have invested a great deal of time and effort in developing
their own frameworks for particular objectives. It will be an arduous process to agree
upon a common scheme; but if successful, it will be an important step toward
coordinating research efforts.
Recent movement toward ecosystem management within Federal land management
agencies has been prompted in large part by biological-diversity issues in forests of
the Pacific Northwest. Following President Clinton's April 1993 Forest Conference in
Portland, Oregon, various Federal agencies proposed initiatives to address problems
with endangered species, remnant old-growth forests, poor forest health, and declining
anadromous fish runs. The Eastside Forest Ecosystem Health Assessment and the
Eastside Scientific Panel are addressing similar forest problems in eastern Washington,
Oregon, Idaho, Montana, and Northern California (Everett and others 1993, Henjum
and others 1994).
The interagency Forest Ecosystem Management Assessment Team (FEMAT) report
(FEMAT 1993) lists four components of an Aquatic Conservation Strategy: the
establishment of riparian reserves and watershed refugia, and watershed analysis as a
foundation for watershed restoration. Ecoregion classifications at State and landscape
scales will contribute to all of the components of the Aquatic Conservation Strategy by
identifying groups of watersheds that are similar in ecosystem structure and by
describing their expected condition.
-------
Clarke and Bryce--10
A possible point of interagency contention is the tendency to consider watershed and
ecosystem frameworks as mutually exclusive approaches. As a framework for
research and planning, ecoregions provide a natural complement to watersheds.
Spatial patterns in ecosystems or environmental resources continue across
topographic divides (Omernik and Griffith 1991). In addition, streams may pass
through multiple ecoregions, and their watersheds often change dramatically in
character from headwaters to midstem to lowland sections. Figure 1 is an example of
the interplay between ecoregions and watersheds (Bryce and Clarke 1996). Because
streams B and C are within the same ecoregion, they probably have more in common
than B has with stream A. Although A and B are located within the same watershed,
they flow through distinctly different landscape types.
Data that offer no particular patterns when stratified by watershed may show patterns
when stratified by ecoregion. Both frameworks should be applied for a full exploration
of spatial patterns and management options; however, neither framework should be
stretched beyond its ability to explain variability and patterns in the data (Bryce and
Clarke 1996).
In this document, we report on the development and application of ecoregions at the
two highest resolution levels of the five-level hierarchy developed at the U.S.
Environmental Protection Agency (Omernik 1995). Level I and Level II classify
ecosystems on a continental scale for the North American continent; Level III
represents national scale ecoregions for the United States; Level IV regions are the
-------
Clarke and Bryce-11
more detailed ecoregions for State-level applications; and Level V are the most
detailed ecoregions for landscape-level or I local level projects. Further discussions of
the development of this ecoregion framework may be found in Bryce and Clarke
(1996), Clarke and others (1991), Gallant and others (1989), and Omernik (1987,
1995).
Ecoregion Methodology
Ecological regionalization is a form of spatial classification, the process by which
boundaries are drawn around relatively homogeneous areas at a specific scale or level
of detail. Ecoregions are developed through an iterative process that involves map
analysis, the collaboration of regional experts, an extensive literature review, and a final
integration of all available information. Physical characteristics such as climate,
geology, geomorphology, historical and present-day vegetation, soil, land use, and
hydrology are studied to determine the factors that reflect ecosystem character.
Map analysis is but one aspect of this approach. An extensive literature review adds
an understanding of regional-level ecosystem processes and guides establishment of
map units. Collaboration with regional experts from State and Federal agencies and
academia is also essential to the process. This collaboration has the dual benefit of
introducing field experience into the process, plus allowing input from those who might
use the final product.
-------
Clarke and Bryce-12
Finally, it is up to the geographer to integrate all of this disparate information.
Innumerable critical judgments are made based on research and experience. The
product is not the result of a mechanical overlay of maps; the computerized
geographic information system is merely a tool to aid in analysis and final display. The
final ecoregion map is an interpretation or model of reality. Boundary decisions may
represent a different suite of landscape characteristics from one area of a region to
another, depending upon the shifting dominance of the landscape characteristics.
Some of the boundaries represent present-day conditions, but some represent
conditions that no longer exist. Modeling the historic condition has applications for
distinguishing how far human impacts have taken the ecosystem from a more
unmanaged condition. Discussions of individual boundary decisions in the chapters to
follow will further address this issue.
The ecoregion delineation is completed by drawing lines directly onto 1:250,000 or
1:100,000 scale topographic maps for digitization, correcting for topography where
appropriate to produce a precise line. An additional map has been developed that
depicts the boundary transition widths (a "fuzzy boundaries" map) (Clarke and others
1991, Griffith and others 1994). An example of a fuzzy boundary map of the Columbia
Plateau Level III and Level IV ecoregions is included in section 1 (fig. 3).
Applications of the Ecoregion Framework to Resource Management
Because they are constructed through the use of many data sources, ecoregions do
-------
Clarke and Bryce-13
not have a specific theme, and thus, they have the potential for general application.
While they may not explain the response of any single ecological element, such as a
particular species distribution, they do represent an area of relative homogeneity in
landscape characteristics (Bryce and Clarke 1996). The ecoregion framework is an
organizational tool that may enable resource managers to apply an ecosystem
approach to assessment, monitoring, and research, and thus transcend administrative
boundaries. Several of the most widely used applications of ecoregions are listed
below:
Sampling design—In order to plan a monitoring program, there may be a need to
group a population of similar sites together from which a representative sample may
be chosen. The size and heterogeneity of a region indicates the appropriate number
of sites required to adequately represent an area. Fewer sites per unit area will be
needed in homogeneous landscapes and more in heterogeneous landscapes.
Identification of ecoregions aid in determining the density of the sampling frame to
ensure that important areas are not missed.
Ecoregions tend to run laterally across large topographic basins, particularly in some
mountainous areas. The headwaters, midstem, and mainstem of larger streams and
rivers may traverse different terrain types represented by unique ecoregions (fig. 1).
To capture the range of conditions, sites from multiple' watersheds within each
ecoregion can be grouped. Though no two sites are exactly alike in physical
characteristics or quality, a group of sites within an ecoregion tend to be similar.
-------
Clarke and Bryce--14
With a hierarchical regional framework, the scale of ecoregion may be chosen to fit a
particular project. Ideally, the scale should fit the type of data being gathered, the
spatial extent of sampling, and the research questions being asked.
Data analysis—The stratification of sites and sampling information into relatively
homogeneous groups reduces apparent variability and increases precision in data
analysis. Once the regional boundaries have been shown through field testing to be
representationally accurate, the regions provide a defensible area in which site-specific
results may be extrapolated. In this way, local project results may be applied to
broader scale resource management. Ecoregions lend a predictive capacity to
management decisions through the assumption that a subpopulation of sites within a
relatively homogeneous region will respond similarly to a specific type of management.
Water quality standards-Ecoregions have been used by states to structure their water
resource monitoring and assessment programs (Heiskary and Wilson 1989, Larsen
and others 1988, Rohm and others 1987). With a spatial framework, water resource
managers are able to set regionally appropriate management goals for lakes and
streams and design programs for nonpoint source pollution abatement and reduction
of effluent discharges. Setting water quality standards regionally means that streams
are less likely to be either over- or underprotected as they may with a single
nationwide or statewide standard.
Reference condition and ecosystem restoration—Ecoregions provide a framework for
-------
Clarke and Bryce-15
locating waterbodies to serve as reference sites. Reference sites can be chosen from
within any scale in the ecoregion hierarchy. As a benchmark for attainable conditions,
a reference site should be both regionally representative and relatively unimpacted
(Gallant and others 1989, Hughes and others 1986). Reference areas, chosen within a
regional framework, are representative of a particular population of streams and serve
as a model of attainable condition. Until now, water resources have been the focus of
the regional reference site concept. However, with the increasing interest in
ecosystem management, there is an opportunity to expand the implementation of the
concept to relatively unimpacted terrestrial sites that are representative of the
landscape character of the ecoregion.
Rehabilitation of disturbed sites can proceed most effectively if there is a measurable
goal represented by the reference site, even if that goal is not completely attainable.
Other sites of varying disturbance can be compared with this model and ranked along
a continuum of condition. Relating present condition to the potential capability of the
systems contributes to the search for the probable causes of condition. Once
probable causes of degradation are identified, managers can set priorities for cost-
effective restoration.
Scope of Work
This introductory section presented the rationale and methodology employed in the
ecoregion process as well as applications of the ecoregion framework. The next two
-------
Clarke and Bryce-16
sections, 1 and 2, describe Level IV ecoregions displayed in the accompanying map
for the Columbia Plateau and Blue Mountain ecoregions. Photographs 1 to 16 in
figure 2 depict the Level IV ecoregions. The complexity and relative importance of
factors considered in the development of ecoregions varied greatly between these two
ecoregions. The same landscape characteristics were considered in each ecoregion
project, but different combinations of factors tended to dominate depending upon the
character of the region For example, the landscape of the Columbia Plateau is a
product of a dramatic series of events during the Pleistocene (Bretz 1969). Its
ecosystems are a reflection of the arid climate, soil, glacial geomorphology, and fluvial
erosion. The clarity of these features made the regionalization of the Columbia Plateau
relatively straightforward. The greater difficulty there arises from applying an
ecosystem framework to an intensively farmed, largely privately owned, arid region. In
the Blue Mountains, on the other hand, the dominant factors are climatic and
elevational gradients imposed upon an extremely complex geology. The complexity of
the landscape and the general lack of biological, geological, and climatic information,
particularly in higher elevations, makes this project a first iteration of an ongoing
refinement of ecological regions there.
Section 3 describes the development of landscape-level ecoregions (Level V) for seven
contiguous watersheds in northeastern Oregon and southeastern Washington located
within both the Columbia Plateau and Blue Mountain Ecoregions. The Bonneville
Power Administration (BPA) is required by law to restore anadromous fish runs to
compensate for fish losses incurred at dams on the Columbia River (Pacific Northwest
-------
Clarke and Bryce-17
Electric Power Planning Act 1980). The BPA funded this project to develop landscape-
level ecoregions as a means to help define expected habitat conditions. The project is
an effort to stratify a complex area into more homogeneous units that may have similar
potentials and reactions to stressors. In this section, we outline the connection
between stream habitat characteristics and landscape characteristics, and we describe
how each of the major landscape characteristics might affect fish habitat. Figure 2
illustrates the connection between the Level IV and Level V ecoregion hierarchy.
Photos 1 to 5 and 6 to 16 depict the Level IV ecoregions of the Columbia Plateau
Ecoregicn and Blue Mountains Ecoregion. To demonstrate the variability within these
subregions, represented by landscape-level ecoregions (Level V), photos 21 to 27
show the diversity within the Maritime Subregion, and photos 27 to 29 show the
diversity within the Canyons and Dissected Highlands Subregion.
The landscape-level ecoregions described in Section 3 were developed specifically for
anadromous fish habitat research and management (Bryce and Clarke 1996).
However, the methodology and types of data sources are the same as those used in
ecoregion delineation at broader scales in the hierarchy. For this reason, we believe
that the classification may have broader application to other resource issues. The full
complement of regions at a particular scale may not be necessary for a specific
project; regions may be aggregated depending upon the objectives of the project.
The rationale for making boundary decisions has been discussed in detail, which
should help in assessing the usefulness of the classification for other resource issues.
-------
Clarke and Bryce--18
Literature Cited
Bailey, R.G. 1983. Delineation of ecosystem regions. Environmental Management. 7:
365-373.
Bailey, R.G.; Avers, P E.; King, T.; McNab, W.H., eds. 1994. Map of ecoregions and
subregions of the United States. Washington, DC: U.S. Department of Agriculture,
Forest Service. Map scale 1:7,500,000.
Bretz, J.H. 1969. The Lake Missoula floods and the Channeled Scablands. Journal of
Geology. 77: 505-543.
Bryce, S.A.; Clarke, S.E. 1996. Landscape-level ecological regions: linking state-level
ecological frameworks with stream habitat. Environmental Management. 20: XX-XX.
Clarke, S.E.; White, D.; Schaedel, A.L. 1991. Oregon, USA, ecological regions and
subregions for water quality management. Environmental Management. 15(6):
847-856.
Everett, R.E.; Hessburg, P.; Jensen, M.; Bormann, B. 1993. Eastside forest ecosystem
health assessment. Volume I. Executive summary. Wenatchee, WA: U.S. Department
of Agriculture, Forest Service, Forest Sciences Laboratory. 57 p.
-------
Clarke and Bryce-19
Forest Ecosystem Management Assessment Team. 1993. Forest ecosystem
management: an ecological, economic, and social assessment. Portland, OR: U.S.
Department of Agriculture; U.S. Department of the Interior [and others]. [Irregular
pagination].
Gallant, A.L; Whittier, T.R.; Larsen, D.P.; Omernik, J.M.; Hughes, R.M. 1989.
Regionalization as a tool for managing environmental resources. EPA/600/3-89/060.
Corvallis, OR: U.S. Environmental Protection Agency, Corvallis Environmental
Research Laboratory. 152 p.
Government of Canada. 1991. The state of Canada's environment. Ottawa, Canada:
Canada Communication Group-Publishing.
Griffith, G.E.; Omernik, J.O.; Wilton, T.F.; Pierson, S.M. 1994. Ecoregions and
subregions of Iowa: a framework for water quality assessment and management.
Journal of the Iowa Academy of Sciences. 10(1): 5-13.
Hart, J.F. 1982. The highest form of the geographer's art. Annals of the Association of
American Geographers. 72: 1 -29.
Henjum, M.G.; Karr, J.R.; Bottom, D.L.; Perry, D.A.; Bednarz, J.C.; Wright, S.G.;
Beckwitt, S.A.; Beckwitt, E. 1994. Interim protection for late-successional forests,
fisheries, and watersheds: national forests east of the Cascade crest, Oregon and
-------
Clarke and Bryce--20
Washington. Bethesda, MD: The Wildlife Society. 245 p.
Heiskary, S.A.; Wilson, C.B. 1989. The regional nature of lake water quality across
Minnesota: an analysis for improving resource management. Journal of Minnesota
Academy of Sciences. 55(1): 71-77.
Hughes, R M.; Larsen, D.P.; Omernik, J.M. 1986. Regional reference sites: a method
for assessing stream potentials. Environmental Management. 10: 629-635.
Johnson, C.G., Jr.; Clausnitzer, R.R.; Mehringer, P.J.; Oliver, C.D. 1994. Biotic and
abiotic processes of eastside ecosystems: the effects of management on plant and
community ecology, and on stand and landscape vegetation dynamics. Gen. Tech.
Rep. PNW-GTR-322. Portland, OR: U.S. Department of Agriculture, Forest Service,
Pacific Northwest Research Station. 66 p.
Larsen, D.P.; Dudley, D.R.; Hughes, R.M. 1988. A regional approach to assess
attainable water quality: an Ohio case study. Journal of Soil and Water Conservation.
43(2): 171-176.
Omernik, J.M. 1987. Ecoregions of the conterminous United States. Annals of the
Association of American Geographers (text and map supplement). 77(1): 118-125.
Omernik, J.M. 1995. Ecoregions: a spatial framework for environmental management.
-------
Clarke and Bryce-21
In: Davis, W.S.; Simon, T.P., eds. Biological assessment and criteria: tools for water
resource planning and decision making. Ann Arbor, Ml: Lewis Publishers: 49-65.
Omernik, J.M.; Griffith, G.E. 1991. Ecological regions versus hydrologic units:
frameworks for managing water quality. Journal of Soil and Water Conservation. 46:
334-340.
Pacific Northwest Electric Power Planning and Conservation Act. 1980. 16 USC = 839-
839H.
Rohm, C.M.; Giese, J.W.; Bennett, C.C. 1987. Evaluation of an aquatic ecoregion
classification of streams in Arkansas. Journal of Freshwater Ecology. 41(1): 127-140.
Thiele, S.A.; Kiilsgaard, C.; Omernik, J.M. 1992. The subdivision of the coast range
ecoregion of Oregon and Washington. Corvallis, OR: U.S. Environmental Protection
Agency, Environmental Research Laboratory. 39 p.
Warren, C.E. 1979. Toward classification and rationale for watershed management and
stream protection. EPA-600/3-79-059. Corvallis, OR: U.S. Environmental Protection
Agency. 143 p.
-------
Clarke and Bryce--22
SECTION 1 - LEVEL IV ECOREGIONS OF THE COLUMBIA PLATEAU ECOREGION
OF OREGON, WASHINGTON, AND IDAHO
by Sandra A. Bryce and James M. Omernik
Columbia Plateau Ecoregion/Subregion Project
The impetus for the Columbia Plateau regionalization project was a mutual concern of
EPA Region X, the Washington Department of Ecology, and the Oregon Department of
Environmental Quality to better frame management decisions about nonpoint-source
(NPS) pollution. The agencies also agreed that the refinement of ecoregion lines, the
delineation of subregions (Level IV ecoregions), and the selection of reference sites
were prerequisites for establishing biocriteria standards in portions of Oregon,
Washington, and Idaho. An ecosystem-based framework also afforded the
opportunity for the three States to share data and assessment results in regions of
similar natural capacities. Though the original project had a water quality focus,
ecoregions are delineated using terrestrial landscape information; thus, they have the
potential for terrestrial applications as well.
Although geographers at EPA's National Health and Environmental Effects Laboratory,
Western Ecology Division (NHEERL-WED) in Corvallis, Oregon, were responsible for
the final delineation of ecoregion boundaries, the Columbia Plateau project was a
collaborative effort with State resource managers and regional experts. Meetings and
-------
Clarke and Bryce--23
phone consultations were held with State and Federal agency personnel and members
of academia to review progress and exchange ideas. Such qualitative input was
critical to the creation of a regional framework that will be meaningful to those using it.
Continued collaboration is also necessary for the evaluation of the boundaries derived
from the geographic data.
Description of the Columbia Plateau Ecoregion
The arid nature and earth-toned landscape of the Columbia Plateau belies the
spectacular origins of its topography. The Columbia Plateau was shaped by episodic
geologic events of epic proportions. During the Miocene epoch, between 17 and 6
million years ago, lava flows erupted from vents in southeastern Washington and
northeastern Oregon to fill the basin with lava up to 2 miles thick. As many as 2(30
separate flows eventually covered an area approximately 160,000 square kilometers
(Baker and others 1991). The earth's crust gradually sank under the weight of the
lava, producing the sloping Columbia Plateau (Allen and others 1986). During the
Pleistocene, a continuous covering of loess was aerially deposited over the basalt.
Because of the area's general saucer shape, sloping from its edges to the low point in
the Pasco Basin, the Columbia Plateau is often called the Columbia Basin. However,
since the term "basin" also indicates drainage area and to avoid confusion hereafter,
the region will be called the Columbia Plateau ecoregion in this report.
The features on the surface of the plateau were sculpted during the Pleistocene.
-------
Clarke and Bryce--24
Those features that are most discernible are thought to have occurred between 12
and 15,000 years ago during the last advance of continental glaciers (Allen and others
1986). Periodically during this time, the Clark Fork River in Montana was dammed by
a lobe of the glacier, forming the huge Lake Missoula, which covered 3000 square
miles (4825 sq km) of intermountain valleys in Montana. Rising waters regularly
breached the ice dam sending huge quantities of water surging across the Columbia
Plateau. These floods varied in intensity, but the largest are said to have contained up
to 10 times the flow of all rivers in the world or 60 times the flow of the Amazon River
(Allen and others 1986, GS 1982). The flood channels cut through the thick deposits
of windblown soil, leaving islands of loess separated by scablands and bedrock
channels.
Attributing the topography of the Plateau to massive flood events was the painstaking
and controversial life's work of geologist J. Harlan Bretz, who crossed much of the
scabland country on foot, mapping as he went and piecing together the evidence of
the Missoula floods (Allen and others 1986). What took Bretz 40 years to compile we
can now visualize in an instant with the aid of satellite imagery—the mammoth braided
channels crossing the entire plateau from northeast to southwest.
-------
Clarke and Bryce--25
The Columbia Plateau is an arid shrub/grassland surrounded on ail sides by forested
mountain ranges: the Cascade Mountains to the west, the Okanogan Highlands to
the north, the Blue Mountains to the south, and the Rocky Mountains to the east. The
climate of the Columbia Plateau is continental with a marine influence. Summers are
hot and dry with most of the precipitation falling during the winter months.
Precipitation increases across the plateau from southwest to northeast. It is lowest in
the basins on the western side of the plateau, where the rain shadow effect of the
Cascade Mountains limits precipitation to 15.2 to 22.9 centimeters per year (6 to 9 in
per yr). Rainfall gradually increases to the east with higher elevations and proximity to
the Rocky Mountains and Blue Mountains.
Temperature, moisture, and mineral content determine soil formation, and the texture
affects the rate at which complex soil horizons, organic matter, and clay layers develop
(SCS 1981a). The lack of moisture on the Columbia Plateau results in soil low in
organic matter and clay. Depending on their location, soils of the Columbia Plateau
may be derived from basalt colluvium on hiilslopes and canyon walls, alluvium from
valley floors, or gravelly flood deposits and glacial outwash. However, the
predominant soil of the Plateau is loess. Loess forms a large deposit called the
Palouse Formation that covers the entire area of the Columbia Plateau to depths of 75
meters (22.9 ft) in the southeastern Palouse Hills (Baker and others 1991). It is of
granitic and metamorphic origin, transported by glaciers from the Rocky Mountains to
the east and north and redeposited by prevailing southwesterly winds over the past
50,000 years (Baker and others 1991, SCS 1981a).
-------
Clarke and Bryce--26
Gradually increasing precipitation toward the northeast influences the dryland
agricultural capability of the loess soil. Cropping techniques vary from a 2-year winter
wheat/fallow rotation in areas with 22.9 to 38.1 centimeters per year of precipitation (9
to 15 in) to a 3-year winter wheat/spring barley/fallow in areas receiving 38.1 to 45.7
centimeters per year of precipitation (15 to 18 in) (SCS 1981b). Annual cropping on -
unirrigated land is only practiced in the Palouse Hills, where moisture levels reach 45.7
to 58.4 centimeters per year (18 to 23 in per yr). Loess is highly productive, but it is
also highly erosive. In the heart of the Palouse River basin, farming on steep slopes
and farming practices such as clean tilling fallow fields leads to average erosion rates
of 44.8 tons per hectare (20 tons per ac) through rill and sheet erosion during the wet
season (Steiner 1987).
The type and amount of vegetation of the Columbia Plateau also varies with
temperature and moisture. The low precipitation amounts and negligible yearly runoff
combined with high evapotranspiration and deep, dry loessial soil create an
environment ill-suited to tree growth. Though the Plateau has been called a desert,
Daubenmire (1970) described it as a steppe. He defined a desert as a region too dry
to support grassland and a steppe as a region with enough moisture to support
grasses but not trees. According to this definition, the Columbia Plateau is a
grassland steppe. Daubenmire has divided the Plateau into nine vegetation zones.
The big sagebrush/bluebunch wheatgrass and big sagebrush/Idaho fescue
associations occur in the driest core of the Plateau. Grasses without the sagebrush
element grow in the slightly wetter eastern portion of the Plateau. The other five
-------
Clarke and Bryce-27
associations appear around the perimeter of the Plateau in the lusher meadows of
grasses, broad-leafed forbs and shrubs that form the transition to forest. The shrub
element varies from bitterbrush in the eastern Cascade foothills to snowberry sp. in the
Rocky Mountain transition zone.
There is no evidence that the plant life of the Columbia Plateau has evolved under
either a grazing or a fire regime. The bunchgrasses will tolerate light grazing after
seed formation. They do not recover after heavy grazing but are replaced by alien
grasses, such as cheatgrass (Bromus tectorum). The brushy species, on the other
hand, are sensitive to fire. It is unlikely that large herds of grazing animals roamed the
Plateau as they did the Great Plains or that the original inhabitants used fire as a
hunting strategy (Daubenmire 1970, Franklin and Dyrness 1974).
In summary then, the Columbia Plateau differs from adjacent mountainous, forested
ecoregions in its aridity, its low relief, and its lack of trees. However, no single
attribute described above, not geology, climate, topography, soil, or vegetation, is
sufficient to determine the ecoregion boundaries of the Columbia Plateau. For
example, the flood basalts that comprise the Columbia Plateau also form the
foundation of the Blue Mountains to the south. The Blue Mountains ecosystem is very
different from the Columbia Plateau, yet large portions of it have the same geology.
As a result, geology is not useful for making the southern ecoregion boundary
decisions. Similarly, vegetation is not a dominant feature in boundary decisions in the
Deschutes River area of Oregon. Similar vegetation associations occur just to the
-------
Clarke and Bryce--28
south in the high desert area of Oregon (as they do in southern Idaho and Montana).
Producing a map of ecological regions requires integrating all of the landscape
characteristics and developing a rationale for the boundaries that is consistent with the
projected uses for the ecoregions. In this way, ecoregion maps differ from thematic
maps which seek to define the distribution of a single characteristic.
Materials
Component maps and data for the Columbia Plateau project were assembled from the
participating States, from EPA's NEERL-WED map library, and from Oregon State
University's map library. Additional information, such as historical documents, satellite
images, and pertinent published articles, contributed to the refinement and evaluation
of subregion boundaries.
The data that was most useful for the subdivision of the Columbia Plateau ecoregion
were maps of geology, soil, potential vegetation, topography, and land use/landcover.
Digital coverages at 1:250,000 scale from a series of soil maps (STATSGO) were
acquired from the U.S. Department of Agriculture's Soil Conservation Service (Note: in
late 1994, USDA changed the name of SCS to the National Resources Conservation
Service [NRCS]). These maps were supplemented by existing county level 1:24,000
scale soil maps and texts (cited at the end of this section under SCS by year). A
present-day vegetation map of Oregon sponsored by the U.S. Department of the
Interior's Fish and Wildlife Service (FWS) was also available in digital form from the
-------
Clarke and Bryce--29
Idaho Department of Water Resources. This map was supplemented with smaller
scale maps of Oregon and Washington vegetation and literature about the vegetation
communities of the Plateau. Geological information was provided by a 1:500,000 scale
map of Oregon, both digital and hard copy (Walker and McLeod 1991) and 1:250,000
scale geologic maps of the northeast and southeast quadrants of the State of
Washington (Schuster 1995, Stoffel and others 1991). U.S. Department of the Interior,
Geological Survey (GS) land-use maps were available in digital form. Finally, at
1:250,000 and 1:100,000 scale, GS topographic maps were indispensable for the
interpretation of land surface form, drainage pattern, contour intervals, gradient,
aspect, and elevation. Topographic maps serve as a reality check for the other maps,
especially those that have been interpreted, such as the soil, geology and vegetation
maps. The ecoregion boundaries were delineated and digitized using the 1:250,000
scale topographic maps as base maps.
Ecoregion Boundary Decisions
During the regionalization process, some boundaries between distinctive ecoregions or
subregions appear obvious whereas others are more difficult to determine. For
example, the Channeled Scablands subregion of the Columbia Plateau is determined
by the clearly visible boundaries of flood channels gouged out of the landscape.
Other boundaries are not so readily apparent; their delineation requires an
accompanying rationale, which is the result of integrating the multiple source materials.
These less-apparent boundaries, such as those at the change from grassland to forest
-------
Clarke and Bryce--30
at the perimeter of the Plateau, represent transition areas of various widths along
ecological gradients. One method of mapping this phenomenon is through the use of
"fuzzy boundaries" (Clarke and others 1991). A boundary transition width map is
included with the ecoregion/subregion map for the Columbia Plateau (plate 3). Sites
near region boundaries, especially in transition areas, tend to have characteristics of
both regions. Field visits are necessary to determine which regional characteristics
dominate at sites of interest.
Other boundary considerations arise because of variations in the extent that human
activities have changed the landscape. In an attempt to have boundaries represent
relatively constant conditions, ecological regions are defined on the basis of potential
capability. As Daubenmire found when trying to relate the distribution of vegetation to
climate patterns (Daubenmire 1956), the correlations between climate and vegetation
were meaningful only where vegetation boundaries were relatively stable. He knew
that fossil-pollen studies indicated that very little climate-induced change had occurred
in plant distributions since the end of the last glacial advance. However, the high
frequency of human changes and the extirpation of species confounded his search for
patterns.
Anthropogenic changes in the Columbia Plateau have continued unabated since
Daubenmire's investigations in the 1950s. Large irrigation projects across the driest
parts of the Plateau, once known as the "Great Inland Desert," have markedly changed
the groundwater and surface-water patterns. Areas irrigated with surface water from
-------
COLUMBIA PLATEAU ECOREGION/SUBREGION BOUNDARY TRANSITION WIDTHS
10d
t10t
St
j i , y i-
' ! \ / ¦ r
-------
Clarke and Bryce-31
the Columbia and Yakima Rivers are experiencing a dramatic increase in the amount
of groundwater and excess irrigation runoff. For example, the extent of marshland has
greatly increased in the Potholes region near Moses Lake, Washington. Long-term
groundwater recharge estimates, covering the period 1956 to 1977, for both
predevelopment and modern land-use conditions in the core area of the Columbia
Plateau, showed an increase of 5.7 centimeters per year (2.23 in per yr) on average in
the rate of groundwater recharge. For the zone covering the Moses Lake area,
recharge increased from 1.45 centimeters per year (0.57 in per yr) to 25.6 centimeters
per year (10.07 in per year). Conversely, in areas that are irrigated from wells, water
withdrawals have caused a decrease in recharge across the zone (Bauer and Vaccaro
1990).
These dramatic changes have not been incorporated into the boundary decisions for
subregion lines. Although it is likely that wide areas of the Columbia Plateau will never
return to their original condition, we cannot gauge the extent of the human induced
changes to ecosystems unless we use a model of the presettlement condition as a
guide. There is enough information available on temperature and precipitation
regimes, soil and native vegetation to reconstruct the boundaries of the presettlement
ecosystems. Human caused changes, such as the modifications in groundwater and
surface runoff in irrigated areas, or the dominance of cheatgrass and other alien plants
across the formerly discrete vegetation zones of the plateau, can be treated as a
separate layer of information which is superimposed upon the model of ecosystem
potential.
-------
Clarke and Bryce--32
Description of the Columbia Plateau Subregions
The ecoregions described in table 1 are presented in the map of the Columbia Plateau
subregions (fig. 3). Native vegetation of the Columbia Plateau has been listed in table
1.
Channeled scablands and Loess Islands (fig. 2, 10a; fig. 3, 10a and 10b)—The
Channeled Scablands were formed as immense floods periodically broke through the
ice dams blocking glacial Lake Missoula during the Pleistocene. The immense
quantity and high speed of the surging water scoured away the loess covering the
Plateau as well as portions of the underlying basalt bedrock. The vertical jointing of
the basalt makes it susceptible to "plucking." This phenomenon can be seen in the
steep vertical walls of the flood channels, in the numerous small lakes, called kolk
lakes, gouged out of the basalt surface, and in the giant plunge pools created at huge
cataracts. In the northwest part of the Plateau, the scabland channels, such as Grand
Coulee and Moses Coulee, also served as outflow channels for glacial Lake Columbia
and the Okanogan lobe of the Wisconsin ice sheet.
Patterned ground covers the basalt plateaus bordering the main flood channels.
These "scabs" are composed of mounds of loess surrounded by rock fragments. The
soil is a thin veneer 0 to 62.5 centimeters (0 to 25 in) on the basalt surface of the
scablands (SCS 1981b).
-------
Clarke and Bryce--33
These scabland tracts cover gently sloping topography. Elevations increase 7.6
meters per kilometer (25 ft per mi) across Lincoln County in the central part of the
Plateau (SCS 1981b). Precipitation amounts range from 17.8 to 45.7 centimeters per
year (7 to 18 in per yr) from west to east. The scablands are too dry to support trees
except in the Cheney area on the far east side of the Plateau, where the absence of
loess and more abundant precipitation allow ponderosa pines to root in the fractured
basalt (Hooper and Reidel 1989). The most common native vegetation of the
scabland channels is the stiff sage/bluegrass association. The present day land use
is grazing.
The Loess Islands are the post-flood remains of the once unbroken mantle of wind'
deposited silt (loess) that covered the entire Plateau. The loess islands range in size
from small scraps to more than 1000 square kilometers (622 sq mi). The loess is
thinner across the central and western portions of the Columbia Plateau than it is in
Palouse Hills to the east. Future investigations of the stratigraphy of loess islands may
clarify the timing of the flooding episodes (Baker and others 1991). Precipitation in the
Loess Islands subregion ranges from 17.8 to 45.7 centimeters per year (7 to 18 in per
yr) from west to east across the Plateau. The distribution of native vegetation
generally follows this change in moisture availability. The big sage/bluebunch
wheatgrass association grades into the bluebunch wheatgrass/ldaho fescue
association as precipitation increases. A different sage species, threetip sage, grows
with Idaho fescue in a wide band around the northern perimeter of the Plateau,
roughly in the 12 to 18 inch (30.5 to 45.7 cm per yr) precipitation range. As in the
-------
Clarke and Bryce-34
other meadow steppe communities on the fringes of the plateau, the threetip sage and
Idaho fescue form a lush complex with a wide variety of broad-leafed forbs.
Present day land use has transformed the loess islands into wheatfields. Cropping
systems vary according to precipitation amount. The drier areas (22.5 to 37.5
centimeters per year [9 to 15 in per yr] precipitation) are farmed in a winter wheat and
fallow rotation; and the wetter areas (37.5 to 45 centimeters [15 to 18 in per yr]
precipitation) are managed in a winter wheat, spring barley, and fallow rotation.
Umatilla Plateau (fig. 3, 10c)—The Umatilla Plateau is bounded on the north by the
basin of glacial Lake Condon, on the west by the eastern slopes of the Cascades, and
on the south by the transition of nonforested loessial soil to the forest soils of the Blue
Mountains. Precipitation increases with increasing elevation, from 20.3 centimeters per
year (8 in per yr) at the margin of the glacial lake basin to 58.4 centimeters per year
(23 in per yr) at the upper elevations. Because the Umatilla Plateau begins above the
drier glacial lake basin, sagebrush is not a significant member of the plant community.
Grasses, bluebunch wheatgrass, and Idaho fescue predominate throughout the
Umatilla Plateau. Fescue dominates in higher areas with greater precipitation and on
steeper northern slopes (Anderson 1956). In the grassland to forest transition zone,
shrubs such as rose, hawthorne, and snowberry mingle with the grasses.
The soil and landform patterns of the Umatilla Plateau follow concentric levels of
terraces rising to the Blue Mountains in the south. From north to south, the deep
-------
Clarke and Bryce--35
loessial soils (Walla Walla and Ritzville) become thinner (Condon and Morrow) as the
rolling wheatfields become interspersed with flatter areas of patterned ground. The
Blue Mountain foothills and piedmont alluvial fans have a very thin loess covering and
steeper slopes. The soils on these upper terraces (Gurdane and Gwin) contain a
higher component of basaltic rock fragments. Here the land-use changes from wheat
farming to grazing on thinner soils (SCS 1988). It is reasonable to assume that the
rest of the Columbia Plateau once resembled the Umatilla Plateau, before the massive
Missoula floods rearranged the loess mantle, creating the channeled scablands and
lacustrine deposits.
In the western section of the subregion, the tablelands isolated by the Deschutes and
John Day river canyons are broad enough to support dryland wheat farming. A
traveler on these isolated tablelands is unaware of the canyons falling away on either
side. The southern boundary on the western end of the Umatilla Plateau is marked by
the change from loess-dominated soils to the clays derived from the ancient
sedimentary deposits of the John Day Formation. Juniper appears on this soil type
and marks the westernmost extension of the Blue Mountains/High Desert area of
central Oregon (Anderson 1956).
Okanogan Drift Hills (fig. 3, 10d)-The Okanogan Drift Hills subregion, located in the
northwest corner of the Columbia Plateau ecoregion, follows the valley of the
Okanogan River north to the Canadian border. The Okanogan lobe of the Wisconsin
Glacier advanced far enough down the valley to dam the Columbia River, creating
-------
Clarke and Bryce--36
glacial Lake Columbia and rerouting the Columbia River through Grand Coulee. As
the glacier melted, it retreated up the Okanogan valley, leaving behind a blanket of
glacial till. The till, up to 15.2 meters (50 ft) thick, is composed of clay, silt, sand,
gravel, cobbles, and boulders. Though the region has a thin veneer of loess on top of
the till, the characteristic deep loessial soil of the Plateau has been transformed by
glacial action. Classic glacial features such as drumlins, kames, eskers, terminal
moraines, and kettle ponds are common.
The Okanogan valley lies in the rainshadow of the North Cascade Mountains. Rainfall
in the valley bottom is 22.9 to 30.5 centimeters per year (9 to 12 in per yr), the same
as in the Pasco Basin to the south. Rolling uplands bordering the valley reach
elevations of 912 meters (3000 ft) with precipitation of 27.9 to 38.1 centimeters per
year (11 to 15 in per yr). In the valley bottoms, present day land use includes irrigated
agriculture and orchards. The uplands support dryland wheat farming and grazing.
Native vegetation includes big sage and bluebunch wheatgrass in the drier areas and
the threetip sage/Idaho fescue association in the moister uplands. West of the
Okanogan River, bitterbrush grows with the fescue as the upper grassland-to-forest
transition association. In the higher elevations of the Okanogan highlands, the
sagebrush is replaced by a fescue/forb groundcover in mountain meadows
(Daubenmire 1970). At elevations where precipitation exceeds 40.64 centimeters (16
in), ponderosa pine grows with a fescue and bitterbrush understory. The boundary
denoting treeline, or the presence of forest soils in areas cleared for agriculture, forms
-------
Clarke and Bryce--37
the east and west boundaries of the Okanogan Drift Hills subregion. The southern
boundary follows the Pleistocene glacier's terminal moraine along Dutch Henry Draw
in Douglas County. This subregion is the southern extension of a glaciated ecoregion
in Canada covering the Okanogan country of British Columbia.
Pleistocene Lake Basins (fig. 2, 10e; fig. 3, 10e)-Vast temporary lakes formed
periodically with the release of flood waters from glacial Lakes Missoula and Columbia
during the Pleistocene. Lake Lewis formed from the damming of the Columbia River
at Walluia Gap on the southern Washington border, and covered 4825 square
kilometers (3000 sq mi) of the Quincy and Pasco basins and Walla Walla and Yakima
River valleys. Downriver, the flood waters ponded again at the entrance to the
Columbia Gorge, creating Lake Condon. High-water marks and faint shorelines mark
the margins of the lake basins at the 304- to 365-meter (1000- to 1200-ft) contour.
The water level at Yakima has been estimated at 61 meters (200 ft) above the present
city. Ice-rafted erratic boulders have been found stranded on ridges surrounding the
basins (Allen and others 1986, Baker and others 1991). Flood waters, tearing through
the channeled scablands, dropped their load of loess, sand, and outwash gravel in the
basins. One of the largest flood-related features, the Priest Rapids bar, formed
downstream of Sentinel Gap and fills the basin north of Richland.
The lake basins are in the driest areas of the rain shadow of the Cascade Mountains
receiving 15.2 to 30.5 centimeters (6 to 12 in) of precipitation per year. Native
vegetation consists of the widespread big sagebrush/bluebunch wheatgrass
-------
Clarke and Bryce-38
association. These sagebrush areas have been transformed by large irrigation
projects that provide Columbia and Yakima River water via pump and canal. The
synclinal Kittitas valley has been included in this subregion even though it was not part
of glacial Lake Lewis because of its position within the Yakima Folds subregion (10g)
and because it has a similar lacustrine history, climate, soil, and land-use capability.
Canyons and Dissected Uplands (fig. 2, 10f; fig. 3, 10f)—In this subregion, the Snake,
Grande Ronde, Clearwater, Imnaha, and Salmon Rivers and their tributaries have cut
the Plateau to depths of 610 to 1524 meters (2000 to 5000 ft). The canyons penetrate
south into Oregon and divide the Blue Mountains from the Rocky Mountains. The
dissected portions of the plateau vary from Palouse-like loess hills in the east and
south of the Snake River near Dayton or Pomeroy, Washington, to the sharp ridges
near the Blue Mountains. In northeastern Oregon, the Grande Ronde River canyon
cuts laterally across the area that is recognized as a major source of the lava flows
that originally formed the Columbia Plateau (Hooper and Reidel 1989).
Rainfall amounts in the subregion vary from 25.4 to 58.4 centimeters per year (10 to
23 in per yr) depending upon elevation. Native vegetation is diverse due to the
proximity of the Blue Mountains and Rocky Mountains and the wide range in
precipitation. Pure grasslands without a sagebrush component cover the subregion in
its driest portions, the 30.5 to 40.6 centimeters per year (12 to 16 in per yr)
precipitation zone. As the Plateau rises toward the Blue Mountains and the northern
Rockies, a shrub component, either rose or snowberry, is added to the association.
-------
Clarke and Bryce--39
Grazing and farming has eliminated much of the original plant cover. The sharptailed
grouse, once abundant in the fescue-snowberry zone, had disappeared completely
(Daubenmire 1970); but it recently has been reintroduced.
Yakima Folds (fig. 2, 10g; fig. 3, 10g)—The Yakima Fold Belt is a series of anticlinal
ridges and synclinal valleys that covers 14,000 square kilometers (8700 sq mi) of the
western Columbia Plateau (Reidel and Campbell 1989). The folds include the
Columbia Hills, the Rattlesnake Hills, Frenchmen Hills, Saddle Mountains, Umtanum,
Yakima, and Manastash Ridges. The ridges are composed of layer upon layer of
basalt. Oil drilling in the Rattlesnake Hills reached depths of 3648 meters (12,000 ft)
without clearing basalt (SCS 1971). The slow deformation of the central Plateau under
tectonic influences as well as the immense weight of the flood basalts caused a
warping of the western end (Allen and others 1986, Baker and others 1991). The uplift
was slow enough to allow the Yakima River to retain its meandering course through
the folds.
The Yakima Folds lie in the rain shadow of the Cascade Range. Precipitation amounts
vary from 15.3 centimeters at 30.5-meter (6 in at 500-ft) elevation to 38.1 centimeters
at 1067.5 meters (15 in at 3500 ft) (SCS 1971). The ridges have steep north-facing
slopes. Soil on the north side of the ridges is thin and derived from basalt. Depth to
basalt ranges from 30.5 to 50.8 centimeters (12 to 20 in). The gentle south-facing
slopes are blanketed with loess to depths of 50.8 to 152.4 centimeters (20 to 60 in).
Land-use capability follows this pattern with dryland wheat farming on the south side
-------
Clarke and Bryce--40
and grazing on the north side of Rattlesnake and Horse Heaven Hills.
The plant cover of the driest portion of the subregion is the big sage/bluebunch
wheatgrass association. It reaches elevations of 820 meters (2690 ft) on the summits
of the eastern Yakima Folds. Other associations include fescue/hawkweed in the
Columbia Hills near Goldendale and a bitterbrush/fescue community in the east
Cascade foothills just below timberline north of Ellensburg and at the eastern end of
the Columbia Gorge. Elsewhere the big sage and wheatgrass association occurs right
to the forest edge. Daubenmire theorized that the sage/pine ecotone was determined
by the change in mean annual precipitation: never more than 36 centimeters (14.5 in)
in the sagebrush zone and never less than 40.8 centimeters (16.5 in) in the pine forest
(Daubenmire 1956).
The synclinal valleys associated with the Folds, such as the Yakima, Ahtanum, and
Kittitas valley, are discussed separately under the Pleistocene Lake Basins subregion.
Palouse Hills (fig. 3, 10h) and Nez Perce Prairie (fig. 3, 10j)—The elliptical Palouse Hills
subregion (10h) ranges from Spokane in the north to the edge of the Snake River
canyon in the south. The eastern boundary is determined by the transition to the
granites of the Rocky Mountains, increased precipitation (up to 55 cm), and the
transitional forest soils. The change from the mesic Palouse soil to the drier Walla
Walla soil, as interpreted from county soil surveys, marks the western boundary. The
loess-covered area southeast of Moscow, Idaho, can be considered the easternmost
-------
Clarke and Bryce--41
extension of the Palouse Hills. However, because this area is cut by deep canyons,
tributaries to the Snake and Clearwater Rivers, it was included in the Canyons and
Dissected Uplands subregion.
Rainfall in the Palouse Hills is in the range of 40.6 to 58.4 centimeters per year (16 to
23 in per yr). Because of the mesic conditions, the soils of the Palouse are classified
as Mollisols rather than Aridosols. They have a higher organic matter and clay content
than other loessial soils on the Plateau. In spite of the increased precipitation, most of
the runoff occurs in January and February, meaning that small, loess-bottomed
streams rising within the subregion are intermittent. The only perennial streams
originate from the Rocky Mountains to the east.
It has been suggested that the shape of the Palouse hills results from a combination of
strong wind action from the southwest that formed sharp dunelike crests toward the
northeast and nivation cirques or amphitheaters caused by the accumulation of
snowdrifts on the north and east sides (SCS 1980). There is also evidence of the
eastward migration of the hill summits during drier times presumably caused by wind
action (Baker and others 1991).
The mesic fescue/snowberry plant association marking the transition through the
Rocky Mountain foothills has been almost entirely supplanted by wheat farms. Bits of
original vegetation survive in the "eyebrows" left where the land is too steep to plow.
The Palouse hills are highly productive, producing up to 100 bushels of wheat/acre
-------
Clarke and Bryce--42
(Sterner 1987).
The Nez Perce Prairie (10j) is a rolling loess plain in western Idaho, southeast of the
Patouse Hills. Though generally similar to the Palouse region, it differs in having
higher elevations overall, colder temperatures, and higher precipitation levels. As a
result, the loess soils have a higher clay development. Also, because the Nez Perce
Prairie was out of the main path of aerial loess deposition, the soil is shallower than in
the Palouse Hills. The Prairie is bounded on all sides by deep river canyons. The
area east of Joseph and Enterprise in northeast Oregon is sometimes included with
the Nez Perce Prairie (Raisz 1965, SCS 1986, Steiner 1987). However, because of its
even thinner loess cover, lower precipitation, and proximity to the Snake River canyon,
it has been included with the Canyons and Dissected Uplands subregion.
Deep Loess Foothills (fig. 2, 10i; fig. 3, 10i)-These foothills of the Blue Mountains
follow an arc on the northwest slopes beginning just north of Pendleton. The
subregion could be a disjunct portion of the Palouse Hills, as it has Palouse soil, the
same range in precipitation (40.6 to 58.4 centimeters per year [16 to 23 in per yr]),
and similar land-use capability with annual cropping possible. The differences lie in
the topography, physiography, and hydrography. The dunelike ridges of the Palouse
Hills are replaced by terraced ridges rising to the forested Blue Mountains. The
sloping basalt ridges are bisected by perennial streams fed by the higher rainfall and
snowpack of the Blue Mountains.
-------
Clarke and Bryce~43
Water Resource Issues of the Columbia Plateau Ecoregion
The initial focus of ecoregion development within EPA was to use ecoregions as a
framework to classify streams and to assess water quality. The rationale behind the
use of terrestrial landscape characteristics to create water-quality regions was that
streams are a reflection of the watersheds which they drain. In moist climatic areas,
information on drainage area size, stream discharge, and typical stream substrate is
incorporated with the terrestrial landscape data layers to test consistency of ecoregion
boundaries with stream types. However, for the semiarid Columbia Plateau, objectives
for water-quality assessment did not influence the boundary decisions as much as
they would in wetter areas. Traditional use of an ecoregion framework for water
quality monitoring and assessment in moist environments carries with it certain
assumptions:
1. Perennial streams are used for sampling and setting biocriteria standards.
2. A number of perennial streams within a region ensures a gradient in stream quality.
3. Reference streams are chosen that have their entire length within a particular
subregion and are considered representative of that subregion.
4. Small, wadeable streams are used for ease of sampling and to assure that their
drainage area lies entirely within a particular ecoregion or subregion.
-------
Clarke and Bryce--44
These assumptions do not apply to a region as arid as the Columbia Plateau.
In the arid and semiarid areas of the Columbia Plateau, most summer precipitation
evaporates or is transpired, leaving little water for streamflow. In the driest parts of the
Plateau, flow is zero or nearly zero. Almost the entire Plateau, with the exception of
the mesic fringes, that is, "meadow steppe," (Daubenmire 1970), has an average
yearly runoff of less than 2.5 centimeters per year (1 in per yr). At Esquatzel Coulee,
near Connell, Washington, average yearly runoff is estimated at .25 centimeter per
year (.10 in per yr) (Nelson 1991). Even in the Palouse Hills subregion, which has the
highest precipitation on the Plateau (40.6 to 58.4 centimeters [16 to 23 in]), of all the
precipitation that falls, 84 percent is lost to evapotranspiration, 15 percent runs off as
surface water, and 1 percent is left and assumed to percolate into the ground (Steiner
1987). Ephemeral streams flow only several days a year or fail to flow at all in
particularly dry years. Perennial streams often originate in higher elevations where the
source includes snowmelt as well as rain. These "exotic" streams reflect mountain
conditions; they are not representative of the lower elevation subregions of the
Columbia Plateau through which they flow.
Streams have disappeared as a result of cultivation practices in steep loess hills.
Small stream channels may be plowed through and removed from the landscape.
Other channels simply fill with soil as it erodes off the hillsides. In the intensively
farmed areas, it is impossible to know the original substrate or riparian cover of
streams or how many streams once ran perennially. Some larger streams of the arid
-------
Clarke and Bryce~45
plateau that do flow are totally appropriated during the irrigation season: water is
withdrawn until the streams are dry. Minimum flow regulations that have been
established to provide fish and wildlife habitat may be secondary to withdrawal rights
and thus are not implemented (Drost and others 1990). In other irrigated areas, the
reverse situation may occur; streams that would naturally not be perennial, run with
irrigation return water
If the Washington Department of Ecology and the Oregon Department of
Environmental Quality are to use the ecoregion framework in the Columbia Plateau, as
they have elsewhere, to classify streams, monitor water quality, and eventually set
biocriteria standards, adjustments will have to be made to adapt bioassessment to the
arid conditions. Possible options are to:
1. Sample only larger systems.
2. Determine the role of springs in the monitoring program.
3. Use reference stream reaches as opposed to reference watersheds.
4. Choose reference reaches of streams within the Columbia Plateau, but with
headwaters in neighboring mountainous ecoregions.
5. Model historic conditions.
-------
Clarke and Bryce~46
6. Use reference sites from similar ecoregions.
Additional field work and interdisciplinary discussion are necessary to reexamine water
quality monitoring and assessment in this arid region.
-------
Clarke and Bryce~47
Literature Cited
Allen, J.E.; Burns, M.; Sargent, S.C. 1986. Cataclysms on the Columbia: a layman's
guide to the features produced by the catastrophic Bretz floods in the Pacific
Northwest. Portland, OR: Timber Press. 211 p.
Anderson, E.W. 1956. Some soil-plant relationships in eastern Oregon. Journal of
Range Management. 9(4): 171-175.
Baker, V.R.; Bjornstad, B.N.; Busacca, A.J. [and others]. 1991. Quaternary geology of
the Columbia Plateau. In: Morrison, R.B., ed. The Geology of North America, vol K-2,
Quaternary nonglacial geology: conterminous U.S., Boulder, CO: The Geological
Society of America.
Bauer, H.H.; Vaccaro, J.J. 1990. Estimates of ground-water recharge to the Columbia
Plateau regional aquifer system, Washington, Oregon and Idaho for predevelopment
and current land-use conditions. Water Resources Investigations Rep. 88-4108.
Tacoma, WA: U.S. Department of the Interior, Geological Survey. 37 p. and two map
sheets.
Clarke, S.E.; White, D.; Schaedel, A.L. 1991. Oregon, USA, ecological regions and
subregions for water quality management. Environmental Management. 15(6):
847-856.
-------
Clarke and Bryce--48
Daubenmire, R. 1956. Climate as a determinant of vegetation distribution in eastern
Washington and northern Idaho. Ecological Monographs. 26(2): 131-154.
Daubenmire, R. 1970. Steppe vegetation of Washington. Tech. Bull. 62. [N.p.]:
Washington Agricultural Experiment Station. 131 p.
Drost, B.W.; Whiteman, K.J.; Gonthier, J.B. 1990. Geologic framework of the Columbia
Plateau aquifer system, Washington, Oregon and Idaho. Water Resources
Investigations Rep. 87-4238. [N.p.]: U.S. Department of the Interior, Geological Survey.
10 p. and 10 map sheets.
Franklin, J.F.; Dyrness, C.T. 1974. Natural vegetation of Oregon and Washington.
Corvallis, OR: Oregon State University Press. 452 p.
GS (U.S. Department of the Interior, Geological Survey). 1982. The channeled
scablands of eastern Washington: the geologic story of the Spokane flood. GPO No.
359-019. Washington, DC: U.S. Government Printing Office. 24 p.
Hooper, P.R.; Reidel, S.P. 1989. Dikes and vents feeding the Columbia River Basalts.
In: Joseph, N.L. [and others], eds. Geologic guidebook for Washington and adjacent
areas. Information Circular 86. Olympia, WA: Washington Division of Geology and
Earth Resources. 369 p.
-------
Clarke and Bryce-49
Nelson, L.M. 1991. Surface-water resources of the Columbia Plateau in parts of
Washington, Oregon and Idaho. Water Resources Investigations Rep. 88-4105. [N.p.]:
U.S. Department of the Interior, Geological Survey. 4 map sheets with text.
Raisz, Erwin. 1941; revised 1965. Landforms of the northwestern states. Map. Scale
approx. 1" = 20 miles. N.p.: N.p.
Reidel, S.P.; Campbell, N P. 1989. Structure of the Yakima Fold Belt, central
Washington. In: Joseph, N.L. [and others]. Geologic guidebook for Washington and
adjacent areas. Information Circular 86. Olympia, WA: Washington Division of Geology
and Earth Resources. 369 p.
Schuster, J.E. 1995. Draft geologic map of Washington-Southeast Quadrant. Scale
1:250,000. In preparation.
SCS (U.S. Department of Agriculture, Soil Conservation Service). 1971. Soil survey of
Benton County area, Washington. In cooperation with Washington Agricultural
Experiment Station. N.p.: N.p.
SCS (U.S. Department of Agriculture, Soil Conservation Service). 1980. Soil survey of
Whitman County, Washington. In cooperation with Washington State University
Agricultural Research Center. N.p.: N.p.
-------
Clarke and Bryce--50
SCS (U.S. Department of Agriculture, Soil Conservation Service). 1981a. Soil Survey of
Douglas County, Washington. In cooperation with Washington State University
Agricultural Research Center. N.p.: N.p.
SCS (U.S. Department of Agriculture, Soil Conservation Service). 1981b. Soil Survey of
Lincoln County, Washington. In cooperation with Washington State University
Agricultural Research Center. N.p.: N.p.
SCS (U.S. Department of Agriculture, Soil Conservation Service). 1986. Major Land
Resource Areas. Map of Oregon. Scale 1:3,500,000. N.p.: N.p.
SCS (U.S. Department of Agriculture, Soil Conservation Service). 1988. Soil survey of
Umatilla County Area, Oregon. In cooperation with U.S. Department of the Interior,
Bureau of Indian Affairs, and Oregon Agricultural Experiment Station. N.p.: N.p.
Steiner, F. 1987. The productive and erosive Palouse environment. Washington State
University Cooperative Extension Bulletin. Pullman, WA: Washington State University.
42 p.
Stoffel, K.L.; Joseph, N.L.; Waggoner, S.Z. [and others]. 1991. Geologic map of
Washington-Northeast Quadrant. GM-39. N.p.: Washington Division of Geology and
Earth Resources. Map 1:250,000 scale and text 36 p.
-------
Clarke and Bryce--51
Walker, G.W.; MacLeod, N.S. 1991. Geologic map of Oregon. 1:500,000 scale.
Denver, CO: U.S. Department of the Interior, Geological Survey. 2 sheets.
-------
Clarke and Bryce--52
SECTION 2-LEVEL IV ECOREGIONS OF THE BLUE MOUNTAIN ECOREGION OF
OREGON, WASHINGTON, AND IDAHO
by Sandra A. Bryce and James M. Omernik
The Blue Mountains of north central Oregon form a transverse bridge between the
Northern Rocky Mountains of western Idaho and the Oregon Cascades. The overall
structure of the Blue Mountains is a large anticline with a steep north flank in
southeastern Washington State, high elevations in the midsection, and gentle south
slopes that join the high lava plains of central Oregon (McKee 1972). The wide arc of
mountains encompasses three different geologies—a continuation of the Columbia
Plateau basalts in the north; a collection of "accreted terranes," that is, ocean crust
and volcanic islands that collided with the western end of the continent, in the central
part; and more recent rhyolites, andesites, and basalts along the southern perimeter.
Individual ranges include the Ochoco, Aldrich, Strawberry, Greenhorn, Elkhorn,
Wallowa, and Seven Devils Mountains. Elevations range from less than 275 meters
(900 ft) in the canyon depths to nearly 3050 meters (10,000 ft) on the Matterhorn, the
highest peak in the Wallowa Mountains.
The ecoregion map presented here (fig. 3) and the report that accompanies it
organize the complexity of this region. The explanations below take the reader first
through the component landscape characteristics. Then we rearrange and integrate
this information in the individual regional descriptions to illustrate the rationale for the
-------
Clarke and Bryce-53
boundary decisions. As one might expect, the available information for any region is
not evenly distributed. The Blue Mountain area is of great interest to geologists; the
available literature is heavily weighted in that discipline. However, we hope the
descriptions will illuminate the distinctions between the regions and serve as a
reference to those who may use the ecoregion map for any of the applications
discussed in the introductory chapter or others not yet invented.
Climate
Water limits ecosystems in the Blue Mountains. It became a limiting factor with the
rise of the Cascade Range 20 to 30 million years ago, which formed a barrier to the
eastward flow of weather systems off the Pacific Ocean. As a result, the semitropical
climate of the area became temperate and continental (Alt and Hyndman 1978).
Precipitation is light, and most of it occurs in the winter months; thunderstorms do
build occasionally over the higher ranges during the hot, dry summer. In winter, the
continental temperature extremes are somewhat reduced because the region is
protected from the full brunt of arctic air flow by the northern Rocky Mountains. In the
summer, there is enough marine influence over the northern and central Blue
Mountains to moderate extremes in heat and drought.
The degree of continentality varies across the Blue Mountains from west to east as
well as latitudinally. The Cascade rainshadow extends 233 kilometers (140 mi) to the
east to the base of the Elkhorn range. In the rainshadow. light precipitation, high
-------
Clarke and Bryce-54
evapotranspiration, and wide temperature fluctuations result in wide expanses of dry
forest dominated by ponderosa pine. The xeric conditions and temperature extremes
are even greater in the south central Blue Mountains and along the southeastern
foothills where the marine influence is entirely blocked.
The continental climate in the northeastern Blue Mountains is masked by a swath of
marine air that penetrates eastern Oregon and Washington from the Pacific Ocean.
Weather systems flowing through the Columbia River gorge, penetrating the Cascade
Range, bring precipitation to the northern Blue Mountain slopes three seasons of the
year. Here, increased cloudiness, higher humidity and precipitation foster a productive
mesic forest. This wetter, maritime-influenced section is bounded by the Wallowa-
Seven Devils Mountains to the east and the Elkhorn range to the south.
The planes of influence delimiting ecosystems in the Blue Mountains are elevational as
well as latitudinal. In general, as moisture increases and temperature decreases with
elevation, forest productivity improves until heavy snowpack and severe winters
shorten growing seasons and limit tree growth at elevations above 1830 meters (6000
ft).
The capabilities of these ecoregions for particular land uses also are determined by
moisture availability, growing season, and temperature. Grazing, mining, logging,
farming, road building, and irrigation have been practiced in the Blue Mountains for
150 years. Many other authors have traced the history of land-use and management
-------
Clarke and Bryce-55
practices that have contributed to the present-day condition of the Blue Mountains
ecosystems (Harvey and others 1994, Hessburg and others 1994, Irwin and others
1994, Johnson 1994, Johnson and others 1994, Lehmkuhl and others 1994, Mcintosh
and others 1994, Oliver and others 1994, Robbins and Wolf 1994, Skovlin 1991).
Major land uses and their effects on particular landscape types are discussed in the
descriptions of the individual ecoregions.
Geology—The Birth of the Blues
In Devonian to early Jurassic time (275 to 200 myr ago), the Blue Mountains began as
a group of offshore islands in the eastern Pacific Ocean. They had formed as island
arcs behind subduction zones in the ocean floor. At a subduction zone, an oceanic
plate (a piece of earth's crust) slides beneath another plate (oceanic or continental).
Intense heat and pressure melt the rocks on the sea floor into magma, which then
resurfaces as volcanic eruptions, forming a continental mountain range or a line of
volcanic islands. Thick ocean sediments and even an occasional piece of ocean crust
may be scraped off and piled against the side of the subduction trench.
At about the same time (geologically speaking), 200 million years ago, the North
American continent separated from Europe, carried westward by the widening rift in
the floor of the Atlantic Ocean. Oregon's land mass did not exist; the Pacific shore of
the continent was then in western Idaho. As the North American continent moved
slowly, but inexorably westward, it engulfed a group of offshore islands about 100
-------
Clarke and Bryce--56
million years ago (White and others 1992). Two island groups have been mapped in
the Blue Mountains: the Wallowa-Seven Devils terrane and the Olds Ferry terrane
near Huntington, Oregon. Remains of limestone coral reefs in the Wallowa Mountains
indicate that the islands formed under subtropical conditions in a warm shallow sea
(Fifarek and others 1994). The offshore sediments and the erosional material from the
islands (shales and fine sandstones) settled between the islands and the mainland as
well as between the islands and the subduction trench. These sediments are evident
in the Aldrich Mountains (the Izee terrane) and in the Wallowa Mountains (White and
others 1992). The island arcs may be remnants of a more extensive system; their
rocks are chemically similar to those in arcs accreted to western Canada as well as
the Aleutian island chain (Brooks 1979, White and Vallier 1994). Here in the Blue
Mountains, the accreted land meets the ancient rocks of the North American plate at
the "suture zone," a north-south trench along the western Idaho border, extending
from Cascade Reservoir north to Riggins, Idaho (Hooper and Swanson 1990).
The chaotic result of subduction zone geology is represented by the Baker and
Grindstone terranes, the formations that form the core of the central Blue Mountains.
They are a patchwork of ocean sediments, sandstone and argillite (shaley mudstone),
and deep seacrust rocks such as pillow lavas, serpentinite, and gabbro, that escaped
the earth's molten maw. Mixed in with these formations are the metasedimentary and
metavolcanic rocks metamorphosed in the superheated contact zones near the
subduction trench (Brooks 1979, Walker and Robinson 1990, White and others 1992).
The complex of sediments, lavas, and ocean crust making up the accreted terranes is
-------
Clarke and Bryce--57
aptly called a melange. Melange formations worldwide signal the presence of
subduction zones and newly accreted land.
The process of accretion also created reservoirs of magma that recrystallized
underground 100 to 130 million years ago to form the granitic core of the Wallowa
Mountains, the northern Elkhorn Mountains, and other scattered peaks. The
metamorphic areas surrounding these intrusions are the source of gold and other
metal deposits that attracted early settlers to the Blue Mountains (Goldstrand 1994,
Thayer 1990).
This addition of new land occurred during pre-Cenozoic time. Then about 60 million
years ago, during the Tertiary period of the Cenozoic era, intense volcanic activity
changed the character of the region yet again. Over a period of 40 million years, the
accreted terranes were buried by volcanic material from the chain of volcanoes of the
Clarno period, the immense ash falls of the John Day formation, the Strawberry
Mountains volcanics, and the Columbia River and Picture Gorge flood basalts.
Erosion has since exposed the older pre-Cenozoic rock at higher elevations (Orr and
others 1992, Thayer 1990).
The Clarno formation is the remnant of a range of volcanoes, formed during the
Eocene period 55 to 35 million years ago, that followed the northeast-southwest axis
of the Blue Mountains. The rocks are mostly andesitic, but they include some basaltic
and rhyolitic rocks. The volcanic cones mostly have eroded away; scattered remnants
-------
Clarke and Bryce-58
may be seen as conical plugs (i.e., White Butte and Black Butte in Wheeler County,
Oregon) (Walker and Robinson 1990).
By the time of the John Day ash falls, approximately 35 million years ago, the sea-floor
subduction zone had changed to a north-south line rather than the northeast trending
Wallowa-Klamath Mountains line (Walker and Robinson 1990). The new location set
the stage for the formation of the Cascade range that would eventually block the
marine air flow from the interior and change the warm, moist climate of Clarno times to
the continental climate of today (Alt and Hyndman 1978).
Between about 17.5 and 6 million years ago, incredible floods of fluid basalt filled what
was once an inland sea in Central Oregon and Washington. This Columbia River
Basalt group constitutes the entire Columbia Plateau and the northern third of the Blue
Mountains. Most of the flows (90 percent) issued from long fissure vents and dike
swarms in the northeastern Blue Mountains, except for the Picture Gorge basalt that
erupted from vents in the John Day River basin in the central Blue Mountains (Hooper
and Swanson 1990).
The northwestward tilting of the Blue Mountains continued throughout this period,
partly due to the rise of the subterranean granitic bodies (the Idaho batholith to the
east and the smaller granitic bodies under the Wallowas, Elkhorns, and the Nez Perce
Plateau) and partly because of regional stresses, faulting, and compression. Faulting
defined the major valleys (Grande Ronde, Baker, and Wallowa), and uplifted the
-------
Clarke and Bryce-59
adjacent mountain ranges (Wallowas, Elkhorns, and Strawberrys). Since these
dramatic geologic happenings, eons of erosion and Pleistocene glaciation have
sculpted the peaks and filled the basins with sediments, producing the familiar,
seemingly stable, landscape of today.
Blue Mountain Soils
One could imagine that such geological mayhem in the parent material would produce
an incredible variety of soil types across the Blue Mountains. And so it does.
However, there are zones of moisture, temperature, elevation, and aspect that create a
discernible pattern of general soil types across the landscape.
In this moisture limited region, much of the lower elevation soils are xeric or aridic; that
is, they are dry for at least 60 to 90 days in the summer months. Early moisture
stress effectively shortens the active growing season. In addition, much of the lower
elevation, drier soil is derived directly from the geologic parent material. This condition
effectively increases the droughtiness of the soil because it is more likely to have rock
fragments throughout that reduce the amount of soil available for rooting as well as
the water storage capability (Andersen 1956). The coarse fragments do make soil
derived from basalt and andesite slightly less erodible than fine-textured soil. In xeric
soil regimes, the lack of water limits the development of soil organic content and slows
soil weathering. Productivity is dependent upon a thin surficial organic layer or upon
regular fires to release nutrients accumulated in woody debris and dried grasses
-------
Clarke and Bryce--60
(Harvey and others 1994). Thus, when moisture is limiting, a number of factors
contribute to a less productive medium for plant growth
The geologic parent material for soils are the rhyolites and tuffs in the southern Blue
Mountains, mixed volcanic debris and ash in the John Day/Clarno region, sedimentary
material in the Aldrich and Elkhorn Mountains, and andesite and basalts in the
Strawberry Mountains. Mixed with these are the droughty soils of the granitic and
metamorphic inclusions Xeric conditions and characteristic plant cover persist into
higher elevations on south-facing slopes and on metamorphic parent material.
At higher elevations, above about 762 meters (2500 ft) in the north and 1525 meters
(5000 ft) in the south, conditions tend to become cooler and moister and so also more
productive. Soil at these elevations is normally udic or moist. In the udic moisture
regime, there are fewer than 45 days of dryness in the 120 days following June 20.
The temperature regime is either frigid or, in the higher elevations above 1525 meters
(5000 ft), cryic.
The most productive soils of the Blue Mountains are those composed of fine materials
that have been aerially deposited. Ash deposits from the eruptions of Glacier Peak
and Mount Mazama (12,000 and 6,600 yr ago, respectively) are as much as 1-meter
thick in the mid- to high-elevation forested areas. Much of the ash has eroded away in
the lower elevation grasslands and on south-facing slopes, but it has been preserved
in more mesic zones, such as north slopes and sites having had a continuous forest
-------
Clarke and Bryce--61
cover. The fine silt loam ash soils have a high water holding capacity, a high water
infiltration rate, and a largely rock free growing medium. Again, most of the nutrients
are near the surface. Without a vegetative cover, ash soils are very vulnerable to
erosion and subsequent loss of productivity (Harvey and others 1994). In general, the
major areas of ashy soil correspond with the regions of Mesic Forest (fig. 4, 111),
especially in the central and northern Blue Mountains.
Aeolian loess (glacial silt deposited by wind during the Pleistocene) is also a significant
addition to Blue Mountain soils, particularly in the northern Blue Mountains adjacent to
the Columbia Plateau (see section 1 above). Like ash, loess deposits can ameliorate
the droughtiness of the soil. Loess soils also are fine-textured, rock free, and high in
nutrients.
Blue Mountain Vegetation
As we walk the landscape and view the patterns of vegetation at that scale, we notice
the variability in plant associations, responding to variations in disturbance,
microclimate, aspect, and local topography. Various species (table 2) dominate over
time as conditions change. How can we generalize about such diversity?
We can compare this conundrum with the phenomenon that happens upon inspection
of a pointillist painting, e.g., the Isle du Gran Jatte by Seurat. From the vantage point
of a few inches away from the canvas, we see only a dizzying array of many-colored
-------
Clarke and Bryce--62
dots. But if we step back away from the painting, the dots fuse to reveal pattern.
Similarly, if we conceptually back away far enough from the landscape, the confusing
collection of plant associations in the Blue Mountains do disclose a pattern, dependent
upon temperature, elevation, growing season, soil type, and depth.
The most apparent pattern in the area is a progression of vegetation types with
increasing elevation, beginning with grasslands at the lowest elevations and changing
to shrublands, ponderosa pine woodlands, true fir mesic forest, subalpine forest and
parkland, and alpine meadows as elevation increases. Tying these patterns to a map
is a more difficult prospect. However, guided by elevation, parent material, soil type,
fire regime, and historic vegetation patterns, we have created a model that includes
vegetation as one element in the integration of landscape characteristics that compose
a map of ecosystems.
In the lowest elevations, grassland communities prevail in the areas of xeric and aridic
soils. They range from bluebunch wheatgrass-Sandberg's bluegrass associations in
the warmest, driest canyons to Idaho fescue associations in deeper, moister soils.
Shrublands and juniper/grassland savannah form the transition from grassland to
forested slopes. Along the southern foothills, bitterbrush, mountain-mahogany, and
sagebrush mix with juniper to provide cover, browse, and berries for birds, elk, and
deer (Johnson and Clausnitzer 1992). In the north-central and northern Blue
Mountains, sagebrush and western juniper are replaced by a mesic shrub
assemblage—ninebark, common snowberry, serviceberry, Rocky Mountain maple, and
-------
Clarke and Bryce--63
oceanspray.
Ponderosa pine grows in a very wide range across much of the Blue Mountains as a
climax species in the warmest and driest forested areas and as a serai dominant in
areas of frequent fire (Harvey and others 1994, Hessburg and others 1994). As a
serai dominant, it grows at the moister end of the xeric soil regime where grand fir is
technically the climax species. However, here the grand fir is at the extreme end of its
range. It is susceptible to fire, moisture-stressed, and prone to disease and insect
infestations (Hessburg and others 1994, Johnson and others 1994). Under a normal
fire regime, with a fire frequency as short as 10 years, the ponderosa pine persists
while the grand fir seedlings continually succumb to fire except near springs or other
moist refugia.
Historically, the forest types were rather sharply bounded by elevation. Ponderosa
pine dominated the xeric soil regions, the fire-prone areas, while true fir associations,
the mesic forest, dominated higher elevations. The mesic forest zone historically
began at about 762 meters (2500 ft) in the north and 1500 meters (4900 ft) in the
southern mountains.
Before widespread fire suppression, these mesic forests had a low fire frequency of 50
to 300 years or more; but the fires were hot, intense, and extensive. Lodgepole pine
and western larch, the serai species at the higher elevations of 1220 to 2285 meters
(4000 to 7500 ft), revegetated the burns to create a patchy forest landscape. Insect
-------
Clarke and Bryce-64
and disease infestations cycled within the limited area of susceptible true fir forest, but
infestations did not often reach epidemic proportions (Hessburg and others 1994).
Fire suppression has blurred the historical-elevational zonation of forest vegetation.
Present day vegetation patterns do not give us a clear picture of the natural affinities of
forest plant associations. Because of 60 years of fire suppression, Douglas-fir, grand
fir, and Engelmann spruce have expanded their range to lower elevations beyond their
normal mesic locations. During this same period, old-growth stands of ponderosa
pine and western larch have been cut, leaving thickets of late successional fir forest.
With the increase in acreage of overstocked mesic forest, insects and disease have
become more common. Over the years, the proportion of forest land in the Blue
Mountains dominated by ponderosa pine has declined on all ownerships from 80
percent in 1936 to 25 percent in 1992 (Hessburg and others 1994).
In the new atmosphere of ecosystem management, proposed management strategies
include uneven-aged management of multiple tree species, reduction of grazing
densities, restoration of understory shrub communities (Irwin and others 1994), and
prescribed burning to emulate the historic fire regimes and to reestablish the serai
forest (Oliver and others 1994). A map like the ecoregion map presented here (fig. 3),
developed as a model of a time when there were minimal human impacts on the
landscape, can serve to guide restoration efforts by suggesting a more "natural"
distribution of vegetation. With that goal in mind, the ecoregions delineated here
reflect the time when ponderosa pine in park-like stands dominated both the lower
-------
Clarke and Bryce--65
elevation sites and the xeric grand fir and Douglas-fir series sites as well.
There is some discussion about exactly how much of the presettlement vegetation
pattern was due to Native American "prescribed burning" (Robbins and Wolf 1994).
Could the park-like stands of ponderosa pine have been "managed" by early Native
Americans? Perhaps early man and the forest community have coevolved since the
Pleistocene. There is evidence to indicate that the forest has evolved under a fire
regime, whether human-caused or not, and that it maintained itself in a healthier
condition than today (Harvey and others 1994, Hessburg and others 1994, Johnson
1994, Lehmkuhl and others 1994, Robbins and Wolf 1994).
The historic view of an area may provide guidance to restoration efforts by serving as
a model or ideal goal. Though we may never reach the ultimate goal (the
presettlement condition), what is attainable will be somewhere between the degraded
condition and the historic condition. How closely we approach the goal depends upon
social values and the availability of funding.
In summary, the climate, geology, soil, historic and present day vegetation, and
important influences on these factors (e.g., the fire regime) were the major categories
of information that contributed to the regional delineations for this project. Map
analysis, consultation with regional experts in meetings and over the phone, and
literature review were all integrated to create the map of Level IV ecoregions of the
Blue Mountain ecoregion of Oregon and Washington. The following regional
-------
Clarke and Bryce~66
descriptions indicate the rationale behind specific boundary decisions.
Descriptions of the Blue Mountains Subregions
Composite subregions—To limit the number of subregions and to strive for
cartographic simplicity, a number of disjunct subregions that are the result of
elevational factors have been combined into two composite subregions, the Mesic
Forest (fig. 3, 111) and the Subalpine Areas (fig. 3, 11m). Thus, there is just one
subalpine subregion instead of three (Wallowas subalpine, Elkhorns subalpine, and
Strawberry Mountains subalpine), graphically displayed as disjunct areas of the same
color. The scattered portions of each composite subregion have a general similarity
across the Blue Mountains ecoregion. However, there is some latitudinal variability
among the disjunct areas. The Mesic Forest and Subalpine Subregions are described
in general below, and each individual area (e.g., Wallowas subalpine) is then described
in more detail in its appropriate subregional description.
Subalpine Areas (fig. 2,11m; fig. 3,11m)—The widest expanses of Subalpine Areas
(fig. 3, 11m) appear in the high Wallowa, Elkhorn, Greenhorn, and Strawberry
Mountains. Elevations at which the subalpine forest becomes broken by high
meadows and parkland vary from north to south. Subalpine parkland, i.e., the mosaic
of forest patches and meadows, begins at approximately 1830 meters (6000 ft) in the
northern Blue Mountains and at 2130 meters (7000 ft) in the south central Blue
Mountains. The high elevation forests and meadows share a cold climate, cold, often
-------
Clarke and Bryce~67
shallow, soils, short growing season (10 to 70 days), heavy snowpack, and high
annual precipitation (85 to 150 cm).
The subalpine lithology varies from Columbia River basalt in the northern Blue
Mountains to granitics in the Wallowas, granodiorites in the Elkhorns and Greenhorns,
and ultramafics and andesites in the Strawberry Mountains. Subalpine fir, and its serai
partners, lodgepole pine, western larch, and Engelmann spruce, reach their highest
potential on silt loam soils derived from volcanic ash and loamy sands from pumice
and ash. Lodgepole pine prefers ashy soil on gentle ridges with slopes less than 15
percent. Western larch can grow in steeper areas, where deeper colluvial soils
provide more stored moisture (Johnson and Simon 1987). In more extreme locations
to 2440 meters (8000 ft), whitebark pine, also serai to subalpine fir at lower elevations,
becomes co-dominant.
Mesic Forest Zone (fig. 2, 111; fig. 3,111)—The Mesic Forest Zone (fig. 3, 111)
appears throughout the Blue Mountains in areas of low moisture stress. The largest
expanses are in the north where marine air flowing through the Columbia River Gorge
brings high rainfall amounts. Elsewhere, in the central and southern Blue Mountains,
elevation increasingly influences the distribution of mesic forest. As the climate
becomes more continental, only higher elevations capture the amount of precipitation
necessary to support the mixed conifer and fir forests. Mesic forests have a high
affinity for ash-derived soils with high moisture-holding capacity, which retard the onset
of summer drought in mountainous areas with cooler temperatures. The mesic forest
-------
Clarke and Bryce-68
starts at about 915 meters (3000 ft) in the north; farther south and west, the true fir
forest is limited to areas above 1830 meters (6000 ft) or to northern aspects.
The subregion map (fig. 3) depicts the Mesic Forest Zone as it might be under a more
"natural" fire regime. The early map by Andrews and Cowlin (1936) for the forest
types of the U.S. Department of Agriculture's Forest Service served as a model for the
delineations of mesic and xeric forest on the subregion map. This 1936 map provides
a snapshot of forest distributions before fire suppression and widespread high grade
logging had changed the sharp elevational patterns of the eastside forest. As a result,
the mesic forest delineated on the subregion map (fig. 3) is an expression of the
higher, moister end of the grand fir series. Areas where grand fir is technically the
climax species, but where it would not normally reach maturity because of frequent
I
fires, appear as xeric forest on the subregion map. Thus, the mapped zones of xeric
and mesic forest do not match the full range of the' true fir forest associations, but they
do correspond more closely to the xeric and udic soil temperature regimes.
Subregions in the Cascade Rainshadow-As the Cascade Mountains began to block
the flow of moist Pacific air to what is now central Oregon 20 to 30 million years ago,
the warm, tropical climate of the Eocene epoch gave way to a drier, cooler, continental
climate. The areas identified on the map as the John Day/Clarno Uplands (fig. 3, 11 a)
and the John Day/Clarno Highlands (fig. 3, 11b) constitute two elevational divisions of
subregions in the Cascades rainshadow.
-------
Clarke and Bryce--69
John Day/Clarno Uplands (fig. 2,11a; fig. 3,11a)—The John Day/Clarno Uplands
(fig. 3, 11a) subregion takes its name from two geologic formations that appear in
several subregions of the Blue Mountains; however, their greatest extent and most
distinctive expression occur within the boundaries of this subregion. The John
Day/Clarno Uplands are a ring of dry foothills that surrounds the western perimeter of
the Blue Mountains and separates the north-central Blue Mountains from the Southern
Blue and Ochoco Mountains. The "John Day Country" is a rough sea of highly
dissected hills, palisades, and colorful ash beds dotted with junipers and cut by the
valleys of the John Day and Crooked Rivers. Elevations range from 457 to 1525
meters (1500 to 5000 ft). Precipitation falls as spring and fall rains and light snow in
the winter. Annual precipitation varies with elevation, from 23 centimeters (9 in) at
lower elevations to 40 centimeters (16 in) at higher elevations.
The Clarno formation of 54 to 37 million years ago (Eocene and lower Oligocene
periods) began as a pre-Cascades chain of andesitic volcanic cones erupting along
the northeast trending axis of the Blue Mountains. The location of the Clarno
volcanoes indicates the past presence of a parallel subduction zone offshore (Walker
and Robinson 1990). The Clarno formation today comprises a wide variety of eroded
remnants of the mountain chain: andesites, basalts, tuffs, breccias, and colorful cliffs
of petrified mudflow (lahars). Red paleosols (fossil forest soils) and fossilized leaves
and nuts give evidence of the tropical climate of the time (Retallack 1991).
By the time of the John Day formation, 37 to 22 million years ago (Oligocene to early
-------
Clarke and Bryce--70
Miocene periods), the ocean floor subduction zone had shifted to the west and taken
up a north-south position west of the rising Cascade range. Several thousand feet of
airborne and waterborne tuffs and ash filled a topographic basin between the western
Cascades and the now-extinct Clarno volcanic chain. The ash flows and airfall tuffs
are assumed to have come from several sources east of the Cascades (Thayer 1990,
Robinson and others 1990). Robinson and others (1990) give us a startling realization
of the immensity of geologic time by relating that the lower unit of the three strata of
John Day ash (350-m deep) spans a period of 12 million years of deposition at a rate
of about 28 mm/1000 year. These formations of red, green, and grey ash are softer
and more easily eroded than the surrounding basalt, and thus, form valleys or bowl-
like depressions. The ashy beds weather to a fine clay. The fossilized remains of a
diverse mammalian fauna have been preserved in the John Day Fossil Beds National
Monument near Picture Gorge, Oregon (Thayer 1990).
Outside of the John Day ash beds, the soils are more varied, reflecting the variety of
volcanic parent materials in the region. The soils are generally deep to bedrock with a
very stony loam surface layer and a clay loam or clay subsoil. The soils at lower
elevations have an aridic soil moisture regime with average annual precipitation
amounts of less than 30 centimeters (12 in). They include Aridic Palexerolls, Chromic
Haploxererts, Xeric Paleargids, Aridic Haploxerolls, and Aridic and Uthic Argixerolls.
The soils at higher elevations have a xeric moisture regime with annual precipitation
amounts of 35 centimeters (about 14 in). These soils include Uthic Argixerolls, Calcic
Pachic Argixerolls, and Pachic Palexerolls. Soils derived directly from basalt differ in
-------
Clarke and Bryce--71
that they are shallow to moderately deep, with a surface layer of very cobbly loam and
a very cobbly clay-loam subsoil. Basaltic soils include Lithic Argixerolls and Aridic
Argixerolls.
The potential natural vegetation of the John Day/Clarno region consists of grassland
and grassland savannah. Bluebunch wheatgrass, Idaho fescue, and Sandberg's
bluegrass with big sagebrush and bitterbrush cover the lower elevation, drier areas.
Juniper woodland forms the transition zone between the dry grasslands and the
ponderosa pine forests of the higher elevations. The juniper zone has an elevation
range between 760 to 1400 meters (2483 to 4593 ft) where the rainfall is between 20
and 30 centimeters (8 to 12 in).
Juniper woodland and savannah have increased markedly over the last 50 years. The
expansion can be traced through historic photographs. The change in spatial extent is
also apparent when comparing a forest type map of the 1930s with a present-day
vegetation map (Andrews and Cowlin 1936, Kagan and Caicco 1992). Though it is
theorized that a combination of fire suppression and cattle grazing caused the most
recent advance in juniper distribution, the fossil record does show regular expansion
and contraction of juniper distribution over the last 5000 years. Increased moisture
and fewer, low intensity fires favored juniper, while drought and increased fire
suppressed it (Johnson and others 1994). Juniper has an affinity for the clay soils of
the region; it benefits from the relatively high moisture-retaining capacity of clays
(Johnson and Simon 1987).
-------
Clarke and Bryce--72
The southern edge of the Columbia River basalt marks the northern boundary of the
John Day/Clarno Uplands (fig. 3, 11a). The boundary is quite clear near Antelope,
Oregon, where the basalt flows of the Plateau crop out in a prominent rim called the
Antelope Scarp. Further east, on Highway 19 between Fossil, Oregon, and Mayville,
Oregon, one can drive 7 miles up a steep grade to the level of the Columbia Plateau.
The change from John Day country to the rolling Columbia Plateau is sharp enough
here to be a "line drawn on the ground." The southwestern boundaries of the John
Day/Clarno subregion end on the juniper flats of the Deschutes River basin. The
transition from the John Day/Clarno Uplands (fig. 3, 11a) to the Continental Zone
Foothills (fig. 3, 11 i) subregion is a combination of a change in geology and climate.
In the Continental Zone Foothills (fig. 3, 11i), the regional aspect is more southerly, the
climate more continental, and the underlying geology changes from John Day and
Clarno deposits to more recent volcanics and tuffs.
John Day/Clarno Highlands (fig. 2, lib; fig. 3,11b)—The change from juniper and
grassland savannah to ponderosa pine woodland marks the boundary between the
John Day/Clarno Uplands (fig. 3, 11a) and the forested John Day/Clarno Highlands
(fig. 3, 11b). The final arbiter of the juniper/pine boundary was again the forest-types
map of Andrews and Cowlin (1936). It was completed when juniper distribution was
much less extensive than today. The elevation of the change from grassland to pine
woodland varies from 762 meters (2500 ft) in the north, to 1525 meters (5000 ft) along
the southern margin of the Ochocos, and 1400 to 1463 meters (4600 to 4800 ft) in the
interior John Day valley. The "fuzzy boundary" or transition area between these two
-------
Clarke and Bryce--73
regions estimates the extent of a juniper/pine mixed woodland (fig. 4).
The John Day/Clarno Highlands span a territory with the same east-west extent as the
John Day/Clarno subregion but at a higher elevational range. This subregion is
bounded by the eastward extent of the rainshadow of the higher peaks of the
Cascade Mountain range. As a result of the mountainous barrier to marine weather
systems, the region exhibits a continental climate regime: little rain with wide extremes
in annual and daily temperatures. However, the continental climate in this area is
moderated by a bit of marine influence spreading southward from the Columbia Gorge
and westward through the low passes of the Cascades. The John Day/Clarno
Highlands subregion is not as dry, nor are the temperature extremes as great, as they
are in the Continental Zone subregion of the southern Blue Mountains and the
Continental Zone Foothills to the south and east (Loy and others 1991).
The soil moisture regime in this subregion is xeric, meaning that the soils are
continuously dry for 60 to 90 days during the summer months. The soil temperature
regime is frigid. In the 6600 years since the deposition of the Mount Mazama ash
layer, erosion has had a greater impact on these dry mountains of moderate elevation.
With the ash layer eroded away and no appreciable loess deposits, the water-holding
capacity of the soil is limited. Representative soils have a surface layer rich in organic
matter, with or without a clay-enriched subsoil. They are Ultic Argixerolls, Ultic
Palexerolls, and Lithic Ultic Haploxerolls.
-------
Clarke and Bryce--74
The entire subregion is characterized by a predominance of ponderosa pine forest as
a potential vegetation cover. In the past, frequent low-intensity fires kept mesic forest
to a minimum distribution. Before fire suppression, true fir forests grew to maturity
only in moist areas that experienced few fires. As a result, serai ponderosa pine, and
to a lesser degree lodgepole pine and Douglas-fir, dominated the forest of this
subregion early in this century. The trees grew in park-like stands, at a low enough
density to avoid excessive moisture stress. Shade-tolerant young growth of true fir
regenerated in the understory but did not survive the fires that recurred every 10 to 15
years (Hessburg and others 1994). Understory vegetation includes bitterbrush,
mountain mahogany, bluebunch wheatgrass, pinegrass, and elk sedge.
The early forest type map (Andrews and Cowlin 1936) shows little late-successional
Mesic Forest (fig. 3, 111) in the John Day/Clarno Highlands subregion compared to
other subregions of the Blue Mountains. Only the Continental Zone of the south
central Blue Mountains has less true fir forest mapped. Along the spine of the Ochoco
Mountains, those peaks over 1830 meters (6000 ft), e.g., Wolf Mountain, Peterson
Point, and Mount Pisgah, have a true fir mantle. True fir also grows on the north
slopes of these and other major Ochoco peaks and near major springs between 1525
and 1830 meters (5000 and 6000 ft). The north slopes are a repository of Mazama
ash that provides the optimum growing medium for the grand fir.
Dry ponderosa pine forest and grassland penetrate almost 30 kilometers (50 mi) east
along the John Day River valley from Picture Gorge to Prairie City, Oregon. There the
-------
Clarke and Bryce-75
John Day/Clarno formations and the Picture Gorge basalts meet the melange geologic
region, a chaotic collection of ocean sediments, deep sea crust, granitic intrusions,
and metamorphosed sediments and basalts, that constitute the core of the central
Blue Mountains. The change in geology plus a pronounced increase in elevation at
the Strawberry, Greenhorn, and Elkhorn Mountains mark the eastern boundary of the
John Day/Clarno Highlands subregion where orographic uplift of weather systems and
expansiv/fe high elevation areas produce a significant extent of mesic forest and
subalpine parkland. Farther north the John Day/Clarno Highlands subregion meets
the Maritime-Influenced Zone of the northern Blue Mountains.
Maritime-Influenced Zone (fig. 2, 11c; fig. 3, 11c)—The Maritime-Influenced Zone (fig.
3, 11c) is that part of the Blue Mountains that directly intercepts the marine weather
systems moving east through the break in the Cascade mountain range at the
Columbia River Gorge. The wet weather is intensified as the clouds rise up the slopes
of the northern Blue Mountains; rain and snow is delivered to these mountains three
seasons of the year. The maritime area receives more precipitation than anywhere
else in the Blue Mountains except the high Wallowas and Elkhorns (Loy and others
1991).
This ecological subregion is a mountainous continuation of the drier Columbia Plateau
to the north; it is composed of a lower elevation xeric forest surrounding large areas of
higher elevation mesic forest and a small area of subalpine parkland in the Blue
Mountains of southeastern Washington and in the northern Elkhorn Mountains.
-------
Clarke and Bryce--76
Geologically, this zone is composed almost entirely of Columbia River basalts, except
for a bit of John Day and Clarno deposits in the far west and south. The uplifted
basalt plateau has an elevation range of 915 to 2135 meters (3000 to 7000 ft). Mean
annual precipitation varies from 58 centimeters (23 in) at the grassland-pine margin to
100 centimeters (40 in) in the upper elevations of the upper Grande Ronde basin to
the south.
The soil of the region, however, is not a direct reflection of its basaltic parent material.
Much of northeast Oregon is covered by an ash mantle as much as 1-meter thick from
the eruption of Mount Mazama in southwestern Oregon 6600 years ago. Glacier Peak
and Mt. St. Helens also contributed ash layers 12,000 and 10,000 years ago.
Elsewhere in the Blue Mountains, in grassland areas and south-facing slopes, much of
this ash layer has eroded away; but this area, because of its adequate moisture, has
had a constant vegetative cover and low fire incidence that preserved the ash layer.
Because of the region's proximity to the Columbia Plateau, there is also a
considerable amount of loess soil present, particularly in the lower elevation xeric
forest and on the north slopes of the northern Blue Mountains. Loess soil also has an
increased water holding capacity, though not as great as ashy soil. Loess soil's high
nutrients, rock-free texture, and moist microclimate mean high productivity for the
Idaho fescue and mesic shrub associations (Johnson and Simon 1987). Douglas-fir
also grows abundantly on loessial soil. It is normally at a competitive disadvantage
with ponderosa pine on soils with little or no loess influence, but on loess soils
-------
Clarke and Bryce-77
Douglas-fir is able to compete successfully (Johnson and Simon 1987).
In the ponderosa pine woodland at the grassland-forest transition and in the adjoining
lower elevation forest, the soil moisture regime is xeric. The soils, Ultic Argixerolls,
Ultic Palexerolls, and Lithic Ultic Haploxerolls, have a surface layer rich in organic
matter with or without a clay enriched subsoil. Because of the increased moisture
availability in the subregion generally, these xeric forests occur at a lower elevation
than elsewhere in the Blue Mountains. Ponderosa pine and bunchgrass associations
occur on south aspects at about 640 meters (2100 ft) and pine/snowberry at 730
meters (2400 ft). In canyons and in the forest understory, ninebark, serviceberry,
oceanspray, and snowberry constitute a diverse shrub community that is able to
regenerate after fire. Tree regeneration is sometimes delayed in the dense shrub layer
(Johnson and Clausnitzer 1992). However, once established, Douglas-fir is resistant
to the constant movement of colluvial soils at canyon sites. The steep canyon slopes
at elevations of 700 to 1770 meters (2300 to 5800 ft) in this subregion are covered
with stands of Douglas-fir on soils formed in ash, loess, and basalt colluvium (Pachic
Ultic Haploxerolls and Vitrandic Argixerolls).
Mesic Forest Section (fig. 3,111)—Above about 762 meters (2500 ft), in the Mesic
Forest (fig. 3, 111) section of the Maritime-Influenced ecoregion, the soil temperature
regime is still frigid, but the soil moisture regime becomes more udic (moist). At the
upper elevations of the mesic forest section, where moisture continues to increase, the
soil temperature regime becomes colder, in the cryic range. At mid- and high
-------
Clarke and Bryce-78
elevations in this region, the soil surface is composed of a thick mantle of volcanic
ash, particularly on north-facing slopes. These ashy soils may have a clay-enriched
subsoil (Alfic Udivitrands and Andic Eutroboralfs). Those without a clay subsoil are
called Typic Udivitrands. Soils on south-facing slopes, with more of the volcanic ash
eroded away, may be drier for a longer period of time. They are the Typic Vitrixerands
and Vitrandic Xerochrepts. Those soils with a clay-enriched subsoil are called
Vitrandic Haploxeralfs, and those with a dark surface layer rich in organic matter are
Vitrandic Haploxerolls.
Grand fir, Douglas-fir, lodgepole pine, Engelmann spruce, larch, and subalpine fir all
grow within the moisture and temperature ranges of the Mesic Forest Zone. Because
of the relative lack of moisture stress and the nutrient value of the ash soil, forest
productivity in the Maritime-Influenced Zone is higher than in other ecoregions of the
Blue Mountains. Larch, Douglas-fir, and lodgepole pine are the serai species;
historically, serai stands covered large areas at these elevations after wildfire. In the
late-seral grand fir forests, understory species range from twinflower and birchleaf
spirea in the drier end of the grand fir series at elevations as low as 762 meters (2500
ft), to Rocky Mountain maple or Pacific yew in the wettest areas, with big huckleberry
and grouse huckleberry associated on the highest, coldest sites (Johnson and
Clausnitzer 1992).
Subalpine Areas (fig. 3,11m)—The Subalpine Areas (fig. 3, 11m) in the Maritime-
Influenced Zone are limited to an area at the northern end of the Elkhorn range and
-------
Clarke and Bryce-79
scattered areas of subalpine fir and ridgebrow grasslands in the Blue Mountains of
southeastern Washington, particularly near Table Rock lookout. Johnson and
Clausnitzer (1992) list the elevation of subalpine fir in the north as 1400 to 1705 meters
(4600 to 5600 ft). At its lower elevational limits, subalpine fir is a component of the
closed canopy mesic forest. It is difficult to find and map the point at which this forest
cover becomes patchy and interspersed with alpine meadows. As a result, subalpine
areas could be shown more extensively in the northern Blue Mountains, but most likely
the areas are too fragmented to be depicted on a map of this scale.
The boundary decisions for this subregion depend upon tracking the limits of the
marine influence coming through the Columbia River Gorge from the west. The
marine influence is evident first at Madison Butte and Tupper Butte in south central
Morrow County, Oregon, with the appearance of areas of mesic fir forest. To the
west, the crest of the ridge is several hundred feet lower, just above 1524 meters
(5000 ft), and the areas of climax grand fir forest disappear (Andrews and Cowlin
1936). This is the point at which we pass out of the path of weather systems moving
east through the Columbia Gorge and into the shadow of the Cascade Mountains.
From here east the northern boundary of this subregion is marked by the transition
from Columbia Plateau grassland to the moister higher elevation ponderosa pine
woodland of the Blue Mountains. To the south and west, the boundary between the
Maritime-Influenced Zone and the John Day/Clarno Highlands follows the convex
slope of the anticline of the northern Blue Mountains as it drops into the dry south-
facing canyon slopes of the North Fork of the John Day River. Farther south and east
-------
Clarke and Bryce--80
the boundary closely follows to the ridgeline above the headwaters of the upper
Grande Ronde River. Though the marine influence would overlap the crest of the
ridge to the south, the geology also changes in this vicinity, from Columbia River
basalts to the much more complex geology of the central Blue Mountains, the
Melange Subregion.
Melange Subregion (fig. 2, 11d; fig. 3, 11d)-The Melange (11d) subregion forms the
central core of the Blue Mountains. Webster's Dictionary defines melange as a
"mixture of heterogeneous and often incongruous elements." These are perfect
adjectives to describe the collection of limestone, marine mudstones, serpentinite and
peridotite, and the metamorphic greenstones and schists that make up the Elkhorn,
Greenhorn, Vinegar, and Aldrich Mountains.
Just as incongruous was the way in which these formations became a part of
Oregon's land mass 100 million years ago. They are "accreted terranes," i.e., chains
of volcanic islands that collided with the continental margin or portions of
dismembered oceanic crust that were scraped off by the conveyor-belt movement of
an oceanic plate diving beneath the continental crust at a subduction zone. In the
central Blue Mountains Melange subregion, a large piece of oceanic crust, an ophiolite
succession, composed of serpentine and peridotite, gabbro, dikes, pillow lavas, and
deep sea sediments (in that order!), constitutes Canyon Mountain above John Day,
Oregon (Hamblin 1985, Thayer 1990).
-------
Clarke and Bryce~81
The soils over serpentine and other metamorphic rocks in the Blue Mountains can be
thin and rocky, particularly in erosional settings. They do not retain moisture, and they
create a hostile environment for plant growth, in part because of their high magnesium
content. Streams draining metamorphic areas dry up early in the season, and xeric
plant associations persist to higher elevations in metamorphic areas.
After their collision with the continental margin, the new lands were partially submerged
by the accumulation of muds and sediments in Cretaceous seas. The sedimentary
rocks in the Aldrich Mountains, rising south of John Day and Dayville, Oregon, are
composed of thick deposits of shales and very fine sandstones (graywackes) that
accumulated to great depths in the ancient sea. They are more crumbly and slightly
more erodible than the argillites of the Elkhorn Mountains (Walker and Robinson
1990). The soil from the graywackes is a well-drained gravelly loam soil (Lithic
Xerochrept) that supports ponderosa pine with fescue and Sandberg bluegrass.
Scattered granitic bodies of Jurassic to Cretaceous age, about 90 to 140 million years
ago, have intruded the matrix of accreted terrane in the Melange Subregion. Granite
intrusions have been exposed at Dixie Butte, Vinegar Hill, and the central and northern
Elkhorns. Gold and silver were discovered in the margins of the granitic intrusions,
and other precious metals have been mined from the metamorphic areas. Placer
mining altered the structure of many of the streams in the region. The town of John
Day is built upon the tailings piles of a mining operation (Thayer 1990).
-------
Clarke and Bryce-82
Above the older Paleozoic and Mesozoic formations in the central Blue Mountains are
several thousand feet of ash and lava from widespread post-Cenozoic volcanism, e.g.,
the Clarno, John Day, Picture Gorge, and Strawberry Mountain volcanics. Beginning
about 15 million years ago, these ranges, the Aldrich and Strawberry Mountains and
ridges to the north, were compressed from both the north and the south and uplifted
along the John Day fault to form the folded topography evident today (Thayer 1990,
Walker 1990a).
By the end of the Pliocene period, only the highest peaks of the original accreted
terranes protruded through the various basalt and erosional layers; the early rocks of
the accreted terranes today constitute 10 to 15 percent of the mountainous area of the
Blue Mountains (Walker 1990a). Glacial activity during the Pleistocene then sculpted
the major peaks into cirques and amphitheaters, shaped valleys above 1525 meters
(5000 ft), and left major sedimentary deposits in valley bottoms (Thayer 1990).
At the lowest forested elevations 1372.5 to 1525 meters (4500 to 5000 ft) in the
Melange Subregion (Map Unit 11d), ponderosa pine and juniper grow in an open
forest above a diverse shrub understory of mountain mahogany, bitterbrush, and
squaw currant. In higher, cooler sites, Douglas-fir is co-dominant with ponderosa pine
with the same understory plants. At the highest elevational ranges for Douglas-fir, on
dry slopes and ridgetops approaching 1677.5 meters (5500 ft), a Douglas-fir/pinegrass
or Douglas-fir snowberry association may be found. Recurring fire favors the
persistence of ponderosa pine with a pinegrass or elk sedge understory (Johnson and
-------
Clarke and Bryce-83
Clausnitzer 1992). All of these associations are adapted to grow on metamorphic
substrate; the droughtiness of the soil may be ameliorated by a thin mantle of ash or
loess.
The eastern Strawberry Mountains differ from the older accreted terranes of the
Melange subregion in that they are composed of mostly younger basalt and andesites.
The residual soil from the basalt and andesite substrate differ from ash and loessial
soil in that they have a finer texture in the upper profile, coarser fragments in the soil
matrix, and a lower moisture-holding capacity, resulting in a wider elevational range of
ponderosa pine and Douglas-fir dominated forest. Where an ash layer is present, the
productivity of the soil increases and grand fir is able to grow. The soils in this zone
have a frigid temperature regime and a xeric moisture regime. The mixed forest soils,
with or without an ash layer or a clay-enriched subsoil, include Typic and Alfic
Vitrixerands, Vitrandic and Ultic Haploxeralfs, and Vitrandic and Dystric Xerochrepts.
Under the more xeric, ponderosa pine forest in the lower elevations, soils form from
basalt residuum and colluvium and have a surface layer high in organic matter; they
may or may not have a clayey subsoil (Vitrandic, Lithic and Lithic Ultic Haploxerolls,
Lithic and Typic Argixerolls).
Mesic Forest Section (fig. 3,111)-In the Elkhorn Mountains, at elevations
surrounding the subalpine and alpine core, the Mesic Forest section here (fig. 3, 111)
resembles that of the Maritime-Influenced Subregion (fig. 3, 11c) to the north. The
difference is that the area is outside of the direct influence of the marine weather.
-------
Clarke and Bryce--84
However, these mountains receive adequate precipitation to support a true fir forest
community (up to 85 cm [35 in]) because of orographic uplift on the west slopes
above the John Day basin. The soil-moisture regime here is drier than in the maritime-
influenced area, but the onset of moisture stress is delayed by a moisture-retaining
ash mantle. Characteristic soils in this Mesic Forest section (fig. 3, 111) are Typic and
Alfic Vitrixerands, Vitrandic and Ultic Haploxeralfs, and Vitrandic and Dystric
Xerochrepts.
Subalpine Areas (fig. 3,11m)—At elevations above 1585 meters (5200 ft) in the
Elkhorns and about 1700 meters (5600 ft) in the Strawberry Mountains, a subalpine
forest of subalpine fir, lodgepole pine, and Engelmann spruce grades into whitebark
pine and alpine sagebrush openings, and finally into alpine meadows at about 2285
meters (7500 ft) (fig. 3, 11m). Soils with a thick ash mantle are classified as Typic
Vitricryands, with a thin ash mantle, Andic Cryumbrepts, and with a thin ash mantle
and a clay-enriched subsoil, Andic Cryoboralfs. These soils are cryic and udic (moist).
In alpine meadows, the soils may be shallow, Lrthic Cryumbrepts and Cryochrepts, or
moderately deep with a thin ash mantle and an organic-matter rich surface (Andic and
Vitrandic Cryumbrepts).
Wallowas/Seven Devils Subregion (fig. 2, lie; fig. 3, 11e)-The Wallowas/Seven
Devils range began as a volcanic island arc in a warm, low latitude sea. As the North
American continent was set in motion westward by the widening of the mid-Atlantic rift
200 million years ago, a subduction zone developed at the western edge of the
-------
Clarke and Bryce-85
continent in what is now western Idaho. As the North American plate moved
westward, it collided with and overran the complex island arc system to form the
earliest Oregon land mass. Geologists have pieced together the evidence of this
genesis from the collage of ocean sediments, limestone reef formations, and slightly
metamorphosed lavas (greenstones) that form the Wallowas and the Seven Devils
Mountains. The Martin Bridge limestone crops out in the cliffs along the Lostine River,
on Marble Mountain, and the Matterhorn. The dark sedimentary rocks and volcanic
greenstones compose many of the intermediate ridges and peaks in the high
Wallowas. Today some of the perimeter sediments are folded from their traumatic
collision with the edge of the continent. Fossil shells attest to their oceanic origins.
The ancient ocean sediments and volcanic debris have been peeled away from the
center of the Wallowas through erosion; they surround the perimeter of a large granitic
intrusion that forms the central core of the Wallowas. Earlier writers, e.g., Baldwin
(1976), suggested that this granitic batholith was related to the Idaho batholith to the
east that is 75 to 100 million years old. However, Goldstrand (1994) places the age of
the Wallowa batholith as 143 to 160 million years, older than the Idaho batholith.
Finally, to complete the geologic setting of the Wallowas and Seven Devils area, the
granitic core and the remains of the island arc terrane surrounding it, rise above a sea
of Columbia River basalt which flooded the area 12 to 15 million years ago. Although
much of the Columbia River basalt has eroded away, it still caps prominent peaks in
the Wallowas, e.g., Chief Joseph, Aneroid, and Twin Peaks.
-------
Clarke and Bryce--86
Following the long episode of accumulation of many strata of flood basalts, the
Wallowas were uplifted by faulting on two sides. The fault system responsible for the
uplift is still active today. The entire Wallowa region is a horst or upthrust block
(Bishop 1994). The fault on the north side of the range is most evident. There the
escarpment is steep, particularly above 2285 meters (7500 ft) (Smith and others 1941).
Except for some overlapping sediments and slumping on the north side, the line
between the Blue Mountains ecoregion and the Columbia Plateau, here represented
by the greater Wallowa Valley, is sharply delineated. The cold air drainage from the
mountains and rainshadow effect gives the local Wallowa Valley climate continental
extremes in temperature and precipitation.
The faulting that raised the domed structure of the Wallowas as much as 1525 meters
(5000 ft) created a radial drainage pattern. Eagle Cap rises 2950 meters (9675 ft) as
the hub and source of the major streams; at center stage, it steals the scene from the
Mt. Sacajawea which at 3000 meters (9838 ft) is actually the highest peak in the
Wallowas. In this varied geology, stream courses tend to follow faults and contacts
between less resistant and more resistant rock. After uplift, the steep gradients
increased the erosive power of the streams, cutting deep canyons. The upper
sections of these deep V-shaped valleys were altered by the onset of the Pleistocene.
Several times over the Pleistocene epoch 11,000 to 500,000 years ago, glaciers
sculpted the major river valleys into deep U-shaped valleys; they carved the peaks into
amphitheaters and aretes and rearranged erosional debris and sediments (Johnson
and Simon 1987). Airline passengers on flights heading east out of Portland, Oregon,
-------
Clarke and Bryce-87
are routinely thrilled to the sight of the perfect terminal moraine enclosing Wallowa
Lake on the north side of the Wallowa range.
The climate of the Wallowa Mountains is temperate and continental, but the
continentality varies from northwest to southeast across the range. The southern
slopes are quite dry and endure continental extremes in weather. The northwest part
of the range, on the other hand, has a marine influence. Winter severity also varies
according to the degree of continentality as well as elevation. Precipitation increases
steadily with elevation; at elevations below 1220 meters (4000 ft), half of the total
precipitation occurs during the winter months. Above 1525 meters (5000 ft), two-thirds
of total precipitation falls as snow (Johnson and Simon 1987).
As in the other ecoregions with high relief, the dome of the Wallowas range is
separated into several elevational zones. In the north, a mesic shrub community of
ninebark, common snowberry, Rocky Mountain maple, and serviceberry grows
beneath the xeric forest canopy. To the south, where the slopes of the Wallowas
meet the Continental Zone Foothills subregion, a xeric shrub community of bitterbrush
and mountain big sage grows beneath ponderosa pine and Douglas-fir at elevations of
1100 to 1220 meters (3600 to 4000 ft). The soils in the xeric forest zone are xeric and
frigid. The Vitrandic and Ultic Argixerolls, Ultic Palexerolls, and Lithic Ultic Haploxerolls
have a surface layer rich in organic matter with or without a clay enriched subsoil.
Those with a thick mantle of volcanic ash are classed as Typic Vitrixerands.
-------
Clarke and Bryce-88
Mesic Forest Areas (fig. 3,111)—The western Wallowas are still influenced by the
marine weather systems coming through the Columbia Gorge. There are large areas
of Mesic Forest (111) in the western Wallowas above about 4000 feet. Mean annual
precipitation ranges from 72 to 125 centimeters (30 to 50 in), most of it falling as
snow. Snow persists late into the spring, shortening the frost-free period to just 60
days. Soils in the Wallowas Mesic Forest subregion are both cool (frigid) and cold
(cryic), and usually moist (udic); they have a significant ash content and are the most
productive soils in the Wallowas. They are classed as Typic Vitricryands, Typic and
Alfic Udivitrands, Andic Eutroboralfs, Vitrandic Haploxeralfs, and Vitrandic Haploxerolls.
The Mesic Forest zone here and surrounding the subalpine core of the Wallowas is
characterized by grand fir and various understory shrubs and forbs, depending upon
slope, aspect, and moisture availability. Subalpine fir becomes more common in the
higher elevations. A grand fir/queen's cup beadlilly association is the most productive,
growing in fine ashy soil. Grand fir with big huckleberry also has a wide range and a
wider tolerance for rockier soil. In general, the character of the mesic forest
community is similar to that in the Maritime-Influenced Zone to the west and north.
Subalpine Areas (fig. 3,11m)-The Subalpine Area of the Wallowa Mountains begins
at the transition to colder soils, deeper snowpack and a shorter growing season.
Subalpine fir grows as low as 1370 meters (4500 ft) along perennial streams and in
drainage headlands. Engelmann spruce is serai or codominant to subalpine fir in
areas with adequate moisture and infrequent fire. Mountain hemlock makes a rare
appearance in the northwest quarter of the Wallowas on sites with northeast or
-------
Clarke and Bryce--89
northwest aspects at elevations of 1830 to 2195 meters (6000 to 7200 ft). Mountain
hemlock is a species that is suited to withstand the deeper snow in that maritime-
influenced portion of the Wallowas.
Above 2286 meters (7500 ft), the subalpine parkland grades into alpine meadows.
Historically, green fescue and sedges covered the high alpine meadows from 1798 to
2408 meters (5900 to 7900 ft). Following intense grazing pressure by sheep early in
the century, many high-elevation plant associations have reverted to serai or exotic
species. For more details, Johnson and Simon (1987) provide a thorough examination
of plant associations and habitats in the Wallowa Mountains and the Snake River
Canyon.
To the east the Wallowas blend with the fragmented remains of the Columbia Plateau
bordering the deep Snake and Salmon river canyons. The ecoregion boundary there
is a combination of the end of the metamorphic geology and the bottom of the eastern
Mesic Forest zone of the Wallowas. East of there the Columbia River basalt
dominates and changes the character of the ecosystems because of the elevation,
aspect, and erosional patterns. These areas are separated out as the Canyons and
Dissected Highlands (11f) subregion.
Canyons and Dissected Highlands (fig. 2, 11f; fig. 3, 11f)—The Canyons and
Dissected Highlands (fig. 3, 11f) subregion is comprised of several disjunct areas: the
eastern and southern edge of the block of mountains in southeastern Washington, the
-------
Clarke and Bryce-90
eastern Wallowas, and the isolated islands of former plateaulands cut by the Snake
River canyons to the east of the main Blue Mountain anticline. The subregion is an
extension at higher elevations of the Canyons and Dissected Uplands subregion of the
neighboring Columbia Plateau ecoregion. At elevations of 1525 to 2135 meters (5000
to 7000 ft), the Canyons and Dissected Highlands subregion is high enough to be
considered mountainous. However, it has the same topographic expression as the
Canyons and Dissected Uplands; that is, the uplifted Columbia basalt plateau has
been eroded to a series of knife-edge ridges cut by deep canyons. The region is in
the lee of the Maritime-Influenced Zone (11c) to the west which creates a rainshadow
effect. However, the climate is still fairly moist; the mean annual precipitation ranges
from 38 to 100 centimeters (15 to 40 in).
The soils are a mixture of colluvial canyon soil and soil with a loess or ash mantle.
Because of the presence of loess and ash, the xeric forest zone is relatively narrow
and soon grades into a mixed ponderosa pine, Douglas-fir, western larch, and grand
fir forest. The plateau soils are frigid and usually moist (udic). They are classed as
Typic Udivitrands if they have a mantle of volcanic ash without a clay subsoil and as
Alfic Udivitrands, Andic, or Vitrandic Eutroboralfs if they have a clay-enriched subsoil.
Often on southern aspects in this region the dominant vegetation changes from
bunchgrasses directly to Douglas-fir without an intervening ponderosa-pine phase.
This is partly due to the loess content of the soil. Douglas-fir has an affinity for loess
and its additional moisture-retaining capacity. Also, Douglas-fir is adapted to growing
in the difficult environment of shifting colluvial soils on steep canyon slopes.
-------
Clarke and Bryce--91
All areas of the Canyons and Dissected Highlands subregion border the Canyons and
Dissected Uplands subregion of the Columbia Plateau except for the east side of
Summit Ridge east of the Wallowas Mountains, where the terrain dives into the canyon
depths of the Snake and Salmon River Canyons (11g).
Snake and Salmon River Canyons (fig. 2, 11g; fig. 3, 11g)—The Columbia River flood
basalts buried the accreted terranes 13 to 16 million years ago, leaving only the
highest Wallowas and Seven Devils Mountains protruding. The basalt dams blocked
the flow of the ancestral Snake River, creating Lake Idaho. About 2 million years ago,
headward erosion and stream capture created the present Snake River channel
between Boise and Lewiston, Idaho. The drainage of Lake Idaho carved Hell's
canyon to expose the foundation rocks of the Blue Mountains, the stranded islands of
the island arc terrane. During the Pleistocene, the emptying of Lake Bonneville
widened and deepened the Snake River canyon (Johnson and Simon 1987). The
topographic relief at Hell's Canyon is 1525 meters (5000 ft) from the river to the tops
of the Seven Devils Mountains, making it the deepest canyon in the United States.
The Snake and Salmon River Canyons (11g) subregion is a continuation of a similar
subregion, the Canyons and Dissected Uplands, in the Columbia Plateau ecoregion to
the north (see section 1 above). It was an option to retain these two canyons as part
of the Canyons and Dissected Uplands subregion of the Columbia Plateau. The two
subregions have a similar erosional pattern, a parallel series of knife-edged ridges cut
into the basalt. They also have similar climate, precipitation pattern, and grassland
-------
Clarke and Bryce-92
vegetation. The climate in the canyon is hot and dry, with mild winters. Precipitation
varies annually from 25 to 50 centimeters (10 to 20 in).
The Snake and Salmon River Canyons have been included in the Blue Mountains
ecoregion in part because of their geographic location, surrounded as they are by the
Wallowas and Seven Devils Mountains. In addition, these canyons differ from those to
the north in the Columbia Plateau ecoregion by the presence, under a thick covering
of Columbia River basalt, of metasedimentary and metavolcanic rock belonging to the
Wallowa/Seven Devils island arc terrane that became part of the continent 100 million
years ago.
These metamorphic rocks may produce deep soils on river benches; however, the soil
is stony, and it retains little moisture. Bluebunch wheatgrass, Sandberg's bluegrass,
and spiny greenbush are adapted to grow under these hot, dry conditions (Johnson
and Simon 1987). Canyon slopes with a favorable northern aspect may have a
covering of loess. Loess retains moisture better than colluvial soil; it fosters Idaho
fescue associations and more mesic shrubs such as rose and snowberry (Johnson
and Simon 1987).
The soils over the Miocene basalt flows are divided by elevation into those below 1067
meters (3500 ft) with a mesic temperature regime and those above 1067 meters (3500
ft) with a frigid temperature regime. Below 1067 meters (3500 ft), the aridic and xeric
soils are dominated by Lithic, Aridic, and Typic Haploxerolls. Above 1067 meters
-------
Clarke and Bryce--93
(3500 ft), the xeric soils in the cooler temperature zone are Lithic and Pachic
Argixerolls and Uthic Haploxerolls. Vegetationally, bluebunch wheatgrass and
Sandberg bluegrass are adapted to the driest areas; Idaho fescue dominates in the
increased moisture of the middle to higher elevations, becoming mixed with snowberry
and scattered ponderosa pine at the upper end of the canyons.
Continental Zone Highlands (fig. 3, 11 h)—The Continental Zone Highlands (fig. 3, 11 h)
comprise the south central and southeastern Blue Mountains, those areas most in the
continental climate regime. Low precipitation, high evapotranspiration, abundant
sunshine, and high temperature extremes of 37.7°C to -23.3°C (100°F to - 10°F),
characterize the climate pattern in this section of the Blue Mountains. The stream
density is less than elsewhere in the Blue Mountains, and intermittent streams are
more numerous. The few major streams flowing south end in the interior drainage of
the Harney Basin. The topography is not as steep as elsewhere in the Blue
Mountains; it is an undulating landscape with broad open flats and stringers of
woodland distributed up the draws. Isolated mountain peaks and buttes dot the
landscape. Elevations range from 1067 meters (3500 ft) to near 2135 meters (7000 ft)
(Paulson 1977).
The geology of the area is also different from regions to the north. Recent volcanics
of Pliocene age, rhyolites and ash flow tuffs, cover these southern mountains as well
as areas far to the south to northern Nevada (Walker 1990b, Walker and MacLeod
1991). Rhyolite is a moderately hard rock that is resistant to weathering. As a result,
-------
Clarke and Bryce-94
the soil formed from residuum and colluvium of rhyolite is often shallow and cobbly or
gravelly. Most of the soils in this subregion are in the xeric soil moisture regime and in
the frigid soil temperature regime. In the lower elevations, where ponderosa pine is
the climax species, the major soil subgroups are Vitrandic and Lithic Haploxerolls as
well as Lithic and Typic Argixerolls.
The vegetation cover of the Continental Zone has similarities to the central Blue
Mountains and the southern Ochocos in the ponderosa pine/xeric shrub/bunchgrass
associations. Ponderosa pine forest occupies the warmest, driest zones; it is the
climax species in the lower elevations. The pine grows on a wide range of soils in a
zone from 762 to 1675 meters (2500 to 5500 ft) where the mean annual precipitation is
41 to 89 centimeters (16 to 35 in). At about 1370 meters (4500 ft), the scattered
pine/grassland savannah becomes a more closed canopy woodland with pine and
juniper mixed with xeric shrubs. The most common xeric shrubs are big sage,
mountain mahogany, bitterbrush, and squaw currant. These pine-shrub associations
are found at elevations of 1220 to 1675 meters (4000 to 5500 ft). Of the understory
plants, big sage indicates a very dry site, mountain mahogany indicates a stony, low
productivity site, and bitterbrush a site higher in productivity (Hall 1973, Johnson and
Clausnitzer 1992).
The especially early drought on xeric, ashy soil also produces several distinctive
associations that are shrubfree. They are the ponderosa pine or Douglas-fir with
pinegrass and elk sedge associations. Both plants are promoted by fire, but
-------
Clarke and Bryce~95
pinegrass prefers ash soil. These associations are found on southern exposures on
droughty, thin soil from 1220 to 1980 meters (4000 to 6500 ft).
There is no appreciable Mesic Forest subregion in the upper elevations of the
Continental Zone Highlands as there is in the other subregions of the Blue Mountains.
In the higher elevations, grand fir is technically the climax with pinegrass and elk
sedge understories. However, normally, the grand fir rarely matures because of the
fire frequency (Hall 1973, Johnson and Clausnitzer 1992). Douglas-fir and grand fir
associations do form a mixed forest at the moist end of the xeric soil moisture
gradient. Douglas-fir climax associations may be found from 1220 to 1705 meters
(4000 to 5600 ft) on all parent materials.
Continental Zone Foothills (fig. 3, 11i)—The Continental Zone Foothills (fig. 3, 11i)
subregion bounds the southern perimeter of the Blue Mountains from a point near the
southeast corner of Crook County on the southern flanks of the Ochoco Mountains,
around the east side of the Elkhorn Mountains and Baker Valley, to the edge of the
massive batholith in western Idaho. Elevations range from 1067 to 1615 meters (3500
to 5300 ft) in the eastern portion and from 1370 to 1585 meters (4500 to 5200 ft) in the
west. Precipitation ranges from 25 to 40 centimeters (10 to 16 in), increasing with
increasing elevation. The precipitation falls as light snow, spring or fall rains, or
occasional summer thunderstorms. Land use is predominantly grazing, except in the
irrigated valleys.
-------
Clarke and Bryce--96
Geologically, the Continental Zone Foothills have three major subdivisions. The
foothills from the Ochoco Mountains to the Willow Creek valley (near the Baker
County/Malheur County line) form the first unit. They are underlain by lava flows and
breccia of Miocene basalt and andesite. These volcanics cover a large area of the
high desert in Oregon south to the Nevada border.
The second geologic subdivision, the hills and terraces bordering the Baker Valley
area, south of the Wallowa Mountains to the Snake River canyon, is a continuation of
the melange geology of the central Blue Mountains. The geologic landscape is a
chaotic collection of volcanic island arc fragments, deep sea crust and sediments, and
granitic intrusions, overlain in the south and east by Miocene basalts and andesites of
the Powder River volcanics.
Finally, an arm of the Blue Mountains east of the Snake River extends south above the
Snake River Plain to just north of Boise, Idaho. Geologically, this area is the
southernmost extension of the Miocene flood basalts that inundated the Columbia
Plateau. The basalt filled the topographic trough that follows the "suture zone" of the
new (200 myr old) accreted terrain and the ancient continental land mass. The flood
basalt also blocked the flow of the Miocene Snake River to create Lake Idaho. The
southern foothills, running south of Huntington, Oregon, and east into Idaho, end at
the margin of the valley sediments deposited in Lake Idaho, what is now the Snake
River Plain.
-------
Clarke and Bryce--97
However, from an ecological point of view, geologic makeup is subordinate to the
other landscape characteristics that make this subregion distinctive. The ecosystem is
a reflection of the consistency of the climate, soil, and vegetation cover. The entire
area is solidly within the continental climate zone. The combined land masses of the
Cascade Mountain Range and the Blue Mountains to the northwest effectively block
any marine influence. As a result, the Continental Zone Foothills subregion
experiences wide temperature ranges, xeric and aridic soil regimes, high
evapotranspiration, and high early season moisture stress for vegetation. The few
perennial streams draining the Ochocos and Malheur Mountains empty into the
marshes of the Harney Basin to the south. To the east, the subregion is surrounded
by higher mountain ranges, the Elkhorns and Wallowas, and thus has more perennial
streams; but up away from the streams, the upland areas have even drier soil regimes
than the foothills farther west.
The soils of the western portion of the foothills region have basalts, andesites, and
rhyolites as parent material. They are shallow to moderately deep soils with xeric
moisture and frigid temperature regimes. The dominant soil families are Typic, Lithic,
and Pachic Argixerolls as well as Lithic Haploxerolls. The surface layer of these soils
is typically very stony loam with a gravelly clay loam subsoil.
The soils of the eastern foothills with mixed geology are deeper in general than those
to the west, except on the steeper hill slopes. The soils on terraces below an
elevation of about 1220 meters (4000 ft) have an aridic moisture and a mesic
-------
Clarke and Bryce--98
temperature regime. The soils above 1220 meters (4000 ft) have a xeric moisture and
a frigid temperature regime. The surface layer is typically a silt loam or loam with a
clay loam or clay subsoil. The soil families on the lower terraces are Aridic and Aridic
Calcic Argixerolls, Xeric Argidurids, and Xeric Haplocambids. On the higher terraces,
the soils often have a hardpan or dense clay layer within 1 meter (40 in) of the soil
surface. Here the major orders are Typic and Pachic Palexerolls and Typic Durixerolls.
The rolling hills between the Baker valley and the Snake River canyon and further east
into Idaho have shallower, more cobbly soils. They are classed as Calcic, Uthic,
Typic, and Pachic Argixerolls.
The vegetation of the Continental Zone foothills is characterized by a diverse desert
shrub community that varies according to the soil depth, texture, and productivity.
The poorest sites, with shallow soil, winter waterlogging, and desert pavement, grow
stiff sagebrush or low sage with some juniper on the scabland "biscuits." Mountain
mahogany grows on rocky sites in a wide elevation range. Somewhat deeper soils
over varied substrate will support bitterbrush, mountain snowberry, and mountain big
sage with bunchgrasses in associations. Both mountain mahogany and bitterbrush
are valuable wildlife resources, providing cover and winter range for mule deer.
Rodents, passerines, and upland game birds utilize the fruits and buds of bitterbrush
(Johnson and Clausnitzer 1992). Bitterbrush decreases to the east where it becomes
scarce on the southern flanks of the Wallowas (Johnson and Simon 1987).
Batholith Contact Zone (fig. 3, 11j)—In westernmost Idaho, an arm of the Blue
-------
Clarke and Bryce--99
Mountains forms the transition from the dry Snake River Plains to the Northern Rocky
Mountains. This area is included with the Blue Mountains because its climate,
geology, soil, and vegetation are consistent with those elsewhere in the Blue
Mountains. There is still a slight marine influence from the northwest in this area; the
climate becomes more extreme to the east. The moderated climate, along with soils
developed from the basaltic parent material, produces a typically Blue Mountain
vegetation. Elevations range from about 1370 to 2530 meters (4500 to 8300 ft), high
enough for a Mesic Forest Zone and a Subalpine Zone. This is the southeastern limit
of the Columbia River basalts. The region is transitional because it bridges the
Columbia River basalts and the massive Idaho granitic batholith; the rock has been
highly metamorphosed to schists, amphibolite, and greenstones (McKee 1972).
Ponderosa pine dominates the forest community between 914 and 1980 meters (3000
and 6500 ft). Its upper limits are determined in part by the depth of snowpack. In this
transitional area, Douglas-fir assumes a similar role to the one it takes farther north in
the Canyons and Dissected Highlands subregion. It is found on north-facing slopes
from 945 to 2164 meters (3100 to 7100 ft) on a variety of substrate with an understory
of ninebark. It is adapted to the more mesic conditions there as well as the colluvial
soils of the canyons (Steele and Geier-Hayes 1989). A Douglas-fir/pinegrass
association is also prevalent in this zone of slight maritime influence (Steele and Geier-
Hayes 1987).
Mesic Forest Section (fig. 3,111)—Above 1525 meters (5000 ft) in the Mesic Forest
-------
Clarke and Bryce--100
Zone, grand fir and blue huckleberry grow on both granitic and volcanic substrate,
between 1525 and 1890 meters (5000 and 6200 ft). Again, due to the slight residual
marine influence, this area is the southeastern outpost for grand fir (Steele and Geier-
Hayes 1987). The major portion of the mesic forest zone occurs on the northwestern
to northeastern aspects. The common serai species in the mesic forest zone are
ponderosa pine, western larch, lodgepole pine, and Engelmann spruce.
Blue Mountain Basins (fig. 2, 11k; fig. 3, 11k)—This ecoregion is a composite
subregion composed of scattered basins large enough to be depicted at the 1:250,000
scale. Structurally, the basins may be the result of depressions or synclinal
downwarps in the flow sequences of the flood basalts (e.g., Ukiah Basin, John Day
Basin), erosion of a soft substrate in ash or tuff areas (Fox Basin), or depressions
defined by faults (Grande Ronde Valley and Baker valleys). Pyroclastic rocks and
tuffaceous sedimentary rocks were deposited in many basins, for example, the John
Day, Fox, Bear, and Logan valleys. The John Day basin was filled with gravelly,
sedimentary rocks from the rising Aldrich and Strawberry Mountains. Airborne ash-fall
tuff characterizes the basins of the southern perimeter of the Blue Mountains, Baker
Valley, the Durkee and Unity Basins, Paulina, Burnt River, and the lower Powder River
valley (Walker 1990b).
The Grande Ronde valley, near La Grande in northeastern Oregon, is a fault-bounded
valley graben. Hot springs, such as those at Hot Lake, are present around the
periphery of the valley, indicating that geologic unrest is continuing. The basin is very
-------
Clarke and Bryce-101
deep to bedrock, filled with up to 610 meters (2000 ft) of sediments to its present day
elevation of 785 to 850 meters (2600 to 2800 ft). The valley fill is a variety of
Pleistocene deposits of gravels and silts, lacustrine deposits, alluvial fans, and loess
from the Columbia Plateau to the north. The lake bed silts are still poorly drained;
they are classified as Haploxerolls, Endoaquolls, and Haplaquolls. Settlers drained
most of the marshland for pasture and hay, but there is a remnant left in the
southwestern valley at Ladd Marsh. The valley also has a significant deposit of loess
running from the central valley to its northern edge. The soils here are very deep to
bedrock; they are classed as Haploxerolls, Argialbolls, and Argixerolls. The potential
natural vegetation on the terraces and loess hills consists of Idaho fescue, common
snowberry, and Sandberg's bluegrass. On the floodplains, tufted hairgrass, redtop,
and sedges are associated with wetter soils. Today, the majority of the valley is
farmed. The Grande Ronde River and its tributary streams, many of which have been
channelized, provide irrigation water to grow hay, commercial grass seed, alfalfa, and
peas.
The Grande Ronde valley is transitional to the dryer, loess covered, Columbia Plateau
to the north; however, structurally the valley is wedded to the Wallowa Mountains.
Also, the climate of the valley is still moderated by the marine influence from the west.
Precipitation amounts vary from 32.5 centimeters (13 in) in the dry southern valley to
60 centimeters (24 in) at the eastern end of the valley. For these reasons, the valley
was included in the Blue Mountains ecoregion rather than in the Columbia Plateau.
-------
Clarke and Bryce-102
The Baker valley of the eastern Blue Mountains is also a fault-bounded depression in
the melange geology of the Baker terrane (see the geologic explanation under the
Melange subregion). Elevations range from 610 to 1675 meters (2000 to 5500 ft).
The valley and surrounding grasslands encompass the alluvial floodplains and terraces
of the Burnt and Powder rivers. The basin is firmly located in an arid area of
continental climate. The mean annual precipitation ranges from 23 centimeters (9 in)
in the lower elevations to 40 centimeters (16 in) on the higher terraces. Soils below
1220 meters (4000 ft) have an aridic soil moisture regime, meaning they receive less
than 25 centimeters (10 in) of precipitation annually. The aridic soils also have a
warmer temperature regime (mesic). The east edge of the basin has abundant
alkaline soils due to poor drainage and high evaporation rates. The shrub
greasewood is abundant there. The floodplain soil is deep silt loam or silty clay loam,
classified as Pachic Haploxerolls or Typic Haplaquepts. Soils on the terraces include
Aridic Argixerolls, Xeric Argidurids, and Xeric Haplocambids.
Finally, many of the scattered depressions filled with airborne ash and pyroclastic
volcanic debris constitute the high elevation meadows. The high meadows are often
also alluvial, with a high water table and with soils containing clay or silt. Streams are
slow and meandering, and if not channelized, they have a dynamic interaction with
their floodplains. These unconstrained sections of stream provide pool habitats so
important to salmonids for refuge and rearing (Mcintosh and others 1994). The
altitude and year-round moisture make these basins unsuitable for cultivation, but they
are grazed heavily by cattle and elk. The vegetation includes sedges and tufted
-------
Clarke and Bryce-103
hairgrass. Meadows in poor condition are dominated by Kentucky bluegrass. The
riparian areas, if still present, support willow. Trampling and overgrazing on the fine
silty substrate may cause the incision of the stream bed, lowering the water table and
stranding the wet meadow without a source of ground water.
-------
Clarke and Bryce--104
Literature Cited
Alt, D.D.; Hyndman, D.W. 1978. Roadside geology of Oregon. Missoula, MT: Mountain
Press Publishing Co. 268 p.
Anderson, E.W. 1956. Some soil-plant relationships in eastern Oregon. Journal of
Range Management. 9(4): 171-175.
Andrews, H.J.; Cowlin, R.W. 1936. Forest type map of Oregon: northeast quadrant.
Scale 1:253,440. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific
Northwest Experiment Station.
Baldwin, E.M. 1976. Geology of Oregon. Dubuque, IA: Kendall/Hunt Publishing
Company. 147 p.
Bishop, E. 1994. The Blue Mountains: a collage of island arcs. Professional Paper
1438. [N.p.]: U.S. Department of the Interior, Geological Survey.
Brooks, H.C. 1979. Plate tectonics and the geologic history of the Blue Mountains.
Oregon Geology. 41(5): 71-80.
Ehmer, L. [Year unknown]. Soil resource inventory. Pendleton, OR: Umatilla National
Forest, Region 6, U.S. Department of Agriculture, Forest Service. 305 p.
-------
Clarke and Bryce-105
Fifarek, R.H.; Juhas, A.P.; Field, C.W. 1994. Geology, mineralization, and alteration of
Red Ledge volcanogenic massive sulfide deposit, western Idaho. In: Vallier, T.L.;
Brooks, H.C. Geology of the Blue Mountains Region of Oregon, Idaho, and
Washington: stratigraphy, physiography, and mineral resources of the Blue Mountains
Region. Geological Survey Professional Paper 1439. Washington, DC: U.S. Department
of the Interior, Geological Survey, and U.S. Government Printing Office: 113-150.
Goldstrand, P.M. 1994. The Mesozoic geologic evolution of the north Wallowa terrane,
northeast Oregon and western Idaho. In: Vallier, T.L.; Brooks, H.C. Geology of the
Blue Mountains Region of Oregon, Idaho, and Washington: stratigraphy,
physiography, and mineral resources of the Blue Mountains Region. Geological Survey
Professional Paper 1439. Washington, DC: U.S. Department of the Interior, Geological
Survey, and U.S. Government Printing Office: 29-53.
Hall, F.C. 1973. Plant Communities of the Blue Mountains of eastern Oregon and
Southeastern Washington. Pacific Northwest Region Area Guide 3-1. Portland, OR:
U.S. Department of Agriculture, Forest Sen/ice, Pacific Northwest Region. 62 p.
Hamblin, W.K. 1985. The earth's dynamic systems: a textbook in physical geology.
Minneapolis, MN: Burgess Publishing. 528 p.
Harvey, A.E.; Geist, J.M.; McDonald, G.I. [and others]. 1994. Biotic and abiotic
-------
Clarke and Bryce--106
processes in eastside ecosystems: the effects of management on soil properties,
processes and productivity. Gen. Tech. Rep. PNW-GTR-323. Portland, OR: U.S.
Department of Agriculture, Forest Service, Pacific Northwest Research Station. 71 p.
Hessburg, P.F.; Mitchell, R.G.; Filip, G.M. 1994. Historical and current roles of insects
and pathogens in eastern Oregon and Washington forested landscapes. Gen. Tech.
Rep. PNW-GTR-327. Portland, OR: U.S. Department of Agriculture, Forest Service,
Pacific Northwest Research Station. 55 p.
Hooper, P.; Swanson, D. 1990. The Columbia River basalt group and associated
volcanic rocks of the Blue Mountains province. In: Walker, G., ed. Geology of the Blue
Mountains region of Oregon, Idaho, and Washington: Cenozoic geology of the Blue
Mountains region. Geological Survey Professional Paper 1437. Washington, DC: U.S.
Department of the Interior, Geological Survey, U.S. Government Printing Office:
63-101.
Irwin, L.; Cook, J.G.; Riggs, R.A.; Skovlin, J.M. 1994. Effects of long-term grazing by
big game and livestock in the Blue Mountains forest ecosystems. Gen. Tech. Rep.
PNW-GTR-325. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific
Northwest Research Station. 49 p.
Johnson, C.G., Jr. 1994. Forest health in the Blue Mountains: a plant ecologist's
perspective on ecosystem processes and biological diversity. Gen. Tech. Rep. PNW-
-------
Clarke and Bryce--107
GTR-339. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific
Northwest Research Station. 24 p. (Quigley, Thomas M., ed; Forest health in the Blue
Mountains: science perspectives).
Johnson, C.G., Jr.; Clausnitzer, R.R. 1992. Plant associations of the Blue and Ochoco
Mountains. Publication R6-ERW-TP-036-92. LaGrande, OR: U.S. Department of
Agriculture, Forest Service, Wallowa-Whitman National Forest. 163 p.
Johnson, C.G., Jr.; Clausnitzer, R.R.; Mehringer, P.J.; Oliver, C.D. 1994. Biotic and
abiotic processes of eastside ecosystems: the effects of management on plant and
community ecology, and on stand and landscape vegetation dynamics. Gen. Tech.
Rep. PNW-GTR-322. Portland, OR: U.S. Department of Agriculture, Forest Service,
Pacific Northwest Research Station. 66 p.
Johnson, C.G., Jr.; Simon, S.A. 1987. Plant associations of the Wallowa-Snake
Province. Wallowa-Whitman National Forest Publication #R-6 ECOL-TP-225A-86.
Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest
Region. 400 p.
Kagan, J.; Caicco, S. 1992. Manual of Oregon actual vegetation. Boise, ID: Idaho
Cooperative Fish and Wildlife Research Unit, University of Idaho. 186 p.
Lehmkuhl, J.F.; Hessburg, P.F.; Everett, R.L. [and others]. 1994. Historical and current
-------
Clarke and Bryce-108
forest landscapes of eastern Oregon and Washington. Part 1: Vegetation pattern and
insect and disease hazards. Gen. Tech. Rep. PNW-GTR-328. Portland, OR: U.S.
Department of Agriculture, Forest Service, Pacific Northwest Research Station. 88 p.
Loy, W.G.; Allen, S.; Patton, C.P.; Plank, R.D., eds. 1991. Atlas of Oregon. Eugene,
OR: University of Oregon. 344 p.
Mcintosh, B.A.; Sedell, J.R.; Smith, J.E. [and others]. 1994. Management history of
eastside ecosystems: changes in fish habitat over 50 years, 1935-1992. Gen. Tech.
Rep. PNW-GTR-321. Portland, OR: U.S. Department of Agriculture, Forest Service,
Pacific Northwest Research Station. 55 p.
McKee, B. 1972. Cascadia. Seattle, WA: Stuart Press. 394 p.
Oliver, C.D.; Irwin, L.L.; Knapp, W.H. 1994. Eastside forest management practices:
historic overview, extent of their applications, and their effects on sustainability of
ecosystems. Gen. Tech. Rep. PNW-GTR-324. Portland, OR: U.S. Department of
Agriculture, Forest Service, Pacific Northwest Research Station. 73 p.
Orr, E.L.; Orr, W.N.; Baldwin, E.M. 1992. Geology of Oregon. 4d ed. Dubuque, IA:
Kendall/Hunt Publishing Co. 254 p.
Paulson, D.J. 1977. Ochoco National Forest soil resource inventory. Portland, OR:
-------
Clarke and Bryce--109
U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station.
347 p.
Retallack, G.J. 1991. A field guide to mid-Tertiary paleosols and paleoclimatic changes
in the high desert of central Oregon - Part 1. Oregon Geology. 53(3): 51-59.
Robbins, W.G.; Wolf, D.W. 1994. Landscape and the intermontane northwest: an
environmental history. Gen. Tech. Rep. PNW-GTR-319. Portland, OR: U.S. Department
of Agriculture, Forest Service, Pacific Northwest Research Station. 32 p.
Robinson, P.T.; Walker, G.W.; McKee, E.H. 1990. Eocene, Oligocene, and lower
Miocene rocks of the Blue Mountains region. In: Walker, G., ed. Geology of the Blue
Mountains region of Oregon, Idaho, and Washington: Cenozoic geology of the Blue
Mountains region. Geological Survey Professional Paper 1437. Washington DC: U.S.
Geological Survey, U.S. Government Printing Office: 29-62.
Skovlin, J.M. 1991. Fifty years of research progress: an historical document on the
Starkey Experimental Forest and Range. Gen. Tech. Rep. PNW-GTR-266. Portland,
OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research
Station. 58 p.
Smith, W.D.; Allen, J.E.; Staples, L.W.; Lowell, W.R. 1941. Geology and physiography
of the northern Wallowa Mountains, Oregon. Bulletin No. 12. Salem, OR: State of
-------
Clarke and Bryce--110
Oregon, Department of Geology and Mineral Industries. 55 p.
Steele, R.; Geier-Hayes, K. 1987. The Grand fir/Blue huckleberry habitat type in central
Idaho: succession and management. Gen. Tech. Rep. INT-228. Ogden, UT: U.S.
Department of Agriculture, Forest Service, Intermountain Research Station. 66 p.
Steele, R.; Geier-Hayes, K. 1989. The Douglas-fir/Ninebark habitat type in central
Idaho: succession and management. Gen. Tech. Rep. INT-252. Ogden, UT: U.S.
Department of Agriculture, Forest Service, Intermountain Research Station.
Thayer, T. 1990. Geologic setting of the John Day country, Grant County, Oregon.
N.p.: N.p. 23 p.
Walker, G. 1990a. Miocene and younger rocks of the Blue Mountains region. In:
Walker, G., ed. Geology of the Blue Mountains region of Oregon, Idaho, and
Washington: Cenozoic geology of the Blue Mountains region. Geological Survey
Professional Paper 1437. Washington, DC: U.S. Department of the Interior, Geological
Survey, U.S. Government Printing Office: 101-119.
Walker, G. 1990b. Overview of the Cenozoic geology of the Blue Mountains. In:
Walker, G., ed. Geology of the Blue Mountains region of Oregon, Idaho, and
Washington: Cenozoic geology of the Blue Mountains region. Geological Survey
Professional Paper 1437. Washington, DC: U.S. Department of the Interior, Geological
-------
Clarke and Bryce--111
Survey, U.S. Government Printing Office: 1-12.
Walker, G.W.; MacLeod, N.S. 1991. Geologic map of Oregon. 1:500,000 scale.
Denver, CO: U.S. Department of the Interior, Geological Survey. 2 sheets.
Walker, G.; Robinson, P.T. 1990. Cenozoic tectonism and volcanism of the Blue
Mountains region. In: Walker, G., ed. Geology of the Blue Mountains region of Oregon,
Idaho, and Washington. Cenozoic geology of the Blue Mountains region Geological
Survey Professional Paper 1437. Washington, DC: U.S. Department of the Interior,
Geological Survey, U.S. Government Printing Office: 119-133.
White, J.D.; Vallier, T.L. 1994. The geologic evolution of the Pittsburgh Landing area,
Snake River Canyon, Oregon and Idaho. In: Vallier, T.L; Brooks, H.C., eds. Geology
of the Blue Mountains Region of Oregon, Idaho, and Washington: stratigraphy,
physiography, and mineral resources of the Blue Mountains region. Geological Survey
Professional Paper 1439. Washington, DC: U.S. Department of the Interior, Geological
Survey, U.S. Government Printing Office: 55-73.
White, J.D.L.; White, D.L.; Vallier, T.L. [and others]. 1992. Middle Jurassic strata link
Wallowa, Olds Ferry, and Izee terranes in the accreted Blue Mountains island arc,
northeastern Oregon. Geology. 20: 729-732.
-------
Clarke and Bryce-112
SECTION 3-LANDSCAPE-LEVEL ECOREGIONS FOR SEVEN CONTIGUOUS
WATERSHEDS, NORTHEAST OREGON AND SOUTHEAST WASHINGTON
by Sharon E. Clarke, Mark W. Garner, Bruce A. Mcintosh, and James R. Sedell
Introduction
Change is constant; streams have undergone millions of years of geological and
climatic processes; however, the rate of change of the present landscape is alarming.
Fifty years ago, spring chinook (Oncorhynchus tshawytscha) spawned in streams
where they may never spawn again. Decreased ranges and declining numbers of
other anadromous stocks and resident species are also evident. The extent,
complexity, and critical nature of this problem has been well documented (Frissell
1993, Henjum and others 1994, Nehlsen and others 1991, Williams and others 1989).
Habitat destruction including loss of large pools, increased sedimentation, loss of
complexity, decreased flows, and increased water temperature, is listed as a major
factor in this rapid decline (Frissell 1993, Henjum and others 1994, Nehlsen and others
1991, Williams and others 1989). These stream habitat characteristics are influenced
by instream and riparian zone processes as well as upland processes (FEMAT 1993,
Frissell and others 1986, Gregory and others 1991, Henjum and others 1994,
Lotspeich and Platts 1982, Platts 1980) operating at many spatial and temporal scales.
To recognize the role of upland processes in modifying stream habitat, we developed
-------
Clarke and Bryce--113
landscape-level ecoregions (Level V). In the text, we use the term ecoregions as a
general term referring to ecological regions defined at any spatial scale; landscape-
level ecoregions refer to regions specifically delineated in this study, that is, an area
defined by specific landscape-level criteria to be smaller in scale than national scale
ecoregions (Level II and III) or a subregion (Level IV). Specific landscape-level criteria
might be uniform soil, climate, topography, geology, geomorphology, or vegetation.
The spatial component of stream ecosystems was recognized by using a hierarchical
landscape stratification. The landscape stratification followed Warren's (1979) ideas of
a classification system for watershed management adapted by Omernik (1987, 1995).
Gallant and others (1989) described the process and uses of landscape stratification at
the national and subregional scale. We used the concepts of hierarchical stratification
for aquatic organisms developed by Frissell and others (1986), Lotspeich and Platts
(1982), Minshall (1988), Platts (1980), and Warren (1979).
Omernik and Griffith (1991) argue against the sole use of a hydrologic framework for
the study and organization of ecosystems. They state that, "although the longitudinal
linkages emphasized in the river continuum concept are important, we must also look
for important lateral linkages with regional characteristics." Ecoregions account for
changes in stream habitat characteristics derived from changes in the surrounding
land and watersheds account for the interconnectedness of water (Whittier and others
1988). Combining landscape stratification with watershed delineations is appropriate
for this study because the pathways for anadromous fish distribution are determined
-------
Clarke and Bryce-114
by watershed linkages and landscape characteristics. To foster this union, our study
area was defined by watershed boundaries. For analysis, finer-scale watershed
delineations can be easily overlaid, with a Geographic Information System (GIS), on
ecoregions using a "cookie cutter" approach. Landscape-level ecoregions form a
bridge between finer-scale stream classifications and coarser-scale ecoregions (Bryce
and Clarke 1996). The choice of which scale in the hierarchy to use depends upon
the question or management problem being addressed (Minshall 1988).
One of the cornerstones of Warren's (1979) classification scheme is the idea of using
the potential capacity of the system. He suggests that "it is capacity of a system not
any particular performance, that most interests us. It is so because performances,
such as structure, of systems are in continuous flux and we want to know what we
can expect of a system." He goes on to say " we cannot have a system whose
essential kind has changed placed in a different class every time its performance
changes, as it will with environmental change. It is useful to distinguish between the
potential capacity of a system and the different realized capacities it could come to
have under different developmental environments and at different stages in a given
environmental system through time." He defines potential capacity "as all possible
performances in all possible developmental environments."
The ecoregion approach assumes that: (1) streams are a reflection of the watersheds
that they drain and (2) ecoregions reflect the variability in the potential capacity of
streams. Climate, geomorphology, geology, topography, potential natural vegetation,
-------
Clarke and Bryce--115
and soil influence substrate, water quantity and quality, habitat complexity, and riparian
vegetation. The relative influence, or importance, of each varies from one region to
another. The grouping of areas having similar landscape characteristics allows a
better comparison of the potential capacities of streams of similar size between each
region.
Ecoregions have been developed at the national (Omernik 1987, 1995) and
regional/state level (Clarke and others 1991, Gallant and others 1989, Griffith and
Omernik 1991, Griffith and others 1994, Griffith and others 1995, Thiele and others
1992) for research, resource management, and regulatory use. Landscape-level
ecoregions are intended to help address issues of research and management at a
local scale by (1) anticipating the effects of different types and intensities of land uses,
(2) providing a target for restoring stream habitats, (3) selecting the monitoring and
research sites, and (4) extrapolating data from research and monitoring sites to other
streams with similar potential capacity.
Many studies are conducted outside of the context of their ecosystems. By using
ecoregions to view the entire fabric of an ecosystem, not just a single thread,
processes operating in one area and not in others may be demonstrated, and some
complex effects and relationships may come to light. For example, meadow and
desert streams obtain much of their energy from within the stream as opposed to
forested areas where the primary energy source is input from litterfall (Murphy and
Meehan 1991). The scientific literature contains many studies attempting to
-------
Clarke and Bryce-116
understand how fish respond to single habitat variables, but attempts to study the
interactions of many habitat variables are few (Petersen and others 1992).
We know that landscape characteristics influence stream habitat. However the effects
of multiple interactions, sometimes acting synergistically and sometimes
antagonistically, are difficult to predict. Instead of trying to knit together all of the
possible contributing factors and predict an outcome or prove a hypothesis
(experimental, bottom-up approach), we have taken a classification (observational, top-
down) approach. However, the knowledge gained by other researchers using an
experimental approach is essential to evaluate the potential influence of any one
landscape characteristic on stream habitat. In this iterative process, the results of the
observational approach help to refine further process-level research. The two
approaches are symbiotic, not competitive.
Objectives of the Project
The goal of this project was to provide a tool to help scientists and managers better
understand the relationship between landscape characteristics and stream habitat for
seven contiguous watersheds within the Blue Mountain and Columbia Plateau
ecoregions of northeast Oregon and southeast Washington. The two objectives of this
research were to use the ecological framework developed for ecoregions and
subregions to define landscape-level ecoregions for seven contiguous watersheds in
the Blue Mountains and Columbia Plateau ecoregions and to describe the rationale
-------
Clarke and Bryce-117
used to delineate the regions.
These landscape-level ecoregions were developed in concert with the refinement and
subdivision of the Columbia Plateau and Blue Mountains ecoregions to ensure the
subregion and landscape-level boundaries coincided (above section 1, above section
2). The quality of these ecoregions were partly dependent upon the quality,
resolution, and scale of the thematic maps used to derive them. Using multiple
landscape characteristics and several sources alleviated some of the problems
associated with an individual map by allowing for comparisons between the layers and
between the sources within the layers. For example, soil maps usually provided
information on vegetation which was compared to vegetation maps. A detailed
description of the merits and drawbacks of each of the map sources was intended to
help potential users of the maps evaluate the usefulness of the classification and point
out where improvements can be made as additional source material becomes
available.
The value of a particular classification scheme is largely a function of the goals and
objectives of a project. In this report, we attempt to explain as explicitly as possible
the reasoning employed in delineating landscape-level ecoregions. This explanation
should help potential users of the map judge the usefulness of this classification for
their purposes.
Connection to Other Approaches
-------
Clarke and Bryce--118
Many approaches to understanding or managing complex interactions between the
uplands and stream habitat have been studied such as:
• Range of natural variability3 (Wissmar 1993).
• Aquatic conservation strategy (FEMAT 1993).
• Biologically based habitat standards (Rhodes and others 1994).
• River continuum (Minshall 1988, Minshall and others 1985, Vannote and others
1980).
• Stream-habitat classification (Bisson and others 1982, Cupp 1988, Frissell and
others 1986, Hawkins and others 1993, Lotspeich and Platts 1982, Montgomery and
Buffington 1993, Nawa and others 1991, Rosgen 1985).
Landscape-level ecoregions are not so much an approach as they are one tool that
can be used in conjunction with these approaches and other tools to assess
ecosystem condition and change through time and the effects of disturbance.
Range of natural variability-The approach using range of natural variability applies
standards that define the limits of acceptable and unacceptable ecosystem conditions
for streams within the spatial and temporal variabilities inherent in dynamic ecosystems
based upon a long-term stream monitoring program (Wissmar 1993). One objective of
-------
Clarke and Bryce--119
the monitoring program is to assess the effectiveness of ecosystem restoration
programs, including defining criteria for success. One of the important components in
the development of a restoration plan and in defining criteria for success is
determination of the present status and historical review of disturbances induced by
natural events and human activities. Wissmar (1993) states that "this information can
be obtained by defining the landscape in terms of bedrock geology, geomorphic
landforms, hydrologic regimes, and distribution of stream and riparian habitats."
Depending upon the objectives, subregions or landscape-level ecoregions may
provide a suitable stratification of the landscape. It may also be useful to embed a
finer-scale stream classification within the coarser landscape-level ecoregions to obtain
another layer of stratification.
Aquatic conservation strategy-FEMAT (1993) propose a system of interim riparian
reserves, key watersheds, watershed analysis, and watershed restoration as their
Aquatic Conservation Strategy. They state "that within a physiographic province,
similar geographic and topographic features control drainage network and hillslope
stability patterns. Riparian reserve design may vary as a result of these differences."
Subregions or landscape-level ecoregions may provide additional information in
assessing drainage network and hillslope stability patterns because of their inclusion of
information on soil, geology, and vegetation.
Watershed analysis uses maps of topography, stream networks, soils, vegetation, and
geology along with sequential aerial photographs, field inventories and surveys,
-------
Clarke and Bryce-120
census data on species presence and abundance, disturbance and land-use history,
and other historical data to provide information on what processes are active within a
watershed, how these processes are distributed in time and space, what the current
upland and riparian conditions of the watershed are, and how all these factors
influence riparian habitat and other beneficial uses. Landscape-level ecoregions
combine most of the map types used in the watershed analysis and may provide a
template on which to place the other types of information for a watershed. Specific
monitoring objectives are derived during the watershed analysis process and are
tailored to each watershed. Riparian areas selected for monitoring are dispersed
among the various landscapes because of the influence of past disturbance,
topography, climate, and other factors on the natural conditions of stream habitat.
Here too, we feel subregions or landscape-level ecoregions may provide guidance in
selecting monitoring sites.
Biologically based habitat standards-Rhodes and others (1994) set standards for
stream ecosystems based on available information about the biological-habitat
requirements of salmon. They state that "meeting the habitat requirements of salmon
species must be the biological bottom line of efforts to protect and restore spawning
and rearing habitat consistent with efforts to stabilize and restore the listed salmon
runs." No further degradation is allowed in stream systems that currently exceed the
biologically based habitat standards. Rhodes and others (1994) agree that
classification is a useful tool for looking at questions of attainability of stream-habitat
criteria and channel response to perturbation. Using subregions or landscape-level
-------
Clarke and Bryce--121
ecoregions may help fine-tune the habitat standards to reflect differences in the
potential capacity of streams. For example, biologically based habitat standards
should be the lowest acceptable standards, but for some systems attaining these
standards still represents a degraded condition, and higher standards should be
imposed. In other systems, natural conditions do not meet the biologically based
habitat standards. Efforts to meet the habitat standards in these streams may be
unrealistic, short-term, and expensive. These efforts could detract from other more
productive efforts. Subregions or landscape-level ecoregions may be useful in
identifying areas where biologically based habitat standards are too low or difficult to
achieve due to the inherent potential capacity of the system.
River continuum—The river continuum concept (Vannote and others 1990) asserts that
stream ecosystems are connected and that conditions change accordingly from
headwaters to the mouth of a river. Because the differences created along a river
continuum are based on the landscape in which it is embedded, the idea of continuity
(the river continuum concept) needs to be merged with the idea of context (regional
classification). The effects of upstream and upslope regions and increasing stream
size need to be recognized in any interpretation of stream ecosystem research.
Landscape classification provides a logical rationale for ordering and testing
relationships.
Stream-habitat classification-Landscape-level ecoregions were developed to
compliment stream-habitat classifications (Bryce and Clarke 1996). Stream-habitat
-------
Clarke and Bryce--122
classification is finer-scale and focuses on the stream channel and valley floor. The
merits of stream-habitat classification are discussed by Bisson and others (1982),
Cupp (1988), Frissell and others (1986), Hawkins and others (1993), Lotspeich and
Platts (1982), Montgomery and Buffington (1993), Nawa and others (1991), and
Rosgen (1985). Stream-habitat classification coupled with a landscape classification of
the entire watershed recognizes the connectedness of the entire drainage network
(Bryce and Clarke 1996, Frissell and others 1986, Lotspeich and Platts 1982). In
addition, O'Neil and others (1986) suggest that using a hierarchical approach allows
for patterns to emerge which may be masked at the next lower level in the hierarchy.
Landscape-level ecoregions may identify patterns masked by the increased variability
of finer-scale stream classifications. However, while stream-habitat classification
generally delineates the current state of the system, landscape-level classification aims
to define potential capacity. This difference makes interpretations and comparisons
between these two methods difficult.
Rationale for Delineating Landscape-Level Ecoregions
Landscape-level ecoregions were developed to aid research and management of
anadromous fish and their habitat.4 To provide a rationale for using a landscape
classification to understand fish distribution and abundance, we described the
connections between landscape characteristics and stream-habitat characteristics and
between stream-habitat characteristics and fish distribution. The primary stream-
habitat characteristics that influence fish distribution and abundance were substrate,
-------
Clarke and Bryce--123
habitat complexity, riparian vegetation, water quality and quantity, and biota.
Landscape characteristics can directly or indirectly influence these stream-habitat
characteristics. By identifying areas that are fairly homogenous in their mix of
landscape characteristics we hypothesized that the potential capacity and the reaction
to both natural and anthropomorphic disturbances of streams will be influenced by the
individual landscape characteristics that constitute landscape-level ecoregions.
Most of the literature discussing stream-habitat characteristics is focused on disturbed
conditions. We have used this information and related it to analogous natural
conditions. In doing so we recognized that a natural system has more resiliency than
a disturbed system.
Substrate-Topography, climate, vegetation, soil, and geology influence sediment
production and erosional processes (Swanson and others 1987). These landscape
characteristics influence the particle size and amount of sediment by affecting the
erosion, transport, and storage of soil particles. Table 3 shows the factors influencing
sediment production along with the relative risk to stream-habitat productivity.
Surface erosion and mass movements contribute sediment to streams. Surface
erosion involves detachment and movement of soil particles. Detachment is influenced
by soil texture and the amount of protection by vegetation. Precipitation intensity and
duration, soil infiltration rate, and slope gradient and length influence soil movement
(Swanson and others 1987, Swanston 1991). Where and when mass movements
-------
Clarke and Bryce--124
occur, their size, and amount of material transported are controlled by slope, parent
material, depth and degree of weathering, and soil saturation (Swanston 1991). The
downstream movement and deposition of sediment are influenced by channel
morphology, water quantity, and amount and size of material (Swanston 1991). Once
the sediment enters the stream, it may move as either suspended sediment or
bedload, depending upon particle size and flow. Small headwater streams are usually
more prone to erosion because both the sideslope and channel gradient are steeper
(Naiman and others 1992). Interaction with the uplands is also greater for these small
streams because there is a less-developed riparian buffer strip to act as a storage
area for sediments.
Erosion and sedimentation are natural processes that contribute to a healthy stream
ecosystem by providing the sources and surfaces necessary for aquatic habitat
(Naiman and others 1992). Both the particle size and amount of sediment in a stream
influence stream habitat. High concentrations of fine sediment may lead to
overloading of substrate material (Heede and Rinne 1990). Fine sediment may directly
reduce the egg-to-fry survival and fry quality of anadromous salmonids by (1)
decreasing water flow through the gravel of the redd therefore reducing dissolved
oxygen and causing suffocation of eggs and alevin and (2) creating a physical barrier
to emergence (Bjornn and Reiser 1991, Everest and others 1987, Murphy and Meehan
1991, Petersen and others 1992, Rhodes and others 1994, Swanston 1991). Rhodes
and others (1994) report that decreased interstitial space from sedimentation results in
the elimination of fry from habitat, alters the food base, and reduces available winter
-------
Clarke and Bryce-125
interstitial habitat and consequently may increase pre-smolt mortality.
Indirect effects of sedimentation on salmonids include changes in channel
morphology, increased turbidity, and decreased food availability (Everest and others
1987, Murphy and Meehan 1991, Petersen and others 1992, Swanston 1991). An
abundance of coarse-grained sediments may increase infiltration and percolation,
thereby reducing surface flows during dry conditions (Everest and others 1987).
Different salmonid species require different gravel substrate sizes for spawning; large
fish can use larger substrate materials than can small fish (Bjornn and Reiser 1991).
Gravel-sized material in riffles provides the primary food-producing areas for fish
(Heede and Rinne 1990).
Habitat complexity-Bisson and others (1982) define habitat complexity as the
distribution and abundance of habitat types. Lichatowich and others (1995) add that
complexity includes connectivity throughout the salmonid's range. They define
connectivity as the ability to migrate at the appropriate time between links in the
habitat chain. Habitat complexity is controlled by discharge, sediment load, bank
characteristics, and structural features such as large woody debris (LWD), bedrock,
boulders, and sediment wedges (Murphy and Meehan 1991, Sullivan and others 1987,
Swanston 1991). The presence of these structural features is determined by local
geology and active hillslope-and-channel erosion processes (Sullivan and others 1987,
Swanston 1991). LWD input varies considerably, depending on species of trees
growing alongside a stream, soil stability, valley form, climate, and lateral channel
-------
Clarke and Bryce--126
mobility (Bisson and others 1987, Maser and Sedell 1994, Sedell and others 1988).
Inputs may be either frequent chronic inputs or intermittent, sizeable inputs (Maser and
Sedell 1994) depending on slope, climate, soil, and bedrock type. Pool-forming
agents are influenced by gradient, vegetation, and geology. For example, in high
gradient streams, pools form around structural features; and in low gradient systems,
most pools are found in meander bends.
Habitat complexity enhances diversity of species and age-classes and translates into a
more resilient ecosystem (Franklin 1992, Hawkins and others 1993, Sullivan and others
1987, Vannote and others 1980). Successional patterns of fish assemblages in
streams are usually spatial rather than temporal. Specialization for temporal stages of
ecological succession are not observed (e.g., pioneering species); periodic
disturbances such as drought or floods cause species to relocate (Marcot and others
1994, Margalef 1960). A complex channel provides the needed refuges for such
relocation (Sedell and others 1990). Floodplains and other seasonally wetted areas
provide access for aquatic organisms to off-channel habitats during high flows
(Sullivan and others 1987). Anchored driftwood and standing vegetation provide quiet
water refuges during floods (Maser and Sedell 1994, Sedell and others 1988).
Lichatowich and others (1995) identify the distribution and quality of salmon habitats in
a watershed as one of the three elements important to the life history-habitat
relationship of individual stocks. Life-history diversity dampens the risk of extinction or
reduced production in fluctuating environments (Lichatowich and others 1995).
-------
Clarke and Bryce-127
Habitat complexity can be organized hierarchically. Gregory and others (1991) and
Frissell and others (1986) use the terms reach, channel unit (pool/riffle), and subunit
(microhabitat) to describe the hierarchical levels. Habitat complexity is a function of
diversity within and between levels. Gregory and others (1991) define reaches as
sequences of channel units with distinct hydraulic and geomorphic structures reflecting
different processes of formation. They describe a reach by the type and degree of
local constraint imposed on the channel and valley floor. Constrained reaches tend to
have relatively straight, single channels; unconstrained reaches in natural systems are
characterized by complex commonly braided or meandering channels and extensive
floodplains. Low gradient, wide valley bottoms, large woody debris, and poor
drainage increase the occurrence of braided channels. In larger rivers where LWD
does not span the channel, debris accumulations along the bank cause meander
cutoffs and may create well-developed braided channels (Bisson and others 1987). A
meandering stream provides diverse flows and increases the probability that the
spawning and rearing needs of different life history stages of fish will be met (Heede
and Rinne 1990).
At a finer scale, the mix of channel units (e.g., pools, riffles, and glides) contributes to
habitat complexity. Channel units are differentiated on the basis of water-surface
slope, width:depth ratio of the channel, and extent of turbulent, high-velocity flow
(Gregory and others 1991). Channel-unit complexity provides favorable water-quality
conditions, food supply, resting and hiding habitat within swimming distance (Heede
and Rinne 1990). Pools provide a feeding area where little effort may be needed to
-------
Clarke and Bryce-128
hold position against the current (Bisson and others 1987, Maser and Sedell 1994).
Migrating adults and juvenile salmon are dependent on pools for shelter from
predators and refuge from low summer flows and high winter flows (Petersen and
others 1992). Deep pools provide a variety of micro-habitats that allow different
species of fish and/or fish of the same species but of different ages to coexist (Maser
and Sedell 1994).
LWD plays an integral role in creating habitat complexity. Debris creates complex pool
types such as dammed pools, plunge pools, lateral scour pools, and backwater or
eddy pools (Bisson and others 1987, Bisson and others 1992). Local reductions in
stream flow caused by LWD provide foraging sites for fish (Sedell and Beschta 1991).
LWD decreases the erosive effects of flood water, including reinforcing meanders
(Beschta 1991, Naiman and others 1992, Petersen and others 1992). In addition, LWD
alters the stream profile and reduces local gradient (Franklin 1992, Swanston 1991).
Gregory and others (1991) define subunits as local hydraulic features created by
boulders, logs, or gravel bars at scales less than the active channel width. They
describe subunits as transitory features over annual hydrological cycles, changing
rapidly with rising or falling water levels. Subunit complexity is provided by structures
such as wood, boulders, and gravel bars. Diversity within subunits provides cover as
protection from predators, competitors, or variation in streamflow. Cover is provided
by LWD, overhanging vegetation, rubble, boulders, undercut banks, and water depth
(Sullivan and others 1987). Low velocity areas downstream of boulders or other flow
-------
Clarke and Bryce--129
obstructions constitute resting, feeding, and spawning habitat (Heede and Rinne
1990). Stored gravel behind debris may provide excellent spawning habitat (Swanston
1991).
Water quantity—The amount, timing, intensity, and type of precipitation are influenced
on a broad spatial scale by climatic processes, modified locally by topography and
vegetation. Precipitation can be either stored as snow or immediately discharged into
the stream as overland or subsurface flow. Infiltration rates, storage capacity, and
transmission rates are related to soil characteristics (Swanston 1991). The major
factors controlling stream flow are channel gradient, watershed size, type and density
of vegetation cover, precipitation characteristics, topography, and soil infiltration rates.
The amount and timing of stream flows affect fish populations and all physical stream
characteristics. Water quantity influences stream temperature, erosion processes,
weathering rates, hillslope and channel sediment transport and deposition, channel
morphology, current velocity, and riparian vegetation. The quantity and duration of
stream flows are important to migrating salmon (Bjornn and Reiser 1991) and for the
stability of spawning gravel (Petersen and others 1992).
Riparian vegetation—The occurrence and type of riparian vegetation are mainly
controlled by channel geomorphology, the spatial position of the channel in the
drainage network, soil, and climate (e.g., hydrologic regime) (Naiman and others
1992). Riparian vegetation contributes terrestrial invertebrates to streams as a food
-------
Clarke and Bryce-130
source for fish and organic matter as an energy base for stream biota (Beschta 1991,
Gregory and others 1987). The indirect role of riparian vegetation in providing fish
habitat is extensive. In summer, riparian vegetation provides shade to streams and
lowers summer temperatures; and in winter, riparian vegetation moderates thermal
temperature losses from streams and retards the formation of ice (Beschta 1991,
Franklin 1992). Riparian vegetation reduces flow velocities, which allows for deposition
of sediments and contributes to the long-term accretion of alluvium on floodplains
(Beschta 1991). The root systems of riparian vegetation help to stabilize the banks
during high flows (Beschta 1991, Franklin 1992, Sedell and Beschta 1991). Riparian
vegetation is the main source for LWD, contributing to habitat complexity by affecting
the dissipation of stream energy and creating local channel scour and deposition
(Beschta 1991). Riparian vegetation slows the flow of water allowing it time to infiltrate
into the bank. This infiltration helps to decrease peak flows, maintain local water
tables and extend base flows through summer months (Wissmar and Swanson 1990).
Biota—The important biotic components of salmonid habitat are inter- and intra-
species interaction and food sources. Species interactions are largely a function of
the other physical components that have been discussed such as habitat complexity.
Salmonids are opportunistic feeders that eat aquatic and terrestrial invertebrates.
Food abundance along with physical habitat and salmonjd behavior interact to
determine the carrying capacity of a stream (Murphy and Meehan 1991). The relative
importance of food is seasonally dependent. Food is more important in summer than
cover, and in winter the opposite is true, although exceptions are numerous (Murphy
-------
Clarke and Bryce-131
and Meehan 1991).
Current velocity, temperature, substrate, vegetation, and dissolved substances
influence the abundance and distribution of aquatic invertebrates. Of lesser
importance are susceptibility to drought and floods, species competition, shade and
zoogeography (Hynes 1972). These factors are influenced by landscape
characteristics
Energy for food is available to a stream from two types of sources: autochthonous
(photosynthesis within the stream) and allochthonous (decomposition of terrestrial
organic matter) (Murphy and Meehan 1991). The three basic autochthonous forms
are phytoplankton, periphyton, and vascular macrophytes. Each form characterizes
streams of different sizes, gradient, and exposure to sunlight. The five main
allochthonous inputs are: (1) streamside litterfall, (2) groundwater seepage, (3) soil
erosion, (4) fluvial transport from upstream, and (5) animal activities (Murphy and
Meehan 1991). Movement of organic matter depends on stream flow, particle size,
and retentive capacity of the channel.
Water quality-In natural streams there is a wide variability in water quality due to the
large number of controlling factors. Fish production can be altered by activities that
affect water quality and water quantity or regimen (Meehan 1991). Naiman and others
(1992) focused on five elements of water quality in the Pacific Northwest Coastal
Ecoregion; nitrogen, phosphorus, turbidity, temperature, and intragravel dissolved
-------
Clarke and Bryce--132
oxygen (DO). Landscape factors are important in controlling water quality in Pacific
Northwest Coastal Ecoregion streams (table 4).
Both nitrogen and phosphorus are important elements in the food chain (Naiman and
others 1992). Turbidity, usually caused by suspended silt and clay particles, has wide-
ranging effects on salmonids, invertebrates, and other aquatic organisms. Intragravel
DO is reduced by turbidity and sedimentation and is critical for salmonid reproduction,
invertebrates, and other aquatic life.
Seasonal and diel stream temperature influences: (1) trophic structure and
composition, (2) habitat selection, and (3) fish metabolism, development, and activity
(Beschta and others 1987). The influence of temperature on habitat selection was
demonstrated in Oregon by Bond and others (1988) who found that when looking at
summer stream temperature, significantly distinct habitat-use patterns appeared for 17
out of the 25 native species analyzed. Stream temperature is very important for bull
trout; they require very cold water for most of their life history. Henjum and others
(1994) found that published temperature tolerances for bull trout are much lower than
for other salmonids 4 to 10°C (39 to 50°F) and egg incubation 1 to 6°C (34 to 43°F).
Effects of higher stream temperature can be both positive and negative, although the
positive effects have only been indicated at the reach level in fairly cold systems.5
Warm temperatures: (1) increase the potential for competition with warm water
species, (2) reduce rearing area availability, and (3) increase susceptibility to disease
(Henjum and others 1994, Rhodes and others 1994). For example, increased
-------
Clarke and Bryce-133
exposure to sunlight can stimulate algal growth which may provide more food for fish;
but if the temperature exceeds the range of efficient metabolism, growth rates are
reduced (Petersen and others 1992).
Stream temperature may also impact substrate, indirectly. Stream temperature is
inversely related to water viscosity. Cooler, denser water can hold more sediment in
transport and may reduce the fine sediment loading of gravel used for spawning
(Heede and Rinne 1990).
Beschta and others (1987) describe the interaction between landscape characteristics
and stream temperature. The routing of water flow (surface versus subsurface)
impacts the temperature of water entering a channel. Subsurface flow reflects the
temperature of the watershed's subsoil environment. In the stream, water temperature
changes because of net radiation, evaporation, convection, conduction, and advection.
Net radiation is the solar radiation that is absorbed by a stream surface. This is
affected by topographic and vegetative shading and cloud cover. Albedo of the
stream may also influence net radiation; a clear stream is a black body, and a turbid
stream reflects solar radiation (Logan and Lammers 1966).
Heat gain or loss from evaporation depends on the vapor pressure gradient between
the water surface and the air immediately above the surface. Convection, internal
movement within a fluid because of differences in density or temperature, increases
with increase in temperature gradient between the water surface and the air
-------
Clarke and Bryce--134
immediately above the surface. Wind speed at the air-water interface is also an
important factor. Conduction of heat between the water in the stream and the
streambed depends on the type and color of material that makes up the bed. A dark
basalt absorbs more heat than a light granite. Bedrock channels are more efficient
than gravel-bed channels at conducting heat.
The importance of groundwater to salmonids is also being recognized. Advection is
the result of heat exchanges as tributaries or groundwater of different temperature
mixes with the main streamflow. Areas of upwelling groundwater and aggraded
floodplains were historically key production areas for salmon. Bedrock outcrops and
encroaching canyon walls are often locations for groundwater discharge (Stanford and
Ward 1992). Springs may also provide important cool water refugia (Bilby 1984).
Channel characteristics and morphology also influence the amount of heat gain or loss
of a stream. The surface area over which energy transfers take place is important;
wide, shallow streams receive more solar energy than narrow, deep ones. Discharge
is also significant in that for the same surface area and energy input, the temperature
change expected of a high-discharge stream will be less than a low discharge stream.
Delineation of Landscape-Level Ecoregions
Data sources-The study area includes the Grande Ronde, Asotin, Tucannon, Imnaha,
Walla Walla, Umatilla, Middle Fork, and North Fork John Day subbasins and the upper
reaches of the mainstem John Day subbasin (fig. 5). Primary types of mapped
-------
Clarke and Bryce--135
information used to delineate landscape-level ecoregions were soil, historical and
present-day vegetation, geology, topography, and climate. Because the quality of the
landscape-level ecoregion map is partly a reflection of the quality of the source
material, a detailed description of each source along with an assessment of its quality
and its utility to the delineation process is provided in the discussion of each
landscape characteristic. Table 5 summarizes the source maps used in the
delineation process.
Data integration—Ecoregion lines were drawn using a synthesis of digital layers, maps,
and descriptive information. We used topographic maps as a base because: (1)
frequently the other landscape characteristics follow topography and (2) these maps
are readily available and meet National Map Accuracy Standards (GS 1989). A GIS
was useful for easily combining map units into groups that we felt were important for
stream habitat and plotting maps at our working scale of 1:100,000. For each
available digital layer, 1:100,000-scale color plots were made to overlay the 1:100,000
scale topographic quadrangles. Lines from each plot were color-coded by landscape
characteristic and transferred to clear mylar. Each topographic quadrangle had a
mylar sheet for vegetation, geology, and soil. A blank sheet of mylar was used to
sketch ecoregion lines.
Topographic maps along with the mylar overlays of soil, geology, and vegetation were
overlain singly and in combinations on a light table. Placement of ecoregion lines was
based upon knowledge of how landscape characteristics might influence stream
-------
Clarke and Bryce-136
habitat. Location of landscape characteristics on nondigital maps of differing scales
and written descriptions were visually estimated and this information was also used in
the delineation process. Greater emphasis was placed on maps of higher quality and
better resolution. Lines were not determined by overlaying source maps within a GIS
because of: (1) lack of necessary information in GIS format; (2) differences in quality,
scale, and resolution of the source maps; (3) use of nondigital maps; (4) use of
descriptive information; (5) spatial differences in importance of the properties of each
of the landscape characteristics; and (6) differences in importance of each of the
landscape properties in influencing stream habitat. Further discussion of the
methodology can be found in Bryce and Clarke (1996), Clarke and others (1991),
Gallant and others (1989), and Omernik (1987, 1995).
After delineating approximate boundaries for landscape-level ecoregions from all
available map and textual sources, we evaluated the boundaries. The first set of draft
lines were transferred to a plastic-coated 1:100,000 scale topographic map for use in
the field. Approximately 3 weeks of road surveys were conducted to evaluate the lines
and questionable areas. Slides and a video were obtained for later use in the lab. In
addition, local biologists, geologists, and soil scientists were contacted to assist in
making boundary determinations in specific areas.
Map production—As a result of this field evaluation, some lines were redefined on the
original mylar copy, and all lines were digitized into an ARC/INFO coverage. After
digitizing, plots were made for each of the 1:100,000 scale topographic maps. Lines
-------
Clarke and Bryce--137
were reassessed by comparing them to the individual maps of landscape
characteristics and checked for accuracy of labels and line digitizing. As a result of
this step, changes were made to the digital map.
A map of landscape-level ecoregions was produced for seven contiguous watersheds
within the Blue Mountain and Columbia Plateau ecoregions (fig. 6). The map depicted
three levels in the ecoregion hierarchy—ecoregions, subregions, and landscape-level
ecoregions.
Each landscape-level ecoregion differed from the adjoining region in one or more of
four landscape characteristics: topography, soil, vegetation, or geology. A short,
descriptive name was developed for each landscape-level ecoregion based on the
defining landscape characteristics. Every map unit was color-coded and given a label
corresponding to the name in the legend. For example, pag refers to a unit named
ponderosa pine argillite. Our smallest map unit was 0.5 square kilometer (.2 sq mi),
and because landscape-level ecoregions were sometimes discontinuous, the smallest
landscape-level ecoregion was 2.6 square kilometers (1 sq mi). In describing each of
the landscape characteristics, this label identifies the map unit associated with that
characteristic. For example, when describing the areas of argillite, the map units
containing ag may be either pag or fag; ponderosa pine argillite or true fir argillite,
respectively. Usually a landscape-level ecoregion occurred within one subregion; but
in specific cases, it may have may occurred within additional subregions as well. For
example, most of the regions comprised of true fir forests are located within the Mesic
-------
Clarke and Bryce--138
Forest Subregion. However, small disjunct patches of true fir also occur in lower-
elevation subregions. These areas were too small to be mapped as subregions but
were large enough to be defined at the landscape-level ecoregion scale.
Landscape characteristics—Individual landscape characteristics rarely (or perhaps
never) independently influence stream habitat, although dominance may vary spatially.
The type and magnitude of influence also varies spatially. This spatial diversity of
influence is why we defined ecoregions based upon multiple landscape characteristics.
However, the prediction of stream-habitat characteristics based upon the stream's
location within an ecoregion is impossible for several reasons. First, we lack the
scientific knowledge to relate an individual landscape characteristic type directly to a
particular stream habitat characteristic. Also, landscape characteristics generally are
mapped as discrete units, although in reality they occur as a continuum. Individual
properties may combine to moderate or increase an effect, usually not in a direct 1:1
relationship. Finally the characteristics of the land adjacent to the stream, in concert
with upslope characteristics and upstream influences, work to determine the
characteristics of the stream.
We make no pretense that by using landscape-level ecoregions we can somehow
predict the stream-habitat characteristics of an individual stream within any delineated
landscape-level ecoregion. However stream-habitat characteristics are influenced by
landscape characteristics and landscape-level ecoregions are delineated using
landscape characteristics, so it follows that stream-habitat characteristics may be
-------
Clarke and Bryce-139
related to landscape-level ecoregions. In an attempt to elucidate the connection
between landscape-level ecoregions and stream habitat, the relationships between
each of the landscape characteristics and stream habitat are discussed. Discussion of
geology/geomorphology and vegetation is by individual types and discussion of
climate and soil is by properties.
Climate—Climate is predominantly influenced by elevation, aspect, availability of
moisture, and the prevailing wind direction. In turn, climate greatly influences
vegetation, soil forming processes, and geomorphology. Every aspect of stream
habitat is directly or indirectly affected by climate.
Climate is a very important landscape characteristic; however, developing meaningful
classes for climate is difficult. As Daubenmire (1956) found out when trying to use
climate classifications to explain patterns of vegetation, "the major finding of the study,
[was that] none of the four universal classifications of climate tested has much
phytogeographical significance in the area under consideration [eastern Washington
and northern Idaho]." He suggested that because vegetation, soil, and climate are
correlated in this area, they should be used to recognize the same landscape units.
Because annual means of temperature and precipitation mask information on daily and
monthly variability that is important to understanding vegetation patterns and erosional
processes the few maps of annual precipitation that were available for the study area
did not weigh heavily in our delineation process. As Daubenmire (1956) points out,
"Ecologists have for some time been aware that annual means of temperature (and to
-------
Clarke and Bryce--140
a certain extent, precipitation also) are scarcely worth even the small amount of
trouble involved in their calculation,
For Oregon, an annual precipitation map developed by the Oregon Climate Service
using PRISM (Precipitation-elevation Regressions on Independent Slopes Model) was
available. This model attempts to correct for the lack of weather stations in
mountainous areas (Taylor 1994). PRISM uses precipitation data collected from
weather stations and a DEM to generate estimates of annual precipitation.
Precipitation data came from two primary sources, NCDC cooperative stations and
NRCS SNOTEL stations for the period 1961 to 1990. Two terrain grids were used with
2.5 and 5 minute latitude/longitude terrain elevations. The 5 min (approximately 6x8
km per grid cell) was better able to resolve orographic effects (Taylor 1994).
For Washington, we used several maps from the State of Washington, Department of
Natural Resources. A very coarse-scale precipitation map (Miller and others 1973)
was marginally useful; much more useful was a map showing rain-on-snow zones
(Brunengo 1991). These zones generally corresponded to elevation.
Climate more than any other factor controls the rate and nature of weathering of
parent material. In arid areas physical forces dominate weathering, decreasing the
size of particles with relatively little change in composition (Brady 1990). Precipitation
increases with elevation and changes physical form while temperature decreases. A
snow-dominated area is more likely to experience flooding because of either an early
-------
Clarke and Bryce-141
spring thaw or a rain-on-snow event. Increased precipitation directly effects the water
quantity. Water is the primary mechanism for transporting substances within and from
forested lands (Gill 1994). Erosion may be increased by higher precipitation, runoff, or
stream discharge. The erosional effects of higher precipitation may be mitigated by
increased vegetation and concurrent increased infiltration and storage capacity of
soils.
Topography has a major influence on climate in this area. The Rocky Mountains block
this region from the continental air masses moving from the east (Franklin and
Dryness 1988) and the Cascade Mountains partially block the maritime influence of the
Pacific Ocean. Precipitation falls mainly in the winter months, although local
thunderstorms are common in the summer. Most of the precipitation in winter falls as
snow in the higher elevations. Rain-on-snow is common in the lower elevations.
Summers are dry and hot, and the winters are comparatively mild for this latitude,
although not as mild as winters west of the Cascade Mountains. Climate in the Blue
Mountains is noticeably different from the climate in the Columbia Plateau.
The Columbia Plateau is dry; precipitation increases as elevation increases. In the
western part of the Columbia Plateau, this precipitation pattern forms a roughly
concentric circle (Franklin and Dryness 1988). The lowest elevations, closest to the
Columbia River receive about 10 to 23 centimeters (4 to 9 in) per year, increasing up
to 40 to 60 centimeters (16 to 24 in) in the foothills of the Blue Mountains. Prevailing
winds are from the west or southwest, which causes the western slopes of the
-------
Clarke and Bryce--142
Columbia Plateau to be influenced by the maritime air tunneling through the Columbia
River gorge. Eastern slopes are drier because they are in the rain shadow formed by
the Blue Mountains (Gentry 1991).
The diverse topography in the Blue Mountains influences local climate (Johnson and
Clausnitzer 1992). Generally cooler temperatures are found as elevation increases.
However, many valleys are colder than the lower slopes of the adjacent mountains
because of cold air drainage. The maritime-influenced portions of the Blue Mountains
have greater cloudiness, increased precipitation, higher relative humidity, and fewer
fluctuations in winter temperature (Johnson and Clausnitzer 1992). Most of the Blue
Mountains, except for the John Day basin, are influenced somewhat by the maritime
climate.
The John Day basin is influenced less by the maritime climate and is in the rainshadow
formed by the Cascade Mountains. Light precipitation, low relative humidity, rapid
evaporation, abundant sunshine, and wide temperature and precipitation fluctuations
characterize this region (Johnson and Clausnitzer 1992). Here the temperature in the
winter is much colder and precipitation ranges from 25 to 51 centimeters (10 to 20 in)
in the grasslands to 43 to 76 centimeters (17 to 30 in) in the higher forested elevations
(Dyksterhuis 1981).
Soil—The importance of soil characteristics to stream habitat is not completely
reflected in the mapping of landscape-level ecoregions. At the time of the mapping,
-------
readily useable soil information was lacking.
Clarke and Bryce--143
The State Soil Geographic Data Base (STATSGO) was obtained from the USDA Soil
Conservation Service (SCS)6 for Oregon and Washington (SCS 1993). This digital
map and database were compiled from detailed soil survey maps and data on
geology, topography, vegetation, climate, and Landsat images. On a STATSGO map,
each map unit contains up to 21 components for which there are attribute data, but
there is no visible distinction as to the geographic location of these separate
components within the map unit. However, the percentage of each component is
given, which was useful for our project. For each component, we looked at drainage,
surface texture, permeability, and depth to bedrock. Within each component, data are
provided for each soil layer. We looked at erodibility, percentage clay and organic
matter, and top-layer depth for the top layer of each component. With this
information, we derived a rough assessment of each map unit.
STATSGO data were too general for this project, and county and forest soil surveys
were too detailed. If the county and forest surveys had been compatible and available
digitally, we could have reclassified the data and made them more useful for our
project. The lack of digital versions with an accompanying database was problematic;
besides not allowing for reclassification, maps could not be produced at our working
scale of 1:100,000. We located soil polygons using many individual map sheets,
usually bound in a book, with descriptive information located on other pages. This
was an error-prone and laborious method that was used mainly in the lower elevation
-------
Clarke and Bryce-144
agricultural areas, where soils played a more dominant role and there were questions
raised from the other data sources.
The National Forest Soil Resource Inventories (SRI) delineate soils on high-elevation
aerial photographs. For our purposes, the maps are too detailed. Each National
Forest has one or more SRI books with maps and soil descriptions. For each map
unit information is provided on plant community types, physiographic position,
geologic materials, aspect, slope, present erosion, major drainage dissection, ground
cover, overstory and understory vegetation, rock outcrops, estimated water holding
capacity, erosion, and hydrology.
County soil surveys delineate soil phases, which are generally subdivisions of a soil
series based on surface layer texture or slope. A soil series is defined as a group of
soils having horizons similar in differentiating characteristics and arrangements in the
soil profile (Soil Conservation Society of America 1982). Each soil phase is described
in the county soil survey. From these descriptions, we obtained information on depth
to bedrock, drainage, parent material, topographic position, potential natural
vegetation communities, land use, and ranges for precipitation, air temperature, slope,
and elevation. These maps also used aerial photographs as a mapping base. Most
of these photos have not been rectified which made input into a GIS a major task.
Although the science of soil classification is fairly sophisticated, the hierarchical
scheme was difficult to use. We found the most useful level of information was the soil
-------
Clarke and Bryce--145
series. However using soil series has several drawbacks: (1) the U.S. Forest
Service's SRI do not use the soil series names; (2) individual landscape-level
ecoregions were frequently comprised of several to many soil series; (3) soil series'
names are not common knowledge, (e.g., Athena soils would not convey deep loess
soils to most people unless familiar with the soils of the area); and (4) some of the soil
characteristics important to stream habitat (i.e., texture and slope) may not be
reflected until the next lower level in the soil classification hierarchy, the soil phase.
However, using soil descriptions, much useful information can be learned which may
influence stream habitat.
Ideally, we would like to have uniform digital coverage of soils at a scale of 1:100,000
or finer with an accompanying database reflecting soil characteristics for each
polygon. This type of soil data would have allowed us to aggregate polygons on the
basis of the characteristics we felt were most important in influencing stream-habitat
characteristics. Both the Natural Resource Conservation Service, NRCS, and the U.S.
Forest Service are heading in this direction. In addition to being difficult to work with,
the available information is difficult to discuss in terms of importance of the different
soil classes to stream habitat.
For these reasons, we chose to describe the properties of soil on the basis of
importance to stream habitat rather than describing individual soil types. Due to the
interrelationship of soil and vegetation, soil is discussed to some extent in the
vegetation section. For people familiar with the soils of the area, table 6 provides a list
-------
Clarke and Bryce--146
of the dominant soil series associated with most of the grassland map units. The soil
series names were listed only for grassland map units because soil was frequently a
dominant characteristic of these map units, and the National Forest SRI's do not use
soil series names. Interpretation of soil characteristics can be improved with the
availability of finer-scale uniform digital soil data.
The properties of soils that we thought to be the most important in influencing stream
habitat were: texture, depth-to-bedrock or other impermeable layer, drainage,
permeability, porosity, percent clay, percent coarse fragments, percent organic matter,
erodibility, and parent material. The effect of parent material on soil will also be
discussed in the section on geology. Harvey and others (1994) report a comparison
of a few physical properties for surface layers of ash-, sandstone-, and basalt-derived
soils in Eastern Oregon and Washington (table 7).
Texture, the relative proportion of sand, silt, and clay in a soil, is an index of the
percentage of voids and the amount of water that can be held. Drainage is measured
by the frequency and duration of periods when the soil is free from saturation with
water. Permeability is the ease with which gases, liquids, or plants roots penetrate or
pass though a layer of soil. In soils, permeability and porosity may be inversely
related. Many sands and gravels have a lower porosity than clay, good drainage and
aeration, but may be drought-prone. Silt is essentially microsand particles and usually
has an adhering film of clay. Silt soils possess some plasticity, cohesion and
absorptive capacity, but less than clay. Most clay, when wet, is sticky and plastic,
-------
Clarke and Bryce--147
which makes it cohesive. Clay soil is fine textured and has slow air and water
movement, giving it a high porosity and low permeability. Gravels have high
permeability, but the porosity differs based on the amount of sorting. Unsorted
gravels have low porosity and sorted gravels have high porosity. Gregory and Walling
(1973) report average ranges for porosity and permeability (table 8).
Clay because of its high cohesion and sand because of its high permeability is less
water erodible than soils with mostly silt. Because of low cohesion, sand is easily
eroded by wind; and on steep slopes, it is prone to dry ravel and erosion by sheetflow
where vegetation is minimal. High amounts of organic matter and coarse rock
fragments decrease the susceptibility of any soil to erosion.
Drainage reflects texture, depth to bedrock or other impermeable layer, and slope.
Poorly drained soil is commonly found in old lake basins, alluvial deposits, and
mountain meadows where the soil contains organic matter, clay and/or silt, and there
is little slope. In these areas, streams are slow and meandering with pool habitat in
the meander bends. Substrate is usually composed of fine materials. The water table
is generally high and water seeps slowly into the streams. Alluvial deposits can also
be excessively drained along with terraces and other areas with sandy or gravelly soil.
These areas usually contain few perennial streams.
Johnson and Clausnitzer (1992) classify soils in the Blue and Ochoco Mountains in the
following broad categories: (1) residual-derived in place from predominately bedrock
-------
Clarke and Bryce-148
or colluvial rock materials; (2) ash/loess-derived from deposited and accumulated ash
and/or loess over older buried soil material; and (3) mixed-derived from colluvium,
ash, and/or loess mixed well in surface layers over buried soil material. Residual soil
differs from volcanic ash and loessial soil in several respects; it has (1) finer textured in
the upper profile, (2) increased structure, (3) higher coarse fragments, (4) lower water-
holding capacity, and (5) higher bulk densities. Ash soils have (1) high water holding
capacity, (2) high water infiltration rates, (3) low compactability, (4) little particle
cohesiveness, and (5) disproportionately high amounts of nutrients in upper surface
layers. Loessial soil (1) is normally high in base saturation (can hold a large amount
of nutrients, (2) has high content of weathered minerals and is thus high in nutrient
reserve, and (3) generally has excellent physical properties. Productivity of plant
communities is closely related to the ash and loess content of the soil. Hall (1973a)
tabulated the stocking in trees per acre at 6 in average diameter at breast height with
10 rings per inch growth for mixed conifer with residual soil to be 275 to 330 versus
340 to 395 for mixed conifer with ash soil. Because southern and southwesterly
slopes were exposed to the prevailing wind in the Blue Mountains, the deposits of
loess were shallower. Also, the loess came from the north, so the south-facing slopes
were sheltered from the deposits. Under some conditions, ash soils support good
vegetation cover that protects the ash from erosion (Johnson and Clausnitzer 1992).
However, the development of soil in the thick ash deposits of the John Day formation
was hampered by lack of moisture. Lack of soil development and moisture increased
erodibility and hampered vegetative cover. Lack of vegetation in turn contributed to
increased erosion and lack of soil development. Lack of moisture seems to be the
-------
Clarke and Bryce-149
key factor in this positive feedback.
Vegetation-Vegetation is the most dynamic of all the landscape characteristics used
to define landscape-level ecoregions. Johnson and others (1994) state that "the
vegetation of eastside Washington and Oregon has a long history of natural
disturbance." Fire, grazing and browsing by ungulates, insect outbreaks and disease
epidemics, windthrow, flooding, and erosion have enhanced biodiversity.
Management-induced disturbances have modified the natural order in ways both
complimentary and detrimental to eastside ecosystems (Johnson and others 1994).
Because vegetation is dynamically influenced by both natural and anthropomorphic
disturbances, vegetation classifications and maps vary widely. An assessment of the
advantages and limitations of the existing vegetation classifications and maps to the
goal of the project was essential. The goal of this project was to define landscape-
level ecoregions as a tool to better understand the potential capacity of streams. To
utilize the best available information, we kept in mind the characteristics of vegetation
that may influence a stream's potential capacity. Ideally, we were interested in
obtaining a classification of vegetation that factored in the role of natural disturbance,
while acknowledging that it was extremely difficult or perhaps impossible to tease apart
natural from anthropomorphic disturbance, especially with the issues of fire and
grazing.
Vegetation directly influences stream habitat by providing shade and large woody
-------
Clarke and Bryce~150
debris and other organic matter to the stream. Vegetation influences erodibiiity, which
in turn influences substrate and habitat complexity. Vegetation also serves as a
climatic indicator, providing valuable clues to moisture and temperature regimes
(Woodward 1987). The amount and importance of precipitation, interception,
transpiration, infiltration, and runoff varies considerably between forest types (Gill
1994). Differences in soil, climate, and vegetation density all contribute to this
variation.
There is a general trend to increased tree stocking with increased elevation, peaking
with the true fir. Hall has recorded tree stocking ranges in the Blue Mountains (table
9).
The ponderosa pine/grassland areas have the lowest stocking, and the true fir has the
highest, although the range is quite variable. The vegetation density of an area affects
sensitivity to erosion, stream temperature, and snow melt. For western forests,
Franklin (1992) found that old-growth forests intercept a large portion of the snow.
The intercepted snow either melts and drips to the ground or is lost to sublimation and
evaporation. In cutover lands, deeper accumulations of snow occur and melting
occurs more rapidly because wind speed and turbulence at the snow surface are
higher. We speculated that the same principle may hold true for closed canopy
versus more open canopy forests, although snow depth would be less for the lower
elevation forests.
-------
Clarke and Bryce-151
Soil organic matter plays a role in soil-water availability, nutrient cycling, and erosion
control (Harvey and others 1994) In lower elevation vegetation zones, the nutrient
rich surface layer is usually washed away. Lower site productivity from reduced
nutrients coupled with a drier moisture regime results in a lower soil organic content.
Generally, soil organic content is highest in the moister zone of true fir and lower in the
drier zone of ponderosa pine. Buckhouse and Gaither (1982) found in a study of
sediment losses from ten natural ecosystems that meadow and forested ecosystems
were statistically similar and that losses for grassland, sagebrush, and juniper were
significantly higher than the meadow and forested ecosystem.
The vegetation information used to define landscape-level ecoregions was a synthesis
of available information. Each piece of information had its utility and its limitations.
The concept of potential natural vegetation (PNV) was the most useful for this project,
attempting to map vegetation as it might appear if humans were removed from the
landscape. However, the only available PNV map of the area was Kuchler's map of
the United States at 1:7,500,000 scale (GS 1970), which was too general for our
purpose.
The concept of plant associations was theoretically useful. If a stand of vegetation is
able to develop and persist in its environment, and if the competitive forces are without
major disturbing influences, then following a relatively long period of time those plants
capable of reproducing in competition will constitute the "climax community." The unit
of classification based on the probable, or projected, climax community type is defined
-------
Clarke and Bryce--152
as the "plant association" (Johnson and Clausnitzer 1992). Hall (1973a, 1973b),
Johnson and Clausnitzer (1992), and Johnson and Simon (1987) have produced
thorough handbooks on the plant associations of the Blue Mountains. Their
descriptions of plant associations were informative and provided some information
useful in making boundary decisions. However, very small-scale dot maps were
frequently the only spatial information found in plant association reports. These dot
maps gave approximate locations where certain plant associations were found. This
type of mapping was inadequate for our purposes. In addition, much of the work in
forested communities (Hall 1973a, 1973b; Johnson and Clausnitzer 1992; Johnson
and Simon 1987), was limited to Forest Service land and did not include private land.
Consequently, we chose to map the forested vegetation from historic forest maps
rather than use plant associations. We assumed that these historic maps reflected a
condition less-impacted than present day, especially in the higher-elevation forested
areas These maps also showed the results of disturbances such as fire, logging, and
grazing. Historic forest county maps were digitized by the Region VI, USDA Forest
Service (PNFES 1949, 1952, 1957). The original maps were at a scale of 1:63,360.
The counties and original date of mapping included:
County
Union
Umatilla
1958
1958
County
Asotin
Columbia
1935
1935
-------
Clarke and Bryce-153
Wallowa 1957 Garfield 1935
Grant 1960
Wheeler 1953
Crook 1952
Baker 1957
These county forest maps were very helpful. In addition to these digital maps, two
paper maps, Forest Type Map of Oregon and Forest Type Map of Washington at a
scale of approximately 1:250,000, were used (Andrews and Cowlin 1936a, 1936b;
PNFES 1936). Both of these surveys were a result of Section 9 of the McSweeney-
McNary Research Act of 1928, which called for a comprehensive and detailed
investigation of the existing timber resources by volume and area. The surveys used
all existing information on distribution of forest types and made type maps of all forest
areas in the region for which no usable data existed (PNFES 1936, 1949, 1952, 1957).
The digital forest maps by county were easier to use because they could be plotted at
our mapping scale of 1:100,000. We thought that they might contain more accurate
information being newer and more detailed than the Andrews and Cowlin maps.
However, the Andrews and Cowlin maps represented a less human-impacted
condition in the forested areas. In many places, logging was confined to the lower
elevations until the 1950s (Oliver and others 1994). Because of registration difficulties
with the 1936 maps, we adjusted some lines using elevation and aspect information
from the literature.
-------
Clarke and Bryce--154
Literature about the vegetation communities (Franklin and Dyrness 1988: Gannett
1902; Hall 1973a, 1973b; Johnson and others 1994; Johnson and Clausnitzer 1992;
Johnson and Simon 1987; Munger 1917) was used to resolve conflicts between the
two maps. For example, juniper has increased its range partly because of fire
suppression. This was very evident in comparing the 1930s and 1950s mapping. In
this case, we used the 1936 map to define the juniper areas. We knew that the
moister ponderosa-pine-dominated sites are often grand-fir-climax sites that historically
were maintained by frequent low intensity ground fires (Hall 1991, Lehmkuhl and
others 1994). Johnson and Clausnitzer's Plant Associations of the Blue and Ochoco
Mountains (1992) describes these areas as grand fir plant associations and provides
the early serai role of ponderosa pine in the section titled "Successional Relationships."
These areas were delineated as ponderosa pine on the landscape-level ecoregion
map because we intended our mapping to reflect the role of fire as a natural
disturbance. Johnson and others (1994) state that "throughout the presettlement
period, fire was an integral part of the maintenance and function for the majority of
eastside ecosystems"
The intent of the 1936 map was to map timber resources, therefore, the extent of
subalpine fir forests was unimportant. Subalpine fir was mapped with noncommercial
rocky areas on the 1936 maps. This mapping convention caused some difficulties for
us. Fortunately in the accompanying forest statistics for each county (Bolles 1937;
Buell 1937; Litchfield 1937; Moravets 1937; Pelto 1937; Sankela 1937a, 1937b; Wolfe
1937), they did not lump these two categories together. We used these numbers and
-------
Clarke and Bryce-155
the later forest mapping to evaluate and adjust our map delineations.
The Oregon Actual Vegetation digital map and manual (Kagan and Caicco 1992),
developed as part of the Oregon Gap Analysis Program, was marginally useful
because it represented actual conditions after decades of land use and vegetation
conversion Mainly it was used for additional information when large patches of larch,
Douglas-fir, lodgepole, or noncommercial rocky areas were in question. The map was
compiled by visually photo-interpreting Landsat MSS false-color infrared positive prints.
There were registration problems with this map, and it is being redone.
We defined five forest zones—Juniper, Ponderosa Pine, Douglas-Fir, True Fir, and
Subalpine Fir. We felt that by using vegetation zones instead of a finer-scale
classification, the variability caused by disturbance history was lessened. Table 10
lists the major species occurring in each forest zone and also gives an approximate
elevation and annual precipitation range for each zone. The primary sources for the
table and descriptions of each zone were Franklin and Dryness (1988), Hall (1973a,
1973b), Johnson and others (1994), Johnson and Clausnitzer (1992), and Johnson
and Simon (1987). Only Franklin and Dryness (1988) provide information on serai and
climax forest species composition. The table is intended only as a general guide to
species composition.
Vegetation in the study area is greatly influenced by the presence or absence of
marine influence. Elevation ranges given for most zones apply to the entire study
-------
Clarke and Bryce--156
area. Although zones will usually start higher in areas not maritime influenced. The
maritime-influenced zone subregion (fig. 4) indicates the extent of maritime influence.
Interfingenng between contiguous forest zones is also important in this area. Certain
zones extend further into dry climates by taking advantage of north-facing slopes and
stream margins and ascend farther on exposed slopes and ridges. The presence of
deep soil with high moisture holding capacity, such as formed from ash, also allows
vegetation to extend into areas with less precipitation.
Occupying the zone between the ponderosa pine and the grassland or sagebrush
zones was the western juniper zone (map units beginning with /). This zone only
occurred in the John Day basin of our study area, perhaps because of continental
influence. This is the driest "forested" zone with most of the precipitation falling during
the winter. Soil is typically light-colored, coarse-textured sandy loam, and low in
organic matter.
Ponderosa pine is climax vegetation on the warmest and driest forest sites and is a
major serai component in the Douglas-fir and grand fir series. Our ponderosa pine
zone contained both climax and serai ponderosa pine forests (map units beginning
with p). The thick bark of ponderosa pine allows it to withstand ground fires better
than the thin-barked true firs. Climax ponderosa pine forest occurs on coarse, sandy
soils and where cracks in the bedrock permit trees to tap underlying moisture. Soils
on these sites are usually dry at depths of 10 to 61 centimeters (4 to 24 in) for 60 or
more consecutive days during summer and autumn. Ponderosa pine forests grow on
-------
Clarke and Bryce--157
three distinctive types of soil. The ash soil is coarse, but it has a high moisture-
holding capacity. Organic matter is concentrated near the surface and declines rapidly
with depth. Soil derived from basalt, andesite, and clayey sediments is moderately
deep and dark-colored, fine and loamy. This soil is easily compacted and puddles
when wet. Surface erosion is a problem on slopes greater than 30 percent. The soils
most sensitive to erosion are derived from rhyolite, andesite, granitics, glacial till, and
outwash. These soils are coarse, loamy, and shallow-to-deep with low organic matter
content and low water-holding capacity. On cooler, moister sites the soil has more
organic matter.
A mixed ponderosa pine/grassland or ponderosa pine/shrubland zone was identified
(map units beginning with m) in areas at the lower edge of the ponderosa pine zone
where the two zones intermix. In this area, ponderosa pine is confined to a fringe of
trees along canyon sides, draws, and north-facing hillsides (Munger 1917). The
geographic extent of this area (about 3500 square kilometers [1351 sq mi]) called for
a new zone rather than dividing it between the ponderosa pine and grassland zones.
A separate shrubland zone was not identified. Species for the shrubland component
of the ponderosa pine/shrubland zone are found in table 11.
A Douglas-fir zone is identified by some authors (Hall 1973, Harvey and others 1994,
Johnson and others 1994, Johnson and Clausnitzer 1992, Johnson and Simon 1987).
Franklin and Dyrness (1988) say that its occurrence is conjectural except in parts of
the Wallowa Mountains. Because of its wide ecological aptitude, Douglas-fir occurs in
-------
Clarke and Bryce--158
both the ponderosa pine and true fir zone; and it was difficult to differentiate based on
the available maps. For this reason, we only delineated a Douglas-fir zone in the
Wallowa Mountains (map units beginning with d).
The upper elevational limit of the ponderosa pine gives way abruptly to a very different,
much denser stand of other species (Munger 1917) comprising the true fir zone (map
units beginning with t). This zone included both grand fir (found in the northeastern
part of the study area) and white fir (found further south). Franklin and Dryness
(1988) treat these areas similarly because "these two species and the zones they typify
occupy analogous positions synecologically and environmentally in their respective
areas." To denote both white fir and grand fir in the name, we used the term "true fir."
We recognized that subalpine fir is also a "true fir." However, because subalpine fir
occupies a higher elevation and colder environment, we mapped it separately from our
"true fir" category. The true fir zone is characterized by neither temperature nor
moisture extremes. It is wetter and cooler than the ponderosa pine zone and has
higher temperatures and less accumulation of snow than the subalpine fir zone. Soil is
usually moderately deep because of the accumulation of volcanic ash. These soils are
very fertile with rapid infiltration, high water-storage capacities, and good aeration.
The subalpine fir zone is the coolest and moistest forest zone with a deep winter
snowpack (map unit beginning with s). Subalpine fir forests may be found below this
elevational range in frost pockets and areas affected by cold air drainage, such as
glaciated valley bottoms. Soils are coarse and stony with well developed, relatively
-------
Clarke and Bryce--159
thin humus layers, and low fertility. They are erodible when exposed to wind. Our
subalpine fir zone contained a whitebark pine zone analogous to the juniper zone at
the lower forest fringe.
Information on vegetation in lower elevation, nonforested land was critically lacking for
this project. Historical forest maps only showed areas outside the forested areas as
"nonforested." The map of Actual Vegetation of Oregon (Kagan and Caicco 1992) was
of little help because much of this area is in agriculture now. Potential-vegetation
community descriptions from the county's soil surveys were helpful. Differentiation
between grass and sagebrush zones was sometimes difficult, especially in the
Pleistocene Lake Basin Subregion. Differentiation within these classes was not
possible with the mapped information available. However, many of these finer
vegetation classes are expected to follow other mapped landscape characteristics,
particularly differences in soil depth and moisture. Literature on nonforested
vegetation was obtained from Daubenmire (1988), Franklin and Dyrness (1988),
Johnson and Simon (1987), and county soil surveys.
We divided the nonforested vegetation into zones starting with the warmest and driest:
aridic grassland, sagebrush, grassland, and alpine. The choice of these classes was
based partly on the review of the scientific literature; greatly influencing the decisions
was the availability of spatial data. The distinction among rigid, low, and big
sagebrushes and between grassland communities dominated by fescue and those that
are not would have been valuable for determination of soil depth and moisture regime;
-------
Clarke and Bryce--160
however, maps showing these distinctions are not commonly available. Table 11
shows the dominant species occurring in these zones. The primary sources for the
table and descriptions of each zone were Daubenmire 1988; Dyksterhuis 1981;
Franklin and Dryness 1988; Hall 1973a, 1973b; Harrison and others 1962, 1973;
Hosier 1983; Johnson and others 1988; Johnson and others 1994; and Johnson and
Clausnitzer 1992. The table is intended only as a general guide to species
composition.
The driest zone was the aridic grassland which occurred mainly in glaciofluviate
deposits (map units beginning with g and containing go/t, es, gfg, aca, loda/h, and
gfs). The mean annual precipitation in this zone is usually less than 30 centimeters
per year (12 in per yr).
The next driest zone, which we called the sagebrush zone, consists dominantly of big
sagebrush and bluebunch-wheatgrass (map units beginning with s). Even in the driest
part of this zone, there is virtually no bare ground in the undisturbed climax
(Daubenmire 1988). Soil is mostly loam or stony loam. In the John Day basin, the
sagebrush zone frequently abuts the juniper or ponderosa pine forested zones,
possibly due to the influence of the colder, drier continental climate.
The grassland zone (map unit beginning with g) rises to an elevation of about 900
meters (2953 ft). This zone is wetter than the sagebrush zone. Most of the Columbia
Plateau and some basins within the Blue Mountains are dominated by grassland
-------
Clarke and Bryce--161
communities. This zone could be broken into the mesic grasslands where Idaho
fescue forms the climax communities and the xeric grasslands where Idaho fescue is
absent and bluebunch-wheatgrass dominates if maps to show this distinction were
available for the study area. Table 11 gives the species composition for these two
grassland zones. The wetter grasslands areas are found at upper canyon elevations
and on north-facing aspects at lower elevations and on deeper soils of the ridges
where moisture is retained longer into the summer drought period.
The alpine zone included the alpine sagebrush and green fescue in the high Wallowas
and alpine fescue communities in the Blue Mountains (map units containing a). The
elevation ranges from 1859 to 2499 meters (6100 to 8200 ft). This zone was large
enough to be mapped only in the Wallowa Mountains. On the granitic soils, rapid
drainage leads to low vegetative productivity making this zone more erosive; on lava
soils, there is more herbage, so erosion is less.
Geology and geomorphology-Geology and geomorphology play a strong role in
controlling stream habitat characteristics, especially substrate, habitat complexity, and
water quantity. Most geology maps are based upon rock type and age and mapping
units may incorporate rocks with vastly different engineering or hydrologic properties.
These groupings may be due to different mapping goals, units that are too complex to
map lithologies, and/or map scale and resolution. Engineering and hydrologic
properties such as joint structure, stratigraphy, erodibility, porosity, and permeability
were more important for our purposes making it necessary to interpret geology maps
-------
Clarke and Bryce--162
for relevance to fishery biology and stream ecology.
A 1:500,000 scale geology map (Walker and Macleod 1991) of the State of Oregon,
available in digital form, was useful. The map was originally produced at 1:250,000
and later reduced to 1:500,000. For the purposes of this project, the detail of the map
was adequate. However, we recognized that this map was not intended to be plotted
at the 1:100,000 scale of this project. Lacking other economically feasible options, we
used the map and made adjustments along the way. The most noticeable problem
was poor registration, especially seen in the alluvial areas. When working with the
geology layer, each 1:100,000 map sheet was manually reregistered to place the
alluvial areas next to rivers, requiring about a 1- to 2-centimeter (.39 to .78 in) shift in
both the x and y direction. A draft geology map of southeast Washington (Johnson
and Derkey 1993) was available in digital form although it was very coarse scale.
Many local 1:24,000- to 1:250,000-scale maps were used to adjust the lines and
resolve questionable areas.
Geologic mapping relies on topographic maps as their base (Weissenborn 1969),
because geology is not usually a surface phenomenon. We frequently used
topography to adjust our lines where it was evident that geology would follow a
topographic feature. We used many reports on geology (Baker and others 1991;
Barrash and others 1980; Bishop 1994; Bishop and others 1992; Brooks 1979; Ferns
1985; Gonthier and Bolke 1993; Hampton and Brown 1964; Hodge 1942; Hogenson
1964; Merriam 1901; Newcomb 1965; Oles and Enlows 1971; Orr and others 1992;
-------
Clarke and Bryce--163
Reidel and Hooper 1989; Smith and others 1941; Taubeneck 1957; Thayer 1990;
Walker 1990a, 1990b; Walker and Robinson 1990; Weissenborn 1969; Whiteman and
others 1994; WRB 1963; WRD 1988) to help us understand the processes and
potential topographic manifestations of the geology. In addition, these reports
frequently had small schematic maps and descriptive information which was invaluable
in delineating boundary lines and describing the geologic properties from which we
could infer potential importance to stream-habitat characteristics. Using existing maps
and literature, we have attempted to delineate and describe the characteristics of the
rocks or formations that potentially influence stream habitat.
The hydrologic properties of a geologic formation are not determined solely by the
permeability or porosity of a rock, but may reflect the joint structure (Gregory and
Walling 1973). Stratigraphy also plays a large role in the movement of water through
the bedrock. For example, basalt is not generally porous or permeable but it
commonly has a hexagonal joint pattern that transmits water vertically and the
Columbia River basalt frequently has extensive, lateral, permeable sedimentary
deposits or paleosols between the flows. Both jointing and intrabeds contribute to the
Columbia River basalts' importance in providing groundwater to streams.
Mineral composition strongly influences the weathering rate, soil productivity, and rate
of erosion. The stratigraphy is important, for example, where the resistant basalt rests
on top of highly erodible ash such as in the John Day formation. Focusing only on
the top layer of basalt might mask the true erosiveness of the area. Stratigraphy at
-------
Clarke and Bryce-164
any single location is impossible to determine from maps where only the top layer of
geology is mapped. The geomorphology of an area strongly influences the movement
of water and erosion. Discussion of geomorphology is linked to both geology and
topography.
The following section describes the major geologic types found in the study area and
summarizes their origin, spatial extent, and properties relevant to stream habitat. In
keeping with geologic convention, the units will be discussed in order of age. The
pre-Cenozoic rocks include the metasedimentary and metavolcanic rocks associated
with exotic island arcs, subduction and collision of the Pacific oceanic plate with the
North American continental plate, and plutonic intrusions. These rocks outcrop in only
about 10 to 15 percent of the Blue Mountains (Walker 1990b). The rocks of the
Cenozoic Era are divided into the Tertiary and Quaternary Periods. The Tertiary
Period was a time of major volcanic activity that formed the Clarno, John Day,
Columbia River Basalt, and Strawberry formations. The Quaternary was a time of
alpine glaciation and fluvial-generated alpine erosion with deposition of sediments in
basins that shaped the current landscape of the study area.
Pre-Cenozoic Era-The pre-Cenozoic rocks (320 to 65 myr ago) are a direct result of
plate tectonics. Brooks (1979) and Bishop (1994) discussed these pre-Cenozoic
deposits. Most of Oregon and Washington were once part of the Pacific Ocean. The
rocks exposed in Hells Canyon and in much of the higher elevations of Wallowas,
Elkhorns, and old mountains in the region were formed as volcanic islands.
-------
Clarke and Bryce--165
Sediments, including limestone, conglomerate, sandstones, and shales, accumulated
from erosion of this island system, eventually burying the old volcanic rocks. The
remnants of the old volcanic arc are called the Wallowa and Olds Ferry terranes,
approximately 300 to 200 million years in age. (A terrane is a coherent package of
rock built in one place through one kind of geologic process.) The subduction zone
and portions of oceanic crust trapped and deformed between volcanic arcs is called
the Baker terrane, and it is approximately 300 to 200 million years in age. The less-
deformed package of sedimentary rocks that overlie the Baker terrane is called the
Izee terrane, and it consists of shale and fine sandstone exposed in the Aldrich
Mountains. As the terranes collided with the North American plate, older deeply
buried and deformed volcanic rocks were heated and subjected to high pressure.
Greenstones and sediments melted and rose into the surface of the overriding plate
producing the granitic plutons that now form the core of the Wallowas and northern
part of the Elkhorn Mountains.
The Baker terrane consists of argillite, chert, conglomerate, limestone, greenstone,
sea-floor and island arc plutonic rock, high pressure schist and serpentinite from both
crustal and supracrustal rocks that have been severely broken up, rearranged, and
deformed. The crustal rocks are called a melange composed of ultramafic rocks,
gabbro, quartz diorite, and albite granite severely deformed by folding and faulting.
Melanges typically have a matrix of serpentinite, a scaly, slippery, green rock (map
units containing sp). The largest exposures of these rocks are found in the
Greenhorn, Elkhorn, Strawberry, and Aldrich Mountains and the Ochoco Mountains
-------
Clarke and Bryce--166
near Antone. The presence of serpentinite may increase the risk of landslides due to
its slippery nature, lack of soil nutrients, and low vegetative productivity.
The Canyon Mountain complex in the Strawberry Mountains is considered island arc-
related, not derived from ocean crust. In its uppermost sections, it consists mainly of
gabbro, the coarse-grained equivalent of basalt. Gabbro weathers and erodes more
rapidly than granitic rocks (Chesterman 1979) and like serpentine produces clays that
are expandable, thus producing slide-prone soils.
Other lithologies common in the chaotic Baker terrane include ribbon chert and
siliceous argillite—hard, silica-rich layers separated by thin, dark, fine-grained layers of
shale (map units containing ag). This shale component makes the argillite erodible.
These rocks are a major part of the Elkhorn and Greenhorn Mountains.
The sedimentary rocks in the Aldrich Mountains are composed of thick deposits of
shales and very fine sandstones similar to the Elkhorn Ridge argillite except less
metamorphosed—a composition that makes them more crumbly. These rocks are
part of the Izee Terrane.
The Wallowa-Seven Devils Terrane consists of volcanic, plutonic, and sedimentary
rocks that have been somewhat metamorphosed. The volcanic rocks were originally
basalt, andesite, and dacite slightly metamorphosed to greenstone (map unit
containing vm). The sedimentary rocks include shales and sandstone of the Hurwal
-------
Clarke and Bryce--167
formation and Lower sedimentary series as well as Martin Bridge Limestone
composed of limestone deposited in shallow water and the Hurwal formation made up
mostly of shale with some sandstone and a few conglomerates (map units containing
svm and se). These rocks were best exposed in the northern Wallowa Mountains.
Large bodies of intrusive rocks called batholiths intruded the Baker and Wallowa-
Seven Devils terranes. These rocks have the appearance of granite, but because they
are rich in feldspar and contain little quartz, they are technically not true "granite." The
name "granitic" or "granitoid" is more correct. The largest, the Wallowa batholith, is
found in the core of the Wallowa Mountains. Other granitic intrusions include Bald
Mountain and Battle Mountain and an unnamed stock along the John Day River near
the Ritter Hot Springs. The intrusive rocks range from gabbro to granite, but feldspar-
biotite-rich lithologies (tonalite and granodiorite) form the vast majority of the exposure.
The intrusions formed contact halos where the surrounding rocks have been strongly
foliated and metamorphosed (Weissenborn 1969). Erosion later stripped off most of
the sedimentary rocks covering these intrusive rocks (Smith and others 1941). The
surrounding sedimentary and volcanic rocks have been metamorphosed to varying
degrees, which made it very difficult to generalize about their properties. These rocks
are more resistant to erosion than they would be if unmetamorphosed. However,
there is a wide range of erodibility for metamorphosed rocks depending upon degree
of metamorphism and origin. Usually the metamorphic volcanic rocks (i.e.,
greenstone, metamorphosed basalt or gabbro) are resistant to erosion, while the
metamorphosed sedimentary rocks (i.e., slate, metamorphosed shale) are not. The
-------
Clarke and Bryce--168
rocks classified as partly metamorphosed-sedimentary and volcanic rocks challenge
even these generalizations. We colored the coded units containing metamorphic
rocks similarly within vegetation zones but labeled them based upon the type of rock
that was metamorphosed, i.e., svm for sedimentary and volcanic rocks partly
metamorphosed, sm for sedimentary rocks partly metamorphosed, and vm for
volcanic rocks partly metamorphosed.
Granitic outcrops are generally bold and well rounded (map units containing in). The
rock is massive, tight and poorly permeable, and weathers to a coarse sandy rock
waste (Hampton and Brown 1964). The abundance of minerals susceptible to
chemical weathering (calcium plagioclase feldspars) and mechanical disaggregation
(biotite) creates granular, friable, easily erodible soil and deep "grus." Megahan (1972)
and Megahan and others (1992) have done several studies on the granitic rocks of the
related Idaho batholith. They note that granitic soils in the western United States are
noted for their high erodibility because of their relatively coarse texture, lack of
cohesion, and occurrence on steep slopes. Their coarse texture accounts for the fact
that most of their sediment loads are carried as bedload. Most of the streams in
granitic areas are clear, only becoming slightly turbid during peak flows.
Cenozoic Era-Tertiary Period-Significant volcanism occurred during the Tertiary
Period. The Cfamo formation, approximately 50 to 35 million years in age, is a very
thick sequence of andesite flows, mudflows, breccias, andesitic to rhyolitic
volcaniclastic rocks, and tuffaceous sedimentary rocks. The Clarno formation is part
-------
Clarke and Bryce--169
of map units containing cl except the tuffaceous facies that is combined with other
map units containing tuff—tu. The formation occupies a large area of more than 4000
square kilometers (2485 sq mi) to the south of the Blue Mountain uplift and near the
western margin of the Blue Mountains ecoregion. A smaller area of about 400 square
kilometers (248 sq mi) is found along the axis of the Blue Mountain uplift in southern
Morrow county (Walker and Robinson 1990). The Clarno formation's bimodal
geographic distribution, as well as its complexity and diversity of ages suggests that
several different volcanic periods may be represented.2 Rock types with different rates
of weathering contribute to the unique terrain found in the Clarno formation. The
volcaniclastic, sedimentary rocks in the Clarno formation show a much higher degree
of induration than the younger, ash-rich, overlying John Day formation. Clarno
mudflow (lahar) deposits frequently form steep bluffs ornamented frequently with
balanced rocks (Merriam 1901). Topography developed on Clarno mudflows yields
large, rough inclined surfaces, hogbacks, knobby hills, buttes, and V-shaped canyons.
Resistant lavas are peeled off in layers because of the alternation of resistant lavas
and yielding sediments. Although Hodge (1942) said that the Clarno formation is very
resistant to erosion, some areas mapped as Clarno include andesite flows separated
by tuff beds. Such areas are susceptible to landsliding. Because of the diversity of
rock types, any generalization about the stability of the Clarno formation is probably
inaccurate. In the headwaters of the Grande Ronde River, there is a large area
mapped as andesite, dacite, and sedimentary rocks by Walker and MacLeod (1991)
and Ferns and Taubeneck (1994). We mapped this unit with the Clarno formation due
to similarity in rock types.
-------
Clarke and Bryce--170
The John Day formation was laid down upon the eroded surface of the Clarno
formation (map units containing tu). The John Day formation is very distinctive, and
much has been written about it. It is an assemblage of tuffaceous sedimentary rocks,
air fall and ash-flow tuff and sparse olivine-rich basalt lava flows (Robinson and others
1990). The John Day formation is restricted to the lowland areas of the John Day
basin (Thayer 1990) and varies in depth from a few feet to more than 610 meters
(2000 ft).
Landslides are common along the steep stream-cut slopes incised into the John Day
formation and some are immense. Water flows through the heavy basalt that caps the
John Day formation and soaks the tuffs below causing the basalt to slip, creep and
landslide (Hodge 1942). Some of the largest landslides were mapped in map units
containing Is. Except in basins, soil does not form on the John Day formation
because it is so pervious to groundwater and blows away easily. Elephant-backs,
gullied slopes, pinnacles, and badlands comprise the topography in these areas
(Hodge 1942).
There were several other pockets of primarily tuffaceous sedimentary rocks in the
study area. Although they may be different ages, structurally we considered them to
be similar and have grouped them into map units containing tu. In the Grande Ronde
basin, Hampton and Brown (1964) report that some tuffs contribute to the rich soils for
which the upland district is famous. They note that the tuff weathers more rapidly than
the basalt and occupies topographic depressions bordered by steep erosional
-------
Clarke and Bryce-171
escarpments.
The Columbia River Basalt Group (CRB) covers most of the Columbia Plateau and is
the most widespread geologic type in the study area. It consists of layer upon layer of
basalt flows which may be interspersed with extensive thin layers of sedimentary
deposits or paleosols. The sedimentary layers and soil were deposited in the time
between flows. In some places, these flows are more than 1524-meters (5000-ft) thick
(Orr and others 1992).
The Columbia River Basalt Group (map units containing cb) is stratigraphically
subdivided into the Saddle Mountain, Wanapum, Grande Ronde, Picture Gorge, and
Imnaha Basalts (Walker and MacLeod 1991). Over 90 percent of the CRB are the
Grande Ronde basalts, erupted between 15.5 and 19.5 million years ago. They are
generally crystal-poor, silica-rich, fine-grained basalt (Reidel and Hooper 1989).
Grande Ronde basalts are about 610 meters (2,000 ft) thick in the Grande Ronde
valley (Bishop and others 1992). The Picture Gorge Basalt with high magnesium and
rapid weathering is exposed in most of the John Day watershed. Because the Picture
Gorge basalt tends to be less resistant to erosion and has a significant exposure in
the study area, we delineated it as a separate category (map unit containing pg).
Future work may find the difference in erosion to be negligible when compared to
other geologic types, and it may prove meaningless to stream habitat.
Many basalt flows displayed a characteristic sequence of layers. The lower section
-------
Clarke and Bryce--172
was characterized by huge vertical six-sided prisms. These prisms have a large
diameter at the base and become smaller near the top. They break at right angles,
and breaks are more common near the tops of the flows. This characteristic fracture
pattern produces large talus slopes of "brickbat" basalt. The tops of each flow are
usually covered by a mantle of vesicular-to-scoriacious lava. The basalt is usually fine
and even grained.
Basalt is not porous, but the porous and permeable tops of some flows, joint patterns,
and incomplete closures of one flow over another make the Columbia River Basalt
Group a very important aquifer (Hampton and Brown 1964, Hogenson 1964,
Newcomb 1965, WRB 1963). Streams without sources outside the Columbia Plateau
only flow in the summer where springs emerge from the basalt; usually along axis of
synclines (Newcomb 1969). During low flow periods in the Upper Grande Ronde River
basin, creeks draining the older metamorphic and intrusive rocks of the Elkhorn Range
are dry, but creeks maintain a small, constant flow where the Columbia River Basalt
overlies these older deposits (Hampton and Brown 1964).
In northern Wallowa and Union counties, there are some sedimentary beds
interbedded with the Columbia River Basalt Group. These deposits were formed in
shallow lakes and peat bogs that formed on tops of the flows (Ferns 1985). These
deposits can be as much as 91-meters (300-ft) thick. They crop out at only one very
limited and remote location. The presence of these lacustrine deposits beneath the
basalt may make slopes more susceptible to landslides. For the time being, we have
-------
mapped them separately as map units containing bl.
Clarke and Bryce--173
Bowen's reaction series shows the sequence in the crystallization of a basaltic melt
(Foster 1975). Olivine is the first mineral to crystallize as the temperature falls; quartz
is the last. Conversely, olivine is the first to break down under chemical weathering
which contributes to unstable clays. This helps explain the differences in volcanic
rocks and associated soil. Basalt generally contains olivine; rhyolite contains quartz
and is the extrusive equivalent of granite. Andesite falls between. Quartz is very
resistant to weathering; olivine is not. So the minerals forming basalt are more
erodible than the minerals forming rhyolite. Generalizations are difficult because the
jointing structure plays a significant part in determining erodibility. Only a small portion
of the study area in the southwestern corner contains significant amounts of rhyolitic
rock (map units containing rh). Significant amounts of andesite are found in the study
area.
The Strawberry Mountain andesite (map units containing an) is a product of volcanoes
rather than the basalt floods that produced the Columbia River and Picture Gorge
Basalt Groups. This platy andesite does not have the well developed vertical joint
system of basalt. It has random vertical joints, not well developed, and thin,
discontinuous horizontal joints instead. Because of this, the andesite is relatively
impermeable and does not have the aquifer properties of the basalt (Hampton and
Brown 1964). The platy structure of this andesite may make it more erodible than the
surrounding basalt.7,8
-------
Clarke and Bryce-174
Concurrent with the volcanism, tectonism played a major role in the formation and
deposits of the basins in the area. Two types of basins were formed. The earliest
basins were a result of synclinal downwarps in flow sequences of the Columbia River
Basalt Group i.e., the Agency, Arlington, John Day, and Fox basins (Walker 1990a).
The Ukiah basin is mapped by Walker (1990a) but not discussed and was probably
formed by faulting as well as folding2 in a structural depression. The southern extent
of this basin was difficult to map. Soil and topographic maps showed the southern
part to be similar to the area around Ukiah. However, the geology maps do not affirm
this. The Arlington, Agency, and Ukiah basins contain partially cemented gravel and
interbedded tuffaceous sand and silt (map units containing ca). The margins of the
Ukiah basin are mapped in units containing rca. They have a thinner veneer of loess
and include some residuum. The lower elevations of the Arlington basin and the Walla
Walla basin are overlaid by thick layers of loess and glaciofluviate deposits, which will
be discussed in the Cenozoic Era-Quaternary Period section. In the John Day basin,
the present Strawberry and Aldrich Mountains were raised 2 to 3 kilometers (1.5 to 2
mi) above the valley floor by faulting along the John Day fault system. Clastic rocks
and tuffs from the Strawberry Mountain volcanoes washed out over the basalt and
produced the Mascall formation. Gravelly, sedimentary rocks from the rising
mountains filled the John Day valley producing the distinctive Rattlesnake deposits
(Thayer 1990, Walker 1990a). A prominent bed of welded tuff or ignimbrite is present
in the upper part of the Rattlesnake formation (map unit containing fs). The Fox basin
is a structural trough confined by faults and a syncline. It contains rhyolitic tuff and
pebble gravels in the northwest and thin basalt flows faulted with tuffaceous beds in
-------
Clarke and Bryce--175
the south (map units containing tu).
The more persistent basins in this area are a result of faulting. The Yakima Folds, a
series of anticlinal ridges and broad, flat-floored synclinal valleys and basins,
influenced the formation of the Walla Walla basin. The Grande Ronde basin is a
graben-containing, fine-grained, poorly drained lacustrine deposits (map units
containing la). Where streams enter the basin bouldery alluvial-fan deposits occur
(map units containing af). Along the edges of the fault scarp (map unit containing fsc)
are the colluvial deposits (map units containing co). The Wallowa Mountains were
uplifted by faulting on two sides causing their characteristic bowl shape and radial
drainage pattern (Smith and others 1941). Glaciation coupled with this uplift caused
the mountain streams to become deeply incised (Bishop 1994) and formed the
Wallowa basin, Joseph Upland. There is also a colluvium slope along the
southwestern edge of the basin.
Nonchannelized streams in the basins generally have a meandering pattern. The
major differences within the basins are in the drainage properties. The areas underlain
primarily by gravels such as the alluvial fans have good drainage. Areas underlain by
cemented alluvium or fine silt have low permeability.
Extensive folding also influenced other areas. Major structures such as the Blue
Mountain anticline and the Walla Walla and Umatilla synclines were formed. From
Prairie City to Dayville, the John Day River flows in a synclinal trough (BR 1985), and
-------
Clarke and Bryce--176
the Grande Ronde River flows through a synclinal trough from Starkey to LaGrande
(Hampton and Brown 1964). This latter area is intersected by many northwest striking
faults (map unit containing f/p). The Columbia River Basalt Group in the Blue
Mountains is the southern upwarped and deformed part of the anticline (Baker and
others 1991). In these uplifted parts, streams have cut narrow, steep-walled canyons
where the rock is resistant. These valleys are marked by the stair-step edges of the
basalt layers. Several large upland areas remain undissected. Slopes on these
plateaus are generally less than 15 percent; in the dissected canyons, slopes generally
range from 30 to 60 percent.
In places where the basalt is thin or underlain by the John Day formation, much wider
valleys are formed (Orr and others 1992). The basalt in the Columbia Plateau region
is less deformed and underlies loess, lacustrine, and glaciofluviate deposits. However,
these deposits have not modified the drainage pattern controlled largely by the surface
of the underlying basalt. The stream gradients are generally high (greater than 50 ft
per mi [15 m per km]), determined by the tilt of the basalt (Harrison and others 1962).
Cenozoic Era-Quaternary Period-Generally, except for the basin fills, the geologic
deposits before this period have been indurated (consolidated and cemented). All the
deposits in the Quaternary Period consist of unconsolidated deposits. They are not
delineated on conventional bedrock geologic maps. These maps do not show
quaternary deposits unless they are extensive, cover all bedrock units, or disrupt
stratigraphy (as in a landslide). Gonthier and Bolke (1993) cite the mapping criterion
-------
Clarke and Bryce--177
of Swanson and others (1987) to map sediments wherever they are sufficiently thick to
obscure the underlying basalt. The subjectivity of this statement makes it clear why
geologic maps vary considerably in the portrayal of these deposits. For our purposes,
we used geologic maps in conjunction with topographic maps to determine the
location of map-unit boundaries.
Glaciation had both direct and indirect effects on the study area. During the early part
of the Quaternary period, the Pleistocene epoch, alpine glaciation carved, eroded, and
built morainal and fluvial deposits in the Wallowa, Elkhorn, Greenhorn, and Strawberry
Mountains. Large amounts of sediment filled the Grande Ronde valley. Indirectly,
flooding from glacial Lake Missoula was responsible for shaping most of the current
landscape of the Columbia Plateau.
In the Strawberry Mountains, glaciers sculpted the principal valleys above 1524 meters
(5000 ft) forming U shaped valleys (Thayer 1990). The radial valleys of the Wallowa
Mountains were widened and deepened by these glaciers. Glacial deposits consisting
of unsorted bouldery gravel, sand, and rock flour are found at the lower end of these
valleys and adjacent lowlands (map units containing gl).
The highest floodwater from Lake Missoula reached almost 366 meters (1200 ft) (Bretz
1969). Backflooding left deposits in many nonglacial valleys of the Snake River such
as the Tucannon and Asotin Rivers. Lake Lewis, formed by hydraulic damming behind
the Wallula Gap is largely responsible for the features of the Walla Walla basin. Lake
-------
Clarke and Bryce--178
Condon, formed from a constriction around the Dalles, strongly influenced the Umatilla
basin. Hogenson (1964) described these deposits in the Umatilla basin and Newcomb
(1965) described them for the Walla Walla basin. Glaciofluviate deposits are found
below 229 meters (750 ft) (map units containing es, gfg, and gfs). This area has been
scoured by wind and water, and the surface of these deposits is dotted by many
blowouts. Deposits consist of undifferentiated gravel, sand, and silt. In the Walla
Walla basin, these are called the Touchet Beds. Lacustrine deposits are found up to
an elevation of about 350 meters (1150 ft). They have been severely dissected into
long narrow ridges having steep north and east slopes and strongly sloping south
slopes (Harrison and others 1962), consisting mostly of stratified lacustrine silt (map
unit containing ca).
During these glacial periods, the weather was cold and windy, and vegetation was
sparse, ideal conditions for wind erosion (Brady 1990). Consequently, extensive loess
deposits blanketed the Columbia Plateau. These deposits are coarser and deeper
near the source, the glacial lake beds. Map units were differentiated using depth and
moisture in these loessial areas (map units containing !o), as discussed in the section
on soil. In the dry areas with sparse vegetation, wind erosion is a major problem
(Harrison and others 1962).
Although the recent alluvial valleys are relatively small (map units containing al), we
have delineated them because of their importance to fish (Reeves 1988). These
alluvial deposits consist of sorted and unsorted silt, sand, and gravel. The low
-------
Clarke and Bryce--179
gradient of these areas and the location of pools along the meanders makes them
important refugia and rearing habitat for fish. Because of their depositional nature,
they are subject to sedimentation from upstream sources. Where county soil surveys
were available, we defined these areas relatively easily. In other areas, we relied upon
their identification from the geology, actual vegetation, and STATSGO soil map. Often,
their width was greatly exaggerated for cartographic purposes. We used topography
to help us better refine the boundaries. Areas less than .3 kilometer (.2 mi) wide were
not mapped. We expect that some small alluvial valleys were missed by this process.
Topography—The preceding sections have covered the characteristics of topography,
therefore, this section only discusses the topographic maps and digital data used in
this project. The components of topography; elevation, slope, and aspect are
discussed in the sections on vegetation, climate, and soil. Because geomorphology is
the science of topography, further discussion of topography are found in the section
on geomorphology/geology.
We used 1:250,000 DEM's (Digital Elevation Models) for Oregon and Washington with
a 60- x 90-meter (66- x 98-yd) resolution. Slope, aspect, and elevation were calculated
from the DEM's. Slope was broken into 6 slope classes, aspect into 8, and elevation
into 10. Slope classes were 0,1 to 14,15 to 29, 30 to 59, 60 to 89, and greater than
90 percent. Elevation classes were 0 to 300 (0 to 328), 300 to 600 (328 to 656), 600
to 900 (656 to 984), 900 to 1200 (984 to 1312), 1200 to 1500 (1312 to 1640), 1500 to
1800 (1640 to 1968), 1800 to 2100 (1968 to 2296), 2100 to 2400 (2296 to 2624), 2400
-------
Clarke and Bryce--180
to 2700 (2624 to 2952), and 2700 to 3000 (2952 to 3280) meters (yd). Aspect classes
were N, NE, E, SE, S, SW, W, and NW. We plotted a coarse-scale map of slope,
aspect, and elevation for the study area showing these broad spatial patterns.
U.S. Geological Survey (GS) topographic maps, at 1:100,000, were indispensable to
interpret land surface form, drainage patterns, slope, aspect, and elevation and were
used as base maps for this project.
Conclusion
The type and distribution of substrate; habitat complexity operating at many spatial
scales; the amount, timing, and quality of water; the type and amount of riparian
vegetation; and interactions among biota are influenced by multiple landscape
characteristics, sometimes acting synergistically to cause a stronger response or
perhaps acting antagonistically to mitigate the influence. It is impossible to predict
stream habitat at any given point based upon a landscape classification because: (1)
landscape properties are not black and white and do not operate independently but
interact either synergistically or antagonistically with other landscape properties; (2) the
effects of upslope processes are not the same throughout a basin, but vary depending
upon the size of the stream, valley morphology, and slope; and (3) the influence of
different types and intensities of land uses needs to be considered.
Landscape-level ecoregions are a tool that may be useful in understanding and
-------
Clarke and Bryce--181
managing stream ecosystems. Years of process-level research will allow us to begin
to understand the direction and intensity of these multiple interacting processes. We
feel landscape-level ecoregions can help focus process-level research and that future
process-level research can help refine these regions in a positive feedback loop.
Because in the interim, which inevitably will be a substantial time frame, we are forced
to make management decisions, we have proposed these landscape-level ecoregions
as a level of stratification to help make local land management decisions. A
landscape-level classification can help identify: (1) potential areas from which to pick
reference sites, (2) areas that may be sensitive to land-use effects, (3) potential
monitoring and research sites, (4) areas to extrapolate the results from finer scale
process-level research, and (5) the potential capacity of a stream.
Before incorporating landscape-level ecoregions into future research or management
strategies, the landscape-level ecoregions should be evaluated. We agree with
Warren's (1979) assessment that "potential capacity is a very abstract theoretical
concept amenable only to very indirect and partial evaluation." He also states that
"although in no ordinary sense can a classification be shown to be true or false,
validation of a sort is important." We have chosen to use the word evaluate instead of
validate because evaluate expresses the element of judgment rather than proof.
By providing the logical underpinnings for the delineation of the map units, potential
users will be able to assess the usefulness of this classification for their specific project
objectives. This critical evaluation is the first step in evaluating a classification. The
-------
Clarke and Bryce--182
next step is a statistical analysis, as was done by Whittier and others (1988). They
used multivariate analysis of biotic assemblages, physicochemical measures, species
richness, diversity, and composition from small, minimally impacted streams to
evaluate the robustness of Omernik's ecoregion classification in Oregon. Three
approximately 100-meter-long (109-yd-long) reaches that encompassed complete sets
of the characteristics stream habitat types for that site were sampled for six streams
within each ecoregion. They found that as a whole streams within an ecoregion tend
to be like other streams in that region and unlike streams in other regions. The
Principal Components Analysis (PCA) ordination for physical habitat shows regional
patterns, however, the distribution of the Blue Mountains and Columbia Basin9
ecoregion's streams do not form a tight cluster pattern. Finer-scale classifications, i.e.,
subregions and landscape-level ecoregions may further explain some of the variability
within ecoregions.
Ongoing research is focusing on evaluating the usefulness of landscape-level
ecoregions for assessing the variability of stream habitat and fish distribution using
survey data. Based upon this and future work, it may be advantageous to combine
some regions if they are physically similar and respond to stresses similarly. Regions
may also be combined to meet specific project goals. Survey data should be both
spatially and temporally representative of potential stream capacity. The best indicator
of a stream's potential capacity results from analyzing fish distribution and stream
habitat survey data from streams in a relatively unimpacted condition. Data can be
from either historic or present surveys, and we recognize that data of this nature is
-------
Clarke and Bryce--183
very limited, especially for low-elevation areas. Warren (1979) states "if a classification
is to serve management purposes, systems within a class must respond similarly to
similar environmental conditions including resource utilization and environmental
management." Following the logic of Warren (1979), data from disturbed streams may
help evaluate the landscape delineation because just as a stream's potential capacity
may vary between regions, reaction to disturbance may also vary between regions.
Because a stream's reaction to disturbance is frequently a simplification of the channel
(Bisson and others 1992, Henjum and others 1994, Hicks and others 1991), evaluating
significant differences in stream habitat characteristics between regions and
understanding a stream's potential capacity is more difficult. However, a stream's
response to disturbance also varies with timing and character of last disturbance as
related to natural disturbance regime. In addition, using data from disturbed streams
requires knowledge of land-use history. For example, the headwaters of the upper
Grande Ronde River have historically been dredge mined (Mcintosh and others 1994).
Stream-habitat survey data from 1941 showed no large pools in this area. Recent
resurveys—(1990), found several large pools in the same stretch of rivers. However,
most of these pools were attributed to the artificial pool-forming structures that were
added to enhance habitat.10 Without knowledge of these events, erroneous
conclusions could be drawn from analysis of this stream habitat data.
Future work should focus on identifying the potential of a system using reference sites,
historical stream data, and expert knowledge and grappling with the issues of
differences in the magnitude of upslope and upstream effects. We hope that more
-------
Clarke and Bryce--184
work will also be done on classifying watersheds from their mix of ecoregions,
identifying appropriate questions for different scales in the hierarchy, and overlaying
information on human disturbances to assess the capability of ecosystems in each
ecoregion to deal with different types and intensities of stressors.
-------
Clarke and Bryce--185
Literature Cited
Andrews, H.J.; Cowlin, R.W. 1936a. Oregon: Northeast quadrant [Forest type].
Portland, OR: U.S Department of Agriculture, Forest Service, Pacific Northwest Forest
Experiment Station. 1:253,440; Lambert projection; colored.
Andrews, H.J.; Cowlin, R.W. 1936b. Washington: Southeast quadrant [Forest type].
Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Forest
Experiment Station. 1:253,440; Lambert projection; colored.
Baker, V.R.; Bjornstad, B.N.; Busacca, A.J. [and others]. 1991. Quaternary geology of
the Columbia Plateau. In: Morrison, R.B., ed. The Geology of North America, vol K-2,
Quaternary nonglacial geology: conterminous U.S., Boulder, CO: The Geological
Society of America.
Barrash, W.; Bond, J.G.; Kauffman, J,D.; Venkatakrishnan, R. 1980. Geology of the La
Grande area, Oregon. Special Paper Number 6. Portland, OR: State of Oregon,
Department of Geology and Mineral Industries. 47 p.
Beschta, R.L. 1991. Stream habitat management for fish in the northwestern United
States: the role of riparian vegetation. In: Colt, J.; White, R.J. Fisheries bioengineering
symposium; Portland, OR 1988, Bethesda, MA: American Fisheries Society: 53-58.
-------
Clarke and Bryce-186
Beschta, R.L.; Bilby, R.E.; Brown, G.W. [and others]. 1987. Stream temperature and
aquatic habitat: fisheries and forestry interactions. In: Salo, E.O.; Cundy, T.W., eds.
Streamside management: forestry and fishery interactions, No. 57. Seattle, WA:
University of Washington, Institute of Forest Resources: 191-232.
Bilby R.E. 1984. Characteristics and frequency of cool-water areas in a western
Washington stream. Journal of Freshwater Ecology. 2(6): 593-602.
Bishop, E.M. 1994. Mafic and ultramafic rocks of the Baker Terrane. In: Vallier, T.L.;
Brooks, H. Geology of Blue Mountains region. Professional Paper 1438. [N.p.]: U.S.
Department of the Interior, Geological Survey. In press.
Bishop, E.M.; Thompson, L.L.; Woodward, J. 1992. Sedimentation and tectonics in the
Grande Ronde Graben, Northeast Oregon: a preliminary assessment. Geological
Society of America, Cordilleran Section, 88th Annual Meeting. Eugene, OR. 7 p.
Bisson, P.A.; Bilby, R.E.; Bryant, M.D. [and others]. 1987. Large woody debris in the
Pacific Northwest: past, present, and future. In: Salo, E.O.; Cundy, T.W., eds.
Streamside management: forestry and fishery interaction. No. 57. Seattle, WA:
University of Washington: 143-190.
Bisson, P.A.; Nielsen, J.L.; Palmason, R.A.; Grove, L.E. 1982. A system of naming
habitat types in small streams, with examples of habitat utilization by salmonids during
-------
Clarke and Bryce--187
low streamflow. In: Armantrout, N.B., ed. Acquisition and utilization of aquatic habitat
inventory information. Bethesda, MD: American Fisheries Society: 62-73.
Bisson, P.A.; Quinn, T.P.; Reeves, G.H.; Gregory, S.V. 1992. Best management
practices, cumulative effects, and long-term trends in fish abundance in Pacific
Northwest river systems. In: Naiman, R.J., ed. Watershed management: balancing
sustainability and environmental change. New York, NY: Springer Verlag: 189-332.
Bjornn, T.C.; Reiser, D.W. 1991. Habitat requirements of salmonids in streams. In:
Meehan, W.R., ed. Influences of forest and rangeland management on salmonid fishes
and their habitats. American Fisheries Society Special Publication 19. Bethesda, MD:
American Fisheries Society: 83-138.
Bolles, W.H. 1937. Forest Statistics for Baker County, Oregon. Portland, OR: U.S.
Department of Agriculture, Forest Service, Pacific Northwest Forest Experiment Station.
10 p.
Bond, C.E.; Rexstad, E.; Hughes, R.M. 1988. Habitat use of twenty-five common
species of Oregon freshwater fishes. Northwest Science. 62(5): 223-232.
BR (U.S. Department of Interior, Bureau of Reclamation). 1985. Planning report
concluding the study of the Upper John Day project, Oregon. [N.p.]: Bureau of
Reclamation. 107 p.
-------
Clarke and Bryce--188
Brady, N.C. 1990. The nature and properties of soils. New York, NY: MacMillan
Publishing Co. 621 p.
Bretz, J.H. 1969. The Lake Missoula floods and the Channeled Scablands. Journal of
Geology, 77: 505-543.
Brooks, H.C. 1979. Plate tectonics and geologic history of the Blue Mountains. Oregon
Geology. 41(5): 71-80.
Brooks, H.C.; Fern, M.L.; Mullen, E.D. 1982. Oregon: Granite quadrangle, Grant
County [geologic and gold deposits]. GMS-25. Reston, VA: U.S. Department of the
Interior, Geological Survey. 1:24,000; Lambert conformal conic projection; colored.
Brown, C.E.; Thayer, T.P. 1966. Oregon: Canyon City quadrangle, northeastern
[Geologic]. I-447. Washington, D.C.: U.S. Department of Interior, Geological Survey.
1:250,000; colored.
Brunengo, M. 1991. Rain-on-snow zones. Olympia, WA: Washington Department of
Natural Resources. Digital data; 1:250,000.
Bryce, S.A.; Clarke, S.E. 1996. Landscape-level ecological regions: linking state-level
ecological frameworks with stream habitat. Environmental Management. 20 (X).
-------
Clarke and Bryce--189
Buckhouse, J.C.; Gaither, R.E. 1982. Potential sediment production within vegetative
communities in Oregon's Blue Mountains. Journal of Soil and Water Conservation.
March/April: 120-122.
Buell, E.D. 1937. Forest Statistics for Wallowa County, Oregon. Portland, OR: U.S.
Department of Agriculture, Forest Service, Pacific Northwest Forest Experiment Station.
9 p.
Carlson, G. 1974. Soil resource inventory, basic soil information and interpretive tables:
Malheur National Forest. Portland, OR: U.S. Department of Agriculture, Pacific
Northwest Region. 177 p.
Carlson, G.; Rother, E. 1974. Soil resource inventory, atlas of maps: Malheur National
Forest. Portland, OR: U.S. Department of Agriculture, Pacific Northwest Region. 33 p.
Chesterman, C.W. 1979. The Audubon Society field guide to North American rocks
and minerals. New York: Alfred A. Knopf. 850 p.
Clarke, S.E.; White, D.; Schaedel, A.L. 1991. Oregon, USA, ecological regions and
subregions for water quality management. Environmental Management. 15(6):
847-856.
Cupp, C.E. 1988. Valley segment type classification for forested lands of Washington.
-------
Clarke and Bryce--190
Timber, Fish and Wildlife Ambient Monitoring Program Report. Olympia, WA:
Washington Department of Natural Resources. 47 p.
Daubenmire, R. 1956. Climate as a determinant of vegetation distribution in eastern
Washington and northern Idaho. Ecological Monographs. 26(2): 131-154.
Daubenmire, R. 1988. Steppe vegetation of Washington. EB 1446. Pullman, WA:
Washington State University Cooperative Extension. 131 p.
Dyksterhuis, E.L. 1981. Soil Survey of Grant County area, Oregon. Washington, DC:
U.S. Department of Agriculture, Soil Conservation Service. 131 p.
Dyksterhuis, E.L.; High, C.T. 1985. Soil Survey of Union County area, Oregon.
Washington, DC: U.S. Department of Agriculture, Soil Conservation Service. 194 p.
Ehmer, L. [Year unknown]. Soil resource inventory. Pendleton, OR: Umatilla National
Forest, Region 6, U.S. Department of Agriculture, Forest Service. 305 p.
Evans, J.G. 1989. Oregon: Desolation Butte quadrangle, Grant and Umatilla counties
[Geologic]. CQ-1654. Reston, VA: U.S. Department of Interior, Geological Survey.
1:62,500; Polyconic projection; colored.
Everest, F.H.; Beschta, R.L.; Scrivener, J.C. [and others]. 1987. Fine sediment and
-------
Clarke and Bryce--191
salmonid production: a paradox. In: Salo, E.O.; Cundy, T.W., eds. Streamside
management: forestry and fishery interaction; No. 57. Seattle, WA: University of
Washington: 98-142.
Ferns, M.L. 1985. Preliminary report on Northeastern Oregon lignite and coal
resources, Union, Wallowa, and Wheeler counties. Salem, OR: State of Oregon,
Department of Geology and Mineral Industries; open-file report 0-85-2. 19 p.
Ferns, M.L; Taubeneck, W.H. 1994. Oregon: Limber Jim, Union County [Geologic].
GMS-82. Reston, VA: U.S. Department of the Interior, Geological Survey. 1:24,000;
Lambert conformal conic projection; colored.
Forest Ecosystem Management Assessment Team (FEMAT). 1993. Forest ecosystem
management: an ecological, economic, and social assessment. Portland, OR: U.S.
Department of Agriculture; U.S. Department of the Interior [and others]. [Irregular
pagination].
Foster, R.J. 1975. Physical geology. Columbus, OH: Charles E. Merrill Publishing Co.
421 p.
Franklin, J.F. 1992. Scientific basis for new perspectives in forests and streams. In:
Naiman, R.J., ed. watershed management: balancing sustainability and environmental
change. New York, NY: Springer Verlag: 25-72.
-------
Clarke and Bryce-192
Franklin, J.F.; Dyrness, C.T. 1988. Natural Vegetation of Oregon and Washington.
Corvallis, OR: Oregon State University Press. 452 p.
Frissell, C.A. 1993. Topology of extinction and endangerment of native fishes in the
Pacific Northwest and California (U.S.A.). Conservation Biology. 7(2): 342-354.
Frissell, C.A., Liss, W.J., Warren, C.E.; Hurley, M.D. 1986. A hierarchical framework for
stream habitat classification: viewing streams in a watershed context. Environmental
Management. 10: 199-214.
Gallant, A.L.; Whittier, T.R.; Larsen, D.P. [and others]. 1989. Regionalization as a tool
for managing environmental resources. EPA/600/3-89/060. Corvallis, OR: U.S.
Environmental Protection Agency, Environmental Research Laboratory. 152 p.
Gannett, H. 1902. The forest of Oregon. Professional Paper No. 4, Series H, Forestry,
No. 1. Washington, DC: U.S. Government Printing Office, U.S. Geological Survey. 36
P-
Gentry, H.R. 1991. Soil survey of Asotin County area, Washington, parts of Asotin and
Garfield counties. Washington, DC: U.S. Department of Agriculture, Soil Conservation
Service. 376 p.
Gill, R.E. 1994. Sediment delivery to headwater stream channels following road
-------
Clarke and Bryce--193
construction and timber harvest in the Blue Mountains, Oregon. Corvallis, OR: Oregon
State University. 59 p. M.S. thesis.
Gonthier, J.B.; Bolke, E.L. 1993. Summary Appraisal of Water Resources of the
Umatilla Indian Reservation. Water-Resources Investigation Report 91-4087-P
Portland, OR: U.S. Geological Survey. 54 p.
Gregory, K.J.; Walling, D.E. 1973. Drainage basin form and process: a
geomorphological approach. New York, NY: John Wiley & Sons. 456 p.
Gregory, S.V; Lamberti, G.A.; Ermann, D.C. [and others]. 1987. Influence of forest
practices on aquatic production. In: Salo, E.O.; Cundy, T.W. Streamside management:
forestry and fishery interaction; No. 57. Seattle, WA: University of Washington: 235-
255.
Gregory, S.V.; Swanson, F.J.; McKee, W.A.; Cummins, K.W. 1991. An ecosystem
perspective of riparian zones. Bioscience. 41(8): 540-551.
Griffith, G.E.; Omernik, J.M. 1991. Alabama/Mississippi project: geographic research
component: ecoregion/subregion delineation. Corvallis, OR: U.S. Environmental
Protection Agency. Environmental Research Laboratory. 27 p.
Griffith, G.E.; Omernik, J.M.; Rohm, C.M.; Pierson, S.M. 1994. Florida regionalization
-------
Clarke and Bryce-194
project. Corvallis, OR: U.S. Environmental Protection Agency, Environmental Research
Laboratory. 76 p.
Griffith, G.E.; Omernik, J.M.; Wilton, T.F.; Pierson, S.M. 1995. Ecoregions and
subregions of Iowa: a framework for water quality assessment and management.
Journal of the Iowa Academy of Science. Forthcoming.
GS (U.S. Department of the Interior, Geological Survey). United States [Potential
natural vegetation], 1970. 1:7,500,000; colored. In: The national atlas of the United
States of America. Washington, DC: Geological Survey: 89-91.
GS (U.S. Department of the Interior, Geological Survey). 1980. Washington and Idaho:
Columbia River Basalt Group, Pullman and Walla Walla Quadrangles, SE Washington
and adjacent Idaho. [Reconnaissance geologic], 1-1139. Reston, VA: GS. 1:250,000;
colored.
GS (U.S. Department of the Interior, U.S. Geological Survey). 1987. Digital elevation
models: data users guide 5. Reston, VA: GS. 38 p.
GS (U.S. Department of the Interior, U.S. Geological Survey). 1989. Digital line graphs
from 1:100,000 scale maps: data users guide 2. Reston, VA: GS [http://www-
nmd.usgs.gov/www/html/2nmdgds.html].
-------
Clarke and Bryce-195
Gulick, C.W. 1994a. Connell Quadrangle, Washington [Geologic]. 1:100,000. [N.p.]:
Washington State Department of Natural Resources, Division of Geology and Earth
Resources; open-file report 94-14. 18 p.
Gulick, C.W. 1994b. Pullman Quadrangle, Washington-Idaho [Geologic], 1:100,000.
[N.p.]: Washington State Department of Natural Resources, Division of Geology and
Earth Resources; open-file report 94-6. 22 p.
Hall, F.C. 1973a. Plant communities of the Blue Mountains of eastern Oregon and
southeastern Washington. Pacific Northwest Region Area Guide 3-1. Portland, OR:
U.S. Department of Agriculture, Forest Service, Pacific Northwest Region. 62 p.
Hall, F.C. 1973b. Plant communities of the Blue Mountains of eastern Oregon and
southeastern Washington. Pacific Northwest Region Area Guide 3-2. Portland, OR:
U.S. Department of Agriculture, Forest Service, Pacific Northwest Region. 51 p.
Hall, F.C. 1991. Ecology of fire in the Blue Mountains of eastern Oregon. Portland, OR:
U.S. Department of Agriculture, Forest Service, Pacific Northwest Region. 27 p.
Unpublished report. On file with: Pacific Northwest Research Station, Forestry
Sciences Laboratory, 3200 SW Jefferson Way, Corvallis, OR 97331.
Hampton, E.R.; Brown, S.G. 1964. Geology and ground-water resources of the upper
Grande Ronde River basin, Union County, Oregon. Geological Survey Water Supply
-------
Clarke and Bryce--196
Paper 1597. Washington, DC: U.S. Government Printing Office. 102 p.
Harrison, E.T.; Donaldson, N.C.; McCreary, F.R. [and others]. 1962. Soil survey of
Walla Walla County area, Washington. Washington, DC: U.S. Department of
Agriculture, Soil Conservation Service. 138 p.
Harrison, E.T.; McCreary, F.R.; Nelson, F.L. 1973. Soil survey of Columbia County
area, Washington. Washington, DC: U.S. Department of Agriculture, Soil Conservation
Service. 88 p.
Harvey, A.E.; Geist, J.M.; McDonald, G.I. [and others]. 1994. Biotic and abiotic
processes in eastside ecosystems: the effects of management on soil properties,
processes and productivity. Gen. Tech. Rep. PNW-GTR-323. Portland, OR: U.S.
Department of Agriculture, Forest Service, Pacific Northwest Research Station. 71 p.
Hawkins, C.P.; Kershner, J.L.; Bisson, P.A. [and others]. 1993. A hierarchical
approach to classifying stream habitat features. Fisheries. 18(6): 3-12.
Heede, B.H.; Rinne, J.N. 1990. Hydrodynamic and fluvial processes: implications for
fisheries management and research. North American Journal of Fisheries
Management. 10: 249-268.
Henjum, M.G.; Karr, J.R.; Bottom, D.L. [and others]. 1994. Interim protection for late-
-------
Clarke and Bryce--197
successional forests, fisheries, and watersheds: National forests east of the Cascade
crest, Oregon and Washington. Bethesda, MD: The Wildlife Society. 245 p.
Hicks, B.J., Hall, J.D.; Bisson, P.A.; Sedell, J.R. 1991. Responses of salmonids to
habitat changes. American Fisheries Society Special Publication Number 19: 483-518.
Hodge, E.T. 1942. Geology of north central Oregon. Oregon State Monographs,
Studies in Geology, No. 3. Corvallis, OR: Oregon State College. 76 p.
Hogenson, G.M. 1964. Geology and ground-water resources of the Umatilla River
basin, Oregon. Geological Survey Water Supply Paper 1620. Washington, DC: U.S.
Government Printing Office. 162 p.
Hosier, R.E. 1983. Soil Survey of Morrow County area, Oregon. Washington, DC: U.S.
Department of Agriculture, Soil Conservation Service. 225 p.
Hynes, H.B.N. 1972. The ecology of running waters. Toronto, Canada: University of
Toronto Press. 555 p.
Johnson, B.R.; Derkey, P.D. 1993. Geologic map compilation of southeast
Washington. Digital data. Draft dated 1993 to Sharon Clarke. On file with: U.S.
Geological Survey, Spokane, WA.
-------
Clarke and Bryce--198
Johnson, C.G., Jr.; Clausnitzer, R.R. 1992. Plant associations of the Blue and Ochoco
Mountains. R6-ERW-TP-036-92. [N.p.]: U.S. Department of Agriculture, Wallowa-
Whitman National Forest. 164 p.
Johnson, C.G., Jr.; Clausnitzer, R.R.; Mehringer, P.J.; Oliver, C.D. 1994. Biotic and
abiotic processes of eastside ecosystems: the effects of management on plant and
community ecology, and on stand and landscape vegetation dynamics. Gen. Tech.
Rep. PNW-GTR-322. Portland, OR: U.S. Department of Agriculture, Forest Service,
Pacific Northwest Research Station. 66 p.
Johnson, C.G., Jr.; Simon, S.A. 1987. Plant associations of the Wallowa-Snake
province: Wallowa-Whitman National Forest. R6-ECOL-TP-255A-86. [N.p.]: U.S.
Department of Agriculture, Wallowa-Whitman National Forest. 399 p.
Johnson, D.R.; Makinson, A.J. 1988. Soil survey of Umatilla County area, Oregon.
Washington, DC: U.S. Department of Agriculture, Soil Conservation Service. 388 p.
Kagan, J.; Caicco, S. 1992. Manual of Oregon actual vegetation. Boise, ID: University
of Idaho, Idaho Cooperative Fish and Wildlife Research Unit. 186 p.
Lehmkuhl, J.F,- Hessburg, P.F.; Everett, R.L. [and others]. 1994. Historical and current
forest landscapes of eastern Oregon and Washington. Part 1: Vegetation pattern and
insect and disease hazards. Gen. Tech. Rep. PNW-GTR-328. Portland, OR: U.S.
-------
Clarke and Bryce-199
Department of Agriculture, Forest Service, Pacific Northwest Research Station. 88 p.
Lichatowich, J.; Mobrand, L.; Lestelle, L.; Vogel, T. 1995. An approach to the
diagnosis and treatment of depleted Pacific salmon populations in Pacific Northwest
watersheds. Fisheries. 20(1): 10-18.
Litchfield, W. 1937. Forest Statistics for Union County, Oregon. Portland, OR: U.S.
Department of Agriculture, Forest Service, Pacific Northwest Forest Experiment Station.
11 p.
Logan, H.; Lammers, D. [1966]. Turbidity of upper Gallatin river determined from IR
photographs; unpublished report. On file with: U.S. Department of Agriculture, Gallatin
National Forest, Montana.
Lotspeich, F.B.; Platts, W.S. 1982. An integrated land-aquatic classification system.
North American Journal of Fisheries Management. 2: 138-149.
Marcot, B.G.; Wisdom, M.J.; Li, H.W.; Castillo, G.C. 1994. Managing for featured,
threatened, endangered, and sensitive species and unique habitats for ecosystem
sustainability. Gen. Tech. Rep. PNW-GTR-329. Portland, OR: U.S. Dept. of Agriculture,
Forest Service, Pacific Northwest Research Station. 39 p.
Margalef, R. 1960. Ideas for a synthetic approach to the ecology of running waters.
-------
Clarke and Bryce-200
Int. Rev. Gesamten Hydrobiol. 45: 133-153.
Maser, C.; Sedell, J.R. 1994. From the forest to the sea: the ecology of wood in
streams, rivers, estuaries, and oceans. Delray Beach, FL: St. Lucie Press. 200 p.
Mcintosh, B.A.; Sedell, J.R.; Smith, J.E. [and others]. 1994. Management history of
Eastside ecosystems: changes in fish habitat over 50 years. 1935 - 1992. Gen. Tech.
Rep. PNW-GTR-321. Portland, OR: U.S. Department of Agriculture, Forest Service,
Pacific Northwest Research Station. 55 p.
Meehan, W.R. 1991. Introduction and overview. In: Meehan, W.R., ed. Influences of
forest and rangeland management on salmonid fishes and their habitats, American
Fisheries Society Special Publication 19. Bethesda, MD: American Fisheries Society:
1-15.
Megahan, W.F. 1972. Sedimentation in relation to logging activities in the mountains of
central Idaho. In: Proceedings, sediment yield workshop; 1972; Oxford, MS. Oxford,
MS: U.S. Department of Agriculture, Sedimentation Laboratory: 74-82.
Megahan, W.F.; Monsen, S.B.; Wilson, M.D. [and others]. 1992. Erosion control
practices applied to granitic roadfills for forest roads in Idaho: cost effectiveness
evaluation. In: Land degradation and rehabilitation, vol. 3. New York, NY: John Wiley
and Sons: 55-55.
-------
Clarke and Bryce--201
Merriam, J.C. 1901. A contribution to the geology of the John Day basin. University of
California, Department of Geology Bulletin. 2(9): 269-314.
Miller, J.F.; Frederick, R.H.; Tracey, R.J. 1973. Precipitation-frequency atlas of the
western United States. Volume IX, Washington. U.S. Department of Commerce,
National Oceanic and Atmospheric Administration. DNR digital data 1:2,000,000.
Minshall, G.W. 1988. Stream ecosystem theory: a global perspective. Journal of the
North American Benthological Society. 7(4): 263-288.
Minshall, G.W.; Cummins, K.W.; Petersen, R.C. [and others]. 1985. Developments in
stream ecosystem theory. Canadian Journal of Fisheries and Aquatic Sciences. 42:
1045-1055.
Montgomery, D.R.; Buffington, J.M. 1993. Channel classification, prediction of channel
response, and assessment of channel condition. TFW-SH10-93-002. Olympia, WA:
State of Washington, Timber Fish and Wildlife. 84 p.
Moravets, F.L. 1937. Forest Statistics for Wheeler County, Oregon. Portland, OR: U.S.
Department of Agriculture, Forest Service, Pacific Northwest Forest Experiment Station.
10 p.
Munger, T.T. 1917. Western yellow pine in Oregon. Bulletin Number 418. Washington,
-------
Clarke and Bryce--202
DC: U.S. Department of Agriculture. 48 p.
Murphy, M.L.; Meehan, W.R. 1991. Stream ecosystems. In: Influences of forest and
rangeland management on salmonid fishes and their habitats. American Fisheries
Society Special Publication 19. Bethesda, MD: American Fisheries Society: 17-46.
Naiman, R.J.; Beechie, T.J.; Benda, L.E. [and others], 1992. Fundamental elements of
ecologically healthy watersheds in the Pacific Northwest coastal ecoregion. In: Naiman,
R.J., ed. Watershed management: balancing sustainability and environmental change,
New York, NY: Springer Verlag: 127-188.
Nawa, R.K.; Frissel, C.A.; Ebersole, J.L.; Liss, W.J. 1991. Life history and persistence
of anadromous fish stocks in relation to stream habitats and watershed classification.
Annual Progress Report, Project F-122-R. Portland, OR: Oregon Department of Fish
and Wildlife. 82 p.
Nehlsen, W.; Williams, J.E.; Ucatowich, J.A. 1991. Pacific salmon at the crossroads:
Stocks at risk from California, Oregon, Idaho, and Washington. Fisheries. 16(2): 4-21.
Newcomb, R.C. 1965. Geology and ground-water resources of the Walla Walla River
basin, Washington-Oregon. Water Supply Bulletin Number 21. [N.p.]: State of
Washington, Division of Water Resources. 76 p.
-------
Clarke and Bryce--203
NRCS (U.S. Department of Agriculture, National Resources Conservation Service
[formerly, Soil Conservation Service]) Wallowa County soil survey. [N.p.]: [N.p.]. In
progress.
Oles, K.F.; Enlows, H.E. 1971. Bedrock geology of the Mitchell Quadrangle, Wheeler
County, Oregon. Bulletin Number 72. Portland, OR: State of Oregon, Department of
Geology and Mineral Industries. 62 p.
Oliver, C.D.; Irwin, L.L.; Knapp, W.H. 1994. Eastside forest management practices:
Historical overview, extent of their applications, and their effects on sustainability of
ecosystems. Gen. Tech. Rep. PNW-GTR-324. Portland, OR: U.S. Department of
Agriculture, Forest Service, Pacific Northwest Research Station. 73 p.
Omernik, J.M. 1987. Ecoregions of the conterminous United States. Annals of the
Association of American Geographers. 77(1): 118-125.
Omernik, J.M. 1995. Ecoregions: a spatial framework for environmental management.
In: Davis, W.S.; Simon, T.P., eds. Biological assessment and criteria: tools for water
resource planning and decision making. Ann Arbor, Ml: Lewis Publishers: 49-65.
Omernik, J.M.; Griffith, G.E. 1991. Ecological regions versus hydrologic units:
frameworks for managing water quality. Journal of Soil and Water Conservation. 46:
334-340.
-------
Clarke and Bryce--204
O'Neill, R.V.; DeAngelis, D.L.; Warde, J.B.; Allen, T.F.H. Allen. 1986. Hierarchical
concept of ecosystems. Monographs in Population Biology 28. Princeton, NJ:
Princeton University Press. 235 p.
Orr, E.L.; Orr, W.N.; Baldwin, E.M. 1992. Geology of Oregon. 4d ed. Dubuque, IA:
Kendall/Hunt Publishing Co. 254 p.
Paulson, D.J. 1977. Ochoco National Forest soil resource inventory. Portland, OR:
U.S. Department of Agriculture, Forest Service, Pacific Northwest Region. 347 p.
Pelto, W.E. 1937. Forest statistics for Grant County, Oregon. Portland, OR: U.S.
Department of Agriculture, Forest Service, Pacific Northwest Forest Experiment Station.
9 p.
Petersen P.H.; Hendry, A.; Quinn, T.P. 1992. Assessment of cumulative effects on
salmonid habitat: some suggested parameters and target conditions. TFW-F3-92-001.
Seattle, WA: University of Washington, Center for Streamside Studies. 75 p.
Platts, W.S. 1980. A plea for fishery habitat classification. Fisheries. 5: 2-6.
PNFES (U.S. Department of Agriculture, Forest Service, Pacific Northwest Forest
Experiment Station). 1936. The forest survey of Eastern Oregon and Eastern
Washington: explanatory text to accompany forest statistics for each county. Portland,
-------
Clarke and Bryce--205
OR: Pacific Northwest Forest Experiment Station. 8 p.
PNFES (U.S. Department of Agriculture, Forest Service, Pacific Northwest Forest
Experiment Station). 1949. Forest type classification of the Pacific Northwest region.
Portland, OR: Pacific Northwest Forest Experiment Station. 15 p.
PNFES (U.S. Department of Agriculture, Forest Service, Pacific Northwest Forest
Experiment Station). 1952. Forest type classification of the Pacific Northwest region,
revised. Portland, OR: Pacific Northwest Forest Experiment Station. 33 p.
PNFES (U.S. Department of Agriculture, Forest Service, Pacific Northwest Forest
Experiment Station). 1957. Plot procedure and forest type classification. Portland, OR:
Pacific Northwest Forest Experiment Station. 35 p.
Reeves, G.H. 1988. Distribution and abundance offish and fish habitat in Upper Elk
River, 1985-1986; draft report. On file with: U.S. Department of Agriculture, Pacific
Northwest Research Station, Corvallis, OR.
Reidel, S.P.; Hooper, P.R. 1989. Volcanism and tectonism in the Columbia River Flood-
Basalt Province. Special Paper 239. Boulder, CO: The Geological Society of America,
Inc. 386 p.
Rhodes, J.J.; McCullough, D.A.; Espinosa, F.A., Jr. 1994. A coarse screening process
-------
Clarke and Bryce--206
for evaluation of the effects of land management activities on salmon spawning and
rearing habitat in ESA consultations. Tech. Rep. 94-4. Portland, OR: Columbia River
Inter-Tribal Fish Commission. 127 p.
Robinson, P.T. 1975. Oregon: John Day formation in the southwestern part of the Blue
Mountains and adjacent areas, north-central [Reconnaissance geologic]. I-872.
Reston, VA: U.S. Department of the Interior, Geological Survey. 1:250,000; colored.
Robinson, P.T.; Walker, G.W.; McKee, E.H. 1990. Eocene, Oligocene and lower
Miocene rocks of the Blue Mountain region. In: Walker, G.M., ed. Geology of the Blue
Mountain Region of Oregon, Idaho, and Washington: Cenozoic geology of the Blue
Mountains region. U.S. Geological Survey Professional Paper 1437. Washington, DC:
U.S. Government Printing Office: 29-62.
Rosgen, D.L. 1985. A stream classification system. In: Johnson, R.R.; Ziebell, C.D.;
Palton, D.R. [and others], eds. Riparian ecosystems and their management:
reconciling conflicting uses; First North American Riparian Conference; 1985 April
16-18; Tucson, AZ. [N.p.]: [N.p.]: 91-95.
Sankela, W.E. 1937a. Forest statistics for Morrow County, Oregon. Portland, OR: U.S.
Department of Agriculture, Forest Service, Pacific Northwest Forest Experiment Station.
6 p.
-------
Clarke and Bryce--207
Sankela, W.E. 1937b. Forest statistics for Umatilla County, Oregon. Portland, OR: U.S.
Department of Agriculture, Forest Service, Pacific Northwest Forest Experiment Station.
8 p.
Schuster, J.E. 1993. Clarkston Quadrangle, Washington-Idaho and the Washington
Portion of the Orofino Quadrangle [Geologic]; open-file report 93-4. [N.p.]: Washington
State Department of Natural Resources, Division of Geology and Earth Resources.
1:100,000. 43 p.
Schuster, J.E. 1994. Walla Walla Quadrangle, Washington [Geologic]; open-file report
94-3. [N.p.]: Washington State Department of Natural Resources, Division of Geology
and Earth Resources. 1:100,000. 18 p.
SCS (U.S. Department of Agriculture, Soil Conservation Service). 1993. State soil
geographic data base (STATSGO). Miscellaneous Publication Number 1492.
Washington, DC: U.S. Government Printing Office. 88 p.
Soil Conservation Society of America. 1982. Resource Conservation Glossary. Ankeny,
Iowa: Soil Conservation Society of America. 193 p.
Sedell, J.R.; Beschta, R.L. 1991. Bringing back the "bio" in bioengineering. American
Fisheries Society Symposium. 10: 160-175.
-------
Clarke and Bryce--208
Sedell, J.R.; Bisson, P.A.; Swanson, F.J.; Gregory, S.V. 1988. What we know about
large trees that fall into streams and rivers. In: Maser, C.; Tarrant, R.F.; Trappe, J.M.;
Franklin, J.F., eds. From the forest to the sea: a story of fallen trees. Gen. Tech. Rep.
PNW-GTR-229. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific
Northwest Research Station: 47-81.
Sedell, J.R.; Reeves, G.H.; Hauer, F.R. [and others]. 1990. Role of refugia in recovery
from disturbances: Modern fragmented and disconnected river systems. Environmental
Management. 14(5): 711-724.
Smith, W.D.; Allen, J.E.; Staples, L.W.; Lowell, W.R. 1941. Geology and physiography
of the northern Wallowa Mountains, Oregon. Bulletin Number 12. State of Oregon,
Department of Geology and Mineral Industries. 55 p.
Stanford, J.A.; Ward, J.V. 1992. Management of aquatic resources in large
catchments: recovery interactions between ecosystem connectivity and environmental
disturbance. In: Naiman, R.J., ed. Watershed management: balancing sustainability
and environmental change, New York, NY: Springer-Verlag: 91-124.
Sullivan, K.; Lisle, T.E.; Dolloff, C.A.; Reid, L.M. 1987. Stream channels: the link
between forests and fishes. Pacific Northwest Forest landscapes. In: Salo, E.O.;
Cundy, T.W., eds. Streamside management: forestry and fishery interaction. No. 57.
Seattle, WA: University of Washington: 39-97.
-------
Clarke and Bryce-209
Swanson, F.J.; Benda, L.E.; Duncan, S.H. [and others]. 1987. Mass failures and other
processes of sediment production in Pacific Northwest Forest landscapes. In: Salo,
E.O.; Cundy, T.W., eds. Streamside management: forestry and fishery interaction. No.
57. Seattle, WA: University of Washington: 9-38.
Swanston, D.N. 1991. Natural processes. In: Meehan, W.R., ed. Influences of forest
and rangeland management on salmonid fishes and their habitats. American Fisheries
Society Special Publication 19. Bethesda, MD: American Fisheries Society: 139-179.
Taubeneck, W.H. 1957. Geology of the Elkhorn Mountains, northeastern Oregon: Bald
Mountain batholith. Geological Society of America Bulletin. (75): 181-238.
Taylor, G. 1994. Oregon Climate Service: a new Oregon precipitation map. Oregon
Cursor. 7(1): 4 + .
Thayer, T.P. 1990. The geologic setting of the John Day Country, Grant County,
Oregon. [N.p.]: Pacific Northwest Parks and Forests Association. 23 p.
Thiele, S.A.; Kiilsgaard, C.; Omernik, J.M. 1992. The subdivision of the coast range
ecoregion of Oregon and Washington. Corvallis, OR: U.S. Environmental Protection
Agency, Environmental Research Laboratory. 39 p.
Vannote, R.L.; Minshall, G.W.; Cummins, K.W. [and others], 1980. The river continuum
-------
Clarke and Bryce--210
concept. Canadian Journal of Fisheries and Aquatic Sciences. 37: 130-137.
Walker, G.W. 1973. Oregon and Washington: Pendleton quadrangle [Reconnaissance
geologic]. Washington, DC: U.S. Department of the Interior, Geological Survey.
1:250,000; colored.
Walker, G.W. 1979. Oregon: Grangeville quadrangle, Baker, Union, Umatilla and
Wallowa counties [Reconnaissance geologic], 1-1116. Reston, VA: U.S. Department of
the Interior, Geological Survey. 1:250,000; Transverse mercator projection; colored.
Walker, G.W. 1990a. Miocene and younger rocks of the Blue Mountain region
exclusive of the Columbia River Basalt Group and associated mafic lava flows. In:
Walker, G.M.. ed. Geology of the Blue Mountain Region of Oregon, Idaho, and
Washington: Cenozoic geology of the Blue Mountains region. Professional Paper
1437. Washington, DC: U.S. Government Printing Office: 101-118.
Walker, G.W. 1990b. Overview of the Cenozoic geology of the Blue Mountain region.
In: Walker, G.M., ed. Geology of the Blue Mountain Region of Oregon, Idaho, and
Washington: Cenozoic geology of the Blue Mountains region. Professional Paper
1437. Washington, DC: U.S. Government Printing Office: 1-12.
Walker, G.W.; MacLeod, N.S. 1991. Oregon [Geologic]. Reston, Va: U.S. Department
of Interior, Geological Survey. 1:500,000; Lambert conformal conic projection; colored;
-------
Clarke and Bryce--211
2 sheets.
Walker, G.W.; Robinson, P.T. 1990. Paleocene, Eocene, and Oligocene rocks of the
Blue Mountain region. In: Walker, G.M., ed. Geology of the Blue Mountain Region of
Oregon, Idaho, and Washington: Cenozoic geology of the Blue Mountains region. U.S.
Geological Survey Professional Paper 1437. Washington, DC: U.S. Government
Printing Office: 13-28.
Wallowa-Whitman National Forest (Art Kreger) 1993. Digital soil survey, dated 1993 to
Sharon Clarke. On file with: Wallowa-Whitman National Forest, LaGrande, OR.
Warren, C.E. 1979. Toward classification and rationale for watershed management and
stream protection. EPA-600/3-79-059. Corvallis, OR: U.S. Environmental Protection
Agency. 143 p.
Weissenborn, A.E., ed. 1969. Mineral and water resources of Oregon. Bulletin Number
64. Washington, DC: U.S. Government Printing Office, U.S. Geological Survey. 469 p.
Whiteman, K.J.; Vaccaro, J.J.; Gonthier, J.B.; Bauer, H.H. 1994. The hydrogeologic
framework and geochemistry of the Columbia Plateau Aquifer System, Washington,
Oregon, and Idaho. U.S. Geological Survey Professional Paper 1414-B. Washington,
D.C.: U.S. Government Printing Office. 73 p.
-------
Clarke and Bryce-212
Whittier, T.R., Hughes, R.M.; Larsen, D.P. 1988. Correspondence between ecoregions
and spatial patterns in stream ecosystems in Oregon. Canadian Journal Fisheries and
Aquatic Sciences. 45: 1264-1278.
Williams, J.E.; Johnson, J.E.; Hendrickson, D.A. [and others]. 1989. Fishes of North
America endangered, threatened and of special concern. Fisheries. 14(6): 2-20.
Wissmar, R.C. 1993. The need for long-term stream monitoring programs in forest
ecosystems of the Pacific Northwest. Environmental Monitoring and Assessment. 26:
219-234.
Wissmar, R.C.; Swanson, F.J. 1990. Landscape disturbances and lotic ecotones. In:
Naiman, R.J.; Decamps, H., eds. The ecology and management of aquatic-terrestrial
ecotones, UNESCO Man and the Biosphere Series, Volume 4. Paris, France: United
Nations Education Scientific and Cultural Organization (UNESCO): 65-89.
Wolfe, H.M. 1937. Forest statistics for Crook County, Oregon. Portland, OR: U.S.
Department of Agriculture, Forest Service, Pacific Northwest Forest Experiment Station.
8 p.
Woodward, F.I. 1987. Climate and plant distribution. Cambridge studies in ecology.
New York, NY: Cambridge University Press. 174 p.
-------
Clarke and Bryce--213
WRB (Water Resources Board). 1963. Umatilla River basin. Salem, OR: State of
Oregon. 107 p.
WRD (Water Resources Department). 1988. Umatilla basin report. Salem, OR: State of
Oregon. 246 p.
-------
Clarke and Bryce--214
ACKNOWLEDGMENTS
Special thanks to the many providers of data. Many people provided draft data so
that we could get started on our project-Steve Beckwitt, Sierra Biodiversity Institute;
Susan Balikov, Wilderness Society; Gary Raines, USGS, University of Nevada, Reno;
Bruce Johnson and Pamela Derkey, USGS, Spokane Field Office; Sue Pierson,
Ogden; Marcia Brett, Oregon State University, Crop and Soil Science Department;
George Taylor, Oregon State University, State Climatologist; Tom O'Neill and Milt Hill,
Oregon Department of Fish and Wildlife; Vi Agnew, Pacific Northwest Region 6 Forest
Service; Art Kreger and Anna Kramer, Wallowa-Whitman National Forest; David Powell,
Keith Walbridge, and Sean McKinney, Umatilla National Forest; Kelly Cassidy,
University of Washington, Cooperative Fish and Wildlife Research Unit; Sandra Bahr,
Washington Department of Natural Resources; Jamie Kienzle, Natural Resource
Conservation Service, Wallowa County; and Robert Weinberger, Boise Cascade.
The integration of source material from a wide range of disciplines was greatly aided
by several people familiar with the natural resources of the study area-Ellen Bishop,
Oregon Watershed Health Program; Mark Ferns and Bob Ottersberg, Wallowa-
Whitman National Forest; Karl Hippie, Natural Resource Conservation Service,
Washington; and Duane Lammers, Pacific Northwest Region 6 Forest Service.
Technical assistance and project support for sections 1 and 2 was provided by Sandi
Azevedo, Francie Faure, Sue Pierson, and Marge Hails. Technical assistance and
-------
Clarke and Bryce~215
project support for section 3 was provided competently and cheerfully by the staff of
the Aquatic/Land Interactions Program at the Pacific Northwest Research Station in
Corvallis, Oregon-Kelly Christianson, Sudhi Gulur, Tami Lowry, Paula Minear, Nathan
Poage, Cathy Baldwin, Kathryn Ronnenberg, Sam Vedanayam, and George Weaver.
The helpful comments of the many reviewers were greatly appreciated-Jim Omernik,
Corvallis, EPA; Chip Andrus and Cynthia Chapman, ManTech Environmental Research
Services; Dan Bottom, Oregon Fish and Wildlife Department; Paula Minear, Oregon
State University, Fish and Wildlife Department; Duane Lammers, Pacific Northwest
Region 6 Forest Service; Charlie Johnson, Malheur, Umatilla, and Wallowa-Whitman
National Forests; Ellen Bishop, Oregon Watershed Health Program; Jonathan Rhodes,
Columbia River Inter-Tribal Fish Commission; Gordie Reeves, Pacific Northwest
Research Station; and Thor Thorsen, Natural Resource Conservation Service, Oregon.
Kelly Burnett and Duane Lammers of the Pacific Northwest Research Station and Glen
Griffith of the Corvallis, EPA participated in many thoughtful discussions about the
utility of the ecoregion approach.
This document has been funded by the U.S. Department of Agriculture, Forest Service,
Pacific Northwest Research Station through BPA Project Number 89-104 to Oregon
State University and U.S. Environmental Protection Agency National Health and
Environmental Effects Research Laboratory, Western Ecology Division, in Corvallis,
Oregon, through Contract 68-C4-0019 to ManTech Environmental Research Services
-------
Clarke and Bryce--216
Corporation. It has been subjected to the EPA's peer and administrative review and
approved for publication.
-------
Clarke and Bryce-217
Table 1 - Native vegetation defining Columbia Plateau subregions (fig. 3 by legend number)
Subregion Genus species Common name
Channeled Scablands (10a) Artemisia rigida Stiff sage
Poa sandbergii Sandberg's bluegrass
Loess Islands (10b)
Artemisia tridentata
Agropyron spicatum
Festuca idahoensis
Artemisia tripartita
Big sage
Bluebunch wheatgrass
Idaho fescue
Threetip sage
Umatilla Plateau (10c) Agropyron spicatum Bluebunch wheatgrass
Festuca idahoensis Idaho fescue
Rosa spp Rose
Crataegus spp. Hawthorne
Symphoricarpus albus Common snowberry
Okanogan Drift Hills (10d)
Artemisia tridentata
Agropyron spicatum
Artemisia tripartita
Festuca idahoensis
Purshia tridentata
Pinus ponderosa
Big sage
Bluebunch wheatgrass
Threetip sage
Idaho fescue
Bitterbrush
Ponderosa pine
-------
Pleistocene Lake Basins (10e) Artemisia tridentata
Agropyron spicatum
Canyons and Dissected Uplands (10f) Rosa spp
Symphoricarpus albus
Yakima Folds (lOg) Artemisia tridentata
Agropyron spicatum
Festuca idahoensis
Hieracium spp.
Purshia tridentata
Clarke and Bryce-218
Big sage
Bluebunch wheatgrass
Rose
Common snowberry
Big sage
Bluebunch wheatgrass
Idaho fescue
Hawkweeds
Bitterbrush
-------
Clarke and Bryce--219
Table 2-Native vegetation defining Blue Mountain subregions (fig. 3 by legend number)
Subregion
-Subalpine areas (11m)
Genus species
Abies lasiocarpa
Pinus contorta
Larix occidentalis
Picea engelmanii
Pinus albicaulis
Common name
Subalpine fir
Lodgepole pine
Western larch
Engelmann spruce
Whitebark pine
Subregions in the Cascade
Rainshadow
-John Day/Clarno Uplands (11a) Juniperus occidentalis
Aretemisia tridentata
Purshia tridentata
Agropyron spicatum
Festuca idahoensis
Poa sandbergii
Western juniper
Big sagebrush
Bitterbrush
Bluebunch wheatgrass
Idaho fescue
Sandberg's bluegrass
-------
Clarke and Bryce-220
-John Day/Clarno Highlands
(11b)
Pinus ponderosa
Pinus contorta
Psuedotsuga menziesii
Purshia tridentata
Cecocarpus ledifolius
Agropyron spicatum
Calamagrostis rubescens
Carex geyeri
Maritime-Influenced Zone (11c)
-Mesic Forest Section (111)
Abies grandis
Psuedotsuga menziesii
Pinus contorta
Picea engelmannii
Larix occidentalis
Abies lasiocarpa
Linnaea borealis
Spirea betulfolia
Acer glabrum
Taxus brevifolia
Vaccinium membranaceum
Vaccinium scoparium
Ponderosa pine
Lodgepole pine
Douglas-fir
Bitterbrush
Mountain mahogany
Bluebunch wheatgrass
Pinegrass
Elk sedge
Grand fir
Douglas-fir
Lodgepole pine
Engelmann spruce
Western larch
Subalpine fir
Twinflower
Birchleaf spirea
Rocky Mountain maple
Pacific yew
Big huckleberry
Grouse huckleberry
-------
Clarke and Bryce~221
-Subalpine areas (11m)
Melange Subregion (1 id)
Abies lasiocarpa
Pinus contorts
Larix occidentalis
Picea engelmanii
Pinus albicaulis
Subalpine fir
Lodgepole pine
Western larch
Engelmann spruce
Whitebark pine
Pinus ponderosa
Juniperus occidentalis
Psuedotsuga menziesii
Cercocarpus ledifolius
Purschia tridentata
Ribes cereum
Symphoricarpos
Calamagrostis rubescens
Carex geyeri
Ponderosa pine
Western juniper
Douglas-fir
Mountain mahogany
Bitterbrush
Squaw current
Snowberry
Pinegrass
Elk sedge
-------
-Mesic Forest Section (111)
-Subalpine areas (11 m)
Clarke and Bryce--222
Abies grandis
Grand fir
Psuedotsuga menziesii
Douglas-fir
Pinus contorta
Lodgepoie pine
Picea engelmannii
Engelmann spruce
Larix occidentalis
Western larch
Abies lasiocarpa
Subalpine fir
Linnaea borealis
Twinflower
Spirea betulfolia
Birchleaf spirea
Acer glabrum
Rocky Mountain maple
Taxus brevifolia
Pacific yew
Vaccinium membranaceum
Big huckleberry
Vaccinium scoparium
Grouse huckleberry
Abies lasiocarpa
Pinus contorta
Picea engelmannii
Pinus albicaulis
Artemisia tridentata
Idaho fescue
Festuca viridula
Carex geyeri
Carex hoodii
Subalpine fir
Lodgepoie pine
Engelmann spruce
Whitebark pine
Alpine sagebrush
Idaho fescue
Green fescue
Elk sedge
Hood's sedge
-------
Clarke and Bryce--223
Wallowas/Seven Devils Subregion
(11 e)
Plnus ponderosa
Psuedotsuga menziesii
Physocarpus malvaceus
Symphoricarpos albus
Acer glabrum
Amelanchier alnifolia
Purschia tridentata
Artemisia tridentata vaseyana
-Mesic Forest Section (111) Abies grandis
Abies lasiocarpa
Clintonia unifiora
Vaccinium scoparium
-Subalpine areas (11 m) Abies lasiocarpa
Picea engelmanii
Tsuga mertensiana
Festuca viridula
Carex hoodn
Canyons and dissected highlands Pinus ponderosa
(1 if) Psuedotsuga menziesii
Larix occidentalis
Abies grandis
Ponderosa pine
Douglas-fir
Mallow ninebark
Common snowberry
Rocky Mountain maple
Western serviceberry
Bitterbrush
Mountain big sagebrush
Grand fir
Subalpine fir
Queen's cup beadlily
Big huckleberry
Subalpine fir
Engelmann spruce
Mountain hemlock
Green fescue
Hood's sedge
Ponderosa pine
Douglas-fir
Western larch
Grand fir
-------
Clarke and Bryce-224
Snake and Salmon River Canyons Pinus ponderosa
(11g) Purschia tridentata
Rosa
Symphoricarpos albus
Glossopetalon nevadense
Agropyron spicatum
Poa sandbergii
Festuca idahoensis
Continental Zone Highlands (11 h) Pinus ponderosa
Psuedotsuga menziesii
Artemisia tridentata
Cecocarpus ledifolius
Purschia tridentata
Ribes cereum
Calamagrostis rubescens
Carex geyerii
Ponderosa pine
Bitterbrush
Rose
Common snowberry
Spiny greenbush
Bluebunch wheatgrass
Sandberg's bluegrass
Idaho fescue
Ponderosa pine
Douglas-fir
Big sagebrush
Mountain mahogany
Bitterbrush
Squaw current
Pinegrass
Elk sedge
-------
Continental Zone Foothills (11 i) Juniperus occidentalis
Artemisia rigida
Artemisia arbuscula
Cercocarpus ledifolius
Purschia tridentata
Artemisia tridentata
Festuca idahoensis
Agropyron spicatum
Batholith Contact Zone (11 j) Pinus ponderosa
Psuedotsuga menziesii
Physocarpus malvaceus
Calamagrostis rubescens
-Mesic Forest Section (111) Pinus ponderosa
Larix occidentalis
Pinus contorta
Picea engelmannii
Clarke and Bryce--225
Western juniper
Rigid sagebrush
Low sagebrush
Mountain mahogany
Bitterbrush
Mountain big sagebrush
Idaho fescue
Bluebunch wheatgrass
Ponderosa pine
Douglas-fir
Ninebark
Pinegrass
Ponderosa pine
Western larch
Lodgepole pine
Engelmann spruce
-------
Clarke and Bryce--226
Blue Mountain Basins (11k)
Symphoricarpus albus
Sarcobatus sp.
Festuca idahoensis
Poa sandbergii
Deschampsia caespitosa
Agrostis diegoensis
Carex sp.
Common snowberry
Greasewood
Idaho fescue
Sandberg's bluegrass
Tufted hairgrass
Redtop bentgrass
Sedges
-------
Clarke and Bryce--227
Table 3-Factors that vary the risk of sedimentation limiting salmonid populations by habitat
degradation4
Factor
Level of risk
Decreasing .. .
Climate
Winter rain
Summer rain,
Spring
Spring-fed
Winter snow
snowmelt
Erosive processes
Surface erosion,
Surface and mass
Mass erosion
channel erosion
erosion
Geology
Metavolcanics,
Sandstone,
Granitic
volcanic
siltstones
Topography lands
Gentle terrain
Moderate terrain
Steep terrain
Stream gradient
High (> 5%)
Moderate (1 -5%)
Low (< 1%)
Channel morphology
Narrow-deep
. . Shallow-wide
Riparian vegetation
Abundant
"From Everest and others (1987).
-------
Clarke and Bryce--228
Table 4—Relative importance of landscape factors controlling water quality elements
in the Pacific Northwest Coastal Ecoregion8
Water quality element
Controlling Intragravel
factor Nitrogen Phosphorus Turbidity Temperature DO
Climatic and
atmospheric
inputs
H
L
M
H
L
Geology and
soil
M
H
H
M
H
Stream order
M
M
M
H
M
Constrained
or
unconstrained
H
H
H
M
M
Vegetation
H
M
M
M
L
Tram Naiman and others (1992); importance is coded as H-high, M-medium, and L-
low.
-------
Clarke and Bryce-229
Table 5—Source maps used to delineate landscape-level ecoregions
Landscape
characteristic Scale
Topography 1 250,000
Subject
Digital Elev
Models (DEM)
1 100,000 Topographic Maps
Source
GS 1987
USGS
Minimum
mapping
Date GIS unit
1970- Y 60 x 90 m
1981 (66 - 98 yd)
grid
1979- N
1982
Soil
1 250,000 STATSGO (OR & WA) SCS
1 63,360 National Forest Soil
Resource Inventories
Carlson (Malheur)
Paulson (Ochoco)
Ehmer (Umatilla)
Wallowa-Whitman NF
1993 Y 625 ha
1544 ac
1974 N
1977 N 20 ha
50 ac
Ny N
1 20,000-
1 31,680
County Soil Surveys SCS*
1993 Y
1962- N 2 ha
1991 5 ac
Climate
1 500.000
Normal Annual
Precipitation
1'2,000,000 Precipitation Map (WA)
Taylor (OSU)
Miller (NOAA)
Digital WADNR
1994 Y 8 km (4.97
mi) gnd
1973 Y
1 250,000 Rain-on-snow Zones (WA) Brunengo (WADNR)
1991 Y
-------
Clarke and Bryce-230
Geology
1.500,000
1 250,000
1:24,000
1 250,000
1 62,500
1 24,000
1100,000
1 100,000
1 125,000
1 100,000
1 100,000
1.250,000
1 250,000
1.250,000
1.100,000
1-250,000
Geology (OR)
Geology (SEWA)
Granite Quad
(OR)
Walker and MacLeod 1991 Y
(USGS)
Johnson and Derkey 1993 Y
(USGS)
Brooks and others (USGS) 1982 N
Canyon City Quad (OR) Brown and Thayer (USGS) 1966 N
Desolation Butte Quad Evans (USGS) 1989 N
(OR)
Limber Jim Quad (OR) Ferns & Taubenack
(USGS)
Connell Quad (WA)
Pullman Quad (WA)
Gulick (WADNR)
Gulick (WADNR)
John Day Formation (OR) Robinson
Clarkston Quad (WA)
Schuster (WADNR)
Walla Walla Quad (WA) Schuster (WADNR)
Pendleton Quad (OR & Walker (USGS)
WA)
Grangeville
Quad (OR & WA)
Columbia River Basalt
Group (WA & ID)
Walker (USGS)
GS
Lignite and coal resources Ferns
(NE OR)
(DOGAMI)
Umatilla Indian Res (OR) Gonthier and Bolke
(USGS)
1994 N
1994
1994
1975
1993
1994
1973
N
N
N
N
N
N
1979 N
1980 N
1985 N
1993 N
-------
1 62,500
1 127,000
(approx)
1.95.000
(approx)
Vegetation 1 63,360
1 253,000
(approx)
1 253,000
(approx)
1 250,000
Clarke and Bryce-231
Upper Grande Ronde River Hampton and Brown 1964 N
basin (OR) (USGS)
Umatilla River basin (OR) Hogenson (USGS) 1964 N
Walla Walla River basin Newcomb (USGS) 1965 N
(OR & WA)
Forest type class (OR &
WA)
Forest type map (OR)
Forest type map (WA)
Actual vegetation (OR)
USFS
USFS
USFS
Kagan and Caicco (U of
ID)
1935- Y
1960
1936 N
1936 N
1992 Y 133 ha
328 ac
'Dyksterhuis 1981, Dyksterhuis and High 1985, Gentry 1991, Harrison and others 1962,1973, Hosier 1983; Johnson and
Makinson 1988, NRCS 1995, unavailable—Wheeler County, Oregon
-------
Clarke and Bryce--232
Table 6—Principal soil series names for map units defined primarily by soil characteristics.
Soil series names from STATSGO and county soil surveys
Soil series
Map unit Description Dominant subregion(s) names
gela/v Grassland eolian lacustrine valley 11 k Blue Mountain Basins Imbler
Alicel
Palouse
gla/v Grassland lacustrine deposits 11k Hot Lake
Conley
gaf Grassland alluvial fans 11 k Catherine
LaGrande
gloa Grassland loess and ash on 11 k
uplands
Watama
Ramo
McMurdie
-------
Clarke and Bryce--233
grca Grassland loess and residuum
over cemented alluvium
ggl/v Grassland gravelly lacustrine
deposits
gel Grassland cobbly loam
glowx/h Grassland warm, xeric loess w/
salts on plateaus and hills
11k Albee
10f Canyons and Lostine
Dissected Uplands Ladd
Langrell
10f Hurwal
Snell
Ateron
10f Nims
Weissenfels
Olical
Stember
Neissenberg
Peola
-------
Clarke and Bryce--234
glox/h Grassland xeric loess on plateaus 10f Neconda
ancl h'"s Ferdinand
Powwahkee
glocx/p Grassland cool, xeric loess on lot Sweitberg
plateaus Sne)l
glod/h Grassland deep loess on hills I0i Deep Loess Foothills Athena
Palouse
Peola
gxca Grassland xeric loess over
cemented alluvium
giomd/h Grassland moderately- deep
loess on hills
10c Umatilla Plateau
11k Blue Mountain
Basins
Bocker
Bridgecreek
Potomus
10c
Gwin
Waha
Rockly
Gurdane
-------
Clarke and Bryce-235
glos/h Grassland shallow loess on hills 10c Gwin
Gurdane
Rockly
glodx/h Grassland deep, xeric loess on
hills
10c Walla Walla
10b Scabland Loess
Islands
gbs/h Grassland biscuit scabland on 10c Morrow
hills Bakeoven
Condon
gloal Grassland deep loess and old 10c Pilot Rock
alluvium McKay
ges Grassland eolian sands I0e Pleistocene Lake Quincy
Basins Sagehill
Hezel
-------
Clarke and Bryce--236
ggfg Grassland glaciofluviate gravel 10e Quincy
deposits Hezel
ggfs Grassland glaciofluviate sand and 10e Ellisforde
silt deposits Sagemoor
ggo/t Grassland aridic glacial outwash I0e Farrell
on terraces Roloff
gaca Grassland aridic loess over lOe Shano
cemented alluvium Burke
gloda/h Grassland deep, aridic loess on 10b Scabland Loess Ritzville
hills Islands Ellisforde
-------
Clarke and Bryce-237
Table 7-Properties for ash-, sandstone-, and basalt-derived soil3
Properties
Soil Bulk Coarse
fragments density Porosity Clay fragment Available water
Ash
Basalt
Sandstone
Percent volume
73-77
65
40
Percent weight
3-10
20
10
Percent volume
0-4
30
0
Percent volume
26-30
13
14
Mg/mm
0.6-0.7
0.9
1.6
'Adapted from Harvey and others (1994).
-------
Clarke and Bryce-238
Table 8—Characteristics of soils by texture8
Soil texture
Porosity
Permeability
Percent
M/d
Clay
45-60
0.00001-0.001
Silt
20-50
0.01-10
Sand
30-40
10-100,000
Gravel
25-40
1,000-10,000,000
"From Gregory and Walling (1973).
-------
Clarke and Bryce--239
Table 9-Stocking per acre in trees at 6 inches
average diameter at breast height (DBH)a
Tree Number stocked
Ponderosa pine/grassland 63-238
Ponderosa pine 214-415
True fir 285-620
Subalpine fir 245-480
"Ten rings per inch of growth. From Hall
(1973a).
-------
Clarke and Bryce-240
Table 10—Characteristics of the forest zones
Forest zone
Elevation range
Annual precipitation range
Tree species
Shrub species
Grass and forb species
Juniper
Western Juniper (Juniperus
Bitterbrush (Purshia
Idaho fescue (Festuca
occidentalis)
tridentata)
idahoensis)
760-1400 m
Big sagebrush (Artemisia
Bluebunch wheatgrass
(2483-4593 ft)
tridentata)
(flgropyron spicatum)
Mountain-mahogany
Elk sedge (Carex geyeri)
20-30 cm
(Cercocarpus ledifolius)
(8-12 in)
Ponderosa pine
Ponderosa pine (Pinus
Mountain-mahogany
Elk sedge (Carex geyeri)
ponderosa)
(Cercocarpus ledifolius)
Idaho fescue (Festuca
900-1500 m
Grand fir {Abies grandis)
Common snowberry
Idahoensis)
(2953-4921 ft)
White fir (Ab/'es concolor)
(Symphoricarpos albus)
Pinegrass (Calamagrostis
Western larch (Larix
Mountain snowberry
rubescens)
35-76 cm
occidentalis)
(Symphoricarpos
Wheeler's bluegrass (Poa
(14-30 in)
Douglas-fir (Psuedotsuga
oreophilus)
nervosa)
menziesii)
Bitterbrush (Purshia
Bluebunch wheatgrass
Lodgepole pine (Pinus
tridentata)
(Agropyron spicatum)
contorta)
Mountain big sagebrush
Western white pine (Pinus
(Artemisia tridentata
monticola)
vaseyana)
-------
Clarke and Bryce-241
Douglas-fir (Psuedotsuga
Rocky Mountain maple
menziesii)
(fleer glabrum)
Ponderosa pine (Pinus
Mallow ninebark
ponderosa)
[Physocarpus
Western larch (Larix
malvaceus)
occidentalis)
Big huckleberry
Lodgepole pine (Pinus
Q/accinium
contorta)
membranaceum)
Mountain snowberry
(Symphoricarpos
oreophilus)
Common snowberry
(Symphoricarpos albus)
Birchleaf spiraea
(Spiraea betulfolia)
Oceanspray (Hoiodiscus
discolor)
Western fescue (Festuca
occidentalis)
Pinegrass (Calamagrostis
rubescens)
Elk sedge (Carex geyeri)
-------
Clarke and Bryce-242
True fir
Grand fir (Ab/'es grandis)
Pacific yew (Taxus
Oakfern (Gymnocarpium
White fir (Ab/'es concolor)
brevifolia)
dryopteris)
1500-2000 m
Ponderosa pine (Pinus
Thimbleberry (Rubus
Coolwort foamflower
(4921-6562 ft)
ponderosa)
parviflorus)
(Tiarella trifoliata
Lodgepole pine (Pinus
Twinflower (Linnaea
unifoliata)
63-115 cm
contorta)
borealis)
Hooker's fairy bells
(25-45 in)
Larch (Larix occidentalis)
Big huckleberry
(Disporum hookeri)
Douglas-fir (Pseudotsuga
Q/accinium
Sword fern {Polystichum
menziesii)
membranaceum)
munitum)
Western white pine (Pinus
Grouse huckleberry
Queen's cup beadlily
monticola)
(Vaccinium scoparium)
(Clintonia uniflora)
Pinegrass (Calamagrostis
rubescens)
Columbia brome (Bromus
vulgaris)
Lupine (Lupinus)
Elk sedge (Carex geyeri)
Ginger flsarum
caudatum)
-------
Clarke and Bryce-243
Subalpine fir
1300-1700 m
(4265-5577 ft)
Subalpine fir {Abies lasiocarpa) Big huckleberry
Engelmann spruce (P/'cea (Vaccinium
engelmannii)
Lodgepole pine (Pinus
contorta)
Western larch (Larix
occidentalis)
Whitebark pine (Pinus
albicaulis)
Douglas-fir (Pseudotsuga
menziesii)
membranaceum)
Fool's huckleberry
(iMenziesia ferruginea)
Grouse huckleberry
(Vaccinium scoparium)
Heartleaf arnica {Arnica
cordifoiia)
Queen's cup beadlily
(Ciintonia uniflora)
Sidebells pyrola (Pyroia
secunda)
Elk sedge (Carex geyeri)
-------
Clarke and Bryce--244
Table 11 —Species composition of the nonforested zones
Zone
Dominant shrub species
Dominant grass species
Aridic grassland Bitterbrush (Purshia tridentata) Needle and thread (Stipa comata)
Basin big sagebrush (Artemisia tridentata spp Sandberg's bluegrass (Poa sandbergii)
tridentata) Indian ricegrass (Oryzopsis hymenoides)
Bluebunch wheatgrass (figropyron spicatum)
Sagebrush Mountain big sagebrush {Artemisia tridentata
spp vaseyana)
Low sagebrush {Artemisia arbuscula)
Stiff sagebrush {Artemisia rigida)
Elk sedge (Carex Geyeri)
Idaho fescue (Festuca idahoensis)
Sandberg's bluegrass (Poa sandbergii)
Bluebunch wheatgrass flgropyron spicatum)
Xeric grassland
Bluebunch wheatgrass {Agropyron spicatum)
Sandberg's bluegrass (Poa sandbergii)
Mesic grassland
Idaho fescue (Festuca idahoensis)
Prairie junegrass (Koeiaria cristata)
Bluebunch wheatgrass {Agropyron spicatum)
Hood's sedge (Carex hoodii)
-------
Clarke and Bryce--245
Shrublands Ninebark (Physocarpus malvaceus)
Common snowberry (Symphoricarpos albus)
Oceanspray (Holodiscus discolor)
Cherries (Prunus)
Mountain-mahogany (Cercocarpus ledifolius)
Bluebunch wheatgrass (Agropyron spicatum)
Idaho fescue (Festuca idahoensis)
Wheeler's bluegrass (Poa nervosa)
Alpine Alpine sagebrush (flrtemisia tridentata spp.
vaseyana)
Green fescue (Festuca viridula)
Idaho fescue (Festuca idahoensis)
Elk sedge (Carex geyeri)
-------
Clarke and Bryce--246
FOOTNOTES
1 Personal communication. 1995. J. Omernik, U.S. Environmental Protection Agency,
National Health and Environmental Effects Laboratory, Western Ecology Division.
Personal communication. 1995. E. Bishop, Oregon Watershed Health Program.
^erm range of natural variability coined by Rhodes and McCullough (1994).
Principles used in defining the landscape-level ecoregions for anadromous fish and
their habitat may make them applicable to other aquatic resource issues as well as
some terrestrial resource issues. An assessment of the usefulness of the classification
for other resources can be made by evaluating the map and reading the rationale
section. Classes can be combined to meet the needs of other resource projects.
5Personal communication. 1995. J.J. Rhodes, Columbia River Inter-Tribal Fish
Commission.
6SCS became the U.S. Department of Agriculture's National Resources Conservation
Service (NRCS) in late 1994.
'Personal communication. 1994. M.L. Ferns, Wallowa-Whitman National Forest.
-------
Clarke and Bryce-247
Personal communication. 1994. E. Bishop, Oregon Watershed Health Program.
9Now referred to as the Columbia Plateau Ecoregion.
10Personal communication. 1995. Bruce Mcintosh, Forest Science Department, Oregon
State University.
-------
Clarke and Bryce--248
FIGURE CAPTIONS
Figure 1 — Example of the interplay between ecoregions and watersheds.
Figure 2—Photos representing the diversity among subregions and landscape-level
ecoregions for the Columbia Plateau and Blue Mountain Ecoregions.
Figure 3—Level IV ecoregions of the Columbia Plateau and Blue Mountain Ecoregions
of Washington, Oregon, and Idaho.
Figure 4—A Boundary transition-width map included with the ecoregion/subregion
map for the Columbia Plateau.
Figure 5—Study area watersheds.
Figure 6—Landscape-level ecoregions.
-------
Mountain*
High Mountain*
ft C\uft I
-------
Figure 2 - Photos
Figure 6 - Landscape-level ecoregions
MAP AND PHOTOS NOT AVAILABLE AT PRESENT.
Available from PNW Research Station
when published
-------
COLUMBIA PLATEAU E C 0 R E G I 0 NIS U B R E G I 0 N BOUNDARY TRANSITION WIDTHS
Sondffl A rhtele", Chris W Knlsgaord1, Jomes M Omeinik: and Suzonne M Piersort1
'US Environmental Praleclion Agency
'Wanlecrt Environmental lechnoloqy Inc
Corralhs. Oregon 973J3
Couallti, Oregon 97333
J
COLUMBIA PLATEAU (10)
Channeled Scablands (IGo)
Scabland loess islands (10b)
Umatilla ploleou (10c)
Okonogon drift hills (lOd)
Pleistocene lake bosin (10e)
Canyons and dissected uplands (lOf)
Yakima folds (lOg)
Polouse hills (10h)
Deep loess foothills (lOi)
Nez Perce prairie (10j)
CASCAOE5 (4)
EASTERN CASCADES SLOPES AND FOOTHILLS (9)
BLUE MOUNTAINS (11)
NORTHERN ROCKIES (15)
ecoregion boundary
subrsgion boundary
I I fully boundary
B 0 n. B 16
STUDY
AREA
I h i s map illuSlroles Ihe 'dative ¦ iij I hi fll ecoregion ond Subreqion
boundaries Rpqionol boundaries art usually portrayed by single
flarrct linn on mops t>ul I he y r cpi e)cnl tronsihon lonej ol »o*
• idtfu on the qrovnd «h«re rherqctei iStiCJ ol one ar«a blend • for to be 01 mdiioltd by component mops and olher sources
ol information In some area; the (honqe i] distinct ond abrupt
such as there the Channeled Scablonds(IOa) metl the Loess Islands(lOb)
in oiner oieai the transition is brood end ihe boundaif is
mare dillxvll to determine Ihe lu//r boundaries define oreai ol
uncerlomly or areas 'here there n a leteroqeneous moiO'C ol
cItaraclerislxi Irom each of Ih« odiocent siibreqions It should
be remembered lint boundaries on o nap are orlilacls abstractions
end oppronmalions Become rl is necessary and dejircbie to dro*
boundaries our in ten I here is to provide the mop user *itfi roore
mlormaiisn oDout the proms nature ond mtcnini ol the ecoreqion
delmeotionj
36 0 km J6 72
ajms egtri
-------
Seven Contiguous Watersheds in
N.E. Oregon and S.E Washington
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
h.8
ORD CLEARANCE FORM
1 EPA Report No.
2. Product Code
3. Lab/Office No.
mivti-ck-tW
4 Copyright Permission
_Yes(Attached) _No _N/A
5A. Original Document Title : Hierarchical Subdivisions of the Columbia Plateau and Blue Mountains Ecoregions
5B Final Document Title, if changed: Hierarchical Subdivisions of the Columbia Plateau and Blue Mountains Ecoregions,
Oregon and Washington
6. Author(s), Affiliation, and Address (Identify EPA authors with Lab/Office)
Editors'
Sharon E. Clarke Sandra A. Bryce
Faculty Research Assistant Dynamac International, Inc.
Oregon State University % USEPA-NHEERL-WED
Corvallis, OR 97331 200 SW 35th Street
Corvalfis, OR 97333
Co-Author Section 1 & 2
James M. Omemik
USEPA-NHEERL-WED
200 SW 35th Street
Corvalts, OR 97333
7 ORD PLbfcatxx^ArYxxjrcerrenl Category
8 Project Officer/Telephone:
Jim Omemik (541) 754-4458
9. Issue/Subissue/Project Area/Project
6^103060^3572
10. Contract/IAG/Assistance Agreement No
Dynamac International, I no 68-C5-0005
11 Product (check one)
A Peer Reviewed Journal Article(See block 12)
A Published Report
A Symposium Paper & Book Chap.(See block 12)
A Internal Report (distribution restricted to EPA)
A Newsletter, Research Brief, and Issue Paper
A Unpublished Report
A Other
_ Special Attention Publication
_ Yes (zinger attached)
No
12 Bibliographic Citation (include Month/Year)
199X
Gen. Tech. Rep PNW-GTR-XXX. Portland, OR- U S
Department of Agriculture, Forest Service, Pacific
Northwest Research Station, xxp.
Accepted.
Published
13 Technical Information (Program) Manager/Sender
'j/2
Stgnahjre J /j
Data
14 Signature/Date
^7 Yj/i« si (2-
------- | |