Macroecology, Paleoecology, andEcological Integrity of TerrestrialSpecies and Communities of theInterior Columbia Basin andNorthern Portions of the Klamathand Great Basins

United StatesDepartment ofAgriculture

Forest Service

Pacific NorthwestResearch Station

United StatesDepartment of theInterior

Bureau of LandManagement

General TechnicalReportPNW-GTR-410October 1998 B.G. Marcot, L.K. Croft, J.F. Lehmkuhl, R.H. Naney, C.G. Niwa,

W.R. Owen, and R.E. Sandquist

AuthorsB.G. MARCOT is a wildlife ecologist, Pacific Northwest Research Station, Forestry SciencesLaboratory, P.O. Box 3890, Portland, OR 97208-3890; L.K. CROFT is a plant ecologist, OchocoNational Forest, P.O. Box 490, Prineville, OR 97754; J.F. LEHMKUHL is a research wildlifebiologist, Pacific Northwest Research Station, Forestry Sciences Laboratory, 1133 N. WesternAvenue, Wenatchee, WA 98801; R.H. NANEY is a wildlife biologist, Okanogan National Forest,1240 S. Second, Okanogan, WA 98840; C.G. NIWA is a research entomologist, Pacific NorthwestResearch Station, Forestry Sciences Laboratory, 3200 S.W. Jefferson Way, Corvallis, OR 97331; W.R.OWEN is a forest botanist, Boise National Forest, 1750 Front Street, Boise, ID 83702; and R.E.SANDQUIST is a regional entomologist, Pacific Northwest Region, P.O. Box 3623, Portland, OR97208-3623.

Macroecology, Paleoecology, and EcologicalIntegrity of Terrestrial Species andCommunities of the Interior Columbia RiverBasin and Northern Portions of the Klamathand Great Basins

B.G. Marcot, L.K. Croft, J.F. Lehmkuhl, R.H. Naney,C.G. Niwa, W.R. Owen, and R.E. Sandquist

Interior Columbia Basin Ecosystem Management Project:Scientific Assessment

Thomas M. Quigley, Editor

U.S. Department of AgricultureForest ServicePacific Northwest Research StationPortland, OregonGeneral Technical Report PNW-GTR-410October 1998

AbstractMarcot, B.G.; Croft, L.K.; Lehmkuhl, J.F.; Naney, R.H.; Niwa, C.G.; Owen, W.R.; Sandquist,

R.E. 1998.Macroecology, paleoecology, and ecological integrity of terrestrial species and com-munities of the interior Columbia River basin and northern portions of the Klamath and GreatBasins. Gen. Tech. Rep. PNW-GTR-410. Portland, OR: U.S. Department of Agriculture, ForestService, Pacific Northwest Research Station. 131 p. (Quigley, Thomas M., tech. ed. InteriorColumbia River Basin Ecosystem Management Project: scientific assessment).

This report presents information on biogeography and broad-scale ecology (macroecology) of selectedfungi, lichens, bryophytes, vascular plants, invertebrates, and vertebrates of the interior Columbia Riverbasin and adjacent areas. Rare plants include many endemics associated with local conditions. Potentialplant and invertebrate bioindicators are identified. Species ecological functions differ among communi-ties and variously affect ecosystem diversity and productivity. Species of alpine and subalpine com-munities are identified that may be at risk from climate change. Maps of terrestrial ecological integrityare presented.

Keywords: Macroecology, paleoecology, ecological integrity, terrestrial communities, ecosystems,wildlife, fungi, lichens, bryophytes, vascular plants, invertebrates, arthropods, mollusks, amphibians,reptiles, birds, mammals, endemism, interior Columbia River basin, Klamath Basin, Great Basin.

PrefaceThe Interior Columbia Basin Ecosystem Management Project was initiated by the Forest Service andthe Bureau of Land Management to respond to several critical issues including, but not limited to, forestand rangeland health, anadromous fish concerns, terrestrial species viability concerns, and the recentdecline in traditional commodity flows. The charter given to the project was to develop a scientificallysound, ecosystem-based strategy for managing the lands of the interior Columbia River basin adminis-tered by the Forest Service and the Bureau of Land Management. The Science Integration Team wasorganized to develop a framework for ecosystem management, and assessment of the socioeconomicand biophysical systems in the basin, and an evaluation of alternative management strategies. Thispaper is one in a series of papers developed as background material for the framework, assessment, orevaluation of alternatives. It provides more detail than was possible to disclose directly in the primarydocuments.

The Science Integration Team, although organized functionally, worked hard at integrating the ap-proaches, analyses, and conclusions. It is the collective effort of team members that provides depth andunderstanding to the work of the project. The Science Integration Team leadership included deputyteam leaders Russel Graham and Sylvia Arbelbide; landscape ecology—Wendel Hann, Paul Hessburg,and Mark Jensen; aquatic—Jim Sedell, Kris Lee, Danny Lee, Jack Williams, Lynn Decker; economic—Richard Haynes, Amy Horne, and Nick Reyna; social science—Jim Burchfield, Steve McCool, andJon Bumstead; terrestrial—Bruce Marcot, Kurt Nelson, John Lehmkuhl, Richard Holthausen, andRandy Hickenbottom; spatial analysis—Becky Gravenmier, John Steffenson, and Andy Wilson.

Thomas M. QuigleyEditor

United StatesDepartment ofAgriculture

Forest Service

United StatesDepartment ofthe Interior

Bureau of LandManagement

Interior ColumbiaBasin EcosystemManagement Project

Executive Summary

Plant Biodiversity

Biodiversity of the interior Columbia River basin assessment area (hereafter referred to as the basinassessment area) is better known than is global biodiversity, but much systematic and inventory workremains to be done on soil micro-organisms, fungi, and invertebrates. Rare plants and lichen groupsoccur throughout the basin assessment area, and rare plants were particularly diverse in the Basin andRange, Columbia Plateau, Blue Mountains, northern Idaho, and the southern regions of the east side ofthe Cascade Range. Most rare plant life forms were hemicryptophytes and cryptophytes. Diversity ofrare plants was not particularly oriented to specific topographic conditions, as many rare plants occurin various mostly azonal (atypical) conditions. Most rare plants are declining with fewer remainingstable; by definition, none is significantly increasing. Many paleoendemic (ancient endemic) andneoendemic (recently endemic) plants occur in the basin assessment area. Lichens disperse mostlyby gravity; rare fungi by wind or vertebrates; and rare vascular plants by gravity, wind, water, andvertebrates. Most of the rare plants are pollinated probably by various invertebrates. Several lichens,bryophytes, plants, and invertebrates can serve as bioindicators of environmental conditions andecosystem health.

Animal Biodiversity

Invertebrates of the basin assessment area disperse by many means. One way involving coevolutionbetween invertebrates and vertebrates is phoresis (dispersal by hitching a ride on vertebrate organisms).Disease and parasites are poorly studied in the basin assessment area and need further work to deter-mine distribution, frequency of occurrence, and effects on plants and animals of interest to manage-ment. Endemic vertebrates include some amphibians, birds, and mammals. The ranges of many verte-brate species extend beyond the basin assessment area, including semiendemics (species occurringonly within the inland West or the basin assessment area for only a portion of the year, as with somemigrants). Thus, to provide for their rangewide viability, conditions beyond the basin assessment areaneed to be known and addressed. Numbers of breeding birds by state match other published predictions.Most bird species were ranked as common, with fewer uncommon, rare, abundant, or irregular. Moreabundant and irregularly occurring bird species occur toward midcontinent than elsewhere in the basinassessment area. Most birds are resident or summer breeders, with fewer being migrants or winteringonly in the basin assessment area.

Ecological Functions of Species Among Communities

We present species function profiles that relate key ecological functions (KEFs)—those major rolesthat each species plays in their environment—of vertebrates to occurrence by vegetation community.Key ecological functions of species can affect ecosystem productivity, diversity, and sustainability;and function profiles of species help summarize the diversity of species functions in ecological com-munities of the basin assessment area. Function profiles of species display the degree of functional re-dundancy (number of species sharing the same KEF) within each community. Some functions, particu-larly cavity excavation in snags and primary burrow excavation in soil, are highly variable among com-munities in terms of number of species. Late seral forests provide the greatest redundancy in carrionfeeding, general nutrient cycling, and primary cavity excavation functions of vertebrates. We discussfunctions and functional redundancy of other communities as well. We pose several hypotheses relatingspecies functional redundancy to ecosystem resiliency and changes in species community structure, asone facet of ecosystem management. The greatest vertebrate functional diversity is found in early seralmontane forest, followed by upland woodlands and riparian woodlands, and upland shrublands. No onecommunity, however, contains all ecological functions of species.

Effects of Climate Change

Near-future changes in regional climates from human activities may be complex, involving increasinginterseasonal and interannual variations in precipitation and temperature. Species preadapted to suchvariations likely will persist during climate changes. The fate of vertebrates within alpine tundra andsubalpine forests may differ individualistically according to latitudinal shifts of these habitats and therange of plasticity of each species. Studies of climate change suggest that patterns of climate, vegeta-tion, and plant and animal species occurrence in the basin assessment area during historic (since 1800)times do not match those over the past four centuries or prehistorically during the Quaternary or Terti-ary periods. This has important ramifications for reinterpreting the range of natural historic conditions(variations), which do not represent conditions under which species evolved.

Prehistoric Conditions

The assemblage of mammalian taxa of the basin assessment area has changed greatly since Tertiaryand Quaternary periods, in accord with changes in climate and vegetation, although overall diversityof species among families and orders has remained more or less constant. Many species, and somefamilies and orders, have declined or become extinct, whereas others have radiated and grown. Weanticipate future losses and gains in species, families, and orders of plants and animals over evolu-tionary time, although prehistoric trends in extinction and speciation (evolution of new species) donot represent short-term ecological conditions.

Ecological Integrity of Terrestrial Communities

We mapped various aspects of terrestrial ecological integrity, including components of species viability(as threatened, endangered and candidate species, locally endemic species, distribution of rare plants,and key habitat corridors); long-term evolutionary potential of species (as disjunct populations and bio-diversity and endemism hot spots); and multiple ecological domains (as peripheral species and uniquespecies assemblages, and the full set of large vertebrate carnivores). Results are shown by watershed.

This page has been left blank intentionally.Document continues on next page.

Introduction

History and Impetus

The Science Integration Team of the InteriorColumbia Basin Ecosystem Management Project(ICBEMP) conducted ecological assessmentsduring 1993-96 as a joint venture among USDAForest Service (FS), USDI Bureau of Land Man-agement (BLM), and many cooperators. The Sci-ence Integration Team developed the first com-prehensive database and assessment of terrestrialplant and animal ecology of the interior ColumbiaRiver basin and northern portions of the Klamathand Great Basins in the United States (collec-tively, and hereafter, referred to as the basinassessment area) (Marcot and others 1997).This document expands on findings related tomacroecology and ecological integrity of ter-restrial species and communities of the basinassessment area.

Our ultimate aim is to contribute to a scientificfoundation for ecosystem management of ter-restrial ecosystems in the basin assessment area.According to the Ecological Society of America(1995), a scientific basis for ecosystem manage-ment should provide adequate information on bi-ological diversity and on the function and dynam-ics of ecosystems, and acknowledge the “open-ness and interconnectedness” of ecosystems.

Other facets of ecosystem management shouldentail study of past and current natural ranges ofconditions (Morgan and others 1994) and theecological roles of species interactions (Willson1996). This report provides information on thesetopics in terrestrial ecosystems of the basinassessment area.

Definitions

In this report,macroecologyrefers to habitat andenvironmental characteristics of individual ter-restrial species or species groups at the broadscale (sensu Brown 1995).Terrestrial speciesrefers to species whose life histories are mostlyconfined to nonaquatic environments, althoughsome largely aquatic vertebrate species such asamphibians and otters (but not fish) are included.

Hydrologic unit codes (HUCs), as used in theICBEMP ecology assessments, are standard delin-eations of watershed and subwatershed land areasbased on river drainage patterns (Jensen andBourgeron 1994). We used two levels of HUCs:4th- and 6th-level hydrologic units, correspondingto subbasins and subwatersheds, respectively.

Broad scaleis defined here as such basin assess-ment area-wide depictions that use characteriza-tions of 1-km2 cells with summarization by4th-code hydrologic units.Mid scale is definedas characterizations at 4-ha resolution with sum-marization by 6th-code hydrologic units.

CHAPTER 1

1

Study Area

Area and Boundaries

The basin assessment area encompasses58 361 400 ha. It is confined on the north by theUnited States-Canada border; on the east by theContinental Divide; on the west by the crest ofthe Cascade Range; on the south by the Oregon-California and Oregon-Nevada borders; andelsewhere by the extent of the Columbia Riverdrainage (fig. 1). One major portion of theColumbia River basin in the United States isthe tributary Snake River basin, which includescentral and southern Idaho and western Montana.

It extends toward the Continental Divide into theGreater Yellowstone Ecosystem and southernYellowstone National Park. The entire ColumbiaRiver basin encompasses about 67 466 000 haand extends north into British Columbia, Canada,(15 percent of the entire area of the ColumbiaRiver basin proper) and west past the crest of theCascade Range to the Pacific Ocean, outside theboundaries of the basin assessment area (fig. 1).

The basin assessment area was defined by the di-rectors of the ICBEMP according to both agencyplanning needs and hydrologic (watershed) cri-teria. The existing Northwest Forest Plan and itspreceding ecological assessment (Forest Eco-system Management Assessment Team [FEMAT]1993) had already addressed the area within therange of the northern spotted owl, including theeastern slope of the Cascade Range. The ICBEMPwas directed to reevaluate the FEMAT assess-ments along the east side of the Cascade Range,given the different composition of forests foundthere as compared to forests west of the crest ofthe Cascade Range.

The basin assessment area includes portions ofseven Western states, including major portions ofWashington, Oregon, and Idaho, as well as west-ern Montana and corners of Wyoming, Utah, andNevada. Although outside the Columbia Riverbasin per se, the northern portions of the KlamathBasin (1.5 million ha of the basin assessmentarea) and Great Basin (4.2 million ha of the basinassessment area) in southern Oregon were in-cluded in the basin assessment area. The purposewas to complete broad-scale land planning andassessments on BLM and FS lands within theregional boundaries of these agencies in Oregon.The portions of the Klamath Basin and northernGreat Basin are therefore also included in theassessments of this report, although they do notform a discrete ecological unit.

Figure 1—Boundaries of the Columbia River basin inNorth America (shaded area), and the basin assess-ment area (heavy solid line) and Western States (lightsolid lines) in the United States.

CHAPTER 2

2

By Land Ownership

Bureau of Land Management and FS lands con-stitute 53 percent of the land within the basinassessment area, with the remainder in private(38 percent), State and other Federal (4 percent),tribal (4 percent), and national park and othernon-Federal wilderness (1 percent) lands (fig. 2).The basin assessment area includes portions ofmajor ecosystems of interest to Federal broad-scale planning concerns, including the GreaterYellowstone Ecosystem, the mountain ranges ofthe North Cascades/Bitterroots, the Selkirks, theCabinet-Yaks, the High Cascades, the KlamathBasin, the Great Basin, and others.

By Type of Management

Most of the basin assessment area is roaded andactively managed for natural resource use (fig. 3).Of all ownerships within the basin assessmentarea, 81 percent of the area is roaded and activelymanaged (43 percent BLM-FS plus 38 percentother lands), 10 percent is unroaded or not ac-tively managed (9 percent BLM-FS plus 1 per-cent National Park Service and other wildernesslands), and 10 percent is tribal, state, and otherpublic lands of varying use, mostly roaded andactively managed. Of all BLM and FS landswithin the basin assessment area, 83 percent isroaded and actively managed for natural resourceuse and extraction, and 17 percent is character-ized as unroaded.

Ecological Reporting Units

The Science Integration Team mapped 13 ecolog-ical reporting units (ERUs) in the basin assess-ment area. Ecological reporting units are geo-graphically defined areas identified by the Sci-ence Integration Team by use of multiple land-scape and watershed criteria and were basedon subsection delineations of 6th-code HUCs(Jensen and others, in prep.). Ecological reportingunits represent large geographic areas with com-mon landforms and drainage patterns, such as theUpper Snake headwaters, northern glaciatedmountains, and the east side of the CascadeRange in Oregon.

Hydrologic Delineations andBiophysical Characterizations

The basin assessment area was partitioned into164 subbasins or 4th-code HUCs, and 7,733 sub-watersheds or 6th-code HUCs. The subbasinsaveraged 356 496 ha and ranged from 4700 to1 080 500 ha. The subwatersheds averaged 7880ha and ranged from 96 to 86 500 ha. The 4th- and6th-code HUCs of the basin assessment area werecharacterized by sundry biophysical parameters,including historic and current vegetation covertypes and structural stages (Hann and others1997), historic and current terrestrial vegetationcommunities (Hann and others 1997), highestand lowest annual precipitation (available datafrom 1989), landform, highest and lowest eleva-tion, and topographic relief.1

The basin assessment area also was characterizedby landform. It contains nine major landformsranging from arid grasslands and lowland plainsand valleys, to intermontane basins and breaks, tosteep mountains and glaciated ranges, and a widevariation in topography, precipitation, and climatepatterns.

Vegetation Cover Types andTerrestrial Vegetation Communities

This section describes the vegetation classifica-tions used in this analysis and summarizes thecurrent and historic vegetation character of thebasin assessment area as presented in variouschapters in Quigley and Arbelbide (1997).

Vegetation characterizations of the basinassessment area—The current vegetation condi-tion of the basin assessment area was character-ized by use of 44 vegetation cover types (table 1)and 24 terrestrial vegetation communities (table 2)(Hann and others 1997). The vegetation com-munities are combinations of 41 vegetation covertypes (table 3) and structural stages (see Hann andothers 1997, for descriptions and methods). Forsome analyses, vegetation communities betterrepresent general conditions than do their compo-nent cover types.

1 Biophysical data from Intermountain Fire SciencesLaboratory, USDA Forest Service, P.O. Box 8089,Missoula, MT 59807.

3

Figure 2—Distribution (hectares) of land ownership and administration in the basin assessment area(source: ICBEMP, GIS 1-km2 raster data).

Figure 3—Distribution of land by management emphasis, and by unroaded or roaded conditions, in the basinassessment area (source: ICBEMP, GIS 1-km2 raster data).

4

Table 1—Names of 44 vegetation cover types denoted forthe basin assessment areaa

Cover-type code Vegetation cover-type name

CRB003 Shrub or herb/tree regenerationCRB005 Alpine tundraCRB006 BarrenCRB007 Herbaceous wetlandsCRB008 Pacific silver fir/mountain hemlockCRBS01 Juniper woodlandsCRBS02 Mixed-conifer woodlandsCRBS03 Juniper/sagebrushCRBS04 Big sagebrushCRBS05 Shrub wetlandsCRBS06 AgropyronbunchgrassCRBS07 Native forbCRBS08 Exotic forbs/annual grassCRBS09 Grand fir/white firCRBS10 White bark pine/alpine larchCRBS11 Red firCRBS12 Cropland/hay/pastureCRBS13 Fescue-bunchgrassCRBS19 UrbanCRBS20 WaterSAF205 Mountain hemlockSAF206 Engelmann spruce/subalpine firSAF208 Whitebark pineSAF210 Interior Douglas-firSAF212 Western larchSAF215 Western white pineSAF217 AspenSAF218 Lodgepole pineSAF219 Limber pineSAF227 Western redcedar/western hemlockSAF233 Oregon white oakSAF235 Cottonwood/willowSAF237 Interior ponderosa pineSAF243 Sierra Nevada mixed coniferSAF245 Pacific ponderosa pineSRM104 Antelope bitterbrush/bluebunch wheatgrassSRM322 Mountain mahoganySRM402 Mountain big sagebrushSRM406 Low sageSRM414 Salt desert shrubSRM421 Chokecherry/serviceberry/rose

a Source: Hann and others (1997). Only 41 cover types are listed here;3 additional cover types were dropped from the overall list after initialanalyses were conducted for this report. The dropped cover types wereeither extremely rare or found to not occur in the basin assessment areaafter subsequent study.

5

Table 2—Names and abbreviations of 24 terrestrial vegetationcommunities denoted for the basin assessment areaa

Abbreviation Terrestrial vegetation community name

AGRICUL AgriculturalALPINE AlpineEXOTICS ExoticsMIXCONES Early seral montane forestMIXCONLM Late seral montane multilayerMIXCONLS Late seral montane single layerMIXCONMS Mid seral montane forestNFHERBUP Upland herbNFUPSHRB Upland shrubPINFORES Early seral ponderosa pine forestPINFORLM Late seral ponderosa pine forest multilayerPINFORLS Late seral ponderosa pine forest single layerPINFORMS Mid seral ponderosa pine forestRIPHERB Riparian herbRIPSHRUB Riparian shrubRIPWDLND Riparian woodlandROCK Rock/barrenSUBFUPES Early seral subalpine forestSUBFUPLM Late seral subalpine forest multilayerSUBFUPLS Late seral subalpine forest single layerSUBFUPMS Mid seral subalpine forestURBAN UrbanWATER WaterWOODLDUP Woodland upland

a Source: Hann and others 1997.

6

Table 3—Crosswalk between terrestrial vegetationcommunities and vegetation cover typesa

Terrestrial vegetation Vegetation cover-typecommunity abbreviation code

AGRICUL CRBS12ALPINE CRB005EXOTICS CRBS08MIXCONES CRB003MIXCONES CRB008MIXCONES CRBS09MIXCONES CRBS11MIXCONES SAF210MIXCONES SAF212MIXCONES SAF215MIXCONES SAF218MIXCONES SAF227MIXCONES SAF243MIXCONLM CRB008MIXCONLM CRBS09MIXCONLM CRBS11MIXCONLM SAF210MIXCONLM SAF212MIXCONLM SAF215MIXCONLM SAF218MIXCONLM SAF227MIXCONLM SAF243MIXCONLS CRBS09MIXCONLS SAF210MIXCONLS SAF212MIXCONLS SAF215MIXCONLS SAF218MIXCONLS SAF227MIXCONLS SAF243MIXCONMS CRB008MIXCONMS CRBS09MIXCONMS CRBS11MIXCONMS SAF210MIXCONMS SAF212MIXCONMS SAF215MIXCONMS SAF218MIXCONMS SAF227MIXCONMS SAF243NFHERBUP CRBS06NFHERBUP CRBS07NFHERBUP CRBS13NFUPSHRB CRBS04NFUPSHRB SRM104NFUPSHRB SRM322NFUPSHRB SRM402NFUPSHRB SRM406

7

Table 3—Crosswalk between terrestrial vegetationcommunities and vegetation cover typesa (continued)

Terrestrial vegetation Vegetation cover-typecommunity abbreviation code

NFUPSHRB SRM414NFUPSHRB SRM421PINFORES SAF237PINFORES SAF245PINFORLM SAF237PINFORLM SAF245PINFORLS SAF237PINFORLS SAF245PINFORMS SAF237PINFORMS SAF245RIPHERB CRB007RIPSHRUB CRBS05RIPWDLND SAF217RIPWDLND SAF235ROCK CRB006SUBFUPES CRBS10SUBFUPES SAF205SUBFUPES SAF206SUBFUPES SAF208SUBFUPLM CRBS10SUBFUPLM SAF205SUBFUPLM SAF206SUBFUPLM SAF208SUBFUPLS SAF205SUBFUPLS SAF208SUBFUPMS CRBS10SUBFUPMS SAF205SUBFUPMS SAF206SUBFUPMS SAF208URBAN CRBS19WATER CRBS20WOODLDUP CRBS01WOODLDUP CRBS02WOODLDUP CRBS03WOODLDUP SAF219WOODLDUP SAF233

a Terrestrial vegetation communities are a combination of vegetationcover types and structural stages (not shown here). See table 2 for theterrestrial vegetation community names associated with the vegetationcommunity abbreviations and table 1 for vegetation cover-type namesassociated with the cover-type codes.

8

Present vegetation conditions—At present, thefive most dominant vegetation cover types of thebasin assessment area (all lands and ownerships)are big sagebrush (16 percent of the assessmentarea), cropland-hayland-pasture (16 percent), in-terior ponderosa pine (Pinus ponderosa)2 forest(10 percent), interior Douglas-fir (Pseudotsugamenziesii) (8 percent), and lodgepole pine (Pinuscontorta) forest (7 percent). The cropland-hayland-pasture has been carved mostly fromnative grassland types.

Terrestrial vegetation communities are a com-bination of cover type and structural stage. Assuch, the terrestrial vegetation communities showsimilar patterns of dominance as those of vegeta-tion cover types and structural stages, but addi-tionally help reveal differences among some ofthe general seral forest conditions. The five mostdominant terrestrial vegetation communities ofthe basin assessment area (all lands and owner-ships) at present are upland shrub (26 percent ofthe assessment area), mid seral montane forest(17 percent), agricultural (16 percent), early seralmontane forest (8 percent), and mid seral ponder-osa pine forest (8 percent) (fig. 4).

Change since historic conditions—Hann andothers (1997) present data on change of vegeta-tion cover types and terrestrial vegetation com-munities since early historic times (about early1800s). The vegetation cover types having under-gone the greatest historic decline in percentageof original area occupied in the basin assessmentarea (that is, less than -50 percent change fromhistoric area), are western white pine (Pinusmonticola)forest (-96 percent change), whitebarkpine (Pinus albicaulis)/subalpine larch (Larixlyallii ) forest (-95 percent), native forb (-92 per-cent),Agropyronbunchgrass (-68 percent),

Fescuebunchgrass (-66 percent), cottonwood(Populusspp.)/willow (Salixspp.) (-64 percent),and shrub wetlands (-61 percent).3 The vegeta-tion cover types having undergone the greatesthistoric increases (that is, > +50 percent changefrom historic area) are mountain hemlock (Tsugamertensiana) forest (53 percent), Pacific ponder-osa pine (80 percent), juniper (Juniperusspp.)/sagebrush (Artemesiaspp.) (163 percent), moun-tain mahogany (Cercocarpus ledifolius) (406 per-cent), western redcedar (Thuja plicata) westernhemlock (Tsuga heterophylla) (853 percent),grand fir (Abies grandis) white fir (Abiesconcolor) (965 percent), and Pacific silver fir(Abies amabilis) mountain hemlock (1,733percent). Many of the increases in these foresttypes have resulted from selective high-gradelogging (for example, of old-growth ponderosapine trees) and fire-suppression activities, result-ing in major changes in tree dominance and standstructure (see Hann and others [1997], for data onabsolute changes of each cover type).

The terrestrial vegetation communities decreasingthe most since historic time (that is, less than -50percent) (fig. 5) include late seral lower montane(including ponderosa pine) single-layer forest(-81 percent change), early seral lower montaneforest (-77 percent), upland herb (-67 percent),and late seral subalpine multilayer forest (-64 per-cent). Terrestrial vegetation communities mostincreasing since historic time (that is, greater than50 percent) include mid seral lower montane for-est (53 percent) and mid seral montane forest (59percent). Urban areas, exotic communities, andagricultural areas also have greatly increased, asthey were virtually absent under early historicconditions. They were carved mostly from nativeupland grassland, herbland, and shrubland com-munities (fig. 6; see Hann and others [1997] fordefinitions of urban, exotic, and agricultural com-munities). Changes in rangeland communitieshave been due to increased fire frequency asso-ciated with the invasion of exotic grasses, primar-ily cheatgrass (Bromus tectorum).

2 Authorities for species can be found in the followingreferences: mosses and nonvascular plants, Schofield(1980) and Vitt and others (1988); plants and allies,Hitchcock and others (1969); invertebrates, Furniss andCarolin (1980); amphibians and reptiles, Collins (1990);birds, AOU (1983 with supplements); mammals, Jones andothers (1992) and Wilson and Reeder (1993). 3 In addition, Sierra Nevada mixed-conifer forest has under-

gone significant decline, although this cover type may havebeen identified as an artifact within the basin assessmentareas and is not included here in the analyses.

9

Figure 4—Current distribution of terrestrial vegetation communities in the basin assessment area (source:ICBEMP, GIS 1-km2 raster data).

Figure 5—Percentage of changes of area covered by terrestrial vegetation communitiesfrom historic (early 1800s) to current time, in the basin assessment area: * = nothistorically present, lowmont = lower montane, subalp = subalpine (source: ICBEMP,GIS 1-km2 raster data).

10

Combining terrestrial vegetation communities ofsimilar seral conditions but different forest typessimplifies the picture (fig. 7). The greatest his-toric declines have occurred in late seral forests(-44 percent) and native upland types (includingnative grasslands and shrublands, collectively-38 percent). The greatest historic increases haveoccurred in mid seral forests (48 percent) as wellas urban, exotics, and agricultural communities.

Since historic times, there has been a slight in-crease in overall forest cover (fig. 8) because firesuppression has encouraged expansion of juniperwoodland and interior Douglas-fir forest into na-tive sagebrush and grassland communities. Of allforest communities combined, early seral forestshave declined slightly, mid seral forests increasedby about half again, and late seral forests havedecreased by half of their original area (fig. 8).Separating out these seral stages, we can see how

forest types have differentially changed (fig. 9).Of the slight overall decline in area of early seralforests, ponderosa pine forest has declined thegreatest, montane forest has declined somewhat,and subalpine forest has increased in area (fig.9a). Of the overall increase in area of mid seralforests, montane forest and ponderosa pine foresthave greatly increased, while subalpine forest hasremained relatively constant in area (fig. 9b). Ofthe overall sharp decline in area of late seral for-ests, ponderosa pine single-layer forest and sub-alpine forest multilayer forest have had thegreatest declines; ponderosa pine multilayer for-est and montane multilayer forest have declinedsomewhat less; and montane single-layer forestand subalpine single-layer forest have slightlyincreased (fig. 9c). We will see later that thesechanges have ramifications for associated plantand animal species.

Figure 6—Historic (early 1800s) and current area of nonforest vegetation communities of the basin assessmentarea, showing increase in areas of agricultural and exotic vegetation (source: ICBEMP, GIS 1-km2 raster data).

11

Figure 7—Percentage of changes of area covered by terrestrial vegetation communities,combined into seral stage categories, from historic (early 1800s) to current time, in thebasin assessment area: * = not historically present (source: ICBEMP, GIS 1-km2 rasterdata).

Figure 8—Historic (early 1800s) and current area of forest vegetation communi-ties combined by seral stage, in the basin assessment area (source: ICBEMP,GIS 1-km2 raster data).

12

Figure 9—Historic (early 1800s) and current area ofindividual forest vegetation communities by (A) early,(B) mid, and (C) late seral stage, in the basin assess-ment area; lowmont = lower montane, subalp = sub-alpine (source: ICBEMP, GIS 1-km2 raster data).

13

Methods

Geographic Area Estimates

Estimates of geographic areas are fromArcInfo®1 geographic information systems (GIS)1-km2 raster data, provided by the ICBEMP.

Species Data Collection

Data on species distribution and ecology weregathered from literature and panels of speciesexperts. Methods for data collection and lists ofexperts consulted were presented in Marcot andothers (in prep.).

Further Explorations of TerrestrialBiodiversity

Global patterns of species richness (number ofspecies) by taxonomic class were compared withthose of the basin assessment area to determinedifferences or similarities. Global richness valueswere tallied from the literature, which presentednumber of known species and low and high esti-mated values. Values of species richness fromthe basin assessment area were tallied as singlevaluesfor number of known species and number of esti-mated species (Marcot and others 1997). Thepoorest known taxonomic classes are those withthe lowest proportion of estimated species thathave been scientifically verified from the assess-ment area. These were compared with global pat-terns. Implications of terrestrial biodiversitypatterns for management are identified basedon similarities of these patterns.

Assessment of Macroecology ofIndividual Species and SpeciesGroups

Species database—A total of 1,347 individualterrestrial species and 93 species groups ofplants and animals were represented in a species-environment relations (SER) database (Marcot1997; Marcot and others 1997). These are not allthe individual species or species groups that occurin the basin assessment area but includes rare orpotentially rare plants, selected invertebrates, andall vertebrates.

The SER database includes descriptions of (a) keyenvironmental correlates of species, which consistof macrohabitats (vegetation cover types andstructural stages) and other microhabitat and en-vironmental factors that affect the presence, dis-tribution, and realized fitness of species; (b) KEFsof species; (c) ecological status, including degreeof endemism and geographic distribution of spe-cies; (d) legal listing status of species by variousFederal and State agencies; and other information.

Species groups included lichens, bryophytes, vas-cular plants, bacteria, protozoa, rotifers, and nem-atodes. Species of lichens and bryophytes weregrouped according to common use of substrates orsoil conditions. Species of vascular plants weregrouped based on various criteria, including gen-era difficult to identify, taxonomic uncertainty,and taxa for which knowledge and experts wereavailable. Some vascular plant groups likely areeither underrepresented or not represented insome geographic areas. Thus, although we sum-marize number of vascular plant groups, we donot include them in subsequent ecological anal-yses. Bacteria, protozoa, rotifers, and nematodeswere each grouped according to their presenceand ecological function in soils.

1 The use of trade or firm names in this publication is forreader information and does not imply endorsement by theU.S. Department of Agriculture of any product or service.

Chapter 3

14

In the SER database, information on rare or po-tentially rare plants also included life form, cate-gories of population trend (as denoted in stateNatural Heritage databases), dispersal mode, pol-linators, indicators of specific soil or environ-mental conditions, and occurrence by state, gen-eral geographic area, substrate, and local topo-graphic condition (azimuth, relief, and slope). In-formation on invertebrates also included dispersalmode. Information on vertebrates also includedpercentage of overall distributional range of eachspecies occurring within the basin assessmentarea, and the breeding, seasonal occurrence, andabundance status of birds.

The SER database was queried to produce eco-logical profiles of species by taxonomic group,habitat condition, ecological function, ecologicalstatus and geographic occurrence, legal status,and other attributes. Because few studies havebeen conducted on most of the basin species, theSER database is incomplete and consists largelyof categorical information based on the judg-ments and experiences of experts (see Marcotand others [1997] and Marcot [1997] for caveatson use and interpretation of the SER databaseinformation).

Plants and allies—Occurrence and autecologyof fungi, lichens, bryophytes, and vascular plantsof the basin assessment area were summarized inMarcot and others (1997) and will be detailed inseveral other publications. In this report, num-ber of taxa and species groups of plants and allieswere tallied by life-form category, populationtrend class, geographic occurrence, substrate,local topographic condition, dispersal mode, typeof pollinator, and type of indicator function. Life-form categories of plants followed the classifica-tion of Raunkaier (1934). Information in the SERdatabase on plant dispersal modes was entered ascomments taken from experts; the commentswere then coded into categories, as discussed in“Results.” Trends of vascular plant species, andof lichen, bryophyte, and vascular plant groupswere denoted in four categories (unknown, de-creasing, stable, and increasing) by the Plant TaskGroup of the Science Integration Team. In thisdocument, only scientific names are used forplants, as common names have not beenstandardized.

Invertebrates—Ecology of invertebrates ofthe basin assessment area was noted in Marcotand others (1997), summarized in Niwa andSandquist (in prep.), and will be detailed by func-tional group in several other publications. In thisreport, dispersal modes of example invertebratespecies of the basin assessment area were catego-rized and cross-tabulated with KEFs of species.This helped determine the role of dispersal ecol-ogy in maintaining ecological functions of theinvertebrate fauna. Examples of invertebrate spe-cies and micro-organism groups dispersing byphoresis (by using another organism as a dis-persal host) were listed, illustrating how speciesinteractions have evolved that potentially influ-ence viability and distribution of some inverte-brates and micro-organisms.

Vertebrates—Ecology of vertebrates of the basinassessment area was summarized in Marcot andothers (1997) and will be detailed by selectedspecies groups in other publications. In this report,patterns of taxonomic diversity were explored bytallying number of species, genera, and familiesthrough the basin assessment area and by vegeta-tion cover types and structural stages.

Key ecological functions of species were furtherexplored by cross-tabulating number of vertebratespecies by ecological functions and by terrestrialvegetation communities. This resulted in “species-function profiles” that were used to determinewhich vegetation communities had the greatestvariation in KEFs of species, and which ecologi-cal function categories varied the most over vege-tation communities.

Species function profiles were developed for allvertebrates combined by tallying the number ofspecies with selected specific KEFs, by terrestrialvegetation community. Greater numbers of spe-cies with the same KEF in a specific communitywere interpreted as greater redundancy of thatfunction. Similarly, Huston (1994:3) referred tospecies with the same KEFs as “functional typesof organisms,” and redundancy of species with aparticular KEF as “functionally analogous spe-cies.” Franklin and others (1989:93) noted that“forest ecosystems which have redundancy instructure and function are more likely to be able

15

to absorb stresses, including species losses, with-out catastrophic damage. . . .” Silver and others(1996:17) similarly concluded that “functionaldiversity, and not just species richness, is impor-tant in maintaining the integrity of nutrient andenergy fluxes,” and that high species richnessaffords alternative and redundant pathways for theflow of resources in an ecosystem. Our speciesfunction profiles helped (1) identify communitieswith the greatest (and the least) redundancy inecological functions of species, and (2) determinevariation in redundancy of each KEF amongcommunities.

Total functional diversity of all vertebrate specieswas calculated by multiplying the number ofKEFs of vertebrate species in each terrestrial veg-etation community by the average number of ver-tebrate species performing each function (afterHuston 1994:4ff). Species function profiles andtotal species functional diversity were not calcu-lated for rare or potentially rare plants and allies,nor for invertebrates, because (a) scientificknowledge and our databases of these taxa are in-complete, and (b) most of these species likely re-spond to microhabitats, microclimates, and sub-strates at far too fine a resolution than affordedby the available broad-scale data in this study.

Broad-scale ecological distribution of birds wasassessed by cross-tabulating number of bird spe-cies by abundance, breeding status, migrationstatus, and U.S. state, and relating to general geo-graphic location. Five abundance categories andfive breeding or migration status categories weredenoted for birds (appendix A). Bird speciesabundance, breeding status, migration status, andoccurrence by U.S. state were examined for pat-terns of local and regional endemics (see Marcotand others [1997] for definitions of endemismclasses used with plants and animals).

We explored potential effects of regional climatechange2 on vertebrates. We first listed speciesassociated with vegetation communities that mayundergo major declines in distribution and areaunder climate-change scenarios. We then talliedthe number of vegetation cover types used byeach species and rank-ordered the species in in-creasing number of cover types used. We iden-

tified species at risk as those that use <20 percent(and also those species at greater risk that use<10 percent) of the total number of vegetationcover types (N = 44). These are species closelyassociated with the primary vegetation com-munity of interest and are thus at greatest riskof decline or extirpation from climate-inducedchanges of that community.

Present and Past TerrestrialCharacterization of the BasinAssessment Area

Present character—The Science IntegrationTeam described broad-scale characteristics ofthe basin assessment area in terms of vegeta-tion cover types and structural stages, geology,weather, land form, and other attributes, by1-km2 cells, 4th-code HUCs, and 6th-code HUCs(chapters in Quigley and Arbelbide 1997).

Historic character—We used results of recon-structions of early historic (early 1800s) vegeta-tion conditions and changes to current conditions,as provided in Hann and others (1997).

Prehistoric character—In this report, we alsosummarize aspects of paleoclimates and paleo-ecology of the basin assessment area by compar-ing reports of prehistoric climates and biota withthose of the present. We compared diversity ofvertebrate Tertiary paleofaunas with that of mod-ern faunas in eastern Oregon, at order, family,genus, and species levels of taxonomy and bygeologic period, and drew conclusions on qualita-tive changes in environments, turnover of varioustaxa, and changes in overall terrestrial communi-ties. Data are lacking on prehistoric rates of spe-cies extinction and speciation in the basin assess-ment area.

Assessment of Ecological Integrityof Terrestrial Species andCommunities

Ecological integrity refers to the degree to whichnative taxa and their ecological functions persistunder human land management (for example,Majer and Beeston 1996). Ecological integritywas assumed by the Science Integration Team tobe a general purpose for managing ecosystems.Ecological integrity reflects the degree to whichall ecological components and their interactions

2 Personal communication. Sue Ferguson, Climate modeldata, Forestry Sciences Laboratory, Seattle WA 98105-6497.

16

are present and functioning. We assumed that thesix goals of managing for ecological integrity inland use planning, as discussed by Haynes andothers (1996), provide important benchmarksagainst which to measure progress. Three of thosegoals dealt explicitly with terrestrial ecologicalconditions: maintaining species viability, main-taining long-term evolutionary potential of spe-cies, and managing for multiple ecological do-mains and evolutionary timeframes. We assumedthat those three ecological goals could be inter-preted in terms of species management objectivesby using spatially referenced information on ter-restrial species ranges and habitat and environ-mental conditions. In this way, conditions in 6th-code HUCs were mapped in a GIS (ArcInfo), foreach component of terrestrial ecological integrity.Specifically, we followed these steps:

1. Reiterate the six goals for ecological integrityas identified by Haynes and others (1996) andidentify those pertinent to terrestrial ecology(goals 1-3).

2. Under each ecological integrity goal, identifycomponents pertinent to terrestrial ecology, par-ticularly those that can be mapped and quantified.Denote these components under each of the threegoals pertinent to terrestrial ecological conditions(as discussed in Marcot and others 1997).

3. For each integrity component, identify corre-sponding species by querying the SER database.Then identify the set of GIS map themes, includ-ing distributions of species, to represent thecomponent.

4. For each integrity component and its set of GISmap themes, quantify conditions for each 4th codeHUC in the basin assessment area. Use these datato build a database in which rows are individual4th code HUCs, columns are the integrity com-ponents, and cells are filled in based on counts,percentages of each HUC covered, or other mea-sures pertinent to each component.

5. Produce a map for each integrity componentshowing, by 4th code HUC, the outcome ofcombining all relevant GIS themes as defined instep 4.

6. Merge the database from step 4 with similardata on 4th code HUCs into an overall databaseexpressing combined conditions for all terrestrialecology components of ecological integrity.

Overall results were expressed in three mapsdepicting current average conditions for each ofthe three ecological integrity goals pertaining toterrestrial species and ecosystems. These mapsare intended to provide a broad-scale overview ofthe basin assessment area that may be useful inplanning policy. They are not scaled for site-levelor midscale management use.

17

Results

Further Explorations of TerrestrialBiodiversity

Marcot and others (1997) present a preliminarysummary of biodiversity patterns of speciesamong taxonomic groups within the basin assess-ment area. In this section, we expand on thosefindings.

First, how does biodiversity within the basin as-sessment area compare with global patterns? Esti-mated global biodiversity suggests that most spe-cies are insects (fig. 10, Wilson 1988). Next inestimated global abundance are algae (not talliedfor the current work), bacteria, fungi, nematodes(roundworms), spiders and mites, and other taxa.The least abundant taxa globally are vertebrates(fig. 10). Estimates of diversity in the basinassessment area suggest that most species arebacteria, followed by fungi, macroinvertebrates(insects), plants and allies (lichens and bryo-phytes), and other species. The least abundant inthe basin assessment area are soil nematodes1 andvertebrates (fig. 11).2 Overall patterns indicatethat, both globally and in the basin assessmentarea, estimated species diversity (number of spe-cies) of invertebrates and micro-organisms rankhighest and vertebrates lowest.

On comparing known with estimated speciesdiversity, it is obvious that the largest gap inscientific knowledge (less than half of estimatednumbers of species actually known) globallyoccurs with meso- and micro-organisms. Thebest known taxa are vertebrates. In the basinassessment area, the least known taxa probablyinclude most of the micro-organisms, as well asinsects, fungi, and mollusks. Only 14 percent ofthe estimated number of insect species in thebasin assessment area have actually been reportedfrom the basin assessment area. The percentagesfor fungi (33 percent) and mollusks (49 percent)3

are a little greater. (Percentage values presentedhere are approximations.) Globally, the percent-age of estimated number of species (from theupper estimate values) that actually are known are0.13 percent of bacteria, 0.95 percent of insects,1 percent of viruses, 5 percent of fungi, 8 percentof spiders and mites, and 35 percent of mollusks.Thus, on a percentage basis, these taxa are betterknown in the basin assessment area than they areglobally, although much basic work on speciesinventory and taxonomic studies remains to bedone.

The best known taxa (at least half of estimatednumber of species actually known) in the basinassessment area probably are vertebrates andlichens (all or nearly all likely occurring specieshave probably been discovered and catalogued inthe basin assessment area), followed by vascularplants (99 percent or perhaps a bit less) and bryo-phytes (94 percent). Globally, the best known taxaare vertebrates (90 percent) and vascular plants(50 percent). Most of the biodiversity patterns of

1 That nematodes rank high in global biodiversity and low inthe basin assessment area likely reflects how poorly knownthey are in the basin assessment area rather than their trueabundance. All the values of known and estimated speciesbiodiversity are subject to change as new studies areinitiated.2 This comparison between global and basin assessmentarea biodiversity is not strictly at parity, because data on alltaxonomic groups as divided for the basin assessment areaanalysis were not available at a global level.

3 No figures for these percentages are available formicro-organisms in the basin assessment area. So little workhas been done on these taxa that virtually all are unstudiedand empirically undocumented in the basin assessment area.

Chapter 4

18

Figure 10—Diversity of terrestrial species globally, showing known, low estimates, andhigh estimates of number of species by taxonomic class, sorted in decreasing order ofhigh estimates (source: Marcot and others 1997). (Note log scale.)

Figure 11—Diversity of terrestrial species and fish in the basin assessment area,showing known and estimated number of species by taxonomic class, sorted indecreasing order of estimated number. (Note log scale.)

19

species numbers by major taxonomic groups inthe basin assessment area follow global patterns.Some groups are better known in the basin as-sessment area because the global estimates areweighted heavily toward largely unstudiedtropical ecosystems.

Macroecology of IndividualSpecies and Species Groups

Plants and Allies

Number of species—In the basin assessmentarea, some 18,946 species of plants and alliesare estimated to exist (Marcot and others 1997).This figure extrapolates for species not yet dis-covered or catalogued and includes projections ofabout 9,000 species of macrofungi, 736 lichens,860 bryophytes, and 8,350 vascular plants. Atpresent, only 12,797 species of plants and allies(68 percent of all estimated species) are actuallyknown from the basin assessment area, includingabout 3,000 macrofungi (33 percent of all esti-mated macrofungi species), 736 lichens (100 per-cent of all estimated lichens), 811 bryophytes (94percent of all estimated bryophytes), and 8,250vascular plants (99 percent of all estimated vascu-lar plants). An additional large set of microfungi,including mycorrhizae, likely occur but arelargely unstudied and are not treated here.

The SER database lists 596 individual taxa(mostly species with selected subspecies or varie-ties) and 82 species groups of plants and allies.These include 394 species of macrofungi; 2 spe-cies and 39 groups of lichens; no species and 11groups of bryophytes; and 200 taxa and 32 groupsof vascular plants (one part of the SER databasealso lists 920 vascular plant taxa, but informationon environmental correlates and ecological func-tions are not available for all of these taxa.) The596 taxa and 82 groups mostly represent rare orpotentially rare but largely unstudied taxa andcommunities; the vascular plants included aremostly perennial forbs or nonwoody perennials.Thus, the analyses that follow pertain only to thisselect subset of rare or potentially rare plants ofthe basin assessment area and do not representconditions of the full flora.

Species and groups by geographic area—Presence by general geographic area within thebasin assessment area was denoted for lichengroups, vascular plant groups, and vascular plantspecies. Numbers of lichen groups and vascularplant groups were relatively evenly distributedthroughout most of the geographic areas, withfewer in Wyoming and Nevada probably becauseof the small total area these states occupy withinthe basin assessment area. Exceptions also in-clude the Columbia Plateau, which is entirelycontained within the basin assessment area butwhich had only two lichen groups represented,and the Basin and Range area (northern GreatBasin), which extends well beyond the basin as-sessment area and had only one vascular plantgroup. The few numbers of these species groupsmay be an artifact of the grouping process.

Numbers of the rare and potentially rare vascularplant species considered in this assessment wereless evenly distributed among general geographicareas (fig. 12) than were lichen groups and vas-cular plant groups. Geographic areas with morespecies (>25 species) included Basin and Range,Columbia Plateau, Blue Mountains, northernIdaho, and the southern part of the east side ofthe Cascade Range. Areas with fewest species(<10 species) included the small portion ofWyoming within the basin assessment area,southern Idaho, and Okanogan Highlands,although the last two of these areas were notreduced in number of species groups.

Life forms—Life forms of most of the rare orpotentially rare fungi were mushrooms (59 per-cent of all 394 macrofungi taxa included in theSER database) and truffles (31 percent), with therest puffballs (7 percent), polypores (2 percent),and resupinate (1 percent) (fig. 13; see defini-tions in table 4).

Life forms of most of the rare or potentially rarevascular plants were hemicryptophytes (36 per-cent of all 280 vascular plant taxa included in theSER database) and cryptophytes (30 percent),with the rest therophytes (16 percent), chamae-phytes (15 percent), and phanerophytes (4 per-cent) (see definitions in table 4). Plants of mostof these life-form categories are widely distrib-uted among vegetation cover types of the basinassessment area. The hemicryptophytes (102 spe-cies) are found in 27 vegetation cover types, most

20

Figure 12—Number of rare or potentially rare vascular plant species by geographic areawithin the basin assessment area.

Figure 13—Diversity of macrofungi in the basin assessment area by life form (see table 4for descriptions of life forms).

21

Table 4—Classes and definitions of fungi and vascular plant life formsa

Class Definition

Fungi:

Mushroom Fungi with any macrofruiting body, typically those with cap and stipemorphology

Resupinate Upside down or recurved growth form

Truffle Fungi that are hypogeous (undergound), rare, odoriferous, and edible, ofAscomycetes

Puffball Fungi of Gastromycetes of the order Lycoperdales (true puffballs and stalkedpuffballs); does not include Hymenogastrales (false puffballs)

Polypore Fungi in the family Polyporaceae

Vascular plants:

Phanerophyte Species with perennating buds or shoot apices on aerial shoots. Groupincludes most woody shrubs or trees

Chamaephyte Species that hold their perennating buds and shoot apices close to theground. Group includes semiwoody and herbaceous species that persistaboveground throughout the year (though not necessarily in aphysiologically active state). Group includes bunch grasses and cushionplants

Hemicryptotphyte Species with perennating buds at ground level, generally protected by snowor organic debris, with all plant parts aboveground dying back at the end ofthe active growing season. Stolons may or may not be present. Groupincludes rosette plants (for example,Taraxacum), partial rosette plants (forexample,Achillea millefolium), and stoloniferous species (for example,Rubus)

Cryptophyte Species bear their perennating buds below ground level or submerged inwater. Group includes rhizomatous and bulb-forming species (for example,Allium), and aquatic species such asAlismaandNuphar

Therophyte Annual species (for example,Bromus tectorum, Stephanomeria malheurensis).Embryonic buds are protected by a seed coat

a Fungi definitions from W. Owen. Vascular plant definitions from L. Croft based on Raunkier (1934); also seeDaubenmire (1974).

22

principally in big sagebrush (26 species), lowsage (14 species), mountain big sagebrush (13species), fescue-bunchgrass (12 species), andalpine tundra (11 species), with the remaining26 species distributed among 22 other vegetationcover types. The cryptophytes (84 species) arefound in 22 vegetation cover types, most prin-cipally in fescue-bunchgrass (20 species),Agropyronbunchgrass (14 species), interiorponderosa pine forest (13 species), and juniper/sagebrush (10 species), with the remaining 27species distributed among 18 vegetation covertypes. The therophytes (45 species) are found in21 vegetation cover types, most principally in bigsagebrush (10 species), juniper sagebrush (9 spe-cies), interior ponderosa pine (9 species), andinterior Douglas-fir (8 species), with the remain-ing 9 species distributed among 17 other vegeta-tion cover types. The chamaephytes (42 species)are found in 21 vegetation cover types, mostprincipally in low sage (13 species), big sage-brush (8 species), and mountain big sagebrush(5 species), with the remaining 16 species dis-tributed among 18 other vegetation cover types.Finally, the phanerophytes (10 species) are foundin seven vegetation cover types, most principallyin Agropyronbunchgrass (two species) and lowsage (two species), with the remaining sixspecies distributed among three other vegetationcover types. These are very general patterns;information on finer classes of habitat typeswould reveal more useful botanical patterns.

Number of rare or potentially rare vascular plantspecies among life-form categories do not showa clear trend at the broad scale in relation to gen-eral topographic position, soil texture, soil depth,soil moisture regime, or soil temperature catego-ries, although correlations would be expected atfiner scales of resolution. This is not surprising,given the very microsite- and substrate-specificnature of most of these rare or potentially rareplants. Based on the initial assignments of plantsto life-form categories, hemicryptophytes mightoccur more frequently in calcareous substratesthan do other life forms; they also might occurmost commonly in alkaline soil conditions. Thismay represent an ecological pattern, or merelyreflect the dominance of this life form in the rareplant flora. These patterns need verification andfurther ecological study.

In general, alkaline or calcareous soils are typi-cally rich in edaphic endemics worldwide. Mostplant species are unable to overcome the physio-logical challenges imposed by such soils, includ-ing Ca++/Mg++ imbalances and the general lowavailability of nitrogen, phosphorus, and potas-sium nutrients in high pH situations because ofspecific solubilities. Alkaline or calcareous sitestend to be azonal; that is, represented by an in-trusion into a matrix of some other parent mate-rial (for example, granite or sandstone). Thoseplants that thrive in such azonal settings with fewother species capable of surviving the local con-ditions often have an opportunity to escape com-petition. Azonal soils almost always have unique,if not rare, flora. Examples include serpentine orultramafic conditions in northwestern Californiaand southwestern Oregon, volcanic ash in easternOregon and western Idaho, and hyperacidic minetailings, hot springs, and sandy substrates else-where. Specialization of rare or endemic plants toazonal conditions can be considered an extremeform of habitat segregation.

Trends of vascular plant species—Of 299 rareor potentially rare vascular plant species, 143 spe-cies (48 percent) had unknown or unreportedtrends, 84 species (28 percent) were coded as de-creasing, and 71 species (24 percent) as stable.One species (<1 percent) might be increasing atleast in some areas. Among the decreasers werevarious species ofAllium, Aster, Astragalus,Botrychium, Calochortus, Cypripedium,Eriogonum, Haplopappus, Howellia, Lepidium,Mentzelia, Mimulus, Penstemon, Ranunculus,Silene, Thelypodium, Trifolium,and other genera.Among the stable species were various otherspecies ofAllium, Astragalus, Botrychium,Castilleja, Claytonia, Erigeron, Eriogonum,Leptodactylon, Mentzelia, Mimulus, Penstemon,Trifolium, and other genera. The sole potentialincreaser was anAstragalusspecies. Species withunknown or unreported trends spanned an evengreater number of genera.

Trends of lichen, bryophyte, and vascularplant groups—Trends were denoted for lichengroups, bryophyte groups, and vascular plantgroups (fig. 14). Trends of most of the lichengroups were decreasing (41 percent of all lichengroups) or stable 36 percent), with fewer groupshaving unknown or unreported trends (18 percent)or increasing trends (5 percent). Trends of

23

bryophyte groups were largely unknown or un-reported (83 percent of all bryophyte groups),with fewer decreasing (17 percent); none wasstable or increasing. Trends of vascular plantgroups were mostly unknown or unreported (44percent of all vascular plant groups), with fewergroups stable (33 percent) or decreasing (22 per-cent); none was increasing.

The only two lichen groups noted as having in-creasing trends were excess nitrogen-indicatorlichens and urban pollution-tolerant lichens.

The 28 stable groups included 16 lichen groupsin various aspen, calcareous, fencepost, leaf, rock,seepage, ledge, and tundra conditions and sub-strates, and 12 vascular plant groups consistingof 2 Allium groups, 5Mimulusgroups, and 5Penstemongroups.

The 28 decreaser groups included 18 lichen, 2bryophyte, and 8 vascular plant groups in variousarid, riparian, and forest conditions. The twodecreaser bryophyte groups included the decayedwood bryophyte group and the epiphytic bryo-phyte group.

The 24 lichen groups with unknown or unreportedtrends included 3 lichen groups of bog, nitrogenfixation, and pioneer soil stabilizer conditions; 9bryophyte groups of aquatic, humid duff, peat-land, rock, and soil conditions; and 12 vascularplant groups, allCarexgroups mostly of mead-ows association, with a few of peatland and for-est conditions.

Lichen, bryophyte, and vascular plant speciesgroups by substrate—Lichen, bryophyte, andplant groups included in the SER database weredenoted with orientations to specific nonvegeta-tion substrates and vegetation substrates.

Patterns of species groups among nonvegetationsubstrates were as follows: lichen groups weremore numerous on basalt rock, other rock, andwelded tuff; bryophyte groups were fairly evenlydistributed among all substrates but were absenton pumice and welded tuff (fig. 15). Among alltaxa, wet and dry soil substrates, and basalt andother rock substrates were used by the most spe-cies groups, and pumice and alkali soils wereused by the fewest species groups (fig. 15). Thecomposition of species groups, however, differedamong these sets. For example, the sole lichen

Figure 14—Number of lichen, bryophyte, and vascular plant species groups in the basinassessment area, by trend category.

24

group found associated with pumice substrateswas the calcareous indicator lichens group. Thisgroup also is found on five other nonvegetationsubstrates. The two lichen groups—the pioneersoil stabilizer lichen group and the soil lichengroup—associated with tuffaceous ash each alsoare found in three other nonvegetation substrates(dry, wet, and alkali soils).

Patterns of species groups among vegetation sub-strates were as follows: lichen groups were mostnumerous as epiphytic conditions and were ab-sent in peatland, fen, and bog substrates (althoughindividual lichen species may occur in such con-ditions); bryophyte groups were most numerousin aquatic submerged substrates but occurred inall but musicolous substrates; and vascular plantgroups were most numerous in peatland, fen, andbog substrates and also occurred in humus andduff substrates but were not identified for theother vegetation substrates (although individualspecies occur on such substrates) (fig. 16).

Patterns of plant endemism—There are manyexamples of endemic plants in the basin assess-ment area. Some examples are Idaho goldenweed

(Haplopappus aberrans) in the Boise River drain-age,Chaenactis evermanniiGreene in centralIdaho, andC. neviiGray in the John Day area.

Not all endemic plants are rare, and not all rareplants are endemics. A large proportion of therare or potentially rare plants considered in thisassessment, however, are locally endemic orregionally endemic (156 taxa or 78 percent of200 vascular plant species in the SER databasewith information on endemism status). We de-fined local endemic as a plant with populationsrestricted to a very small area such as one portionof a mountain range or one canyon. Local endem-ics also may be specialized on highly restrictedhabitats. The entire range may lie within the basinassessment area. A regional endemic is a plantwith populations that inhabit a larger geographicarea than that of local endemics such as southeastOregon or the Palouse. The range of regional en-demics may extend beyond the basin assessmentarea boundaries. Regionally endemic plants alsomay be closely associated with specific habitats.In general, about a third of all vascular plants(including common and rare species) are endemicat least regionally.

Figure 15—Number of lichen, bryophyte, and vascular plant species groups by soil androck (nonvegetation) substrate condition in the basin assessment area.

25

Gentry (1986) tallied endemic plants (mostlybased on candidate taxa for Federal listing, simi-lar to our consideration of plants of conservationconcern) by U.S. state. We compare our tallieswith his in table 5. His definition of endemictaxa included plant species with distributions of50 000 km2 or less, which corresponds to ourcategories of local and regional endemics. Histallies included entire states, whereas ours in-cluded only portions of states within the basin as-sessment area. We counted more endemic plantsin Idaho and Montana than he did, however;whereas his figures for Oregon, Washington,Nevada, and Wyoming were greater than ours.(We did not characterize plant endemism in theUtah portion of the basin assessment area.) Onlya small portion of Nevada and Wyoming are with-in the basin assessment area, so these differencesare not surprising. But we were surprised to seeour greater tallies for Idaho and Montana, con-sidering particularly that only the western portionof Montana (west of the Continental Divide) wasincluded in the basin assessment area, and only aportion of the entire flora of the basin assessmentarea is included in the SER database. Our greater

tallies of endemic plants in Idaho and Montanapossibly reflect more current data from stateNatural Heritage program databases.

Gentry (1986) suggested two kinds of endemics.The first kind ispaleoendemics, which are spe-cies that were formerly more widespread but nowhave restricted distributions because of climaticchanges and natural or human-caused reductionof favorable habitats. Some paleoendemics maybe archaic taxa with primitive traits; some maybe close to natural extinction. Pielou (1991:240)notes that the present alpine flora of the westernmountains contains a mix of paleoendemic plantspecies, some presumably derived from Pleisto-cene refugia south of the Cordilleran ice sheetand others from northern Beringian tundra. Shealso noted that the nearest unglaciated drainagebasin to the Cordilleran ice of the Pleistocene wasthat of the Columbia River, which must havebeen the refuge for many paleoendemic species.

The second kind of endemic isneoendemics,which are species that have recently evolved andhave restricted ranges because they have not yethad time to spread over a larger geographic area.Examples of neoendemics might include species

Figure 16—Number of lichen, bryophyte, and vascular plant species groups by vegetationsubstrate condition in the basin assessment area.

26

that have recently arisen on serpentine and ultra-mafic substrates. Most species ofAstragalusinthe basin assessment area likely are neoendemics.Allium aaseae(Ownbey) is a “classic”neoendemic.

Kruckeberg and Rabinowitz (1985) suggestedthat paleoendemics tend to have more thanone disjunct population, whereas endemicsconfined to a single population can be eitherpaleoendemics or neoendemics. This contentionis untested for the regional or local endemicplants of the basin assessment area. Of course,one cannot assert the converse; plants with morethan one disjunct population cannot be assumedto be a paleoendemic, any more than plants witha single population are endemics of either type.

In general, paleontology explains much aboutcurrent plant species distributions, especiallywith respect to historic climates. Some plants ofnorthern Idaho occur because of the ClearwaterMaritime refugium found there. Southern Idahoand the portion of Wyoming in the basin assess-ment area are geologically young, and the fewendemics found there are classic neoendemics

of species-rich genera (for example,Astragalus).Also, climate fluctuations over the last severalmillion years in southern Idaho have been broad,thereby making it difficult for either mesophytesor xerophytes to persist over the entire period(Tausch and others 1993).

Overall, genetic data are needed to determinewhich other vascular plant taxa within the basinassessment area are paleoendemic or neoendemic.Chemical, physical, and biological properties ofsoils often constrain endemic taxa (Kruckebergand Rabinowitz 1985). In some cases, rarity orrange-restriction of a coevolved symbiont mayaccount for endemism of plant taxa, particularlywith a rare obligate pollinator but, again, thisneeds empirical study in the basin assessmentarea.

We explored the question of environmental deter-minants of endemic plants by comparing 13 bio-physical factors between the 156 plant taxa fromthe SER database that were denoted as local orregional endemics to the remaining 44 plant taxain the database denoted as neither local nor re-gional endemics. This analysis is pertinent only

Table 5—Tallies of locally and regionally endemic plant taxa inU.S. states of the basin assessment area, compared with estimatesof endemic plant species from Gentry (1986:155)a

This studyEndemics

Local Regional as listed byState endemics endemics Total Gentry (1986)

Oregon 47 30 77 109Washington 24 13 37 49Idaho 29 22 51 37Montana 6 7 13 6Nevada 2 1 3 19Wyoming 1 3 4 90

Entire basinassessmentarea 93 63 156 —

— = no dataa We defined local endemic as plants with populations restricted to a very small areasuch as 1 portion of a mountain range or 1 canyon, and regional endemic as plantswith populations that inhabit a larger geographic area such as southeast Oregon orthe Palouse. Gentry defined endemic as plant species with distributions of 50 000 km2

sor less, which corresponds to our categories of both local and regional endemics.

27

to rare or potentially rare plant taxa and does notdescribe more common or widespread plant spe-cies. The biophysical factors included in this anal-ysis were maximum elevation, minimum eleva-tion, aspect, slope angle, slope position, geo-graphic area, geology, landform, soil pH, soil tex-ture, soil depth, soil moisture, and soil tempera-ture. These factors were coded for each species inthe SER database as categorical values. We useda log-likelihood ratio contingency test to analyzethe null hypotheses that local and regional endem-ics do not differ from nonendemics, in number ofspecies distributed among categories of each ofthese biophysical factors.

Results are presented in table 6. Only two bio-physical factors—slope angle and slope posi-tion—were significantly different (P < 0.05)between endemic and nonendemic rare plants.Six other factors were marginal in significance(0.05 < P < 0.10): maximum elevation, aspect,geology, landform, soil pH, and soil moisture.Given the broad scale of the basin assessmentarea and coarse resolution of the categorical datadenoted in the SER database, the marginally sig-nificant factors might be seen as working hypoth-eses of biophysical conditions that influence num-bers of rare endemic plants. Finer grain and morequantitative data, as well as species-specific anal-yses, are needed to further unravel the factorsinducing plant endemism in the basin assessmentarea. Also needed are further studies to explainthe exclusion of endemics from more commonhabitats (Kruckeberg and Rabinowitz 1985).

Plant dispersal modes—Dispersal (dissemina-tion) modes of all rare or potentially rare plantsof the basin assessment area are not thoroughlystudied, but those known were denoted in theSER database. Seven general dispersal modecategories were identified: gravity, wind(anemochory), water, insects, vertebrates(zoochory), growth or reproduction, and un-known. The vertebrates dispersal category wasfurther divided into dispersal by birds, smallmammals, large mammals, and not specified.The growth or reproduction dispersal categorywas divided into dispersal by dehiscence, bulbettscattering, spore dispersal, stolon, rhizome, vege-tation growth, seed dispersal, and turion. Disper-sal mode was denoted for lichen, bryophyte, andvascular plant groups, and for the rare or poten-tially rare species of fungi and vascular plants.

Any one group or species could have been codedas having one or more dispersal modes.

Gravity plays the major role in dispersal of nearlyall the lichen groups, with one other group (aquat-ic rock lichens) dispersing by water and by verte-brates (birds) (fig. 17). Dispersal modes of bryo-phyte groups were evenly split between water andwind.

About two-thirds (69 percent) of the 394 rare orpotentially rare fungi species included in theassessment disperse by wind, and the remainderdisperse by vertebrates (that is, mycophagy) (fig.18). Mycophagy, or the ingesting and distributingof belowground fungal bodies (truffles) includingspores, is an important ecological role of somelarge and small vertebrates in several forest typesthroughout the inland West. Vertebrates engagedin mycophagy and dispersal of fungi includeRocky Mountain elk (Cervus elaphus nelsonii),western red-backed vole (Clethrionomyscalifornicus), southern red-backed vole (C.gapperi), northern flying squirrel (Glaucomyssabrinus), least chipmunk (Tamias minimus),Douglas squirrel (Tamiasciurus douglasi),American pika (Ochotona princeps), andColumbian mouse (Peromyscus keenii).

Dispersal modes of vascular plant groups in-cluded all seven major dispersal categories. Mostvascular plant groups, however, disperse mainlyby gravity, wind, water, and vertebrates (fig. 17).

The rare or potentially rare vascular plant speciesdisperse (in order of decreasing number of spe-cies per dispersal category) by gravity (27 percentof species), wind (26 percent), growth or repro-duction (21 percent), vertebrates (13 percent),water (7 percent), and insects (2 percent). Howthe remainder disperse (4 percent) is unknown(fig. 18). The 55 vascular plant species that dis-perse by vertebrates more or less evenly disperseby birds and small mammals, with fewer speciesdispersing by large mammals, and a third notspecified (fig. 19a). Over half of the 89 vascularplant species that disperse by growth or repro-duction do so by seed set; the remainder disperseby many other vegetative and reproductive means(fig. 19b). It must be remembered, though, thatthese patterns pertain only to the set of rare orpotentially rare plants considered in this as-sessment, and that patterns of the full flora—

28

Table 6—Results of log-likelihood ratio contingency tests (G statistic) of bio-physical factors and endemic plantsa

Biophysical factor Value ofG df P

Maximum elevation: 6.678 3 0.05 < P < 0.10 +<1219 meters1219 to 1828 meters1829 to 2438 meters2439+ meters +

Minimum elevation: 1.443 1 0.10 < P < 0.25<1219 meters1219+ meters(Categories combined because ofsmall sample sizes)

Aspect: 7.781 4 0.05 < P < 0.10 +NorthSouthEastWestFlat

Slope angle: 7.929 3 0.025 < P < 0.05 *Flat (0 to 10 percent)Gentle (11 to 30 percent)Steep (31 to 50 percent)Very steep (>50 percent)

Slope position: 7.155 2 0.025 < P < 0.05 *Lower slopeMiddle slopeUpper slope

Geographic area: 11.743 7 0.10 < P < 0.25Basin and rangeBlue MountainsColumbia basinEast Cascade Range northEast Cascade Range southMontanaNorthern IdahoSouthern Idaho(Other states omitted because ofsmall sample sizes)

Geology: 16.178 10 0.05 < P < 0.10 +BasaltCalcareousClayGlacialGranitePlaya

29

Table 6—Results of log-likelihood ratio contingency tests (G statistic) of bio-physical factors and endemic plantsa (continued)

Biophysical factor Value ofG df P

RhyoliteSandSedimentary or metamorphicSlides and talusTuff(Serpentine was omitted becauseof small sample sizes) +

Landform: 7.111 3 0.05 < P < 0.10 +CliffRidgetopScablandsValley bottom

Soil pH: 3.075 1 0.05 < P < 0.10 +Acidic to neutral (combinedbecause of small sample sizes)

AlkalineSoil texture: 1.363 2 0.50 < P < 0.75

Coarse (sand)MediumFine

Soil depth: 3.896 3 0.25 < P < 0.50Very shallow (<25 centimeters)Shallow (25 to 50 centimeters)Moderately deep (51 to 102centimeters)

Deep to very deep (>102centimeters)

Soil moisture:b 6.434 3 0.05 < P < 0.10 +AndicUdicUsticXeric

Soil temperature:b 0.144 2 0.90 < P < 0.95Cryic (0 to 8 ºC)Frigid (diff. mean summer/winter ≥ 5 °C)

Mesic (8 to 15 ºC)

a The null hypothesis for each test was that biophysical characteristics of locally and regionallyendemic rare or potentially rare vascular plant species do not differ from those of nonendemic rareor potentially rare species.b See Harvey and others (1994:3) for definitions of terms.

30

Figure 17—Number of lichen, bryophyte, and vascular plant species groups by dispersalmode.

Figure 18—Number of rare or potentially rare fungiand vascular plant species by dispersal mode.

31

Figure 19—Number of rare or potentially rare vascular plant species by dispersal mode, showing (A)dispersal by means of vertebrates and (B) dispersal by means of growth or reproduction.

32

including many more trees, shrubs, and otherforms—are likely to be very different.

Plant pollinators—Along with assuring suitablehabitat conditions and protection of specific loca-tions, successful pollination of the rare or poten-tially rare vascular plants of the basin assessmentarea may be one of the most important ecologicalfactors that can help ensure their long-term via-bility. Pollination ecology of these species, how-ever, is essentially unstudied and unknown. Whatfollows are general patterns from initial presump-tions as to pollination ecology. These patternsneed to be tested with field studies.

As Buchmann and Nabhan (1996) note, plant-pollinator relations exemplify connections be-tween endangered species and threatened habi-tats. Human-induced changes in pollinator popu-lations—caused by overuse of chemical pesti-cides, unbridled development, and conversion ofnatural areas into monocultural cropland—canhave a ripple effect on disparate species, ulti-mately leading to a “cascade of linked extinc-tions” (Buchmann and Nabhan 1996).

Pollinators also can be influenced by habitat frag-mentation, sometimes differentially benefittingexotic insect pollination vectors to the detrimentof native vectors. Aizen and Feinsinger (1994)studied native and honey bee (Apis mellifera; anexotic) pollinators of native flowering plants insubtropical dry forest of northwestern Argentina.They found that habitat fragmentation adverselyaffected native flower visitors but facilitated theuse of floral resources by honey bees.

Loss of pollinators also can result in different se-lection pressures for reproductive strategies ofsurviving plants. In a modeling study of an iso-lated population ofPrimula sieboldii, Washitani(1996) reports that loss of pollinators may resultin strong fertility selection for a self-fertile,homostyle morph. He concluded that active man-agement of plants with absent or moribund pol-linators could include hand pollinations and re-introduction and reestablishment of suitable pol-linator populations. This would help maintain amore natural gene pool in future plantgenerations.

In the basin assessment area, by far most of therare or potentially rare vascular plants are pol-linated by invertebrates (66 percent of plant spe-

cies). Additional modes of pollination includeautogamous pollination (10 percent), wind (5 per-cent), hummingbirds (2 percent), and generalist(1 percent), and others are unknown (16 percent)(fig. 20). Most plants are serviced by multiplepollinator species. For example,Eriogonumprociduum(prostrate buckwheat) likely is pol-linated by beetles (Cleridae: Coleoptera) butalso can be pollinated by flies (Tachinidae,Sarcophagidae, Bombyliidae: Diptera), wasps (sixfamilies: Hymenoptera), and moths (Sesiidae:Lepidoptera).4

Of the plants pollinated by invertebrates, abouthalf are pollinated by bees, wasps, and allies(Hymenoptera), with others pollinated by flies(Diptera), butterflies and moths (Lepidoptera),beetles (Coleoptera), spiders (Araneida), andother insects (not specified) (fig. 21). Insects as awhole play a vital function in reproduction andviability of populations of angiosperms.

Long-term viability of many plant populationslikely is affected by viability of plant pollinatorand plant disperser mutualists. Many of theserelations are typically highly specific (Menges1991). Terbourgh (1986) discussed keystonemutualists, which are species that, if extinct, likelywould result in secondary extinctions of manyassociated obligate species. Keystone mutualistsmight exist among insect-plant pollinator rela-tions in the basin assessment area (and possiblyalso among plant-disperser relations as well).Thus, understanding basic pollination and dis-persal ecology of plants is key to projecting andplanning for long-term population viability.

Indicators—Plants and allies can serve asbioindicators of environmental conditions andecosystem health. The concept is far from new,beginning perhaps with Clemment’s (1920) usageof plant indicators to denote vegetation communi-ties.5 Others have extended the concept in variousdirections. For example, Nygaard (1949) indexedthe trophic health and degree of eutrophication ofwater bodies by calculating the ratio in number ofalgae species of different taxonomic groups.

4 Source: OSDA-BLM Challenge Cost Share Project 89-6.5 Actually, well into prehistory, many indigenous peopleslikely used plants to locate underground water, gameanimals, and good places to build camps. Modern sciencedoes not have the corner on observing nature.

33

Figure 20—Number of rare or potentially rare vascular plant species by type of pollinator.

Figure 21—Number (in parentheses) of rare or potentially rare vascular plant species by type ofinvertebrate pollinator.

34

Westveld (1954) used plants to index site quality.And, Anderson (1986) (and many others) iden-tified plant indicators of range condition.

The Plant Task Group of the Science IntegrationTeam identified a number of lichen, bryophyte,and vascular plant groups and vascular plantspecies that can serve to index specific environ-mental conditions of the basin assessment area.Such environmental conditions include air qual-ity, carbonate levels, water flow levels (low,high), high nitrogen conditions, metal-rich rockconditions, old-growth forest conditions, soiltemperature (low, high), and soil texture (coarse,fine). Specific indicators, and the environmentalcondition for which they may be useful as anindicator, are listed in appendix B.

Overall, many lichens can be used to index airquality (fig. 22; see appendix B), a functionalready finding utility in management of manyforests and other geographic areas (for example,Geiser and others 1994, Stolte and others 1993).Lichen groups along with bryophyte groups alsocan be used to index most of the other categoriesof environmental conditions listed above (fig. 22;see appendix B). In addition, vascular plantgroups and rare or potentially rare vascular plantspecies can be used as bioindicators of alkaline(carbonate) soil conditions, overgrazing, metal-rich rock conditions, old growth, low soil tem-perature, coarse soil textures, and other condi-tions (fig. 22; appendices B and C).

Nonvascular plant indicators of old-growth for-est conditions may be of particular interest tomanagers. Tibell (1992) identifies crustoselichens as indicators of temporal continuity ofold boreal conifer forests. The term temporalcontinuity refers to the degree to which specifickinds of habitats—in this case, old boreal coniferforests—continuously occupy a local geographicarea over time. Some studies outside the basinassessment area have indicated that interruptingthe temporal continuity of old forests has led toloss of sensitive species closely associated withsuch habitats because there was insufficient timeor lack of sources for recolonization.

Lichens also are identified that indicate foreststand temporal continuity in northern boreal for-ests (Esseen and others 1996, Nilsson and others1995, Selva 1994). Pike and others (1975) identifyepiphytic lichens and bryophytes associated withold-growth forests of western Oregon. Söderströmand Jonsson (1992) identify bryophytes that in-dicate fragmentation of old forests; that is, loss ofold-growth bryophyte species in the more isolatedforest fragments. In the current study, we identi-fied 15 lichen groups and 9 bryophyte groups thatmay indicate old-growth forest conditions in thebasin assessment area (see appendices B and C).The rotten log and tree base lichen group inparticular indicates ecological continuity of oldgrowth.

Many of the bioindicator species and groups iden-tified here, including those of old-growth forests,can be useful to managers interested in quicklysurveying potential environmental conditionswithout investing in long-term ecological researchstudies. Such indicators can serve to track effectsof alternative management activities, such as var-iations in grazing, timber harvesting, or restora-tion regimes.

Invertebrates

Number of species—In the basin assessmentarea, some 24,270 species of macroinvertebratesare estimated to exist (Marcot and others 1997).This figure extrapolates for species not yet dis-covered in the basin assessment area and somenot yet described, and comprises about 770mollusks (425 freshwater gastropods, 35 fresh-water bivalves, 30 slugs, and 280 land snails;T. Frest in Marcot and others 1997) and 23,500arthropods, including insects. At present, only3,780 species of macroinvertebrates (16 percentof all estimated species) have actually been re-ported from the basin assessment area, including380 mollusks (200 freshwater gastropods, 30freshwater bivalves, 25 slugs, and 125 land snails;T. Frest in Marcot and others 1997) (49 percentof all estimated mollusks) and 3,400 arthropods(14 percent of all estimated arthropods). An ad-ditional, large set of micro-organisms, includingbacteria, protozoa, rotifers, viruses, and othertaxa, occur. They may number in the hundredsof thousands of species but are mostly unknownand are treated here only by taxonomic groups.

35

The SER database lists only a small example setof 346 invertebrate species. Of these, environ-mental correlates and ecological functions aredenoted for 195 individual species of arthropodsand 11 species groups of micro-organisms (6 spe-cies groups of soil bacteria, 1 group of soil proto-zoa, 1 group of soil rotifers, and 3 groups of soilnematodes). Thus, the analyses that follow pro-vide only examples of invertebrate species andmicro-organism groups.

Dispersal modes and KEFs—Even for the rela-tively tiny set of example species explicitlyincluded in this assessment, various dispersalmodes and related KEFs were identified. Disper-sal modes included various forms of independentlocomotion (crawling and other flightless terres-trial movement), sedentary habits, soil digging,

aerial movement (flying, ballooning, and drift-ing), budding, and phoresis (dispersal using otherhost species).

Phoretic dispersal exemplifies some interestingrelations that have evolved among species. Pho-retic dispersal has appeared in many micro-organisms and invertebrates, including bacteria,rotifers, protozoa, nematodes, and insects (table7). Host species can include seeds, plants, organicdebris, other insects, birds, mammals, and evenhumans. Soil bacteria are ubiquitous and disperseby means of most of these hosts. Rotifers and soilprotozoa can hitch rides on arthropods, smallmammals, birds, and even soil nematodes.Bacterial-feeding soil nematodes in turn cantravel by arthropods, birds, small mammals, andeven by humans via tools that redistribute soil,such as shovels.

Figure 22—Number of lichen, bryophyte, and vascular plant species groups by bioindicator cate-gory. See appendices B and C for lists of groups, and text for explanation of categories.

36

Tab

le7—

Exa

mpl

es o

f mic

ro-o

rgan

ism

s an

d in

vert

ebra

tes

that

disp

erse

by

phor

esis

(fac

ulta

tive

asso

ciat

ions

am

ong

spec

ies,

such

as

bein

gca

rrie

d an

ddi

sper

sed

by o

ther

org

anis

ms)

Sci

entif

icna

me

orC

omm

on n

ame

orge

nera

lspe

cies

grou

psp

ecifi

c sp

ecie

sgr

oup

Dis

pers

alm

ode

Dis

pers

alre

quire

men

ts

Bac

teria

:

Erw

inia

cart

ovo

raB

acte

rialp

atho

gens

Car

ried

byse

eds,

plan

tdeb

ris,o

ral

tern

ativ

eho

sts,

Pre

senc

e of

bact

eriu

min

dise

ased

plan

tor

anim

albi

rds,

orhu

man

sm

ater

ial;

disp

ersa

lage

ntof

ten

requ

ired

tom

ove

the

path

ogen

from

dis

ease

dho

stto

susc

eptib

leho

st

Mic

roco

leu

sB

enef

icia

lbac

teria

Pho

retic

onar

thro

pods

orsm

allm

amm

als

and

bird

sP

rese

nce

ofth

e ba

cter

ial/l

iche

nsin

undi

stur

bed

soil;

feed

ing

by d

ispe

rsal

agen

t

Ba

cillu

sth

erm

op

hilu

sC

ompe

titiv

eba

cter

iaP

hore

ticon

arth

ropo

dsor

smal

lmam

mal

sor

bird

s,or

Pre

senc

e of

bact

eriu

min

undi

stur

bed

soil;

feed

ing

byby

inoc

ulat

ion

disp

ersa

lage

nt

Pse

ud

omo

na

sflu

ore

sce

nsCom

petit

ive

bact

eria

inP

hore

ticon

arth

ropo

dsor

smal

lmam

mal

sor

bird

s,or

Pre

senc

e of

bact

eriu

min

undi

stur

bed

soil;

feed

ing

byth

erh

izos

pher

eby

inoc

ulat

ion

disp

ersa

lage

nt

Kle

bsi

ella

pla

ntic

ola

Bac

teria

:N-im

mob

ilize

rsP

hore

ticon

arth

ropo

dsor

smal

lmam

mal

sor

bird

s,or

Pre

senc

e of

bact

eriu

min

undi

stur

bed

soil;

feed

ing

byby

inoc

ulat

ion

disp

ersa

lage

nt

Rh

izo

biu

m m

elli

otii

Bac

teria

Pho

retic

onar

thro

pods

orsm

allm

amm

als

orbi

rds,

orP

rese

nce

ofba

cter

ium

inun

dist

urbe

dso

il;fe

edin

gby

by in

ocul

atio

ndi

sper

sala

gent

Rot

ifers

:

Rot

ifers

Rot

ifers

:no

info

abo

utP

hore

ticon

arth

ropo

ds,s

mal

lmam

mal

s,bi

rds,

orP

rese

nce

inun

dist

urbe

dso

il;pr

esen

ceof

disp

ersa

lro

tifer

spp.

inba

sin

nem

atod

esag

ent

Pro

tozo

a:

Am

oe

ba

am

oe

baP

roto

zoa:

bact

eria

l-pre

dato

rsP

hore

ticon

arth

ropo

ds,s

mal

lmam

mal

s,bi

rds,

orP

rese

nce

inun

dist

urbe

dso

il;pr

esen

ceof

disp

ersa

lne

mat

odes

agen

t

Nem

atod

es:

Acr

ob

elo

ide

sB

acte

rial-f

eedi

ngne

mat

odes

Pho

retic

onar

thro

pods

,sm

allm

amm

als,

orbi

rds

Pre

senc

e in

undi

stur

bed

soil;

pres

ence

ofdi

sper

sal

agen

t

Ap

he

len

chu

sa

ven

ae

Fun

gal-f

eedi

ngne

mat

odes

Pho

retic

onar

thro

pods

,sm

allm

amm

als,

orbi

rds

Pre

senc

e in

undi

stur

bed

soil;

pres

ence

ofdi

sper

sal

agen

t

Pra

tyle

nch

us

pe

ne

tra

ns

Pla

nt-f

eedi

ngne

mat

odes

Pho

retic

onar

thro

pods

,sm

allm

amm

als,

bird

s,or

Pre

senc

e of

undi

stur

bed

soil;

pres

ence

ofd

ispe

rsal

hum

anto

ols

(sho

vel)

agen

t

37

Tab

le7—

Exa

mpl

es o

f mic

ro-o

rgan

ism

s an

d in

vert

ebra

tes

that

disp

erse

by

phor

esis

(fac

ulta

tive

asso

ciat

ions

am

ong

spec

ies,

such

as

bein

gca

rrie

d an

ddi

sper

sed

by o

ther

org

anis

ms)

(co

ntin

ued)

Sci

entif

icna

me

orC

omm

on n

ame

orge

nera

lspe

cies

grou

psp

ecifi

c sp

ecie

sgr

oup

Dis

pers

alm

ode

Dis

pers

alre

quire

men

ts

Fun

gi-r

elat

ed:

Hys

tera

ng

ium

Ect

omyc

orrh

izal

mat

-C

arrie

dby

nem

atod

es, a

rthr

opod

s,or

smal

lmam

mal

sD

ispe

rsal

agen

tsfo

rmin

gfu

ngi

Rh

izo

cton

iaso

lan

iF

unga

lpat

hoge

nsC

arrie

dby

seed

s,pl

antd

ebris

,or

alte

rnat

ive

host

s,P

rese

nce

ofba

cter

ium

inun

dist

urbe

dso

il;fe

edin

gby

bird

s,or

hum

ans

disp

ersa

lage

nt

Glo

mu

sm

oss

ae

Ves

icul

arfu

ngi

Car

ried

byse

eds,

plan

tdeb

ris,a

rthr

opod

s,an

ts, o

rD

ispe

rsal

agen

tsw

ind

Pe

nic

illiu

mci

trin

um

Sap

roph

ytic

fung

iP

hore

ticon

arth

ropo

ds,s

mal

lmam

mal

s,bi

rds,

orby

Pre

senc

e of

bact

eriu

min

undi

stur

bed

soil;

feed

ing

byin

ocul

atio

ndi

sper

sala

gent

Inse

cta:

Cim

ex

latip

en

nis

Bed

bug

(Cim

icid

ae:

Flig

htle

ss—

disp

erse

onl

yon

host

anim

alP

rese

nce

ofba

ts w

hen

prop

erlif

e st

ages

ofbu

gH

emip

tera

)av

aila

ble

38

Phoretic invertebrates play various roles in con-tributing to trophic structures and ecologicalprocesses of ecosystems. Many phoretic inverte-brates consume soil micro-organisms and areprey for other consumers; they aid soil nitrogencycling through nitrogen immobilization and byacting as a source for nitrogen mineralization,which can be vital functions in forest nitrogenconservation (Davidson and others 1992, Millerand others 1989). A few phoretic invertebratesare parasites or pathogens, or cause disease. Pho-retic invertebrates also include soil-oriented spe-cies whose functions aid soil structure, enhancesoil stabilization, detoxify xenobiotics, sequester(accumulate) soil metals, and influence vegeta-tion succession and presence and diversity ofplant species.

The SER database afforded the opportunity to ex-plore some examples of how dispersal modes ofinvertebrates relate to their KEFs. Invertebrateswith independent locomotion provide many eco-logical services in their ecosystems. Most of theseexample species included in the SER databasewere insectivorous and provide food for otherinsectivores, with a few being carrion feeders,detrivorous, or herbivorous on trees or shrubs, orfeeding in water on algae or plankton. The foursedentary species were folivores.

Species dispersing by digging enhance the phys-ical structure of soil and turnover of soil nutrientsand layers. Earthworms are a prime example (seediscussion in Marcot and others 1997). In onestudy of eight species of earthworms in a Kansastallgrass prairie, James (1991) reported that totalannual soil consumption by all earthworms was 4to 10 percent of organic matter in the top 15 cm;100 to 300 percent of plant annual belowgroundproduction passed through the earthworms eachyear; and mineral nitrogen processed was 10 to12 percent of annual plant nitrogen uptake, com-parable to half of the input from precipitation,whereas the phosphorus processed was equiva-lent to 50 percent of annual uptake. Native andintroduced earthworm species, however, do notnecessarily perform the same functions. Jamesfound that exotic, introduced earthworm specieshave a negative effect on soil turnover and nutri-ent mineralization because of the lower soilthroughout and their relative intolerance ofsummer soil temperatures, as compared withthe native earthworms.

Invertebrate species that disperse by flying, bal-looning, or drifting cover an even wider array oftrophic functions than those with independentlocomotion, and include folivores, spermivores,sap feeders, root feeders, insectivores, soil micro-organism feeders, carrion feeders, aquatic herbi-vores, detritovores, and moss feeders. Many ofthese aerial-dispersing species (many Hymenop-tera) serve as important pollination vectors criticalfor reproduction of flowering plants. They alsoinclude forest beetles, some of which can changethe structure of terrestrial and aquatic habitats bycausing tree mortality, breaking down standingdead trees and large down logs, creating canopygaps in forest stands, and causing increasedstreamflow. Thus, providing suitable conditionsand means of dispersal for invertebrates also pro-vides for many ecological processes and speciesfunctions in the ecosystem.

Disease and parasites—The ecology of diseaseand parasites is not well studied in the basin as-sessment area and is poorly represented in theSER database. Disease can strongly mediate bio-diversity and viability of many kinds of wildplants and animals and needs to be considered inwildlife conservation biology programs (Dobsonand Mat 1986, May 1988, Scott 1988). Aguirreand Starkey (1994) conclude that proximity ofdomestic ungulates to wild ungulates can transmitdisease and parasites and complicate wild un-gulate management; they cite several examples,including bovine brucellosis (Brucella abortus) inbison (Bison bison) and Rocky Mountain elk inYellowstone National Park, and lungworm(Pasteurellaspp.)-pneumonia complex in bighornsheep (Ovis canadensis) in several Western na-tional parks (also see Peterson 1991). Simonetti(1995) cites other similar examples, andCunningham (1996) warns of secondary transmis-sion of parasite pathogens to wildlife during trans-location projects. Disease transmission can ex-acerbate recovery of endangered species (Thorneand Williams 1988) and is a controversial facet ofwild game ranching. Certainly, too, human healthconcerns warrant attention to wild pathogens,such as transmission of lyme disease by ticks(Ixodes scapularis) throughout the basin assess-ment area (Ostfeld and others 1995).

Parasites, however, also play ecological roles inecosystems and serve as an ecological “glue” tobind many species’ relations. Windsor (1995,

39

1996) provides many fascinating examples ofthis, including the following. Hybrid cottonwoods(Populus fremontiix P. angustifolia) are moresusceptible to aphid parasites than are their parentspecies. Thus, aphids concentrate in the hybridzone between the parents. This aphid sink dis-courages aphids from adapting to the more nu-merous parents, and it keeps the parents as sepa-rate species. Peterson (1991) calls for controlledexperiments, or at least an adaptive managementapproach (learning by trial), to provide a sounderscientific basis for management of parasite-hostinteractions.

Many invertebrate parasite-host relations awaitdiscovery. For example, Boonstra and others(1980) found that botfly parasitism can affectvole populations. Dobson (1988) suggests thatparasites can be used as an effective conservationbiology weapon to control introduced mammalsthreatening fragile isolated or island biota.

Vertebrates

Number of species—In the basin assessmentarea, 468 species of vertebrates (excluding fish)are both estimated and known to exist (Marcotand others 1997). These include 26 species ofamphibians, 27 reptiles, 283 birds, and 132 mam-mals. An additional 79 bird species are accidentalor casual species in the basin assessment area.

The SER database lists 476 species of vertebrates,consisting of all the known species and including291 species of birds (excluding the accidentals orcasuals, but including 8 bird species that likelyoccur only sporadically in the basin assessmentarea). Thus, the analyses that follow provide a fullaccounting of all regularly occurring vertebratespecies.

Endemic species—Findings on vertebrate speciesendemism within the basin assessment area arepresented in Marcot and others (1997). To recap,two amphibian species are local endemics: Idahogiant salamander (Dicamptodon aterrimus) andLarch Mountain salamander (Plethodon larselli);and seven other amphibian species are regionalendemics: northwestern salamander (Ambystomagracile), Cope’s giant salamander (Dicamptodoncopei), Dunn’s salamander (Plethodon dunni),Coeur d’Alene salamander (P. idahoensis), Cas-cade torrent salamander (Rhyacotriton cascadae),

red-legged frog (Rana a. aurora), and Cascadesfrog (R. cascadae). The locally endemic amphib-ian species occur in 9 vegetation cover types,and the regionally endemic species occur in 20.Locally or regionally endemic amphibians areparticularly rich (more than six species) in severalmontane forest vegetation cover types, includingPacific silver fir/mountain hemlock forest, moun-tain hemlock forest, interior Douglas-fir forest,and western redcedar/western hemlock forest.Other vegetation types, however, also may benecessary to ensure the long-term persistenceand distribution of some of these species.

No reptile was determined to be locally endemicor regionally endemic. Many reptile species occuronly along the margins of the basin assessmentarea and beyond, such as in the Great Basin andKlamath Basin to the south.

There are no locally endemic birds, and four areregionally endemic: sage grouse (Centrocercusurophasianus), rufous hummingbird (Selasphorusrufus), northern spotted owl (Strix occidentaliscaurina), and Columbian sharp-tailed grouse(Tympanuchus phasianellus columbianus).Collectively, these taxa occur widely through-out most vegetation cover types, although spe-cific ranges of some of these taxa (especiallyColumbian sharp-tailed grouse) are quite limitedwithin the basin assessment area.

Among mammals, local endemics include twospecies (Idaho ground squirrel (Spermophilusbrunneus) and Washington ground squirrel(S. washingtoni)) and two subspecies (potholesmeadow vole (Microtus pennsylvanicus kincaidi)and White Salmon pocket gopher (Thomomystalpoides limosus)). There may be other locallyendemic subspecies of mammals not recognizedin this study. Some 22 other mammal species areregional endemics (see Marcot and others 1997for list). Collectively, local endemics and regionalendemics occur widely throughout most vegeta-tion cover types of the basin assessment area butare concentrated somewhat more in drier forest,shrub, and herb communities. Vegetation covertypes that are particularly rich in locally orregionally endemic mammals (>12 species) in-clude mixed-conifer woodland, big sagebrush(Artemisia vaseyana), native forb, mountain bigsagebrush, and low sage. High numbers ofendemics occur in mixed-conifer forests probably

40

because such forests occupy a wide range ofmoisture and temperature conditions and theiroverall species diversity is typically high (Harveyand others 1989, Sayler and Martin 1996).

The number of species that are local and regionalendemics among amphibians, reptiles, birds, andmammals was not significantly different than thenumber of all nonendemic vertebrates by taxo-nomic class. That is, local and regional endemicswere not more apt to be of any particular verte-brate class (Marcot and others 1997).

Species by percentage of range within thebasin assessment area—A complementary wayof considering patterns of endemism is the per-centage of a species’ range that lies beyond thebasin assessment area. We identified five cate-gories for percentage of range: (A) nearly all

(>79 percent) of the species’ range lies beyond thebasin assessment area; (B) more than half butnot nearly all (70 to 79 percent) lies beyond; (C)about half (69 to 40 percent) lies beyond; (D) lessthan half (20 to 39 percent) lies beyond; and (E)only a very small portion (<20 percent) liesbeyond.

By taxonomic class, about two-thirds of the am-phibian species have more than half their rangesbeyond the basin assessment area (range cate-gories A and B); only two amphibian species (thetwo local endemics) are locally confined (rangecategory E) (fig. 23). Reptiles and birds are evenmore broadly distributed. Among mammals,about 95 percent of the species have half or moreof their range beyond the basin assessment area(range categories A, B, and C); and only six taxa

Figure 23—Number of vertebrate species by taxonomic class, by proportion of species’ range lyingoutside the basin assessment area. Species range categories: A = nearly all (80 to 99 percent) ofthe overall species range occurs outside the assessment area; B = more than half, but not nearlyall (60 to 79 percent) occurs outside; C = about half (40 to 59 percent) lies outside; D = less thanhalf, but not only a very small portion (20 to 39 percent) lies outside; and E = only a very smallportion (0 to 19 percent) lies outside.

41

have their range mostly or entirely within thebasin assessment area (range category E). Thesesix mammal taxa include all four local endemicsplus two other sciurids: Uinta ground squirrel(Spermophilus armatus) and red-tailed chipmunk(Tamias ruficaudus).

Species that migrate and spend part of theiryear elsewhere but that restrict their distribu-tion during one part of the year to a specific re-gion (such as the basin assessment area), arecalledsemiendemic speciesby Garza (1996).Semiendemics can include Neotropical migratorybirds and other species that occur only seasonallyin a given region. Garza argues that species thatare confined to a specific geographic region, evenif only for a portion of the year, still should beconsidered as restricted to that area for purposesof habitat conservation. Because our range mapsshow neither season of use nor migration rangebeyond the basin assessment area, our designa-tion of migratory species into endemism and per-centage of range classes approximates the conceptof semiendemics. Thus, some migratory speciesmight occur as semiendemics in the basin assess-ment area but rank only as range categories A, B,or C.

Another way of looking at abundance patternsamong range classes is shown in figure 24. Mostspecies that extend beyond the basin assessmentarea (range categories A, B, and C) are birds andmammals, with a smaller proportion of amphib-ians and reptiles. Those somewhat more confinedto the basin assessment area (range category D)are all mammals, and those most confined to thebasin assessment area (range category E) are am-phibians and small mammals.

We were surprised to find so few vertebrates, es-pecially species with small body size, to be localor regional endemics, and so few vertebrates withranges mostly confined to the basin assessmentarea. It is evident that the administrative and hy-drologic boundaries of the basin assessment areado not particularly circumscribe unique vertebratecommunities and species ranges. This is becausethe basin assessment area is “land-locked,” andmuch of its ecological character along the marginspills into adjacent areas, including the westernslope of the Cascade Range in western Washing-ton and Oregon, the Klamath Basin and GreatBasin in northern California and Nevada, the

Greater Yellowstone Ecosystem in Wyoming, thenorthern glaciated mountains in the CanadianRocky Mountains, and the Okanogan Highlandsin the Okanagan6 Desert of British Columbia,Canada. Marcot and others (1997) list verte-brate species closely associated with many ERUs,and many of these assemblages consist of speciesthat range well beyond the basin assessment areain many of these directions.

Abundance, breeding status, and migrationstatus of birds—Abundance, breeding status,and migration status of birds in particular werecategorized by state and for the entire basin as-sessment area. The total number of birds (in-cluding accidentals and casuals) occurring bystate was remarkably constant, ranging betweenonly 360 and 362 species. The total number ofbreeding birds (migrants plus residents) by stateaveraged 221 species and ranged from 192 spe-cies in Wyoming to 242 species in Oregon. (Itshould be remembered that these species talliesrefer only to those portions of these states that liewithin the basin assessment area.) This breeding-bird richness is quite close to that suggested byHuston’s (1994:25) plot of latitudinal gradientin species richness of breeding birds in Northand Central America. The basin assessment arealies between latitude 42º and 49º N., for whichHuston’s bird species richness curve predictsabout 200 species (but the data scatter on thecurve suggests upward of 300 speciesempirically).

For the basin assessment area as a whole, mostbird species were ranked as common, with fewerspecies uncommon or rare, fewer still as abun-dant, and the fewest as irregular (fig. 25).7 Thisgeneral bird abundance pattern also held for mostof the five western states included in this part ofthe assessment (Washington, Oregon, Idaho,

6 A word on spelling: the United States has the OkanoganHighlands and Canada has the Okanagan Desert, but they doconstitute a single shared ecosystem.7 If the abundance categories can be taken as approximateoctaves or geometric series, then the bird speciesabundance histograms match well those reported in othernorth temperate ecosystems, such as nesting bird species inQuaker Run Valley, New York (Krebs 1978:452). Suchspecies abundance patterns are somewhat stable attributesof continental communities.

42

Figure 24—Percentage of all vertebrate species by taxonomic class, showing proportionof species’ range lying outside the basin assessment area (see fig. 23 for description ofspecies range categories).

Figure 25—Number of bird species by abundance category (see text for definitions ofcategories).

43

Montana, and Wyoming; data for Nevada andUtah were not available for those small portionspresent in the basin assessment area), with a fewdeviations (fig. 26). Most of the irregular speciesoccur in western Montana, which is locatedtoward midcontinent and which picks up manyspecies found more commonly in other, adjacentcontinental ecosystems. Examples of irregularspecies in western Montana are McCown’s long-spur (Calcarius mccownii), which is more of aprairie species farther east, black-throated bluewarbler (Dendroica caerulescens), which occursmore commonly in Eastern U.S. states, and hoaryredpoll (Carduelis hornemanni), which is an ir-regularly irruptive species from northern latitudeboreal and subboreal forests. On the other hand,likely for the same reasons of being more centralin the continent, western Montana has the greatestnumber of abundant species, and northwesternWyoming has the greatest number of commonspecies in the basin assessment area (fig. 26).

As for breeding and migration status of birdsthroughout the basin assessment area, most spe-cies are resident (36 percent) or summer breeders(26 percent), with fewer species occurring as mi-grants (14 percent), winter only (7 percent), ornonbreeding summer residents (2 percent) (the re-maining 15 percent were not specified) (fig. 27).Proportions of bird species in these breeding andmigration classes generally were consistent acrossthe states but with a decreasing proportion ofresidents and increasing proportion of migrantstoward midcontinent (fig. 28).

Tallying number of bird species by abundancestatus and by breeding and migration statusrevealed interesting geographic trends by state(fig. 29). Throughout the five states included inthis part of the analysis, most of the abundant,common, and uncommon bird species were resi-dents and summer breeders. Far fewer abundantbird species occurred in Idaho and Wyoming thanin the other three states. Rare bird species in-cluded various mixes of breeding status categoriesby state with no clear geographic pattern,

Text continues on page 48

Figure 26—Number of bird species by abundance category and U.S. state, within thebasin assessment area (see text for definitions of categories).

44

Figure 27—Number of bird species by breeding status category, within the basinassessment area (see text for definitions of categories).

Figure 28—Percentage of bird species listed by breeding status category and U.S. state,within the basin assessment area (see text for definitions of categories).

45

Figure 29—Number of bird species listed by abundance category, breedingstatus category, and U.S. state, within the basin assessment area (see textfor definitions of categories).

Figure 29—continued

46

Figure 29—continued

Figure 29—continued

47

although Wyoming and Washington had morerare migrants than did other states. Among allfive states, irregular species consistently com-prised mostly migrants. Because of their geo-graphic mobility, migrants, especially long-distance migrants, are more apt to appear ir-regularly in individual states than are birds ofother breeding status categories.

Brood parasitism—Among vertebrate parasites,brown-headed cowbirds (Molothrus ater) are nestparasites of Neotropical migratory songbirds andare a factor in their decline (Brittingham andTempe 1983, Rothstein and others 1980; seeMarcot and others 1997 for a list of host spe-cies of the basin assessment area). Not all hostsmay respond the same, however. Uyehera andNarins (1995) experimentally demonstrate thatwillow flycatchers (Empidonax traillii) can dif-ferentially recognize female and male cowbirdsnear their nests, and that the behavior of the fly-catchers acts as a counteradaptation to thwartbrood parasitism.

Species function profiles—Numbers of specieswith selected KEFs were tallied by terrestrialvegetation community. We term these depictions“species function profiles” (fig. 30), which depictvariations in functional redundancy among ter-

restrial vegetation communities (table 2). For thisanalysis, we selected a range of KEFs that seemedparticularly pertinent to vertebrates and for whichvariations in numbers of vertebrate species amongcommunities might prove instructive (table 8).

We explored patterns of two trophic relationsfunctions. The first was presence of heterotrophs(table 8), which can include primary consumersor herbivores, secondary consumers or primarypredators, tertiary consumers or secondary preda-tors, omnivores, carrion feeders, cannibals, andcoprovores (feed on fecal material). As expected,the number of vertebrate heterotroph speciesdid not differ significantly among terrestrial veg-etation communities (fig. 31a) because hetero-trophy is a generalized function of all vertebratespecies (380).

We then narrowed the focus to carrion feeders,which include at least seven species: Great Basinspadefoot (Spea intermontana), which also feedson insects, small invertebrates, and sometimeseach other (cannibalism); coyote (Canis latrans),an opportunist scavenger for which carrion ismost important during winter; wolverine (Gulogulo); turkey vulture (Cathartes aura), whosefeeding behavior significantly contributes to de-composition of carcasses; bald eagle (Haliaeetus

Figure 29—continued

Text continues on page 52

Figure 30—The general form of species function profiles show-ing functional redundancy and total species functional diversityin a community. Each of the KEFs in a community has a num-ber of species with that function. Species function profiles aregraphs displaying the numbers of species with each specificKEF across communities. Functional redundancy is the numberof species with a specific KEF in a given community. Total spe-cies functional diversity is the number of KEFs times the meannumber of species per KEF (adapted from Huston 1994:5).

Table 8—Codes used in the species function profiles (fig. 31) for key ecological functions (KEFs)of speciesa

KEF code Description of KEF

1.2 Heterotrophic consumer (380)1.2.5 Carrion feeder (7)2.1 Ungulate herbivory (may influence rate or trajectory of vegetation succession and

presence of plant species) (15)3.1 Nutrient cycling relations: aids in physical transfer of substances for nutrient

cycling (C,N,P, other) (32)4.1 Interspecies relations: potential insect population control (22)4.2 Interspecies relations: vertebrate population control (10)4.3 Interspecies relations: pollination vector (6)4.4 Interspecies relations: transportation of seed, spores, plant or animal disseminules (153)4.8 Interspecies relations: primary cavity excavator in snags or live trees (17)4.9 Interspecies relations: primary burrow excavator (fossorial) (39)6.1 Soil relations: physically affects (improves) soil structure, aeration (typically by

digging) (10; combined with KEF 6.2)6.2 Soil relations: aids general turnover of soil nutrients and layers (10; combined with KEF 6.1)7 Wood relations (breaks down wood) (4)

a Total number of vertebrate species with each KEF within the basin assessment area is shown in parentheses. Taken from alonger classification presented in Marcot and others (1997).

49

Figure 31—Vertebrate species function profiles. These show thenumber of vertebrate species that have specific KEFs, listed by ter-restrial vegetation community, within the basin assessment area. Seetable 8 for description of KEFs. (A) KEF 1.2 = heterotrophic consumer,KEF 4.4 = transportation of plant or animal disseminules; (B) KEF1.2.5 = carrion feeder, KEF 2.1 = herbivory; (C) KEF 3.1 = nutrientcycling, KEF 4.1 = potential insect control, KEF 4.2 = vertebratepopulation control; (D) KEF 4.3 = pollination vector, KEF 4.8 = primarycavity excavator; and (E) KEF 4.9 = primary burrow excavator, KEF 6.1or 6.2 = soil structure relations, KEF 7 = wood relations (breaks downwood). Note that three terrestrial vegetation codes included in table 2are not displayed here; these are urban, water, and rock-barrencommunities.

Figure 31—continued

50

Figure 31—continued

Figure 31—continued

51

leucocephalus), which can also take live prey;common raven (Corvus corax); and gray jay(Perisoreus canadensis), an omnivore. Other ver-tebrate species doubtless feed on carrion as well.Presence of the seven species listed here wasmore variable among terrestrial vegetation com-munities (fig. 31b) than was the general functionof heterotrophy. The fewest number of carrion-feeders—that is, the least functional redundancyof the vertebrate community for this particularecological function—occurred in early seral pon-derosa pine forest and early seral subalpine forestvegetation communities, and reached its maxi-mum in late seral forests and woodlands.

Herbivory is another subset of heterotrophy, andan ecological function that displayed a profile dif-ferent from that of carrion feeding. Herbivoryreached its greatest functional redundancy in earlyseral montane forest and in upland shrub vegeta-tion communities (fig. 31b). As an ecologicalforce, herbivory can alter the vegetation structureand succession of habitats. For example, moose(Alces alces) heavily browse early seral speciessuch as aspen and willow, which can result in in-creases of conifers in habitats otherwise domi-nated by deciduous shrubs and trees (also seePastor and others 1993). In semipalustrine habi-tats, plant communities can be altered by herbi-vory of several species of waterfowl, includinggreater white-fronted goose (Anser albifrons),

brant (Branta bernicla), snow goose (Chencaerulescens), and Ross’ goose (C. rossii).Montane voles (Microtus montanus) and mead-ow voles (M. pennsylvanicus) can modify grass-land structures through herbivory. Mule deer(Odocoileus hemionus), white-tailed deer (O.virginianus), Rocky Mountain elk (Cervuselaphus nelsonii), and mountain goat (Oreamnosamericanus) can have major influences on com-position and succession of vegetation throughherbivory. Mountain goats can alter alpine andsome subalpine plant communities. Herbivorescan also interact with pollinators and affect plantreproduction and the evolution of plant traits(Brody 1996).

Another KEF is that of nutrient cycling relations,particularly species that substantially aid in phys-ical transfer of substances for nutrient cycling.Essentially, all organisms cycle nutrients. At leastsome 32 vertebrate species, however, affect nutri-ent cycling substantially in their environments(see Marcot and others 1997 for discussion). Thevertebrate species function profile for generalizednutrient cycling relations suggests a fair amountof variability of functional redundancy amongcommunities (fig. 31c). Communities with thegreatest redundancy in nutrient cycling func-tions among vertebrate species include late seralstages of most forest types, upland and riparianwoodlands, and early seral montane forest. Com-

Figure 31—continued

52

munities with the least redundancy include agri-cultural, alpine, exotic, upland herb and shrub,and riparian herb and shrub communities, pos-sibly because of overall lower vertebrate speciesrichness in at least some of these communities.

We developed species function profiles for sixcategories of interspecies relations: potential in-sect population control, vertebrate populationcontrol, pollination vectors, transportation ofplant or animal disseminules, primary cavity ex-cavation, and primary burrow excavation (fig.31). Some 22 species of insectivorous amphib-ians, birds, and mammals have the potential forcontrolling some nonirruptive insect populations;some woodpeckers can dampen the amplitude ofirruptive insect populations (Koplin 1969). Func-tional redundancy of potential insect control wasgreatest in upland and riparian woodlands andleast in upland and riparian herbland, exotic, andagricultural communities.

Predatory activities of at least 10 species can con-trol vertebrate populations. The species includefive mammalian carnivore predators and two rap-tors. Examples include American kestrel (Falcosparverius), which may aid in population controlof some insects and rodents, and Americanbadger (Taxidea taxus), which may help con-trol ground squirrel populations. Redundancyof vertebrate population control was less variableamong communities but reached its peak in up-land and riparian shrub, upland woodland, earlyseral montane forest, and—interestingly—alpine communities; and was least in most midand late seral forest communities.

At least six birds—five hummingbirds and anoriole—serve as pollination vectors for floweringplants (see Marcot and others 1997 for discus-sion). This species assemblage did not showmuch variation in redundancy among vegetationcommunities (fig. 31d). Another, much larger setof 153 species serve to transport plant or animaldisseminules; this set showed a fair amount ofvariation in redundancy among communities,with the greatest redundancy (most species) inearly seral montane forest, upland shrub, riparianwoodland, and upland woodland communities,and the least redundancy in early seral ponderosapine forest and early seral subalpine forest (fig.31a).

Primary cavity excavators—mostly woodpeckers,nuthatches, and chickadees—number 17 species(see Marcot and others 1997 for discussion) andshowed by far the greatest variation in functionalredundancy among vegetation communities of allecological functions explored here (fig. 31d). Pri-mary cavity excavation reached its greatest redun-dancy by far in all late seral forests and uplandand riparian woodland communities; and itslowest redundancy in agricultural, alpine, exotics,upland and riparian herbland, upland and riparianshrubland, and early seral forest communities.

Primary burrow excavators, a set of 39 species(see Marcot and others 1997 for discussion), alsoshowed significant variation among communities.The greatest functional redundancy was reachedin early seral montane forest, upland and riparianshrub, and alpine communities, and the least func-tional redundancy was reached in late seral forestcommunities, a rather inverse pattern to that ofprimary cavity excavators (fig. 31e). As withinvertebrate soil excavators, among vertebratesthe primary burrow excavators can be importantto soil turnover and mixing of organic matter.Gophers, for example, can process an enormousquantity of soil material. In one study in coastalSan Diego County, California, Cox (1990) foundthat the total soil mined by pocket gophers(Thomomys bottae) in 10 Mima-type moundsamounted to 8.23 Mg-1•ha-1•yr-1 and subsurfacedeposition was 20.31 Mg-1•ha-1•yr-1, so that totalsoil mining equalled 28.54 Mg-1•ha-1•yr-1.

Finally, soil relations and wood relations showedrelatively little variation in functional redundancyamong communities (fig. 31e). Soil relationsfunctions showed the greatest redundancy in earlyseral montane forest, and upland and riparianshrub communities, and the least redundancy inearly seral ponderosa pine forest and early seralsubalpine forest communities.

What general trends can be summarized from thepatterns of these selected ecological functions?First, no one vegetation community harbors thegreatest redundancy of all ecological functions.Most vegetation communities contribute togreatest redundancy of some specific ecologicalfunction. Thus, collectively, all vegetation com-munities combined help provide for redundancyof all ecological functions. Exceptions seem to be

53

agricultural and exotic (and urban) communities,which typically provide poorly for most func-tions explored here.

As a corollary, late seral forest communities seemto provide the greatest redundancy in carrionfeeding, general nutrient cycling, and primarycavity excavation functions. Late seral forestcommunities provide poorly for vertebrate popu-lation control and primary burrow excavationfunctions, which reach their peaks of redundancyin other communities, principally early seral for-est communities.

Alpine communities provide for high redundancyin vertebrate population control and in primaryburrow excavation functions. Out of six mammalspecies that regularly function as secondary bur-row users (that is, use burrows created by primaryburrow excavators), however, only two species(ermine [Mustela erminea] and long-tailed weasel[M. frenata]) occur in alpine communities, andthese species occur widely throughout other vege-tation communities as well. Thus, although alpinecommunities provide for high functional redun-dancy of primary burrow excavation, it does notprovide for high redundancy of secondary burrowuse.

Upland and riparian herblands do not provide formaximum redundancy of any of the functions ex-plored here. Upland and riparian shrublands pro-vide for high levels of redundancy of herbivory,vertebrate population control, pollination, primaryburrow excavation, and soil-relations functions.Upland and riparian woodlands provide for highlevels of redundancy of carrion feeding, generalnutrient cycling, potential insect population con-trol, vertebrate population control, pollination,and primary cavity-excavation functions. Thus,shrublands and woodlands provide for comple-mentary sets of ecological functions, just as dolate and early seral forest communities.

Early seral montane forest seems to provide foran anomalously wide array of conditions and eco-logical functions of all forest types in the basinassessment area. It provides for high redundancyof general nutrient cycling, vertebrate populationcontrol, pollination, primary burrow excavation,and soil-relations functions, whereas early seralponderosa pine forest and early seral subalpineforest communities do not. This may have to dowith the overall greater habitat and biotic diver-

sity found in early seral montane forests than inearly seral stages of other forest types, but thisneeds verification.

Which KEFs considered here are the most vari-able among terrestrial vegetation communities interms of number of vertebrate species (redundan-cy of species)? The answer is found by plottingthe standard error (SE) of number of speciesamong vegetation communities for each KEF. Re-sults (fig. 32) show that, of the functions consid-ered here, primary cavity excavation is by far themost variable in functional redundancy of speciesamong all terrestrial vegetation communities. Thismeans that only specific vegetation communitiesare likely to support most of the species with thisfunction. The next two functions with high varia-tion in redundancy are the carrion feeder functionand the primary burrow excavator function. Thus,the vegetation communities in which these most-variable functions occur are likely to be importantfor maintaining these species’ ecological func-tions within the basin assessment area. On theother hand, KEFs with the lowest SEs includetransport of plant or animal disseminules, hetero-trophic consumer, and pollination vector func-tions. These are the least variable in redundancyand the most reliable among communities; noneof the vegetation communities is particularly crit-ical for maintenance of these functions within thebasin assessment area (although specific com-position of species with these functions may dif-fer among communities).

Which terrestrial vegetation communities are themost variable in terms of redundancy among eco-logical functions? The answer is found by plot-ting the SE of species richness (number of spe-cies) among KEFs for each vegetation commu-nity. Results (fig. 33) suggest that early seralmontane forest has the greatest variation, andthree communities (mid seral montane forest,mid seral ponderosa pine forest, and mid seralsubalpine forest) have the least variation in re-dundancy among the ecological functions dis-cussed above. This does not mean that low-variation vegetation communities contain allKEFs, but only that the functions they do providefor tend to vary the least in number of species perfunction; there is a greater “evenness” (relativeproportion) but not necessarily “richness” (num-ber of different kinds) of species’ ecologicalfunctions.

54

Overall, it is unclear how the degree of func-tional redundancy of KEFs, either among speciesor among vegetation communities, specificallyaffects long-term resiliency of ecosystems to per-turbations of the environment and to perturba-tions of species community structures. A workinghypothesis might suggest that a greater redun-dancy promotes greater resiliency in ecosystemecological processes to stochastic environmentalevents and to short-term or localized losses ofsome species. This, however, is largely unstudiedin the basin assessment area. Our approach pro-vides an analytic framework and a repeatablemeans of posing such testable hypotheses onecosystem processes and functional redundancyamong communities.

Total species functional diversity—Huston(1994) characterized total species functional di-versity as the number of different ecological func-tions performed by all species in a communitytimes the mean number of species per function.Figure 34 presents estimates of total species func-tional diversity of vertebrates for terrestrial veg-

etation communities of the basin assessment areaby using the ecological functions discussed above.Over all communities, the greatest vertebratefunctional diversity is found in early seral mon-tane forest, followed by upland woodlands andriparian woodlands, and upland shrublands. His-torically, most vegetation cover types associatedwith early seral montane forest and upland wood-lands have increased in total area in the inlandWest since early historic times, whereas those ofriparian woodlands and upland shrublands havedecreased. Vertebrate species within these covertypes have shifted in relative abundance even ifvertebrate functional diversity has remained moreconstant.

The lowest vertebrate functional diversity isfound in early seral ponderosa pine forest, earlyseral subalpine forest, and agricultural lands(fig. 34). The vegetation cover types associatedwith early seral ponderosa pine forest communi-ties have decreased in area since early historic

Figure 32—Variation in functional redundancy of KEFs among terrestrial vegetation communities,displaying SE of number of species among vegetation communities, by KEF. Greater SE valuesdenote greater variation in functional redundancy among communities (see fig. 31 and table 8 fordescriptions of KEF codes).

55

times; those associated with early seral subalpineforest communities have both increased and de-creased; and agricultural lands have greatly in-creased. Interestingly, exotic vegetation com-munities do not rank particularly low in overallvertebrate functional diversity (fig. 34), althoughthere are many vertebrate species that do not findsuitable conditions in such habitats (Marcot andothers 1997). Thus, levels of vertebrate func-tional diversity among vegetation comunities arecomplementary to—not coincident with—patternsof vertebrate species diversity (see Marcot andothers 1997 for species diversity descriptions).

To what extent can we expect functional redun-dancy among vertebrate species with the sameKEF? This question was addressed by Marcotand others (1997) who conclude that generalfunctions, such as herbivory, cavity excavation,and pollination, might be considered redundant ifperformed by different species, but because eachspecies has its unique combination of habitat as-sociations and life history patterns, no two spe-cies can be expected to be exactly interchange-able. Thus, patterns of species functional diver-

sity presented here must be interpreted only asbroad-scale geographic and macroecologicaltrends and not interpreted as evidence for allow-ing any particular species loss or replacement incommunities.

Climate Change and Species at Risk

Results of climate change modeling suggest po-tentially complex local and regional climate re-sponses in the basin assessment area (see Hannand others 1997).8 Potential changes in climatepatterns among geographic locations within thebasin assessment area are not consistent in in-dicating warming and drying trends, as suggestedby Covington and others (1994) for the inland

Figure 33—Variation in functional redundancy of KEFs among functions, displaying SE of numberof species among KEF, by terrestrial vegetation community. Greater SE values denote greatervariation in functional redundancy among KEFs (see table 2 for descriptions of terrestrialvegetation communities).

8 Output from only one climate model—RegCM2, aregional climate model linked with a general circulationmodel to test regional responses to 2xCO2—is currentlyavailable for the basin assessment area. Hann and others(1997; also see footnote 2 in chapter 3) summarized resultsof that model along with knowledge of patterns that controlregional climate.

56

West,9 and by general circulation models for bo-real forests of high latitudes. The growing seasoncould become +1oC warmer with slightly greaterspring precipitation (+5 percent) and slightly lessautumn precipitation (-5 percent). Above 1000 melevation, however, winter might experienceslight cooling (-1oC minimums and -2.5oC max-imums). This, along with significant increases inwinter precipitation (+30 to 50 percent) couldresult in increased snow depth at high elevations,although the general circulation models reportedby Covington and others (1994) suggest overallregional warming trends. Below 1000 m eleva-tion, winter might experience significant warming(up to +3oC minimums and maximums) (seefootnote 2 in chapter 3).

Over all elevations, temporal changes in tempera-ture and precipitation would not occur smoothly.Thus, individual plant species adversely affectedwould be those unable to adapt to changes or in-creases in interseasonal and interannual variationin temperature and precipitation. Mostly, climate

changes and species-specific vegetation responsein the basin assessment area are poorly knownand not well modeled.

Given this uncertainty, we inspected the potentialresponse of species to one possible aspect ofclimate change, namely, high-elevation warmingduring the growing season. Knoll (1984) notedthat the effect of climatic change on extinctionsof plant species depends less on absolute changesof mean annual temperature and precipitation, andmore on successful migration of populations overgenerations to suitable areas. If suitable areasdo not exist, such as caused by warming trendspushing upper elevation climate and life zonesinto more restricted distributions (that is, intoareas of barren rock or ice, or even off the topsof mountains), or if access to suitable areas isblocked by dispersal barriers, then local extinc-tion rates greatly increase. We do not know howthese factors specifically will play out in the basinassessment area under climate-change scenarios,but we can surmise some general possible out-comes for upper elevation communities.

Figure 34—Total species functional diversity (number of KEFs per community times the meannumber of species per function), by terrestrial vegetation community of the basin assessmentarea (see table 2 for description of terrestrial vegetation communities).

57

Upper elevation vegetation communities, particu-larly alpine tundra and subalpine forest communi-ties, may experience higher summer temperaturesand precipitation regimes. Under the “individualresponse” hypothesis (Gleason 1926, Grime1977), plant species would respond individuallyand differently to climate changes (Covington andothers 1994, Mehringer, n.d.). Mehringer (1995,n.d.) notes that upper elevation vegetation canrespond rapidly to climate change; spruce (Piceaspp.) and subalpine fir (Abies lasiocarpa) abovetreeline in the Canadian Rocky Mountainsshowed rapid upslope advances to >110 m abovepresent treeline between 9,000 and 5,000 yearsago and again near 1,000 years ago. Chapin andothers (1995) note that arctic tundra has re-sponded to climate changes by altered nutrientavailability, growth-form composition, netprimary production, and species richness.

Alpine tundra and subalpine forest communitiescurrently are distributed among many mountainranges with differing degrees of geographic isola-tion (fig. 35). At present in the basin assessmentarea, alpine tundra covers 95 100 ha (83 percentof this occurs on BLM or FS lands and 17 percenton other lands), and subalpine vegetation com-munities cover 6 633 600 ha (94 percent of thisoccurs on BLM or FS lands and 6 percent onother lands). Upward elevation shifts of thesecommunities may result in increasing isolation ofclosely associated plant and animal species, re-duction in overall distribution and area of occur-rence in the basin assessment area, and local elim-

ination of these communities and closely associ-ated species. Mehringer (n.d.) notes that, becauseof great topographic relief in the Columbia Riverwatershed, even slight changes of temperatureregimes and vegetation response can have a greatinfluence on total areas of vegetation types.

Some of the hot spots of biodiversity and speciesrarity and endemism identified in Marcot andothers (1997) occur on isolated mountain tops,and thus may be at risk from rapid climatechanges. The degree of risk depends on howquickly the climate changes, how variable theclimate becomes during the change period, andhow well individual species can respond to suchvariability either through genetic plasticity or bychanging aspect, elevation, or topographic loca-tion, as was found for Mojave species (Spauldingand others 1983). Topographic situation—suchas mountain tops or upper elevation zones—maynot be limiting if changes occur slowly enough topermit plant species to respond. One example ofa hot spot of biodiversity in upper elevation ormountain settings is the eastern edge of the north-ern Cascade Range of Washington. In this area,glaciated landscapes, rounded peaks, high eleva-tions, and subalpine climates provide for uniqueassemblages of amphibians and large carnivores,in contrast with the surrounding landscapes, andhold wetlands, bogs, and meadows that housemany rare plants. Another biodiversity hot spotin mountain top settings is Glacier National Parkand the higher elevations of Flathead NationalForest in northwestern Montana, where rare largecarnivores and intact rare plant communitiesoccur in relatively undisturbed and isolated sub-alpine and alpine Rocky Mountain environments.Three examples of hot spots of species rarity andendemism in mountain top settings in Oregoninclude the crest of the Umatilla Mountains,where many west-side species are isolated anddisjunct on these mountain tops; Wallowa Moun-tain peaks and Sacajewea peak, which hold dis-junct Rocky Mountain and boreal species; andthe Three Sisters area in Deschutes National For-est, which has unique volcanic substrates andgeologic landforms that house locally endemicplants. Other examples are listed in Marcot andothers (1997).

Other ecological aspects of upper elevation com-munities may be affected by shifts in precipita-

9 Covington and others (1994:45) report that “generalcirculation models (GCM) predict an increase of 1.5-2.5o Cin mean annual temperature within the next 30 to 50 years....Actual meteorological records for the 1980 decadeand for the century indicate the Inland West region isparticularly vulnerable to global warming and to extrememoisture stress.” Results of the regional climate modelsimulations conducted for the present study, however,suggest a more complex picture for the basin assessmentarea (see footnote 2, chapter 3). The difference between themodel predictions may be because of the difference inspatial resolution of the models. Covington and others doconcur with transition conditions to a new, futureequilibrium of (debatably) higher temperatures beingmarked by extreme climatic variability and discontinuities.Obviously, the final word on climate modeling and regionaleffects has not been written, and much work remains to bedone, particularly in reconciling general circulation climatemodels with regional climate models.

58

tion, temperature, or disturbance (especially fire)regimes (Tausch and others 1993). As comparedwith lower elevation forests, subalpine forestsare characterized by colder annual temperatures,heavy winter snowpacks, forest floors as impor-tant “sinks” of total ecosystem nitrogen (“loose”nitrogen cycles), greater importance of mycor-rhizae in nutrient uptake, slower growth rates oftrees, greater carbon allocation to root biomass,long retention of foliage, and long lifespan oftrees (Vogt and others 1989), as well as high ar-boreal epiphyte loads and occurrence of someclosely associated plant and animal species, in-cluding endemics. Some of these characteristics,particularly nitrogen sinks, retention of foliage,and lifespan of trees, may help buffer otherwiseadverse community responses to quick changesin environmental conditions, including climatechange. But other characteristics, particularly my-corrhizae importance, and occurrence of closeassociates and endemics, may impede quick“adaptation” to new climatic and disturbance re-gimes. Overall, potential changes in climate and

disturbance patterns, and potential communityresponse, are likely to be complex.

Local climate changes in lowland environmentsalso may result in changes in fire regimes. Plantsthat respond differently to occurrence, intensity,frequency, and seasonality of fires likely wouldshow differential increases or decreases underchanging fire regimes. Marcot and others (1997)present a list of rare or potentially rare plants ofthe basin assessment area showing their orienta-tion to fire regimes; this list can be consulted tohypothesize how species might respond. Resultsare quite mixed. For example, overstory lethalfires may benefit 7 lichen groups, 9 plant groups,and 9 vascular plant species, but harm a differentset of 5 bryophyte groups, 11 lichen groups,2 plant groups, and 11 vascular plant species. In-creasing fire frequency, increasing fire-suppression activities, and changes in the sea-sonality of prescribed and natural fires—all ofwhich may occur from regional changes inclimate—also have mixed results.

ICBEMP

Legend

Alpine

Subalpine

State boundaries

Basin boundary

Figure 35—Current distribution of alpine tundra and subalpine forest communities in thebasin assessment area.

59

Invertebrate responses to changes in fire regimesis poorly studied and largely unknown, exceptfor forest insect pests (Hann and others 1997).Among vertebrates, stand-replacing fires can cer-tainly change community composition. Hutto(1995) documented differential response by birdspecies to stand-replacement fires in the northernRocky Mountains, and noted that one species—black-backed woodpecker (Picoides arcticus)—was nearly restricted to sites with standing fire-killed trees. Hutto also listed other bird speciesthat may benefit over the long term from stand-replacement fires.

In some settings, vertebrate species may showclearer responses to changes in climatic condi-tions than would plants. We identified vertebratespecies closely associated with upper elevationvegetation communities and thus possibly atgreatest risk from upward altitudinal shifts ofthese life zones. In the basin assessment area,some 53 species—5 amphibians, 1 reptile, 22

birds, and 25 mammals—are associated withalpine tundra communities. Many of these speciesalso occur in other vegetation communities. Eightof these species (5 birds and 3 mammals) arecloselyassociated with alpine tundra com-munities (that is, occur in <20 percent of all 44vegetation cover types) (fig. 36) and thus may beat greatest risk from warming trends. Of theseeight species, white-tailed ptarmigan (Lagopusleucurus) is found solely in alpine tundra, andAmerican pipit (Anthus rubescens) occurs in onlyone other vegetation cover type (that is, occurs in<10 percent of all 44 vegetation cover types).Ptarmigan and pipits are especially vulnerable tochanges in alpine tundra communities.

Subalpine forest communities are potentiallyvulnerable to upper elevation warming trends(Graumlich 1991, Romme and Turner 1991,Schullery 1995), although Peterson (1995) reportsthat subalpine forests have been increasing in theWest because of warmer climates. Peterson

Figure 36—Vertebrate species at risk of regional warming trends, through decreases in total area of alpinetundra. Species shown are closely associated with alpine tundra vegetation communities (that is, occur inalpine tundra and <20 percent of all 44 vegetation cover types) in the basin assessment area.

60

(1995) also concludes that growth response ofsubalpine forests varies considerably accordingto geographic climate settings and microsite con-ditions. In the basin assessment area, subalpineforest communities include Engelmann spruce(Picea engelmannii)/subalpine fir forest, moun-tain hemlock forest, whitebark pine forest, andwhitebark pine/subalpine larch forest cover types.Whitebark pine forest may be particularly vulner-able to climatic warming. Schullery (1995) pro-jected that existing whitebark pine in Yellow-stone National Park could decline by about 90percent under a modest increase in warmth anddryness (also see Whitlock 1993).

A total of 205 vertebrate species—20 amphibians,6 reptiles, 106 birds, and 73 mammals—are asso-ciated with subalpine forest communities. Ofthese, 36 species (8 amphibians, 16 birds, and 12mammals) arecloselyassociated with subalpineforest communities (that is, occur in <20 percentof all 44 vegetation cover types) and thus may beat particularly greater risk from upper elevationclimatic warming (fig. 37). Of these 36 species,13 occur in few other vegetation types (<10 per-cent of all 44 vegetation cover types) and areparticularly vulnerable. Among these 13 highlyvulnerable species, the salamanders include theregionally endemic Cascade torrent salamander

Figure 37—Vertebrate species at risk of regional warming trends, through decreases in total area of subalpineforests. Species shown are closely associated with subalpine forest vegetation communities (that is, occur insubalpine forests and <20 percent of all 44 vegetation cover types) in the basin assessment area.

61

and Dunn’s salamander, and the locally endemicLarch Mountain salamander; the birds includespruce grouse (Dendragapus canadensis), borealowl (Aegolius funereus), black scoter (Melanittanigra), surf scoter (M. perspicillata), white-winged scoter (M. fusca), oldsquaw (Clangulahyemalis), three-toed woodpecker (Picoidestridactylus), and boreal chickadee (Parushudsonicus); and the mammals include black-tailed deer (Odocoileus hemionus columbianus)and California myotis (Myotis californicus).

The three aforementioned regionally or locallyendemic salamanders may be the most vulnerableof all 13 species in terms of changes of subalpineforest communities within the basin assessmentarea. The three aforementioned scoters occur inthe basin assessment area as occasional migrants;the surf scoter is an occasional winter visitor,although the white-winged scoter has been re-corded nesting in northeastern Washington. Thereis likely very little danger to the viability of anyof the overall populations of scoters caused byhabitat changes within the basin assessment area.Black-tailed deer ranges mostly west of theCascade Range and occurs in the basin assess-ment area only along the high Cascade Range.Climate change might reduce the area of suitablehabitat for black-tailed deer within the basin as-sessment area, although they likely would persistwest of the crest of the Cascade Range. The restof the subalpine forest community species listedabove occur more regularly or more extensivelywithin the basin assessment area.

We expect that increasing isolation of upper ele-vation environments might result in differentiallocal isolation or extinctions of some animals,much as past rising temperatures likely drovesome mammal species and their habitats upslopeinto disjunct distributions on mountaintop islandsin the Great Basin (Diamond 1984). Likely survi-vors would include small mammals over bigmammals of the same trophic status and habitatpreference, habitat generalists over habitat spe-cialists, and habitat generalist herbivores oversimilarly sized habitat generalist carnivores (Dia-mond 1984). In alpine tundra communities in thebasin assessment area, such likely survivors mightinclude the following herbivorous small mammals

(≤ squirrel size) that are habitat generalists (oc-cur in≥50 percent of all 44 vegetation covertypes): deer mouse (Peromyscus maniculatus),Columbian ground squirrel (Sperm ophiluscolumbianus), golden-mantled ground squirrel(S. lateralis), long-tailed vole (Microtuslongicaudus), and masked shrew (Sorexcinereus).

The converse of these criteria might describemammal species more ecologically at risk ofclimate-induced changes in alpine tundra habitat.These species include big mammals (> squirrelsize) over small mammals, habitat specialists(<50 percent of all 44 vegetation cover types)over habitat generalists, and habitat generalistcarnivores over similarly sized habitat generalistherbivores. In the basin assessment area, the oneat-risk mammal of alpine tundra fitting these cri-teria is wolverine.

Neilson (1991, 1993) and Noble (1993) note thatecotones between biomes may be sensitive areasto climatic change and therefore useful for moni-toring change. Specifically, two types of change,boundary shifts of regions and physiognomicshifts within regions, are potentially independentbecause of differential species responses. Eachtype of change may require different strategiesfor monitoring.

The response of plant, invertebrate, and verte-brate communities to any directional changes inclimate regimes likely will be made by individualspecies rather than by entire communities. Insome cases, however, coadapted species com-plexes may suffer greater disruption than indi-vidual species’ habitat associations may suggest.These complexes may include insect-plant polli-nation relations, invertebrate phoresis (especiallydependence on other organisms for dispersal),mychorrhizal fungi-tree relations, and othermutualistic, commensal, and symbiotic relations.

For example, Clark’s nutcracker (Nucifragacolumbiana) is a key element in regeneration ofwhitebark pine—on the decline in the assessmentarea—by caching (burying) seeds in the ground.Clark’s nutcracker occurs in 16 vegetation cover

62

types in the basin assessment area. It is unclearhow complex shifts in plant species, vegetationcover types, and disturbance regimes resultingfrom climate change might affect Clark’s nut-cracker populations. Wells (1983) hypothesizedthat Clark’s nutcracker (along with pinyon jay(Gymnorhinus cyanocephalus)) might have fa-cilitated brisk redistribution of pinyon pine (Pinusedulis) in the intermountain West following rapidHolocene (Recent) climate changes. However, iflocal, high-elevation populations of Clark’s nut-crackers become reduced and isolated because ofwarming-induced declines in total area of mon-tane and subalpine forests (the species does nottypically occur in alpine tundra communities ormuch below montane forest zones), there may bea compounding adverse affect on whitebark pineredistribution and viability (see Marcot and others1997 for further discussions of other vertebratespecies relations).

Another relation that might complicate speciesresponse to changing climatic and disturbanceregimes is that of mycorrhizal fungi, their hosttree species, and soils. Disturbances that reduceaeration or soil organic matter reduce mycorrhizalactivity (Amaranthus and Perry 1994). Reductionin mycorrhizae reduces growth of many speciesof trees and retards productivity of soils. Somestudies suggest that sites with harsh, continentalclimates (such as parts of the basin assessmentarea) may have lower fungal diversity thanmarine, coastal climates, and thus may be poorlybuffered against changes in fungus populationscaused by disturbance (Powers 1989) or by ad-verse climate change.

Climate change also has the potential to alter de-composition processes in grasslands and coni-ferous forests (Anderson 1991, Breymeyer andMelillo 1991, Esser 1992), plant respiration(Ryan 1991), phenology and growth of alpinevegetation (Walker and others 1994, 1995), bogformation (Foster and Wright 1990), and the roleof soil in carbon sequestration (Van Veen andothers 1991). Van Veen and others (1991) reportthat an increased level of CO2 (a so-called“greenhouse gas”) in the atmosphere likely willresult in an increased input of organic carboninto the soil because of the expected increase inprimary production. Whether this will lead toaccumulation of greater amounts of organiccarbon in soil depends on the flow of carbon

through the plant into the soil, the tightness of thecarbon cycle, and its sub-sequent transformation in the soil by micro-organisms.

In this section, we have discussed potential futurechanges of terrestrial ecological communities andprocesses resulting from climate shifts. In the nextsection, we peer backwards and discuss prehistor-ic changes in climates and in the diversity ofbiota.

Paleoecology and Trends inVertebrate Species andCommunities of the BasinAssessment Area

Late Quaternary Conditions andChanges of the Basin AssessmentArea

The Quaternary period (Pleistocene and Holocene[Recent] epochs of 1.64 million years ago to thepresent) represents a time of remarkably variableclimates and biota in the inland West (McDonald1984, Robbins 1993, Robbins and Wolf 1994).Climate cycles during this period were a majorforce in the evolution of plant response to change,so that, according to Tausch (1993), current plantcommunities today may be far less stable thanthey appear to be from our perspective.

Using pollen records from the late Quaternary,Mehringer (1985) reconstructed the followingvegetation changes in selected sites of the inlandWest. Pollen sequences from the Holocene sug-gest fluctuations in vegetation resulting fromshort, acute climatic changes and from fire andvolcanic episodes. Continental ice, alpine gla-ciers, and vast pluvial lakes had covered much ofthe area during the last glacial episode. Most ofthe northern Great Basin and adjacent provinceswere covered by cold sagebrush-steppe conditionswith montane conifers growing at lower eleva-tions than they occur today.

Then, before 9,000 years ago during the earlyHolocene, a warming trend caused shrinkinglakes, wasting glaciers, and catastrophic flooding(Allen and others 1986). Pioneer plants such asbuffalo-berry (Shepherdia canadensis(L.) Nutt.)increased and forests spread northward, while

63

sagebrush-steppe spread southward as montaneforests retreated to upper elevations. Douglas-firbegan its spread to become a dominant elementin conifer forests (Hermann 1985).

By 7,000 years ago, conifer forests lost ground tograsslands and sagebrush, and shadscale and sage-brush communities replaced grasslands. By 5,400years ago, this trend slowed, and by 4,000 yearsago, it reversed. A moister maritime climate de-veloped in northern Idaho by 2,500 years ago orsooner, and fire regimes changed (Mehringer1985). By the late Quaternary, lodgepole pinehad spread north from Pleistocene glacial refugesfarther south (Critchfield 1985). Relict lodgepolepine populations persisted in the Rocky Moun-tains (MacDonald and Cwynar 1985). Eventually,Douglas-fir became an important part of North-western forests during interglacial periods of theQuaternary (Hermann 1985).

The Quaternary fauna—known mostly frommammal records—was characterized by thespread of many large mammals throughout theinland West, as summarized by Potts andBehrensmeyer (1992). These included genera(1) now globally extinct, such as the mammoth(Mammathus), the bovidSoergelia, and saber-toothed cat (Smilodon); (2) extinct from the basinassessment area but persisting today elsewhere,such as jaguar (Panthera); and (3) persistingwithin the basin assessment area to this day, in-cluding wolf (Canis), mustelids (including er-mine), moose (Alces), caribou (Rangifer), horse(Equus), cricetid rodents of genusClethrionomys,hare (Lepus), and vole (Microtus). Five Quater-nary families of artiodactyls occurred, includingbovids, camelids, and antilocaprids, as well asxenarthrans, including glyptodonts, armadillos,and ground sloths. Proboscideans includedgomphotheres and mastodonts in addition tothe mammoths. Cricetid rodents underwentexceptional adaptive radiation during the Plio-Pleistocene as savanna and grassland ceded tosteppe environments.

During the Pleistocene, rodents diversified, whileother groups (proboscideans, perissodactyls,xenarthrans, and camelids) reduced. Immigrants

across the Bering land bridge to the area duringthis time included bison (Bison), mountain sheep(Ovis), muskox (Ovibos), moose (Alces), lion(Panthera), and humans (Homo sapiens) (Pileou1991). Climatic fluctuations served to blendbiotas during this time, mixing these newcomerimmigrants with the existing resident saber-toothed cat, dire wolf (Canis), horse, mammoth,many squirrels (sciurids), New World rodents(cricetids), and bovids (Potts and Behrensmeyer1992).

In the late Pleistocene, large mammals sufferedwaves of extinctions between 15,000 and 9,000years ago. This “megafaunal extinction” took outmammoths, mastodonts, gomphotheres, saber-toothed cats, dire wolves, horses, sloths, camels,and many species of bears, deer, antilocaprids,and others. It may be important to understand thereasons for these extinctions—as yet, they are notfully understood—to better interpret potentialeffects of our modern day changes in climatesand habitats. Causes of the megafaunal extinctionepisodes have been variously attributed to humanhunting, climate changes and associated habitatshifts during the early interglacial periods, inter-specific competition, and differential resorting ofspecies into increasingly disjunct communities(Martin and Klein 1984).

Glacial cycles in the Holocene caused manymammalian taxa to separate in distribution ashabitats became more patchy and disjunct andas mammal species responded individually tochanging environments (Potts and Behrensmeyer1992). The resulting communities have beencalled nonanalogue communities or disharmoni-ous faunas. Such faunas had high beta diversityof species among geographic locations, such asdifferent mountain ranges. Some authors attrib-uted these changes and the persistence of non-analogue communities to periods of more equableclimates (Potts and Behrensmeyer 1992).

Several important lessons for interpretingpresent-day biota in the basin assessment areacan be learned from this brief review of Quater-nary period changes over the span of a dozenmillennia:

1. Climate shifts have occurred unpredictably andhave signalled great changes in both distributionand species composition of floras and faunas.

64

2. Episodic extinctions of at least large mammalshave occurred, perhaps associated with climatechange, human hunting pressure, or other factors.

3. Differential extinctions and originations(speciation and immigration of existing species)have occurred mostly with unpredictable patternsbut over periods of millennia. Examples are thegreat reduction in artiodactyls (especially horses)and the increase in diversity of cricetids (miceand rats) and sciurids (squirrels and chipmunks).

4. Separation and isolation of species because ofclimatic shifts resulted from differential responseand sorting by species during periods of climatechange and equilibration.

At least as a working hypothesis, we might con-jecture that today’s biotas of the basin assessmentarea might respond with such similar, general pat-terns to increasing environmental pressures,changing habitats, and potentially shifting cli-mates. Unknowns in this hypothesis include (a)rates of response of species, assemblages, andecological communities, which may be unable toadequately respond and thereby might vanish; (b)species-specific changes that may ensue, althoughwe can now draw from databases on species-environment relations to begin to frame specificpredictions (for example, see the section aboveon potential near-future climate change); and (c)how effects of human activities might compoundwith background natural variations in climatesand environments in affecting species and ecolog-ical communities. In this vein, historic and cur-rent human activities of greatest concern on Fed-eral lands include short-term changes of naturaldisturbance regimes such as fire, and humanactivities directly affecting populations or hab-itats including hunting, urbanization, livestockgrazing, mining, and timber harvesting.

Two still-unanswered questions about climatechange during the Quaternary within the basinassessment area may prove of interest tomanagement:

1. An explicit test is needed of the null hypothesisthat the range of historic conditions (past 50 to200 years) does not reflect the range of prehis-toric (Holocene) conditions, in terms of climateand vegetation. To test this hypothesis, one mightuse pollen diagrams of vegetation change (for ex-ample, Mehringer 1985) as compared with the

projections of historic vegetation conditions(Hann and others 1997).

2. An explicit test is needed of the null hypoth-esis that there is no long-term (Holocene or Qua-ternary) set of repeating conditions of climate andvegetation patterns within which an idealized,stable range of natural variability (sensu Morganand others 1994) can be estimated. This hypothe-sis can be tested by calculating a running SE ofclimate parameters or vegetation conditions (astaken from pollen profiles), as a function of time,and determining if it reaches an asymptote, thussignifying a stable range of natural variability.

The first question above is important in determin-ing if the range of historic conditions truly pro-vides for environmental conditions in which spe-cies immigrated or evolved, and in which com-munities developed. The second question is im-portant for testing the very concept of range ofnatural variability.

Assuming that understanding the past and presenthelps us project the future, additional work on thefollowing ecological topics would prove useful:

1. Identify Quaternary source pools of species,zones and directions of spread of species, andspecies dispersal dynamics, rates, and barriers.

2. Identify evolutionary adaptations and selectionmechanisms during the Quaternary, includingallopatric speciation, character divergence amongsympatric species, isolating mechanisms, and spe-cies swamping.

3. Study reproductive ecology pertinent to under-standing adaptations, including hybridization(Samson and Knopf 1994).

4. Study development of keystone roles of speciesand their expression in changing environmentsand communities.

5. Identify changes in endemism by taxonomiclevel and proportion of taxa, and mechanismscausing endemism (such as Pleistocene refugia,allopatric or sympatric speciation, and isolationby physiographic conditions such as montaneislands, flood-carved gorges, and scablandlandscapes).

65

Tertiary Biota and Comparison With ExtantVertebrates in Eastern Oregon

Work at John Day Fossil Beds National Monu-ment has described the ancient (pre-Plioceneepoch, Tertiary period) flora and fauna of easternOregon (Retallack and others 1996).10 The bedsoccur in several arms of the John Day River sys-tem in what is now mainly grazed rangeland, ag-riculture land, and some sagebrush-steppe habitat.Fossil records from the beds span more than 40million years and date from 45 to 5 million yearsago. Although a few amphibians, reptiles, and oneraptor have been identified in fossil records at thebeds, the most diverse fossil evidence is that ofmedium- to large-sized mammals. At the beds,some 12 orders, 47 families, 155 genera, and 298species of mammals have been identified span-ning the 40-million-year period (fig. 38; seefootnote 10).

Of these Tertiary taxa, 8 orders (67 percent),16 families (34 percent), 5 genera (3 percent), andno species (0 percent) remain extant in the basinassessment area (fig. 39). The Tertiary mam-mal genera still exant from the Tertiary are twocarnivora,Martes(Mustelidae) andCanis(Canidae);two rodentia, “Sciurus” andSpermophilus(both Sciuridae); and a chiropteranbat,Myotis(Vespertilionidae). In addition, stillextant in the basin assessment area are theamphibian (frog)Rana(Ranidae), and the turtleClemmys(Testudinidae). The current speciesfound within the general basin assessment area re-presenting these ancient forms are listed in table9. These are taxa of particular paleoecologicalinterest because of the antiquity of the genera.Fossils of all these ancient genera were takenfrom formations representing the mid-lateMiocene (8 to 6 million years ago), the mostrecent of the Tertiary formations in John DayFossil Beds. An exception was “Sciurus,” whichcame from the much older formations spanningthe late Eocene, Oligocene, and early Miocene(39 to 20 million years ago). In addition,ClemmysandMartesalso were found in formations fromthe mid-Miocene (15 to 12 million years ago).

Many other ancient families are still extant in thebasin assessment area even though none of theirTertiary genera has survived. These ancientfamilies, listed in table 10, include those givenabove as well as additional families of turtles,lizards, avian raptors, marsupials, carnivores,horses, pronghorns, rodents, and rabbits.

In comparison, the total extant mammal faunaof the basin assessment area consists of 8 orders,24 families, 67 genera, and 132 species (fig. 38).Compared to all mammalian fossil taxa dis-covered at John Day Fossil Beds, the number ofextant orders is 67 percent of the total number ofTertiary orders; families are 51 percent; generaare 43 percent, and species are 44 percent. Interms of taxonomic richness (number of taxa),then, the Tertiary mammal fauna was far richerthan that at present, at all taxonomic levels.

This comparison, however, is uneven in severalways. (1) The fossil record differentially pre-serves evidence of species with various life forms,body size, and habitat associations. For example,bats are represented by only a single species inthe fossil records of the beds, whereas the currentbat fauna includes some 15 species; more than asingle species probably existed prehistorically,but other species have not been preserved ordetected yet in the fossil record. As well, rare spe-cies probably often go undetected in fossil records(Koch 1978). (2) The mammalian fossil recordsat the beds represent an enormous span of time ascompared with today’s “snapshot,” which is takenover a much broader geography of the basin as-sessment area. Results of the above comparisonswould be even more exaggerated if confined tojust the extant sagebrush-steppe fauna, becausethe current local biota of just the sagebrush-steppeenvironment of eastern Oregon is substantiallyless diverse than that of the full basin assessmentarea. (3) Taxonomic problems of comparisoninclude the fact that species are more difficulttoidentify in the fossil record than are orders orfamilies, and prehistoric morphospecies do notexactly coincide with current species in termsof equivalent systematics. Still, the differencesbetween prehistoric and extant number of taxa

10 Personal communication. 1996. T. Fremd, John DayFossil Beds National Monument, John Day, Oregon.

66

Figure 38—Biodiversity (number of taxa) of prehistoric (Tertiary) mammals ofJohn Day Fossil Beds, eastern Oregon; and extant (still living) mammals of thebasin assessment area. (Source of Tertiary biodiversity information: seefootnote 10 in chapter 4.)

Figure 39—Biodiversity of prehistoric (Tertiary) mammals of John Day FossilBeds, eastern Oregon, showing number and percentage of prehistoric taxapersisting to the present.

67

Table 9—Currently extant species or subspecies of vertebrates occurring in the basin assessmentarea that are represented by Tertiary genera discovered in John Day Fossil Beds NationalMonument, eastern Oregon, including the formation in which each genus was discovereda b

Tertiary genus Extant species or subspecies Order, family Formationc

Rana R. aurora, red-legged frog Anura, Ranidae AR. boylii, foothills yellow-legged frogR. cascadae, Cascades frogR. catesbeiana, bullfrogR. clamitans, green frogR. luteiventris, spotted frogR. pipiens, northern leopard frogR. sylvatica, wood frog

Clemmys C. marmorata, western pond turtle Chelonia, A, BTestudinidae

Martes M. americana, American marten Carnivora, A, BM. pennanti, fisher Mustelidae

Canis C. latrans, coyote Carnivora,C. lupus, gray wolf Canidae A

“Sciurus” S. griseus, western gray squirrel Rodentia, CS. niger, eastern fox squirrel Sciuridae

Spermophilus S. armatus, Uinta ground squirrel Rodentia, AS. beecheyi, California ground squirrel SciuridaeS. beldingi, Belding’s ground squirrelS. brunneus, Idaho ground squirrelS. columbianus, Columbian ground squirrelS. elegans nevadensis, Wyoming ground squirrelS. lateralis, golden-mantled ground squirrelS. saturatus, Cascade golden-mantled ground squirrelS. townsendii, Townsend’s ground squirrelS. washingtoni, Washington ground squirrel

Myotis M. californicus, California myotis Chiroptera, AM. ciliolabrum, western small-footed myotis VespertilionidaeM. evotis, long-eared myotisM. lucifugus, little brown myotisM. thysanodes, fringed myotisM. volans, long-legged myotisM. yumanensis, Yuma myotis

a See footnote 10 in chapter 4.b These living taxa belong to ancient genera and are of particular evolutionary and biogeographical interest. All species andsubspecies listed here are currently found within the basin assessment area although not necessarily at John Day Fossil Beds.c Formations: A = mid to late Miocene, 8 to 6 million years ago; B = mid to Miocene, 15 to 12 million years ago; and C = lateEocene, Oligocene, and early Miocene, 39 to 20 million years ago.

68