PETITION TO LIST THE

RIO GRANDE CUTTHROAT TROUTOncorhynchus clarki virginalis

AS AN ENDANGERED SPECIESUNDER THE U.S. ENDANGERED SPECIES ACT

February 16, 1998

Southwest Center for Biological DiversityEndangered Species Petition #37

February 16, 1998

Mr. Bruce BabbittSecretary of the InteriorOffice of the SecretaryDepartment of the Interior18th and "C" Street, N.W.Washington, D.C. 20240

The Southwest Center For Biological Diversity, Southwest Trout, the Biodiversity LegalFoundation, Carson Forest Watch, Ancient Forest Rescue, Rex Johnson and Kieran Sucklinghereby formally petition to list the Rio Grande cutthroat trout (Oncorhynchus clarki virginalis) asendangered pursuant to the Endangered Species Act, 16 U.S.C. 1531 et seq. (hereafter referred toas "ESA"). This petition is filed under 5 U.S.C. 553(e) and 50 CFR 424.14 (1990), which grantsinterested parties the right to petition for issue of a rule from the Assistant Secretary of theInterior.

Petitioners also request that Critical Habitat be designated concurrent with the listing, pursuant to50 CFR 424.12, and pursuant to the Administrative Procedures Act (5 U.S.C. 553).

Petitioners understand that this petition action sets in motion a specific process placing definiteresponse requirements on the U.S. Fish and Wildlife Service and very specific time constraintsupon those responses.

Petitioners

The Southwest Center For Biological Diversity is a non-profit public interest organizationdedicated to protecting the diverse life forms of the American Southwest and northern Mexico. Ithas offices in New Mexico, Arizona, and California.

Southwest Trout is a non-profit public interest organization dedicated to restoring the coldwaterfish habitats of the American Southwest and the Sierra Madre Occidental of northern Mexico.

The Biodiversity Legal Foundation is a non-profit public interest organization dedicated to thepreservation of all native wild plants and animals, communities of species, and naturallyfunctioning ecosystems in this country. Through visionary educational, administrative, and legalactions, the BLF endeavors to encourage improved public attitudes and policies for all livingthings.

Carson Forest Watch works to protect and restore the forests, streams and rivers of northern NewMexico where the Rio Grande cutthroat trout lives.

Ancient Forest Rescue is dedicated the protection of the forests, streams and wildlife of southernColorado.

1

TABLE OF CONTENTS

I SPECIES DESCRIPTION 3

II TAXONOMY 4

III DISTRIBUTION 4

IV NATURAL HISTORY 9

V HABITAT REQUIREMENTS 9

VI RANGE REDUCTION 10

VII HISTORIC POPULATION DECLINE AND CURRENT INSTABILITY 14

VIII HABITAT FRAGMENTATION 19

IX HABITAT LOSS- LIVESTOCK GRAZING 21

X HABITAT LOSS- WATER DIVERSIONS 24

XI FLOOD, FIRE AND DROUGHT 25

XII ROAD BUILDING AND LOGGING 25

XIII DISEASE AND PREDATION 27

XIV COMPETITION & HYBRIDIZATION 28

XV POOR WATERSHED CONDITION 30

XVI INADEQUACY OF EXISTING REGULATORY MECHANISMS 30

XVII CRITICAL HABITAT DESIGNATION RECOMMENDED

35

XVIII REFERENCES CITED 37

2

FIGURES

Figure 1 Rio Grande Cutthroat Trout (Oncorhynchus clarki virginalis) 3

Figure 2 Current and Historic Range of the Rio Grande Cutthroat Trout in NewMexico and Colorado.

5

Figure 3 Suitable Trout Streams and Current Populations of Rio Grande CutthroatTrout Within Its Known Historic Range

8

Figure 4 Extirpated, At Risk/Declining, At Risk/Stable, Secure/Stable, andUnknown Status of Hydrologic Sub-basins Within the Historic Range ofthe Rio Grande Cutthroat Trout.

17

Figure 5 Livestock Have Destroyed Riparian Cover on the Rio Tusas, Tres PiedrasRanger District, Carson National Forest. This Was Formerly ExcellentTrout Habitat.

22

Figure 6 Recent logging on the Conejos Peak District of the Rio Grande NationalForest near Cumbres Pass. Erosion from this tributary severely damageddownstream trout habitat.

26

Figure 7 Angostura Creek, Carson National Forest. This spot is marked for loggingin the proposed Angostura Timber Sale. New roads for the sale would alsodamage Rio Grande cutthroat trout population and habitat.

26

TABLES

Table 1 Historic distribution and current status of the Rio Grande cutthroat trout byUSGS hydrologic sub-basin (from Stumpff and Cooper 1996)

11

Table 2 Total Rio Grande cutthroat trout habitat currently available and currentlyoccupied on selected federal land in New Mexico.

12

Table 3 Status of 68 Recorded Stream and Lake Populations of Rio GrandeCutthroat Trout in Colorado.

15

Table 4 Population Stability Definitions. 18

Table 5 Sub-basin and Population Status. 18

3

Figure One. Rio Grande Cutthroat Trout (Oncorhynchus clarki virginalis)

I. SPECIES DESCRIPTION

The Rio Grande cutthroat (Oncorhynchus clarkivirginalis) is a strikingly beautiful fish (seefigure one). It is closely related to two othercutthroat subspecies originally found indrainages adjacent to the Rio Grande, to thenorth and west: the large-spotted greenbackcutthroat trout (Oncorhynchus clarki stomias),which inhabited the upper Arkansas Riversystem; and the Colorado cutthroat(Oncorhynchus clarki pleuriticus), whichoccupied the headwaters of the Colorado River.Two distinct forms of virginalis are known toexist, one from the Pecos River basin (the"Pecos form") and the other from the remainderof the Rio Grande system (the "Rio Grandeform"). The trout from the Pecos drainage haveextremely large spots up to an inch (30 mm.) indiameter, typical of the stomias, and more scalesalong the lateral line, typically 175. The otherform more closely resembles the pleuriticussubspecies.

The number of vertebrae of both forms ofvirginalis is usually 61 or 62; pyloric caeca are30-50, a slightly higher total than found in bothstomias and pleuriticus. Basibranchial, or hyoidteeth are weak or vestigal. Slight hybridizationwith rainbow trout will cause virginalis

populations to lose all signs of these teeth in 50percent or more of the individuals (Behnke,1979). Hybridization with the Yellowstonecutthroat (O. c. bouvieri) causes the hyoid teethto become more prominent.

Coloration of virginalis is similar to pleuriticusand stomias, though not as intense, with olive togreenish yellow background on the head andback, olive to yellow just below the lateral line,often overlaid with high colors of burnishedorange and red on the lower body up to thelateral line, more often on gill covers andventral fins. The club-shaped spots are highlyconcentrated on the caudal peduncle, occurringprogressively more sparingly forward, above thelateral line. Spots on the forward half of thebody are scarce, an unusual pattern among thecutthroat trouts. One peculiar feature of the RioGrande form of virginalis is the black-rimmedborder of the adipose fin (Behnke, 1967).Another peculiarity is the bluish tinge in thelining of the mouth and on the membrane underthe maxillary (Smith, 1984). The carmine-colored slashes in the gular folds on both sidesof the bottom jaw give the cutthroat trout itsname. In virginalis these markings can extend tocover the entirety of the lower jaw and gillcovers (Johnson, personal observation).

4

II. TAXONOMY

The first description, Salar virginalis by Girardin 1856, was based on specimens collected fromthe eastern end of the San Luis Valley during aPacific Railway survey in 1853. The typelocality was present-day Ute Creek (mis-namedUtah Creek by Girard), then the site of FortMassachusetts (later Fort Garland), in theTrinchera Creek drainage. In 1872 Copeexamined other specimens from nearby Sangrede Cristo Creek, also within the Trincheradrainage, and named them "Salmo spilurus."Apparently, he thought they differed from theUte Creek population in being more slender. Onthe basis of Girard's misnomer, Jordan in 1891mistakenly thought that the virginalis populationlay in Utah, and identified it with the presentlydescribed Bonneville cutthroat trout (O. c.utahi), while identifying spilurus with the RioGrande basin. Except for its having fewerscales, he considered the virginalis to be"wholly identical" to the Colorado Rivercutthroat (i. e. pleuriticus). Jordan and othersbelieved the virginalis to be derived from thegreenback cutthroats (O. c. stomias) found inand transferred from the Arkansas Riverdrainage to the north, later giving rise to thepleuriticus by yet another headwater transferinto the Colorado River basin, but yet the theremaining virginalis populations of theintervening Canadian River system more closelyresemble pleuriticus, while the virginalis sub-type from the more southerly Pecos drainageitself very closely resembles stomias. Thiswould argue against a direct prehistoric transferof stomias stocks southwards from the Arkansaswatershed. At any rate, stomias, pleuriticus, andvirginalis are a very closely linked geneticsubgroup, which is in turn affiliated mostclosely with the Yellowstone cutthroat, O. c.bouvieri (Shiozawa and Williams 1992).

III. DISTRIBUTION

Historic Distribution. The first writtendescription of a North American trout dates to1541, when Pedro de Castañeda de Najera, amember of Coronado's expedition, wrote of "alittle stream which abounds in excellent troutand otters". This was in all likelihood GlorietaCreek southeast of present-day Santa Fe- now abarren, ephemeral wash for most its length andin recent years holding a few brown trout withinthe enlarged boundary of Pecos NationalMonument (Trotter and Bisson 1988; Pittenger,personal communication 1997). Thedisappearance of virginalis from this stream,and such places as Tijeras Canyon near present-day Albuquerque and the Rio San Jose withintoday's Acoma Pueblo, typifies the loss of trouthabitat in the upper Rio Grande Basin over thepast two centuries.

Native virginalis populations occurred inmountain ranges in the extreme upper reaches ofthe Rio Grande system in present-day NewMexico and Colorado. Populations likelyinhabited the headwaters of the Canadian Riversystem, for instance in the upper Vermejo River.Due to a scarcity of first-hand accounts,however, a full picture of the historicdistribution of the Rio Grande cutthroat has notbeen definitively established. Stumpff andCooper (1996) believe the species formerlyoccurred in "all waters presently capable ofsupporting trout in the Rio Grande drainage"including the Chama, Jemez, and Rio San JoseRivers, along with the Bonito-Ruidosos-Hondoand other trout streams of the Pecos system,plus a number of the upper Canadian Riverdrainages. Sublette et al. (1990) concur.Stumpff and Cooper suggest it may haveinhabited 40 hydrologic sub-basins in Coloradoand New Mexico, and possibly streams innorthern Mexico as well (see figure two).

5

Figure Two. Current and Historical Range of theRio Grande cutthroat trout (excluding Texas).

6

The contention that virginalis once occupied allthe waters in the upper Rio Grande andCanadian River basins currently capable ofsupporting trout is sound, but notcomprehensive. The number of trout waters hasdiminished over the past century, indicating thatvirginalis likely occupied a range larger thanthat currently capable of supporting trout. Thislost habitat is considerable. A good portion ofthe Rio San Jose watershed in New Mexico'sCibola and McKinley counties, for example,once supported healthy populations ofvirginalis, including Cottonwood, Bluewater,Seboyeta and Seboyetita Creeks, the Rio SanJose, and Water and San Jose Canyons (vid.Behnke 1988). Only traces of trout habitatremain in this system today.

Colorado Historic Range. The Rio Grandecutthroat trout formerly occurred within allsouthern Colorado Rio Grande basin streamscapable of supporting trout (Stumpff andCooper 1996). This would have included theentirely of the mainstem and hundreds oftributaries.

Virginalis currently occupies tributaries to theCanadian River in southern Colorado andnorthern New Mexico. It is not clear, however,that the Canadian River was within the species’natural historic range, since the Canadian andRio Grande systems are not connected.

New Mexico Historic Range. The historic NewMexico range of the Rio Grande cutthroatextended over a very broad area, with a greatconcentration in northern New Mexico wheremost populations could communicate directlywith the Rio Grande, which in Coronado's timeretained its coldwater nature perhaps as farsouth as present-day Albuquerque or Belen.Some populations, such as those of the RioHondo system draining from the SacramentoMountains into the Pecos River, were possibly

isolated from their ancestral parent stream bywarm summer water temperatures, but stillsurvived in complex watersheds of third orfourth order, immune to the effects of droughtor fire.

Ranges in New Mexico which likely held stablepopulations of virginalis include the San Mateoand Zuni Ranges of west central New Mexico,the Manzano and Sandia Ranges east and southof present-day Albuquerque, the SacramentoMountains of south central New Mexico, andthe Black Range of southwestern part of thestate. The only river system outside the RioGrande drainage in which virginalis appears tohave occurred is the Canadian River."Wagonloads" of adult virginalis were reportedto have been taken by anglers from theheadwaters of the Canadian River (a tributary ofthe Mississippi rather than the Rio Grande) inthe nineteenth century (Behnke 1967).

Mexico Historic Range. One early report byCope in 1886 of a trout from southernChihuahua, Mexico, near the borders of Sinaloaand Durango, may well have been a Mexicangolden trout (Oncorhynchus chrysogaster),known to occur west of Mexico's ContinentalDivide. The Rio Conchas forms the majorstream system east of this divide draining intothe Rio Grande from the south, from which nonative trout specimens or sightings of any kindhave ever been reported. The Rio Grande has noother major or even permanent Mexicantributary upstream from the Conchas, while thestreams draining the interior of Mexico east ofthe Continental Divide and north of the RioConchas drain into closed desert basins. One ofthese stream systems, the Rio Casas Grandes,does contain a native trout in its extremeheadwaters near the Continental Divide, the so-called Yaqui trout (Oncorhynchus sp.,undescribed), but it is not in the cutthroat group,

7

and most likely traces to a headwater, trans-divide transfer from the Rio Yaqui system.

No Rio Grande tributaries downstream from theRio Conchas above Nuevo Lauredo areextensive enough to reach into the higherelevation mountains to the south which might becapable of supporting trout. The one exceptionto this pattern is the watershed of the 9000-ftSierra del Carmen, south of Big Bend NationalPark in Texas, a high range which nearlyadjoins the Rio Grande but from which trouthave never been reported. The mountains of theSierra Madre Oriental south of Nuevo Lauredoare isolated from the western mountains byextensive desert, communicating more closelywith the warmwater Gulf of Mexico, and havenever shown records of trout.

Texas Historic Range. Though there is a lackof convincing evidence of the occurrence ofvirginalis in Mexico, there have been a numberof historic reports of trout found in streamsdraining into the Rio Grande from the north, inTexas. J. W. Daniel in 1878 wrote that he hadcaught "speckled trout" in the Devils River inVal Verde County, Texas during the Civil War,and other trout from the nearby San FelipeSprings, and, earlier, from the Limpia Riverdraining the 8,400-ft. Davis Mountains. Thisrecollection was reinforced in 1878, by N. A.Taylor, who conferred with a formerConfederate surgeon stationed at Fort Davis andwrote that "brook trout" occurred in the LimpiaRiver in Texas, a "clear, cool, sparkling streamflowing through a region about 5,000 feetelevation," and also in unnamed streams to thenorth (Behnke 1967). According to Garrett andMatlock, these reports are bolstered by thepresent-day occurrence of Rio Grande chubs inthe Little Aguja Creek, a tributary of the LimpiaRiver. This species is closely associated withpopulations of virginalis in New Mexico and

Colorado, and is today found nowhere else inTexas (vid. Behnke 1992).

The only self-reproducing population of troutremaining in Texas lives near the New Mexicoborder in McKittrick Canyon within theGuadalupe Mountains National Park. These arerainbow trout, probably stemming fromintroductions made in the 1930s. Current Distribution. Today there are some520 streams in New Mexico and 230 streams inColorado feeding the Rio Grande still capable ofsupporting trout populations. These watersamount to perhaps 4,500-5,000 stream miles(see figure 3). Nearly all are concentrated inextreme northern New Mexico, in the Jemez,Tusas, and New Mexico's portion of the Culebraand Sangre de Cristo Ranges, and in theCulebra, San Juan La Garita, and CochitopaMountains of southwestern Colorado.

This figure was higher within the historic past,for many streams in the lower and middleelevations, particularly in New Mexico, havebecome unsuitable for trout within the past 150years, causing cutthroat to disappeared fromentire mountain ranges. It is virtually certain thevirginalis occupied streams and streamsegments which are not currently trout waters,and that this lost habitat is considerable. Nearlythe entire Rio San Jose watershed in NewMexico's Cibola and McKinley counties oncesupported healthy populations of virginalis,including Cottonwood, Bluewater, Seboyeta andSeboyetita Creeks, the Rio San Jose, and Waterand San Jose Canyons (vid. Behnke 1988).

Among the still substantial number of availabletrout habitats within its former range, there are92 known populations of the virginalissubspecies, 53 in New Mexico and 39 in

8

TABLE 2. Total Rio Grande cutthroat trout habitat currently available and currently occupied on selectedfederal land in New Mexico (from Stumpff and Cooper 1996, percentage columns corrected).

AGENCY TOTALAVAILABLE

STREAM LAKEMILES ACRES

TOTAL OCCUPIED

STREAM LAKEMILES ACRES

PERCENTOCCUPIED

STREAM LAKEMILES ACRES

BLM 247 5000 0 0 0 0Carson National Forest 576 81 126 0 21.9 0Santa Fe National Forest 1100 429 83 0 7.5 0Cibola National Forest 0 0 0 0

Lincoln National Forest 90 740 0 0 0 0Gila National Forest 330 70 3 0 0.9 0TOTAL 2343 6320 212 0 9 0

Figure 3. Current Streams Capable ofSupporting Rio Grande Cutthroat Trout in NewMexico and Colorado, Current Populations ofRio Grande Cutthroat.

9

Colorado (Stumpff and Cooper 1996). All thesepopulations are within the Rio Grande, Pecos,and Canadian River systems. Only threepopulations occur south of Santa Fe. Mostpopulations are in isolated segments of small,first or second order headwater streams whosetotal length is perhaps 212 miles in New Mexico(Stumpff and Cooper 1996).

Oncorhynchus clarki virginalis is now absentfrom up to 95% of its historic range (Stumpffand Cooper 1996).

IV. NATURAL HISTORY

Typical of all members of the cutthroat series,virginalis lives in clean, cold mountain streams,preferably of moderate (6 % or less) gradient. Itis insectivorous, feeding on the invertebrate driftwhich is also the typical diet of streamcutthroats found across the American West.Virginalis in the Rio Chiquito in CarsonNational forest were found to feed in June onmidge (Diptera) larvae, caddisflies (Tricoptera),and mayflies (Ephemeroptera), according to aNew Mexico Game and Fish D. J. projectperformed in the 1960s (F-22-R-788, February,1968). Smaller individuals will eat zooplankton;it has been observed in another cutthroatsubspecies, O. c. henshawi, that trout smallerthan 6 centimeters prefer Daphnia pulex, aminute freshwater crustacean, while largerindividuals focus on the typical benthicinvertebrates (Luecke 1986). Terrestrial insectsare also consumed during summer months,while other freshwater crustaceans are alsosupplemental. (vid. Sublette et al. 1990). Thefood chain upon which virginalis depends isgenerally allochthonous (Sublette et al. 1990)--that is to say, its ultimate energy source comesfrom outside the stream itself, generally fromdeciduous leaves and leaf litter which fall into

the stream. Thus, riparian vegetation plays avital role in providing food for the fish.

Also typical of other stream cutthroats,virginalis spawns in the spring during snowmelt(from April to July, depending upon latitude andelevation). Some populations or individualfemales spawn every other year. This patternhas been thought by some observers to putvirginalis at a disadvantage to two of itscompetitors, the brook char and the brown trout,which spawn in the fall when water flows arestable (Hubbard 1976). Virginalis lives up tonine years, more often six, maturing sexually atfour years, as compared to three years for exotictrout species (McClane 1963). This delay inspawning age and consequent short spawninglife is also felt to put virginalis at a disadvantageby retarding its natural rate of populationincrease (Hubbard 1976). Egg production byfemales depends on their size and varies from200 to 4000.

Virginalis evolved with minnow species inmiddle to lower stream elevations --Rio Grandechub and long-nose dace in the Rio Grande,Canadian, and Pecos River drainages, plus thewhite sucker and creek chub in the Canadianand Pecos River drainages.

Predators included garter snakes, great blue andother herons, otters, and raccoons (Rinne 1995).There is a tendency to piscivorousness as thesize of the fish increases (McAffe 1966; Baxterand Simon 1970).

V. HABITAT REQUIREMENTS

Virginalis typically requires high oxygencontent in its stream habitat, low summer watertemperatures, and clean gravel for its spawningbeds. It requires riffle areas for food productionand habitat for young, and pools for

10

overwintering, and summer rest, and the numberof pools and riffles should be roughly equal.Vegetation in the riparian zone needs to beabundant enough to provide shade and cover(Propst and McInnis 1975). In the headwaterstreams, overwintering habitat must includepools of sufficient size, and enough thermalcover to prevent sheet ice in hypercooled water.

In the minds of anglers and biologists alike, thecutthroat trout is synonymous with wilderness,and for good reason. In the Southwest thisassociation is particularly noticeable; there is adefinite, positive correlation between roadlessareas and virginalis populations in New Mexicoand Colorado (see below, "Logging and RoadBuilding"). Roadless surroundings appear to bea necessary but not sufficient condition forsurvival of healthy stream populations of thesefish.

VI. RANGE REDUCTION

New Mexico Reduction by Sub-Basin.Stumpff and Cooper (1996) identified 40hydrologic sub-basins within the historicdistribution of the virginalis (see Table 1).Virginalis is currently absent from 57.5% (= 23)of the historic sub-basins. The actual sub-basindecline is likely greater, however, since Stumpffand Cooper did not list the Limpia River systemin Texas as a historically occupied sub-basin.

That the Rio Grande cutthroat trout has beenextirpated from over 23 entire sub-basins isalarming. Sub-basin analysis, however, is acorse measure of range reduction. Finer scaleswhich detect additional range loss within sub-basins is even more disturbing.

Reduction By Stream Miles and Lake Acreson Selected New Mexico Federal Lands.Stumpff and Cooper (1996) identified 2,343miles of stream and 6,320 lake acres on selectedfederal lands in New Mexico as being currentlysuitable for virginalis occupation (see Table 2).Only 9% of this area is currently occupied,including 0 lake acres, 0 BLM stream miles, 0stream miles on the Lincoln and Cibola NationalForests, and only 3 stream miles on the GilaNational Forest. Of all occupied habitat on theseselected federal lands, 98.6% is in northern NewMexico, 0.9% is in central New Mexico, and1.4% is in southern New Mexico.

By comparing stream miles currently occupiedby virginalis on selected federal lands, withtotal available trout stream miles on those samelands, Stumpff and Cooper conclude that theRio Grande cutthroat’s range has declined by91%, not only on these lands, but throughout itsrange in New Mexico. They assume, therefore,that the decline is representative of other landsin New Mexico. The actual decline in NewMexico, however, is likely higher for severalreasons:

1. The calculations by Stumpff and Cooper arebased on trout stream mile figures in NewMexico which are only readily available forthe national forests (vid. Tripp andRockland 1988). Their calculation,however, does not include the CibolaNational Forest, National Park Serviceunits, state land, private land, NativeAmerican Nations, or private inholdingswithin the National Forests themselves.

11

TABLE 1. Historic distribution and current status of the Rio Grande cutthroat trout by USGS hydrologicsub-basin (from Stumpff and Cooper 1996).

SUB-BASIN USGSHYDRO.UNIT #

OWNERSHIP OCCUPIEDSTREAMS

OCCUPIEDLAKES

STATUS

COLORADORio Grande Headwaters 13010001 Rio Grande NF 2 0 At Risk, StableAlamosa-Trinchea 13010002 Rio Grande NF 4 1 Secure, StableRio Grande Headwaters 13010003 Rio Grande NF 2 1 At Risk, StableRio Grande Headwaters 13010004 Rio Grande NF 6 0 At Risk, StableConejos 13010005 Rio Grande NF 4 0 At Risk, StableRio Chama 13020102 Rio Grande NF 1 0 At Risk, StableUpper Rio Grande 13020101 Rio Grande NF 0 0 ExtirpatedUpper Canadian 11080001 Rio Grande NF 0 0 Extirpated

NEW MEXICO

Upper Canadian 11080001 Carson NF 1 0 At Risk,Declining

Cimarron 11080002 Carson NF 1 0 At Risk,Declining

Mora 11080004 Carson NF 3 0 At Risk,Declining

Conejos 13010005 Carson NF 4 0 At Risk, StableUpper Rio Grande 13020101 Carson NF 24 0 At Risk, StableRio Chama 13020102 Carson/Santa Fe NF 9 0 At Risk, StableJemez 13020202 Santa Fe NF 3 0 At Risk, StableRio Puerco 13020204 Santa Fe/Cibola NF 3 0 At Risk, StablePecos Headwaters 13050001 Santa Fe NF 9 0 At Risk, StableRio Grande-Caballo 13030101 Gila NF 1 0 At Risk, StableTularosa Valley 13050003 Lincoln NF 1 0 UnknownRio Grande-Albuquerque 13020205 Cibola NF 0 0 ExtirpatedArroyo Chico 13020205 Cibola NF 0 0 ExtirpatedRio San Jose 13020207 Cibola NF 0 0 ExtirpatedRio Salado 13020209 Cibola NF 0 0 ExtirpatedRio Grande-Elephant Butte 13020211 NA 0 0 ExtirpatedEl Paso-Las Cruces 13030102 Gila NF 0 0 ExtirpatedJornada Draw 13030103 NA 0 ExtirpatedRio Grande-West Estancia 13050001 Cibola NF 0 0 ExtirpatedRio Grande-Sacramento 13050004 Lincoln NF 0 0 ExtirpatedUpper Pecos 13060003 NA 0 0 ExtirpatedAlamosa Creek 13060004 NA 0 0 Extirpated

12

TABLE 2. Total Rio Grande cutthroat trout habitat currently available and currently occupied on selectedfederal land in New Mexico (from Stumpff and Cooper 1996, percentage columns corrected).

AGENCY TOTALAVAILABLE

STREAM LAKEMILES ACRES

TOTAL OCCUPIED

STREAM LAKEMILES ACRES

PERCENTOCCUPIED

STREAM LAKEMILES ACRES

BLM 247 5000 0 0 0 0Carson National Forest 576 81 126 0 21.9 0Santa Fe National Forest 1100 429 83 0 7.5 0Cibola National Forest 0 0 0 0

Lincoln National Forest 90 740 0 0 0 0Gila National Forest 330 70 3 0 0.9 0TOTAL 2343 6320 212 0 9 0

SUB-BASIN USGSHYDRO.UNIT #

OWNERSHIP OCCUPIEDSTREAMS

OCCUPIEDLAKES

STATUS

Pecos-Artesia 13060007 NA 0 0 ExtirpatedRio Hondo 13060008 Lincoln NF 0 0 ExtirpatedPecos-Rio Felix 13060009 NA 0 0 ExtirpatedRio Penasco 13060010 Lincoln NF 0 0 ExtirpatedUpper Pecos-Black River 13060011 Lincoln NF 0 0 ExtirpatedOcate 11080003 Carson NF 0 0 ExtirpatedConchas 11080005 NA 0 0 ExtirpatedTucumcari 11080006 NA 0 0 ExtirpatedUte 11080007 NA 0 0 ExtirpatedPlaza Largo 11080008 NA 0 0 Extirpated

TOTAL 78 2

13

There is considerable trout habitat onprivate lands, especially in northern NewMexico and in the Sacramento Mountains.For example, there are 20 to 30 miles ofprivate trout habitat on the Rio Peñasco.There is a good deal of trout habitat onNative American Nations as well. By ourcalculations (see attached map), there areabout 600 miles of available trout habitat inNew Mexico and Colorado outside thefederal lands tallied by Stumpff and Cooper.Yet these lands are known to support only 7virginalis populations, averaging aboutthree miles stream miles, for a total of 20 to25 miles. The decline rate on these lands isabout 95%, therefore, exceeding the 9% ratecalculated by Stumpff and Cooper.

2. Stumpff and Cooper rely on conservativeForest Service records of available troutstreams. These were based on surveysconducted with sole emphasis on recreationand sport fisheries. As a result, manysmaller streams not considered to haverecreation potential but nonethelesscontaining stable populations of wild troutwere omitted from the list. Accordingly, themileages contained in the Gila and LincolnNational Forests, to cite only two examples,only show roughly two thirds of the streammiles actually occupied by trout (vid. Trippand Rockland 1988). As a result, thecalculations overlook numbers and miles ofstreams currently supporting trout evenwithin the selected federal lands in table 2.While Stumpff and Cooper over-report troutstream miles of historical virginalis habitaton the Gila National Forest by includingstreams west of the Continental Divide, theyoverlook perhaps 500 additional miles oftrout stream habitat on the federal lands (seefigure 3). This mileage omission inflatestheir percentage figure, and as a result theyover-estimate the percentage of current trout

stream miles currently occupied byvirginalis.

3. The equation of historic range with currentavailable misses a significant percentage ofhistorical trout habitat that is no longercapable supporting either virginalis or anyother trout form. For instance, only nine orso miles of the entire Rio San Josewatershed are currently capable ofsupporting trout- much it marginally. Theremainder of this watershed and other lostwaters account for perhaps a 20 to 30percent increase over currently availabletrout stream miles, possibly even more.Indeed, under natural conditions, nearlyevery permanent stream of 6,000 feetelevation in the mountain ranges of NewMexico feeding the Rio Grande should havebeen capable of supporting virginalis. Agreat number of these streams are nowephemeral. A better index would be streammiles currently occupied by virginalisversus historical permanent stream miles of5,500 feet elevation and above withinappropriate river basins (including the RioLimpia in Texas). Such an accountingwould reveal declines of well over 91%percent.

4. The stream mileage figure itself ismisleading, for one mile of the upper RioGrande, Vermejo, or Pecos Rivers, all lostto virginalis for many decades, wouldsupport as many trout as perhaps 20 miles ofany of the tiny headwater tributaries wherevirginalis still survives. Even streams ofintermediate size, such as the Rio Ruidosonear Cloudcroft, are capable of producingthousands of trout per mile of stream, ascompared to 100-300 per mile in the typicalheadwater tributaries still holding virginalis(vid. Fish Population Estimates For theSanta Fe National Forest). Some of the first-

14

order streams hold even fewer virginalis.For instance, the entirety of Peralta Creek,roughly three miles of trout habitat and oneof the 53 remaining cutthroat populationsalluded to by Stumpff and Cooper, wasreported to contain just 50 adult virginalis in1973 (New Mexico Game and Fish 1973;Stork and Behnke 1975).

5. There is a distinct possibility that the qualityand thus productivity of most trout habitatshas declined significantly in historic times.Numbers of virginalis probably have beenimpacted by this lessening quality as well asby the accounted-for lessened numbers ofstream miles. Indeed, it is clear that none ofthe virginalis habitats available today couldprovide for the "wagonloads" of capturedtrout referenced by Behnke (1967).

6. A number of the somewhat larger streamsstill containing virginalis also holdcompeting exotic salmonids, meaning thatthe virginalis only occupy a minor portionof the habitat currently available within thestreams themselves. For example, in the RioChiquito in 1968, the New MexicoDepartment of Game and Fish found browntrout to comprise 78 percent of the trout inpopulation samples, while virginalis madeup only 22 percent (New Mexico Game andFish 1973). Propst's study of the remainingvirginalis habitats in the Canadian Riverheadwaters (these are among thosereferenced by Stumpff and Cooper) foundvirginalis outnumbered by brown trout byratios as high as 10 to 1 (Propst 1977).Studies reported by Little and McKurdy in1968 found brown trout to outnumbervirginalis by a ratio of more than 9 to 1 inRito de la Olla, another of the 53 virginalishabitats referenced by Stumpff and Cooper.In effect, virginalis has lost the use ofconsiderable portions of its original habitat

within these streams, which were formerlyfully occupied by the native subspecies.

Reductions in Colorado. Stumpff and Cooper(1996) did to not prepare an equivalent analysisfor Colorado. Rinne (1995), however, statesthat virginalis might have declined to 5-7percent of its total historical range, inclusive ofColorado.

Harig and Fausch (1996) documented writtenaccounts of 68 stream populations and 25 lakepopulation from historic and recent records (seetable 3). Of the 68 stream populations, 18% (2)have been extirpated, 43% (29) are insecure,18% (12) are secure1, and one is of unknownstatus. Of the 25 lake populations, all areintroduced, 32% (8) have been extirpated, 8%(2) are insecure, 60% (15) are unknown, andnone are known to be secure. In all, only 13%(12) all historic and recent populations areknown to be secure.

VII. HISTORIC POPULATIONDECLINE AND CURRENTINSTABILITY

Historic Population Decline. O. c. virginalisprobably existed in the time of Coronado as arelatively small number of distinct and isolatedintra-breeding population groups, each ofconsiderable size. It is likely that nearly all ofthe virginalis stocks from the Rio Grandewatershed, at least from the Jemez River mouthto the extreme headwaters in Colorado, werecapable of genetic exchange, making one great

1Secure is defined by the petitioners as beingfree of exotics, protected to some degree by natural orartificial barriers, and demographically “stable” asdefined by Harig and Fausch (1996).

15

TABLE 3. STATUS OF 68 RECORDED STREAM AND LAKE POPULATIONS OF RIOGRANDE CUTTHROAT TROUT IN COLORADO. from Harig and Fausch 1969 (69th

stream unaccounted for).

POPULATIONS DOCUMENTED SINCE 1990 (25 STREAMS)

STABLE UNSTABLE FISH BARRIERPRESENT

EXOTICTROUTPRESENT

STABLE,BARRIER, NOEXOTICS

15 (60%) 10 (40%) Most 11 (44%) 10 (40%)

POPULATIONS NOT DOCUMENTED SINCE 1990 (28 STREAMS)

PREVIOUSLYSTABLE, STILLPRESENT

PREVIOUSLYUNSTABLE,UNKNOWN TODAY

PREVIOUSLY HADEXOTICS

HABITAT DEGRADED

5 (17%) 22 (79%) 17 (61%)including 3/5stable populations

11 (40%)

EXTIRPATED IN LAST 20 YEARS (10 STREAMS)

NATIVEPOPULATION

INTRODUCEDPOPULATION

CAUSE: INVADEDBY EXOTIC TROUT

CAUSE: HABITATDEGRADATION

8 2 7 3

“OTHER” STREAMS (5)

UNSTABLE UNKNOWN EXTIRPATED

Archuleta- Stocked in 1970South Fork Saguache Creek- present in 1984,since stocked with Colorado River cutthroat

Ricardo- CanadianRiver basin

Rio Grande #2- found 1889,extirpated prior to 1970.

La Jara Creek- naturalrecolonization attempt in 1977 fromTorsido Creek; reclaimed, stockedwith brook trout in 1996.

25 LAKES: ALL STOCKED, DOCUMENTED SINCE 1990

STILL EXISTING STABLE OR REPRODUCING UNSTABLE

17 likely Unknown Haypress Lake: yesLower Dome: likely

genetically fluid population. Populations in the Canadian River system likely formed a second

16

group, the upper Pecos Region another, perhapsincluding the Rio Hondo system. The RioPuerco and possibly the various isolatedmountain ranges in southern New Mexico andfrom the Davis Mountains in Texas, may wellhave held separate and recently isolatedpopulations.

In New Mexico, roughly 50 of the 53 remainingvirginalis populations are physically isolatedand cannot recruit new members or interbreedwith other populations (Stumpff, personalcommunication 1997). The situation inColorado virtually the same. Thus, the broadhistoric populations of virginalis have beenreduced in number over the past two centuries,then split into 80 or so discrete, geneticallyisolated groups, each containing a few hundredto at most a few thousand individuals (Stumpff,personal communication 1997). This isolation,a result of man-caused habitat destruction andintroductions of exotic salmonids, is a profoundthreat to the integrity and continued existenceof the subspecies. Prior to such fragmentation,the virginalis would have exhibited meta-population structure which is no longerrecoverable, for undoubtedly the remainingstocks cannot account for all the originalgenetic variability of the subspecies. It isirrefutable, however, that the sharp decline inthe range of virginalis was accompanied bydisappearances of original populations as wellas tremendous declines in the numbers ofindividual fish.

Current Population Instability. The U.S.Forest Service conservation assessment for theRio Grande cutthroat trout (Stumpff andCooper 1996) is based on responses to a

standardized survey sent to wildlife biologists(see Appendix A in Duff 1996). The surveyused the status definitions described in Table 4.

Based on the surveys, Stumpff and Cooper(1996) reported 92 extant populations, of which4 are secure-expanding, 29 are secure and/orstable, 8 are declining and 51 are unknown.Stable populations are not necessarily secure,they have simply not experienced recentdeclines. Stumpff and Cooper (1996) report thatmost populations are at risk, primarily due topresence of non-native trout.

Three of the 17 currently occupied sub-basinsare at risk-declining, twelve are at risk-stable,one is stable-secure, and one is unknown (seeTable 5). Of the occupied sub-basins then,88.2% are at risk, 5.8% are stable-secure and5.8 percent are unknown

Genetic isolation. Rinne (1995) states thatsince most stocks are isolated in headwaters,gene flow among populations is "virtually non-existent." Cutthroat trout do occasionally movedownstream from headwater tributaries intolarger streams, then ascend other tributaries(Young 1996). This shows how genes might beexchanged in montane stream systems, andunderscores the importance of connectingdiscrete populations. As mentioned above,virginalis now exists as roughly 80 small,discrete populations. The possibility of geneticdrift and inbreeding must be assumed until theminimum safe number of individuals forpopulations of virginalis has been determined.

Of the 92 existing populations, only 69 arenative remnants.

17

Figure 4. Sub-basin map

18

Status Colorado

Sub-Basins Populations # % # %

New Mexico

Sub-Basins Populations# % # %

Secure-expanding

3 38 4 19 0 0 0 0

Secure-stable 3 38 5 24 4 36 6 11

At risk-stable 1 13 4 19 7 64 16 30

At risk-declining

3 38 5 24 3 27 9 17

Unknown 3 38 4 19 6 54 19 35

TOTAL 13 22 20 50

Exotic speciesPresent

8 100 13 62 10 91 44 82

Table 4. Population Stability Definitions Used By Stumpff and Cooper (1996)

population- distinct aquatic system or interconnected drainage basin with no barriers to genetic interchange

stable- abundance or distribution has remained relatively constant over the last 10 years

secure-expanding- adults relatively abundant; recruitment occurring; abundance and/or distribution in upwardtrend; no major threats to habitat quality; no threats from competition or hybridization withexotic fish.

secure-stable- adults relatively common; recruitment occurring; most recent abundance and distributiontrends stable; no major threats to habitat quality; no threats from competition or hybridizationwith exotic fish.

at risk-stable- adults relatively common; recruitment occurring; most recent abundance and distributiontrends stable; major threats to habitat quality or from competition or hybridization with exoticfish.

at risk-declining- adults are relatively rare; recruitment not occurring; most recent abundance and/or distributiontrends are downward; major threats to habitat quality or from competition or hybridizationwith exotic fish.

unknown- information insufficient to classify

TABLE 5. Sub-basin and Population Status (from table 3 & 4).

19

VIII. HABITAT FRAGMENTATION

A fragmented landscape is one in which habitat"islands" are separated from one another bymarginal or unsuitable habitat. Fragmentationmay be caused by biogeographical influences(in the case of virginalis, dry stream stretches,waterfalls, etc...), natural disturbances(unusually high natural flooding, landslides,etc...), or by human intrusion (livestockgrazing, non-native species introduction, roadbuilding, logging, etc.). Habitat fragmentationin the Rio Grande basin has been greatlyexacerbated through human intrusion.

The very small size of current virginalispopulations, limited genetic diversity, and theircomplete isolation from other populations,increases the chances of local extirpation andtotal extinction of the subspecies (see forexample, Propst et al. 1992). Human inducedevents, as well as stochastic processes such asfloods and fires, unseasonal weatherfluctuations, and disease, could easily decimateany single or many remaining. Anthropogenicproblems such as hybridization with introducednon-native trout species also threatenpopulations with extinction.

If populations were larger, more numerous, andconnected to one another, a proportionallylarger number of populations would beexpected to survive stochastic and systematicthreats (though not with hybridization), and beable to recolonize an extirpated or reducedpopulation (see for example, Propst et al.1992). Because of the presence of rainbow andbrown trout in nearly all waters between currentRio Grande cutthroat populations, the latter areno longer able to establish new populations inlower elevations outside of their currentlyoccupied isolated tributaries due tohybridization. Simply put, all virginalis washeddown from the isolated tributaries potentially

breed with the non-native trouts, resulting in theloss of the pure strain and the creation ahybridized population.

The total population of a species is oftencomprised of metapopulations maintained by adynamic balance of local extinction andcolonization (Wilson 1992). By analyzingpopulation dynamics in terms ofmetapopulation interaction, Levins (1969,1970) has shown why species may not occupyall available suitable habitat, and whypopulations may become extinct even thoughsuitable habitat patches are available. Assumingthat patches of suitable habitat are randomly orevenly distributed across a large region, but areseparated from one another by unsuitable areas,Lande's model (1987) predicts the equilibriumoccupancy of suitable habitat by females as afunction of the proportion of the landscapewhich is suitable, and the demographicpotential of the species (i.e. life history anddispersal behavior).

Lande (1988) has pointed out several problemswith his model which are applicable to O. g.gilae. The model "depends strongly" on theassumption that suitable habitat patches arerandomly or evenly distributed across a region,that suitable habitat is at equilibrium, that initialpopulations are large enough to be immune toextinction due to demographic or environmentalstochasticity, and that there is no loss of fitnessdue to inbreeding depression. Lamberson et al.(1992) have extended Lande's analysis bydeveloping a model which accounts forenvironmental stochasticity. To predict long-term population dynamics of northern spottedowls in fragmented landscapes, they determinedthat populations tend towards stableequilibrium when initial populations aresufficiently large and their search efficiency isrelatively high. "If search efficiency was low,however, even very large initial populations

20

crashed." This is known as the Allee effect: aspopulations become smaller and more scattered,or as the habitat becomes more fragmented,dispersing males become less successful atfinding suitable territories and females becomeless successful at finding potential mates.

Patterson (1978) has noted that the non-randomprocess of extinction inevitably leads to theextirpation of rare and local species from smallhabitat "islands". Small habitat patches in thecase of birds and mammals, or severely isolatedpopulations in the case of fish, even in greatnumbers, do not serve the biological needs metby larger patches or connected populations.Specifically, isolation induced impacts to fishspecies such as the Rio Grande cutthroat, aremany times worse due to lack of ability todisperse, and include such impacts asinbreeding depression. Species in small habitatfragments go extinct in a very predictablesequence (Cutler 1991, Patterson 1978). WhilePatterson is concerned with the sequence ratherthan the causes of extinction, it is clear thatsmall populations have a very limited geneticpool from which to draw upon in the face ofchanging environmental conditions. They arealso more likely to be wiped out in a singlestroke, leaving no regenerative base.

In Minimum Viable Populations: Coping withUncertainty, Shaffer (1987) agrees:

“...that extinction may often be the result of chanceevents, and that the likelihood of extinction mayincrease dramatically as population size diminishes.”

But Shaffer goes on to point out that predictablehuman intrusions greatly increased the chanceof extinction. It is probable that the most severethreat to virginalis within the foreseeable futureare the highly predictable human inducedprocesses of extirpation and extinction whichare complementary to stochastic processes.Recent models have shown that extinctionoccurs much faster than genetic/stochastic

models predict due to the addition of non-stochastic causes (Lamberson et al. 1992, Pimm1993) such as grazing.

Were the separate populations of virginalisconnected to one another by segments ofsuitable habitat, local extinction due tostochastic and more systematic processes couldbe overcome by re-colonization fromunaffected sites within the same watershed ordrainage. Isolated populations are alsogenetically separated, narrowing the geneticvariance necessary to enable dynamicresponsiveness to environmental changes. Ifhuman induced extinction processes wereeliminated, and populations were allowed torecover within an entire watershed, it is likelythat there would be ample genetic variation toeliminate or greatly reduce the risk of disease orinbreeding, and greatly reduce the possibility ofchance extinction or extirpation of the entiresub-population. But again, those geneticvariants must be given a chance at full recoverywithin separate headwater stream watershedsthrough reclamation, and elimination ofdestructive habitat alteration activities.

By not addressing the issue of protection forentire watersheds in the protection orreclamation of virginalis habitat, landmanagement agencies such as the U.S. ForestService have artificially perpetuated theisolation of the subspecies. Replication ofremaining indigenous populations is a validfirst step towards recovery, but by noteliminating destructive habitat impacts such asgrazing of livestock and fire suppression, andby only reintroducing populations to tiny,isolated stream stretches, management agencieshave accomplished relatively little towardsfacilitating the recovery of the subspecies.

IX. HABITAT LOSS- LIVESTOCK

21

GRAZING

Habitat loss has been and continues to be theprimary threat to the Rio Grande cutthroat trout(Stumpff and Cooper 1996). Even non-habitatrelated threats such as hybridization,competition and predation are exacerbated byhabitat degradation which favors exotic,competing and predatory species overvirginalis. Livestock grazing is the primaryfactor degrading virginalis habitat (Stumpff andCooper 1996).

Destruction of Trout Populations. Thetremendous negative effect of cattle on troutand trout habitat has been documented indozens of scientific studies.

• New Mexico Game and Fish (1974)determined that cattle grazing wassuppressing Rio Grande cutthroatpopulations on Cañones and East PolvaderaCreeks.

• New Mexico Game and Fish issued a reportentitled “Status and Future of the Rio Grandecutthroat trout” (Hubbard 1976). Itconcluded:

“Since Spanish days, the range and numbers of thistrout have probably been declining in New Mexico,as degradation of habitat (e.g. through overgrazing)and use and misuse of water and later theintroduction of exotics...streams with limitedriparian growth and erosional problems (e.g.degraded banks, rapid run-off form watershed orstreambed) tend to lack proper shade (and therebyhave higher temperatures), sufficient feeding andshelter areas, and a suitable diversity of invertebratelife. Such streams harbor few or no native trout,whereas streams without these did sustain fish.”

• Behnke and Zarn (1976), Sublette et al.(1990), and Behnke (1992) concluded thatlivestock grazing on National Forests andother lands was harming Rio Grande

cutthroat trout populations.

• Sublette et al. (1990) concluded:“Most streams occupied by Rio Grande cutthroattrout have been impacted by overgrazing andlivestock. Limited vegetation in the watershed,especially in riparian areas, has led to alteredstream nutrient and sediment loads, and hasmodified flow regimes along with the morphologyof the stream course. Trampling of streambanks bylivestock has further accelerated habitatdestruction. Trout survival in many of these streamsis impaired because of the lack of productive riffleareas, suitable spawning sites, undercut banks (toescape predation), pools (for resting, feeding andoverwintering), and shade (in proper proportionswhich reserves cold water temperatures yet allowsadequate solar gain essential for primaryproduction).”

• Parson and Wilson (1991) determined thatApache trout were ten times more abundanton ungrazed streams on the Apache-Sitgreaves National Forest and other areas inthe White Mountains, AZ than on grazedstreams.

• Rinne and LaFayette (1991) found thatungrazed streams on the Tonto and Santa FedNational Forests had twice as many trout,trout populations, and trout biomass thangrazed streams.

• Propst and McInnis (1975) found that SantaFe National Forest streams with little riparianhabitat and erosion problems, such asdegraded banks or sign of rapid run-off,sustained few or no cutthroat trout. Streamswithout this deficiency.

• Platts (1991) reviewed 21 studies, findingonly one that did not concluded that cattledegrade trout populations and habitat.

Figure 5. Grazing Photo

22

• Chaney et al. (1990) report that degradedcutthroat spawning habitat in MahoganyCreek, ID recovered when cattle wereremoved from the riparian area.

• Chaney et al. (1990) report that populations ofcutthroat trout in Huff Creek, Wyomingincreased from 36 per mile to 444 per milewhen cattle were excluded from the streamarea, as a result of better in-stream coverlower water temperature, and decreasedsedimentation.

• Chaney et al. (1990) also found that cattleexclusion from the riparian zone of BearCreek in Oregon converted an ephemeralreach of the stream into a permanent flowsupporting a wild trout population.

• Twenty years of cattle exclosure on CampCreek in central Oregon turned an ephemeralwash into permanent stream capable ofsupporting redband trout (Hunter 1991).

• Behnke and Zarn (1976) concluded thatwestern trout streams cannot be recovered oreven stabilized unless livestock impacts aregreatly reduced.

• Armour (1977), Marcuson (1977), Crispin(1981), Kennedy (1977), and Duff (1979)found increases in trout population andindividual trout size when cattle are fencedout of riparian areas.

Habitat degradation facilitates competitionpressure by providing the conditions in whichbrook and brown trout can dominate the RioGrande cutthroat. Griffith (1988) states that inoptimal habitat, native cutthroat may be able towithstand competition from exotic trout,leading Stefferud (1988) to conclude that“habitat protection and enhancement may beextremely powerful tools in the futuremaintenance of Rio Grande cutthroat troutstocks.” Rio Grande cutthroats have dominatedbrown trout in remote, coldwater streams such

as the Rio Chiquito and Upper Trinchera Creek(USFWS 1973). Logging, road building andlivestock grazing, by increasing siltation andraising water temperature, will favor browntrout over virginalis (USFWS 1973).

Destruction of Stream Morphology andVegetation. The mechanics of how domesticlivestock harm western trout are described byClarkson and Wilson (1991):

“A majority of the ungulate damage to stream banks isundoubtedly caused by the domestic cow. Although wecannot unequivocally assure that our models explain acause-and-effect relationship between cattle use andtrout standing crops, the literature abounds withtreatises that document the debilitating effects oflivestock use on bank morphology and troutpopulations (see reviews by Szaro 1989, Marcus et al.1990, Chaney et al. 1990, Armour et al. 1990, Platts1991). The avenues of impacts include (Platts andRaleigh 1984): 1) Increased stream temperature due toloss of overhanging vegetation that is less suitable forthe biology of trouts; 2) Increased sedimentation frombank and upland erosion that trap and suffocate eggsand fry; 3) Increased channel width due to hoof-inducedbank sloughing and consequent erosion that reducestrout cover, decreases winter stream temperatures, andincreases susceptibility to formation of anchor ice; 4)Stream channel trenching or braiding that degradesinstream habitats and increases the stream'ssusceptibility to catastrophic floods; and 5) Plantcommunity alteration and/ or vegetation loss that reducebank cohesiveness, cover attributes, and terrestrial foodinputs.”

Destruction of Upland Soils and Vegetation.By removing herb and grass cover, and bycompacting soils, livestock grazing slows therate of water infiltration, leading to unnaturallyhigh and frequent runoff events (Dasmann1972, Holechek et al. 1989, ). Dasmann, forexample, reports that a heavily overgrazedwatershed on Utah’s Wasatch Frontexperienced severe flooding, while an adjacent,ungrazed watershed suffered little or noflooding (Dasmann 1972).

23

Livestock grazing also results in thereplacement of native grasses and herbs by non-native grass species, salt-cedar, juniper,mesquite, rabbit brush, and other shallowrooted vegetation less adapted for soilstabilization, thereby increasing runoff. Erosionand unnaturally heavy and frequent floodevents destroy trout habitat by silting in pools,uprooting riparian vegetation, widening andaggrading stream channels, and lowering watertables (Bock et al. 1992).

The direct connection between the health ofupland vegetation or habitat conditions, and thehealth of riparian or aquatic habitat conditionsis well illustrated in the EnvironmentalProtection Agency report Managing Change,Livestock Grazing on Western Riparian Areas(Chaney et al. 1993):

“Improper grazing of upland vegetation increases theamount, and concentrates and increases the speed ofoverland runoff to streams. Accelerated runoff fromuplands can trigger downcutting by streams with softbottoms. Downcutting lowers the streambed and watertable, dries out the riparian area, destabilizes streambanks, increases erosion, and further acceleratesrunoff.”

In the Diamond Bar Allotment ManagementPlan Draft Environmental Impact Statement(1993), the Forest Service agrees:

“The amount of residual forage in the uplands andriparian areas is critical for maintaining a healthywatershed. Riparian and upland areas provide groundcover which holds soil in place, cushions the impactsof raindrops, and provides for sediment retention.Cumulative watershed effects are the results ofdownslope and/or downstream interaction of runofffrom the management of activities that reduce theproductive land and water base. The primary physicalcauses of increased peak flow and increased runoffare soil compaction and removal of vegetativecover.”

Continued disturbance to soils, vegetation,hydrologic regimes and stream channelmorphology often leads to the completedessication on trout streams:

“The greatest damage from erosion on range landsoccurs where the areas have been overgrazed and theground cover destroyed or seriously impaired. Beforethe ranges had been overstocked and the groundcover impaired, erratic runoff and erosion werepractically unknown. After the breaking up of thevegetative cover in the early (eighteen) nineties,however, many streams originally of steady year-longflow and teaming with trout became treacherouschannels with intermittent flow through which thewater from rainstorms was plunged, or rose and fellaccording to the size and frequency of the storms andcarried so much sediment in the water that fish andsimilar life could not exist.” (Weyl, 1918)

X. HABITAT LOSS- WATERDIVERSIONS

Water diversion and withdrawal has severelyaffected trout habitat in many parts of NewMexico and southern Colorado. For example,water diversions from Rio Bonito forAlamogordo and Holloman Air Force base havecaused several miles of stream below BonitoLake and other reaches below Fort Stanton tobecome dry regularly during summer months.The city of Ruidoso draws heavily from thewatershed of the Rio Ruidoso, causing lowsummer flows. Several tributaries are now drybelow storage dams, due to intake forresidential and industrial use. As a result, theRuidoso-Hondo-Bonito river system no longerhas the capability to hold one interconnectingtrout population, and trout habitat in the systemhas become fragmented into several pieces(Smorynski, personal communication).Irrigation diversion accompanying theimmigration of early settlers into northern NewMexico resulted in the loss of streams thatlikely had provided historical virginalis habitat(Sublette et al. 1990).

XI. FIRE, FLOOD, ANDDROUGHT

24

Since virtually all remaining virginalispopulations exist in isolated streams, thepotential for catastrophic loss of any of thesepopulations and for an accumulation of suchlosses over time is very great. Once lost, thereis no natural means by which any of thepopulations can recruit new members andreturn. Such catastrophic losses of populationsof other species of trout have occurredrepeatedly over the Southwest over the pastseveral decades and continue today. A 1989 firein the Gila National Forest followed by heavyrains resulted in 100 percent mortality of theGila trout (O. gilae gilae) in Main DiamondCreek, which had held over 50 percent of theworld's population of the species (Propst et al.1992). A 1994 fire in the Chiricahua Mountainsexterminated the entire trout population of fourseparate streams. Fires in 1951 and 1954 causedthe extinction of trout in McKnight Canyon andLittle Turkey and upper Little creeks in the GilaNational Forest (Campbell, 1994; BruceAnderson, personal communication 1990).Flood conditions in the Prescott National Forestin the early 1980s followed by drought in 1989and 1990 resulted in the disappearance of theintroduced population of Gila trout in GapCreek by 1991 (Propst et al. 1992).

Drought and fire, often immediately followedby heavy rain and flooding, are part of thenatural scene in the forests of the Southwest.When trout populations exist, find passage, andinter-connect genetically in larger, morecomplex watersheds, such as the Rio Bonito-Ruidoso-Hondo system in the SacramentoMountains of Lincoln County New Mexico orthe upper Pecos or Chama river systems, troutalways remain in some portion of the watershedafter drought or fire. These are able to re-colonize the affected areas either by migrationor gradual augmentation and encroachment

surviving stocks. Isolated populations in first orsecond order streams can never recover aftersuch stochastic events, however. Thus, oversufficient time, the 92 remaining wild virginalispopulations can be expected to disappear oneby one if left isolated and unaided.

XII. ROAD BUILDING ANDLOGGING

Logging and road building are similar in theireffects on water quality and trout habitat in thatboth create extreme local disturbances in theform of unstable bare soil subject to erosion. Infact road building is one of side effects oflogging. Both logging and road buildingactivity causes short term sedimentation of trouthabitat, which in turn produces similardestruction of trout spawning gravels, riffles,and pools outlined in the above discussion ofgrazing effects. In the long term, runoff fromroads and road deterioration causesedimentation, while culverts can act as barrierspreventing upstream movement whichfragments trout habitat.

Aside from the obvious issue of degraded waterquality as a result of road building and roadrights of way, which affects all species ofcoldwater fishes, for virginalis there is theadded issue of increased fishing pressureresulting from increased road access. It must benoted that the great majority of remainingvirginalis populations live in streams notdirectly accessible by roads. There is evidencethat when roads are built along such streams,the populations of virginalis greatly diminish,while other species seem to take advantage andincrease. For instance, shortly after private landwas acquired by the Forest Service in 1967 and

25

Figures 6 & 7: logged and unlogged trout habitat.

26

opened to fishing along a newly-constructedforest service road, the ratio of virginalis tobrown trout in a portion of the stream along thenew road plummeted from 98 percent \2 percentto 10 percent \90 percent (New Mexico Gameand Fish 1973). This may be due to increasedfishing pressure, owing to the fact thatvirginalis are exceedingly easy to catch byangling and suffer from increased fishingpressure when stream access is improved; or itmay be that virginalis suffers another sort ofcompetitive disadvantage when water quality islessened by the results of road building. At anyrate, roads appear to cause steep declines invirginalis populations.

Road building in itself appears to be sufficientcause for the eventual local disappearance ofvirginalis, and probably has the most immediateand pronounced harmful effect. It is worthnoting that of the 53 remaining virginalispopulations in New Mexico, only three can bedirectly accessed by the public in motorvehicles, and the virginalis populations in twoof these streams have suffered noticeably frominvasion and competition provided by exoticbrown trout. Among the other fifty populations,twenty flow within officially designatedwilderness areas, ten flow within wildernessstudy areas and proposed wilderness areaadditions, and the rest are accessible only byfoot, pack, or jeep trails, or are on private landrestricted from full public use. In Colorado thepicture is much the same. Only two of thestate's 39 remaining virginalis populations aredirectly accessible to the public by a passablemotor road.

Habitat degradation facilitates competitionpressure by providing the conditions in whichbrook and brown trout can dominate the RioGrande cutthroat. Griffith (1988) states that inoptimal habitat, native cutthroat may be able towithstand competition from exotic trout,

leading Stefferud (1988) to conclude that“habitat protection and enhancement may beextremely powerful tools in the futuremaintenance of Rio Grande cutthroat troutstocks.” Rio Grande cutthroats have dominatedbrown trout in remote, coldwater streams suchas the Rio Chiquito and Upper Trinchera Creek(USFWS 1973). Logging, road building andlivestock grazing, by increasing siltation andraising water temperature, will favor browntrout over virginalis (USFWS 1973).

XIII. DISEASE AND PREDATION

Human Predation. The virginalis is a popularsport fish (Stumpff and Cooper 1996).Sportfishing always causes at least somedecline in populations of cutthroat trout, andthis may be exacerbated An Idaho studyshowed that 32 man hours of angling removed50 percent of the cutthroat trout above 6 inchesin one small, previously unfished stream.Currnet sportfishing regulations in New Mexicoallow capture and possession of two virginalis.,while the number of licensed New Mexicoanglers compares unfavorably with the totalpopulation of the sub-species. As mentionedabove, the inaccessibility of most remainingvirginalis habitats affords some protection fromangling take, but for many streams thisinaccessibility can by no means be expected tocontinue indefinitely. For instance, in 1994 theU. S. Forest Service tried to develop a"flatwater" fishery and high-density recreationarea along McCrystal Creek, a virginalis habitat(see below). If this effort had been successful,angler impacts along the stream would mostlikely have increased, along with the possibilityof unauthorized stocking of exotic species.

Even under the current regulatory scheme,enforcement of the two-fish limit can bedifficult. Stumpff and Cooper (1996) state that"since all subspecies of cutthroat trout seem to

27

be extremely vulnerable to angling, specialfishing regulations to protect populations of RioGrande cutthroat trout are needed. Regulationsmay include variations of angling methods,size, bag limit, and season. Strict enforcementof regulations will be required "(emphasisours).

Whirling Disease. Hatchery trout are notoriousin the West for spreading whirling diseaseamong wild trout stocks. Recent introduction ofthe disease eliminated 90 percent of the wildrainbow trout stocks in the Madison River andreduced populations in the Dolores River inColorado..

XIV. COMPETITION &HYBRIDIZATION

Hybridization and competition are second onlyto habitat loss as a past and current threat to thevirginalis (Stumpff and Cooper 1996). Habitatdegradation exacerbates non-habitat threatssuch as hybridization and competition byfavoring exotic and competing species overvirginalis. The continued survival of the RioGrande cutthroat trout requires the removal ofexotic trout from occupied and unoccupiedstreams (Sublette et al. 1990).

Historically, the Rio Grande cutthroat trout wasthe only trout species present within its range.Beginning in the 1800s, settlers introducedexotic trout species into trout streams inColorado, New Mexico and Texas. The Game& Fish Departments of New Mexico andColorado began organized trout-stockingprograms in the early 1900s (Stumpff andCooper, 1996). Exotic trout spread rapidly andare now present throughout the range of thevirginalis, with the exception of a few smallheadwater streams.

The most isolated and southerly population ofvirginalis, in the Animas Creek-Holden Prongdrainage in the Black Range of southern NewMexico, is currently jeopardized by rainbowtrout planted a small pond communicating withthe stream just outside the Gila National Forestboundary. The Rio Costilla in northern NewMexico continues to be stocked by the NewMexico Game and Fish Department with exoticrainbow trout, in spite of the fact that threevirginalis populations lie upstream withoutobvious barriers to protect them (Stumpff,personal communication 1997). Thoughcontamination of pure stocks generallyprogresses upstream, studies in GlacierNational Park indicate that impure, hybridizedor exotic populations of cutthroat trout inheadwater lakes can eventually affect thegenetic makeup of pure-strain populationsdownstream (Allendorf and Leary, 1988). Thisis significant in that many formerly barrenheadwater lakes in Colorado and New Mexicohave been stocked with hybrid or exoticcutthroat, while virtually all remainingvirginalis stocks extend upstream into highheadwater areas.

Hybridization. Rainbow trout and Rio Grandecutthroat trout spawn at the same time and inthe same habitat. The two species readilyhybridize. Since rainbows are actively stockedby state agencies, are better able to withstandhabit degradation, and successfully compete forspawning reds, they tend to genetically swampmixed species populations over time.

Until 1972, the New Mexico Department ofGame and Fish raised “native” cutthroat at theSeven Springs Hatchery and stocked themthroughout the state, often in high-elevationlakes draining into watersheds containingvirginalis. The stock was derived from fishcollected from the Costilla River. They weremost likely a hybrid form containing some non-

28

native cutthroat and rainbow ancestry (Behnke1979; Behnke 1981). Both these hybridcutthroats and the rainbow trout stocksintroduced into virtually every major coldwaterstream system in the state have accounted to agreat degree for the replacement of virginalis inhundreds of trout waters in New Mexico byhybridization alone. Although no virginalisstocks remain in the Pecos River basin of theSacramento Mountains, for example, many ofthe naturalized rainbow trout populationscontain specimens with tell-tail cutthroatmarkings indicating some remaining cutthroatinfluence, the vast majority of which has beenlost by repeated rainbow stockings andhybridization (Smorynski, personalcommunication).

Competition. Brown trout do not hybridizewith virginalis because they spawn earlier inthe year. This gives them a competitiveadvantage over Rio Grande cutthroats,however, because the latter lose a much higherpercentage of their eggs to seasonal flooding.Brown trout often replace Rio Grande trout atmiddle and lower streams, but are not assuccessful in higher elevation streams.

Like the all cutthroat subspecies, virginalisdoes not compete well with non-cutthroat troutspecies. It readily yields feeding spots torainbow and brown trout, even those of smallersize.

Brook trout also appear to displace cutthroatsfrom higher elevation streams and lakes.Wherever brook or brown trout are introducedinto cutthroat waters, the cutthroat troutpopulations have almost always been displaced(Young 1996).

Habitat Degradation. Habitat degradationfacilitates competition pressure by providing

the conditions in which brook and brown troutcan dominate the Rio Grande cutthroat. Griffith(1988) states that in optimal habitat, nativecutthroat may be able to withstand competitionfrom exotic trout, leading Stefferud (1988) toconclude that “habitat protection andenhancement may be extremely powerful toolsin the future maintenance of Rio Grandecutthroat trout stocks.” Rio Grande cutthroatshave dominated brown trout in remote,coldwater streams such as the Rio Chiquito andUpper Trinchera Creek (USFWS 1973).Logging, road building and livestock grazing,by increasing siltation and raising watertemperature, will favor brown trout overvirginalis (USFWS 1973).

Fish Barriers. Eighty to ninety percent of allvirginalis habitats in New Mexico are upstreamfrom brown trout populations. Many of theremaining populations of virginalis areprotected from rainbow trout introgression andbrown trout displacement because of naturaldownstream barriers in the streams, such asfalls or downstream desiccation. When suchnatural barriers are not available, artificialbarriers are regularly placed in virginalisstreams (20 of the streams have had suchbarriers erected in them).

While barriers are an important short-termstrategy to protect existing populations whilehabitat is restored and exotics are removed, it iscounterproductive as a long term strategy orsubstitute for habitat restoration. Barriersexacerbate genetic isolation and fragmentationof the stream populations into dangerouslysmall interbreeding groups. Failure to restoretrout streams and connect virginalis populationswill result in a permanently endangered speciesat best, stochasticly driven extinction at worst,and a very expensive, high maintenanceconservation strategy.

29

To extend virginalis populations downstreamand to allow genetic exchange among thesetrout in a number of tributary streams, the triplehabit of blocking, poisoning, and stocking,either with wild fish or the offspring of certifiedbroodstock, is employed. This practice isexpensive, time-consuming, often unsuccessful,and generally unpopular with the public.Attempts are currently under way to reclaimJack's Creek and the head of the Pecos River inthis way. If this is successful, it will provide theonly complex drainage holding virginalis andthe most stable habitat for the subspecies.

Roughly a dozen virginalis habitats in NewMexico are not protected by downstreambarriers of any sort (Stumpff, personalcommunication 1997).

XV. POOR WATERSHEDCONDITION

According to Stumpff and Cooper (1996), 38%of occupied virginalis watersheds are in stablecondition while 55% are in an unknowncondition. Current habitat conditions areinsufficient to expand virginalis populations:

"Habitat restoration is essential to re-establishingcutthroat trout to their former range...overhangingbanks, riparian vegetation, instream boulders, logjams, pools, water volume, and water depth need tobe restored. Watershed conditions need to berestored." (Stumpff and Cooper 1996)

A 1991 report by the Forest Service’s air andwatershed department (USFS 1991), indicatesthat degradation is likely worse outsidecurrently occupied sub-basins:

“...there are still millions of acres of land andthousands of miles of stream that remain in anunsatisfactory condition. Riparian areas, instead ofbeing lush green oases in the hot, dry climate, arevoid of vegetation, eroding, and, frequently, as dry asthe uplands.”

Many of the 50% of Forest Service watershedsidentified by the report as being in degradedcondition, will have to be restored if the RioGrande cutthroat is to ever to adequatelyexpand its current range.

XVI. INADEQUACY OFEXISTING REGULATORYMECHANISMS

Livestock Grazing. There are five NewMexico National Forests within the current andhistoric range of the Rio Grande cutthroat trout(Gila, Lincoln, Cibola, Santa Fe and Carson).Despite the fact that scientists have expressedconcern over the impacts of grazing onvirginalis, none of the New Mexico NationalForest have adopted a specific conservationplan for the Rio Grande cutthroat trout.

In May, 1996, the Regional Forester signed aRecord of Decision amending the Forest Plansfor the five forests. The amendment includednew grazing direction apparently intended toreverse an admitted trend toward federal listingfor aquatic and riparian species under theunamended Forest Plans. The Rio Grandecutthroat was not analyzed in the accompanyingEnvironmental Impact Statement. The Recordof Decision contained no virginalis specificpolicy changes. The new grazing directionconsists of a utilization table which is supposedto be used in the absence of site specificinformation. Since some site specificinformation exists for almost all grazingallotments, it is likely that the utilization tablewill not be implemented. Even if implemented,there is no evidence to indicate whether or how,the table would improve grazing in Rio Grandecutthroat trout habitat. The table does notrequire the removal of cattle from riparianareas, trout habitat, cold water fisheries, or anyother aquatic designation.

30

In 1997, the Southwest Region of the ForestService, to ward off a threatened jeopardydecision on seven species endangered bylivestock grazing, developed a Seven SpeciesPlan. The Rio Grande cutthroat trout was notincluded, nor is there any substantial overlapbetween the seven species and the trout.

On April 1, 1997, four fisheries biologistspresented a briefing on the impacts of grazingon imperiled fish on Southwestern NationalForests to the Regional Forester. Theirconclusion were remarkably candid anddamning:

“The cumulative and synergistic effects of ForestService management is causing long-termdegradation of the habitats of these species, andcontributing to their endangerment and downwardtrend in range and abundance. Many of these effectsare due to irreversible activities that occurred in thedistant past. But some are due to current anddeliberate action. During our interviews we heardtime and again that the needs of the species were notfully considered during NEPA analysis. We heardthat terms and conditions of the programmatic BAEfor grazing weren’t being followed. We heard thatbiologists were pressured into changing effectsdeterminations so that targets could be met withouthaving to undergo consultation. We heard thatmitigation measures weren’t applied. But we werealways assured that there actually was no problem.”

“There are several hundred other riparian dependentspecies in the region, wildlife that will become thesubject of listings and lawsuits if we don’t effect achange. We need incentives for line officers tocommit to riparian area and endangered speciesmanagement. We need to commit to management forforest health. Above all, we need a change inmanagement attitude.”

“For example, we found that range management is achronic abuser of riparian habitats. Now rangemanagers truly believe in their hearts that degradedriparian areas can be restored with cattle. And theyhave come up with an amazing variety of grazingsystems to accomplish that...But evaluations ofriparian area condition 5 or 10 years later seldomshow an upward trend. Why is that? It’s becausecattle grazing is a core value of the agency, and

riparian area health and endangered speciesmanagement is not. Prescriptions are developed andapplied to meet the needs of the rancher, the cattle, orthe agency. Soil, vegetation, water, and wildliferesources are secondary considerations.”

“Recovery of riparian areas with cattle hasn’t workedin the past, is not working now, and won’t work inthe future. And this is where a change in managementattitude is necessary. The only practical way torestore riparian areas supporting endangered speciesis through removal of cattle impact. And based onexperience, we advocate that prescriptions that callfor complete rest or nonuse be the first step. Achange in attitude to recognize that other multipleuses in riparian areas are more beneficial to thegreatest number than a few AUM’s is necessary.”

“But, management in this region has traded off itslove and passion for the land in order to indulge ineconomically questionable targets. Gifford Pinchot’sphilosophy of “...providing the greatest good for thegreatest number...” has been distorted to a doctrine ofproviding the most economic use for the few. Andthis has resulted in the current situation: the FWSthreatening a jeopardy call on our management,outside groups taking us to court (and winning) onthe same issue, and we being the subject ofwidespread ridicule and derision.”

Overgrazing and habitat destruction continues.Streams flowing through high-elevationmeadows are often impacted by cattle or sheepgrazing, even within officially designatedwilderness areas. In particular, the ninevirginalis streams arising within the San PedroParks Wilderness of the Santa Fe NationalForest- nearly ten percent of the remainingvirginalis habitat- are largely of the open vega,or meadow type, and cattle utilize the riparianzones of a number of these streams. Most areheavily degraded. Riparian areas within thePecos Wilderness are also degraded.

Logging and Road Building. None of the NewMexico National Forest contain specificmanagement direction to prohibit or limit theimpact of logging or road building on the RioGrande cutthroat trout. Though the newlyamended Forest Plans contain logging

31

restrictions to benefit the Mexican spotted owland Northern goshawk, these restrictionsprovide little benefit for virginalis because theydo not include aquatic reserves or riparianbuffers.

If carried out, it is likely that road constructionfor the proposed Angostura Timber Sale alongAngostura Creek in Carson National Forest willcause a noticeable decline or evendisappearance in the creek's survivingpopulation of virginalis. It is also clear thatnearly half of New Mexico's remainingvirginalis populations exist in de factowilderness areas open to road-building and thusare unprotected from the potentialdisappearance or severe impact represented byroad building alone.

In Colorado, only three populations are inprotected wilderness areas, 15 in unprotected defacto wilderness, and 20 in lands under privatecorporate ownership subject to possible futureroad or subdivision development.

Logging on private lands in New Mexico andsouthern Colorado has the potential to harm oreven extirpate virginalis populations. Neitherthe State of Colorado nor New Mexico possessadequate environmental regulations tosignificantly control private lands logging. Theowners of the Taylor Creek Ranch, forexample, have destroyed Rio Grande cutthroattrout habitat and harmed existing populationswith very little regulatory resistance from thestate of Colorado.

U.S. Fish & Wildlife Service. The Rio Grandecutthroat trout should have been listed andprotected as an Endangered Species decadesago. Had this been done, it may already be wellon its way to complete recovery today.

The Rio Grande cutthroat trout was listed asendangered in the 1967 Red Data Book of theU.S. Bureau of Sports Fisheries and Wildlife(the precursor to the U.S. Fish & WildlifeService). Virginalis did not appear in the 1968edition, however, probably because of a reportby Behnke suggesting that unknown would bea better description given the lack of conclusivedata about the subspecies taxonomy, historicaldistribution, and status (Behnke 1967).Strangely, Behnke’s study did establish withfair accuracy the historic and currentdistribution. Perhaps this why the Bureau’sDivision of Fisheries, based on a newrecommendation by Behnke, suggested thatvirginalis be listed in the 1969 edition as rare(King 1968).

The 1972, 1973, and 1974 Red Data Books listvirginalis as rare and endangered because ofrange declines associated with “drasticenvironmental changes wrought by use andmisuse of water; the introduction of exotic fishspecies, particularly rainbow trout and othersubspecies of cutthroat trout...” (USBSF&W1972). Indeed, the 1972 book states thatvirginalis is not viable, establishing two sets ofshort term goals to maintain the species and torestore it as a “viable component of theecosystem.” Intensive management of livestock(including fencing of streams), protection fromlogging and road building, and cessation ofexotic trout introductions were recommended.

The Rio Grande cutthroat is listed as athreatened species on the March 1973 list of“Threatened Wildlife of the United States”(USBSF&W 1973). It was apparently listed asendangered on the 1971 list (USFS 1971).Strangely, the U.S. Fish and Wildlife Servicefailed to place the Rio Grande cutthroat on theendangered species list of 1974 following thepassage of the Endangered Species Act inDecember, 1973. That Act mandates that all

32

species previously listed as threatened betransferred to the new list.

The Rio Grande cutthroat trout was made acandidate for listing under the EndangeredSpecies Act in 1985 (USFWS 1985).

U.S. Forest Service Conservation Plans. TheForest Service manages the vast majority of RioGrande cutthroat trout populations andtherefore has an extraordinary responsibility torecover the species. Sixty seven of the currentlyexisting Ninety two populations are on publicland, most Forest Service (Stumpff and Cooper1996). Though the Forest Service has a legalmandate under the National Forest ManagementAct to maintain viable, well-distributedpopulations of vertebrate species, and thoughthe Rio Grande cutthroat was declared to be nolonger viable as far back as 1972 (USBSF&W1972), none of the National Forests in Coloradoor New Mexico have developed a conservationplan to preserve the species.

In 1971, the Santa Fe National Forestdeveloped a draft Rio Grande cutthroat troutconservation plan (USFS 1971). Though theplan states:

“It is essential that the existing fishery habitat bemaintained, improved, and protected to insure futuresurvival of this species for the benefit of mankind,”

The plan was never finalized. In 1993, theForest stated that it was developing a RioGrande Cutthroat Restoration Plan with theCarson National Forest and the New MexicoDepartment of Game and Fish (Tunberg 1993).The plan was never completed.

In 1993, the Lincoln National Forest stated thatthe Sacramento division had begun “feasibilitysurveys” as a first step toward re-introducingthe subspecies to the Forest (Delorenzo 1993).No populations were ever introduced.Virginalis remains extirpated from the Lincoln

National Forest which believes that the absenceof the species means its actions can not harm it(Martinez 1997). No management plan forvirginalis has been created.

In 1994, the Carson National Forest announcedthat a draft multi-agency virginalis plan wouldbe completed in early 1994 (Lucero 1994). Theplan was to be produced by the Forest Service,Bureau of Land Management, and the NewMexico Department of Game and Fish. In 1997,the Forest promised that the plan would becompleted in 1997. Thus far, no plan has beenreleased (Lucero 1997).

In 1992, the Rio Grande National Forestproduced a draft virginalis conservation plan. Itwas never finalized.

The Gila National Forest, though it supports avirginalis population in Animas Creek, has noRio Grande cutthroat trout conservation plan.

In 1988, the Rio Grande cutthroat trout waslisted as a Forest Service sensitive species inNew Mexico. It is also a Forest Servicesensitive species in Colorado.

In 1992, the Chief of the U.S. Forest Serviceasked Regional Foresters in the IntermountainWest to develop inter-regional HabitatConservation Assessments (HCAs) for allinland cutthroat trout species. In 1994, theChief reiterated the need for HCAs. The ForestService has published two volumes ofconservation assessments for the Rio Grandeand other inland cutthroat trout, but has thus farnot developed a regional or inter-regionalconservation plan.

State of New Mexico- Habitat Protection.Since the majority of virginalis sites andhabitats in New Mexico are on National Forest,National Park, Indian Nation, and private lands,

33

the New Mexico Department of Game and Fishis limited in its capacity to regulate landmanagement to protect and restore the RioGrande cutthroat trout.

The Department’s repeated warnings about theimpact of habitat degradation have not resultedin significant changes in land managementagencies. In 1976, the Department issued areport entitled “Status and Future of the RioGrande cutthroat trout” (Hubbard 1976). Thereport outlined the serious condition of the RioGrande cutthroat, and the steps necessary torestore it:

“Since Spanish days, the range and numbers of thistrout have probably been declining in New Mexico,as degradation of habitat (e.g. through overgrazing)and use and misuse of water and later theintroduction of exotics...by 1972 the subspecies haddeclined to the point of being listed as a threatenedspecies by the U.S. Fish and Wildlife Service...thedecline is unlikely to reverse itself without humanhelp...”

“...streams with limited riparian growth and erosionalproblems (e.g. degraded banks, rapid run-off formwatershed or streambed) tend to lack proper shade(and thereby have higher temperatures), sufficientfeeding and shelter areas, and a suitable diversity ofinvertebrate life. Such streams harbor few or nonative trout, whereas streams without these didsustain fish...it appears certain that sufficientinformation is now known, to correct— or at leastattend to— deficiencies in streams to insure bettertrout survival in them.”

“To date, on a limited amount of management hasbeen carried out on behalf of the native cutthroat.With proper attention, this situation can be reversed,and the process should begin soon. For too long, thissubspecies has survived in spite of, instead of at thebehest of man, and without concerted efforts, we getfarther behind.”

The Department recommended reduction orelimination of livestock from the five knownnorthern New Mexico metapopulations,establishment of fishery and a supportinghatchery program, a management plan for eachmetapopulation, restoration of streams occupied

by each metapopulation, and the cessation ofstocking exotic trout in streams connectingpopulations.

In 1989, the Department published the resultsof a statewide fisheries investigation (Subletteet al. 1990) It identified the same ongoingproblems 13 years after its previous report:

“Most streams occupied by Rio Grande cutthroattrout have been impacted by overgrazing andlivestock. Limited vegetation in the watershed,especially in riparian areas, has led to altered streamnutrient and sediment loads, and has modified flowregimes along with the morphology of the streamcourse. Trampling of streambanks by livestock hasfurther accelerated habitat destruction. Trout survivalin many of these streams is impaired because of thelack of productive riffle areas, suitable spawningsites, undercut banks (to escape predation), pools (forresting, feeding and overwintering), and shade (inproper proportions which reserves cold watertemperatures yet allows adequate solar gain essentialfor primary production). “”Other factors such as lumbering, mining, and firesalso has had adverse effects on Rio Grande cutthroatsimilar to what grazing has done as far as destroyinghabitat.”

“To insure its survival, it is imperative that exoticsalmonids be excluded from the streams where O. c.virginalis persists.”“Considering the restricted areas, habitat destruction,competition with exotics, and hybridization there isa definite need to stabilize the present populations ofRio Grande cutthroat and to try to renovate lost areasto avoid losing the species altogether.”

State of New Mexico- ReintroductionPrograms. All re-introductions of virginalis inNew Mexico to date have been accomplishedby relocation of captured wild fish. Efforts byNew Mexico Game and Fish to secure ahatchery or naturalized broodstock of virginalisfor possible re-introduction efforts have thus farbeen unsuccessful.

Plans to develop a breeding population ofvirginalis in a small reservoir on McCrystal

34

Creek were abandoned in 1994 when the damforming the reservoir developed a number ofleaks and had to be destroyed. The New MexicoDepartment of Game and Fish is currentlynegotiating with the U. S. Fish and WildlifeService to make some use of the new federalfacility in Mora, New Mexico for the rearing ofvirginalis (Stumpff, personal communication1997).

State of New Mexico Fishing Regulations.The Rio Grande cutthroat trout is regulated asa game fish in New Mexico and Arizona.Permitted fishing is allowed. There are no non-fishing refugia. New Mexico has not listedvirginalis under its state protected species list.

Federal Reintroduction Programs. Nonumbers of pure-strain virginalis had beenreared at a hatchery within New Mexico until1987, when fish of the large-spotted, Pecosstrain of virginalis from Indian Creek on theMescalero Apache Reservation were placed inthe Mescalero National Fish Hatchery facilitywith the intent to stock reservation waters.These fish failed to thrive, and by 1993 allstock had died, without any re-introductionshaving occurred.

Since then, efforts have been made to developa broodstock at the New Mexico Seven Springsfacility and in the Rio Cebolla upstream fromthe facility, with fish of the smaller-spotted RioGrande form taken from Cañones Creek, upperRio Las Vacas, and Rio Puerco in the JemezMountains. These efforts are continuing.

XVII. CRITICAL HABITATDESIGNATION RECOMMENDED

Petitioners strongly recommend the designationof critical for the Rio Grande cutthroat troutcoincident with its listing. Critical habitatshould be designated in all areas where it iscurrently located and in key unoccupied areaswhere restoration is necessary for theconservation of the species.

35

Respectfully submitted,

____________________________Kieran SucklingExecutive DirectorSouthwest Center for Biological DiversityP.O. Box 710Tucson, AZ 85702520-623-5252

_____________________________Rex JohnsonDirectorSouthwest Trout713 W. 6th StreetSilver City, NM 88061505-388-9334

___________________________D.C. “Jasper” CarltonDirectorBiodiversity Legal FoundationP.O. Box 18327Boulder, CO 80302303-442-3037

_____________________________Joanie BerdeDirectorCarson Forest WatchP.O. Box 15Llano, NM 87543505-587-28

____________________________Mike McGowenAncient Forest RescueP.O. Box 762San Louis, CO 81152719-672-3012

Signed this 5th day of February, 1998

36

REFERENCES CITED

Allendorf, F.W. and R.E. Leary. 1988.Conservation and distribution of geneticvariation in a polytypic species, the cutthroattrout. Conservation Biology 2:170-184.

Anderson, J. R. 1993. Letter to Peter Galvin,Greater Gila Biodiversity Project dated 15December, 1993 From Creede RangerDistrict, Rio Grande National Forest.

Armour, C.L., D.A. Duff, and W. Elmore.1991. The effects of livestock grazing onriparian and stream ecosystems. Fisheries 16:7-11.

Baxter, G. T. and J. R. Simon 1970. Wyomingfishes, Wyoming Game Fish Comm. Bull.No. 4, 168.

Behnke, R. J. 1968. Progress Report: CutthroatTrout of the Rio Grande and Colorado RiverBasins. Cooperative Fishery Unit, ColoradoState University, Fort Collins, CO.

Behnke, R. J. 1988. About Trout, Rio GrandeCutthroat. Trout, Summer 1988, 29(3): 43-7

Behnke, R. J. 1992. Native Trout of WesternNorth America. American Fisheries Society,Monograph No. 6.

Behnke, R. J. 1967. Rare and EndangeredSpecies: The Rio Grande Cutthroat Trout.Colorado Co-operative Fishery Unit,Colorado State University, Fort Collins, CO.pp. 6-11.

Behnke, R. J. 1979. Monograph of the NativeTrouts. BLM, Denver.

Behnke, R. J. 1981 The Cutthroat Trout ofIndian Creek, Mescalero Apache Indian

Reservation, Tularosa Basin, Otero county,New Mexico.

Behnke, R. J. and M. Zarn. 1976. Biology andmanagement of threatened and endangeredwestern trouts. Gen. Tech. Rep. USDA ForestService, RM-28: 1-45. Rocky MountainForest and Range Experiment Station, FortCollins, CO.

Bock, C.E., V.A. Saab, T.D. Rich, and D.S.Dobkin. 1992. Effects of livestock grazing onneotropical migratory landbirds in westernNorth America. USDA Forest Service Gen.Tech. Rep. RM-229. Rocky Mountain Forestand Range Experiment Station, Fort Collins,CO.

Campbell, D. 1994. Letter to WildernessRanger District, Gila National Forest.Published IN: Addendum to the March 24,1980 Environmental Analysis, Gila TroutRecovery Projects on the Gila Nationalforest, 1994 Restoration of Gila Trout. USDAForest Service, Gila National Forest,Wilderness Ranger District.

Chaney, E., W. Elmore, and W. S. Platts 1990.Livestock Grazing on Western RiparianAreas. EPA report. 14-7, 26-7.

Chaney, E., W. Elmore, and W.S. Platts.. 1993.Managing change; livestock grazing onwestern riparian areas. US EnvironmentalProtection Agency. Denver, CO. 31 pp.

Cutler, Alan. 1991. Nested faunas andextinction in fragmented habitats.Conservation Biology Vol. 5, No. 5: 496-505.

Dasmann, Raymond F. 1972. EnvironmentalConservation. John Wiley and Sons, NewYork, NY. 473 pp.

37

DeLorenzo, D. 1993. Letter from DonDeLorenzo, Lincoln National Forest, to PeterGalvin, Greater Gila Biodiversity Project,dated December 23, 1993.

Duff, D. 1996. Bonneville cutthroat trout(Oncorhynchus clarki utah) IN Duff, D.A.,technical editor. 1996. ConservationAssessment for Inland Cutthroat Trout:Distribution, Status and Habitat ManagementImplications. USDA Forest Service,Intermountain Region, Ogden Utah.

Duff, D.A. 1995. Fish stocking in U.S. FederalWilderness Areas: Challenges andopportunities. International Journal ofWildlife 1:17-19

Duff, D.A., technical editor. 1996.Conservation Assessment for InlandCutthroat Trout: Distribution, Status andHabitat Management Implications. USDAForest Service, Intermountain Region, OgdenUtah.

Griffith, J. S. 1988. Review of competitionbetween cutthroat trout and other salmonidsIN Robert E. Gresswell ed., Status andManagement of Interior Stocks of CutthroatTrout, American Fisheries SocietySymposium 4:134-140.

Hall, J. D. and C. O. Baker. 1982.Rehabilitating and enhancing stream habitat:1. Review and evaluation. U.S. ForestService General Technical Report PNW-138.

Hanske I. and Gilpin, M. 1991. Metapopulationdynamics: Brief history and conceptualdomain. Biological Journal of the LinneanSociety 42:3-16.

Harig, Amy L. and Kurt D. Fausch. 1996.Compilation of data on Colorado waters

containing Rio Grande cutthroat trout: FinalReport to Colorado Division of Wildlife(MOU Project No. 904-96). ColoradoDivision of Wildlife, Montrose, Co.

Holechek, J.L., R.D. Piper, and C.H. Herbel.1989. Range management: principles andpractices. Prentice Hall, Englewood Cliffs,New Jersey.

Hubbard, J. P. 1976. Status and Future of theRio Grande Cutthroat Trout. New MexicoDepartment of Game and Fish, EndangeredSpecies Program Report.

Hunter, C. J. 1991. Better Trout Habitat. IslandPress, Washington, D. C.

King, W. 1968. Memo from Willis King, Chief,Division of Fishery Services, to Office ofEndangered Species, U.S. Bureau of SportFisheries and Wildlife, dated August 16,1968.

Krueper, D.J. 1992. Effects of land usepractices on western riparian ecosystems.USDA Forest Service Gen. Tech. Rep. RM-229. Rocky Mountain Forest and RangeExperiment Station, Fort Collins, CO.

Lamberson, R.H., R. McKelvey, B.R. Noon,and C.Voss. 1992. A dynamic analysis ofnorthern spotted owl viability in a fragmentedforest landscape. Conservation Biology 6/4:505-512

Lande, R. 1987. Extinction thresholds indemographic models of territorialpopulations. American Naturalist 130: 624-635.

Levins, R. 1969. Some demographic andgenetic consequences of environmentalheterogeneity for biological control. Bulletin

38

of the Entomological Society of America 15:237-240.

Little, R. G. and H. J. McKirdy. Statewidefisheries Investigations, Rancho del RioGrant, Carson N. F., Project Nos. F-22-R-7 &8, Job No. A-1. 14 pp.

Lucero, L. 1994. Letter from Leonard Lucero,Carson National Forest Supervisor, to PeterGalvin, Greater Gila Biodiversity Project,dated January 6, 1994.

Lucero, L. 1994. Letter from Leonard Lucero,Carson National Forest Supervisor, to

Luecke, C. 1986. Ontogenetic changes infeeding habits of cutthroat trout. Trans. Am.Fish. Soc. 115:703-10.

MacPhee, C. 1966. Influence of differentialangling mortality and stream gradient on fishabundance in a trout-sculpin biotype. Trans.Amer. Fish. Soc. , 95(4):381-387

Marcus, M.D., M.K. Young, L.E. Noel, B.AMullen. 1990. Salmonid-habitat relationshipsin the western United States: A review andindexed bibliography. USDA Forest Service,Rocky Mountain Forest and RangeExperiment station Gen. Tech. Rep. RM-188.84 pp.

Martinez, J. M. 1997. Letter from Jose M.Martinez, Lincoln National ForestSupervisor, to Landi Fernley, SouthwestCenter for Biological Diversity, dated March17, 1994.

May, B.E. 1996. Yellowstone cutthroat trout(Oncorhynchus clarki bouvieri) IN Duff,D.A., technical editor. 1996. ConservationAssessment for Inland Cutthroat Trout:Distribution, Status and Habitat Management

Implications. USDA Forest Service,Intermountain Region, Ogden Utah.

McCaffe 1966. Lahontan cutthroat trout,pp.225-31. In: Calhoun, A (ed. ), InlandFisheries Management, California Dept. ofFish and Game.

McClane, A. J. 1965. McClane's StandardFishing Encyclopedia. Holt, Rinehart andWinston. 210-212.

McGurrin, J.C., C. Ubert, and D. Duff. 1995.Use of cultured salmonids in the federal aidin sport fishing restoration program.American Fisheries Society Symposium15:12-15.

Murry, M.I. 1996. Natural processes:Wilderness management unrealized. NaturalAreas Journal 16(1):55-61.

New Mexico Department of Game and Fish.1973. Rio Grande Cutthroat Trout. StatusReport. 10 pp.

New Mexico Department of Game and Fish.1968. D. J. project (F-22-R-788, February,1968)

Pimm, Stuart L., Jared Diamond, Timothy M.Reed, Gareth J. Russel, and Jared Verner.1993. Times to extinction for smallpopulations of large birds. Proc. Natl. Acad.Sci., Vol 90: 10871-10875.

Platts, W.S. 1991. Livestock grazing. Pp. 389-423 In: W.R. Meehan, editor. Influences ofForest and Rangeland Management onSalmonids Fishes and their Habitats. Amer.Fish. Soc. Spec. Pub. 19: 389-423. Bethesda,MD. 751 pp.

39

Platts, Bill, and R.F. Raleigh. 1984. Impacts ofgrazing on wetlands and riparian habitat. Pp.1105-1117 In: Developing Strategies forRangeland Management. Committee onDeveloping Strategies for RangelandManagement , Na t iona l ResearchCouncil/National Academy of Sciences.Westview Press, Boulder, CO.

Propst, D. L. and M. A. McInnis 1975. Ananalysis of streams containing native RioGrande cutthroat in the Santa Fe NationalForest. WICHE Report for the Santa FeNational Forest, Region 3, Albuquerque,NM.

Propst, D. L. 1977. Taxonomic Analysis ofCutthroat Trout, Salmo Clarki Ssp., Collectedin Northern New Mexico, 1976. Final Report,Contract No. 516-64-22, New Mexico Gameand Fish.

Propst, D. L., J. A. Stefferud and P.R. Turner1992. Conservation and Status of Gila Trout,Oncorhynchus Gilae. The SouthwesternNaturalist 37(2)117-125.

Pulliam, H.R. 1988. Sources, sinks andpopulation regulation. American Naturalist132(5):652-661.

Reiman, B.E., D.C. Lee, J. McIntyre, C.K.Overton, and R. Thurow. 1993. Considerationof extinction risks for salmonids. USDAForest Service, Fish Habitat RelationshipsTechnical Bulletin No 4.

Reiman, B.E. and K.A. Apperson. 1989. Statusand analysis of salmonid fisheries: Westsidecutthroat trout synopsis and analysis offishery information. Idaho Department ofFish and Game, Boise. Job performancereport, project F-73-R-11, subproject No. 11,Job No. 1.

Rieman, B.E. and J.D. McIntyre. 1993.Demographic and habitat requirements forconservation of Bull trout. USDA ForestService General Technical Report INT-302.

Rinne, J. N. and R. A. Lafayette 1991.Southwestern Riparian-Stream Ecosystems:Research Design, Complexity, andOpportunity. USDA Rocky Mountain Forestand Range Experiment Station, Fort Collins,CO. 9pp.

Schaffer, M.L. 1987. Minimum viablepopulations; coping with uncertainty INViable Populations for Conservation, M.E.Soule, editor, 69-86.

Shiazoa, D. K. and R. N. Williams 1992.Geographic variation, isolation and evolutionof cutthroat trout (Salmo clarki). NationalGeographic Society Final Report, Grant #3653-87. 28 pp.

Smith, R. H. 1984. Native Trout of NorthAmerica. Frank Amato Publications,Portland, OR.

Stacey, P.B. and M. Taper. 1992.Environmental variation and persistence ofsmall populations. Ecological Applications2(1):18-29.

Steffurd, J. 1988. Rio Grande cutthroat troutmanagement in New Mexico IN Robert E.Gresswell ed., Status and Management ofInterior Stocks of Cutthroat Trout, AmericanFisheries Society Symposium 4:90-92.

Stumpff, B. and J. Cooper. 1996. Rio GrandeCutthroat Trout, Oncorhynchus clarkivirginalis IN Duff, D.A., technical editor.1996. Conservation Assessment for InlandCutthroat Trout: Distribution, Status andHabitat Management Implications. USDA

40

Forest Service, Intermountain Region, OgdenUtah.

Sublette, J. E., M. D. Hatch and M. Sublette1990. The Fishes of New Mexico. Universityof New Mexico Press, Albuquerque, NM.

Szaro, R. C. 1989. Riparian forest andscrubland community types of Arizona andNew Mexico. Desert Plants 9: 69-138.

Thompson, P.D. 1995. Evaluating theeffectiveness of electro-fishing and man-made barriers for controlling Brook troutpopulations in streams containing cutthroattrout. MS Thesis, University of Wyoming,Laramie. 272 pp.

Tripp, L. and D. B. Rockland 1988. FisheryResources of the National Forests: Extent,Uses, and Economic Benefits. USDA ForestService Wildlife and Fisheries, Washington,D.C.

Trotter, P. C. 1987. Cutthroat, Native Trout ofthe West. Colorado Associated universityPress, Boulder, CO. pp.1-19.

Trotter, P.C. and P.A. Bisson. 1988. Historyand discovery of the cutthroat trout.American Fisheries Society Symposium 4:8-12.

Tunberg, G. 1993. Letter from Gail Tunberg,Wildlife Biologist, Santa Fe National Forest,to Peter Galvin, Greater Gila BiodiversityProject, dated December 20, 1993.

U.S.B.S.F. & W. 1972. Rio Grande cutthroat(Salmo clarki virginalis). Red Data Book,U.S. Bureau of Sports Fisheries and Wildlife,Washington, D.C.

U.S.B.S.F. & W. 1973. Threatened Wildlife ofthe United States. Office of EndangeredSpecies and International Activities, Bureauof Sport Fisheries and Wildlife, ResourcePublication 114, March 1973 (RevisedResource Publication 34).

U.S.F.S. 1971. Draft Rio Grande cutthroathabitat management plan. Santa Fe NationalForest, Santa Fe, NM.

U.S.F.S. 1994. Letter from Chief to RegionalDirectors reiterating importance ofdeveloping Habi ta t Conservat ionAssessments for inland cutthroat trout.

U.S.F.S. 1992. Letter from Chief to RegionalDirectors requesting preparation of HabitatConservation Assessment for inland cutthroattrout.

U.S.F.S. 1993. Diamond Bar grazing allotmentdraft environmental impact statement; GilaNational Forest. USDA Forest Service.Southwestern Region, Albuquerque, NewMexico. pp.

U.S.F.W.S. 1973. Rio Grande cutthroat (Salmoclarki virginalis). Red Data Book, U.S.Bureau of Sports Fisheries and Wildlife,Washington, D.C.

Utah Division of Wildlife Resources. 1991.Dewatered stream miles. Instream flow taskforce, Utah Division of Wildlife Resources,Salt Lake City, UT.

Van Eimeren, P. 1996. Westside cutthroat trout(Oncorhynchus clarki lewisi) IN Duff, D.A.,technical editor. 1996. ConservationAssessment for Inland Cutthroat Trout:Distribution, Status and Habitat ManagementImplications. USDA Forest Service,Intermountain Region, Ogden Utah.

41

Warren & Burr. 1994. Status of fresh waterfishes. Fisheries 19(1):6-18.

Williams et al. 1989. Fishes of North America:Endangered.. Fisheries 14(6):2-20.

Wilson, Edward O. 1992. The diversity of life.The Belnap Press of Harvard UniversityPress, Cambridge, MA. 424 pp.

Young, M. K., R.N. Schmal, T. W. Kohley andY. G. Leonard. 1996. Colorado Rivercutthroat trout (Oncorhynchus clarkipleuriticus) IN Duff, D.A., technical editor.1996. Conservation Assessment for InlandCutthroat Trout: Distribution, Status andHabitat Management Implications. USDAForest Service, Intermountain Region, OgdenUtah.

Young, M. K. 1996. Summer movements andhabitat use by Colorado River cutthroat trout(Oncorhynchus clarki pleuriticus) in small,montane streams. Canadian Journal ofFisheries and Aquatic Sciences 53:1403-1408.