Great Basin Naturalist

Volume 52 Number 3 Article 14

12-18-1992

Full Issue, Vol. 52 No. 3

Follow this and additional works at: https://scholarsarchive.byu.edu/gbn

Recommended Citation (1992) "Full Issue, Vol. 52 No. 3," Great Basin Naturalist: Vol. 52 : No. 3 , Article 14. Available at: https://scholarsarchive.byu.edu/gbn/vol52/iss3/14

This Full Issue is brought to you for free and open access by the Western North American Naturalist Publications at BYU ScholarsArchive. It has been accepted for inclusion in Great Basin Naturalist by an authorized editor of BYU ScholarsArchive. For more information, please contact [email protected], [email protected]. dbkddbmdb

T H E

GREATG R E AT BASINB A S I1 N naturalistNATafuf A I1ST

A

VOLUME 52 NR 3 SEPTEMBER 1992

BRIGHAM YOUNG university GREAT BAINBASIN naturalist editor JAMES R BARNES 290 MLBM brigham young university provo utah 8484602602

associate editors MICHAEL A BOWERS BRIAN A MAURER blandy experimental farm university of department ofzoology brigham young university virginia box 175 boyce virginia 22620 provo utah 84602

J R CALLAHAN JIMMIE R PARRISH museum of southwestern biology university of BIOWESTBIO WEST inc 1063 west 1400 north logan new mexico albuquerque new mexico utah 84321 mailing address box 3140 hemet california 92546 PAUL T TUELLER department of range wildlife and forestry JEANNE C CHAMBERS university of nevada reno 1000 valley road USDA forest service research university of reno nevada 89512 nevada reno 920 valley road reno nevada 89512 ROBERT C WHITMORE JEFFREY R JOHANSEN division of forestry box 6125 west virginia uni- department of biology john carroll university versityversity Morganmorgantowntown west virginia 2650661252650626500612500 6125 university heights ohio 44118 PAUL C MARSH center for environmental studies arizona state university tempe arizona 85287

editorial boardrichardwboard riehardrichard W baumann chairman zoology H duane smith zoologyzoologoolo claciaclaytonon M white zoology berranjerranjprran T flinders botany and range science william hess botany andan range science all are at brigham young university ex officio editorial board members include clayton S huber dean college of biological and agricultural sciences norman A darais director university publications james R barnes editor great basin naturalist the great basin naturalist founded in 1939 is published quarterlquartellquarquarterlyterlteri by brigham young university unpublished manuscripts that further 0ourur biological understanding oftthelieitelle great basin and surrounding areas in western north america are accepted for publication subscriptions annual subscriptions to the great basin naturalist for 1992 araree 25 for individual subscribers 15 for student and emeritus subscriptions and 40 for institutions outside the united states 30 20 and 45 respectively the price ofsingle issues is 12 all back issues are in print and 11 erint available for sale Aallali matters pertaining to subscriptions back issues or other business shouldsfrintouldouid be directed to the editor great basin naturalist 290 MLBM brigham young university provo UT 84602 sehoscholarlscholarisohoSchoscholarlylarllariiari exchanges libraries or other great yroughbrough organizations interested in obtaining the basin naturalist through a continuing exchange of scholarly publications should contactconeontact the exchange librarian harold B lee library brigham young university provo UT 84602 editorial production staff joanne abel technical editor jan spencer assistant to the editor natalie miles production assistant copyright 0 1992 by brigham young university ISSN 001736140017 3614 official publication date 18 december 1992 129212 92 750 2473 the great basin naturalist PUBLISHED AT PROVO UTAH BY BRIGHAM YOUNG university

ISSN 001736140017 3614

VOLUME 52 SEPTEMBER 1992 no 3

great basin naturalist 523 appp 195 215

PLANT adaptation IN THE GREAT BASIN AND COLORADO PLATEAU

1 jonathan P comstock and james R ehleringerEhlennger

abstractABSIRACTr adaptive features of plants of the great basin are reviewed the combination of cold winters and an and to seinisemiaridand precipitation regime results in the distinguishing features of the vegetation in the great basin and colorado plateau the primary effects of these climatic features arise from how they structure the hydrologic regime water is the most limiting factor to plant growth and water is most reliably available in the early spring after winter recharge of soil moisture this factor determines many characteristics of root morphology growth phenology of roots and shoots and photosynthetic physiology since winters are typically cold enough to suppress growth and drought limits growth during the summer the cool temperatures characteristic ofthe peak growing season are the second most important climatic factor influencing plant habit and performance the combination of several distinct stress periods including low temperature stress in winter and spring and high temperature stress combined with drought in summer appears to have limited plant habit to a greater degree than found in the warm deserts to the south nonetheless cool growing conditions and a more reliable spring growing season result in higher water use efficiency and productivity in the vegetation of the cold desert than in warm deserts with equivalent total rainfall amounts edaphic factors are also important in structuring communities in these regions and halophytic communities dominate many landscapes these halophytic communities of the cold desert share more species in common with warm deserts than do the nonsalinenonsahne communities the colorado plateau differs from the great basin in having greater amounts of summer rainfall in some regions less predictable rainfall sandier soils and streams which drain into river systems ratherrathel than closed basins and salt playasolayas one result of these climatic and edaphic differences is a more important summer growing season on the colorado plateau and a somewhat greater diversification of plant habit phenology and physiology

key words cold desert plant adaptation water stress phenology salinity great basin colorado plateau

several features arising from climate and nevada and increase both to the north and east geology impose severe limitations on plant and to the southeast moving into the colorado growth and activity in the great basin and col- plateau fig 1 table 1 the fraction of annual orado plateau the climate is distinctly conti- precipitation during the hot summer months nental with cold winters and warm often dry june september varies considerably from summers annual precipitation levels are low in 10 20 in northern nevada to 30 40 along the basins ranging from 100 to 300 mm 4 12 the boundary of the cold and mojave deserts in inches and typically increasing with elevation southwestern nevada and southern utah and to 500 mm 20 inches or more in the montane 35 50 throughout much of the colorado pla- zones precipitation levels are lowest along the teau winter precipitation falls primarily as southwestern boundary of the great basin in snow in the great basin and higher elevations

department of biology univumveisityity of utah salt lake city utah 84112

195 196 GREAT BASIN naturalist volume 52

TABITABLE 1 selected climatic data for low elevation sites in different regions of the great basin mojave desert and colorado plateau values aiealeare based on averages for the USU S weather bureau stations indicated the three divisions of the year presented here reflect ecologically relevant units but are unequal in length the five months of october february represent a period of temperature imposed plant dormancy and winter recharge of soil moisture the spring months of march may represent the potential growing period at cool temperatures immediately following winter recharge the summer and early fall from june through september represent a potential warm growing season in areas with sufficient summer rain or access to other moisture sources

total precipitation mean temperature

region map weather elevation annual oct feb marmaymar may jun sep annual oct feb marmaymar may jun sep fig 1 station m mm ICC ICC ICC ICC

northern I1 fortfortbidwellport bidwell 1370 402 63 24 13 90 30 80 173 great basin 2 reno 1340 182 61 24 15 95 33 84 180 3 elbelmeimelkoeiko 1547 230 52 29 19 76 01 71 175 4 snowvilleSnowville 1390 300 43 33 24 74 07 62 184

southern 5 sarcobatus 1225 85 45 22 33 135 64 125 231 gigreateat basin 6 caliente 1342 226 47 24 29 117 41 112 215 7 fillmore 1573 369 44 34 22 110 30 100 217

mojave desert 8 trona 517 102 70 19 11 190 113 184 290 9 beaverdamBeaverdam 570 169 50 23 28 183 110 169 286

colorado 10 Hankshanksvilleville 1313 132 36 19 45 114 21 115 228 plateau 11 grand junction 1478 211 39 25 36 113 24 109 229 12 blanding 1841 336 48 19 33 97 21 87 199 13 tuba city 1504 157 38 21 41 126 48 120 228 14 chaco canyon 1867 220 35 20 45 103 26 94 206

of the colorado plateau which is thought to be internal drainage typical of the great basin in a critical feature ensuring soil moisture recharge this paper we address the salient morphological and a reliable spring growing season west physiological and phenological specializations 1983 caldwell 1985 dobrowolski et al 1990 of native plant species as they relate to survival during the winter period precipitation is gen- and growth under the constraints of these erally in excess of potential evaporation but low potentially stressful limitations we emphasize temperatures do not permit growth or photo- 1 edaphic factors particularly soil salinity and synthesis and exposed plants may experience texture and 2 the climatic regime ensuring a shoot desiccation due to dry winds and frozen fairly dependable deep spring recharge of soil soils nelson and tiemantiernan 1983 strong winds moisture despite the overall aridity as factors can also cause major redistributions ofthe snow molding plant adaptations and producing the pack sometimes reversing the expected unique aspects of the regional plants and vege- increase in precipitation with elevation and tation the majority of the great basin lies at having important consequences to plant distri- moderately high elevations 4000 ft and above butions branson et al 1976 sturges 1977 west and it is occupied by cold desert plant commu- and caldwell 1983 the important growing nities reference to the great basin and its season is the cool spring when the soil profile is environment in this paper will refer to this high recharged from winter precipitation growth is elevation region as distinct from that comer of usually curtailed by drying soils coupled with the mojave desert that occupies the southwest- high temperatures in early or midsummermid summer A ern corner of the great basin geographic unit clear picture of this climatic regime is essential fig I1 our emphasis will be placed on these to any discussion of plant adaptations in the cold desert shrub communities in both the region great basin and the colorado plateau ranging A second major feature affecting plant per- from the topographic low points of the saline formanceformance is the prevalence of saline soils caused playasolayas or canyon bottoms up to the pinyon juni by the combination of low precipitation and the per woodland the lower elevation warmer 199211992 PLANT adaptation 197

in the southern great basin but summer pre- cipitation is substantially greater on the colo- rado plateau table 1 soils and drainage patterns also differ in crucial ways the high- lands of the colorado plateau generally drain into the colorado river system in many areas extensive exposure of marine shales from the chiclechinle mancos and morrison brushy basin formations weather into soils that restrict plant diversity and total cover due to high concentra- tions of nas04 and the formation of clays ththatat 888 do not allow water infiltration potter et al 7 1985 in other areas massive sandstone out- 10 11 crops often dominate the landscape shrubs and 12 trees may root through very shallow rocky soils into natural joints and cracks in the substratum 13 14 deeper soils are generally aeolian deposits forming sands or sandy loamscoams in contrast high elevations of the great basin drain into closed valleys and evaporative sinks this results in great basin greater average salinity in the lowland soils of the great basin with nacl being the most mojave common salt flowers 1934 and a more exten- sive development of halophytic or salt tolerant vegetation soils tend to be deep especially at colorado plateau lower elevations and vary in texture from clay to sandy loamscoams summer active species with fig 1 distribution of the major desert vegetation zones different photosynthetic pathways such as c4ca4 in the great basin and colorado plateau numbers indicate locations of climate stations for which data are presented in grasses and CAM succulents are poorly repre- table 1 most of the mojave desert indicated is geologically sented in much of the great basin but the part of the great basin but due to its lower elevation and combination of increased summer rain sandier armerwarmer temperatures it is climatically distinct from the rest soils and milder winters at the lower elevations of the region ofthe colorado plateau has resulted in a greater expression of phenological diversity and drier mojave desert portion of the great the interactions of edaphic factors and cli- basin will be considered primarily as a point of mate are complex and often subtle in their comparison and for more thorough coverage of effects on plant performance furthermore that region we recommend the reviews by plant distributions are rarely determined by a ehleringer 1985 macmahon 1988 and single factor throughout their geographic range smith and nowak 1990 for the higher mon- even though a single factor may appear to con- tane and alpine zones of the desert mountain trol distribution in the context of a local ecosys- ranges the reader is referred to reviews by tem species specific characteristics generally vasek and thomethorne 1977 and smith and knapp do not impart a narrow requirement for a spe- 1990 we are indebted in our own coverage of cific environment but rather a unique set of the cold desert to other recent reviews includ- ranges of tolerance to a large array of environ- ing caldwell 1974 1985 west 1988 mental parameters in different environmental dobrowolski et al 1990 and smith and contexts different tolerances may be more lim- nowak 1990 iting both abiotic and biotic interactions may be the great basin and the colorado plateau altered and the same set of species may sort out share important climatic features such as overall in different spacialspaciel patterns A further conse- aridity frequent summer droughts and conti- quence of this is that a local combination of nental winters yet they differ in other equally species which we might refer to as a great important features temperatures on the colo- basin plant community represents a region of rado plateau are similar to equivalent elevations overlap in the geographically more extensive 198 GREAT BASIN naturalist volume 52

and generally unique distributions of each com- alienailenAllenallenrotiaallenrolfiaroTia occidentoccidentalisoccidentalistalis and salicornia sppapp ponent species in fact the distributions of spe- and greasewood sarcobatus vermiculatus cies commonly associated in the same great may themselves show zonation due to degrees basin community may be strongly contrasting of tolerance they tend to occur in close prox- outside the great basin this is an essential imity however on the edges of salt playasolayas saline point in attempting to discuss plant adaptations flats with shallow water tables and near saline with the implication of cause and effect seeps over a wide range of elevations tempera- because few species discussed will have a strict tures and seasonal rainfall patterns in both the and exclusive relationship with the environment great basin and southern warm deserts of interest unless we can show local ecotypic macmahon 1988 this relative independence differentiation in the traits discussed we need of distribution from prevailing climate is a spe- to take a broad view of the relationship between cial characteristic of strongly halophytic plant environment and plant characters in a few communities throughout the world and reflects instances a small number ofedaphicofedaphic factors and the overriding importance of such extreme plant characters such as tolerance of very high edaphic conditions species found on better salinity in soil with shallow groundwater seem drained moderately saline soils where ground- to be of overriding importan6eimportance in most cases we water is not found within the rooting zone need to ask what are the common elements of include winterwinterfatfat Ceratceratoidesoides lanata and climate over the diverse ranges of all these spe- shadscaleshadscale atriplex conferticonfertifoliafolia these species cies one such common element which will be are in turn replaced at higher elevations and on emphasized throughout this review is the moister nonsaline soils by big sagebrush anthantearte- importance of reliable winter recharge of soil misia tridentatdtfidentatatridentated rabbitbrushrabbitbrusbrabbitbrush chrysothamnus moisture in an arid to semiarid climate by iden- sp bitterbitterbrushbrush purshia sp and spiny hopbop tifying such elements common and focusing on sage crayiagrayiagrabia spinosa shadscaleShadscale is often found do them we not fully describe the autecology of in areas where soils are notably saline in the but a any species we attempt cogent treatment lower half of the rooting zone but not in the to the great ofplant adaptations basin environ- upper soil layers such a tolerance of moder- and an the ment explanation of unique features ately saline soils seems to control its distribution of its plant communities around playasolayas especially in the wetter eastern portion of the great basin western utah and CLIMATE EDAPHIC FACTORS AND PLANT lower elevations in the warm mojave desert in distribution PATTERNS the more andaridarndannd regions of western and central nevada however the shadshadscalescale community typical zonation patterns observed in spe- occurs widely on nonsaline slopes lower in ele- cies distributions around playasolayas the saline flats vation warmer and drier than those dominated at the bottom of closed drainage basins are by big sagebrush these higher bands of quite dramatic reflecting an overriding effect of shadshadscalescale have been variously interpreted in salinity on plant distribution in the great basin terms of aridity tolerance and climate billings moving out from the basin center is a gradient 1949 or an association with limestone derived of decreasing soil salinity often correlated with calcareous soils beatley 1975 the latter progressively coarser textured soils along this author points out that even on nonsaline soils gradient there are conspicuous species replace- percent cover in the shadscaleshadscale community is ments often resultresuitresultingresultingininginin concentric zones of lower than expected for the level of precipita- contrasting vegetation flowers 1934 vest tion and arguesthatargues that this indicates stress from 1962 in the lowest topographic zone saline edaphic factors thus shadshadscalescale distribution is groundwater may be very near the surface soils most correlated with salinity tolerance in some are very saline fine textured and subject to regions and other edaphic and climatic toleran- occasional flooding and anoxic conditions in ces in other regions this environment the combination of available where the higher elevations of the great moisture with other potentially stressful soil basin come in contact with the lower elevation characteristics seems to be more important than generally drier and warmer mojave desert climatic factors of temperature or seasonal rain- region communities dominated by creosote fall patterns species found here such as desert larrea dentatatritrltridentatatridentate replace sagebrush commu- saltgrasssaltgrass distichlis spicataspicatespicata pickleweedspickleweeds nities on nonsaline slopes and bajadasbajbajamdasadas 199219921 PLANT adaptation 199 shadscaleShadscale can be found both on saline soils at cal and phenologicalphonological traits fountinfoundinfound in the domi- very low elevations in the mojave and as a tran- nant shrubs reflect the primary importance of sitsitionalional band at high elevations between creo such a cool spring growing season sote and sagebrush elements of the cold desert shrub communities adapted to continentalcontinentaladn win- photosynthesis ters and a cool spring growing season can be found throughout the winter rain dominated photosynthetic PATHWAYS pro- mojave desert region as a high elevation band the cess of photosynthesis in plants relies on the on arid mountain ranges they also extend to the acquisition ofoccoa from the atmosphere which southeast at high elevations into the strongly ofcoac02 when coupled with solar energy is transformed bimodal precipitation regime of the colorado into organic molecules to make sugars and pro- plateau and northward at low elevations into vide for plant growth in moist climates plant idaho washington and even british columbia I1 communities often achieve closed canopiescanopies and from bajadasbajamdas of the southern warm moving up 100 cover of the ground surface under these deserts there appears to be no suitable combi- conditions competition for light may be among nation of and at any temperature precipitation the most important plant plant interactions in elevation to floristic elements of the support the deserts total plant cover is much less than cold desert As with alti- precipitation increases 100 and in the great basin closer to 25 tude with total zones equivalent precipitation photosynthesis is not greatly limited by available do not yet have cold winters and are occupied light but rather by water mineral nutrients by warm desert shrub communities grading into needed to synthesize enzymes and maintain chaparral composed of unrelated taxa higher metabolism and maximum canopy leaf area elevations with cold winters have sufficient pre- development cipitation to support forests other elements three biochemical pathways of photosyn- common in shadshadscalescale and mixed shrub commu- thesis have been described in plants that differ of the great such as winterfatfat and nities basin winter in the first chemical reactions associated with budsagebuddage artemisia spinosa are often found the capture of coz from the atmosphere the outside the great basin in cold winter but most common and most fundamental of these largely summer rainfall grasslands pathways is referred to as the c3ca3 pathway from these patterns of community distribu- because the first product of photosynthesis is a tion within the great basin and colorado pla- 3 carbon molecule the other two pathways c4ca4 teau and also from a consideration of the more and CAM are basically modifications of the extensive ranges and affinities ofthe component primary c3ca3 pathway osmond et al 1982 the we can isolate a few key features of the species C4 pathway first product is a 4 carbon mole- environment that are largely responsible for the cule is a morphological and biochemical unique plant features seen in the great basin arrangement that overcomes photorespirationphotorespiration the most obvious features are related to the a process that results in reduced photosynthetic of soil salinity and texture by patterns generated rates in c3ca3 plants the c4ca4 pathway can confer a the overall aridity combined with either internal much higher temperature optimum for photo- drainage basins or the in situ weathering of synthesis and a greater water use efficiency As specific rock types the most important climatic water use efficiency is the ratio of photosyn- variables are slightly more subtle there is thetic carbon gain to transpirationtranspirationalal water loss clearly not a requirement for exclusively winter c4ca4 plants have a metabolic advantage in hot dry rainfall but there is a requirement for at least a environments when soil moisture is available in substantial portion ofthe annual rainfall to come grasslands c4ca4 grasses become dominant at low during a continental winter this permits winter elevations and low latitudes where annual tem- accumulation ofprecipitation to a greater depth peraperaturestures are warmest in interpreting plant in the soil profile than will occur in response to distribution in deserts the seasonal pattern of less predictable summer replenishment of rainfall usually restricts the periods of plant drying soil moisture reserves under an overall growth and the temperature during the rainy andaridarndannd climate winter recharge maintains a pre- season is thus more important than mean annual dictablydictably favorable and dominant spring growing temperature the c4ca4 pathway is often associated season even in many areas of strongly bimodal with summer active species and also with plants rainfall most of the physiological morphologimorphology of saline soils while c3ca3 grasses predominate in 200 GREAT BASIN naturalist volume 52

most of the great basin c4ca4 grasses become temperatures ehleringer and bjorkman 1978 increasingly important as summerrainsummer rain increases mooney et al 1978 comstock and ehleringer to the south and especially on the colorado 1984198419881988 ehleringer 1985 this presumably plateau halophytic plants are often c4ca4 such as reflects the specialization of these great basin saltbush atriplex sppapp and saltgrasssalt grass distichlis shrubs towards utilization of the cool spring sppapp and this may give the plants a competitive growing season positive photosynthetic rates advantage from increased water use efficiency are maintained even when leaf temperatures on saline soils are near freezing which permits photosynthetic the third photosynthetic pathway is CAM activity to begin in the very early spring depuit photosynthesis crassulaceancrassulaceaeCrassulacean acid metabolism and caldwell 1973 caldwell 1985 CAM plants open their stomata at night capture unusually high maximum photosynthetic coa and store it as malate in the cell vacuole rates of 46 mol cog m 2 s 1 have been reported and keep their stomata closed during the day for one irrigated great basin shrub rabbitbrushrabbitbrush osmond et al 1982 the coa is then released chrysothamnus nauseosus davis et al 1985 from the vacuole and used for photosynthesis this species is also unusual in having a deep tap the following day because the stomata are open root that often taps groundwater unusually high only at night when it is cool water loss associ- levels of summer leaf retention branson et al ated with CAM photosynthesis is greatly 1976 and continued photosynthetic activity reduced this pathway is often found in succucuccu throughout the summer drought donovan and lents such as cacti and agave and although ehleringer 1991 all of these characters sug- CAM plants are present in the great basin they gest greater photosynthetic activity during the are a relatively minor component ofthe vegetation warm summer months than is found in most photosynthetic rates ofplants like most met- great basin shrub species abolic processes show a strong temperature SHOOT ACTIVITY AND PHENOLOGY gener- dependence usually photosynthetic rates are ally speaking there is a distinct drought in early reduced at low temperatures because of the summer june july in the great basin cold temperature dependence of enzyme catalyzercatalyzed desert the mojave desert and the sonoran reaction rates increase with temperature until desert all of these deserts have a substantial some maximum value which defines the tem- winter precipitation season but they differ in peraperatureture optimum and then decrease again at the importance of the summer and early fall higher temperatures the temperature optima rainy season july october in supporting a dis- and maximum photosynthetic rates of plants tinctinctivetive period of plant growth and activity show considerable variation and they generally macmahon 1988 the relationship between reflect the growing conditions of their natural climate and plant growing season is complex and environments includes total rainfall seasonal distribution of photosynthetic adaptation in the rainfall and predictability of rainfall in different spring the photosynthetic temperature optima seasons as important variables furthermore in of the dominant shrub species are typically as very andaridanidannd areas the seasonalityseasonably of temperatures low as 15 C 40 F caldwell 1985 correspond- may be as important as that of rainfall in the ing to the prevailing environmental tempera- great basin cold winters allow the moisture tures midday maxima generally less than 20 QC from low levels of precipitation to accumulate As ambient temperatures rise 10 15 C in the in the soil for spring use while hot summer summer there is an upward shift of only 5 10 C temperatures cause rapid evaporation from in the photosynthetic temperature optima of plants and soil high temperatures can prevent most shrubs coupled with a slower decline of wetting ofthe soil profile beyond a few centime- photosynthesis at high temperatures the max- ters depth in response to summer rain even imum photosynthetic rates measured in most when summer rain accounts for a large fraction great basin shrubs under either natural or irri- of the annual total caldwell et al 1977 As total gated conditions range from 14 to 19 moimolLMOI cog annual rainfall decreases the relative impor- 2 1 m s depuitanddepuit and caldwell 1975 caldwell et tance of the cool spring growing season al 1977 evans 1990 these rates are quite increases as the only potential growing period in modest compared to the high maxima of 25 to which available soil moisture approaches the 45 LMOI coagoa m 2 s 1 observed in many warm evaporative demand thornthwaiteThomthwaite 1948 com- desert species adapted to rapid growth at higher stock and ehleringer 1992 finally reliability 199219921 PLANT adaptation 201

of moisture is important to perennials and as expansion or contraction of vegetative phases average total precipitation decreases the vari- and even the omission of reproductive phases ance between years increases ehleringerehleni nger most species initiate growth in early spring 1985 variability ofannnalof annual precipitation is dis- march when maximum daytime temperatures cussed in more detail later in the section on are 5 15 C and while nighttime temperatures annuals and life history diversity summer rain are still freezing delayed initiation of spring is more likely to be concentrated in a few high growth is generally associated with greater intensity storms that may not happen every year summer activity and may be related to an ever- at a given site and may cause more runoffwhenrunoff when green habit a phreatophytic habit or occupa- they do occur the ability of moisture from tion of habitats with greater summer moisture winter rain to accumulate over several months availability from regional rainfall patterns greatly enhances its reliability as a moisture runoff or groundwater higher than average resource thus most plants in the great basin winter and spring precipitation tends to prolong have their primary growing season in the spring vegetative growth and delay reproductive and early summer some species achieve an growth till later in the summer sauer and uresk evergreen canopy by rooting deeply but few 1976 cambellcambellandCambellandbeilandand harris 1977 strong vegeta- species occur that specialize on growth in the tive dormancy may be displayed in midsummermid summer hot summer season branson et al 1976 cald- everett et al 1980 evans 1990 although root well et al 1977 everett et al 1980 ehleringer growth hodgkinson et al 1978 and increased et al 1991 measured the ability of common reproduction west and gastro 1978 evans perennial species in the colorado plateau to use black and link 1991 may still occur in moisture from summer convection storms response to rain at that time in years with they observed that less than half of the water below average spring and summer precipita- uptake by woody perennial species was from tion leaf senescence is accelerated and flower- surface soil layers saturated by summer rains ing may not occur in many species prevalence of summer active species increases the time taken to complete the full annual along the border between higher basins and the growth cycle including both vegetative and southeast mojave desert where summer rain is reproductive stages is strongly related to rooting more extensive and especially on the colorado depth in most of these communities with deep plateau where summer rain is greatest summer rooted species prolonging activity further into temperatures are also lower on the colorado the summer drought pitt and wikeemwakeem 1990 plateau than in the eastern mojave table 1 the exact timing of flowering and fruit set shows allowing more effective use of the moisture considerable variation among different shrub most phenology studies especially from the species some especially those that are more northern areas emphasize the directional drought deciduous flower in late spring and sequential nature of the phenological phases early summer justjust prior to or concurrent with branson et al 1976 sauer and uresk 1976 the onset of summer drought many evergreen cambell and harris 1977 west and gastro shrub species begin flowering in midsummer 1978 pitt and wikeemwikeernwakeem 1990 A single period of ATteartemisiamisiamisla or in the fall Gutiergutierreziagutierrezbarezia and spring vegetative growth is usually followed by chrysothamnus these late flowering species a single period of flowering and reproductive generally do not appear to utilize stored reserves growth many species produce a distinct cohort for flowering but rely on current photosynthe- of ephemeral spring leaves and a later cohort of sis during this least favorable period in the case evergreen leaves daubenmire 1975 miller and of ATteartemisiamisiamisla dentataedentatatrimdentatatridentatafhifrn it has been shown that schultz 1987 for most species multiple carbohydrates used to fill fruits are derived growth episodes associated with intermittent exclusively from the inflorescences themselves spring and summer rainfall events do not occur while photosynthate from vegetative branches in years with unusually heavy august and sep- presumably continues to support root growth tember rains a distinct second period ofvegeta- summer rain during this time period does not tive growth may occur in some species west promote vegetative shoot growth but does and gastro 1978 climatic variations from year increase water use by and the ultimate size of to year do not change the primary importance inflorescences evans 1990 evans black and of spring growth or the order of phenological link 1991 have argued that this pattern of events in particular years they may cause flowering based on residual deep soil moisture 202 GREAT BASIN naturalist volume 52

and the unreliable summer rains may contrib- 1980 fall and winter precipitation is the most ute to competitive dominance within these important in promoting good spring growth of communities the more favorable and much perennials beatley 1974 comstock et al more reliable spring growing season is used for 1988 looking at one yearyearss growth in 19 vegetative growth and competitive exploitation mojave species described an important cohort of the soil volume while reproductive growth is of twigs initiated during the winter period that delayed until the less favorable season and may accounted for most vegetative growth during be successful only in years with adequate the following spring although new leaves were summer precipitation produced in response to summer rain summer most grasses in the northern part of the growth in many of the species was largely a great basin utilize the c3pathway3 pathway and begin further ramification of spring initiated floral growth very early in the spring these species branches in most species summer growth made complete fruit maturation by early or mid- little contribution to perennial stems despite summer often becoming at least partially dor- the preferred winter spring orientation of many mant thereafter on the colorado plateau with shrubs winter recharge is much less effective much greater amounts ofsummer precipitation and reliable in the mojave desert than in the there is a large increase in species number and great basin due to warmer temperatures less cover by c4ca4 grasses such as blubiubluestembluesteinesteinestern winter dormancy may be complete but andropogon and grama bouteloua espe- vigorous growth rarely occurs until tempera- cially at warmer lower elevations and on deep tures rise further in the early spring rapid sandy soils many of these species occur in growth may be triggered by rising spring tetem-in peraturestures or may be delayed until mixed stands with thecathec3the 3 species but delay ini- pera major spring tiation of growth until may or june they can be rains provide sufficient moisture beatley 1974 considered summer active rather than spring ackerman et al 1980 furthermore a shal- lower soil recharge often results active in contrast somec4grassessome 4 grasses such as sand moisture in dropseed sporobolus cryptcryptandrusandrus galleta fluctuating plant water status and multiple grass hilariajamesiiihilariajamesffl and three awn aristida episodes of growth and flowering during the purpurpurpureapurpureanea are widespread in the great basin spring and early fall some great basin species where summer rain is only moderate in long- that also occur in the mojave such as winterwinterfatfat term averages and very inconsistent year to year and shadscaleshadscale commonly show multiple growth and spring growth of these widespread species reproductive episodes per year under that tends to be slightly or moderately delayed com- climate ackerman et al 1980 but not in the great basin west and gastro 1978 the pared to co occurringoccurringc3 3 grasses but they are which this is still able to complete all phenological stages degree to difference due entirely based on the spring moisture recharge they to environmental differences as opposed to eco- typic differentiation does not appear to have show a greater ability than the 3 species to thecathec3 been studied respond to late spring and summer rain with renewed growth everett et al 1980 however which compensates in some years for their later WATER RELATIONS development other c4ca4 grasses in the great basin may be associated with seeps adaptation TO LIMITED WATER stoma- streamstreamsidessides or salt marshes and show a tal pores provide the pathway by which atmo- summer activity patternc4shrubspattern 4 shrubs such as salt- sphericc02diffusesspheric 602coagoa diffuses into the leaf replacing the bush AtTiatriplexplex show similar spring active 602coagoa incorporated into sugar molecules during growth patterns to the c3ca3 shrubs but may show photosynthesis because water vapor is present slightly greater tolerance of summer drought at very high concentrations inside the leaf caldwell et al 1977 opening stomata to capture coa inevitably phenology in the mojave desert shows both results in transpirationtranspirationalal water loss from the similarities and strong contrasts to the great plant thus leaf water content is decreased basin although rainfall is largely bimodal in the resulting in a decrease in plant water potential eastern mojave absolute amounts are very low ar4rP thus plant water status transpiration and the summer is so hot that little growth normally acquisition of water from the soils have a tre- occurs at that time unless more than 25 minmm J1 mendous impact on photosynthetic rates and inch comes in a single storm ackerman et al overall plant growth 199219921 PLANT adaptation 203

many soils in the great basin are fine tex- much of the evidence for this is quite indirect tured which has both advantages and disadvan- nonetheless the osmotic potential of the cyto- tages for plant growth infiltration of water is plasm must also be balanced or suffer dehydra- slower in fine textured soils increasing the like- tion the cytoplasmic soluteskolutes must have the lihood of runoff and reduced spring recharge special property of lowering the osmotic poten- especially on steeper slopes they are also more tial of the cell sap without disrupting enzyme prone to waterloggingwater logging and anoxic conditions function or cellular metabolism and are hence the deep root systems of great basin shrubs are termed compatible soluteskolutes wyn jones and very sensitive to anoxia and this can be a strong gorham 1986 the use of specific molecules determining factor in species distributions shows considerable variation across species lunt et al 1973 groeneveld and crowley apparently due to both species specific varia- 1988 unusually wet years may even cause root tions in cell metabolism and taxonomic relation- dieback especially at depth once water enters ships some frequently encountered molecules the soil profile the extremely high surface areas thought to function in this manner include of fine textured soils with high clay and silt amino acids such as proline and also special content give them a much higher water holding sugar alcohols some plants appear to generate capacity than that found in sandy coarse tex low osmotic potentials by actively manufactur- turedaured soils much of this water is tightly bound ing larger quantities of dissolved organic mole- to the enormous surface area of the small cules per cell in response to water stress a particles however and is released only at very process referred to as osmotic adjustment negative matric potentials plants must be able this process may be costly however requiring to tolerate low tissue water potentials to make the investment of large amounts of materials use of it new sosoluteskoluteslutes at a time when water stress is As soil water is depleted during summer largely inhibiting photosynthetic activity an plants reduce water consumption by closing sto- alternative method seems to involve dramatic mata depuit and caldwell 1975 cambell and changes in cell water volume several desert hartisharrishardis 1977 caldwell 1985 miller 1988 and species have been observed to reduce cell water reducing total canopy leaf area to a minimum volume by as much as 80 without wilting when branson et al 1976 evergreen species shed subjected to extremely low soil water potentials only a portion of the total canopy however moore et al 1972 meinzer et al 1988 evans maintaining the youngest leaf cohorts through- et al 1991 this allowed the leaf cells to have out the drought miller and schulz 1987 sufficiently low osmotic potentials for water although physiological activity is greatly uptake even though solute content per cell was reduced by water stress evergreensevergreens such as actually reduced although total soluteskolutes per leaf sagebrush can still have positive photosynthetic and presumably per cell decreased the rela- rates at leaf water potentials as low as 50 bars tive concentrations of specific soluteskolutes changed evans 1990 and may survive even greater evans et al 1991 such that compatible levels of stress in contrast crop plants can soluteskolutes made larger contributions to the osmotic rarely survive prolonged TP ofless than 15 bars potential under stress thus tolerance of low remaining functional at low water potentials leaf water potentials was achieved by a combi- requires the concentration of soluteskolutes in the cell nation of anatomical and physiological special- sap and this appears to be accomplished by izationsizations the anatomical mechanisms involved several mechanisms in many mesic species in this magnitude of volume reduction and the increases in organic soluteskolutes may account for implied cell elasticity have not been studied but most ofthe change in osmotic potential in other the process has been shown to be fully reversible species and especially those that experience soil texture is also an important factor in very low leaf water potentials a large fraction of determining the ability of plant communities in the soluteskolutes is acquired by the uptake of inor- a cold winter climate to respond to summer ganic ions such as K wyn jones and gorham rain in areas with moderate levels of precipita- 1986 high concentrations of inorganic ions tion sandy soils because of their low water may be toxic to enzyme metabolism however holding capacity usually hold a sparser more and they are widely thought to be sequestered drought adapted vegetation than finer textured largely in the central vacuole which accounts ones in warm andaridarld areas however what has for 90 of the total cell volume even though been called the reverse texture effect results 204 GREAT BASIN naturalist volume 52

in the more lush vegetation occurring in the well et al 1977 sturges 1977 of the total root coarse textured soils this occurs because the biomass the very large annual root invest- high water holding capacity of fine textured ments therefore are not primarily related to soils allows them to hold all the moisture large storage compartments but to the turnover derived from a single rainfall event in the upper- of fine roots and root respiration necessary for most layers of the soil profile where it is highly the acquisition of water and mineral nutrients subject to direct evaporation from the soil the the fine root network thoroughly permeates same amount of rainfall entering a sandy soil the soil volume root densities are generally precisely because of its low water holding quite high throughout the upper 050.5 meters of capacity will penetrate to a much greater depth the profile but depth of maximum root devel- it is also less likely to return to the drying surface opment varies with depth of spring soil mois by capillary action less of the rain will evapo- ture recharge species and lateral distance from rate from the soil surface and a greater fraction the trunk or crown A particularly high density of it will await extraction and use by plants this in the first 0100.1oi 1 m may often occur especially inverse texture effect further restricts the effec- immediately under the shrub canopy branson tivenesstiveness of summer rains in the fine soils of the 1976 caldwell et al 1977 dobrowolski et al great basin the effect is less operative in 1990 alternatively maximal density may occur respect to winter precipitation in the great at depths between 020.2 m and 050.5os m sturges basin however because ofthe gradual recharge 1980 and sometimes a second zone ofhigh root of the soil profile to considerable depth under density is reported at depths of 080.8os m conditions where surface evaporation is mini- daubenmire 1975 to 151.5 m reynolds and mized by cold temperatures the combination fraley 1989 regardless ofthe precise depth of of much sandier soils and greater amounts of maximum rooting shrub root density is usually summer rainfall in the region of the colorado high throughout the upper 050.5 m and then plateau is largely responsible for the major flo- tapers off but may continue at reduced densi- ristic and ecological differences between the ties to much greater depth the radius of lateral two regions at lower elevations on the south- spread is usually much greater for roots 1 2 m east edge of the plateau shrub dominated than for canopiescanopies oiol010.1 050.5 m percent plant desert scrub may be replaced by grassland dom- cover aboveground is often in the neighborhood inated by a mix of spring active c3ca3 and summer of 25 with 75 bare ground but belowgroundbelowground active c4ca4 species the interspacesinterspaces are filled with roots throughout ROOTING DEPTH morphology AND PHE- the profile and root systems of adjacent plants NOLOGY one of the unique and ecologically will overlap the perennial grasses that are most important features of the great basin potentially dominantcodominantco with shrubs in many of shrub communities is not apparent to above- these communities such as wheatgrasswheatgrass ground observers this is the investment of the agropyron sp wildrye elymus sp vast majority of plant resources in the growth squirreltail sitanionSitanion hihisthisttixhistriohistrixstrixtix indian ricegrassricegrass maintenance and turnover of an enormous root oryzopsis hymenoideshymenoides and galleta grass system the dominant shrubs ofthe great basin hilafiajamesiihilariajamesii generally have somewhat shal usually root to the full depth ofthe winter spring lower root systems than the shrubs branson et soil moisture recharge depending on soil tex- al 1976 reynolds and fraley 1989 dobro- ture slope aspect and elevation this is gener- wolski et al 1990 root densities of grasses are ally between 101.0iolo and 303.0 m dobrowolski et al often as high as or higher than those of shrubs 1990 although this represents a wide range of in the upper 050.5 m but taper off more rapidly absolute depths many of the qualitative pat- such that shrubs usually have greater density at terns of rooting behavior are similar on most of depth and greater maximum penetration these sites ratios of rootshootroot shoot standing bio- the moisture resource supporting the great- mass range from 4 to 7 and estimates of est amount of plant growth is usually the water rootshootroot shoot annual carbon investment are as high stored in the soil profile during the winter this as 353.5 most of the shrubs have a flexible gen- profile usually has a positive balance with more eralizederalized root system with development of both water being added by precipitation than is with- deep taproots and literalslateralslaterals near the surface drawn by evapotranspiration between october moreover it is the category of fine roots 303.0 and march As temperatures warm in march mm in diameter that constitutes 50 95 cald evergreen species may begin drawing on this 199219921 PLANT adaptation 205 moisture reserve and most species begin active only in the spring are shallower rooted and growth in march or april due to both plant lateral root spread of the evergreen species is water use and soil surface evaporation soil greatest in the shallower soil layers the more moisture is depleted first in the shallow soil dominant perennials usually use more water layers As the upper layers dry plants withdraw over the whole seasonlyseasonbyseaseasonsonbyby tapping deeper soil moisture from successively deeper soil layers a layers cline et al 1977 sturges 1980 and are process that continues in evergreen species characterized by higher water use efficiencies throughout the summer until soil moisture is delucia and schlesinger 1990 smedley et al depleted throughout the profile this progres- 1991 sion of seasonal water use beginning in superfi- shrubs appear to be better than grasses at cial layers and proceeding to deeper soil layers withdrawing water from deep soil layers for is facilitated by the pattern of fine root growth several reasons in shallow soils underlain by which shows a similar spatial and temporal pat- fractured or porous bedrock the flexible mul- tern fernandez and caldwell 1975 caldwell tiple taproot structure of a shrub root system 1976 root growth generally precedes shoot may be better suited than the more diffuse growth in the early spring and continues fibrous root system of grasses for following throughout the summer in evergreen species chinks and cleavage planes to indeterminate which may appear quiescent aboveground in depths this should allow shrubs to better cap- annual budgets of undisturbed communities italize on deep localized pockets of moisture soil moisture withdrawal almost always unfortunately such dynamics are rarely studied approaches measured precipitation over a wideaardeaa4de quantitatively because of the difficulty of mea- range of annual fluctuations in total precipita- suring soil moisture budgets or root distribu- tion and very little moisture is lost to runoff or tions under such conditions but they are deep drainage beneath the rooting zone implicated by plant distribution patterns in daubenmire 1975 caldwell et al 1977 many areas indeed despite the visual austerity cambell and harris 1977 sturges 1977 calcu- of such habitats rooting into major joints and lations ofthe portion evapotranspirationofevapotranspirationof actu- cracks in bedrock outcrops can create such a ally used by plants in transpiration are quite high favorable microlitemicrositemicrosite by concentration of runoff for a desert environment with low percent in localized areas that elevational limits of tree cover they range from 50 to 75 caldwell et and shrub distributions may be substantially al 1977 cambell and harris 1977 sturges 1977 lowered as would be expected along riparian competition for soil moisture is not neces- corridors or other unusually mesic habitats sarily limited to any single season plant water loope 1977 even in deep soils shrubs tend stress is highest in late summer and survival is to have deeper root systems than grasses but in most likely to be influenced at this time espe- addition to this shrubs may be more efficient cially if one plant can deplete residual soil mois- than grasses at extracting deep water shrubs are ture below the tolerance range of another this sometimes able to draw on deep soil moisture has been shown in several cases with regard to to a greater extent than would be predicted from seedling establishment harris 1977 delucia root biomass distribution alone sturges 1980 and schlesinger 1990 henbergerreichenbergerReic and pyke and this is due in part to an intriguing phenom- 1990 growth and productivity are more likely enon reported by richards and caldwell 1987 to be affected in the spring and early summer and named by them hydraulic lift during the growing season this is because most of the late spring and early summer most of the year s water resource is already stored in the soil remaining soil moisture is present in deeper soil in early spring and all plants are drawing on a layers where rooting density may be relatively dwindling reserve which ultimately determines low due to low densities deep roots alone may growing season length competitive ability is be unable to absorb water as quickly as it is lost often found to be associated with an ability to by the transpiring shoot during the night water begin using the limiting water resource earlier is actually redistributed within the soil flowing in the spring eissenstat and caldwell 1988 from deep soil layers through the roots and back miller 1988 but spatial partitioning is also out into shallower soil layers this is the phe- important competition may be most intense in nomenon named hydraulic lift and the shallower soil layers because grasses and advantage to the plant is thought to be a reduc- drought deciduous shrubs which are active tion in the rooting density necessary to fully 206 GREAT BASIN naturalist volume 52

exploit the resources of the deepest soil layers evaporative pans where salinisalinitiesties reach extreme during the day rates of water movement from levels and salts precipitate forming a surface the soil into the roots can be limiting to shoot crust the water table near these evaporative activity moistening the upper soil layers by noc- pans is often at or very near the surface through- turnal hydraulic lift increases the root surface out the year but ifthere is no groundwater flow area for absorption during the periods of high out of the basin it too is often quite saline transpiration daytime water use can be sup- dobrowolski et al 1990 salinitiesSalinities are lowest ported by the entire root system and not just by on slopes and at higher elevations where precip- the low density deep roots caldwell and rich- itation exceeds evaporative loss and they ards 1989 increase on more level terrain and in lower ele the root systems of great basin shrubs and vation basins where evaporation exceeds pre- mojave desert shrubs differ strongly in several cipitation salinitiesSalini ties may also be higher in areas ways 1 mojave desert shrubs often have max- where wind bobebomeborne materials are transported imal rooting densities at soil depths of 010.1oi 030.3 from saline playasolayas to surrounding slopes young in and maximum root penetration of 040.4 050.5 m and evans 1986 these patterns of soil salinity wallace et al 1980 these shallower roots are are important in determining plant distribu- due to lower rainfall and warmer winter temper- tions with more specialized salt tolerant spe- atures resulting in shallower wetting fronts in cies halophytes replacing less tolerant species the soil and the development of shallow calichebaliche repeatedly along gradients ofincreasing salinity layers 2 great basin species have more roots in general species diversity is low on saline in the shallowistshallowestshallowest 010.1oi m soil layer wallace et al soils the vast majority of tolerant shrub species 1980 dobrowolski et al 1990 differences in in our deserts and all the shrubs specifically soil temperatures have been used to explain mentioned in this section belong to a single these patterns but the link between cause and plant family the chenopodiaceae goosefoot effect is less obvious in this case and conjec- family most other important taxa in the saline tures should be viewed cautiously much hotter communities are grasses soil temperatures in the mojave may be detri- in the most extreme case of hypersalinehypersaline salt mental to roots in the uppermost few centime- flats and pans there may be standing water in ters during summer and the rapidly drying soil the wet season with saturating salt concentra- surface may be too ephemeral a moisture tions under such conditions only microflora resource to favor large investments in roots in consisting of a few species of photosynthetic contrast snowbeltsnowmeltsnowmelt and cooler spring tempera- flagellates cyanobacteria and halobacteriahalobacteria are tures may result in less rapid evaporation from commonly found the halobacteriahalobacteria appear to be the soil surface in the great basin and thus favor unique in having adapted in an obligate manner more rooting by perennials in that zone 3 to the high salinitiessalinities of these environments because of the greater soil volume exploited as they not only tolerate but require high well as the high root density of great basin cytoplasmic salinitiessalinities for membrane stability species their ratios of rootshoot biomass are and proper enzymatic function brown 1982 about twice that of mojave desert species in strong contrast to this algae and all higher bamberg et al 1980 dobrowolski et al 1990 plants growing in hyper saline environments adaptation TO SALINITY when annual show severe inhibition of enzyme function at precipitation levels are much lower than poten- high salinity and they must compartmentalize tial evaporation salts are not leachedbeached to any salt sensitive metabolic processes in cellular great depth and they can accumulate within the regions of low ionic strength munns et al 1982 root zone this is especially important in associ- the best definition of a halophyte is simply ation with particular bedrock outcrops such as a plant tolerant of soil salinitiessalinities that would the mancos and chinie shales which release reduce the growth of unspecialized species this high concentrations of salts during weathering is somewhat circular and that reflects our lim- potter et al 1985 precipitation increases with ited understanding of how halophytes do what elevation and following major storms or spring they do halophytes are more likely to use na snowbeltsnowsnowmeltmelt there may be surface runoff and in their tissues for osmotic adjustment while recharge groundwaterofgroundwaterof systems that transport glycophytes are more likely to have high K water from high elevations into closed basins hellebust 1976 flowers et al 1977 but there streams in the great basin usually terminate in are numerous exceptions other differences 199211992 PLANT adaptation 207 may be more quantitative than qualitative var- for osmotic adjustment and positive photosyn- ious aspects of mineral nutrition in halophytes thetic rates can be measured in the leaves of are less sensitive to high soil salinitiessalinities but many halophytes at leaf water potentials as low unique mechanisms to achieve this tolerance as 90 to 120 bars caldwell 1974 well below have rarely been identified it is widely held that the range that would result in death of even the ability to compartmentalize salts and restrict desert adapted glycophytes this is accom- high na concentrations to the vacuole is of plished in part by transforming the available crucial importance caldwell 1974 flowers et salts in the environment into a resource and al 1977 briens and larherdarher 1982 this conclu- using them for osmotica in plant tissues moore sion is based primarily on indirect evidence of et al 1972 bennertbeimert and schmidt 1984 many low enzyme tolerance of salinity however halophytes actually show stimulation of growth rather than direct measurements of actual salt rates at moderate environmental salt levels compartmentalization munns et al 1982 halophytes too must deal with the problem jefferies and rudmikrudmin 1984 of salt buildup in the leaves and they do so by a halophytes differ in which ions reach high wide variety of processes there is a great deal tissue concentrations when all plants are grown ofinterspecific variation in which methods are on the same medium caldwell 1974 some used all the methods appear to incur substan- will concentrate cl for instance while others tial energetic costs associated with maintaining concentrate s04 2 these differences do not high ion concentration gradients across key necessarily determine plant distributions such membranes kramer 1983 exclusion of salts at as occurrence in soils dominated by nacl versus the root is possible this is the method most nas04 but rather seem to reflect different reg- employed by winterwinwinterfatterfatterratfatfht Ceratceratoidesceratoideslanataoidesoldes lanata salt- ulatory specializations in plant metabolism bush atriplex sppapp has specialized hair blad moore et al 1972 A strong requirement for a ders on the leaf surface into which excess salts unique composition of soil salts is the exception are actively pumped the hairs eventually rup- rather than the rule and the most important ture excreting the salts to the outside other effect of soil salinity seems to be a disruption of plants may transport salts back to the root via plant water relations from low soil osmotic the phloem many plants exhibit increased leaf potentials rather than toxic effects of specific succulence when grgrownown under high salinity and ions halophytes tolerate these conditions by this increase in cell volume can create a sink for having better regulatory control over ionlon move- salts within the leaf without raising salt concen- ment within the plant ion compartmentaliza- trattrationsions or further lowering leafosmotic potential tion at both tissue and subcellularsubcellular levels and in strong contrast to the evident importance better homeostasis of other aspects of mineral of temperature and rainfall pattern in favoring nutrition in the presence of very high na c3ca3 versus c4ca4 grasses c4ca4 shrubs tend to belong salinity poses three major problems for to edaphic communities associated with saline higher plants first salts in the soil solution soils the same species may occur in both warm contribute an osmotic potential depressing the and cold deserts and in areas with both winter soil water potential and this may be aggravated and summer rainfall patterns this is an intri- as salts become concentrated with soil drying guing difference but the physiological basis even when substantial moisture is present linking C 4 shrubs with high salinity is less well plant tissues must endure very low water poten- understood than the tradeoffs associated with tials to take it up and this requires a specialized temperature and controlling c3ca3 and c4ca4 grass metabolism second any salts entering the plant distributions species number and percent with the transpiration stream will be left behind cover by shrubs such as saltbush atilAtTlatriplexplex sppapp in the leaf intercellular fluids as water evapo- and inkweed sueda sppapp which possess the c4ca4 rates from the leaf this can result in salt pathway usually increase dramatically with buildup in the intercellular solution causing increasing salinity on well drained soils in water movement out of the cells and leading to marshy habitats or soils with a shallow saline cellular dehydration third salts entering the water table however halophytic c3ca3 species such cytoplasm in high concentration willvaslivadil disrupt as picklepickleweedsweeds allenrotlaallenrolfia sppapp and salicomiaSalicomia enzyme function halophytes are able to deal sppapp and greasewood sarcobatus ver with all of these factors over a wide range of soil micumiculoidesloides regain dominance it has been sug- salinitiessalinities halophytes show a greater capacity gested that higher water use efficiency by C 4 208 GREAT BASIN naturalist volume 52

species may be advantageous on saline soils to NUTRIENT RELATIONS help avoid salt buildup in the leaf tissues how- ever it has not been shown that transpiratiotranspirationn acquisition OF MINERAL NUTRIENTS rate is an important factor controlling sasaltsaitt apart from the very high elevation montane buildup in the leaves of halophytes when com- zones water appears to be the most limiting pared with other regulatory mechanisms resource in the great basin and colorado pla- osmond et al 1982 nor does this hypothesis teau communities productivity is usually well explain the dominance of cg species in wet correlated with yearly fluctuations in precipita- saline soils in the greasewood and pickleweed tion and spring moisture recharge over a wide communities soil salinitiessalini are extreme ties but range of values daubenmire 1975 kindschy soils remain wet throughout the season growing 1982 and competitive success has more often or else groundwater is availablearithinavailable within the root- been associated with soil water use patterns ing zone detling 1969 hesla 1984 As a con- than nutrient budgets nonetheless addition of water do sequence plant potentials not reach mineral fertilizer sometimes does result in the extreme low values of the saltbush commu- modest over increases in productivity and studies nities a wide range of soil salinitiessalinities plants have shown strong effects of neighboring plants such as greasewood appear to draw on readily on nutrient uptake rates caldwell et al 1987 available deep soil moisture and high leaf con- these dynamics have been less studied than ductancesductances are maintained throughout the have plant water budgets and broad ecological 1984 summer hesla romo and hafercamp relationships are just now being worked out in 1989 the highest whole plant water use rates detail nutrient acquisition has been shown to may occur in the presence of high soil salinity in be a major factor determining community com- midsummermid summer hesla 1984 the communities in position only in very special habitats such as which 04 shrubs are most prevalent have large sand dunes bowers 1982 or unusual bed- summer stress from both high soil salinity and rock delucia and schlesinger 1990 mid soil midsummersummer water depletion combined microphytic CRUSTS throughout the reach lower these species much plant water great basin and colorado plateau it is common potentials during summer than either nonsaline for the exposed soil surface to be covered by a communities or wet saline communities the complex community of nonvascular plants role of c4ca4 photosynthesis in tolerating these including dozens of species of algae libenslichens conditions remains to be determined but it and mosses west 1990 these organisms often could be related to avoiding excessively low leaf form a biotic crust in the upper few centimeters water potentials either by 1 retarding soil of the soil and when undisturbed may result in moisture depletion which both lowers the soil a very convoluted microtopographymicrotopography of the sur- matrix potential and concentrates soil salts or face while a detailed discussion of the 2 avoiding exacerbation of low soil water microphytic crusts is beyond the scope of this potentials due to high flux rates and large water review it is important to realize that percent potential gradients between the leaf and root cover by such crusts often exceeds that of the water movement in the xylem occurs under vascular plants and their contribution to total tension and anatomical features that avoid cav- ecosystem productivity is considerable perhaps itation in the xylem at extremely low water most important to co occurring vascular plants potentials usually reduce the hydraulic conduc- are the nutrient inputs to the soil by nitrogen tivity of the xylem per unit cross sectional area fixing crust organisms cyanobacteria and davis et al 1990 sperry and tyree 1990 low lichens these inputs will be particularly specific conductivity of the xylem will in turn important in the cold desert where few vascular predispose the plant system to large water plants form symbiotic relationships with nitro potential gradients between roots and shoots gen fixing bacteria causing an eveneyen greater depression of leaf water NURSE PLANTS AND FERTILE ISLANDS in potential this problem could be ameliorated many desert areas including both the mojave either by increased cross sectional area of the and the great basin establishment of new indi- xylem by increased allocation to wood growth viduals may occur preferentially under the exist or by features such as c4ca4 photosynthesis that ing canopiescanopies of already established individuals reduce the flux rate of water associated with these previously established individuals may photosynthetic activity under warm conditions then be referred to as nurse plants preferential 199219921 PLANT adaptation 209 establishment under nurse plants may occur in photosynthesis in the cool spring and earlier spite of the fact that 75 or more of the ground growth on limited soil moisture reserves may be area may be bare interspacesinterspaces between plant competitively advantageous occupying warm canopiescanopies the phenomenon can be important in micrositesmicrosites maybe favored substantial increases both steady state community dynamics and also in air temperature and reductions in wind speed successional patterns following disturbance will exist in the lowest meter next to the ground wallace and romney 1980 everett and ward and especially in the lowest decimeter low 1984 two somewhat distinct factors contribute cushion plants or low dense shrub canopiescanopies to the nurse plant phenomenon the first has to should have warmer spring leaftemperatures by do the with beneficial effects of partial shading virtue of being short and by virtue ofleafing out and reduced wind under existing canopiescanopies first in a dense clump of old dead leaves and resulting in cooler temperatures and possibly twigs smith et al 1983 wilson et al 1987 this moister soil conditions in the surface layers advantage may be partially offset by overly high these interactions depend directly on the pres- temperatures in summer for species remaining ence of the nurse plant in creating a favorable active all summer stature is also likely to affect microlitemicrositemicrosite and have been studied with particular aeolian deposit of materials under the shrub reference to pinyon and juniper establishment canopiescanopies wood et al 1978 young and evans in the great basin A second factor involves the 1986 snow accumulation branson et al 1981 creation of fertile islands by the prolonged occu- west and caldwell 1983 and the likelihood of pation of the same microlitemicrositemicrosite by many genera- winter desiccation under cold vindywindy conditions tions of plants this can make the fertile island a nelson and tiemantiernan 1983 all of these could preferred site even if the previous occupant is be important factors but few detailed studies already deceased this microlitemicmicrositerosite improvement have been done occurs due to preferential litter accumulation having considered the relationships of the and more extensive root growth directly under dominant plant habits and phenolphenoiphenologiespherrologiesogiesogles to cli- a plant canopy and deposition of aeolian mate- mate it is perhaps instructive to consider why rials under reduced wind speeds in plant cano some of the other famous desert life forms are soils pies in time under existing plants may so poorly represented in this region three life come to be slightly raised above the interspace forms which are prominent features ofthe warm level have a fighterhighter loamierfoamier texture lighter higher desert but inconspicuous elements of the cold matter content and better soil structure organic desert are 1 large CAM succulents eg cacti less surface and compaction better aeration and agave 2 opportunistic drought decidudecidua water infiltration more rapid andor higher ous shrubs specialized for rapid leaf flushing levels available mineral than of nutrients and 3 annuals definitive work explaining the immediately adjacent interspace soil vest 1962 structural uniformity of the vegetation is not wood et al 1978 romney et al 1980 hesla available but the environment is well enough 1984 west 1989 dobrowolski et al 1990 understood to identify at least some of the likely effects of nurse plants and indirect direct causes effects of fertileoeloeifey tiletlletiie islands should complement and CAM most ofthe large CAM reinforce each other in maintaining selective succulents succulents are not tolerant of freezing spacielspacial patterns of seedling establishment sur- temper- atures and most extant species would be face soil under halophytes may also show excluded from the great by this factor increased salinity richard and cline 1965 due basin alone are however a sufficient number to excretion of excess salts by the canopy or there of which have cold translocation and re excretion from the roots species adapted to tolerate temperatures that we are justified in asking why they have not undergone more adaptive radia- DIVERSITY OF GROWTH FORMS tion or claimed a more prominent role in these communities the most important factor limit- one ofthe striking features ofthe cold desert ing this life form is probably the importance of vegetation is the uniformly low stature of the the cool spring growing season CAM succucuccu vegetation this is undoubtedly due to several lents generally 1 allocate very little biomass to factors and few studies have specifically root rootshootroot shoot ca 010.1oi 2 are shallow rooted addressed the role of plant stature in these com- 3 store moderate sized compared to soil munitiesmunities since low temperatures may limit water holding capacity water reserves inside 210 GREAT BASIN naturalist volume 52 their tissues when water is available in the sur- opportunistic DROUGHT DECIDDECIDUOUSUOU S face soil layers and 4 use their stored water in MULTIPLE LEAF FLUSHING SPECIES this photosynthesis with unparalleled water use effi- habit like that of the succulents is favored by ciency by opening their stomata only at night 1 intermittent rainfall wetting only shallower when temperatures are cool nobel 1988 they soil layers and 2 warm temperatures allowing are favored by 1 very warm days 30 40 C for rapid leaf expansion in response to renewed which allow them to have higher photosynthetic soil moisture again these requirements are not rates and cause competing species to have very well met in the great basin the primary mois- low water use efficiencies 2 large diurnal tem- ture resource is a single deep recharge in the peratureperature fluctuations allowing for cool nights winter most shrub species are deep rooted and 10 20 C which allow them to have high rates rather than experiencing vacillating water avail- of coagoa uptake with high water use efficiency ability they have active root growth shifting to and 3 intermittent rainfall which only wets the deeper and deeper soil layers during the season upper soil layers so that the limitations of their thus producing a gradual and continuous shallow roots and water hoarding strategy are change in plant water status this allows many compensated for by the ephemeral nature of the spring active shrubs to remain partially ever- soil water resource these conditions are some- green throughout the summer and in regions what poorly met in the cold desert the impor- where it occurs they are able to make rapid use tant water resource is one of deep soil recharge of any moisture available from summer precip- that favors deep rooted species and confers itation without the need for renewed leaf pro- much less advantage on internal water hoarding duction the only shrub reported to have multiple leaf flushes in response to late spring freezing tolerance in CAM succulents appears great to be associated with low tissue water contents or summer rain in the basin is the dimin- utive and shallow rooted ATtemisiamisla cens and this may inhibit uptake of water when it is artemisia spinescentspinescensspines everett et al 1980 some found the plentiful in the surface layers in the thermally species in great basin are reported to have multiple vacillating early spring littlejohn and williams growth cyclesyearcycles year where they occur in the 1983 furthermore water use efficiencies 0of mojave ackerman et al 1980 c3ca3 and c4ca4 species are quite high in the cool ANNUALS AND LIFE HISTORY DIVERSITY spring spectacular wildflower shows displayed in nonetheless even moderate amounts of the favorable years in the mojave desert do not in southern and summer rain the eastern por- occur in the cold desert of the great basin tions of the great basin result in numerous ludwig et al 1988 annual species are few in of the of species cacti due to open nature the number and except in early succession after of have a understory many these species large fire in woodlands or on very disturbed sites they elevaelevational1 and often elevationaelevationstionatlona range they are more rarely constitute a major fraction of total com- the common in the pinyon juniper or even mon munity biomass this is undoubtedly related to tane zone than on the desert piedmont slopes several complex factors but various aspects of almost all of these cacti are small usually 5 20 precipitation patterns are likely to be among the pediocactus cm high with a small globose eg Pediocactus most important to begin with the paucity of simpsosimpsoniinii prostrate eg opuntia poly summer rain in some parts of the great basin cantha or low caespitose habit eg may largely eliminate an entire class of c4ca echinocereus tfiglochidiatustriglochidiatus this allows them summer annuals important in the floras of other to take advantage oftheodtheof the warmer daytime tem regions including the colorado plateau other peraperaturestures near the ground in the spring and aspects than seasonalityseasonably are also crucial how- facilitates an insulating snowcoversnowcover during the ever very low means of annual precipitation are coldesta4ntercoldest winter periods the number of and total commonly associated with large annual floras cover by cacti increase considerably with but correlated with low mean precipitation is increased summer rainfall on the colorado pla- high year to year variation in precipitation teau but only in the eastern mojave with both which some authors have argued is equally summer rain and warm spring temperatures do important the coefficient of variation CV in we find the larger barrel cactus eg ferocactus precipitation shows a relationship to mean pre- acanthacanthoidesacanthoiclesoidesoicles and tall shrubby challaschollas eg cipitation in the great basin and colorado pla- opuntia acanthocarpa teau fig 2 very similar to that found in warm 1992 PLANT adaptation 211

07 loo100 CZ Z 066 80 r2ra 95 0 2 z3za i r 57 0 ZC Q 05 0 0 0 015 00 60 0 0 a 0 2 04 a D 05 0 00 0 0 15 0 8 0 0 40 a 03 cb 0 P 0waw 0 0 00 000 5u.5 0 0 0 0 0 0 20 02 0 0 000ooo 0 0 0 3 C0 0 0 01 0 0 01 02 03 04 05 06 00 CV for mean annual 100 200 300 400 500 precipitation mean annual precipitation mm fig 3 the relationship between leliabilityreliability of annual fig 2 the relationship between mean precipitation and precipitation and life history strategy of herbaceous plants the variability of rainfall between years as measured by the the site with greatest representation of annuals isis death coefficient of variationvanation in annual precipitation the data valley in the mojave desert the second highest is canyon include points scattered throughout the great basin in utalutaiutah lands in the colorado plateau of southeastern utah and the and nevada and the colorado plateau in utah and arizona other three sites are great basin cold desert or shrub the line shown is the least squares best fit for the data CV steppe data were collected by kimkirn haihalhalperharperper and previously 127 04030 403 s logmeanlog mean annual precipitation mm n published in schaffer and gadgil 1975 69 sites p 001

arizona monsoon and the region where the deserts ehleringer 1985 although mean pre- fraction of summer rain increases substantially cipitation has the greatest single effect there moving southward this zone also has some of are additionally important geographic influ- the most andaridarndannd sites of the entire region located ences on the CV of precipitation which are along the transition to the mojave desert in independent of mean precipitation A multiple southern nevada and the canyon country of regression of the CV of precipitation on southeastern utah and these sites can be logmeanlog mean annual precipitation latitude and expected to have the highest variability due to elevation in the great basin has an r2ra of si8181.81 and both low mean rainfall and geographic position indicates that each variable in the model is correlated with regional weather patterns highly significant p ooi001001.001 or better for a because the great basin and colorado plateau given mean precipitation the CV increases with are only semiarid the CV of annual precipita- decreasing altitude in the great basin but an tion is not usually as high as in many ofthe more independent effect of elevation was not signifi- andaridanidannd warm deserts beatley 1975 ehleringer cant in the colorado plateau the CV also 1985 but particular sites may be both andaridarld and increases from north to south in the great basin highly unpredictable and increases from south to north in the colo- harper cited in schaffer and gadgil 1975 plateau rado which results in a latitudinal band found that the prevalence of annuals was posi- of greatest annual variability running through tively associated with the CV in annual precipi- southern nevada and utah this band is related tation for five sites located in the great basin to two major aspects ofregional climate moving colorado plateau and mojave desert fig 3 southward in the great basin temperatures the largest annual populations occurred in gradually increase favoring moister air masses death valley mojave followed by canyon and more intense storms but sites are more lands colorado plateau in southeastern utah removed from the most common winter storm one interpretation of this relationship is that tracks and the number of rainy days per year high variability in total precipitation between decreases houghton 1969 moving northward years may be associated with high rates of mor- from arizona and new mexico the southern tality and therefore favor early reproduction and nevada and utah band of highest precipitation an annual habit schaffer and gadgil 1975 variability also corresponds to the northernmost many desert annuals are facultatively perennial extent of summer storms associated with the in better than average years and some have 212 GREAT BASIN naturalist volume 52 perennial races or sister species ehleringer between short and long lived perennials may 1985 the dynamics and distributions of these be influenced by very similar climatic parameters closely related annual and perennial taxa should sometimes operating over different time scales receive further study in regard to their expected other factors that may be important in the life span reproductive output and relationships ecology of great basin annuals include the to climatic predictability another perspective is effects of the very well developed cryptogam to ask how competition between very distinct soil crusts or vesicular horizons on seed preda- shrub and annual species is affected by precip- tion ability of seeds to find safe sites seed itation variability while in many respects com- germination and seedling establishment the plemenplementarytary with the optimal life history restriction of winter growth by cold tempera- arguments this approach emphasizes how large tures could also be ofcrucial importance inhib- differences in habit affect resource capture and iting the prolonged establishment period competition rather than focusing on subtler dif- enjoyed by winter annuals in warm deserts fall ferencesferences in mortality and reproductive sched- germination followed by low levels ofphotosyn- ules the lower variability of precipitation in thesis throughout the milda4ntermild winter is essential for much of the great basin compared to the vigorous spring growth of winter annuals in the mojave and sonoran deserts as well as the more mojave and while heavy spring rains may cause reliable accumulation of moisture during the germination such late cohorts rarely reach winter recharge season may favor both stable maturity beatley 1974 annuals are common demographic patterns and growth ofperennials in transition zone sites of the ecotone between annuals tend to be shallow rooted most roots mojave desert and great basin plant commu- in upper 0010.1oloi 1 in depth and they are poorly nities in southern nevada but associated with equipped to compete with shrubs for deep soil changes in perennial species composition along moisture if shrub density is high and years of decreasingdecdeereasingbeasing mean temperature gradients in that unusually high mortality are rare then shrubs region are decreases in annual abundance may largely preempt the critical water and min- beatley 1975 eral resources and suppress growth of annuals the dominant shrubs ofthe warm deserts do not literature CITED have high root densities in the upper 10 cm of the soil profile wallace et al 1980 have lower ALKFRMANACKER MAN T LLEE M romROMNFYromneyNEY A wallageWALLACEWALLACF andaandjand J E total root densities and have lower total cover KINKINNEARNEAR 1980 phenology of desert shrubs in south- when compared with great basin perennials ern nye county nevada great basin naturalist mem- annuals are therefore likely to oirsairs 4 4 23 experience more BAMBERG S A A WALLACEWALLACF E M ROMNFY and R E intense competition from shrubs in the great HUNTERHUN 1980 further attributes of the perennial basin this conjecture is further supported by vegetation in the rock valley area of the northern considering that perennials in the great basin mojave desert great basin naturalist memoirs 4 373937 39 generally transpire 50 or more of the annual BEATLEYBFATLEY J C 1974 phenological events and their environenviron- moisture input over a wide range of yearly vari- mental triggerstuggers in mojave desert ecosystems ecology ations in the mojave this fraction may average 5585655 856 863 only 27 and vary between years from 15 to 1975 climates and vegetation patterns across the Mojave Great of southern al mojavegreat basin desert transition 50 at the same site lane et 1984 or even nevada american midland naturalist 93 53 70 be as low as 7 sammis and gay 1979 the BFNNERIBENNFRT H and B SLIIMIDTSCHMIDT 1984 on the osmoregula- reduced overlap in rooting profiles and the tion in atriplex hymenelytra torr wats greater availability of unused moisture chenopodiaceae oecologia 62 80 84 BILLINGS D 1949 shadscalescale of have favored the of W the shad vegetation zone resources may development nevada and eastern california in relation to climate annual floras in the mojave desert more than in and soils american midland naturalist 72 87 109 the great basin with severe disturbance from BOWERS J E 1982 the plant ecology of inlandidland dunes in 4farid grazing and other anthropogenic activities western north america journal of and environments 5 199 220 exotic annual species have invaded many great BRANSON F A G F GIFFORD K G RENARDRFNARD and R F basin communities once established following HADLEYHADLFY 1981 rangeland hydrology kendallhuntkeddallhuntKendall Hunt disturbance these annuals are not always easily publishing dubuque iowa displaced by short term shrub succession while BRANSON F A R F MILLER and I1 S mcqueen 1976 moisture relationships in twelve northern desert shrub this discussion has been presented in the con- communities near grand junction colorado ecology text of annuals versus perennials tradeoffs 57110457 1104 1124 199211992 PLANT adaptation 213

BRIENS M and F LARHER 1982 osmoregulation in halo- and range ecological studies 80 springer verlag phytic higher plants a comparative study of soluble berlin carbohydrates polyolspolyols betainesbebevainestaines and free proline plant depuitDFPUIT E J and M M CALDWFLLCALDWELL 1973 seasonal pat- cell and environment 5 287 292 tern of net photosynthesis in Arterartemisianisia ttltntrltridentata BROWN A D 1982 halophytic prokaryotes pages 137 american journal of botany 60 426 435 162 L S B in 0 lange P nobel C osmond H 1975 gas exchange of three cool semisemidesertdesert spe- ziegler eds encyclopedia of plant physiology new cies in relation to temperature and water stress journal series vol 12c springer verlag new york of ecology 63 835 858 CALDWELL M M 1974 physiology of desert halophytes DFDFTLINGFLING J K 1969 photosynthetic and respiratory pages 355 377 inmrR J reimoldreimoldandwand W H queen eds response of several halophytes to press moisture stress ecology of halophytes academic new york unpublished doctoral dissertation university of utah 1976 root extension and water absorption pages salt lake city 63 85 in L L Kap E D schulze ino0 lange kappenkappinekappenepenE water DONOVAN L aandyA and J R ehleringerEHL eringerFRINGER 1991 ecophysio and plant life aandj ecological studies analyses and synthe- logical differences among juvenile and reproductive sis vol 19 springer verlag new york plants ofseveralwoodyofseveral woody species oecologiaoecologia8686594594 597 1985 cold desert pages 198 212 B F chabot in dobrowolski J P M M CALDWELLCALDWFLL and J H RICH andh A mooneyedsmooney eds physiological ecology anah ofnorth ARDSAHDS 1990 basin hydrology and plant root systems american plant and communities chapman hall ltd pages 243 292 C B osmond L F pitelka and london in G M hidy eds plant biology of the basin and range CALDWFLLCALDWELL M M and H RICHARDS 1989 manajmandjJ hydraulic ecological studies 80 springer verlag berlin lift water efflux from upper roots effective- improves ehleringerEHLFRINGER J R 1985 annuals and perennials ofofwarmwarm ness of water uptake by deep roots oecologia 79 1 5 deserts pages 162 180 in B F chabot and H A CALDWFLLCALDWELL M M H RICHARDS H MANWARING in J J mooney eds physiological ecology of north american anadanddand D M EISENSTAT 1987 rapid shifts inm phosphate plant communities chapman and hall london acquisition show direct competition between neigh- ltd ehleringer R and BJORKMAN 1978 AAcom boring plants nature 327 615 616 J randarandorand00 comparisonpanson of photosynthetic characteristics of encelia CALDWELL M M R S WIIITFWHITE R T MOORE and L B species possessinvpossessing glabrous and pubescent leaves plant phys- CAMP 1977 carbon balance productivity and weiterwaiterwaterwetter iology 62 a5185aa 190 use of coidcoldeold winter desert shrub communities domi- ehleringer J R S L philllpsPHILLIPS W S F SCHUSTER and nated by c3ca3 and c4ca4 species oecologia 29 275 300 D R SANDQUISTSANDQUISF 1991 differential utilization CAMBELL G S and G A HARRIS 1977 water relations of summer rains by desert and water use patterns for artemisia dentata nutt rains plants oecologia 88 430 434 tntritrl D M and M M in wet and dry years ecology 58 652 658 eissenstat mandmmanam CALDWELLCALDWFLL 1988 competi- ability isis linked to rates of CLINE J F D W URESK D W RICHARD and W H tive water extraction oecologia 75 1 7 RICHARD 1977 comparison of soil water used by a EVANS R D 1990 drought tolerance mechanisms and sagebrush bunchgrass and a cheatgrasscheatgrass community journal of range management 30 199 201 resource allocation in artemisia tntritrldentata nutt sspasp dentata unpublished doctoral CoMCOMSTOCKsrocK J P T A COOPFRCOOPER and J R ehleringer tntri dissertation wash- pullman 1988 seasonal patterns of canopy development and ington state university EVANS R D R A BLACK and S LINK 1991 carbon gain in nineteen warm desert shrub species 0 repro- oecologia 75 327 335 ductive growth during drought inm artemisia dentatattltntritrltndentatatridentatetridentata COMSIOCKCOMSTOCK J and J R ehleringer 1984 photosyn- functional ecology 5 676 683 thetic responses to slowly decreasing leleafadwateredwaterwater poten- EVANS R D R A BLACK W H loesciierLOESC lierllerIIFR and R J tials in enceliafrutescensenceha frutescentfrutescens oecologia 61 241 248 FELLOWSFFLLOWS 1991 osmotic relations oftheodtheof the drought tol shrub 1988 contrasting photosynthetic behavior in leaves erantarant artemisia ttltntrltridentatatridentatatridentate in response to water and twigs of hymenocleaHymen oclea salsola a green twigged stress plant cell and environment 15 49 59 warm desert shrub american journal of botany 75 EVERETT R L P T TUELLERTUFLLFR J B DAVIS and A D 1360 1370 BRUNNERBRUNNFR 1980 plant phenology in galleta shadscalesbadscale 1992 correlating genetic variation in carbon isoto- and galleta sagebrush associations journal of range pic composition with complex climatic gradients pro- management 33 446 450 ceedingsce of the national academy of science 89 EVERFTTEVERETT R lanaklandkL and K WARD 1984 early plant succession 7747 7751 on pinyon juniper control bums northwest science DAVIS S DDAA PAUL andland L MALLAREMALLARF 1990 differential 58 57 58 resistance to water stress induced embolism between fernandefernanoeFFRNANDFFERNANDEZ 0 A and M M calCALDWELLCAI dwnawn i 1975 phenol- two species of chaparral shrubs rhus laurina and ogy and dynamics of root growth of three cool semi- ceanothus metacarpusmegacarpus bulletin of the ecological desert shrubs under field conditions journal of society of america program and abstracts ecology 63 703414703 714 DAVIS T M N SANKHLA W R ANDERSEN D J WEBER FLOWERS S 1934 vegetation oftheofodthethe great salt lake region andaandband B N SMITH 1985 high rates of photosynthesis in botanical gazette 95 353 418 the desert shrub chrysothamnus nauseosus sspasp FLOWERS T J P F TROKETROKF and A R YFO 1977 the albicauhsalbicaulisalbialblcaulis great basin naturalist 45 520 521 mechanism of salt tolerance in halophytes annual daubenmire R 1975 ecology of artemisia tntrltrityl dentata review of plant physiology 28 89 121 subsp tndentata in the state ofwashingtonofwashington northwest groeneveld D P and D E CROWLEYCROWLFY 1988 root science 49 24 36 system response to flooding in three desert shrub spe- DELUCIA E Hhandashandwsand W S schlesinger 1990 ecophys ciescleseles functional ecology 2 491 497 iologylologybiology of great basin and sierra nevada vegetation on HARharnisHARRISililiiIII111 S G A 1977 root phenology as a factor of compe- contrasting soils pages 143 178 incin C B osmond L F tition among grass seedlings journal of range man- pitelkaandpitelka and G M hidy eds plant biology of the basin agement 30 172 177 214 GREAT BASIN naturalist volume 52 hrllfbuslHELLE BUST J A 1976 osmoregulation annual review of MUNNS R H GRFENWAYGREENWAY and G 0 KIRSIKIRST 1982 plant physiology 27 485 505 halotolerantHalo tolerant eukaryotes pages 59 135 in 0 L lange hiHESLAsllSLA B 1984 the implications of spatial variations in P S nobel C B osmond H ziegler eds encyclo- adult performance for the distribution patterns of two pedia of plant physiology new series vol 12c perennial halophytes of thoeletooele valley utah unpubl- springer verlag new york ished mastermasterss thesis university of utah salt lake city NELSON D Llandclandaand C F TIERNANTIFRNAN 1983 winter injury of hodgkinson K C P S JOIINSONJOHNSON and B E NORTONnonNOR loslonION sagebrush and other aildlandwildwildlandwindlandland shrubs in the western 1978 influence of summer rainfall on root and shoot united states USDA foiestforest service research paper growth of a cold desert shrub atriplex conferticonfertifohaconfertifoliafolia IMT 314 intermountain forest and range experi- oecologiaoecologia3434 353353362362 ment station ogden utah 17 appp houiitonHouIHOUGH lonIONITON J G 1969 characteristics of rainfall in the nobelNOBFL P S 1988 environmental biology of agacesagaves and great basin university of nevada desert research cacti cambridge university press new york institute report reno OSMOND C BBKK winferWINTER andandhanahH zieclerZIEGLERZIFGLER 1982 funct- jefferies R Llandtand T RUDMIK 1984 the responses of ional significance ofdifferentofdifferent pathways ofcoaofco2occoa fixation halophytes to salinity an ecological perspective pages in photosynthesis pages 479148479 448 in 0 L lange P S 213 227 in R C staples ed salinity tolerance in nobel C B osmond and H ziegler eds water plants stiategiesstrategies foiformol crop improvement john wiley relations and carbon assimilation encyclopedia of and sons inc new york plant physiology new series vol 12b springer ay KIKINUSCN DSC liyllyIIYI1 iy R R 1982 effects of precipitation variance on verlag new york annual growth of 14 species of browse shrubs in south- pinPITT M D and B M WIKFEM 1990 phonological pat- eastern oregon journal of range management 35 terns and adaptation in an artemistaagropyronattemisialagropyvon plant 265 266 community journal of range management 43 350350358358 KRAMKRAMFRF R D role transfer cells the kramerkramen 1983 the possible of in POHERPOTTER L D R C REYNOLDS JR and E T plantarumplantarium adaptation of plants to salinity physiologic Plantarum louderbough 1985 mancos shale and plant com- 5854958 549649 555 munity relationships field observations journal ofaridofariaof and laniianeLANELANF L JJEE M ROMNEYROMNFY andandtT E HAKONSON 1984 environments 9 137 145 water balance calculations and productionduchon of net pro REICreichenbergerreichenbergfrREICHENBERGHENBERGERFR G and D A PYKF 1990 impact of in the northern mojave perennial vegetation in desert early root competition on fitness components of four journal of range management 37 12 18 semiarid species oecologia 85 159 166 R JR andandgG J WILLIAMS 1983 diurnal liiilfjohnlittlfjoiin 0 anda REYNOLDS T D and L FRALFYFRALEY JR 1989 root profiles of and seasonal variations in activity of crassulationcrassulation acid some native and exotic plant species in southeastern metabolism and water status in a northern plant in idaho environmental and experimental botany 29 latitude population of erinefinennaceaacea oecologia 59 opuntia erinacea 241 248 83 87 RICHARD W H and J F CLINECLINF 1965 mineral transfer in loope WWLL 1977 relationships of vegetation to environ- handahandj a greasewood community an ion uptake by grasses ment in canyonlandsCanyon lands national park unpublished doc- health physics 11 1371 1374 toral dissertation utah state university logan RICHARDS J H and M M CALDWELLCALDWFLL 1987 hydraulic LUDWIG J A G L cunninchamcunningham and P D WIIHSONWHITSON lift substantial nocturnal water transport between soil 1988 distribution of annual plants in north american layers byatternisiaby artemisiatndentatatridentata roots oecologia 7348673 486 489 deserts journal of andarid environments 15 221 227 ROMNEY M A WALLACE H KAAKAAZ andandvanavV Q HALE 1980 LUNILUNT R J leteyleffylefeyLETFY and S B CLARK 1973 oxygen 0 RJ role of shrubs on redistribution of mineral nutri- requirements for root growth in three species of desertofdesert the soil the great shrubs ecology 54 1356 1362 ents in in mojave desert basin natural- ist memoirs 4 122 131 MACMAIIONMAC malionmallonMAIION J A 1988 warm deserts pages 209 230 inm ROMO and M R HAFFRKAMP 1989 water relations M G barbour and D W billings eds north ameri- J Ttandmstandm of dentata asp and can terrestrial vegetation cambridge university press artemisia tntritrl ssp wyomingensis new york sarcobatus vermiculatus in the steppe of southeastern oregon 121 155 164 meinermelnerMEINZER F C M R SIIARIFI E T NIELSENNIFLSFN andandaandpP W american midland naturalist sammisSAM MIS and L GAY 1979 RURUNDFLND F L 1988 effects of manipulation of water and mls T W W evapotranspiration nitrogen regime on the water relations of the desert from an and zone plant community journal of andaridarld shrub larrea tndentata oecologia 77 480 486 environments 2 313321313 321 MILLERMILLFR R F 1988 comparison ofwaterofwater use byartemisiaby ArtemisiaArtantannenisia SAUFR R Hhandahanddand D W URESK 1976 phenology of steppe tndentata subsp wyomingensiswuomingensis and chrysothamnus plants in wet and dry years northwest science 50 viscidiflonisviscidiviscidiflomyflomy sppapp viscidiflorusviscidiflorus journal of range man- 133 139 agement 41 58 62 SCIIAFFFRSCHAFFER W M and M D GADCILGADGIL 1975 selection for millenMILLERMILLFR R F and L M SCIIULI 1987 development optimal lifeilfe historieshistones in plants pages 142 157 inm M L and longevity of ephemeral and perennial leaves on cody and J M diamond eds ecology and evolution pipress artemisia hitndentata nutt sspasp wyorningensiswyomingensis great ofcommunitiesofcommunities harvard university ess cambridge basin naturalist 47 227 230 massachusetts moonjyMOONkoonjymoonly jYY H A 0 BJOBJORKMANRKMAN andaandgand G J COLLAICOLLATZ 1978 SMEDLEY M P T E DAWSON J P COMSIOCKCOMSTOCK L A photosynthetic acclimation to temperature in the DONOVAN D E SIIERRILLSHERRILL C S COOK and J R desert shrub larrea divaricatedivaricatadivaricata 1 carbon dioxide eijleringfrEHLF ningerRINGER 1991 seasonal carbon isotope discrim- exchange characteristics ofintactof intact leaves plant physiolphysioc ination min a grassland community oecologia 8531485 314 320 agyogyogy6161 406 410 SMITHSMITIL S D and R S NOWAK 1990 physiologyecophysiologyEco of mooiumojiu R T S W BRFCKLF and M M CALDWELLCALDWFLL plants in the intermountain lowlands pages 179 243 1972 mineral ionlon composition and osmotic relations of in C B osmond L F pitelka and G M hidy eds AtTiatnplexatriplexattiplexplex conferticonfertifohaconfertifoliafolia and eurotiaeudotia lanata oecologia 11 plant biology of the basin and range ecological stud- 67 78 ies 80 springer verlag berlin 199219921 PLANT adaptation 215

SMITH wkandkW K and K KNAPP 1990 physiologyEcoecophysiology of high west ed temperate deserts and semi deserts elseaelsev elevation forests pages 87 142 in C B osmond L F lerier amsterdam pitelka and G M hidy eds plant biology of the basin 1988 intermountain deserts shrub steppes and steetsteetpespes and range ecological studies 80 springer verlag woodlands pages 209 230 in M G barbarbourour and berlin D W billings eds north american terrestrial vege- SMITH W K A K KNAPP J A PEARSON J H VARMAN tation cambridge university press new york J B YAVITT and D BR YOUNG 1983 influence of 1989 spatial pattern functional interactions in microclimate and growth form on plant temperatures shrub dominated plant communities pages 283 306 of early spring species in a high elevation prairie in C M mckell ed the biology and utilization of american midland naturalist 109 380380389389 shrubs academic press new york SPERRYSPERKY J S and M T TYREE 1990 water stress induced 1990 structure and function of microphytic soil crusts in embolism in three species of conifers plant cell and wikwlkwildlandwikllandwildwindlandlandllandliandiland ecosystems of and to serniaerni andarid regions pages environment 13 427 436 179 223 in advances in ecological research vol 20 press STURGES D L 1977 soil water withdrawal and root char- academic new york acteristicsacteristics of big sagebrush american midland natu- WEST N eandieandmE and M M CALDWELL 1983 snow as a factor ralist 98 257 274 in salt desert shrub vegetation patterns in curlew 1980 soil water withdrawal and root distribution valley utah american midland naturalist 109376109log 376 378078 WEST N E and CASTROGASTRO 1978 aerial under grubbed sprayed and undisturbed big sage- J phenology of of shadscaleshadscale and winterfatwmterfatfat in curlew brush vegetation great basin naturalist 40 157 164 portion winter in valley utah journal of range management 31 43 45 thornthwaite C W 1948 an approach to a rational WILSON C GRACE S ALLEN and F SLACK 1987 classification of climate geographical review 385538 55 94 J temperature and stature a study of temperatures in VASEK F C andrand R F THORNE 1977 transmontane conif- montane vegetation functional ecology 1 405 413 erous pages 797 832 M G barbour and vegetation in WOOD M K E H BLACKBURN R E ECKFRTECKERT and eds terrestrial of ohn KE JR J major vegetation california Jlohnjohn F F PETERSONPETFRSON 1978 interrelations of the physical wiley and sons inc new york properties of coppice dune and vesicular dune inter- VEST E D 1962 the great biotic communities in salt lake space soils with grass seedling emergence journal of desert institute of environmental biological range management 31 189 192 research and zoology no 73 ecology Epiepizoology series uni- WYN JONES R G andaandjand J corhamGORHAMGORIIAM 1986 osmoregulation versity of utah salt lake city pages 35 58 in 0 L lange P S nobel C B WALLACE aaldeaandeA and E M ROMNEYBOMNEY 1980 the role ofpioneerofpioneer osmond H ziegler eds encyclopedia of plant phys- species in revegetation of disturbed areas great basin iology new series vol 12c springer verlag new naturalist memoirs 4 29 31 york WALLACE AEA E M ROMNEYROMNFY andandaandjJ W CHA 1980 depth YOUNG J A and R A EVANS 1986 erosion and deposi- distribution of roots of some perennial plants in the tion of fine sediments from playasolayas journal of andaridarld nevada test site area of the northern mojave desert environments 10 103 115 great basin naturalist memoirs 4 199 205 WEST N E 1983 overview of north american temperate received 17 august 1992 deserts and semi deserts pages 321 330 in N E accepted 25 october 1992 great basin naturalist 523 appp 216 225 LIFE HISTORY ABUNDANCE AND distribution OF MOAPA DACE MOAPA CORIACEA

G gary scoppettone 1 howard L burge12 and peter L tattletuttletuttie13

abstracabstractABSIKACTr moabamoapa dace moabamoapa coricorlcoriaceaconoonconaceacopaceaacea is a federally listed endangered fish endemic to the spring fed headwaters oftheodtheof the muddy river clark county nevada species life history abundance and distribution were studied from march 1984 to january 1989 reproduction which was observed year round peaked in spring and was lowest in fall it occurred in headwaterheadwatelwater tributatributariesries of the muddy river within 150 m of warm water springspnngspang discharge in water temperatures ranging from 30 to 32 C females matured between 41 and 45 mm in fork length FL egg abundance increased with female size 2 r 93 counts ranged from 60 for a 45 mm FL female to 772 for one 90gommmm FL the oldest of eight fish aged by the opercle method was a 90 mm FL 4 year old female adults are omnivorous but tended toward carnivorecarnivorycarnivory 75 of matter by volumevolurne consumed was invertebrates and 25 plants and detritus fish size was generally commensurate with flow the largest fish occurring in the greatest flow adults were near bottom in focal velocities ranging from 0 to 55 cmsams juveniles occupied a narrower langetangerange of depths and velocities than adults and larvae occupied slack water fromflom december 1984 to Septemseptemberbeibel 1987 the total adult population ranged from 2600 to 2800 although these numbers are higher than previously believed for moabamoapa dace they are still sufficiently low to warrant its endangered status the dependency of moabamoapa dace s different life history stages to various areas and habitat types of the warm springs area suggests that all remaining habitat is necessary foifolfor thentheir survival

key words moabamoapa coriaceacoricorleoriconaceaconcopaceaacea moabamoapa dace life history reproduction biology fecundity age growth food habits habitat use body size muddy river nevada

the moabamoapa dace moabamoapa coticoriaceacotiaceacoriacea is a ther- prior to this study la rivers 1962 identified mophmophilicilic minnow endemic to the muddy river them as methodical schoolers a cursory gut system clark county nevada first collected examination by him indicated that they foraged in 1938 it has historically been relegated to the primarily on arthropods and some vegetative headwater area where the muddy river origi- matter in a systematic sampling effort deacon nates from a series of warm springs hubbs and and bradley 1972 collected moabamoapa dace in miller 1948 la rivers 1962 called the moabamoapa 28 30 C water one specimen was collected in dace and its coinhabitant moabamoapa white river 19519.5 C water within the confines of its limited springfish crenichthys baileyibaileyy moapaemoapae ther- distribution moabamoapa dace have been captured in mal endemicsendemics because of their apparent affinity a variety of habitats including spring pools and for warm water rarely exceeding 12 cm in fork slow to fast moving water and in association length FL moabamoapa dace have morphological with various substrates and submergentsubmergent legetavegeta similarities to roundtail chub gila robusta and tion hubbs and miller 1948 speckled dace rhinichthys osculus which also past ichthyofaunal surveys suggested a inhabit the muddy river hubbs and miller declining moabamoapa dace population deacon and 1948 they are more similar however to the bradley 1972 cross 1976 these surveys were genus agosia which occurs in other lower col- qualitative and produced neither an estimate of orado river drainages the two genera are spec- the number of dace remaining nor the relative ulated to have a common ancestor hubbs and population decrease between surveys ono et al miller 1948 moabamoapa dace are distinguished by 1984 thought that only several hundred small embedded scales and a bright black spot moabamoapa dace persisted and that their distribution at the base of the caudal fin had been further restricted within the already little was known of moabamoapa dace life history limited historic habitat confining them to the

abubUS fishfisli andmclmci wildlife serviceSe ivice national fisheriesFisliedes researchRe seaichsealch center reno substation reno nevada USA 89502 21ePiepiesentsentn 1 addiddiess fish and wildlife Scserviceivice dw01slialdwoisliak fisheries assistance office alisahkaAlisahka idaho USA 83520 niesent ress US fisli 31pitscntaddicssesen adresI1 US fisandbisandfish and wiwildlifeclifedlife Serservicevicevicc cleatgleatgreat basin complexCoinplex reno nevada USA 89502

216 199211992 MOAPA DACE 217 main stem of the upper muddy river and a arrow weed pluchea sericeasenceksencea two normative semi isolated headwater spring system about fishes successfully established inm the warm 130 m long the purpose of this study is to springs area mosquitmosqmtofishmosquitofishofish gambusia afafffinisims expand information on moabamoapa dace life history present when moabamoapa dace were discovered in abundance and distribution life history infor- 1938 hubbs and miller 1948 and shortshortfinshortfmfinhinbin mation includes reproductive biology habitat molly Poepreciliapoeciliapoecihacihaciliaclila mexicansmexicanamexicana introduced inm the use food habits and age and growth early 1960s hubbs and deacon 1964 besides moabamoapa dace and springfish roundtail chub and speckled dace are the only fishes STUDY AREA native occur- ring within the warm springs area but they are rare and in greater abundance downstream the muddy river is at the northern edge of cross 1976 deacon and bradley 1972 mohave where annual the desert average pre- in 1979 the moabamoapa national wildlife refuge is 15 cm usually in form of cipitation the rainfain NWR was established in histonehistoric habitat at the carpenter 1915 described historic terrestrial southern edge of the warm springs area for the which included greasewood vegetation preservation and perpetuation of the moabamoapa sarcobatus venniculatusvermiculatus shadshadscalescale atriplex dace fig 1 the refuge stream originates from creosote bush triden confettifoliaconferticonfetticonfertifoliafolia larrea five small springs occurring in a radius of 70 in tata and mesquite prosopis sp stream banks and having a cumulative discharge of abut 0090.090 09 were lined with willows salix sp screwbeanscrew bean ms fig 2 fan palms are the predominant prosopis pubescenspubescentpubes cens cottonwood populus sp ns sl riparian vegetation in 1984 moabamoapa dace larvae and mesquite carpenter 1915 harrington and adults were reintroduced into the upper prior 1930 to the completion of hoover dam refuge stream and by january 1986 there was aka boulder dam in 1935 the muddy aka a stable reproductive population of 120 adults moabamoapa river was about 48 km long and dis- authors unpublished data they were isolated charged into the virgin river which joined the by a 75 cm high waterfall springfish were the colorado river hubbs and miller 1948 only other fish present and they were abundant today it is about 40 km long and discharges into the overton arm of lake mead fig 1 source springs of the muddy river probably originate MATERIALS AND METHODS from paleozoic carbonate rocks garside and schilling 1979 and occur within a 2 km radius reproductive BIOLOGY among our As is typical of warm springs the water is rela- objectives was to quantify duration ofthe repro- tively rich in minerals garside and schilling ductive period and the season of peak larvae 1979 list sodium and calcium as predominant recruitment to this end a segment ofthe upper cationscatcanionsions and carbonate and sulfate as predomi- refuge steam system was snorkeled at 30 to nant anions total dissolved solids were 854 ppm 90godayday intervals from february 1986 to january and ph was 77tt7.7 water emerges at 32 C and 1989 and larvae were enumerated fig 2 this cools and increases in turbidity downstream is the area in which virtually all reproduction on cross 1976 although spring discharge is rela- the moabamoapa NWR occurred dace 7 15 mm TL tively constant at about 11ll1.1 msm3sns the muddy were considered larvae this range approximates river flow fluctuates because of rain agricul- the proto- to metametalarvaelarvae stages of the similar tural diversions evaporation and transpiration sized speckled dace snyder 1981 snorkeling eakin 1964 the headwater region the his- enabled us to locate reproduction sites in the toric range of the moabamoapa dace is known as the headwater muddy river system and to deter- warm springs area fig 1 during our study mine the abundance and distribution of adult the area was used primarily for agriculture and moabamoapa dace as well as to quantify habitat use for up to 0250.25 msm3s of river discharge was being all life stages areas with larvae close to swimsna4m up diverted to irrigate alfalfa barley and pasture size about 7 minmm TL were considered repro- spring outflows had been channelized and sev- duction sites fish used for food habit analysis eral were converted into irrigation ditches and aging were also used to determine fecundity some lined with concrete earthen tributary HABITAT USE we defined habitat use inm channels had scant to thick riparian corridors of terms of stream depth and velocity at foraging fan palm washingtonia filfiliferafiligeraifera tamarisk sites and at suspected spawning areas depth tamarisk sp ash trees frazinusfraxinusFrazinus sp and measurements included focal and total while 218 GREAT BASIN naturalist volume 52

N warm springs area & t 0 05 KM

stream 01 cx 001 diversion ditch soosou P concrete ditch 0 01 spring

accaapca property

be uge 7 warm springs moabamoapa national wildlife refuge road bridge

meadow valley ash

I1 warm k sp rings virgin I1 areaanea overton river muddy river 0

las 0 vegas

mead 350 N colorado riverriven 0 20 1150 km

fig 1 map showing relationship of the muddy to the virgin river andlindcind lake mead nevada and relationship of the warinwaiwar in springs aleaarea to the muddy river below warm springs area or headwaters of the muddy river showing tributary streams to the upper muddy riverrivel and relationship of the moabamoapa national wildlife refuge above 199211992 MOAPA DACE 219

N

0 loo100

meters

upper refuge stream 11 snorkeled transect abandoned swimming 0 poolpooi 0 spring

75 cm falifall re ound2u J we residence

k

PO

fig 2 map of moapdmoapamoaga national wildlife refuge shaded site indicates the reach of the upper refuge stream where larvae snorkel counts were made from february 1961986 to january 1989 velocity measurements included focal and mean yellow springs instrument model 57 dissolved water column as prescribed by bovee 1986 oxygen meter for temperature and dissolved dissolved oxygen and temperature were also oxygen sampling occurred from 1984 to 1986 measured fish were located using mask and adult habitat was also defined by contrasting snorkel A marsh and mcbirney model 201d body sizeadthsize with quantity of stream flow it was our digital flow meter mounted on a calibrated rod subjective evaluation that larger fish were was used to measure depth and velocity and a inhabiting larger water volumes we tested this 220 GREAT BASIN naturalist volume 52

200

a 2.2a 150 2g aD E 33 C UJ a 100

cz jaj3j 0 0 50 F

S S 0 fmamjjasondjfmamjjasondjfmamjjasondjFMAMJ jasondjfmamj jasondjfmamj JASONDJ 0 is-1 co 0 00CO 00CD COCD 00CO 03 0 0 0 monthyearmonth year

fig 3 abundance of moapdmoapamoaga dace larvae from february 1986 to january 1989 in the muddy river system on the moabamoapa national wildlife refuge nevada bars represent a single days count for the month NS indicates not sampled hypothesis in the summer of 1986 when samples they were preserved in 10 formalin solution of adults were minnow trapped from the contents in the anterior third of the gut were muddy river muddy spring stream refuge examined using a dissecting microscope and stream and apcar stream and their length fre- quantified by frequency of occurrence windell quenciesquencies compared discharge for each stream 1971 and by percent composition hynes 1950 was measured using standard US geological ABUNDANCE AND distribution the survey methods rantz et al 1982 near each abundance and distribution of adult moabamoapa fish sample A one way factorial ANOVA was dace 40 mm FL were determined by snor- used to test whether there was a significant keling the upper muddy river system begin- difference between length frequency among ning from 200 in downstream of warm springs fishes and different water volumes road bridge fig 1 except for 1984 the sur- AGE AND GROWTH the opercle bone was veys included 535.3 kinkm of the upper muddy river used for estimating age as described by cassel- and 757.5ts linkinkm of its spring fed tributatributariesries refuge man 1974 eight specimens collected in stream system apcar stream muddy spring summer 1985 and 1986 were aged flesh was south fork and north fork in 1984 the scraped with a scalpel and the bone allowed to survey area was the same except that only the dry glycerin was used to highlight the more upper 130 in ofthe apcar stream was snorkeled transparent region of the bone which was rather than its entire stream length snorkeling assumed to have the greatest calcium concen- was conducted over periods of four to six days tration and to have been formed in the winter when turbidity was low between 141.4 and 505.0so when food is scarce the more opaque region NTU because no agricultural return flows were signifies greater concentration of protein asso- entering the stream counts were made 6 10 ciated with growth casselman 1974 december 1984 6 10 june 1986 and 16 22 FOOD HABIT food habit analyses were september 1987 each observer enumerated made from 10 moabamoapa dace taken 9 11 novem- moabamoapa dace twice at three areas of relatively ber 1984 from each ofthree upper muddy river high concentrations 30 60 fish and the range tributatributariesries apcar south fork and muddy of results was then calculated these sites were spring they were captured by seiningbeining and chosen because the greatest variation among with unbaited minnow traps fished no longer observers was expected among them for the than 10 minutes ranging from 42 to 71 minmm FL three sites variation was less than 15 in counts 1992 MOAPA DACE 221

1000

2 r 93 800 n 23 el

U 0 LP W0 600goo ED 40

E 400 3 Z E

el 200 ll11

0 30 40 50 60 70 80 90 100 fork length mm

fig 4 moabamoapa dace fecundity as a function of fork length between individuals thus we conservatively females were not readily apparent and spawning estimated a 15 variation in our population was not observed during our study however we counts indirectly identified and quantified spawning habitat the presence of hundreds of proto RESULTS AND discussion larvae in a concrete irrigation channel immediately downstream of the baldwin reproductive biology springhead fig 1 indicated that reproduction had taken place progenitors apparently came moapa dace larvae were found year round moaba from the south fork entering baldwin spring indicating year round reproduction on the moabamoapa outflow through a diversion channel 1 NWR peak larval recruitment was in spring the fig concrete channel had homoge- low in autumn fig 3 fish at other reproduc- the irrigation neous water and tive sites in the warm springs area exhibited this depth and velocity substrate was sandy silt several sand same general trend seasonal fluctuation in larval depressions in the similar nedds recruitment was probably linked to availavailabilityability were to redds described for longfin of food in the upper muddy river system the dace agosia chrysogasterchrysogaster minckley and wil- abundance ofbenthic and drifting invertebrates lard 1971 depth and velocity at the suspected is much lower in winter than in spring scop- reddsnedds were representative of the outflow chan- pettonepettone unpublished data naiman 1976 nel and similar to other suspected spawning documented substantial seasonal fluctuation in areas in the warm springs area depth ranged primary productivity in another southwestern from 15015.0 to 19019.0igo cm near bed velocities from warm springs where production is lowest in 373.7 to 767.6tg cmseccusecemsee and mean water column veloc- winter presumably most invertebrate popula- ity from 15215.2 to 18318.3 cmseccusecemsee tion fluctuates with anpnprimaryi mary production similar to the longfin dace which repro- recently emerged larvae were found within duces during much of the year kepner 1982 150 in of spring discharge over sandy silt bot- eggs in the skein of moabamoapa dace were in differ- toms in temperatures of 30 32 C and dissolved ent stages of development all visible eggs were oxygen of 383.8 737.3 eglmgl whether spawning counted but because they are intermittently occurs only at these headwater sites or is suc- deposited and develop throughout a given year cessful only at these sites is unknown visual our counts do not represent absolute annual cues such as sexual dichromatism pronounced fecundity however egg production increased male spawning tuberclestubercledtubercles or overtly gravid with fish size rar2T 9393.93 n 25 fig 4 counts 222 GREAT BASIN naturalist volume 52

daceoace dace 80 h adults 80 adults n 564 n 564 60 n564 60 n564 40 40 20 20 0 0 i dace dace 80 juvenile 80 juvenile n148n 148 n 147 860 60 40 40 cr20 20 0 LL 0 dace dace 80 larvae 80 larvae nn201201 n 199 60 60- 40 40 20 20 0 0 0 10 20 30 40 50 60 70 80 90 100110 0 10 20 30 40 50 60 70 total depth cm mean water column velocity cuseccmseccm sec dace dace 80 adult 80 adult n 452 60 60 n 544 40- 40 20- 20 0 0 dace dace juvenile 80 juvenile 60 n 148 60 n 147 C Q 40 40 cr CD 20OL 20 L U- 0 dace dace so larvae bo larvae 60 n 201 60- n 201 40 40 20 20 0 AL 01 0 10 20 30 40 50 60 70 80 9010011090 100110 0 10 20 30 40 50 60 70 focal depth cm focal velocity cuseccmseccm sec

fig 5 mean water column and focal point velocities total depth and focal point depth used by moabamoapa dace adults juveniles and larvae in the upper muddy river system warm springs area nevada 1984 through 1986 199211992 MOAPA DACE 223

TABLE 1 fork length sex and estimated age of eight TABLETABLF 2 food items ingested by 21 moabamoapa dace by moabamoapa dace collected from the upper muddy river system percent composition hynes 1950 and percent frequencyfieflequency nevada in 1985 and 1986 age was determined by the of occurrence windell 1971 nine other guts examined opercle method were empty

FL collection food items composition of occurrence mm sex date age gastgastropodaGASI ROPODA 45 unknown 486 0 tyronia clathrataclathrate 11 48 55 unknown 786 1 oligoliiafieoligochafteoligochaete 270 238 61 unknown 786 1 ampiiipodaampilipoda 67 female 486 2 hyallelaHyallela azteca 17 95 69 female 042286 2 HFMIPTFRA 77 unknown 100985 3 Pelopeloconspelocotisphiocotiscons shoshone 45 48 80 unknown 101185 3 homovreraHOMOPI eraFRA 90 female 100885 4 aphiidaeaphndaeaphididae 90 48 triciiofieratbiciio1rera dolophilodes 51 95 nectopsyche 45 95 ranged from 60 in a 45 mm FL individual to 772 lepidofieralepidorreralepinoLEPIDO flerapleraFIERA in a 90 mm FL dace eggs were just developing paragyractis 45 95 coleopteraCOLFOPI F HA a 41 female and in mm FL were mature in a stenchnisStenstenelmissteneStenc hniscimiselmis calida 11 48 45 mm FL fish suggesting that females mature dytiscidae larvae 90 48 at lengths in this range DIPTERA chironomidae 45 48 habitat use unidentified insect paitsparts 33 95 filamentous algae 18518 5 42342 3 again moabamoapa dace larvae were found exclu- vascular plants 34 95 sively in the upper reaches of spring fed tribu detritus 28 14314.314 3 tarnestannes while juveniles occurred primarily in tributatributariesries but were more far ranging adults were present in tributatributariesries and in the main river 27 to 32 C and dissolved oxygen from 353.53 5 to 848.4 with larger fish generally found in the largerlar9reraer eglmgl water volumes there were significant differ- age growth ences in length frequencies among adults from different water volumes p 006006.006 in the annulus formation is typically associated muddy river in a flow of about 0500.50 nsms mean with an annual period of slower growth caused FL was 73 mm n 78 SD 16 mm muddy by seasonal changes in environmental condi- spring had a flow of 0200.20 msm3s and the mean FL tions such as temperature or food resources was 64 mm n 72 SD 14 mm the refuge tesch 1971 although seasonal water temper- stream flowed at 017olt0.17 msm3s and mean FL was atures do not change substantially in the warm 56 mm n 64 SD 8 mm the apcar stream springs area there is an apparent reduction of flowed at 0060.06 msm3s and mean FL was 51 mm n potential food during the winter scoppettone 89 SD 5 mm unpublished data we were unsuccessful in larvae occurred and fed in the mid to upper aging moabamoapa dace by the scale method because region of the column they were found most scales were small embedded and extremely frequently in zero water velocity fig 5 As size difficult to remove from live specimens also increased individuals tended to occupy faster environmental conditions in waters ofthe warm water and occur lower in the water column springs area were sufficiently constant that juvenile moabamoapa dace occupied focal and mean annuli were not readily apparent assumed water column velocities ranging from 0 to 46 annuli on opercular bones were presumed to be cmsemsams adults were found in a wide range ofwater associatedassociatedwithwith slower growth during the winter depths and velocities but they tended to orient ages of the eight fish examined ranged from 0 at the bottom in low to moderate current water for a 43 mm FL individual to 4 for a 90gommmm column depth ranged from 15 to 113 cm and FL female table 1 focal depth from 9 to 107 cm mean water point food column velocity ranged from 2 to 77 cmsamsems and habit focal point velocity from 0 to 55 cmsemsams water nine of 30 guts examined were empty and temperatures within adult habitats ranged from the remainder generally contained few items 224 GREAT BASIN naturalist volume 52

takeeTABLETABLF 3 estimated number of moabamoapa dace adults in six tributary streams in the warm springs area muddy river sysystemsteinstern nevada 6 14 december 1984 13 18 june 1986 and 16 22 september 1987

stream december variation june variation september variation name 1984 in count 1986 in count 1987 in count muddy river 475 71 1230 185 1165 175 refuge system 370 56 406 61 806 121 apcar 200 30 565 85 475 72 south fork 300 45 185 28 100 15 north fork 15 2 30 5 60 9 muddy spring 1450 218 160 24 200 30 total 2810 422 2581 387 2806 421 only fiethelleile ttpprnppci 130j inm ofstrcainstiestic nn vismiswiilwisi s spiedimplccl in 9841984 but what had been consumed indicated moabamoapa summer 1986 79 of the observed dace in the dace to be omnivorous tending toward carniv main stem muddy river were in groups of 10 or ory 75 by composition was invertebrates more and 37 were in groups of 30 or more in while 25 was plant material and detritus tributariestributa ries groups were generally smaller with table 2 among 21 dace guts oligochaetes 52 of the adults in groups of 10 or more and represented the largest volume 27027027.0 of food- only 13 in groups of 30 or more stuffss consumed followed by filamentous algae 18518518.5 in terms of frequency of occurrence filamentous algae occurred in 42342.3 of the guts conclusion while oligochaetes were in 23823.8 the structure of the pharyngeal teeth also suggests an omniv- moabamoapa dace are dependent upon the link orous diet they are strongly hooked but have a between the upper river and its tributariestributa ries the well developed grinding surface la rivers main stem river typically harbors the largest 1962 the presence of detritus and gastrogastropodsgastropodapods and presumably the longest lived and most indicates at least some foraging from the ben- fecund fish yet tributatributariesries are important for thos and we observed fish in the field occasion- reproduction and as larvae and juvenile nursery ally age growth information suggests pecking at substrate however the greatest habitat 0 and 0 00 time in foraging is expended on drift feeding that three years is the mean age of fish in the authors unpublished data although our data river and that adults in smaller tributariestributa ries are set does not strongly support this observation one to two years old although the moabamoapa dace population is abundance and distribution more widespread and abundant than previously moabamoapa dace were more widespread and believed its existence remains in jeopardy numerous than had been previously reported widespread movement and obligatory spawn- ono et al 1984 they were in five headwater ing near warm water spring discharge suggest tributatributariesries and the upper muddy river to about that species survival depends on access to the 100 in downstream from the warm springs entire headwater muddy river system warm road bridge fig 2 numbers ranged from springs area river and tributatributariesries alike every about 2600 in 1986 to 2800 in 1984 and 1987 effort should be made to preserve all of its the numerical distribution for the three years remaining habitat suggests movement by the adult population table 3 in 1984 the muddy spring stream supported about 50 of the population 1450 acknowledgments adults with only 16 450 adults found in the river in june 1986 we could account for only 7 william burger and dana winkleman of the population in the muddy spring stream assisted in snorkel surveys and michael parker while almost 50 of the total was in the river in and nadine kanim assisted in estimating fish 1987 the mainstream river again supported populations peter rissler helped to determine most adult moabamoapa dace 1200 the distribution habitat use michael parker conducted gut analy- of adult moabamoapa dace was patchy and clumped sis glen clemmer randy mcnatt and tom for example during the snorkel survey in strekal reviewed the manuscript linda hallock 199211992 MOAPA DACE 225

helped with editing and stephanie byers with HUBBS C Llandrand R R MILLFRMILLFH 1948 two new relict graphics genera of cyprinid fishes from nevada university ofot michigan museum of zoology occasional papers 507 1 30 CITED HYNES H B N 1950 the food of freshwater sticklebacksstickle backs literature gasterosteus acuaculeatusleatus and pyspsteuspygosteus pungipungitiustius with a leviewreview of methods used in studies of the food of boveeBOVEBOVEFE K D 1986 development and evaluation of habitat fishes journal of animal ecology 19 35 58 suitability criteriaentena for use inm the instream flow incre- KFPNFRKE P N E 11 WWGG 1982 reproductive biology of longfin dace mental methodology instreaminstrearnInstream flow information agosia chrysogasterchryso gaster in a sonoran desert stream paper 21 USU S fish and wildlife service biological arizona unpublished master s thesis arizona state report 867 235 appp university tucson CARPFNFICARPENTER K E 1915 ground water in southern nevada LA RIVERS I1 1962 fishes and henesfisheriesfis of nevada nevada U S geological survey water supply paper 365 1 86 fish and game commission reno 782 appp CASSELMAN J M 1974 analysis ofhaidhardbard tissue of pike esox MINCKLEY W Llandalandwand W E WILLARD 1971 some aspects lucius L with special reference to age and growth ofbiology of the longfin dace a cyprinid fish character- pages 13 27 in T B bagenalBagenal ed proceedings of an istic of streams in the sonoran desert southwestern international symposium on the agoingageing of fish naturalist 15 459 464 european inland fisheries commission ofoffaooffakFAOpao the NAIMAN R J 1976 primary production standing stock fisheries society of the british isles and the fish and export of organic matter in a mohave desert biological association unwin brothers thermal stream limnology and oceanography 21 60 CROSS J N 1976 status of the native fauna of the moabamoapa 73 river clark county nevada transactions of the ONO R DDJJ D wllliamwilllamsWILLIAMWILLIAMS and A WAGNER 1984 van- american fisheries society 105 503 508 ishing fishes of north america stone wall press inc DEACON J E and W G BRADLEYBRADLFY 1972 ecological dis- RANTZ S E and othersOTHFRS 1982 measurement and com- tributiontribution of the fishes of the moabamoapa muddy river in putation of streamstreamflowstrearnflowflow volume 1 measurement of clark county nevada transactions of the american stage and discharge geological survey water supply fisheries society 101 408419408 419 paper 2175 EAKIN T E 1964 groundwaterground water appraisal of coyote SNYDFRSNYDER DED E 1981 contributions to a guide to the cyprini- springs and kane spring valleys and muddy river form fish larvae of the upper colorado river system springs area lincoln and clark counties nevada in colorado USU S bureau of land management nevada department of conservation and natural denver colorado contract YA 5612 ctrct8 129 81 appp resources groundwaterground water resources reconnais- TESCH F W 1971 age and growth in W E ricker ed sance series report 25 methods for assessment of production in fresh waters GARSIDE L jandjjandaJ andaandj H SCHILLING 1979 thermalthermalwaterswaters IBP handbook no 3 blackwell scientific publication of nevada nevada bureau of mines and geology oxford and edinburgh bulletin 91 mackay school of mines university of WINDFLLWINDELL J T 1971 food analysis and rate of digestion nevada reno 163 appp in W E rickelrickerrieker ed methods for assessment of pro- harrington M R 1930 archaeological exploration inm duction in fresh waters IBP handbook no 3 black- southern nevada southwest museum papers no 4 well scientific publication oxford and edinburgh reprinted inm 1970 126 appp HUBBS C and J E DEACON 1964 additional introduc- tions of tropical fishes into southern nevada south- received I1 august 1991 western naturalist 9 249 251 accepted 15 september 1992 great basin naturalistnatur ahstabst 523 appp 226 231

CONDITION MODELS FOR WINTERING NORTHERN PINTAILS IN THE SOUTHERN HIGH PLAINS

2 loren M siniffilSinifsmithfilooi douglas G sheeley and david B wester

aiwrmcABSI nacinaclKACIr three condition models for wintering northern pintails anas abutaacuta were tested for their ability to predict fat mass logarithm of fat mass or a condition index CI incorporating fat mass equations generated to predict fat mass and the logarithm affatoffatof fat mass accounted foifolfor moiemolemore than 69 oftheofodthethe variation in these dependent variables log transforma- tions of body mass wing length and total length explained at least 60 of the variation in CI all models performed better on an independent data set mean prediction error was minimal 588 of measured variables and negative for all models regression models apply to live andlindtind dead pintails and thus represent tools that have utility in a wide variety of studies on pintail condition

key words northern pintails anas abutaacuta body condition predictive models texas waterfowl

biologists have used various indices for regions in the western hemisphere bolen et assessing waterfowl nutritional status initially al 1989 twenty thousand playasolayas are present in only body mass was used hanson 1962 folk et the SHP providing winter habitat for waterfowl al 1966 street 1975 flickinger and bolen haukos and smith 1992 at least one third 1979 but later structural variables were incor- 300000 of the northern pintails wintering in porated to adjust for individual size differences the central flyway winter on the SHP bellrose owen and cook 1977 bailey 1979 wishart 1980 1979 ringelman and szymczak 1985 and johnson et al 1985 reviewed avian condition METHODS indices and noted the value of an accurate index of lipids in migratory bird management these northern pintails were collected using studies noted that scaling morphological vari- decoys and by jump shooting on playasolayas and ables with body mass provided useful indices to associated tailwatertailtailgaterwater pits in the SHP from octo avian body condition ber through march of 1984 85 and 1985 86 northern pintails anas abutaacuta are one oftheodtheofthe tarsal length measured from the junction of most widespread waterfowl species in north the tibiotarsus and tarsometatarsus to the point america bellrose 1980 but recently their pop- of articulationofarticulation between the tarsometatarsus and ulatulationsions have declined making them a species middle toe ooi0010.01 mm flattened wing chord of special concern smith et al 1991 our measured from the insertion of the alula to the 0.1oi01 1 objectives were to provide an equation to pre- tip of the tenth primary 001 cm I1 and total body dict total carcass fat body condition of north- length measured from the tip of the bill to the ern pintails and to test that index on an end of the pygostyle thus avoiding complica- independent data set the anatomical variables tions due to tail feather growth 010.1oi cm were tested are suitable for field studies recorded for each bird during 1985 86 an additional wing measurement was recorded from the insertion of the alula to the tip of the STUDY arfarfaARPAAREAARP A ninth primary because the ninth primary maymaybebe slightly longer than the tenth birds were the study was conducted in the southern plucked and frozen high plains SHP oftexasof texas an 82880 km2 area ingesta and intestinal contents were that is one of the most intensively cultivated removed in the laboratory birds then were

I1 deudedDCU linelitlinelet ol01 kanroranro mdanilaqil viivildiavildiowildlifeVil diolielio maniremcntmaiiageineiit texistexas techtbell universityuniver sitygity lubbock texas 79409 213oxbouboi 464 eldora toiowa 50627

226 199219921 PINTAIL CONDITION MODELS 227

TABLE 1 variables used in predictive models ofbody condition for northern pintails anas abutaacntaacuta on the southern high plains texas

adult adult juvenile juvenile males n 140 females n 69 males n 58 ferfemalesnales n 49 variable Y SE Yx SE 7.7 SE 7.7 SE massgmassmasog g 96393 1094 83507 1260 91197 1641 78668 1490 tarsal length mm 411541 15 017 386838 68 0230 23 411341 13 0250 25 389038 90 0300 30 wing length cm 2658 006 2469 008 2592 010 2423 009 total length cm 4972 012 4337 014 4953 022 4314 019 lipidmassglipid mass g 17157 627 17320 833 14793 1107 14821 956 reweighed nearest ooiool0010.01 g to determine a net stant 1 was added to smooth the function CI can carcass mass and refrozen table 1 frozen be simplified to birds were sectionedaithsectioned Aithwith a meat saw and passed CI log DMFFDM twice through a meat grinder the homogenate was dried to a constant mass in either a forced because air oven 60 QC or freeze dryer dried pintails DM fat FFDM were reground to insure a uniform mixture FFDM was modeled as a function of lipidlipidwaswas extracted from 10 15 g samples using log the logarithms of structural variables petroleum ether solvent in a soxhlet apparatus LTOTAL LWING and 36 48 hrs fat free dry mass FFDM was LTARSAL and log DM as a func- calculated by subtracting water and lipid from tion of these plus the logarithm of body mass S al total carcass mass body mass minus feathers LMASLMASS johnsonetaljohnson etalet 1985 unlike mallards and anas latyrhynchosplatyrhynchosplatyrhynchousP ringelman and szszymczakymezak ingesta total carcass mass minus water geese mass yielded dry mass DM 1985 and canada branta canadensis raveling 1979 water content of three models were evaluated to predict 1 wintering northern pintails fluctuated widely smith and fat mass 2 a condition index CI incorporat- 1993 did ing fat mass and 3 the logarithm of fat mass of sheeley therefore we not test fat free as index wintering northern pintails first pintails were mass an to structural size ringel- and sorted by sex age was not significant multiple man szymczak 1985 johnson al 1985 used logarithms regression P 05os05.05 A predictive model for fat et of structural variables model was generated for each sex using total body to logarithms of car- fat fat length TOTAL wing length WING tarsal cass mass log A separate equation was estimated for length TARSAL and body mass MASS as each agesenagesex group model 3 explanatory variables using dummy variables for age DAGE and sex DSEX because regression coefficients for in model 1 regression coefficients of explan- explanatory variables differed P 0505.05 atory variables between sexes were not different among these four P 05os05.05 A predictive applicable to groups equation predictive both sexes was therefore constructed which equations were validated on a 40 included a dummy variable for sex DSEX as data set of randomly selected pintails not well as structural variables included in the generation of models percent- of each class the second model was constructed follow- ages agesexagesen of pintails in the inde- pendent sample were with ing johnson et al 1985 a lipid index was consistent their defined occurrence in the sample collection prediction error PE was calculated as an lipid index fat FFDM additional test of model performance PE is fat free dry mass is included to correct for size defined as differences between individuals lipid index PE measured Y predicted Y was transformed to where Y is the dependent variable mean PE is CI log index 1 lipid an average value for all members of the valida- because the structural measurements are allo tion data set finally predicted fat CI and log metric and because logarithms can be used to fat were correlated with lipid index in the linearize ratios johnson et al 1985 the con validation data 228 GREAT BASIN naturalist volume 52

tabieTABIFTABLE 2 regression equations and associated statistics for predicting carcass fat model 1 content g in northern pintails anas acuto collected on the southern high plains of texas october march 1984 86

explanatoryE vaivalnablesnabiesvariables

equation Rfiafi22 intercept MASS WING TOTAL DSEX

11 779 parameter estimate 191854igligi191 854 0560 13386 4136 male n 198 SE 0022 3894 1901 variance inflation factor 1181 1231 1221 partial Rfiafi22 0741 0013 0005 12 711 parameter estimate 145570145 570 0570 9516 4953 female n 118 SE 0035 5561 2994 variance inflation factor 1125 1212 1174 partial Rr2ra2 0691 0007 0007 13 757 parameter estimate 190494190igo 494 0563 12068 4409 2251322.51322 513 combined n 316 SE 0018 3178 1600 10536 variance inflation factor 1492 3164 6842 5987 partial Rr2ra2 0726 0011 0006 0004

not signifyignifiwntint P 05

tabletailleTABLF 3 regression equations and associated statistics for predicting condition index model 2 in northern pintails anas abutaacuta collected on the southern high plains of texas october march 1984 86

explanatory variables

2 equation R intercept LMASS LWING LTOTAL DSEX

21 673 parameter estimate 0816 1371 1025 0909 male n 198 SE 0069 0343 0312 variance inflation factor 11901 190 1233 1229 partial Rr2ra2 0656 0015 0014 22 599 parameter estimate 0725 1316 1179 0710 female n 118 SE 0101 0512 0486 variance inflation factor 11231 123 1206 11761 176 partial r2Rra2 0595 0019 0008 23 657 parameter estimate 0761 1350 1080 0834 0041 combined n 316 SE 0057 0286 0264 0016 variance inflation factor 1496 3207 7035 6141 partial Rflafl22 0610 0016 0011 0007

stepwise multiple regression maximum r2Rra2 RESULTS improvement technique was used to generate and test all models SAS institute inc 1985 in model 1 table 2 body mass explained a variables were eliminated that did not contrib- major portion of variation in carcass fat content ute significantly P 0505.05 to a model partial Rr2ra2 in males equation 11iili1.1 and females equation values were calculated for each variable in a 121.2 total length did not account for a signifi- cant FP 0505.05 of variation in fat content model A sum of squares type II11 for each portion for females as it did males based on low vari- model variable was divided by the total sum of ance inflation factors VIF regression coeffi- squares in the model A partial r2Rra2 value for a cient estimates for each sex were stable when variable the contribu- given represents unique sexes were combined through use of a dummy variable all tion of that when other variables are variable 1.313 the VIF for TOTAL and 2 equation 13 already present in the model partial Rfiafi2 values DDSEXSEX were relatively high this is largely attrib- are not additive and therefore their sum will utable to the high correlation between length 2 not equal the total model Rfiafi2 differences in and sex of bird point biserial correlation coeffi- variation accounted for by ninth versus tenth cient equal to 091ogi0.91 primary length were evaluated using the Rfiafi22 pro- LTOTAL LWING and LTARSAL ex- cedure SAS institute inc 1985 plained variation in log FFDM for modeling 199211992 PINTAIL CONDITION MODELS 229

TABLE 4 regression equations and associated statistics for predicting log carcass fat model 3 in northern pintails anas abutaacuta collected on the southern high plains of texas october march 1984 86

I1explanatory variablescanabvanablesies equation Rr2ra2 intercept LMASS LWING

31 727 parameter estimate 3410 3412 3209 adult male n 140 SE 0182 0993 variance inflation factor 1156 1156 partial r2ra 0697 0021 32 693 parameter estimate 1611 3687 4998 adult female n 69 SE 0303 1472 variance inflation factor 1034 1034 partial r2ra 0687 0054 33 722 parameter estimate 11066 5028 1223 juvenile male n 58 SE 0422 2009 variance inflation factor 1015 1015 partial r2ra 0719 0002 34 745 parameter estimate 5444 3968 2834 juvenile female n 49 SE 0348 1844 variance inflation factor 1109 11091 109lag partial r2rafiafi2 0720 0013

2 TABLE 5 coefficients of determination fi and predic- only variables that contributed significantly tive error estimates from the validation n 40 of predic- P 0505.05 but they were not homogeneous tive equations to measured variables and lipid index for P agog05.05 between agesexagesen groups therefore pintails anas the southern wintering northern abutaacuta on four estimated 4 DAGE high plains of texas october march 1984861984 86 equations were table explained variation in log fat but not CI mean prediction a1 lipid index given other model variables body mass 2 2 equation Rr2ra error SE Rr2ra MASS and LMASLMASSS consistently accounted for the largest portion of variation in carcass fat 5850b 11ilii1.11 1 and 121 2 785 99219 921 5 850 662 table 2 CI table 3 and log fat table 4 of fat 616 pintails 13 765 90439 043 t 58535.8535 853 659 wintering northern wing length fat 6246 24 WING and LWING explained 1 5 of the 212 1 and 222 2 697 0 0192001920.0192 000910.00910 0091 671 variation in carcass fat log fat and CI when condition index 7877 87 other variables were already in the models 23 700 0019 00092 675 condition index 7797.797 79 TARSAL did not contribute to any model vari- 31 34 733 0050 00009 634 ation accounted for by ninth and tenth primary log fat 2412 41 lengths always differed by less than 1 conse- quentquentlyly ninth length was not tested in Piepredictiondiction error expressed as a percentagepercent ige of the memeanin inin the validationv ilidalid itionaition datad it i primary set any model negative prediction eierrorror indicates overestimation of the true sainesaluevalue in the validation data set all models accounted for 69 or more of variation in car- log DM LMASS LWING and DSEX were cass fat mass CI and log fat table 5 all significant P 05os05.05 thus CI was modeled with models explained less than 70 of the variation LTOTAL LWING LTARSAL and LMASS for in lipid index for validation datasetdata set birds bias sexes separately and combined table 3 As in in all models was relatively low and negative model 1 regression coefficient estimates were predictive equations overestimated fat mass stable in equations 212.12 1 and 222.2 when sexes were CI and log fat of validation datasetdata set pintails combined multicollinearity between TOTAL and DSEX resulted in relatively high VIFs for discussion these variables age and sex effects were significant when log A useful condition index will save funds by fat was regressed on the same explanatory vari- eliminating the need for expensive laboratory ables used in model 2 furthermore the struc- analyses and willwiil411 lessen the need to sacrifice tural variables LMASS and LWING were the birds for direct nutrient analyses the problems 230 GREAT BASIN naturalist volume 52 associated with using body mass alone as an graphical range in which they were developed index to condition of migratory birds have been for comparisons between age and sex classes noted bailey 1979 wishart 1979 iverson and we encourage use of model 3 research also may vohs 1982 johnson et al 1985 because indi- require knowledge of absolute fat content viduals vary in structural size body mass will importance of accuracy and precision will affect reflect that variability in muscle and bone in model selection care should be exercised to addition to variation in lipids restrict model use to winter when changes in models have been developed that predict fat body mass primarily reflect fluctuations in fat content in waterfowl but these require sacrifice not fat free dry mass ie protein and mineral and dissection of the bird woodall 1978 chap- fractions pell and titman 1983 thomas et al 1983 whyte and bolen 1984 these equations may acknowledgments incorporate skin subcutaneous abdominal omental andor intestinal visceral fat mass we offer our thanks to A P leif C D and often account for most of the variation in olawsky D G cook P J grissom and PR N total body fat content our study was designed gray for field assistance E G bolen L D to develop models using explanatory variables vangilder D H johnson and C B ramsey that could be applied to live as well as dead provided comments on the manuscript the pintails project was supported by the caesar kleberg miller 1989 developed regression models foundation for wildlife conservation and the to predict carcass fat on live pintails from sac- texas state line item for noxious brush and ramento valley california but cautioned weed control this is manuscript T 9 488 ofthe against their use outside that region our regres- college of agricultural sciences texas tech sion models for carcass fat provided better esti- university mates offat raRr2 7171.71 for live pintails than those 2 developed for california birds raRr2 66gg66.66 how- literature CITED ever similar to millermillerss 1989 study body mass 2 alone accounted for most of the variation fiaRfi2 balleyBAILEYBAILFY R 0 1979 methods of estimating total lipid con- 69gg69.69 in pintail carcass fat tent in the redhead duck aythya anwricanaamencanaamencana and an evaluation of condition indices canadian journal of the possibility of a condition bias among zoology 57 1830 1833 waterfowl captured in baited traps versus the BELLROSEBFLLHOSF ECF C 1980 ducks geese and swans of north general population has been addressed america ard3rd ed stackpole books harrisburgHarnsburg penn- weatherhead and ankney 1984 1985 sylvania BOLENBOLFN E G L M smitnSMIIIISMITH and H L SCIIKAMMSCHRAMM 1989 burnham and 1985 GL nichols hypothetically playa lakes prairie wetlands of the southern high birds in poor condition may be hungrier less plains biosciencebioscience3939 615623615 623 wary and more likely to enter a trap containing BURNIIAM K Ppandjpandlpandaand J D NICHOLS 1985 on condition food condition models could be used to test for bias and band recovery data from large scale waterfowl evidence banding programs wildlife society bulletin 13 345 of a body condition bias given that 349 samples ofpintailsofpintails captured both in baited traps CHAPPELL W A and R D TITMAN 1983 estimating and by presumably less biased methods eg reserve lipids in greater scaup aythya manlamaniamarila and net gun are available lesser scaup aythya affinis canadian journal of zoology 61 35 38 models could be used to test for annual flickinger E L and E G BOLENBOLFN 1979 weights of variation in body condition and for changes in lesser snow geese taken on their winter range jour- condition across the winterminter ringelman and nal ofwildlifeofwildlife management 43 531 533 szymczak 1985 demonstrated the potential of FOLK CKC K HUDFC andaandjand J TOUFAR 1966 the weight of the mallard anas platyrhynchosplatyrhynchous and its changes in condition indices in determining spatial differ the course of the year zoology listy 15 249 260 encesances in body condition and the preferability of HANSON 11 C 1962 the dynamics of condition factors in condition indices to use of body mass alone canada geese and their relation to seasonal stresses paper no et al 1986 also used condition indices arctic institute north america technical hepp to 12 fairbanks alaska document a positive relationship between con- HAUKOS D aanalaandlA and L M SMIHISMITIL 1992 ecology of playa dition and survival in mallards lakes in waterfowl management handbook USU S fish these pintail condition models should be and wildlife service leaflet 133713 3 7 hepp R BLOIIM R REYNOLDS useful waterfowl G R J F J F HINFSHINES to biologists however models and J D NICHOLS 1986 physiological condition of should be verified when used outside the seogeo autumn banded mallards and its relationship to hunting 1992 PINTAIL CONDITION MODELS 231

vulnerability journal ofwildlife management 50 177 condition of northernnor them pintails in the southern high 183 plains journal of wildlife management 57 in piessplesspress inersonIVFHSONIVERSON G C andandaandpP A loiisvoiis liililJK 1982 estimating lipid sybSTHSIRFFIE fr M 1975 seasonal changes in the diet body weight content of sandhill cranes from anatomical measure and condition of fledged mallaidsmallardsmalimaiiMallaids in eastern england ments journal of wildlife management 46 47883478 483 international congress of cainecalnegainegame biologists 7 339339347347 JOHNSONJOIINSON D H G L KRAPU K RFINFCKFRFINECKE and D G J reinecke tuomasTHOMAS V G S K mainguyMAINCUY and J P PKIpm vi rri i 1983 lordeJORDEJORDF 1985 an evaluation of condition indices for predicting fat content of geese frornfromtrombrorn abdominal fat birds of 569 journal wildlife management 49 569575575 weight journal of wildlife management 47 1115 MILLERMILLFR M R 1989 fat botein estimating carcass and piproteinotein in 1119 northern pintails during the norihorihorlnonbreeding season breeding wfarny111iead P J and C D ANKNFY 1984 A critical journal of wildlife management 53 123 129 wrallinuifad assumption of band recoveryrecoreeo ery models may often be OWFNOWEN M and W A COOK 1977 variations in body violated wildlife society bulletin 12 198 199 weight wing length and condition of mallards anas 1985 bias platyrhynchosplatyrhynchous and thentheir relationship to environmental condition and band recovery data a changes journal of zoology 183 377 395 reply to burnhamBurn hainharn and nichols wildlife society bulle- tin 13 349351349 351 RAVELINGRAVFLING D G 1979 the annual cycle of body compesicomposi tion of canada geese with special reference to control wiivmWHY IF R J and E G BOLENBOLFN 1984 variation in winter of reproduction auk 96 234 252 fat depots and condition indices of mallardsmalimaiiMallaids journal of RINGELMAN J K and M R SYMCAKSZYMCZAK 1985 A physiaphysio wildlife management 48137048 1370 1373 logical condition index foifolfor wintering mallardsmalimaiiMallaidsards journal WISHARI R A 1979 indices of structural size and condi- of wildlife management 49 564 568 tion of american digeonwigeon anas amei icanaacana canadian SAS INSTITUTE INCIN 1985 SAS uleisuseisusers guide statistics journal of zoology 57 2369 2374 version 5 edition SAS institute inc carygary noinorthth WOODALL P F 1978 omental fat a condition index for carolina red billed teal journal of wildlife management 42 smitnSMIIIISMITH GWG W F A JOIINSONJOHNSON J B bohtnfrBORTNFR J P BLADENBLADFN 188 190 and PR D keywoodKTYVWOD 1991 trends in duck breeding populations USU S fish and wildlife service administrative report laurel maryland received I1 june 991iggi1991 SMIIIISMITH L M and D G SIIFFLFYSIIEFLEY 1993 factors affectaffectingin accepted 22 iunelunejune 1992 greatgi eat basinscismbcism naturalist 523 appp 232 236

evaluation OF ROAD TRACK SURVEYS FOR COUGARS FELIS CONCOLOR

walter D van sickle I1 and frederick G lindzey I1

aj3sniacABS I1 RA rI1 road track surveys were a poor index of cougar density in southern utah the weak relationship we found between track finding frequency and cougar density undoubtedly resulted in part from the fact that available roads do not 2 sample properly frornfrombrorn the nonuniformly distributed cougar population however the significantly positive relationship r 73 we found between track finding frequency and number of cougar home ranges crossing the survey road suggested the technique may be of use in monitoring cougar populations where roadabundanceroad abundance and location allow the population to be sampled properlyproper ly the amount of variance in track finding frequency unexplained by number ofhome ranges overlapping survey roads indicates the index may be useful in demonstrating only relatively large changes in cougar population size

key words cougar felis concolor track survey utah

sign left by animals has been commonly STUDY AREA used by wildlife managers to make inferences about population characteristics neff 1968 the boulder Escalanteescalante study area comprises lindzeylindzeyetLindzeyetet al 1977 novak 1977 this approach 4500 kmofkm2kjof garfieldofofgarfield and kane counties in south is appealing because it seldom requires special- central utah boulder escalante and canaan ized equipment and is usually much less costly mountains dominate the area topographically than other more intensive techniques the and elevation ranges from 1350 rn to 3355 m approach requires however that the relation- hot dry weather is characteristic of june and ship between sign and the population character- july with rains beginning in august and contin- istic of interest eg size composition be uing through september annual precipitation understood ranges from 18 cm at low elevations to 60 cm at track counts have been used to indicate high elevations average temperatures for cougar felis concolor abundance or change in escalante in january and july are 282.8 C and abundance but population estimates were 24524.5945 C respectively US department of com- seldom available to evaluate the validity of these merce 1979 indices koford 1978 shaw 1979 fitzhugh and desert grass and shrub communities domi- smallwood 1988 van dyke et al 1986 how- nate the vegetation with a sparse overstory of ever conducted road track surveys in an area of pinyon pine pinus edulisadulis and juniper known cougar density and found a weak rela- juniperus osteosperma between 1350 m and tiontionshipship tat2r2 isls18.18 between track finding fre- 1800 m dense pinyon juniper stands with a quency and density because of the potential sagebrush atreayreATteartemisiamisia tridentata understory value of this technique to agencies chargeda4thcharged with dominate the vegetation between 1800 rn and management of cougars our objective was to 2400 m ponderosa pine pinus ponderosa and test again the relationship between track find oakbrushoakbrush quercus gambehigambeliigambgambelinehiebiebleliieill are prominent ing frequency and cougar density following pro- above 2400 m where rocky vertical walled can- cedures of van dyke et al 1986 additionally yons with large areas of bare sandstone charac- we examined the influence cougar distribution terize the topography subalpine meadows with patterns as measured by cougar home ranges small stands of engelmann spruce picea had on track finding frequency engelmanii quaking aspen populus tremultremuloidesoides

wyomingwy ing cooprativcoopeialive FIIfilFpil II tildibidiiidd wildlifwildlife Rri seilseiisellseiicliseiichiwhclielleil unit boxB 3166 university statistation lailallaLiramiei wywyomingjg 820782071

232 199211992 COUGAR TRACK SURVEY 233 and white fir abies concolor occur above 2700 substrate surface condition survey areas dif- in river canyons transverse the area with asso- fered in density independent adult cougars per ciated vegetation consisting primarily of fre- kmkm2 and the number of home ranges that inter- mont cottonwood populus fremonfremontiifremontiatii and sected road sections 2 3 in the first 4 5 in the willow salix sppapp ackerman 1982 hemker second and 6 7 in the third 1982 roads were surveyed from a pickup truck at the human population of about 800 is con- 8 12 aphkph each road including both shoulders centcentratedrated in the towns of escalante and boul- was dragged with a conifer tree pulled from the der livestock grazing timber harvesting and rear of the truck the following day both sides energy exploration are the primary land uses in of the road were searched for cougar track sets the area road density is about 25 lulkmkni of road by driving on one side and returning on the per 100 kmkm2 van dyke et al 1986 hunting of other A track set was defined as a continuous cougars is prohibited on the study area set of tracks created by one cougar on a single occasion three to 10 days later each road was METHODS again surveyed and dragged we felt that after 3 days the effect of dragging would be minimal capture and monitoring procedures and movements of cougars in the area hemker et al 1984 suggested this interval would be cougars were tracked on horseback treed sufficient to provide independent sampling with the aid of trained hounds and immobilized periods dust ratings determined from imprint with an intramuscular injection of ketazineketamineketamine characteristics of the observerobserverss shoe van dyke hydrochloride and xylazinexylazine hydrochloride et al 1986 were conducted every km before hemker et al 1984 each immobilized cougar and after dragging to quantify road surface con- was fitted with a collar containing a motion sen dition at each stop the observer took 10 steps sitiveshtive radio transmitter telonics inc mesa 5 on each shoulder then each impression was arizona radio collared cougars were moni- given a point value from I1 to 4 simple regres- tored with portable radio telemetry equipment sion analyses were used to examine the relation- on the ground and from the air all radioloca ship between track sets per km surveyed and tionseions were assigned UTM coordinates and both measures of density track sets per km recorded to the nearest 100 in an attempt was surveyed were considered the independent made to locate all radio collared cougars a min- variable because only these data would be avail- imum of once each week able to the manager the boulder Escalantescalsealescalanteante study area including the random systematic road track survey areas occupied by collared cougars was involved dividing the study area into four survey searched periodically for sign of new cougars areas again the four areas were spatially and eg tracks seats scratches when detected behaviorally isolated from each other two uncollared cougars taking up residence and survey areas had 2 4 cougar home ranges over- transients were captured and radio collared lapping roads and two had 5 7 each area had a different density of cougars 001700170.017 00320.032 00420.042 road track surveys 001700320042001700320.042 0050.0500570.057 7 cougarskcougarcougarskm2skm2ma A 16 km stretch ofroad was cougar density was measured as both the randomly selected in each area and the first number of known cougars per kkm2m2ma in the survey area to be surveyed was randomly chosen sur- area and the number of home ranges of inde- veys were run as described for systematic sur- pendent cougars overlapping the survey road veys except that an all terrain vehicle was used we conducted both systematic fitzhugh and and only one shoulder of the road was dragged smallwood 1988 and random systematic van once all four areas had been surveyed we dyke et al 1986 road track surveys only dirt returned to the first area randomly selected roads were surveyed different 16 km survey routes for each area and for the systematic survey the study area was began the sequence again surveyed roads were divided into three survey areas spatially and not eligible for resampling until all dirt roads behaviorally home range boundaries isolated within an area had been sampled once for from the others one 113 km section of road analyses each 16 kmkin section of road was was chosen in each area roads were similar in divided into segments varying in length from I1 elevation change habitat type and condition to 10 kinkm depending on the number of home 234 GREAT BASIN naturalist volume 52

7 deformed traffic categories were no traffic traffic on one half the length of the road and 6 traffic half Q on more than one the length 1 ICC 5 RESULTS

0- 4 the systematic road track surveys were con- 02hji may this 0 ducted june 1988 during period 407 3 r2ra 073 SZr ro73o73 km of road was surveyed and two track sets were ce df3d found one hundred thirty five km 12 surveys CC 21t p0066 of road was surveyed in an area where 2 3 ranges overlapped the survey road 146 km 13 8 surveys where 4 5 ranges overlapped and 126 km 11 surveys where 6 7 ranges overlapped 0 0 0005 001 00150075 002 the survey road unequal survey numbers resulted from weather or equipment problems TRACK SETS PER KILOMETER SURVEYED precluding surveys being run each road 11311.3 km surveyed hours with fig 1 relationship between cougar track sets per kilo- was in three two areas meter and cougars with home ranges overlapping the survey being surveyed the first day and the third the road on the boulder escalante study area utah 1988 next day the two track sets were found on a road overlapped by 4 5 cougar home ranges ranges overlapping the segment each segment because ofthe small number of track sets found then had a home range overlap value 2 7 and these results were not regressed against either was assigned one of the four density values measure of density we examined the relationship between track random systematic road track surveys were sets found per linkinkm surveyed and the two mea- run in july and august 1988 during this period sures of density with simple regression analysis 684 linkinkm was surveyed and seven cougar track road segments with the same home range over- sets were found three hundred fifty km 37 lap values were combined to obtain km sur- road segments was located in an area of low veyed as were road segments representing the home rangeroad overlap and 334 kmkin 42 road same densities data points entered into the segments in high the number of linirnkmkin searched regression equations were the sum of tracks per day was 16 found in each of the six home range overlap or we identified no relationship between den- divided kmkin 2 four density categories by the sum oflin sity as measured in cougars per linkinkm2 and track surveyed in the respective categories finding frequency 00oo00.00 P 886886.886 n 4 we evaluated whether dragging would however the relationship Y 2232.23 197x rr2ra improve survey roads with a simple regression 7373.73 P 066066.066 ROOT MSEM S E 1 n 5 between of predragpre drag dust ratings against post drag rat- number of cougars known to have home ranges ings data from both road track surveys were overlapping the road and track finding frequency combined to increase sample size and regres- was positive fig 1 the data point associated sion slopes were tested against 1 the number with the home range overlap value of 7 was drop- of track sets found on dragged and undragged ped because 20 km of road was surveyed roads was also compared by dividing the total results from both one day periods and three or number of track sets in each by the total kmkinkiu more days were combined for these analyses searched in each because of the small number of track sets multiple regression analysis was used to found we did not statistically evaluate the rela- examine the effect of rainfall and traffic on tiontionshipship between track finding frequency and one day post drag dust ratings predragpre drag dust dust rating categories or dragged and ratings rainfall and traffic were the indepen- draggedunundraggedundrugged roads we found a positive relation- dent variables considered we used two indica- ship between post drag dust ratings Y and tor variables to code the three levels of rainfall prepredragdrag ratings after one XI and three or and two to code the three levels of traffic the more xax2 days Vrar2 5454.54 Y 6056.05gos 08750875xi0875x1XI P three road surface categories related to increas- ooi001001.001 ROOT MSE 10410.4 n 43 rar2 3434.34 Y ing rainfall intensity were unchanged dimpled 3143.14 0707x2 P oi0101.01 ROOT MSE 464.6 individual raindrop impressions distinct and n 20 however we failed to reject the null 199219921 COUGAR TRACK SURVEY 235 hypothesis slope 1 in both cases indicating location in determining number oftracks found that our method of road dragging did little to use of index values to compare cougar density improve tracking medium or that dust ratings between areas in tenuous the probability of were not sensitive enough to detect changes in existing road networks in two areas sampling the tracking medium data associated with similarly from the two populations seems small heavy rainfall were omitted from these analyses use of track surveys to document cougar pres- multiple regression analysis one day relating ence is feasible but again the approach ulti- post drag dust ratings to predragpre drag dust ratings mately relies on roads intersecting a cougar rainfall and traffic yielded a three variable home range model that contained only predragpre drag dust ratings ideally roads with suitable tracking surface XI and rainfall xax2 x3xa as the independent should be abundant as in parts ofthe northwest variables rarr2 6767.67 Y 7657.65 0838xi0838x1 76x2076x20 where logging is common and located so that 565x3 FP ooo000000xi000x1XI P 583x2 P the home range of each cougar would be inter- 001x3 ROOT MSE 9 n 43 moderate cepted even in an ideal situation however the rainfall had little effect on post drag dust rat- index may prove sensitive only to relatively large ings however heavy rainfall resulting in road changes in cougar population size twenty surface deformity had a deleterious effect on seven percent of the variance in number of post drag dust ratings the effect of traffic on tracks found was unexplained by number of post drag dust ratings was not significant P 0505.05 cougar home ranges overlapping survey roads

discussion acknowledgments research was funded by the the utility of road track surveys for monitor this utah divi- sion of wildlife resources and administered ing cougar abundance is limited by the generally sionslon by the and poor relationship between cougar density and wyoming utah cooperative fishery 1 and wildlife research units we thank W J track finding frequency both our results raTr2 bates for coordinating our thrughthough the 00oo00.00 although based on a small sample and project UDWR H harlow R A powell L L those ofvan dyke et al 1986 rar2 lsis18.18 indicate J G reviewed initial a weak relationship between cougar density and mcdonald and D bonett drafts of the we offer track finding frequency the strongest signifi- manuscript special thanks 2 to C S mecham and M C mecham for field cant relationship found by van dyke et al raTr2 assistance and functioning as houndsmenhoundsmen glgi61.61 resulted from a multiple regression model with track finding frequency the dependent variable and female density good tracking con- literature CITED ditionsditions and proximity of cougars to survey road the independent variables As the authors ACKERMAN B B 1982 cougar predapredationpredd tion and ecological energetics in southern utah unpublished master s noted however a biologist would seldom have thesis utah state university logan 95 appp knowledge of cougar distribution in regard to FITZHUGH E lanaklandkL and K S SMALLWOOD 1988 techniques survey roads for monitoring mountain lion population levels pages relationship documented between 69 71 in R H smith ed proceedings of the third the poor mountain lion workshop arizona game and fish track finding frequency and cougar density department appears the result of sampling problems largely hemkerHEMKEHFMKFRii T P 1982 population characteristics and move- beyond the control of the biologist cougars are ment patterns of cougars in southern utah unpub- rarely uniformly distributed hemker et al lished mastermasterss thesis utah state university logan 66 1984 and available roads the PP sampling strata HEMKEHFMKERii T PPFF G lind7fyGLINDZEY andandaandbB B ackermanbackerman&ackermanACKFRMAN 1984 are seldom abundant enough or optimally population characteristics andlindtind movement patterns of located to sample from a nonuniform distribu- cougars in southern utah journal ofwildlife manage- tion available roads for example could fail to ment 48 1275 1284 KOFORDKOFORDC C B 1978 the welfare of the puma in california intercept any cougar home ranges or could be carnivore 1 92 96 found only in the areas occupied by cougars in LINDEYLINDZEY F GGSS K tilTiiTHOMPSONomPSON andandaandjJ I1 hodgesHODGFSLHODGES 1977 both scenarios the index tracks found could scent station index ofblack bear abundance journal of wildlife management 41 151 153 easily prove to be a poor measure of change in NEFF D J 1968 the pellet group count technique for big cougar numbers over time in an area likewise game trend census and distribution a review journal because of the potential importance of road of wildlife management 3259732 597 614 236 GREAT BASIN naturalist volume 52

NOVAK M 1977 determining the average size and compo presence journal of wildlife management 50 102 sitionaition of beaver families journaljouinal of wildlife manage 109 ment41ment 41 751 754 US department OF commerceCOMMFRLFCOMMFECE 1979 climatological SHAW II11 G 1979 A mountain lion field guide arizona data game annual summary climatological data utah andlindtind fish department special report no 9 27 8113 PP VAN DYKIDYKE F G R H BROCKEBRO kr and H G SHAW 1986 use of t oadoddroad track counts as indices of mountain lion received 10 november 1991 accepted 16 april 1992 great basin naturalist 523 appp 237 244

LEAF AREA RATIOS FOR SELECTED RANGELAND PLANT SPECIES

1 2 1 mark A weltz wilbert H blackburn and J1 roerroger simanton

ABSTRACT leaf area estimates are required by hydrologic erosion and growthyieldgrowth yield simulation models and are important to the understanding of transpiration interception coagoa fixation and the energy balance fortor native plant 2 communities leaf biomass g to leaf area mm linear regression relationships were evaluated for 15 perennial grasses 12 shrubs and 1 tree the slope coefficient 30po of the linear regression equation is a ratio of leaf area to leaf biomass and is defined as the leaf area ratio LAR one sided leaf area mm2ovenbovenoven dry leaf weight g LAR represents Pp0pa in each 2 regression equation where Y powpombox3ox linear regression relationships for leaf area were computed r 849884 98 for all 28 native range species after full leaf extension within plant estimates of leaf area for mesquite prosopis glandulosaglandulose torr var glanduloseglandulosa torr cockllcockal or lime pricklyashprickpacklypncklylyashash zanthoxylumfagarazantjwxylumfagaraZanthoxylumfagara L sarg were not significantly different P 05 LARs for three of the shrubs and the tree were established at four different phenological stages there were no significant differences P 05 in LARs for lime packlypncklyprickly ash mesquite and texas persimmon diospyros teanateonatexana scheele after full leaf extension during the growing season the LAR relationship for texas persimmon changed significantly after full leaf extension LAR relationships for texas colubrinacolub nna colubrinaColubnna texensis T & G gray changed in response to water stress

key words leaf area index drought response learleaficarleafbiomassbiomass

eighty percent of the world s rangeland is wight 1986 kc is defined as the ratio of actual classified as andaridarndannd or semiarid branson et al evapotranspiration to evapotranspiration when 1981 ie precipitation is less than evapotrans water is nonlimiting this empirical method is pirationpispirationration under these conditions water avail- extremely difficult to parameterize for range- ability is the most important environmental lands because water is often limiting and esti- factor controlling plant production and survival mates of transpiration are confounded by soil brown 1977 evapotranspiration ET is the water evaporation wight and hansen 1990 major component of the water balance and is thus wight and hansen 1990 reported that estimated to account for 96 of annual precip- kc values were not transferable across range itation for rangeland ecosystems branson et al sites the second method is based on leaf area 1981 carlson et al 1990 with surface runoff index LAI ritchie 1972 LAI is defined as the accounting for most of the remaining 4 foliage area per unit land area watson 1947 gifford 1975 lauenroth and sims 1976 carl- the LAI method is more process based than the son et al 1990 kc approach and has been successfully used in evapotranspiration has been measured for several rangeland hydrologic erosion and selected rangeland plant communities with growthyieldgrowth yield simulation models wight and lysimeters and the bowen ratio method wight skiles 1987 lane and nearingnearinoNegearinoarino 1989 arnold et 1971 hanson 1976 gay and fritschen 1979 al 1990 carlson etalet al 1990 estimates ofET for unmea- A limitation in using natural resource sured rangeland plant communities are usually models like the water erosion prediction proj- simulated from hydrologic models lane et al ect WEPP lane and nearing 1989 is in 1984 wight 1986 for hydrologic simulation developing LAI coefficients for rangeland models to be biologically meaningful improved plants LAI is difficult to measure because ofthe methods of simulating evapotranspiration from drought deciduous nature of certain shrubs in rangeland plant communities are needed two which several cycles of leaf initiation and defo- different approaches are currently being used liation occur within a single growing season one approach is to use a crop coefficient kc ganskopp and miller 1986 and seasonal

I1 USDA agriculturalagnlulturilresnrcliresearch seaceservice southwest W itu&hedwatershed research centergenter 2000 eastF ist alienallenailenahlenablen roadro id on arizona 85719159685719 1596 2 tiontituon anoni nolnoithernnorthernNoiortheinplunsaieiadministntiveofficcthern plains area adininistrative office 2625 redwing roadho id suite 350 fort collinsColl his coloradocoloichloi do 80526

237 238 GREAT BASIN naturalist volume 52

tahltabiTAHITABLE 1 1 description of study sites iangelangerange sites and soil seriesserlessenes of species evaluated for leaf area to leaf biomass relationships

frost mean free PPT period location range site mm days soil seriesserlessenes soil family tombstone AZ limey upland 356 239 Strongstrongholdholdboldhoid coarse loamy mixed thermic ustouicUstustollicollic calciorthidCalciorthid meekelmeeker CO0 clayey slopes 200 180 debaterdegater clay montmonlloniticmontmorillonitic mesic typic camborthidcambortbidCamborthid sidney MT silty 300 130 vida fine loamy mixed typic argboroll chickashacidekasbaChicghiekasha OK loamy prairiepianplanprah ie 927 200 grant fine silty mixed udic argiustollaigiustoll chickashaChickasha OK liodedpranieeroded prairie 927 200 eroded fine silty mixed udic grant aigiustollargiustoll ft supply OK dune 597 200 pratt sandy mixed thermic psammentic haplustalf woodwardWoodwaidwald OK shallow piprairiean ie 584 200 quinlan loamy mixed thermic shallow typic ustochrept alice TX fine sandy loam 710 280 miguel fine mixed hyperthermic udic Paleupaleustalfstalf sonolasonora TX shallow 609 240 purves finefincpincpine loamy mixed thermicthermietheither micmie typic calciustollCalciu stoll changes in leaf size shape andor thickness is with the leaf area ratio LAR method rad- resultresuit iromfrom water nutrient and chemical ford 1967 LAR is defined as the ratio of leaf stresses cutler et al 1977 curtis and luchlibuchli area per unit weight of plant material the slope 1987 foliar surface area of irregular shaped coefficient 30PO of the linear regression equation tree leaves has been estimated by coating the is a ratio of leaf area to leaf biomass and is leaves with a monomonolayerlayer of glass beads and mea- defined as the leaf area ratio LAR one sided suring displacement thompson and leyton leaf area mmoven dry leaf weight g LAR 1971 and by estimating from photographs represents p0paB in each regression equation miller and schultz 1987 miller et al 1987 where Y pox LAI can be calculated as the estimated total surface area ofjuniperof juniper foliage product of LAR and live biomass per unit area ffiomaiomrom projected leaf area determined from a leaf the objective of this study was to determine area meter miller et al suggested this method LARs for selected rangeland species underestimatedundeidundei estimated leaf area by 10 due to leaf overlap cregg 1992 reported that leaf area MATERIALS AND METHODS could be satisfactorily estimated from leaf weight or volume for juniperusJumperus virginiajavirginianavirgimanavirgvirginianaimana and the study area included nine range sites in J copulorumpopulorumscopulorum however leaf area relationships five states and was part of the USDA water differed by crown position and seed source erosion prediction project WEPP table 1 sapwood area stem diameter tree height the dominant plants on each range site were canopy area and canopy volume have been evaluated LARs for 15 grasses 12 shrubs and correlatedcolcoieol related to total shrub biomass and leafbioleaf bioblo I1 tree were developed table 2 selected mass ludwig et al 1975 brown 1976 ritten rangeland species were sampled once during house and sneva 1977 whisenant and burzlaff the summer of 1987 near tombstone arizona 1978 ganskopp and miller 1986 hughes et al and in 1987 near meeker colorado sidney 1987 in contrast only a few studies have esti- montana chickashaChickasha ft supply and wood mated leaf area and LAI for langelandrangeland plant ward oklahoma and sonora texas sites sea- communities coffgoff1985goff 1985 ganskopp and miller sonal fluctuations in LAR for three shrubs and 1986 and ansley et al 1992 one tree were evaluated near alice texas in an effective method is needed to improve 1985 and 1986 LAI estimates for natural resource models one for leaf area determination grass leafbiomass potential approach for improving LAI estimates from 10 randomly located 025 mm2ma quadratsquadquadransrats was 199211992 RANGELAND LEAF AREA RATIOS 239

TABLETABLF 2 location of study sites sample dates height class number ofot samples and species evaluated toitolfolfor leaiealeafarealeaffareaarea to leaf biorbiomassnass irelationshipselation ships

leightheight class m speciessp beles

location sample 0 111 1 2 2 3 343 44 4 common name scientific name date tombstone AZ aug 1983 6 6 little leaf sumac rhus micromicrophyllaphylla engelm aug 1983 7 8 tarbush flourensiaFlour ensia cemuacamua DC aug 1983 8 broom snakeweed cutierreiacutlerGutiergutierreziagutierrezbareziafeelaveela sarothraesarothrae pursh britt &rusbybusbybrusby&rusby aug 1983 10 10 Creosotecreosotebushbush larrea tntritrldentata DC coville aug 1983 15 desert zinnia zinnia pumila daymaygray aug 1983 15 manolamariola parthenium incanumincamtin HBK meeker CO june 1987 10 shadscaleShad scale saltbush atnpkxatriplexatrlplex confertifoliaconfertifolia tootontorr & frem wats june 1987 10 wyoming big sagebrush artemisia tntritrldentata subsp wyomingensis beetle & young sidney MT july 1987 10 needleneedie and thread stipa comatacamata trin & duprrupr july 1987 10 western wheatgiasswheatgrasswheatgrass agmpynnagi ranron sinitsmithasmithnsinithiihiihilhll redbrydb chickashaChickasha OK june 1987 10 indiangiassindiangrassIndian grass sorohastrumsorghastruin hutansnutans L nash june 1987 10 big biublubluestemestern andropogon gerardugerardiigerardiagerardii vitman june 1987 10 little bluestem schiachynumschizachyritun panumscopariusicopanumscopariumico michxmicha nash chickashaChickasha OK june 1987 10 buffalograssBuffalograss buchloe dactyldactyloidesdactyloulesoidesoulesoldes nutt engelm june 1987 10 scribners dichantheliurndichanthelium dichanthelhondichanthelium ofiloligosanthesrosanthesoosgosanthesgos anthes schult guild var scribnerscnhnenanumscribnerlantanlantan nash couldgould june 1987 10 sand paspapaspalumlurn paspalum setaceumcetaceum michamichx vaivalvar stramineum nash D banks ft supply OK june 1987 10 sand sagebrush artemisiafififoliaartemisia filiflitfilifoliafoliapolla torr june 1987 10 tall dropseed sporobolus asper michamichx kunth june 1987 10 sand lovegrasslonegrasslovegrass eragrostiserogrostis chodestntritrltnchodestrichodes nutt wood woodwardWoodwaidwald OK june 1987 10 hairy grama bouteloua hirsutahirsute lag june 1987 10 sideoatsSideoats grama bouteloua curtipendulactcrtipendula mieMicmiclixmichxmiclialixkix tontootorr alice TX may 1985 4 4 4 4 4 honey mesquite prosopis glanduloseglandulosa tontorrtorn vaivar glandulosaglandulose torr cockllcockal aug 1985 2 2 2 2 2 nov 1985 2 2 2 2 2 jan 1986 NAnau api 1986 2 2 2 2 2 may 1985 5 5 5 5 5 lime prickly ash zanthovdumfagarazanthoxylmnfagara L sarg aug 1985 3 3 3 3 nov 1985 3 3 3 3 jan 1986 3 3 3 3 aarapr 1986 3 3 3 3 may 1985 5 5 texas colubrinacolubnna colubrinaColubnna texensis T & G dayoaygray aug 1985 5 5 nov 1985 5 5 jan 1986 5 5 api 1986 5 5 may 1985 5 5 texas persimmon diospyros texana scheele aug 1985 5 5 nov 1985 5 5 jan 1986 NA api 1986 5 5 sonolasonora TX june 1987 10 white tndenstedenstridens ttckmtedenstndens albescemalbascensalbesalbascemcens vasey woot & standl june 1987 10 curly mesquite hilaria belangeriabelangenabeibelbelanbeianangenageria steudstaud nash june 1987 10 texas wintergrasswmtergiass stipa leucottkhaleucotnchaleucotTkhamha trin & rupi no saniclesanipleunbitunpit collected toiforfoifol deciduousclecicln shrubs andinclticestives 240 GREAT BABASINSIN naturalistNATU RALIST aohmohvohvolumejime 52

a1 TABLITABLE 3 mean and standardstan daiddald errorarorerror of leaf biomass and leafaiealeafaleaf areaiealea aiandid linear regressionregressigressl on model slopesiopesllslisilope coefficientcoefficients s larb i elatingrelating leaiealeafarealeaffareaarea to leaf biomass lorforf or selected rangeland grassesgibassesrasses and shrubss sampled afterabterabheraiitereter full leaf cextension

leaf biomass SE leaf area SE LAR r2 2 2 1 species g mm mm 9

GRASSESGKASSFS needle and thread 36 080 3580 900 1040 98 western wheatwheatgrassgrass 20 033 5760 902 2910 98 indiangrassIndiangrass 85 156 82670 1350 9440 96 little bluestemblubiuestern 27 038 28030 4710 10780 98 big blubiubluestemestern 13 045 11290 2213 12970 86 buffalo grass 15 022 6820 1091 5680 97 scribners dichanthehumdichanthchum 13 021 15300 2601 16110 96 sand paspalum 15 023 7580 1136 6890 95 1aaatalltalitaii dropseed 09 015 8500 1334 9390 99 sand lovegrasslonegrasslovegrass 08 012 8650 1383 11380 98 hairyhalryiiairygramagrama 07 013 4360 769 5890 99 sideoatsSideoats grama 06 022 5240 2836 10210 98 white tedenstndenstridens 07 016 3980 1007 5830 98 texas wintergrass 12 024 8320 1361 6720 95 curly mesquite 08 015 5270 925 6620 99

slikubssilliulisSilli ulis desert zinniamniaamnia 16 010 9440 580 5700 89 manolamariola 35 040 19410 1280 5690 84 bloombroom snakeweed 37 051 11160 920 2700 96 little leaf sumac 39 071 22050 331 4700 91 tarbush 37 100 23360 203 6100 97 Creosotecreosotebushbush 30 019 16790 910 3660 86 sand sagebrush 32 058 5950 1257 2010 98 shadscaleShad scale saltbush 39 081 10530 2047 2640 98 wyoming big sagebrush 53 083 18220 2715 3340 97

all areimeighttic i wcilil regressionslegiessions wenwere signifisignifysignihnntit atit P 05 1 kawvaratioI il in 11 tholahicpilsliihlarrpvs t P3iiineliin leicionleilerierlefcioneloncionelod whwilenwilca Y PXP X

used grass biomass in each quadrat was clipped sample cube was placed in an area considered to a 20 mm stubble height and separated by representative of the entire canopy and the species into live or dead leaves live leaves were leaves within the area were removed by hand placed in plastic bags on ice for later determina- LARs were determined in the same manner as tion of leaf area the leaves were flattened and for grasses placed between clear plastic sheets and then within plant variability of LARs was evalu- processed through a leaf area meter leaf area ated for four mesquite trees and four lime was determined with a lilicorcor 30003 leaf area prickly ash shrubs in may 1985 near alice meter to the nearest I1 mmmm2mma the samples were texas fifteen sample cubes were randomly then oven dried at 60 C for three days and dry located and sampled from each of the four mes- mass determined quite trees for the lime prickly ash shrubs 12 to ensure that samples of shrubs and trees sample cubes were harvested from each of the represented the full range of size of plants pres- four shrubs LAR was determined in the same ent a stratified random sampling procedure was manner as previously described A one way used height classes of I1 m were arbitrarily analysis of variance was used to test for differ chosen and plants were selected randomly from ancesences P 0505.05 among the slopes of the regres- each class As a result total number of plants sion equations within plant canopy by species sampled varied among species depending upon steel and torrie 1980 within plant LARs the range of plant heights table 2 were not significantly different for lime prickly side an open ended cube 250 mm on a was ash and mesquite in may 1985 based on these used biomass to sample shrub and tree leaf the relationships one sample per plant was utilized during the remainder of the study tbtiie uuse ofaof a tirletraa 0 r finn nanienanlename iniii thistins papelpaper isis floltoiiol reader inforinatiinfoi matlonmotion andadd three shrubs lime prickly ash texas per- doesdo notot iimplylayidyJAY endoiseinentadondo nt by ththe USU S decartidepartideputinentt of aadagdagncnltineiture ofayof any prodpioclnclt or01 servicei simmon and texas colubrina and one tree 199211992 RANGELAND LEAF AREA RATIOS 241

TABLETABLF 4 mean and standard error ofofleafbiomassleaf biomass and leaf area and linear regression model slope coefficients lablarlLARILAR relating leaf area to leafleafbiomassbiomass for selected rangeland shrubs and tree on a fine sandy loam range site near alice texas

species date leafL eaf biomass SE leaf area SE LAR r2 2 2 I1 g mm imnmmyammyg

lime prickly ash may 1985 47 073 45180 1450 8760 aca 99 aug 1985 42 063 40330 1530 8730 a 98 nov 1985 56 089 43360 1460 8670 a 98 jan 1985 49 076 44310 1450 8870 a 98 aarapr 1986 53 065 52730 1580 8690 a 98 mesquite may 1985 65 087 57830 1610 8990 a 98 aug 1985 57 064 56040 1470 8780 a 98 nov 1985 55 070 4846048.460 14101.410 86308.630 a 98 jan 1985 nainad aarapr 1986 64 081 59100 1470 9290 a 98 texas persimmon may 1985 46 064 49960 1940 10590 b 96 aug 1985 41 065 41670 1780 10360 b 98 nov 1985 48 059 51060 1790 1010130130 b 98 jan 1986 46 068 44720 1900 10020 b 98 aarapr 1986 47 069 64150 2070 12660 a 97 texas colubrinacolub nna may 1985 49 078 55070 2020 10310 b 98 aug 1985 52 089 57010 1720 10110 b 98 nov 1985 3818 065 55380 2090 1336013 360 a 98 jan 1986 NA aarapr 1986 41 071 41760 1880 10230 b 98

all neiarea weight regressions weieweresignificantwere significantsigmfic mt atit P 05015 leleaff arealieuieule 11ratioitlo LAR icpresentsrepresents PIpo min eeachtchichteh regressionlegie&sion where Y poxrox pvparajnctersimeteismin the column by species sharing a corcommonninon letterietter lielleileare not significantly different PF 505sos05 basedbisect on homogeneity1 gerritygerwity ofot slope test no impiesarnples witsw is collected forfoimoloor deciduousshrubsdeciduous shrubs honey mesquite were selected for evaluation of among the slopes of the regression equations seasonal fluctuation in LAR honey mesquite LAR between sample periods within species texas persimmon and texas colubrina are steel and torrie 1980 drought deciduous while lime prickly ash is an evergreen sample dates were selected to cor- RESULTS AND discussion respond to the phenologicalphonological stages oflkoflof 1 maxi- mum leaf area 2 peak drought defoliation 3 leaf area of gramingraminoidsoids was highly corre- autumn just prior to winter leaf fall and dor- lated with leaf biomass for all species within mancy and 4 after winter leaf fall for the sample dates table 3 the LAR for perennial mma deciduous shrub grass leaf area ranged from 2910 to 16110 mm2 g LAR for shrubs and from the statistical analysis system SAS 1982 the trees ranged 2010 to 13360 mm2mma g goff 1985 also was utilized to evaluate linear regression rela- reported significant linear regression relation- tiontionshipsships Y po pix between leafbiomassleaf biomass bix ships faf2 8397838397.839783.83 htgt97.97 for LAR for 11 native grass and leaf area where Y is estimated leaf area species in southern arizona goff reported that mm2mma po is the Ppi is the slope LAR intercept the linear regression coefficients for stem area coefficient as defined by 1967 mm2mma radford in to stem biomass SAR ranged from 32 to 73 1 and X is leaf biomass 9 leafbiomass g the intercept was of the LAR and the mean SAR was 44 of the tested to determine if it was significantly differ- mean LAR ent FP 0505.05 from zero the intercept was not there was no significant seasonal variation significantly different from zero for all species in LAR for lime prickly ash and mesquite table 4 therefore the data were reanalyzed and pre- although there was no significant seasonal dif- sented using a linear regression model Y ference between mesquite LAR relationships a fonPONpomlox similar to that reported by coombs et al gradual decrease in the LAR from may through 1987 and ansley et al 1992 for estimating november was apparent in 1985 furthermore LAR all statistical tests were judged significant the LAR was larger in april 1986 though it was at P 05os05.05 unless otherwise stated A homogene- not significantly different from 1985 sampling ity of slope test was used to test for differences dates mooney et al 1977 found that the specific 242 GREAT BASIN naturalist volume 52

leaf density mg mm 2 of mesquite leaves are initially light green in color and become increased over the growing season the density glabrous after elongation ceases As the leaf ranged fromfroni 000040.0004 mg mm 2 in the spring to matures the xylem and bundle fibers become 2 00170.017 mg mm in the fall this corresponds with increasingly lignifiedsignified and the leaf turns dark mma 1 a leaf area change of 5880 to 25000 mm2 g greencreen with the underside becoming densely et al 1992 north central ansley working in covered with trichomes leaf modification is texas reported that LAR of mesquite ranged 1 complete by early july the lower LAR of texas from 9916 to 5944 mm2mma gg1ga LAR mesquite persimmon leaves in 1986 was attributed to the declined from may through august 1987 but leaves not being fully elongated with stabilized from august through september fol- incomplete development of trichomes and lig- lowing substantial precipitation in 1988 precip- itation was substantially less than in 1987 and nification the mean LAR was significantly lower than in LAR relationships for texas colubrina varied 1987 LAR followed the same pattern in 1988 seasonally LAR was similar during the early declining from a high of 6877 in the spring to a growing seasons in may 1985 and april 1986 low of 4996 inmainm2mm2 g 1 in october ansley et al and in august 1985 in november the LAR was 1992 speculated that the decline in LAR was 33 greater than during other sample dates caused by cell wall thickening in response to table 4 basal leaves of texas colubrina are drying conditions based on the work ofofkramerkramer approximately 10 times larger than the outer and kozlowski 1979 canopy leaves in response to an extended dry the similarity in LAR across sampling dates period in july and august texas colubrina from this study may be partially explained in that dropped 95 of its leaves the only leaves sampling was not initiated until all leaves were retained during this dry period were the large fully expanded for approximately foulroulfour weeks in basal leaves in the center of the shrub the may addition april june and september pre- significant difference in LAR between the was significantly above cipitation the longtermlongiong terillterm sample dates was attributed to the different average precipitation and no noticeable water proportion of leaf types and not the change in stress was apparent in the trees sampled nilsen specific weight of the leaves et al 1986 indicated that relative leaf area of and miller phreatophytic mesquite FP glandulosaglanduloselandhlandulosalandulosaiosa var tor ganskopp 1986 reported sim- rewinarewindreyanaregana in the sonoran desert of southern cal ilar significant seasonal changes in LAR for iforniaifornia remained nearly constant from may wyoming big sagebrush they speculated that through november maximum leaf area was the greatest proportion of seasonal variation was maintained throughout the hottest and driest due not to the development or alterations in months of the year via access ofdeepof deep stored soil starch and sugar accumulations but rather to water by taproots when water availability to the changes in the proportion of larger persistent normally phreatophytic mesquite was reduced leaves to smaller ephemeral leaves total leafarea was reduced nilsen virginia and shrub leaf biomass to leaf area was highly jarrell 1986 we hypothesized that mesquite correlated for the nine other shrubs sampled leaves reach a stable weight at maturity and the table 3 the LAR for shrub leaf area ranged lack of stress the 1 water during growing season from 2010 to 6100 mmmm2mma g other researchers the changes in leaf weight to leaf prevents area have also reported satisfactory results in relating reported by ansley et al 1992 in leaf changes leaf biomass to leaf area schilesingerschllesingerSchilesinger and weight as a result of translocation of sugars chabot 1977 kaufmann et al 1982 ganskopp starches other compounds and insect damage and miller 1986 within date could not be detected or separated from cell sample based on wall thickening from water stress within the the seasonal variability in LAR for texas mersimpersim precision of sampling in our study mon and texas colubrina in this study and the texas persimmon LAR in april 1986 was findings ofganskopp and miller 1986 in eastern significantly greater than for sampling dates in oregon for wyoming big sagebrush we can state 1985 meyermeyer19741974 reported that texas persim- that seasonal variability in these and other mon produces two types of leaves a large leaf in drought deciduous shrubs is an important source the center of the canopy and a smaller leaf of variation that needs to be accounted for when around the perimeter of the plant the leaves simulating LAI over the entire growing season 199219921 RANGELAND LEAF AREA RATIOS 243

conclusion series no 1 ind ed society of range management denver colorado kendallhuntKendall Hunt publishing co dubuque iowa for the species sampled leaf biomass is a broynBRO N J K 1976 estimating shrub biomass from basal reliable estimator of leaf area however for stem diameters canada journalloinjoin nalnat of forest researchreseal ch 6 some shrub species seasonal differences in 153 158 of BROWN R W 1977 water relations of i angerange plants pages development and shedding different types of 98 140 in R E sosebee ed Ranaranoranaelandrangeland11eland plant physi- leaves and leaf morphological development can ology range science series no 4 society of range produce significant temporal fluctuations in management denver coloradocoloichloi ado LAR caldwell et al 1981 reported that for CALDWFLICALDWELL M MMJJ H RICHARDS D A JOIINSONJOUNSON R S semiarid bunchgrassesbuncligrassesbunch grasses leaf blades of regrowing novvakNOWAK andandrR S durecdzurec 1981 coping with herbicheibivherbiv ory photosynthetic capacity and resource allocation in tillers had greater photosynthetic capacity than two semiarid agropyron buncligrassesbunchgrassesbunch grasses Occooecologiaoccologialogia 50 blades on undippedunclipped plants this resulted in 14 24 greater carbon gain for clipped plants and an CARLSON D HHTT L TIIURO R W KNICIIIKNIGHT andandrR K increased photosynthesistranspirationphotosynthesis transpiration ratio HEhenhehHFH1tsc11ms IIMIDIDT1 1990 effect of honey mesquite on the nal of range the water balance of texas i oilingrolling plains journaljoin nowak and caldwell 1984 reported that management 43 486 490 photosynthetic rate for both clipped and un- COOMBS J D 0 HALL S P LONG andandlJ M 0 scurS UR dippedclipped plants decreasedaithdecreased Aithwith age ofthe leaves LOCK 1987 techniques in productivitybioproductivitybiopioductivitybioblo and photo- current rangeland hydrologic simulation synthesis and2nd edition pergamon piessplesspress new york crecoCRECRFCCcrego RB M 1992 area of foliage of models do not account for changes in LAR or leaf estimation in juniperus forest science 38 61 67 as a function of evapotranspiration rates of age cuCURTISRTIS PPSS and A luctillluceilllucluo IILI 1987 the effect of moderateofmodcrate the leaf proportion of leaf type or bompensacompensacompensa salt stiessstress on leaf anatomy in hibiscushibiHihi icus cannacannabinuscannahiniisbinus tory photosynthesis rate increases following kenaf and its relation to leaf area american journal defoliation due to grazing models currently of botany 74 538 542 CUILFRCUTLER J R D W RAINS andandrR S ILOOMISkomis 1977 the utilize a fixed coefficient for calculating LAI culler RD if importance of cell size in water relationslelaielarelahons of plants significant advances in modeling evapotranspi physiology of plants 40 255 260 ration on rangelands are to be made GANSKOPP D and R millekMILLERMILLFR 1986 estimating leaf area improvements in the relationships used to sim- of big sagebrush from measurement of sapwood jour- nal range 39 338 340 ulate evapotranspiration that incorporate these of management GAY L wandlW and L J FRIISCHFN 1979 an energy budget processes will be needed the LAR method of analysis of watelwater use by salt cedar water resource calculating LAI evaluated in this study provides bulletin 26 1589 1597 a fast reliable method of estimating LAI neces- GIFFORD G F 1975 approximate annual water budget of range sary to parameterize these hydrologic simula- two chained pinyon juniperjumper sites journal of management 38 73 74 for seasonal tion models to account the GOFF B F 1985 dynamics of canopy structure and soil differences in LAR for texas persimmon and surface cover in a semiarid grassland unpublished texas colubrina a weighted average based on mastelmaster s thesis universityumveiumpei litysityvity of arizona tucson of is recommended for HANSON C L 1976 model foifor predicting cvapotranspiraevapotranspna season year parameter- great plants like tion from native rangelands in the northern izing the WEPPWEPP model for mesquite plains transactions of the american society of agri- and lime prickly ash one LAR value can be used cultural engineering 19 471 477 in non drought years for years with significant huchucheshughesHuGHUGHFSilesliFs H GLG L W VARNFRVARNER and L H blankenship dry periods a decrease in LAR of 10 40 may 1987 estimating shrub productionpioploduction from plant dimen- need to be accounted for with non phreato sions journal of range management 40 367369367 369 KAUFMANN M R C B EDMINSIFREDMINSTER and C A TROEN- as indicated by this work and RC phytic mesquite DLFD LE 1982 leaf area detdeteimmationsdetermiderminermiD ationsactions for subalpmetieesubalpine tree that ansleyofansleyof et al 1992 species in the central rocky mountains USDA foiestforest service research paper RM 228 KRAMERKRAMFR P J and T T KOLOWSKI 1979 physiology of literature CITED woody plants academic press new york ianeLANELANF L J andandmanamM ANEARINGANFARING 1989 USDA water ero- prediction ANSLFYANSLEY R J D L pricePRIC F S L dowhowerDowil OWER and D H sion project hillslopehillslope profile model docu- CARLSON 1992 seasonal trends in leaf area of honey mentation USDAARSUSDA ARS national soil erosion mesquite trees determination using image analysis research laboiatorylaboratoryLaboi atory purdue university report 2 journal of range management 45 339 344 west lafayette indiana ARNOLD J G J R WILLIAMS A D NICKS and N B ianeLANELANF L JEJ E M ROMNEYROMNFY and T E HAKONSON 1984 SAMMONS 1990 SWRRB a basin scale simulation water balance calculations and net production of model foifolformol soil and water resource managementmanagernent texas perennial vegetation in the noinorthernnor thern mojave desert aama&m university press college station journal of range management 37 12 18 BRANSON F A G F GIFFORD K G RENARDRFNARD and R F LAUFNROIHLAUENROTH W K and P L SIMS 1976 evapotranspira hauleyHADLEYHAULFY 1981 rangeland hydrology range science tion fromflom a shortshortgrassgrass prairie subjected to water and 244 GREAT BASIN naturalist volume 52

nitrogen treatments water resource research 12 SAS 1982 SAS useruserss guide basics SAS institute inc catygatycarygary 437 442 north carolina ludwicLUDWIGLUUWIC J AAJJ F REYNOLDS andeand P D WIIHSONWHITSON 1975 schilesingfrsc iiilesingfr W H and B F CHABOTCHABCH 1977 the use of siesize biomass relationships of several chihuahuancbihuahuan water and minerals by evergreen and deciduous shrubs desert shrubs american midland naturalist 94 451 in the okefenokee swamp botanical gazette 138490138 490 461 497 mlmeyeiimeyeisYI R R E 1974 morphology and anatomy of texas SIFFLSTEEL R G Ddandldanajdandjand J H TORRIFTORRIE 1980 principles and persimmon diospyros texana scheele texas agricul- procedures of statistics a biometric approach tural experiment station bulletin 1147 texas a&maam mcgraw hill book company new york universityumveiumpei sity college station THOMPSON F bandlB and L leylonLEYIONLEYTON 1971 method for mea- MILLER 11 FFLL E EUDLFMANEDDLEMAN and R F ancelANCFLANGEL 1987 suring the leaf surface area of complex shoots nature relationship ofwesternjuniperwestern juniper stem conducting tissue 29 572 and basal cucumferencecircumference to leaf area and biomass WATSON D J 1947 comparative physiological studies on great basin naturalist 47 349 354 the growth of field crops 1 variation in net assimilation

1 and leaf and mhitiimilmii 1 K R F and L M schultzSLIIULI 1987 water relations rate area between species and varieties fandl within and leaf morphology of juniperus occidentoccidentalisocciclentalisoccidentalistalisallsails in the and between years annals of botany 11 41 76 northernn great basin forest scienceselenceseledee 33 690 706 WHISENANT S gandaganddG and D F BURZLAFF 1978 predicting moonlyMOONIYMOONEY H ABA B B SIMPSON and 0 T SOLBRIG 1977 green weight of mesquite prosopis glandulosaglandulose torr of 31 395 397 pages 26 44 in B B simpson ed mesquite its biol- journal range management WIGHT R 1971 of and ogy in two ecosystems dowden hutchinshutehinshutchasonhutchmson and J comparison lysimeter neutron ross inc stroudsburgStroudsburg pennsylvania scatter techniques for measuring evapotranspiration from semiarid rangelands of range manage- NIISINH senN E T M R SHARIFISIIARIFI P W RUNDFL and R A journal 390 VIIINIA 1986 influences of microclimatic condi- ment 24 393 1986 ERHYMIIII11 model description and user tions and water i elationrelations s on seasonal dimorphism of ERHYM guide for the basic USDA agricultural prosopis glandulosaglanduloseianlanlandulowdulomdulow var torrtorreuanatorreyanareyanareganaeuana in the sonoran version tor research service ARS 59 infor- desert california oecologiaoecologia6969 95 100 national technical mation service springfield virginia NILSENNIL SIN E T R A VIRGINIA and W M JARRELLJARRFLL 1986 WIGHIWIGHT J R and L HANSON 1990 crop coefficients water relations and growth characteristics of prosopis C for rangeland journal of range management 43 482 glandulosaglandulose var torreyanatorreyana in a simulated phreatophytic in 485 enviionmentenvironment american journal of botany 7342773 427 433 WIGIIIWIGHT J R C L elansonHANSONPIANSON and K R COOLEY 1986 NOWAK R sanamS and M MCALDWELLM CALDWF LL 1984 A test of com- sandm modehngmodebdg evapotranspiration from sagebrush grass pensatorypen satory photosynthesis min the field implications for rangelands journal of range management 31 81 85 herbivoreherbivoryherbheihel bijorybivoryivory tolerance oecologia 61 311 318 WIGHT J R and J W SKILESSKILFS edskusFUSFDS 1987 SPUR simulation RAOIORDRADFORD P 1967 growth analysis formulae their use wlglirj randjranda PJJ of production and utilization of ranglandsranrangelandsglands documen- and abuse crop science 7 171 175 clop tation and user guide USU S department of agricul- T 1972 A model for rrriiierilrii lliralir J predicting evaporation ture agricultural research service ARS 63 boise fromflom a row crop with incomplete cover water idaho resource research 8 1204 1213 rllllnlfousrRITTFIN I1 OUSE L R and F A SNFVASNEVA 1977 A technique foifollorlurfor estimating big sagebrush production journal of received 5 june 1991 range management 30 68 70 accepted 10 september 1992 great basin naturalist 523 appp 245 252

ECOLOGY AND management OF medusaheadMEDUSAHEAD taeniatherumTAENIATHERUM CAPUT MEDUSAE SSPSSE ASPERUM SIMK MELDERIS

james A young

ABSTRACT medusaheadMedusahead is another in the extensive list of annual herbaceous species to invade the temperate desert rangelands of the great basin medusaheadmedusalieadMedusahead is not preferred by large herbherbivoresivores and apparently is not preferred by gramvoiesgranivoresgrangramivoresvoiesvoles herbage of this annual grass enhances ignition and spread of wildfireswildfires medusaeadmedusaeanmedusaheadMedusMedusA aheadead is highly competitive with the seedlings of native species and isis probably the greatest threat to the diversitybiodiversitybio of the natural vegetation that has yet been accidentally introduced into the great basin despite the obvious biological disruptions that are associated with medusaheadmedus ahead invasion the species offers a wealth of opportunities for students to examine the mechanism by which this species is so successful students of evolution plant physiology and ecology may find this species to be an excellent model for colonization

key words medusaheadmedusahead taeniatherumtaematherumTaeniaTaema therum caput medusaemedusan annual grass colonizing species wildfireswildfires grazing

in the management of natural resources cies there has been confusion about the correct there are certain problems that by their persis- scientific taxon for medusaheadmedusahead the first tence magnitude of ecological disruption and description of medusaheadmedus ahead in a north american economic impact refuse to dissipate as a result flora used the taxon elymus caput medusaemedusan L of being ignored and neglected unfortunately howell 1903 there is apparent agreement for range management medusaheadmedus ahead that medusaheadmedus ahead is a member of the tribe taeniatherumTaeniatherum caput medusanmedusae LLJ nevskianevski is triticeae of the grass family there is also appar- that of type problem during the 1950s ent agreement among morphologists and catocyto medusaheadmedusahead was considered among the most geneticists that medusaheadmedus ahead does not fit in the pressing problems on the rangelands of califor- genus elymus various authors have placed and A nia idaho oregon great deal of research medusaheadmedus ahead in hordeum or hordelymusHordelymus effort was devoted to solving the medusaheadmedus ahead nevskianevski 1934 proposed that medusaheadmedusahead was problem valuable information was learned truly a different genus and published the name about the ecophysiologyeco and synecology of physiology taeniatherumTaeniatherum jack major of the university of medusaheadmedus ahead control methods were developed california suggested in 1960 that material intro- usingusinausino herbicidesherbicides fatal link in integrated b the duced to the united states was taeniatherumTaeniatherum programs for the suppression of medusaheadmedus ahead asperumesperum major et al 1960 based on the populations proved to be artificial revegetation european and russian literature major technologies after medusaheadmedusabeadmedusmedusa beadheadahead was controlled reported that taeniatherumTaeniatherum contained three the nature of the sites infested had more to do geographic and distinct taxa with this failure than the weed itself especially morphologically T caput wdusaemedusanmedusae T as and T in the intermountain area the recent discovery perum crinitumcrinitum three found of medusaheadmedusahead in northern utah has renewed these species are in the mediterra- interest in suppressing this rangeland weed nean region and extend eastward into central the european my purpose in this review is to refresh our asia after examining material collective memories about medusaheadmedus ahead ecology growing in place major decided the united and management states introduction was T asperum the danish scientist signe frederiksen TAXONOMY revised the genus in 1986 he kept the same three taxa but reduced them to subspecies of As is often the case with an introduced spe taeniatherumTaeniatherum caput medusanmedusae positive identifiidentify

agricultural restuchreseuchReseaRe seuchml seasemservice US decartidepartidepartmentit Agricultureofagricultureof 920vilieyroicl920 vaileyvalley road henorenoremo nevada 89512

245 246 GREAT BASIN naturalist volume 52 cation to the lowest level possible is absolutely south of steptoe butte sharp and tisdale essential for any proposed biological control 1952 fred renner told jack major he had seen program for medusaheadmedus ahead according to medusaheadmedus ahead near mountain home idaho as frederiksen s revision subspecies crinitumcrinitum has early as 1930 and lee sharp had reports from a very strict spike subspecies caput medusaemedusannwdusae ranchers that the species occurred in idaho as has a large open spike with straight awnsagns the early as 1942 the medusaheadmedus ahead infestation in spike of subspecies esperumasperum is intermediate idaho increased to 30000 acres by 1952 min with angled awnsagns subspecies asperasasperamasperum is the hironaka estimated that 150000 acres were only one of the three with pronounced barbs infested by 1955 and the bureau of land man coated with silica on the awnsagns apparently the agementargement estimated 700000 acres were infested correct taxon for the medusaheadmedus ahead of western by 1959 at that rate of spread it appeared that north america is taeniatherumTaeniatherum caput medusanmedusae all of idaho would be infested by the end of the sspasp aspenimaspemm simk melderis frederiksen next decade the spread of medusaheadmedus ahead slowed 1986 and nearly continuous infestations remained taeniatherumtaeniathentmTaeniatherum caput medusanmedusae sspasp caput confined to gem payette and washington twdusaemedusanmedusae is mostly restricted to portugal spain counties in southwestern idaho there were southerns outhernbern france morocco and algeria it has several spot infestations in surrounding counties been collected outside this area in europe and hironaka and tisdale 1958 asia but frederiksen considers it adventitious medusaheadMedusahead spread south in california to in these areas subspecies crinitumcrinitum is found santa barbara on the southern coast and fresno from greece and yugoslavia eastward into asia county in the interior valleys the rapid spread subspecies asperumesperum completely overlaps the from southwestern oregon through northern distribution of the other two subspecies all and central california occurred in annual dom three subspecies integrate with each other inated grassland oak quercus woodland and apparently only the one subspecies occurs in chaparral communities these areas have a north america does this indicate one or very mediterranean type climate with hot dry sum- limited introductions mers and cool moist falls winters and springs medusaheadMedus ahead is predominantly self polli germination occurs in the fall and flowering natedbated genetically the genus appears to stand and seed set in the spring alone in genomic relations within the triticeae in northeastern california east of the sierra schooler 1966 sakamoto 1973 apparently nevada cascade rim medusaheadmedus ahead invasion taeniatherumtaeniathenimTaeniatherum has a genome that is distinct but occurred at a much slower rate in the pitt river faintly related to those of psathyrostachys drainage vegetation is an intergrade of oregon eremopyrum or hordeum daypyntmdasypyrumDasypyrum Eremopyrum white oak quercus oafcarruanaoarruanaoargarryanagarruanafyanaryana woodlands frederiksen and bothner 1989 cismontane california species western juniper juniperus occidentoccidentalisoccidentalistalis ponderosa pine pinus HISTORY IN NORTH AMERICA ponderosa woodlands and sagebrush artemi siabunchgrass5fabunchgrass communities more typical of medusaheadmedusalreadMedus aheadalread was first collected in the the intermountain area united states near roseburg oregon on 24 medusaheadMedusahead was discovered in the great june 1887 by thomas jefferson howell 1903 basin at verdi nevada in the early 1960s iso- it was next collected near steptoe butte in east lated infestations were subsequently found ern washington in 1901 by george vasey piper along the eastern front of the sierra nevada in and beattie 1914 followed by a collection near areas where range sheep bands used to concen- los gatos california in 1908 by charles hitch- trate while waiting for mountain summer pas- cock jepson 1923 medusaheadMedusahead certainly tures to be free of snow attracted the noted agragrologistagriologistologist mckell robin- in northeastern california in the great basin son and major 1962 commented on this during the early 1960s there were two small strange initial distribution reaching 390 miles infestations in city lots in Susanville and a small north and 450 miles south from the point of infestation at the old sheepshearingsheep shearing site of initial collection early herbarium specimens viehlandviewlandViewland along the railroad above wendel cal- show a rapid spread to the south into california ifornia another isolated infestation occurred at J F pechanec made the first collection in the mouth of fandango pass in surprise valley idaho in 1944 near payette or about 180 miles by the early 1970s medusaheadmedus ahead was nearly 199219921 ECOLOGY AND management OF medusaheadMEDUS AHEAD 247 continuous over about 60000 acres of the seeds touching a moisture supplying substrate willow creek tablelandsTablelands northeast of susan in this situation germination of medusaheadmedusahead ville currently after four years of extreme seeds is controlled by the relative humidity drought medusaheadmedusahead spot infestations occur within the litter and the incubation tempera- over perhaps an additional million acres on the ture which of course influences the relative western margin of the great basin humidity the needlelike vitreous caryopses of medusaheadmedusabeadmedusmedusabeadheadahead appear hydrophobic rather than medusaheadmedus ahead seeds BIOLOGY OF medusaheadMEDUS AHEAD hygroscopic not only can germinate under these conditions but they can be dried until the primary root is dead then medusaheadMedusahead in some ways is a rerun of following remoisteningre a new adventitious root cheatgrasscheatgrass bromus tectoriumtectorumtectorum invasion moistening will develop cheatgrassCheat grass dominates secondary succession in evans and I1 demonstrated what a a majority of sagebrushbunchgrass communi- raymond modifying influence litter cover can be to ties in the great basin and provides a significant great the surface of seedbeds on temperate desert portion of the forage base for livestock grazing of however there are highly significant differ- rangelands in terms reducing extremes in temperature and conserving moisture evans ences in the ecology of the two grass species ori caryopses harris andwilsonand wilson 1970 al dakheel 1986 and young 1970 1972 of squirreltail elymus hystrix are similar in germination the caryopsis of medusa very to those of head is less than a millimeter wide with a very morphological appearance As I1 adliwill discuss later sharp callus and an elongated non geniculated medusaheadmedus ahead squirreltail seedlings are one of the few native that awn the medusaheadmedus ahead caryopsis is covereda4thcovered with species small barbs of silica vicious is the best desendescrip-p can become established in undisturbed and tion for this grass caryopsis bovey et al 1961 medusaheadmedus ahead stands both taeniatherumTaeniatherum determined that medusaheadmedusahead had a much elymus are members of the tribe triticeae but higher ash content over 10 than other grass they do not share the same genome species and the ash was about 75 silica heavy medusaheadMedus ahead populations easily exceed 1000 deposition of silica occurs on the barbs of awnsagns plants per square foot and they are phenotypi- and the epidermis of leaves cally plastic enough that a population of I1 plant for the vast majority of collections of per square foot can exceed the seed production cheatcheatgrassgrass from the intermountain area seeds of 1000 plants per square foot unpublished are ready to germinate when they are mature research ARS reno nevada huhuehugee seed no germinationpregerminationprogerminationpre treatments are necessary banks develop in medusaheadmedus ahead communities in young and evans 1982 for collections from the litter and soil medusaheadMedusahead seed acquires a the great plains and perhaps the columbia dormancy in the field similar to that of basin seeds may have a brief afterripening dor- cheatgrasscheatgrass see young et al 1969 these dor- mancy in contrast seeds of medusaheadmedusahead have a mant seeds respond to enrichment of the seed- temperaturetemperature related afterripening and germi- bed with nitrate and gibberellin evans and nation will not occur except at cold incubation young 1975 temperatures for about 90 120 days after matu- LIFE CYCLE medusaheadMedusahead seeds can ger- rity young et al 1968 nelson and wilson minate in the fall winter or spring and seed- 1969 found this dormancy was controlled by lings from all seasons can produce flowers and materials located in the awn seeds early in the summer the striking thing the high silica content on the herbage of about the medusaheadmedusahead life cycle is that it medusaheadmedusahead makes the litter very slow to matures from 2 to 4 weeks later than other decompose harris 1965 described the chok- annual grasses all those famous botanists and ing accumulations of medusmedusaheadahead litter that range scientists who were out on the range dis- built up for several years we evaluated the covering new infestations of medusaheadmedus ahead were germination of seeds of various annual grass led to the populations by the bright green color species in medusaheadmedusahead litterutter young et al 1971a when all other annuals in either cismontane allelopathyAllelopathy was not suspected but rather the california or the great basin were brown physical holding of seeds out of contact with the R L piemeisel recognized the dominance surface of the seedbed medusaheadMedusahead seeds ger- of alien plant species in the secondary succes- minate very well without the callus end of the sion of disturbed sagebrush communities in the 248 GREAT BASIN naturalist volume 52

intermountain area piemeisel 1951 working in the intermountain area maynard on the snake river plains of idaho during the fosbergofthefosberg ofthe university of idaho reported that 1930s piemeisel enumerated dominance from the medusaheadmedus ahead infestations along the colum- russian thistle salsola australis to tumble bia river in washington idaho and oregon mustard sisymbrium altissimum to cheat were restricted to clay textured soils fosberg grass continued disturbance tended to per- 1965 he suggested that the greater soil mois petuate cheatgrasscheatgrass dominance according to ture holding capacity of these soils allowed piemeisel the annual species that germinates medusaheadmedus ahead to complete its life cycle first reaches maximum growth and maturity building on the work of fosberg and first has the capacity to withstand crowding hironaka I1 sampled the plant communities in and has high seed production is the one that will the medusaheadmedus ahead invasion area along the western occupy and persist in seraiseral sagebrush plant com- edge of the great basin young and evans munitiesmunities Pierpleiplervierneiselpiemeiselpierneiselneisel always noted that no one 1970 medusaheadMedusahead was found on the margins species had a clear dominance on all these char- of many degraded meadows where moisture acteristicsacte ristics but on balance cheatgrasscheatgrass was the relationships probably favored it over clear winner cheatgrasscheatgrass A much larger area of infestation medusaheadMedus ahead contradicts several of was sagebrushgrasssagebrush grass communities the sage- piemeisel s criteria medusaheadMedus ahead seeds are ini- brush communities consisted of mountain big tially dormant with temperature related sagebrush artemisia dentatatritridentatatridentate sspasp vastyanavaseyanavaseyana afterripening requirements while cheatgrasscheatgrass on soils with sandy loam to loam textured sur- seeds have no such restraints this works only face horizons and often well developed argillic for initial establishment because once seed horizons A second series of sagebrush commu- banks are established with seeds with acquired nities consisted of low sagebrush A arbuscula dormancy our research indicates that growing on soils with clay textured surface hori- cheatcheatgrassgrass and medusaheadmedus ahead seeds have equal zons harry summerfield retired soil scientist chances of germination with the initial moisture soil conservation service and forest service event in the fall medusaheadMedus ahead does take much USDA suggests the low sagebrush soils share longer to mature than cheatgrasscheatgrass and perhaps the same development as the big sagebrush tumble mustard min hironaka and his students soils but the surface horizons have been have conducted a series of excellent experiexpert removed by erosion personal communication ments comparing the cumulative growth curves on the modoc plateau of northeastern califor- for roots and aerial structures of medusaheadmedus ahead nia these two series ofplant communities divide and other grasses hironaka 1961 hironaka and the landscape about equally young et al 1977 sindelar 1973 1975 dr hironaka concluded in the northern great basin low sagebrush con- from these studies that the comparative growth stitutesstitutes only about 10 of the total sagebrush phenology restricts medusaheadmedus ahead to areas with vegetation surplus soil moisture after cheatgrasscheatgrass normally on the western edge of the great basin matures medusaheadmedus ahead in nonmeadownonmeadow situations is largely restricted to low sagebrush potential SOILS plant communities would this restriction to clay soils change over time as appears to have happened in cismontane california remem- evans noted the 1950s raymond in when ber the studies of raymond evans that showed medusaheadmedus ahead first invaded glenn and colusacalusa competition in the cismontane portion of the counties in the northern sacramento of valley california annual grasslands is initially for light california that medusaheadmedusahead appeared to be while in cheatgrasscheatgrass communities of the inter- restricted to clayelaychay textured soils personal com- mountain area competition is overwhelmingly municationmunicatloncation mallory 1960 thiss reported on thithl for soil moisture evans et al 1970 1975 relationship at the 1960 meeting of the californicaliforniaa section of the society for range management burgess kay made the chilling observation that WILDFIRES after a couple of decades this relationship disap- peared and medusaheadmedusahead occupied many sites accumulations of litter on areas where with coarser textured soils personal communi- medusaheadmedus ahead is established will burn mckell cationscations wilson and kay 1962 had initial results that 199211992 ECOLOGY AND management OF medusaheadMEDUS AHEAD 249

seemed to indicate that burning was the answer not know what the influence of medusaheadmedus ahead to the control of medusaheadmedus ahead the idea was to invasion would be on other granivoresgranivores seeds of bum stands while competing annual grasses other recently introduced weeds in temperate were fully mature and medusaheadmedus ahead seeds were desert communities such as those of barbwire still in the inflorescences this study showed russian thistle salsola paulseniipaulsenii are heavily burned seeds would not germinate however preyed upon by granivoresgranivores ifcheatgrassifcheatgrasscheatgrass popu- the burned seeds were apparently incubated at lations crash because of replacement by 20 C and unburned fresh seed would not have medusaheadmedusahead what happens to cheatgrasscheatgrass seed germinated at that temperature we tried a predators series of burning experiments on the pitt river A study conducted at washington state uni- indian reservation and found burning favored versity illustrates that granivore preference medusaheadmedus ahead young et al 1972 we helped works both ways in plant succession bird pop- forest service range conservationists evaluate ulationsulations prefer the seeds of native perennial burning treatment on low sagebrush communi- grass species over those of cheatgrasscheatgrass and ties on the silver lake district of fremont medusaheadmedus ahead goebel and berry 1976 national forest in oregon the off season burns utilization of medusaheadmedus ahead by large cerbiherbi appeared to favor remnant perennial grasses vores of infested ranges results in increased over medusaheadmedusahead incidence of injury from the seeds data on the low sagebrush communities because of level of injury are not available for domestic lack of herbaceous cover are relatively resistant livestock and certainly not available for wildlife to the spread of wildfireswildfires big sagebrush com- munitiesmunities especially those with cheatgrasscheatgrass CONTROL OF medusaheadMEDUS AHEAD underunderstoriesstories are very subject to the spread of wildfireswildfires invasion of medusaheadmedus ahead into low sagebrush communities introduces wildfireswildfires to kay developed highly technical and very suc- cessful control these communities perhaps for the first time and revegetation techniques for the annual dominated of since they were in pristine condition perennial rangelands eismoneiscisels mon- grass forb and shrub cover are all negatively tane california using the herbicide paraparaquatquat 111 I1 dimethyl bipyridinium ion and correlated with medusaheadmedus ahead cover in the west- llI spe- ern great basin young and evans 1970 cializedcialized seeding equipment kay 1963 1966 kay and mckell 1963 this technique was not successful in the GRAZING preference intermountain area because medusaheadmedus ahead plants were not susceptible to paraparaquatquat in the it is obvious from the above discussion that temperate desert environment and the annual preference by grazing animals plays an impor- legumes that proved so adapted to eiscisels montanecismontane tant part in the successional dynamics of california were not adapted to the sagebrush medusaheadmedus ahead communities one ofthe few stud- environment young et al 1971b herbicidal ies of medusaheadmedus ahead palatability was conducted fallow techniques using atrazineatrazine 6 chlorochloronN on the northern coast of california using sheep ethylnethyl N 135 triazine 24 di in small hurdle plots lusk et al 1961 under amine or dalabondalapon 22 dichloropropanoic the confined conditions of the study sheep uti- acid and mechanical fallow techniques were lized medusaheadmedusahead when it was green when developed for use in the great basin hilken faced with no choice they used some herbage and miller 1980 provide a summary ofherbiof cerbiherbi after the medusmedusaheadahead matured how much uti- eldaleidal control measures applied experimentally lizationlization of medusaheadmedusahead would occur in temper- for the control of medusaheadmedusahead A large part of ate desert situations is unknown the area infesteda4thinfested with medusaheadmedusahead in the west- cheatgrassCheat grass stands put a tremendous produc- ern great basin was never adapted to these tion of grass caryopses into a local ecosystem treatments because of surface rock cover that vertebrate granivoresgranivores have adapted to this food prohibited tillage or seed drilling techniques source savage et al 1969 showed in feeding the current mass cancellation of federal regis- trials that chukar partridges alectoris graeca tration for uses of herbicidesherbicides on rangelands and could not utilize the caryopses of medusaheadmedus ahead the failure of federal land management agencies as a food source these birds are dependent on to adopt the use herbicidalofofherbicidal revegetation tech- cheatcheatgrassgrass seeds in the fall and winter we do niques have made the use of these techniques 250 GREAT BASIN naturalist volume 52 impossible landformsLandforms and soils of the sites limits establishment of perennial grass seed- where medusaheadmedus ahead is spreading into temperate lings desert rangelands are critical factors in the eco- the area of medusaheadmedus ahead invasion in the logical suppression of this species western great basin is a microcosm where events in soil and plant ecology that influence millions of acres in the intermountain area are NATURE OF medusahead INFESTED landscapes brought by fortuitous combinations of physical and biological parameters into sharp focus in the medusmedusaheadahead invasion area lake deposited landscape of the western great basin the red clay is in obvious discontinuity with the thin has is of where medusaheadmedus ahead invaded composed grayish surface soil in undisturbed profiles of flows com- a series of fairly recent basalt that this situation the influence of alleviation of sub- plateau and the extreme south- prise the modoc aerial deposited material is apparent on the basaltsBabasalessalts ern extension of the columbia river structure of the clay subsoil indicating the the flows clays from a superimposed on are antiquity of this process personal communica- lake lake much older than tertiary age this was tion robert blank soil scientist ARS USDA which at the pluvial lake lahontan lapped accumulations of medusaheadmedusahead litter change lower of the flows old lake left margins the wildfire characteristics and the shrub compo- thick beds of clay textured sediments occasion- nent ofthe plant community is eliminated con- ally interbedded with diatomaceous earth the tinued grazing of inedusaheadmedusahead dominated clay minerals are predominantly double lattice grasslands is extremely deleterious on remnant forms expand and contract with moisture that perennial grasses because of differential grazing content and shrinkage has this expansion preference in contrast to medusaheadmedusabeadmedusmedusa beadheadahead sorted basalt rock from the buried flows into cheatgrasscheatgrass is seasonally preferred forage spe- polygons and pressure ridges until por- giant cies and even the dry herbage of cheatcheatgrassgrass is resemble packs tions of the landscape arctic ice utilized by livestock this dilutes the effect of that are black instead of white grazing as far as the native perennials are con- there are a host of topotopoedaphicedaphic situations cerned lack of preference for medusaheadmedus ahead within this wilderness that specific support concentrates the effects of herbivoreherbivoryherbivory subaeri- assemblages of plants however the landscape is ally deposited surface soil is extremely erodible characterized areas of residual soils by upland once protection of the shrub canopy and its surface soils big with loam textured that support dependent microphytic crust is lost loss of the surface soils sagebrush and clay textured that surface leads to exposure of the clay sediments low vast nearly level support sagebrush that then function as vertisolsVertisols shrinking crack- benches of lake sediments swirling support ing and swallowing the surface and reexpand of basin big sagebrush Arteartemisiamisla mosaics ing with moisture edusaheadmedusaheadMedusM edusahead is one ofthe few and a recently discov- tridentata sspasp dentatatritridentatatridentate plant species adapted to these vertisolsVerti sols perhaps low ered type of sagebrush a subspecies of some of the soils of these landscapes were known sagebrush as lahontan sagebrush the always Vertivertisolssols where in wet years annual where basin big sagebrush occurs in depressions sunsunflowersflowers helianthus annausannuus and turkey erosional accumulate on soils with products mullein Eremoeremocarpuscarpus setigerussetisetigerousgerus formed the unusual clay textured surface horizons a very only native vegetation perhaps excessive graz- for the great basin lahontan occurrence the ing converted some of these soils to vertisolsVertisols on the lake bed sagebrush communities occur before medusaheadmedus ahead arrived the important sediments that are veneveneeredveneererered with thin clay point is that medusaheadmedus ahead is actively attacking textured layers of subaerially deposited coarser assemblages of native vegetation and changing soil the physical and biological potential of the sites wind erosion products accumulate under the shrub canopiescanopies and coupled with organic matter from leafleaffallfall build mounds under the management OF medusaheadMEDUS AHEAD shrubs while miniplayas develop in the inter infestations spaces eckert et al 1989 have described and experimented with the seedbeds of these it is difficult to revegetate Vertivertisolssols in desert mound interspace situations particularly the environments with both seedlings ofwoody and vesicular crust that forms in the interspacesinterspaces and herbaceous species native and exotic not only 1992 ECOLOGY AND management OF medusaheadMEDUS AHEAD 251 establishment but also subsequent growth are and microphytic crust that covers the mounds problems on these soils despite both brementremen to extend down to minglea4thminglemingie with vesicular crust in dous cation exchange capacity and moisture the interspacesinterspaces the thallophytic crust of holding capacity the tremendous matric mosses lichens and liverwortsliverworts is obviously potential ofthese fine clay soils is always surpris- gone and we can only speculate on the fate of ing moisture is not available for normal plant the microscopic crust of algae fungi and bacte- growth when soils still stick to your boots ria prolonged medusaheadmedusahead dominance may decrease populations of mycorrhizalmycorrhizae spores in NATURAL succession the soil and thus influence growth of artificially established perennial seedlings personal com- municationmunication jim trent soil microbiologist ARS min hironaka suggests that over pro- dr USDA reno nevada longed periods perennial seedlings might estab- specific plant pathogens developed and lish in medusahead infested sites especially the marketed by biotechnological companies may short lived perennial grass squirreltail hiro- have a role in weed control a naka 1963 hironaka and his students fol- range perhaps dr fusarium species exists that would be highly lowed this aspect of medusaheadmedusahead succession in specific for medusaheadmedus ahead several studies he demonstrated that squirrel- personal communica- tion joe antognini national program scientist tail can establish in medusaheadmedusahead communities weed science ARS USDA but he found the perennial grass populations to and who be cyclic when the squirreltail plants die they taxonomists geneticists have worked with medusaheadmedus ahead are replaced by medusaheadmedus ahead not longer lived have commented on how variable individual collections perennial grasses personal communication may be common garden studies have shown this to be in the western great basin hironaka s dr true for collections from the american west work is borne out by gradual increases in mckell robinson and major 1962 young et squirreltail plant density as grazing manage- al 1971b we found in common garden stud- ment systems have been implemented this has ies a collection from northern california that been especially noticeable during the past four iesles matured 4 weeks earlier than the for years of extreme drought densities of one average other collections or on or before the for squirreltail plant per 10 square feet began to maturity cheatcheatgrassgrass As medusaheadmedusahead evolves we have change the aspect of medusahead dominated yet to see the limits of its potential on the western sites but the fragile nature of this improvement range recent discovery of medusaheadmedusahead in is apparent when bioassaybioassay of seed banks shows the utah illustrates that portions of the eastern 250 500 viable medusaheadmedusahead seeds per square great basin have the potential to be invaded foot down from 1000 per square foot before the by this weed horton 1991 drought and fails to detect any viable squirrel- tail seeds unpublished research ARS USDA reno nevada literature CITED As you look at medusahead infested areas on the vertisolsVertisols of the western great basin you AL DAKIIEELDAKFIEEL A J 1986 interference growth and physio- have a nagging thought that something is miss- logical response of downy brome and medusaheadmedusahead unpublished doctoral dissertation university of cali- ing the lahontan and big sagebrush commu- fornia davis and san diego state university san nities of the ancient lake sediments have as their diego california most frequent perennial grass sandberg blue- BOVEYBOVFY R WDW D letourneau and L C ERICKSONEHICKSON 1961 grass this species is completely absent from the the chemical composition of medusmedusaheadahead and downy brome weeds 93079 307 311 medusaheadmedusahead stands and is from missing the EERT R E jilJR F F petersonPFTFRSONPFTFIISON M K WOOD W H stands where squirreltail has begun to return BLACKBURN and J A SFFPIIENS 1989 the role of what factors of seedbed quality exclude the soil surface morphology in the function of semiarid native invader sandberg bluegrass and are the rangelands TB 890189 01 nevada agricultural experi- ment station university of nevada reno same factors related to the failure of level higher ENSEVANS R aandjaandyA and J A YOUYOUNGNC 1970 plant litter and estab- perennial grasses to become established in lishmentlish ment of alien annual species in rangeland commu- squirreltailmedusahead communities nities weed science 18 697 703 the striking difference between native and 1972 microlitemicrositeMicrosite requirements for establishment of annual rangeland weeds weed science 20 350 356 loss 350356 medusaheadmedusahead communities other than of 1975 enhancing germination of dormant seeds of shrub canopiescanopies is loss of subcanopysubcanopy mounds downy brome weed science 23 354357354 357 252 GREAT BASIN naturalist volume 52

EVANEVANSs R A B L KAY andandaandjJ A YOUNGYO uncUNG 1975 microenvi mckell 1961 medusaheadmedusabeadMedusMedusa beadheadahead palatability journal of ronironinentrosmentronmentnent of a dynamic annual community in relation to range management 14 248 251 range improvement hilgardiaHilgardia 43 79 102 MAJOR JCJ C M mckellMKFLL andland L J BFRRYBERRY 1960 improve- EVANS R AAHH R HOLBO R E eckertECKFRI JRJB and J A ment of medusaheadmedusahead infested rangeland california YOUNCYOUNG 1970 functional environment of downy agricultural experiment station extension service brome communities in relation to weed control and leaflet 123 vegetationlerevegetation weed science 18 154 162 MALLORY J 1960 soil relations with medusaheadmedusahead pages fosbergfostirfosiir R M A 1965 relationship of cheatgrasscheatgrass and 39 41 in proceedings of the california section of the medusmedusaheadrnedusaheadahead to soils in the columbia river basin society for range management fresno california proceedings of the cheatgrassCheat grass symposium vale mckellMKFLL C MMJJ P ROBISON andland J MAJOR 1962 eco- oregon USU S department of the interior bureau of typic variation in medusaheadmedusahead an introduced annual land management washington DCD C grass ecology 43 686 698 fluorrlksfnmieieriksen S 1986 revision of taeniatherumtaematherumTaeniaTaematherum mcMCKFLLKE LL C M A M w1WILSONwa L S 0 N andandaandbB L KAY 1962 effec- poaceae nordienordicnordle journal of botany 6 389 397 tive burning of rangeland infested with medusmedusaheadmedusabeadmedusabeadheadahead flufiupliE dl1 1 KIKSIRIKSEN N S and R VON BOIIINFRBOTIINFR 1989 inter weeds 10 125 131 genericgenerie hybridizationhybl idizationiodization between taeniatherumtaematherumTaeniaTaematherum and dif NELSON J R and A M WILSON 1969 influence of age ferent geneiaegenerahgenerae oftnticeanaof triticeana poaceae nordic journal and awn removal on dormancy of medusaheadmedusahead seeds of botany 15 229 240 journal of range management 22 289 290 GOEBELcoi601goi nelbel C J and G brrbyrryrbyRRYBBY 1976 selectivity of range NEVSKINFVSKI S A 1934 schedatschedae ad herbarium florae ariaeasiae glassgrass seeds by local budsbirds journal of range manage- mediaemediac acta umu ariaeasiae med viiibv111b botanica 17 ment 29 393 395 1 94 harkishankisHARHISHA 11 R IS G A 1965 medusaheadMedusahead competition pages 66 69 PIFMFISFLPIEMEISFL R L 1951 causes affecting change and role of in proceedings of the cheatgrassCheatgrass symposium vale change in a vegetation of annuals in idaho ecology 32 oregon buteaubureau of land management portlandPoitpoltland 53 72 olegonoregon piperPIPFR C Vvandrand R K beaolebeattleBEATTIE 1914 flora of southwest- narrisharbisHARRIS G A and A M WILSON 1970 competition for ern washington and adjacent idaho the new era moisture among seedlings of annual and perennial printing co lancaster pennsylvania S patterns gigrassesasses as influenced by i ootroot elongation at low temper- SAKAMOSAKAMOTOfO 1973 of phylogenetic differentia- atures ecology 51 530 534 tion in the tnbetnticeaetribetriticeae seik zihoziboaihobibo 24 113111 31 hilkenHIIKIN T 0 and R F MILLERMILLFR 1980 medusaheadMedus ahead SAVAGESAVAGF D E J A YOUNG and R A EVANS 1969 utili- taemathenim7aeniathennn aspeasperiniafpemmaspenininini nevskianevski a review and anno- zation of medusaheadmedusahead and downy brome caryopses by tated bibliography station bulletin 664 agricultural chukar partridges journal of wildlife management experiment station oregon state university 3397533 975 978 corvallis S IIOOLFR A B 1966 elymus caput medusaemedusan L crosses with aegilops cyhndncacylindfica host crop science 6 79 82 lilikoHIRONAKA111110 NA KA M 1961 the relative i aterate of root development of cheatgrasscheatgrass and medusaheadmedusahead journal of range man- SHARP L A and E W TISDALETISDALF 1952 medusaheadMedusahead a agementagernent 14 263 267 problem on some idaho ranges research note 3 HIRONAKAHIHONAKA M and B W SINDELARSINDFLAR 1973 reproduction forest wildlife and range experiment station uni- idaho moscow success of squirreltail in medusaheadmedusahead infested ranges versity of journal of range management 26 219 221 YOUNG J A and R A EVANS 1970 invasion of great 18 1975 growth characteristicschaichal actenstics of squirreltail seed- medusaheadmedusahead into the basin weed science 89 97 lings vs competition with medusaheadmedusahead journal of range management 28 283 285 1982 temperature profiles for germination of cool HIRONAKAHI RONAKA M and E W TISDALETISUALF 1958 relative role of season range grasses ARRWARR W 27 agricultural root devedevelopinedevelopingdevelopmentlopinent of medusaheadmedusahead and cheatgrasscbeatgrasscheatgrass page research service USDAUS DA oakland california 28 in progresspi ogress report western weed control confer- YOUNG J ARA R A EVANS and R E ECKFRIECKERT JR 1968 ence germination of medusaheadmedusahead in response to tempera- ture and afterripening weed science 16 92 95 1963 secondary succession in annual vegetation in southern idaho ecology 44 810 812 1969 population dynamics of downy brome weed science 17 26 horlonHORIONHORTON W H 1991 medusaheadMedusahead importance distribu- 20 YOUNG J A R A EVANS and B L KAY 1971a germina- tion and control pages 387394387 394 in L F james J 0 of of annual litter evans M H ralphs and R D childs eds noxious tion caryopses grasses in simulated milmii agronomy journal 63 551 555 i angerange weeds westviewWestview press boulder colorado 1971b response of medusaheadmedusahead to paraquatpara quat jour- HOWELLhowl i L T 1903 A flora of northwest america vol 1 nal of range phanorogamaephanorogarnae binford and mort portland oregon management 24 41 43 YOUNG J A R A EVANS and J MAJOR 1977 sagebrush jebsonJEPSONjr PSON WLW L 1923 annals of flowering plants ofcalifornia andl pages 763 797 in M G barbour and M sather gate bookshop berkeley california steppe in barbourandjJ majorelorajor eds terrestrial of california john wiley & KAY 13 L 1963 effects of ofdalapon medusaheadmedusahead vegetation dalabondalapon on a com- sons new york munity weeds 3 207 209 YOUNG A R A EVANS ROBISON 1972 influence 1966 paraquatParaquat for seeding without cultivation J AR andaandj of repeated annual burning on a medusaheadmedusabeadmedusmedusa ahead commu- california agriculture 20 2 4 beadhead nity journal of range management 24 451 454 KAY B L and C M MCKELLme kellKFLL 1963 piepre emergence her bicides as lintinan aid on seedling annual langelandrangeland weeds 1126011 260 264 received 23 may 1991 LUSK W C M B JONES D T torelltobellTORFLL and C M acceacceptedptedapted 22 june 1992 great basin naturalist 523 appp 253 261

ROOST SITES USED BY SANDHILL CRANE STAGING ALONG THE PLATTE RIVER NEBRASKA

1 1 1 bradley S norling stanley H anderson and wayne A hubert

ABSIRACIABSTRACT we assessed the influence of water depth extent of unobstructed view and human disturbance features on use of roost sites by sandhillSandbill cranes along the platte river nebraska during spring migratory stopover aerial photos taken near dawn were used to determine areas of flock use and habitat availability in four sample reaches and measurements were made on the ground at flock roost areas in general depths of 1 13 cm were used by sandhill cranes in greater proportion than those available exposed sandbars and depths 20 cm were avoided while depths of 14 19 cm were used in proportion to their availability sites 11 50 in from the nearest visual obstruction were used significantly greater than their availability while sites 0 4 and 50 in from visual obstructions wereavoidedwere avoided sandhill cranes avoided sites near paved roads gravel roads single dwellings and bridges when selecting roost sites however they did not appear to be disturbed by private roads groups of residential buildings gravel pits railroads or electrical transmission lines

key words sandhill crane grus canadensis river boostsroosts habitat selection water depth disturbance sandbars platte river

the impact of water resource development americana also uses the area during migration on the platte river is well described and the threatened bald eagle haliaeetus kroonemeyer 1978 williams 1978 eschnereschneretEschneretet leucocephalus is a common winter resident al 1981 kircher and karlinger 1981 US fish US fish and wildlife service 19819811 the area and wildlife service 1981 krapu 1987 sidle et is also important habitat for the endangered al 1989 the major impact has come from interior population of least tern sterna antil irrigation projects along the north platte river larum and the threatened piping plover krapu et al 1982 which remove approxi- charadrius memelonusmeloduslodus both of which nest along mately 70 of the annual flow of the platte the platte river US fish and wildlife service river before reaching south central nebraska 1981 sidle et al 1989 kroonemeyer 1978 concomitant with chan considerable attention has been given to the nel shrinkage woody vegetation has encroachencroachedea impact of changing channel conditions on the on thousands of hectares of former channel midcontinentmidcontinent population of sandhill cranes area contributing to further changes in channel crusgrus canadensis that congregate along the features and altering habitat for numerous spe- river from early march to mid april during their cies of migratory birds in the big bend reach of annual spring migration lewis 1977 krapu the platte river in nebraska US fish and 1978 US fish and wildlife service 1981 wildlife service 1981 the big bend reach of during this time approximately 400000 sand- the platte river in nebraska is an area of hill cranes use this area while enroute to their importance to numerous species of migratory breeding grounds in canada alaska and eastern birds ofthe central flyway US fish and wild- siberia US fish and wildlife service 1981 life service 1981 in nebraska various facets of sandhill crane this area is an important stopover area for roosting habitat requirements have been stud- most of the midcontinentmidcontinent population of sandhill ied frith 1974 lewis 1974 US fish and cranes crusgrus canadensis 400000 600000 wildlife service 1981 krapu et al 1982198219841984 birds which roost in the river and feed in however these studies have not considered the nearby corn fields krapu et al 1981 krapu influence of habitat availability in relation to 1987 the endangered whooping crane CG habitat use the purpose of this study was to

wyoming cooperative fish and wildlife research unit box 3166 university station laiamielaramieLaiamielamielamle wyoming 82071

253 254 GREAT BASIN naturalist volume 52

us Z 0 281

rneqneanevenver study kearea

reach 1 reach 2

reachbeach 3

N reachbeach 4 intensive study area AMON

0 1 2 3 4 5 krnarn

fig 1 study sites in the plattepiatte river nebrasnebraskaa determine the influence of habitat availability great plains to their confluence near north as well as habitat use on the selection of roost platte nebraska sites by sandhill cranes the study area is characterized by numerous this study was designed to assess the influ- braided channels interspersed with unvege ence of three types of habitat features on roost tatedbated sandbars that frequently shift most of the sites used by sandhill cranes 1 water depth land within and adjacent to the study area is in 2 magnitude of unobstructed view and 3 private ownership land use in the area is pre- disturbance features dominantly agriculture and includes approxi- mately 60 cropland mostly corn 5 tame grassland and 15 STUDY AREA pasture 20 native riparian woodland reinecke and krapu 1979 riparian woodland comprises eastern study is located in south central the the area is in cottonwood populus deltoidesdeltoides forests with nebraska in hall and buffalo counties in the in dominant understory species of red cedar eastern half ofthe big bend reach of the platte virginianavirginiana and rough leaf dogwood encompasses a 36 km stretch of the juniperus virginiaja rivelriverrivet it comuscornus drumnwndii on low islands and veg- platte beginning 4 km west of shelton to river etated sandbars peach leaf willow salix grand island 1 all field measurements fig amygdalamygdaloidesoides coyote willow S exigualex and ffour 1 6 km reaches along the igual were in our 16 main bush amorphafiruticosaamorpha are the domi- channel of the platte river indigo fruticosafruticose nant species US fish and wildlife service min nebraska contributes spring precipitation 1981 currier 1982 to the platte river basin flow but most of the flow is derived from spring runoffthatrunoff that originates as snowsnowbeltsnowmeltmelt in the rocky mountains eschner METHODS et al 1981 spring runoff flows into both the north and south platte risersrivers which flow north- aerial photography was used to determine east and southeast respectively across the flock locations and delineate flock boundaries of 199219921 CRANE ROOST SITES 255

ostingroostingboostingro sandhill cranes along a 36 km stretch into general areas of equal size with two to five of the platte river photography was restricted transects depending upon flock size A flock was to mornings with less than 10 cloud cover and defined as a continuous distribution of birds or ceilings above 975 m flights were begun 30 an aggregation of birds spatially independent of minutes before sunrise because of the need to other birds separated by a distance 20 m photograph sandhill cranes before they leave flocks usually occurred in configurations that the roost in early morning light was adequate appeared distinct from other flocks in the vicinity to permit photography 10 15 minutes before after transects were located on photographs sunrisesunnsesunnee they were measured and laid out on the ground A hasselblad 500 EL 70 mm camera was in relation to marker locations using vinyl flag- used to photograph the study area the camera ging placed on each side of the channel water was mounted in a standard camera hatch in a depths were measured to the nearest 3 cm at cessna 172 fixed wing aircraft and was 3 m intervals and plotted on acetate overlaid on equippeda4thequipped with an 80sommmm focal length zeiss lens aerial photographs with delineated flock bound- exposures were made at 160 and 1125 second aries width and depth data were combined to at f28 using kodak trixtri X 640 AFS aerographic livetivecivegive mean estimates for each of the four reaches film the camera was equipped with a 70 expo- each 16 km reach was sampled as soon as sure back loaded with 555.5 m of film allowing 80 possible after each flight always within three exposures laysdays staff gauges were placed in each area to the aircraft was flown at approximately 140 measure any changes in water level between the kahrkmhr at an initial altitude of 790 m above time each reach was photographed and the time ground level for the first two flights during the it was sampled detectable changes in water last two flights the altitude was increased to 910 level were recorded and used to correct depth m above ground level these altitudes provided distributions a 048 km and 0640.64 linkinkm2 coverage on each discharge was measured on each flight day frame respectively frame rate was controlled in close proximity to the study areas followingfolloia4ng by an intervalometerinterval ometer calibrated for 30 over- the technique of buchanan and somers 1969 lap to provide continuous photographic cover- contact prints were made from each roll of age of the study area film individual frames were cut out and glued shortly after each flight the film was custom onto posterposterboardboard to form a mosaic providing a processed by hand agitation in a single solution continuous coverage of the river channel scale tank varying time and developer temperature was determined by comparing bridge segments to obtain optimum development approxi- and transect locations on the contact prints with mately 150 frames were exposed from each measurements of these locations made on the flight frames were examined under 8xax magni- ground scale estimates were made along 2 to ficationfication to identify crane flocks and were 3 km segments of river photograph scales enlarged to 41 X 51 cm 16 X 20 in and printed ranged from 18681 to 110334 for the first two on kodak poly contract RC paper processed flightsflights and 105951105951 to 111857111857 for the last two photographs were stored for later analysis of lightsflightsf visual obstructions and disturbance features A binocular zoom microscope 1 4xax was each of the four 16 km reaches was marked used to identify flocks and delineate flock on both sides of the river bank with 16 im2ima1 m2ma boundaries on the contact prints covered with markers made of white cloth the markers acetate flocks were delineated and subse- placed 100 m apart at the edge of the river bank quently numbered on the acetate overlays on were positioned in such a way that markers on contact photos the distance from the edge of the opposite sides of the channel were parallel each flock to the nearest visual obstruction was to the channel the markers enabled accurate measured to the nearest 050.5 mm on the photos scale measurements to be taken from photos ground distance 4 6 m using a drafting cal- and provided position reference for transects iper visual obstructions included vegetation a across the channel when sampling water depths river bank or any other visually solid object aerial photographs covering each reach were 1 I1 m in height used to determine the position of transects random points were plotted on contact through flocks transects were positioned so photos to estimate the features of available hab- that each flock studied on a photo was divided itat random points were determined by a series 256 GREAT BASIN naturalist volume 52

of random numbers identifying point coordi- being selected habitat availability use and nates on gridded overlay covering contact selection were summarized within reaches prints points outside the river channel were across flight dates and from data pooled across discarded only random points located in water reaches and flight dates data were pooled to were used because points on sandbars islands generalize the selection of depths over the or the river bank were not considered poten- course of the sampling period tially usable roosting habitat A total of 339 the chi square of homogeneity marcum random points within the river channel were and loftsgaarden 1980 was used to test identified on the contact prints grid squares whether differences existed between the distri- were 1251.25 mmmm2mma to ensure a representative bution of random points and those locations sample of locations on the river As with flock used by sandhill cranes relative to visual locations the distance from each random point obstructions and disturbance features it was to the nearest visual obstruction was measured also used to determine if there were differences on the photos to the nearest 050.5os mm using a between the proportion of used and available drafting caliper water depths among and within reaches confi- for analysis of human disturbance features dence intervals were calculated using the flock locations and random points along the Bonbonferroniferroni Z statistic to test which intervals entire 36 km study area were transferred from within the distributions were used more or less 70 mm contact prints to acetate overlays of color than expected byers et al 1984 differences infrared aerial photographs scale 125595 between selection functions were tested with a using a zoom transfer scope the photographs Z test analysis of variance ANOVA was used taken in april 1989 were obtained from the to determine ifvisual obstructions had an effect bureau of reclamation in grand island on the disturbance potential created by various nebraska distances were measured from the types of disturbance features significance for edge of each flock and individual random points all statistical inferences was P 0505.05 selected by placing a card over the photograph to the nearest human disturbance features these features included paved roads gravel RESULTS roads private roads urban dwellings single dwellings railroads commercial development A total of four sampling flights were made highways and bridges distances were mea- one each on 21 and 31 march and 4 and 10 april os sured to the nearest 050.5 mm on photos ground 1989 A total of 285 flocks were identified distance 13 m with a drafting caliper during the four flights following the flights 20 flock sites were selected and sampled and a total data analysis of 5109 depth measurements were recorded in frequency histograms were plotted for mea- the field sured distances from the edge of a flock and for SAMPLING AREAS reaches I1 and II11 were random distances to the nearest visual obstruc- the narrowest with mean channel widths of 254 tion and disturbance features frequency distri- m range 225 319 m and 249 m range butions were plotted for available and used 241 263 m respectively while reaches iliIII111 and selected water depths frequency distributions IV located upstream werewereadderwiderAdder reach iliIII111 had of available and used selected water depths for a mean channel width of 413 m range 387 each 161 6 kmkinkiuklu reach were determined by combin- 440 m while reach IV had a mean channel ing flock data for each reach for a given flight width of 357 m range 296 445 m available depths were defined as all depth mea- reaches I1 and II11 had similar discharge 17 surementssure ments taken along a transect and used nsms while reaches liiIII111 and IV had greater depths were those depths where birds were values 27 and 44 ms on 21 march table 1 present along a transect habitat selection was discharge in reach illlilIII111 was typically twice as computed by dividing the proportion of habitat high as reaches I1 and II11 reach IV had the used within a depth interval by the proportion highest discharge of the four reaches often of depths available in that same interval bovee three times greater than in reaches I1 and II11 1986 depths used less than their availability table 1 reaches I1 II11 and iliIII111 were located in were defined as being avoided while those used a braided portion of the surface along the south more than their availability were defined as channel and contained only partial river flow 1992 CRANE ROOST SITES 257

3 TABLETABLF 1 discharge in cubic meters per second m for sample reaches on different flight dates along the platte river nebraska during spring 1989

flight date reach I1 reach II11 reach iliIII111 reach IV

a 21 marchmarch11 17417.4 174 275 446 31 march 111 186 321 4 april 106 106 288 10 april 79 79 137 217

discharges for all reachesleareaellesclieseiles on 21 marchmaichmatch were medineasmeameasuiedmeasuredsuieded on 24 march thusrhus a thieethree dayclay lag period existed between the iuneinnetimetune each reachleach was lownhownflown andwidmid thetiietile timetune each reach was ineameasuredsured fortoroor dischargediscdisehaige

reach IV was located along the mainmaln channel study season progressed depths 20 cm were and contained total rivernver flow used significantly less than expected during the HABITAT availability the distribution first flight but by the last survey only depths of available water depths differed among 29 cm were used less than expected P 05os05.05 reaches on 21 march 1989 82 of the avail depths of 0 cm were generally avoided by able habitat in reaches I1 and II11 consisted of sandhill cranes during the last two surveys and depths 0 25 cm in contrast 53 and 66 of were used less than would be expected by the available habitat in reaches iliIII111 and IV chance P 05os05.05 respectively consisted of depths 0 25 cm habitat selection was assessed using both an increased frequency of shallow depths habitat use and availability data for specific 0 19 cm and a decreased frequency of deeper water depths the most frequently occurring depths 20 cm occurred over the study depth intervals for which selection occurred period this division is made because cranes were 5 7 cm followed by 1 4 8 10loilloll101111 13 and seldom used depths greater than 20 cm the 14 16 cm in decreasing order of preference increase in exposed sandbars depth 0 cm was VISUAL obstructions there was a sig- most pronounced in reaches I1 and II11 which nificantnificant difference between the distribution of showed increases of 13 and I111 I1 respectively flock locations and random points relative to the reaches II11 and liililIII111 showed increases of 12 and distance from the nearest visual obstruction 19 respectively in available depths of 1 4 cm P ooiool001.001 proportional use of sites 0 50 m between the first and last flight reaches iliIII111 and from the nearest visual obstruction was signifi- IV showed decreases of 10 and 7 respec- cantly greater than availability P 05os05.05 while tively in depths 38 cm for the same period sites 50 m from a visual obstruction were during the study period a progressive decrease avoided P 05os05.05 in discharge occurred table I1 causing more the 0 25 m interval was divided into six shallow areas 0 19 cm increments 0 1 4 5 10 11 15 16 20 and HABITAT USE frequency distributions of 21 25 m there was a significant difference roosting habitat use by cranes indicated the between the distribution of flocks and random highest proportions of used water depths were point distances P ooi001001.001 sites as close as 5 10 from the 1 4 and 5 7 cm increments this range m from the nearest visual obstruction were used of water depth accounted for 65 of the mea by sandhill cranes only sites 0 4 m from a sured depths there was no discernible varia- visual obstruction were avoided P 05os05.05 while tion in the frequency of water depths used sites 11 25 m from a visual obstruction were among the four reaches used more than expected P 05os05.05 there was a small but significant difference visual obstructions were divided into three in the distribution of depths used between the categories 1 unvegetated bank 2 vegetated beginning and end of the study period P 05 bank and 3 vegetated island there were no depths ofoof 0 cm showed a significant decrease in significant differences in the distribution of dis- use while depths 20 22 cm showed a significant tances between an vegetatedunvecretatedunvegetatedun and vegetated increase in use P 05 the data showed a bank but there were significant differences for significant difference between the distribution the distribution of distances between vegetated of used and available water depths for all four banks and vegetated islands and between sampling periods FP 001 sandhill cranes unvegetated banks and vegetated islands P used progressively deeper water depths as the 005oos005.005 sandhill cranes roosted a mean distance 258 GREAT BASIN naturalist volume 52 of 45 inm from vegetatedunvegetatedun banks 50 in from presence of visual obstruction between a roost vegetated banks and 27 in from vegetated ing flock and the nearest gravel road did not islands appear to reduce the disturbance potential cre- CHANNEL WIDTH there was a relation- ated by gravel roads ship between the minimum unobstructed chan- SINGLE DWELLINGS there was a signifi- nel width and distance to the nearest visual cant difference between the distribution ofused obstruction the distance to the nearest visual and random locations relative to the distance to obstructions was a function of less than one half the nearest single dwelling P oi0101.01 in general the minimum unobstructed channel width sandhill cranes showed an avoidance for sites there was a significant difference between closer than 400 in from a single dwelling P the distribution of flock locations and random 05os05.05 sites 501 600 in from the nearest single points relative to minimum unobstructed chan- dwelling were used more than expected P nel width P 005 sandhill cranes used chan- 05os05.05 the presence of a visual obstruction nels 100 200 in wide inm greater proportion than between a flock and the nearest single dwelling those generally available channels narrower did not affect the disturbance potential created than 100 in were avoided while those 200 in by single dwellings wide were used in proportion to their availabil- BRIDGES sandhill crane flocks were not ity the mean minimum unobstructed channel distributed randomly with respect to distance width used by roosting flocks was 196 in range from bridges P ooi001001.001 they showed avoid- 34 445 in nearly 100 of the flocks were in ance of sites closer than 400 in from the nearest channels with a minimum unobstructed chan- bridge P 0505.05 similarly they used sites 400 nel width of 50 in and over 97 and 80 of in from the nearest bridge the flocks were in channels with a minimum OTHER disturbances no significant unobstructed width of 100 and sisiso150150 in differences were found between urban dwell- respectively the mean relative flock size sur- ings gravel pits commercial development face area was 3883 inm2ma range 19 55354 m2ma transmission lines and the distribution of there was no relationship between flock size sandhill crane flocks and minimum unobstructed channel width both large and small flocks were located in wide discussion as well as narrow channels human disturbance features DEPTH distribution this study indicated that sandhill cranes prefer water depths of PAVED ROADS sandhill crane flocks were 1 13 cm for roosting but roost in greater depths not distributed randomly with respect to dis- lataka and yahnke 1986 developed a predic- tance from paved roads P 001 sandhill tive model for sandhill crane roosting habitat cranes showed avoidance of sites closer than and stated that the majority roosted in water 500 in from the nearest paved road P 05 depths between 0 and 12 cm which is presum- but used sites as close as 301 400 in sites ably the optimal depth for roosting similarly located 701 900 in from the nearest paved road frith 1986 suggested a water depth of 2 15 cm were used more than expected P 05 as optimum for roosting sites currier 1982 sandhill cranes roosted a mean distance of reported a slightly deeper range of depths from 1260 in from the nearest paved road when a 10 15 cineincm as optimum for roosting lewis 1974 visual obstruction was present but a mean dis suggested that roost sites be characterized by tance of 1575 in from the nearest paved road in depths 10 20 cm and folk 1989 reported an the absence of visual obstructions even greater range of depths used for roosting GRAVEL ROADS there was a significant 010.1oloi 21021.0 cm for sandhill cranes along the difference between the distnbutiondistribution ofused sites north platte river in nebraska and random locations relative to distance from despite a change in the availability of water gravel roads P 01 sandhill cranes showed depths with over a 50 reduction in discharge avoidance of sites that were closer than 400 in over the period of study table I1 only slight fromfroin the nearest gravel road P 05 but flocks differences were detected in the overall use of were located as close as 301 400 inm sites that specific water depths the fact that habitat use were 601 800 in from the nearest gravel road remained the same despite a change in habitat were used more than expected P 05 the selection suggests that selection indices more 1992 CRANE ROOST SITES 259

strongly refreselectrefelectelect changes in habitat availability flows through these channels latka and yahnke than habitat preference if habitat selection had 1986 reflected habitat preference then habitat selec- our findings corroborate the results of tion indices would have been more similar krapu et al 1984 who reported that over 99 between the beginning and end of the study of all roosting sandhill cranes were in unob- period strucstructedted channels over 50 in wide and almost VISUAL obstructions this study indi- 70 were in channels 150 in wide in contrast cated that sandhill cranes will not roost closer data from nighttime aerial thermography by than 5 in from a visual obstruction and that pucherelli 1988 suggested that almost half of distances from I111I1 to 25 in are the most fre- all boostsroosts were in channels 150 in wide and quently used latka and yahnke 1986 reported that the greatest proportion of boostsroosts were in that sandhill cranes did not roost 15 in from channels 51 150 maedema4dem wide the bank folk 1989 suggested that sandhill folk and tacha 1990 studied roosting cranes preferred to roost 25 in from a visual along the north platte river in nebraska and obstruction but he observed roosting as close as reported a channel width criterion that was dif- 4 in from a visual obstruction our results indi- ferent from this study they reported that 82 cate that various forms of visual obstructions of the boostsroosts were in channels 48 in wide and have different impacts on roost site selection 18 were in channels from 16 47 in wide overall vegetated islands have little influence HUMAN disturbance our study demon- on the selection of roost sites whereas vege- strated that human disturbance features influ- tated banks have greater influence ence selection of roost sites by sandhill cranes it is generally believed that sandhill cranes in general the greatest disturbance potentials maintain an optimum distance from a visual were attributed to roads paved and gravel single obstruction to increase their security from ter- bridges and dwellingsb where irregular rerestrialstrial predators primarily candcandiascandidsids this is evi- but considerable human activity might occur gravel roads denced by the fact that the majority of flocks are pits private railroads and power lines had infrequent disturbances and did located in closer proximity to vegetated islands not affect than to vegetatedunvegetatedun or vegetated banks seem to roost site selection in all likeli- hood some form of acclimation occurs between channel morphology may also be a factor the constant disturbance on commercial and influencing the distribution of ro areas osting urban development relative to banks or islands this assertion is is little supported by observations from depth measure- there literature that objectively describes the zones of influence exerted ments which suggest that water depths and by var- ious human disturbance features selec- velocities near banks are deeper and faster than on the tion of roost sites by sandhill cranes along the depths near islands due to bank undercutting platte river folk 1989 suggested that riparian thus sites near islands may contain a greater forest along the river provides a visual barrier of suitable ro than proportion osting depths sites against most types ofpotential disturbances and adjacent to banks that sandhill cranes roost in sections ofthe river CHANNEL cranes WIDTH sandhill selec- as close as 80 in from a bridge in contrast our tively used channels 100 200 in wide while study indicates that sandhill cranes roost in channels narrower than 100 in were avoided sections of the river that are 400 in from the nearly 100 of the roosting sandhill crane nearest bridge we feel that our results provide flocks were located in channels with an unob- an objective description of potential zones of strucstructedted channel width 50 in and over 80 influence exerted by various disturbance fea- were located in channels 150 in wide wide tures and the effect these features have on roost channels potentially provide more space for site selection by sandhill cranes along the roosting sandhill cranes more security from platte river predators and more available water depths to in summary our study shows the importance choose from however since channel width was of sandbars with water less than 20 cm in depth evaluated independently of channel depth it is surrounded by deeper water these sandbars possible that use of narrow channels 100 in must be at least 5 in from some form of visual wide is limited not so much by a requirement obstruction such as dense vegetation this for wider channels but by deeper water that apparently allows the sandhill cranes to see 260 GREAT BASIN naturalist volume 52 approaching predators As a result sandhill eschnerESCIINER TRT R HADLEYHADLFY and K CROWLEYCROWLFY 1981 hydrol- platte cranes normally roost in channels 100 200 m ogic and morphologic changes in the river basin a historical perspective USU S geological survey wide these sites are generally away from report 8112581 125 denver colorado 57 appp human disturbances such as roads bridges and FOLK M J 1989 roost site characteristics of sandhill private dwellings sandhill cranes could toler- cranes in the north platte river of nebraska unpub- s illinois car- ate irregular disturbances such as private roads lished master thesis southern university bondale 58 appp and railroads FOLK M J and I1 C TACIIATACHA 1990 sandhill crane roost the fact that 80 of the midcontinentmidcontinent pop- site characteristics in the north platte river valley ulation of sandhill cranes uses this area for journal of wildlife management 54 48086480 486 platte staging in the spring indicates its importance it iliillFRIIIIit11 C R 1974 the ecology ofthe river as related to sandhill cranes and other waterfowl in south central is during this period that the birds apparently nebraska unpublished master s thesis kearney state build up energy reserves allowing them to con college kearney nebraska ilg116116ppappp tinuedinue their northward migration if the area 1986 crane habitat of the platte river pages were to become unfit for sandhill cranes the 151 156 in J C lewis ed proceedings of the 1981 crane workshop national audubon society tavenier population would likely suffer decline florida kircherkirtKIRC llerllenIIFR J E and M R KARLINGERKARLINGFR 1981 changes in surface water hydrology for the south platte river in acknowledgments colorado and nebraska the north platte river and platte river in nebraska USU S geological survey we appreciate the help of delmar holz report 8181881 818 77 appp laura smith and craig schwieger of the grand KRAPU G L 1978 sandhill crane use of staging areas in nebraska pages 1 6 in J C lewis ed proceedings island nebraska office of the US bureau of of the 1978 crane workshop colorado state univer- reclamation who supplied needed equipment sity fort collins and assisted in the field the US fish and 1987 use of staging areas by sandhisandhill cranes in the wildlife service personnel in grand island midcontinentmidcontinent region of north america pages 451 462 archibald and R eds proceed- brabander and john sidle who in G W F pasquier especially jerry ings of the 1983 crane workshop international crane piloted the plane on most flights were of great foundation baraboo wisconsin assistance tierny parrish gene maddox and KRAPU G L D E FACFYFACEY E K fiiitzellFRIIVFLL and D H kevin Clauglaugiauglaubiusbiusblus helped gather field data we are jonnsonJOIINSONJOHNSON 1984 habitat use by migrant sandhill cranes in nebraska of wildlife management grateful to the landowners in the platte river journal 4840748 407117407 tit417117 valley who allowed us access to their properties KRAPU G L K J reineckeRFINFCKFRFINECKE and C R FRIIII 1982 and we appreciate the help of kenneth and sandhill cranes and the platte river transactions of marie strom we also wish to thank ronald the north america wildlife natural resource confer- ence 47 542 552 marrs for his help in of aerial interpretation kroonemeyer K E 1978 the USU S fish and wildlife S photographs funding was provided by the UUS serviceservicess platte river national wildlife study pages bureau of reclamation in grand island 29 32 in J C lewis ed proceedings of the 1978 nebraska and the US fish and wildlife ser- crane workshop fort collins colorado LAIKA YAIINKFYAIINKE 1986 the vice national ecology research center in fort lalkalarka D C and J W simulating ostingroostingboostingro0sting habitat of sandhill cranes and validation of collins colorado suitabilitysu of use indices pages 19 22 in J veinervemerverner M L morrisonmornson and C J ralph eds wildlife 2000 modeling habitat relationships of terrestrial verte- literature CITED brates university of wisconsin press madison LFWIS J C 1974 ecology of the Sandsandhillbill crane inm south- bovboarbovrE 1 K D 1986 development and evaluation of habitat eastern central flyway unpublished doctoral disser- suitability criteriaentena for use in the instream howflow incre- tation oklahoma state university stillwater 213 appp mental methodology USU S fish and wildlife service 1977 sandhill cranes grus canadensis pages biological report 86 7 14 appp 5 43 in G C andersonandersonedAnder soneded management of migratory buciiananBU ilananIIANAN T J andwandanawW PR SOMERSSOMI KS 1969 discharge mea- shore and upland game birds in north america inter- surementssuiesure ments of grazinggracing stations book 3 chapter 8aaa in national association of fish and wildlife agencies techniques of water resource investigations US geol- washington DCD C ogical survey washington DCD C 65 appp MARCUM C L and D 0 loffscaardenloftsgaarden 1980 Aanonnon BYERSBYF KS C RRRR K siriniiorsisteiniiwisr and P R KRAUSMAN mapping technique for studying habitat preferences 1984 classification of a technique for analysis of utili- journal of wildlife management 44 963 968 zation availability data journal of wildlife manage- PUCHERFLU M J 1988 measuring channel width variables inmentent 48 1050 1053 of sandhill crane roosting habitat sites along the platte cuCUKKIIii ii i i liii PPJJ 1982 the floodplainfloodplamfloodplainplampiam vegetation of the platte river using nighttime aerial thermography applied river phytosociology forest development and seed- science reference memo no AD 884588 4 5 USU S ling establishment unpublished doctoral dissertation department of interior bureau of reclamation iowa state university amesarnes 332 appp denver colorado 1992112 CRANE ROOST SITES 261

REreineckeRFINECKFINECKE K and G L KRAPU 1979 KJjandajandg spring food habits US FISH AND WILDLIFE SFRVICES ERVICE 1981 the platte river of sandhill cranes pages in nebraska 13 19 in J C ecology study special research report jamestown lewis ed proceedings of the 1978 crane workshop north dakota 187 appp collins fort colorado WILLIAMS G P 1978 the case oftheodtheofthe shrinking channels SIDLE G E D MILLER and P J J CURRIERCURRIFR 1989 the north platte and platte rivers in nebraska circu- changing habitats platte chmgsng in the river valley of lar 781 US geological survey reston virginia 48pp48 appp nenebraskara ka frameprairieprame naturalist 21 91 104 received 6 december 1991 accepted 30 julyluigjudgjudy 1992 great basin naturalist 523 appp 262 268

POST pleistocene DISPERSAL IN THE MEXICAN VOLE MICROTUS MEXICANUS AN EXAMPLE OF AN APPARENT TREND IN THE distribution OF southwestern MAMMALS

1 2 russell davis and J R callahan

ABSIRA 1 the present distribution of the mexican vole microtus meicmeicanusmexicanusmexic anus is not entirely the product of post pleistocene forest fi agmentationaugmentationagfragmentationmentation and extinctextinctionionlonioD recent dispersal also is indicated literature records further suggest that this phenomenon may reflect a general pattern of northward range expansion in many southwestern mammal species

keykegkelkei words microtus vole dispersal geographybiogeographybiohio vicanancevicariancevicavicariancenanchnance pleistocene

traditional biogeographic theory attributes DISPERSAL A BRIEF REVIEW the modern distribution of small nonflying montane mammals in the southwest to post popostst pleistocene dispersal has been verified pleistocene climatic change brown 197119781971 1978 primarily in 1 conspicuous diurnal mammals patterson 1984 patterson and atmar 1986 such as sciurids and 2 mammals colonizing restriction of woodland and forest habitat to regions that were previously well sampled by higher elevations is assumed to have stranded collectors for species and groups that do not such species on isolated patches of montane fall into either category the geographerbiogeographerbiogeograpberbioblo is habitat although it is recognized that local left to interpret broader distribution patterns extinction has caused further range reductions andor small bits of indirect evidence post pleistocene range expansion generally has As an example of the first situation davis been discounted brown 197119781971 1978 this relict and brown 1989 and davis and bissell 1989 model satisfactorily explains the distribution of showed that recent dispersal has significantly great many basin species but evidence from altered the distribution of abertabertss squirrel elsewhere in the southwest strongly supports sciurus abertiabenti another example involves the recent dispersal davis and dunford 1987 dusky chipmunk tamias obscuresobscurus which was davis and ward 1988 davis et al 1988 davis absent from thomas mountain in southern cal- and bissell 1989 davis and brown 1989 ifornia at least between 1974 and 1976 calla- lomolino et al 1989 han 1977 by 1979 the species had recolonized in this paper we will review evidence indi- this peak which is isolated from the san jacinto cating that many southwestern mammals range by a lomile10 mile stretch of semiarid grass including the mexican vole and other montane landsagebrushland sagebrush habitat callahan in prepara- mammals as well as nonmontanenonmontane species tion the second scenario is illustrated by davis have shown a striking northward range shift and dunford 1987 and davis and ward during the past several decades for some spe- 1988 who found evidence of recent montane cies this pattern appears to reflect milder win- colonization by Sigsigmodonsigmodontmodon ochrognathus in a ters or humanbuman influences for others the trend is well studied area of southeast arizona harder to explain ifverified however this trend since many small mammals are not readily presupposes among other things a greater dis- trapped and many localities have not been sam persal capability than is typically attributed to pled extensively it is easy for critics to shoot small mammals down new distribution records on the grounds of inadequate prior sampling in such cases it is

deortdejrtocpti hnentanenttofeologyof ecology wd evolutionary bialogbiologbiology universityuniveisityotofaiizonaanona tucson anoniarizona 85721 milsolunofsolithwesternmuseum of southwestersouthwestemSouthw estem biology univesityuniversityuniversityoUnive sity of new mexico albuquerque new mexico 8718713131 mailing addre box 314013140 1 iletimetlemet calificiliforniiia 92546

262 199211992 VOLE DISPERSAL 263

prospect navaho mt 0 valley 36 36 0 0 Mmusicmusieusic black mtsatsomtso0 S kaibab B mesa 0 C

eluhualapaikualapaipiualapaiHuPIu alapai mts bradshaw 4 mts P X A sierra C ancane a 00 M s n 7 Y CD nantanesmantanesNan tanes A plateau 20 20

I1 J 0 250 km papago spsaps 0 75km

fig 1 distribution of Microtus mexicanusrrwxtcanusmexicanus at this scale only the two most isolated populations in the united states fig 2 details oftheodtheof the distribution of Microtus mexlcanusmexicanusmexicanus are distinguished modified from findley et al 1975 hall in arizona showing isolated populations and three subspe- 1981 finleyfinleyetet al 1986 hoffmeister 1986 cies A B and C modified from hoffmeisterHoffineisterlster 1986 open circles indicate records added by spicer et al 1985 and spicer 1987 subspecific relationships ofofthesethese populations necessary to look at broader distribution pat- are unknown papago springs is a late pleistocene fossil site terns and draw some reasonable inferences which includes a tentative record for this species harris davis et al 1988 analyzed southwestern mon- 1985 tane mammal distributions and found that dis- tance from the source was a significant predictor jones 1962 harlisharrisharnisharrns 1985 questions a fossil of species richness a relationship suggesting record from southeast arizona that would con- dispersal lomolino et al 1989 in a study firm this past distribution but the present dis- encompassing much of the southwest con- junct range of the species fig 1 implies its firmed the relationship between species rich- former presence in southeast arizona regardless ness and isolation and proposed recent of the fossil record post pleistocene climatic dispersal by several montane species including changes fragmented this distribution and local microtus meticmexicanusinexicanusmexicanusanns extinctextinctionsions in southeast arizona apparently sep- arated the mexican and northern populations MEXICAN VOLE distribution this scenario is consistent with the historical legacy hypothesis but there is also evidence that the range of the mexican vole fig 1 pres- the pattern has been modified by recent dis- ently extends from mexico into arizona new persal as discussed below mexico southern colorado and utah durrant EVIDENCE FROM ARIZONA the mexican 1952 armstrong 1972 findley et al 1975 hall vole now occurs in the continuous forests of 1981 hoffmeister 1986 the species typically central arizona and on isolated mountains to the inhabits meadows in ponderosa pine and mixed south southwest and north figs 1 2 four conifer forests but can occupy pinyon juniper populations occur on mountains connected to woodland if suitable understory is present the central high country by pinyon juniper harris 1985 hoffmeister 1986 in arizona it woodland and interior chaparral brown and occurs less often in interior chaparral and great lowe 1983 through which the species could basin desertscrubdesertscrub hoffmeister 1986 disperse the Nannantanesmantanestanes plateau the sierra the late pleistocene distribution of this spe- ancha the bradshaw mountains and the south cies probably was continuous from the mexican kaibab fig 2 three other populations occur plateau to the southwest UUSS findley and at sites that are isolated by grasslands but 264 GREAT BASIN naturalist volume 52

1986 recent dispersal is not the only possible mesa verde explanation for this pattern but it is the most 0 0 parsimonious one ancient relicsrelicts in dissimilar habitats would be expected to show more evi- 0 0 dence of divergence after several thousand 0 years there is evidence of a recent range expan- sion in northeast arizona the mexican vole was first recorded in the navajo mountains in south- sandiasandla mts 0 ern utah and northern arizona in 1933 benson manzano mts 1935 although this locality seems isolated since 1986 the species has turned up at several other sites on black mesa in northeast arizona spicer 1987 these sites fall on a line southeast from navajo mountain to the southwest foot- hills of the chuska mountains at black mesa fig 2 the habitat is pinyon juniper with ponderosa pines and a few doug las firs on north facing slopes draws and other protected areas spicer 1987 again this is 0 160km relatively poor habitat for this species and it seems unlikely that the population could have fig 3 details oftheodtheof the distribution Microtusofmicrotusof mexicanusmcvicanusmexicanus for in new mexico and southern colorado showing some iso- survived in isolation several thousand years lated populations modified from findley et al 1975 some between these sites and navajo mountain is data from hall 1981 open circles indicate records listed mostly pinyon juniper with narrow strips of byfinleyetal1986by finley et al 1986 northern grassland and great basin desertscrubdesertscrub brown and lowe 1983 the mexican vole interconnected by pinyon juniper woodland occupies these habitats elsewhere and presum- and interior chaparral prospect valley the ably can disperse through them this scenario music mountains and the hualapaikualapaiHualapai mountains implies that the chuska mountains now unoc- fig 2 cupied by the species hoffmeister 1986 will since the hualapaikualapaiHualapai mountains and prospect eventually be colonized or recolonized from valley still contain small patches of forest the the northwest vole populations at these sites might be EVIDENCE FROM NEW MEXICO AND COLO- pleistocene relicsrelicts in forest refugia but the RADO findley et al 1975 suggested that the population in the music mountains a site range of microtus mexicanusmexic anus in new mexico midway between the other two consists of only could have expanded as a result of recent dis- pinyon juniper woodland spicer et al 1985 persal in the sandia mountains trapping from this habitat interconnects all three localities 1950 to 1970 revealed only M longicaudus and is more likely to serve as a dispersal corridor mexican voles were first taken there in 1970 and than as a post pleistocene refugiumrefugium the spe- soon became the dominant species while the cies was recorded in the hualapaikualapaiHualapai mountains in species could have been overlooked earlier dis- 1923 and in prospect valley in 1913 but it was persal from the manzano mountains fig 3 is not found in the music mountains until 1981 an equally likely scenario until 1975 these were spicer et al 1985 the northernmost records east of the rio when the rate of dispersal exceeds that of grande river in new mexico the mexican vole extinction a species should be present on those has since been recorded from five sites in montane islands closest to the source assuming extreme northeast new mexico dalquest 1975 the species can cross the intervening habitat finley et al 1986 macarthur andandwilsonwilson 1967 the distribution in colorado the first specimens were taken of the mexican vole in the southwest generally in 1956 at mesa verde rodeck and anderson fits this model fig 2 lomolino et al 1989 in 1956 later the species was found at seven arizona the most closely related isolate popula- more colorado sites fig 3 mellott and choate tions occur in geographic proximity hoffmeister 1984 finley et al 1986 A trapping study in 199211992 VOLE DISPERSAL 265

TABLE 1 southern mammal species for which there is evidence of a recent northward range expansion unless indicated otherwise evidence is based on directionality and chronology of records 1 arizona distribution in cockrum 1960 vs Hoffrhoffmeisterneister 1986 2 distribution in hall and kelson 1959 vs hall 1981 3 texas distribution in taylor and davis 1947 davis 1960 and davis 1974 nomenclature follows jones et al 1986

region and direction species of expansion evidence and references

didelphis virginiajavirginianavirgimanavirgvirginianaimanaimand N through E USU S N into udvardy 1969 mcmanus 1974 Y petryszyn S arizona from N mexico personal communication Mornmornwopsmornuwpswopsmops megalophyllauwgalophyllamegalophylla N in texas 3 taylor and davis 1947 davis 1960 davis 1974 mollhagen 1973 choeronyderischoeronyctens mexicansmexicananwxicanamexicanagana now a winter resident in S R sidner personal communication probably due arizona to hummingbird feeders leptonycterisLeptonycteris sanbornsanbomisanborm now a winter resident in S R sidner personal communication probably due arizona to hummingbird feeders lasiurusLasiurus ega N inm texas spencerspenceretaletalet al 1988 idionyctensldionycteris phyllitisphylphyllotislotis Nnmswusin SW US to utah first USU S record was in 1955 in SE arizonaanzona cock- rum 1956 2 Tatadaridafietnorosaccacaridadarida femorosacca N in arizona landI1 and 2 Tadarida macrotismamacrotincrotis N in arizona also texas r 1 mollhagenMollhagenbagen 1973 dasypus novemcinctus N from S texas into okla- charmanchatmanbuchannanBu and talmage 1954 udvardy 1969 homa colorado kansas humphrey 1974 meaney et al 1987 and nebraska lepus allem limitedly NE in arizona 1 lack of records in N cochise co until 1976 alienallenailen 1895 roth and cockrum 1976 ScittscittrusscitirusScitirusrusnus abertiabenti NW in colorado N into davis and bissell 1989 known dispersal ability wyoming W into utah and history of ponderosa pine distribution davis and brown 1989 baiomys tayloritaylora N from SE texas into diersing 1979 stangl and dalquest 1986 oklahoma and NE in taylor and davis 1947 vs davis 1974 recent new mexico record in luna co new mexico W cannongannon personal communication choate et al 1990 sismodonSissigmodonsigmodontSigmodon hispidus N in the US through cockrum 1952 mohlennchmohimohlenrichMohlenMohlmohlhenrichenrichnch 1961 jones 1960 kansas to nebraska and cameron and spencer 1981 N in rio grande valley in new mexico sigmodonfulviventergigsigmodonSigsigmodontmodon fulviventerfulviventer N in new mexico mohlennchmohimohlenrichMohlenMohlmohlhenrichenrichnch 1961 Sigsigmodonstgmodonsigmodontmodon ochrognathus NW in arizonaanzona and N in davis andlindkind dunford 1987 davis and ward 1988 texas davis etalet al 1988 hollander et al 1990 stangl and dalquest 1991 microtus mexicanusmexicnwxicanusanusunus various in arizonaanzona N in this study new mexico into S colo- rado nasua nasua NW in alizonaarizonaanzona and per not reported by early explorers davis 1982 not haps in new mexico recorded in arizona until 1892 in extreme S hoffmeister 1986 no late pleistocene record hamsharris 1985 tabor 1940 wallmo and Gallizioli 1954 but see kaufmann et al 1976 conepatustwsoleucusconepatusConeconcpatus mesomesoleucusleucus NW in arizona 1 recent records Hoffmeistehoffmeisteri 1986 tayassu tajacu N in arizonaanzona and new indian name for peccary is of spanish originonginsowlsonginSowlsowissowlssowis mexico 1984 rarely encountered by early explorers davis 1982 no use by early prehistoric cultures crosswhite 1984 sowls 1984 266 GREAT BASIN naturalist volume 52

1938 and others prior to 1975 found no mexi- the first such events that have been recorded in can voles near cimarron new mexico although the literature ifthese animals were able to cross other vole species were taken armstrong 1972 unsuitable habitat once then they could have findley et al 1975 the mexican vole is now done so repeatedly in the course of centuries common in the area finley et al 1986 thus our suggestion of recent dispersal by the the northward range expansion by this species mexican vole should be evaluated in the context may be continuing into northeast new mexico of a more general pattern involving many and southeast colorado mammal species post pleistocene dispersal has influenced montane species assemblages throughout much of the southwest lomolino discussion AND conclusions et al 1989 in addition we propose a second pattern of recent northward range expansion historical legacy hypothesis requires the indolinvolinvolvingving at least 19 north american mammal widespread late pleistocene distribution the species all primarily austral in distribution but fossil record documents the late pleistocene occupying a wide range of habitats table 1 presence of microtus mexicanusmexicanus in southern this pattern of northward dispersal is not new mexico adjacent portions of texas and easily explained and there is unlikely to be a perhaps southeast arizona despite the admit- single causative factor some species the tedly weak fossil record however there is no for shift appears to result from climatic change evidence that the species range formerly andor habitat modification by humans alternat- included the entire area where populations now ively the pattern can be viewed as one small exist harris 1985 several lines of evidence recognizable northward surge in a continuing pleistocene dispersal for this support post holocene cycle of northsouthnorth south distribution species shifts I1 distance as a predictor of pres enceabsenceence absence lomolino et al 1989 acknowledgments 2 the close relationship of adjacent ari- zona populations isolated by theoreti- preliminary drafts of the manuscript were cally crossable habitat read by R sidner M V lomolino and E L 3 its presence in isolated habitats unlikely cockrum T van devender and an anonymous to have served as post pleistocene reviewer also provided helpful comments refugia some of the fieldwork cited was conducted by 4 recent records suggesting dispersal in R B spicer and C larue with support from northwest and northeast arizona cen- the arizona department of game and fish tral and northeast new mexico and southern colorado literature CITED although the distribution of the mexican vole undoubtedly has been influenced by historical ARMSFRONG D M 1972 distribution of mammals in col- events and by local extinctionsextinctions it is difficult to orado university of kansas museum of natural his- evidence of and post tory monograph 3 1 415 ignore the past continuing ALLEN J A 1895 on a collection ofmammals from arizona pleistocene dispersal and mexico made by mr W W price with field notes A reviewer of this paper asked why the mex- by the collector bulletin of the american museum of ican vole and other small mammals took 4000 natural history 7 193 258 we claim BENSONBFNSON S B 1935 A biological reconnaissance of navajo years to reach certain localities that mountain utah university of california publications were colonized within the past few decades in zoology 40 439 456 this point requires clarification first there BROWN D E and C H loweLOWF 1983 map biotic commu- have been local changes in vegetation and cli- nities ofthe southwest general technical report RM range the 50 100 78 rocky mountain forest and experiment mate in the southwest during past to station USDA forest service fort collins colorado years and these conditions may have favored bownBROWN J H 1971 mammals on mountaintopsmountain tops non- recent dispersal even though the broader pic- equilibrium insular biogeography american natural- ture has remained constant for some 4000 years ist 195467195 467 478 1978 the theory of insular biogeography and the second we do not claim that these recent distribution of boreal birds and mammals great basin records represent thethefirstfirst colonizations by the naturalist memoirs 2 209 227 mexican vole or other species they are simply buchananbucnananbuc ilananIIANAN G D and R V talmacetalmageTALMACFTALMAGF 1954 the geo 1992 VOLE DISPERSAL 267

graphical distribution of the armadillo in the united 1975 mammals of new mexico university of new states texas journal of science 6 142 150 mexico press albuquerque 360 appp CALLAHAN J R 1977 diagnosis of eutamiasEutamias obscurusobscures FINLEY R BBJJ R CIIOATF and D F HOFFMhoffmfisifrhoffmfisterhoffmeisterFISTER rodentia sciundaesciuridae journal of mammalogy 58 188 1986 distribution and habitats of voles in southeastern 201 colorado and northeastern new mexico southwest- in preparation status of the lodgepole chipmunk ern naturalist 31 263 266 tamias speciosusspeciosus in the san jacinto mountains cali- HALL E R 1981 the mammals of north america 2 vol- fornia umes john wiley and sons new york CAMERONCAMEKON G N and S R SPENCERSPFNCFR 1981 Sigsigmodonsigmodontmodon HALL E R and K R KELSON 1959 the mammals of hispidus mammalian species no 159 9 appp north america 2 volumes ronald press new york CIIOAIECHOATE L LLJJ K JONESJONFS JR R W MANNING and C HARRIS A H 1985 late pleistocene vertebrate paleoeco- JONESJONFS 1990 westward ho continued dispersal of the logy of the west university of texas press austin pygmy mouse baiomys taytagtayloritaylorztaylorialoriiorilorz on the llano estocadoestacadoEstacado hoffmeisterHOFFM eisterelsterFISTER D F 1986 mammals of arizona univer- and in adjacent areas of texas occasional papers 134 sity of arizona press tucson and arizona department the museum texas tech university of game and fish 602 appp COCKRUM E L 1952 mammals of kansas university of HOLLANDER R RRBB N HICKS andandaandjJ F SCUDDAYsc UDDAY 1990 kansas publications of the museum of natural history distributional records of the yellow nosed cotton rat 717 1 303 SignisigniodonsigmodonSigsigmodontsogodonmodon ochrognathus bailey in texas texas journal 1956 new genus and species Idioidionycterisidionyctensnycteris phyl of science 42 101 102 lotis recorded from the united states journal of HUMPHREY S R 1974 zoogeography of the nine banded mammalogy 37 546 547 armadillo dasypus novenicinctusnovemcinctus in the united 1960 the recent mammals of arizona their taxon- states biosciencebioscience2424 457 ff omy and distribution university of arizona press JONES J K JR 1960 the hispid cotton rat in nebraska tucson 276 appp journal of mammalogy 41 132 crosswhitfcrosswhitecrosswiiite C D 1984 the significance of cacti in the JONES J K D C CARTER H H GFNOWAYSgenoways R S HOFF diet of the javelina tayassu tajacu desert plants 6 MAN D W RICE andandcanacC JONES 1986 revised check- 3 48 list of north american mammals north of mexico DALQUFSIDALQUEST W W 1975 the montane and mexican vole 1986 occasional papers 107 1 22 the museum in northeastern new melcomeleomexico and adjacent colorado texas tech university southwestern naturalist 20 138 139 KAUFMANN J HDH D V LANNINCLANNING andsand S E POOLEPOOLF 1976 DAVIS G B 1982 man and wildlife in arizona the amer- current status and distribution of the coati in the ican exploration period 1824 1865 depaitmentdepartment of united states journal of mammalogy 57 621 637 game and fish phoenix arizona LOMOLINO M VVJJ H bbownBROWN andrand R DAVIS 1989 island DAVIS R and S J BISSELL 1989 the distribution of biogeography of montane forest mammals in the abert s squirrels in colorado evidence of a modern american southwest ecology 70 180 194 expansion of range southwestern naturalist 34 306 macaktiiurmacaiitiiur R handeH and E 0 WILSON 1967 the theory 309 of island biogeography princeton university press DAVIS R and D E BROWN 1989 the role of post princeton new jersey pleistocene dispersal in determining the modern dis- MCMANUSMLMANUS J J 1974 didelphis virginiajavirgvirginianavirgimanavirglvirginianaimana mammalian tributiontribution ofabert s squirrels great basin naturalist 49 species no 40 6 appp 425 434 MEANEYMEANFY C AASS J BISSELL andland J S SLAIERSLATER 1987 A DAVIS Rrandcgrandcand C DUNFORD 1987 an example of contem- nine banded armadillo dasypus novemcinctus porary colonization by small nonflying mammals of dasypodidae in colorado southwestern naturalist montane islands in the american southwest american 3250732 507508sot507 508sos naturalist 129 398 406 MFLLOTIMELLOTT R sandasandjS and J R CHOATECHOATF 1984 sciurus abattiabertiabetti and DAVIS R C DUNFORD and M V LOMOLINOLk mollnoMOLINO 1988 mon- microtus montanus mexicanusmexic anus on foothills on the tane mammals of the american southwest the possible culebra range in southern colorado southwestern influence of post pleistocene colonization journal of naturalist 29 135 137 biogeography 15 841 848 mohlenrichmohlhenrichMOHL ENRICH J S 1961 distribution and ecology of the DAVIS R and 0 G WARD 1988 A vacant microtus hispid and least cotton rats in new mexico journal of niche now occupied by the yellow nosed cotton rat mammalogy 421342 13 24 SigsigmodonsigniodonSignisigmodontmodonodon ochrognathus on an isolated mountain in MOLLHAGEN T 1973 distributional and taxonomic notes southeastern arizona journal of mammalogy 70 362 on some west texas bats southwestern naturalist 17 365 427 429 DAVIS W B 1960 the mammals oftexas texas game and PATTEpaytePAITERSONRSON B D 1984 macmahanmammalianmammahanMammahan extinction and bioge- fish commission bulletin no 41 ography in the southern rocky mountains pages 247 1974 the mammals of texas texas parks and 293 inm M H nitecki ed extinctionsExtinctions university of wildlife department bulletin no 41 chicago press DIERSINCdl ersEPS incING V E 1979 additional records of baiomys PATTERSONpsiPAIpst 1 ersonEPSON B dandwalmarDandD and WWAlMARATMAR 1986 nested subsets tabiontaylon taylon thomas in texas southwestern natu- and the structure of insular mammalian faunalfaunas and ralist 24 707 708 archipelagosarchipelagoes biological journal of the linnaean soci- DURRANT S D 1952 the mammals of utah taxonomytaxonomyandand ety 28 65 82 distribution university of kansas museum of natural RODFCK H G and S ANDERSON 1956 sorex mesamimemamimerriainimerriaini history publications 6 1 549 and microtus mexicanusmexic anus in colorado journal of mam- faf1FINDLEYN D LEY J sandcsanacS and C lonksJONFSJO N F S 1962 distribution and variavarlavana malogy 37 436 tion of voles of the genus microtus in new mexico and ROIHROTH E L and E L COCKRUM 1976 A survey of the adjacent areas journal of mammalogy 43 154 166 mammals of the fort bowie national historical site findleFINDLFFINDLEY J S A H HARRIS D E WILSON andandcanacC JONES in cockrum et al eds survey of the vertebrate fauna 268 GREAT BASIN naturalist volume 52

of foitfoltfortport bowie national site arizona university of worthy records of oklahoma mammals southwestern arizonaanona projectpioject no 50102849025010 2849 02 sponsored by USU S naturalist 31 122 124 park service project no CPSUUA 005 1991 historical biogeography of Sigsigmodonstgmodonsigmodontmodon SOWLS L 1984 the peccariespeccaries university of arizona press ochrognathuschronchronathusathus in texas texas journal of science 43 tucson 127 131 SPENCERSPI N 1 K S G PCR C chioucjiouCIIOUCAIRaiiiaidi and B R CHAPMAN taborTABWTABOK F W 1940 range of the coati in the united states 1988 northward expansion of the southern yellow bat journal of mammalogyMarnma logy 21 11 14 Lasilatinislawinislasiunisunis ega in texas southwestern naturalist 33 TAYLOR W P and W B DAVIS 1947 the mammals of 493 texas texas game fish and oyster commission bul sisphSPKi c I1 ii R B 1987 status of the navajo mountain mexican letin no 27 volevolcvoievoic microtusmexicanusmicrotus mexicanusmexicanus navaho benson report pre UDVARDYUDVAKDY M D F 1969 dynamic zoogeography van paredpaipal ed foifolfor the USU S fish and wildlife service nostrand reinhold new york 445 appp SPIC 1 K R B R L GLINSKI C DEVOS 1985 status siiciii andaandj WALLMO 0 C and S GALLIZIOGALLIIOLILL 1954 status of the of vole thetiietile hualapaikualapaiHuiiualapaivolealapai microtusMicro tuftus mexicanusmexicanus hualpaiensis coati in arizona journal of mammalogy 35 48 54 goldman report prepared for the office ofofendanendan- gered species USU S fish and wildlife service stanclSIANCLSTANGL F BBJRlitjil andandwanawW W dalquestDALQUFSIDALQUFST 1986 two note received 12 october 1990 accepted 17june17 iunelunejune 1992 great basin naturalist 523 appp 269 277

CAN TOWNSENDS GROUND SQUIRRELS SURVIVE ON A DIET OF EXOTIC ANNUALS

I1 2 eric yensen and dana L quinney

ABSTRACT southwestern idaho desert shrub bunchgrassbunch grass rangeland is being invaded by fire prone exotic annuals that permanently dominate the landscape following wildfireswildfires this study was undertaken to describe diets oftownsend s ground squirrels spermophilus townsendtownsendiitownsendiaii idahoensis at foulfour study sites with varying degrees of exotic annual invasion to determine if the squirrels could utilize high proportions of exotic annuals in their diets townsend s ground squirrels were collected in march and may of 1987 and 1988 and stomach contents were analyzed using a micromicrohistologicalhistological technique grasses comprised 37 87 of townsend s ground squirrel diets at the four sites native species especially sandbergsandbergszandbergsSand bergss poa bluegrass secunda winterwinterfatfat ceratoidesCerat oides lanata big sagebrush artemisia tntritrldentata and six weeks fescue vulpicvulpia octofloraoctoflora constituted 7 96 x 47247 2 of the diet whereas exotic species especially cheatgrasscheatgrass bromus tectoriumtectommtectorumtecthoteotorumtomm tumbleweed salsola iberica and tansytansymustardstansymnstardsmustards descurainia sppapp made up 4 68 x 48048 0 of the diet at each site 2 A4 species comprised 90 ofthe diet there was no apparent correlation between the importance values of exotic species at a site and their importance in townsend s ground squirrel diets

key words spermophilus townsendtownsendiitownsendiaii food habits dietary analysis idaho ground squirrels

the snake river birds of prey area is a bunchgrassesbunch grasses and shrubs apparently provide a 243000 ha tract of multiple use shrub steppe more constant stable food source than exotic rangeland administered by the US bureau of annual species that may vary in productivity land management townsend s ground squir- between wet and dry years by several orders of rels spermophilus townsendii idahoensis are magnitude young et al 1987 important prey of rapraptorsraptorestors and continued exis- like other ground squirrels of subgenus tence of the area s dense breeding populations spermophilus townsend s ground squirrels eat of raptorsraptores depends upon dense townsendTown sendss green vegetation early in their four to five ground squirrel populations US department month active season then eat seeds of grasses of interior 1979 and forbs to fatten up for hibernation howell invasion of southwestern idaho rangeland 1938 rickart 1982 in southwestern idaho by exotic annuals such as cheatgrasscheatgrass bromus townsendtownsendsTownsendss ground squirrels are in estiva tectorumtectoriumtectorum tumblemustardtumble mustard sisymbrium altis tionhibemation from june or july until the fol- sibumsimum pinnate tansytansymustardmustard descurainia lowing january or february with low survival pinpinnatapinnatenata and tumbleweed salsola iberica has rates ca 28 smith and johnson 1985 food resulted in frequent and destructive wildfireswildfires quantity and quality could influence overwin that kill native shrubs and weaken native bunch teringbering survival as well as reproductive success grasses over time fires have resulted in the the following spring permanent replacement of native shrub and townsendTownsendss ground squirrels are known to bunchgrass dominated communities by exotic eat native forbs sphaeralcea davis 1939 bunch annual dominated communities yensen 1980 grasses poa sp june grass koeiKoelkoeletlakoeleriaKoeleriaeTla sp davis kochert and pellant 1986 1939 and desert shrubs big sagebrush arte- townsend s ground squirrel populations are misia dentatatritridentatatridentate budbudsagebuddagesage atreATteartemisiamisiamisla spinescensspinescentspines cens much less stable in exotic annual dominated shadscaleshadscale atriplex conferticonfertifoliafolia davis 1939 commmunities than in native shrub communi- johnson 1961 as well as insects such as grass- ties yensen et al 1992 native perennial forbs hoppers and cicadas and occasionally carrion

I1 museum of natural history albertson college caldwellCil dwell idaho 83605 213nreaubuieauoflindof land management 3948 developmentdp lolmentcolment aeave boise idaho83705idaho 83705 present addressaddless idahoid ihoibo anny nationalN itionaition 1 guard departinentdepirtinentoffnviioninentof friviron nt box 45 gowen field boise idahoidalio 83705

269 270 GREAT BASIN naturalist volume 52

takiTABLETABI 1 1 vegetation importance values relativerelalelareiative cover plus frequency in may 1987 and 1988 at four study sites nealnear coyote butte in the snake river birds of prey area southwesternsouth westein idaho

studystudy site

big native exotic rehabilitation species sagebrush grasses annuals seeding

1987 1988 1987 1988 1987 1988 1987 1988 gnassiGHASSIGRASSES s breviu13broviuBro13ronalsronalshinninviu tectortcctorumtectoruntuhtuhiunt 25 2 11 33 86 35 0 31 poa secundawcuncla 67 60 90 58 45 45 85 60 vulpiavulpicviiipui octofloraoctoflora 16 7 24 0 2 8 12 3 sitanionSitanion hystrixmyernmysrn 16 11 14 14 8 21 0 28 Aoaopironagropynnipironpinon deserdesertorumdesertoruindesertoyesertorumruin 0 0 0 0 0 0 29 2 slihubss1111u13s Ceratclrntoidtceratoidesoides lanata 29 47 3 5 0 0 7 0 Artenarteiniwiartentisiatisia tridentata 33 39 0 0 0 4 0 0 atriplex nuttalfiiuuttnllii 0 0 0 0 0 0 0 20 foisfols saiidiolaSatxdiolasalsolasatsolasolasoia ibericail erica 0 0 33 26 0 0 40 18 1dccnrainiidescurainiasophiaophia 0 0 3 0 0 0 0 0 twjinhriumsisymbriton altiswnumaltisshnuin 0 11 8 31 34 14 0 5 nacinlactucaserriolaLacinlacingdGd arnolacrnola 0 0 0 0 0 0 4 0 otlieioffiereffier forbs 0 5 0 9 0 2 0 2 tolatulaTOIALCOVIL COVEgove 11 35 24 26 15 21 10 18 14 ofieofic speckspechspeekspeeksspeckss

howell 1938 alcorn 1940 however they do proportion of exotic annual species in the habi- eat introduced cheatgrasscheatgrass tumblemustardtumble mustard tat however the study was not designed to peppergrass lepidium perfohatumperfoliatum davis study dietary preference as such 1939 as well as crop species like alfalfa wheat barley potatoes beets carrots and lettuce STUDY SITES howell 1938 johnson 1980 and rogers and gano 1980 study sites were located near coyote studied diets of spermophilus townsentownsenddnii four butte approximately 19 km south ofkunaof kuna ada townsentownsenddnii in washington and found in native county idaho in the snake river birds of prey bluegrass poa sp 26 29 and lupine area sites described below were selected 11 25 the lupinus laxifloruslaxiflorous to be dietadietanlydietarilydiedletanlyrily for progressively greater deviation from undis- whereas the important descurmmadescurainia was only turbed native vegetation exotic eaten in quantity 15 33 cheatgrasscheatgrass UNBURNED BIG SAGEBRUSH this site tumbleweed tumblemustardtumble mustard and peppergrass TIS RIW sec 24 elev 850 m is a big sage- constituted 0 4 of the diet al tis 01 johnson et brush winterwinterfatfat mosaic and represents the 1977 estimated the percent volume of food unburned condition of the other three sites big categories in 174 townsendtownsendsTownsendss ground squirrel sagebrush winterainterfatwinterfatfat and native grasses stomachs prey in the snake river birds of area sandbergsandbergszandbergsSandbergss bluegrass poa secundalsecunda squirrel- they found glassesgrasses including cheatgrasscheatgrass were tail sitanionSitanion hystrix and six weeks fescue most important followed by forbs and winterwinterfatwmterfatfat vulpicvulpia octofloraoctojloraoctofloravlora dominate the site cheatgrasscheatgrass Ceratceratoidesoides lanata is the main exotic annual table 1 because cheatgrasscheatgrass tumbleweed tumble NATIVE GRASS this site tstist1s r1wraw sec mustard and peppergrass are becoming 13 elev 850 m is 1 km northwest of the increasingly dominant in the snake river birds unburned big sagebrush site in a former big of prey area this study was designed to learn if sagebrush wintera4nterfatwinterfatfat community burned by a townsendtownsendsTownsendss ground squirrels were substituting human causeda41dfirecaused wildfire on 26 august 1983 the these exoticsexotica for native species in their diets we fire killed the shrubs and the site was domi- also wished to learn if consumption introduced nated subsequently by native sandberg s blue- plant species increased with increases in the grass six weeks fescue and squirreltail with 1992 TOWNSENDS GROUND SQUIRREL DIETS 271 some introduced tumbleweed cheatgrasscheatgrass and stomachs were removed from the animals other exotic annuals present table 1 immediately postmortem and preserved in 70 EXOTIC ANNUALS this site tisTIS RIW ethanol in the lab stomach contents were sec 13 elev 850 in is adjacent to the native removed from ethanol diluted 50 with water grass site and was similar to it prior to the 1983 and homogenized I1 min in a waring blender to burn D L quinney unpublished data both produce fragments of uniform size the homog- sites were burned by the same fire however enate was washed through a immI1 mm sieve since the fire the exotic annuals cheatgrasscheatgrass and hansen 1978 and collected in a 010 immI mm screen tumble mustard with some remnant native to remove tiny unidentifiable fragments the grasses especially sandberg s bluegrass table material was then mounted on microscope I1 have dominated the site slides using hertwig s and hoyerhover s media rehabilitation SEEDING this site is sparks and malechek 1968 located 6 km east 252.5 km south tstisTt1sIS rieR I1 E sec plant species in the diet were identified by 27 elev 885 in of the unburned big sagebrush comparisons to a reference collection of micro site the area burned in 1981 was seededrereseeded scope slides using micromieromicrohistologicalmicrobistologicalhistological characters with desert wheatgrasswheatgrass agropyron desertorumdesettorumdeserdeseTt torumorum all reference slides were made from cataloguercataloguedcatalogued in 1982 but burned again in 1983 in 1987 and specimens in the albertson college harold M 1988 the area was dominated by sandberg s tucker herbarium and werewere prepared using the bluegrass desert wheatwheatgrassgrass tumbleweed and technique described above other native and exotic forbs table 1 for food habits analysis one slide was exam- ined per stomach occurrence of food catego- METHODS ries frequency was recorded from each of 20 microscope fields per slide using a phase contrast microscope at loox frequency2020 to determine the degree of exotic annual Frequency fields was then converted to percent relative invasion at each site vegetation analysis was density sparks and malechek 1968 conducted in early june 1987 and late may 1988 using a table developed for frequency to density con- while townsendTownsendss ground squirrels were being version fracker and brischle 1944 collected at each site we used a transect with importance of each dietary category was forty 1 mm2ma quadratsquadquadransrats spaced at 10lom m intervals the calculated in three ways 1 percent relative daubenmire 1959 percent cover of each spe density a standard dry weight conversion from cies was estimated using a I1 mm2ma quadrat frame im frequency data sparks and malechekM 1968 2 divided into tenths to facilitate estimation to frequency in stomachs the give a better approximation of the availability of percent percentage of stomachs from a site with the item and 3 each plant species percent relative cover and frequency fields percent relative frequency were converted to percent in microscopic the of all fields from importance values cox 1990 percentaget microscopic a site with the item squirrels were collected by trapping and twenty fields shooting at all four sites in may and june 1987 microscopic were examined from each slide a n 75 and in march and may 1988 n 42 using predetermined pattern and frequency of occurrence of each except from the rehabilitation seeding site in species was recorded frequency of each dietary may 1988 squirrels were aged in the field using the fields pelage and body weight criteria bureau of category20category 20 on one slide was compared with slides land management unpublished data repres- other or replicate counts of the same slide using the kulcyznski index entativesentative specimens were prepared as 1 stan costingoosting 1956 also well known as the simi- dard study skins with skulls n 12 2 bray curtis larity index and 1957 skeletons n 3 or 3 skulls only n 25 and bray curtis deposited in the albertson college museum of 2waw a b yensen natural history tooth wear patterns the index was calculated as a dissimilarity index 1991 were consistent with the acreage assignments b aw for all specimens based on these criteria all I1 2w a b 1987 specimens were juveniles since they were using a BASIC microcomputer program pro- collected late in the active season while the vided by ludwig and reynolds 1988 adults were entering seasonal torpor all 1988 weather data were from the national oce- specimens were either yearyearlingsyearningslings or adults anic and atmospheric administration monthly 272 GREAT BASIN naturalist volume 52

TABU 2 late season 25 may 19 june 1987 townsend s ground squirrel diets data are from stomachs ofjuvejuvenilenileDile TGS at louiloulfour sites in the snake river birds of prey area adults were entering torpor and none were collected during this period dietary composition is given as percent relative density RD percent frequency in microscope fields MF and percent frequencyliellefiequency in stomachs PS for each dietary category other symbols 1 absent and n number of stomachs

unburned native exotic rehabilitation big sagebrush grasses annuals seeding dietary category RD MF PS RD MF PS RD MF PS RD MF PS n 21 20 15 19 GRASSESGKASSFS broinusbromufgroinus tectoruintectoriumtectorumtectectortorumuin 22 41 71 62 93 100 31 45 87 57 74 95 poa secunda 24 35 86 2 25 7 8 40 11 sitanionsitamonsidamonSitanionanlon hystrix 2 10 13 Onzooryzopsisoncopus himehimenoideshymenoideshymenoides 5 grass seed 5 19 1 5 1 13 classGglassrass lootroot 2 9 10 2 5 3 13 total glassesgrasses 49 64 39 57 sllkubssljiluls ccratmdeiceratoklesCeratokles lanata 3 9 52 4 5 10 43 59 67 11 17 32 temisiaartemisiaar tnkntatah1dentata 2 14 1 3 10 6 19 67 2 5 atriplex nuttalnuttathihi 5 2 5 13 4 12 37 chnwthamnusc&71sothainnus viscidiflorusvi&cidiflorufviscidiflorus 2 5 total shrubs 3 5 51 16 forbsFOKBS salelasdwlasalwla ibeticaihencaiberica 39 69 91 3 40 1 3 20 7 17 63 sisyndrlumisiiiihnum altissimum 5 5 19 60 2 5 21 dewiruimadescurainia 2 sppapp 2 24 5 lepidiuntLepidilefndiuinpcrfohatumuhtuhlunt perfoliatuin 2 6 20 1 13 1 3 21 cryptanthaCryptantha mterrwptainteri tiptacipta 5 ranunculus testiculatustcsticulatus 1 5 lactuca semiolajemiolawn alawla 1 5 25 2 4 5 chenopodiaceae 5 7 2 21 unidentified forb 6 7 11 total forbs 40 11 2 18 misceuaneousmls I1 lianiLLANI OUS insects 8 17 62 19 44 90 7 21 87 3 11 53 fungi 1 4 10 1 7 2 5 unknown 5 4 7 4 5 5 unidentified seed 1 5 5 total miscellaneous 8 20 8 8 idaho climatological data reports for the kuna the vegetation at each of the four sites 2 NNE weather station ca 20 km N ofthe study varied significantly all p oi0101.01 R X C G tests sites of independence between years table 1 importance values averaged 65 similar range 48 77 at a site between years RESULTS total percent cover decreased on all sites in 1988 in 1988 when there was less herbaceous cover the sites vegetation analysis were slightly more similar x 61361.3 range 47 74 thus each site differed almost as vegetation at each site table 1 vanedvaried the much between years as the sites differed among significantly from the other sites all p 01 each other in a given year R X C G tests of independence sokal anand rohlf 1981 using the kulcyznski index the stomach analyses similarity among the four sites averaged 48748.748 7 range 27 73 in 1987 the unburned sage- although the three measures of dietary brush site was more similar 60 to the native importance percent relative density percent grass site and less similar to the exotic annual dry weight percent frequency in microscope and seeding sites 44 and 47 respectively fields percent frequency in stomachs gave 199211992 TOWNSENDS GROUND SQUIRREL DIETS 273

different numerical results the rank orders duceddeuced annuals bristly crcryptcryptanthaypanthatantha cryptanthaCrypt antha among categories were generally consistent interruptintermptainterruptainterruptsa was the only native forb found in tables 2 4 however percent frequency in townsendtownsendsTownsendss ground squirrel stomachs although stomachs was very sensitive to sample sizes 1988 sample sizes were small the importance of there were 1 9 food categories per stom- forbs in the diet increased in the samples ach site means varied from 383.8 to 444.4 categories between march and may 1988 while the per- per stomach the total number of food catego- centage of grasses and shrubs decreased tables ries used by all townsendtownsendsTownsendss ground squirrels 3 4 thus suggesting large seasonal differences sampled at a site varied from 4 to 17 on the three between march and may diets sampling occasions may june 1987 march A surprising number of insects were eaten 1988 may 1988 however if species used in especially in may june 1987 3 19 table 2 trace amounts 5 relative density are elim- however insects were not important in 1988 inated only 3 6 x 404.0 categories were used trace amounts at the big sagebrush site only per site and only 2 4 species comprised 10 insect remains were so fragmentary that identi- ofthe diet species comprising 1 I100 of the diet ficationfication was not usually possible however at one or more study sites included sandberg s abundant lepidoptera larvae could be recog- bluegrass cheatgrasscheatgrass six weeks fescue winter nized by the soft exoskeleton and prprolegsprologsolegs and fat big sagebrush tumbleweed descurainiadescwrainia fragments recognizable as beetle antennae and sppapp seeds of bur buttercup ranunculus tes elytraclytra were found ticulticulatusatus and insects the importance values of exotic species grasses were important constituents of the were lowest at the unburned big sagebrush site diet in both 1987 and 1988 and often comprised in both years and highest in the exotic annual over 50 of the diet 37 88 relative density site in 1987 and at the native grass site in 1988 tables 2 4 sandberg s bluegrass and however there was no correlation between the cheatgrasscheatgrass were both heavily utilized especially importance values of all exotic annuals at a site in march 1988 55 87 of diet late in the and their importance in the diet at that site r townsendtownsendsTownsendss ground squirrel active season may 454454.454 tables 1 4 and june use of grasses declined except at the exotic annual site in 1988 most of the grass discussion eaten in may june consisted of seeds especially of cheatcheatgrassgrass sandberg s bluegrass leaves were the data show that for sites with varying utilized slightly more than cheatgrasscheat leaves grass degrees of exotic annual invasion sampled over tables 2 4 and the two far together were a two year period townsendtownsendsTownsendss ground squirrels more important than all other grasses com- can and do utilize introduced species in their bined squirreltail was little used although it diets and that cheatgrasscheatgrass tumbleweed and was the third most abundant grass tumblemustardtumblemustard are the most important of these winterfatWinterfat 0 43 relative density and big both the vegetation at a site and townsend s sagebrush 0 21 were both eaten and ground squirrel diets varied considerably winterwinterfatfat was especially important at the exotic between years and among sites differences in site where least it was abundant Winterwinterfatfat was amount of precipitation most likely account for utilized at all sites 1987 in even though it was the differences in vegetation importance values abundant not enough to be sampled by the between years there was less september may vegetation analysis at the exotic annual site in precipitation 192 mm in 1986 87 and 170 mm 1988 it was eaten only at the unburned big in 1987 88 at kuna ca 20 km N the sagebrush winterwinterfatfat site and its use declined daubenmire quadratsquadquadransrats were taken on the same between march and may 1988 table 2 big transect in both years by the same technicians sagebrush was used all in march at sites in both the substantial annual differences in but less years was important in may townsendtownsendsTownsendss ground squirrel diets may be the tumbleweed and tumblemustardtumble mustard were the result of 1 vegetation differences between most important forb species consumed tansy years 2 the fact that juveniles were sampled mustards descurainia sophia and D pinnatepinnatapinnata in 1987 and adults and yearyearlingsyearningslings were collected peppergrass seeds ofbur buttercup and leaves in 1988 3 differences in collecting dates 25 of prickly lettuce lactuca sefsersetiolarriolafriola were of may 19 june 1987 versus 16 19 may 1988 or secondary importance all of these are intro 4 small sample sizes volume 52 274 GREAT BASIN naturalist

s squirrel diets are from stomachs of adult and yearling TAISI 1 3 early season march 1988 townsend ground data available in march dietary composition is TGS at four sites in the snake river birds of prey area juveniles were not in in fields MF and percent frequency of stomachs given as percent relative density RD percent frequencyfieflequency in microscope 1 absent number of stomachs PS containing each dietary category other symbols n unburned native exotic rehabilitation seeding big sagebrush grasses annuals RD MFME PS RD MFME PS dietary category RD MFME PS RD MEMF PS

5 7 16 ti 4 gnassiGKASSIGRASSES s 39 70 100 39 35 75 67 46 100 51 16 71 Broibrommrufsruts tectonontectommtectonon 39 71 100 poasecunda 48 73 100 16 92 100 4 86 100 PoaPW secunda 6 vulpicvulpiav111pia octofloraoctoflora 1 25 1 144 sitanionSitanion hystrix 1 1 13 Aagroptjronropyron deertonimdvsettontin 55 79 total glassesgrasses 87 83

SHRUBSS I1 I1 RU ils 86 oides lanata 24 55 Ceratceratoides 3 6 13 dentata 11 41 50 15 35 60 21 46 100 artemisia tntrltri 3 8 19 nutunuutnuttalliinuttallnnuttalliaallnalinaliuliiill atriplex nuttalnutt 6 total shrubs 11 15 45 FOKBS salwin iberica 1 2 4 56 sisyndrimn altissimumaltis&imum 1 5 20 14 sistuthnum 10 25 75 descurainiadescuraima 2 sppapp 2 6 25 testiculatuitesticulatus ranunculus 2 4 13 cnptanthaCryptcryptanthaantha internipta 1 7 13 loni halogetonhaioHalo eton ionigloineratusgloinlonieratuseratus 6 lepidbonjjcpidiuin perfoliatumperfoliatumr 6 creotscrepiscrepts acuininataacwninata 6 LacnacnoctuLi serriolamolar lactuca 4 iai313 chenopodiaceae 15 total forbs 2 1

were found differences in age classes were probably at each site several plant species the stomachs that did not important fitch 1948 found no differ- in townsend s ground squirrel the analysis for that ences in adult and juvenile diets in california not appear in vegetation 1959 method of vege- ground squirrels S beecheyibeecheyi nor did hansen site the daubenmire acceptable and johnson 1976 find differences in wyo- tation analysis gave an intuitively of dominant vegetation but for estab- ming ground squirrel S eleganseldelegansegans diets by sex or estimate close link between plant abundance age class dyni and yensen in preparation lishing a scale method of found no dietary differences between and herbivore diets a finer is since individual adultyearlingadult yearling and juvenile age classes in idaho resource analysis necessary s squirrels have large home S bmnneushrunneus or columbian S columbianuscolumbianus townsendtownsendsTownsends ground m2ma and johnson ground squirrels on the other hand the 1988 ranges mean 1357 in smith radius is not sur- data do show a strong seasonal component 1985 with a wide foraging it thus the observed annual dietary differences prising that townsendTowntownsendssendss ground squirrels were by the vegetation be a result of later collecting dates inm 1987 eating species not recorded inaymay were relatively coinbinedvithcombined with annual vegetation differences analysis even though the sites although 117 stomachs were examined the homogeneous no correlation between the total sample siesslessizes were too small to draw many con there was abundance of exotic annuals at a site and their iiiliidiisions111siolisdivisions111 siolis about interintersitesite and between season in the diet number of plant lictslichtsIK ts milllillii 1988 this amount of collecting had a importance the the diet not recorded by the vegeta- diclctcrioiis1 lenonstenons effect on local townsendtownsendsTownsendss ground species in determining dietary I1 tion analysis precluded sliiiiSIIIIKlellei densities and we recommend use of annual dietary preference indices for townsendTownsendss ground olicrglicrotlnotan i iiietliodsUK tliods if seasonal or examination of vegetation slnllss1lifts aiarearc ofdillterestdillterest squirrels however 199219921 TOWNSENDTOWNSENDs S GROUND SQUIRREL DIETS 275

TABLETABLF 4 late season may 1988 townsendtownsendsTownsendss ground squirrel diets data are from stomachs of adult and yearling TGS at four sites in the snake river birds of prey area dietary composition is given as percent relative density RD percent frequency in microscope fields MF and percent frequency of stomachs PS containing each dietary category other symbols 1 absent n number of stomachs site 4 was not sampled in 1988

unburned native exotic big sagebrush grasses annuals

dietary category RD MEMF PS RD MFME PS RD MFME PS

n 1 5 4 GRASSFSGRASSES poa secunda 2 10 100 19 40 80 bromus tectoriumtectorumtectorum 22 65 100 35 57 80 24 34 75 vulpicvulpia octofloraoctoflora 29 75 100 21 36 100 6 18 100 sitanionSitsitamonhystnxanion hystrix 1 5 100 4 6 40 5 18 50 agropyron desertorumdesertorum I1 5 100 grass seed 2 10 100 4 10 20 2 1 25 grass root 5 13 60 total grasses 57 88 37 SHRUBS Ceratceratoidesoides lanata 13 25 50 artemisia tntridentata 3 8 50 atnplexatriplex nuttalnuttalliinuttallialii 3 15 100 6 14 20 2 4 25 chrysothamnus viscidiflorusviscidiflorus 1 25 total shrubs 3 6 18 FORBS salsola ibehcaibenca 22 60 100 sisymhnumsisyn7hrium altissimum 6 18 60 1 4 75 descurainiadescuraima 2 sppapp 15 33 50 ranunculus testiculatus 17 55 100 28 43 50 forbforbrootforbrookroot 1 20 totalforbstotal forbs 39 6 44 miscellaneous insect 1 25 unknown 7 25 total miscellaneous 0 0 abundance table 1 in comparison to consumptconsump was not recorded by the vegetation analysis tion tables 2 4 indicates that most of the small amounts of it were found in two stomachs abundant plant species were also important in at the rehabilitation site in march 1988 table the diet and that rare plants were being used 3 halogeton is poisonous to livestock but only in trace amounts there were some inter- sheep can eat it with impunity in winter proba- esting exceptions to this however cheatgrassCheatgrass bly because rains have leachedbeached the oxalates out was dietarilydietarily important 39 relative density of the dried leaves cook 1977 presumably but not recorded in the vegetation analysis at the townsend s ground squirrels were eating dried rehabilitation site rather than fresh leaves in march diets became more diverse in may probably idaho and columbian ground squirrels have as a result of grasses curing and seeds becoming highly varied diets of 11 25 plant species per available ground squirrels eat large amounts of fecal pellet group dyni and yensen in prepa- seeds prior to entering torpor rickart 1982 E ration however in that study only 2 4 plant yensen personal observation perhaps if insuf- species usually grasses contributed 10 to ficient seeds are available during a drought year the diet rogers and gano 1980 found that townsendtownsendsTownsendss ground squirrels turn to insects as a only three plant species poa sppapp descurainia fat source however at the exotic annual site pinnatapinnatepinnata and lupinus laxifloruslaxiflomslaxiflorous contributed where insect use was highest in 1987 cheatgrasscheatgrass 10 of the diet of townsend s ground squir- mostly seeds was the major constituent of the rels in southeastern washington hansen and diet this relationship should be explored further ueckert 1970 found 1 5 species contributed although halogeton halogeton glomeratusglomeratus 10 in the diverse 47 plant species diets of 276 GREAT BASIN naturalist volume 52 wyoming ground squirrels in colorado hansen herbarium specimens and we especially appre- and johnson 1976750 concluded that ciate M P luscher s assistance in preparing the slides R G anthony D R johnson S knick rieharichardsonRicha idsonldson wyomingWyorning ground squirrels graze on A K and B van a variety of plants as they fill their stomachs rather E rickart steenhof homehorne than selecting only pipreferredefelebel red foods when their stom- made helpful comments on an earlier draft of achs are nealnearlyly empty this may be an evolutionary the manuscript strategy developed to allow them to consume vetchdetchesvetchesvetchcscs the dilution of toxic foods by nontoxicnon toxic foods decreasesdecidecldeel eases the probability of plant poisoning literature CITED freeland and janzen 1974 reviewed strat egles of herbivoreherbherbivoryivory by mammals in response to ALCORN J R 1940 life history notes on the plute ground of mammalogy 21 160 170 secondary plant compounds they suggested squirrel journal BRAY J R andandaandjJ T CURTIScunCUR us 1957 an ordination of the that a generalist herbivore should feed predom- upland forest communities of southern wisconsin 349 inantly on one or two foods I1 but continue to ecological monographs 27 325349325 sample other foods present when an herbivore COOK C W 1977 effects of season and intensity of use on desert utah agricultural experiment sta a nutritional deficiency it should vegetation experiences tion bulletin 483 utah state university logan 57 appp sample all available foods until it finds some- cox G W 1990 laboratory manual ofgeneral ecology ath6th thing which supplies that nutrient ed wm C brown publishers dubuque iowa 251 appp feeding strategies proposed by freeland daubenmire R F 1959 A canopy coverage method of the 66 and janzen 1974 and hansen and johnson vegetation analysis northwest science 33 43 DAVIS W B 1939 the recent mammals of idaho caxton 1976 appear to occur in several members of printers ltd caldwell idaho 400 appp the subgenus spermophilus the data indicate FIILII H S 1948 ecology ofthe california ground squirrel that ground squirrels specialize on 2 4 highly on grazing lands american midland naturalist 90 nutritional species but supplement them with a 334 340 frackerFRACKFRFRACKFII S B andandaandjJ A biiisiileBRISC HLF 1944 measuring the wide variety of other species apparently as poi- local distribution of ribes ecology 25 283 303 soning insurance in this study townsend s FREELANDFRFFLAND W J and D H lanenlanzenjanenJANFNJANZEN 1974 strategies in ground squirrels similarly depended on only a herbivoreherbivoryherbivory by mammals the role of plant secondary 269 289 few species for the bulk of the diet but a wide compounds american naturalist 108 shasta aground sloth foodhabitsfood habits ram available HANSENHANSFN R M 1978 Shastshastagroundground variety of trace species was not if any part cave arizona paleobiology 4 302 319 of these species should provide insufficient HANSENHANSFN R M and M K JOIINSONJOHNSON 1976 stomach con quantities of a key nutrient eg linoleic acid tent weight and food selection by richardson ground necessary for hibernation then the limited squirrels journal of mammalogy 57 749 751 HANSENHANSFN R M and D N UFCKFRIUECKFRT 1970 dietary simi could have negative ueckert selection of food species lantylarity of some primary consumers ecology 51 640 population consequences 648 question ofwhether townsend s ground HOWELLHOWFLL A H 1938 revision of the north american the amen squirrels can utilize exotic annuals as dietadietaryry ground squirrels with a classification of north ameriamerl can sciuridae north american fauna 56 1 256 is answered the affirmative by this staples in JOIINSONJOHNSON D R 1961 the food habits of rodents on study native forb species were of minor impor- rangelands of southern idaho ecology 42 407110407 410 tance in the diet but this does not necessarily JOIINSONJOHNSON D R G W SMIHISMITH R M OLSON 1977 popu reflect native forbs are now so rare lation ecology and habitat requirements of townsend preference birds by the ground squirrels pages 203 225 in snake river at the four sites that none were recorded of prey research project annual report 1976 vegetation analysis and thus they may not have JOHNSONJOIINSON M K 1980 food of townsend ground squirrels been available for consumption only one native on the andaridarld land ecology reserve washington forb Cryptcryptanthaantha was found in the stomachs great basin naturalist 37 128 KOCIIFRI M N and M pellanipellantPFLLANIPFLLANT 1986 multiple use in limited dietary on the consequences of variety the snake river birds of prey area rangelands 8 the longtermlong term nutrition of townsend s ground 217 220 squirrel are unknown LUDWIGLUDWIC J A and J F REYNOLDS 1988 statistical ecolacol oey john wiley & sons new york OosOOSTINGriNC H J 1956 the study of plant communities and2nd acknowledgments ed W W freeman san francisco RICKAIHRICKART E A 1982 annual cycles of activity and body we thank J weaver and other bureau of composition in spenwphilusspemwphilus townsendtowntownsendutownsendiitownsendiasenduii mollis cana land management personnel who collected the dian journal of zoology 60 3298 3306 rogriisboorocrooRoG FRSergeng L E and K A GANO 1980 townsend ground data B and riiseilsells squirrels and the vegetation dyni squirrel diets in the shrub steppe of southsoutheentralsouthcentralcentral T foppe for discussion of the micromicrohistologicalhistological washington journal of range management 33 463 analysis technique P L packard for access to 465 199219921 TOWNSENDS GROUND SQUIRREL DIETS 277

SMITH G landdwandd R JOHNSON 1985 demography of a area progress report USU S department of interior townsend ground squirrel population in southwestern bureau of land management boise idaho 82 appp idaho ecology 66 171 178 YENSFNYENSEN E 1991 taxonomy and distribution of the idaho SOKAL R R and F J ROHLF 1981 biometry and2nd ed ground squirrel Spennspermophilusophilus brunbrunneusbrunneousneus journal of W H freeman & co san francisco 859 appp Marnmamammalogylogy 72 583 600 SPARKSSPAR Ks D R and E K randarandjJ C MALECHMALFCIIEK 1968 estimating per- YENSENYFNSEN EDE D L QUINNFY K JOIINSONKJOHNSON K timmermantim mer MAN centage dry weight in diets using a microscopic tech- and K STFFNIIOFSTEFNHOK 1992 townsend s ground squirrel nique journal of range management 21 264 265 population fluctuations in southwestern idaho amer- US departmentDEPARIMENT OF INTFRIOR 1979 snake river birds icanlean midland naturalist in press of prey special research report US bureau ofland YOUNG J A R A EVANS R E ECKFRTECKERT JR andaandband B L KAY management boise district boise idaho 142 appp 1987 cheatgrassCheat grass rangelands 9 266 270 YENSENYFNSFN D L 1980 A grazing history of southwestern with idaho emphasis on the snake river birds of prey received 5 may 1991 accepted I1 september 1992 gloatgleatgreat basin naturalist 523 appp 278 283

AVIFAUNA OF CENTRAL TULE VALLEY WESTERN bonneville BASIN

peter hovingh

keykelkei words birds avifauna desert aquatic habitat great basin wetlands

fautin 1946 described the flora and fauna ponds three cornered bulrush scirpus amer of several northern desert biotic communities in icarusicanus and salt grass distichlis spicatespicataspicata are the tule valley located 80 km west of delta utah dominant emergent species with phragmites in millard county of western bonneville basin australis typha domingensisdomingensis and scirpus his study during 1939 june to september and acutisacutus occurring in highly localized stands 1940 april to september included a descripdescript tamarisk tamarix ramosissima is the only tion of greasewood sarcobatus vermiculatus shrub growing within some springs wetlands and pickleweed alienailenAllenallenrokeaallenrolfearokearonea occidentoccidentalisoccidentalistalis com but was not noted by fautin 1946 munitiesmunities from 1980 through 1991 while invenanven toryingstorying the aquatic habitats of tule valley I1 METHODS noted the avifauna utilizing wetlands springs greasewood and adjacent pickleweed commu- A total of 36 visits were made to tule valley and saline flats nities this note reports on the between 1980 and 1991 with 10 visits of two avifauna within a occurringvithinoccurring the two communities day durations occurring in 1981 inventories and the 1980 91 faunal with compares lisitng were conducted duringeachduringduningduringeacheach month exceptjanexcept jan- that reported previously by fautin 1946 com uary with emphasis during march may and pariparnsonssons are also made with fish springs june birds were inventinventoriesinventoriedoried by random national wildlife refuge located 50 kmkin north encounters and unidentified species were not of the tule valley springs this study identifies pursued nomenclature follows that ofpetersonofpeterson changes in raptorsraptores and songsongbirdsbirds that have 1990 occurred over 40 years and notes the differences between natural springs and wetlands and those dedicated to waterfowl management RESULTS AND discussion

table I1 lists the 80 species ofbirds identified description OF THE TULE VALLEY during 1980 91 the months they were encoun- AQUATIC environments tered and those species also reported by fautin 1946 mallard scientific names noted in table within the greasewood and pickleweed 1 northern harrier horned lark common communities of central tule valley are some 25 raven and marsh wren were encountered fissure fault springs and associated wetlands year round and are considered permanent resi- saline flats covered in part by water from saline dents almost half 31 of the species ilventoinvento seepage springs occur to the east and west of ried during this study were observed two or these fissure fault springs the springs wetlands fewer times dates included in table 1 and are vary in size from 100 m2ma to over 97000 mm2ma2 considered casual or transient visitors the coyote springs with a total of 195000 m2ma single palm warbler a casual bird in utah conductivity of the aquatic systems varies from behle et al 1985 was identified by its charac- 1200 spring sources to greater than 93000 teristicte tail movement as previously observed by umhos per cm some wetlands and saline me on numerous occasions during annual

72721 sondsecondS ond avenue salt laklake city utah 84103

278 199211992 NOTES 279

migrations in the midwest the saline ponds valley which consists of 5 permanent residents west of the fissure fault springs hosted gulls and a total of 17 summer residents the larger numerous waterfowl and shore shorebirdsbirds during number of species at fish springs national migration wildlife refuge probably reflects the availabil- fifteen species great blue heron turkey ity of surface water the presence of trees and vulture sharp shinnedchinned hawk coopercooperss hawk buildings and the proximity ofthe springs wet swainsonswainsonsswainsonaSwainsonss hawk red tailed hawk burrowing lands to the mountainous fish springs range owl common nighthawk western kingbird tule valley springs wetlands are noinorthemnorthernnor thebthem mockingbird yellow warbler undeveloped and lack the man made features an additional yellow breasted chat green tailed towhee factor that may contribute to the brewerbrewerss sparrow and lark bunting observed difference in avifauna of by fautin 1946 were not encountered in this constituency tule valley and fish springs is the contribution study burrowing owls while nesting in the over many years of field ornithologists adjacent shadshadscalescale community were at fish springs national not wildlife observed in the greasewood community refuge the two birds western absence of raptorsraptores in particular the swainsonswainsonsswainsonaSwainsonss sandpiper and lincoln s sparrow have hawk and the turkey vulture only occasion not been reported in this region ally seen but observed throughout the summer in the latilongLatilong study walters and sorenson 1983 in one community or another fautin 1946 and the lincoln s sparrow was not 285 could reflect the rangeland predator con- reported at fish springs US department of trol programs occurring in tule valley since the interior 1988 fish springs and tule valley fautin did his studies absence of other species area in the same latilongLatilong region and fish mentioned above could reflect the loss of wil- springs observations overwhelm the tule valley lows salix exigiaexigua which fautin 1946257 had observations within the latilongLatilong study noticed as being prevalent most of the birds fautin reported for the greasewood community that were not observed during the present study conclusions were considered transients by fautin 1946 over 157 species with 41 permanent resi- A listing of the avifauna for central tule dents those species that can be found in all valley is reported comparisons are made to the seasons and 54 nesting species have been avifauna list reported by fautin 1946 and to reported for fish springs national wildlife the species list prepared by the fish springs refuge US department ofthe interior 1988 national wildlife refuge differences in spe- this contrasts sharply with the avifauna of tule cies are noted and explanations are offered

TABLF 1 distribution of birds in the greasewood wetland community of tulethletuie valley

month of year

J F M A M J J A S 0 N D specific dates podicipedidae pied billed grebe x x podiceps nignnigrinigricollisnigncolliscollis 8881 61282 eared grebe x podilymbus auritusanntusauretus 62081 ardeidaeARDFIDAFARDEIDAF american bittern x botaurus lentiginosus 92984 great blue heron ardea heroberobevoherodiaschasohms snowy egret x egretta thula 61382 black crowned night heron x x nycticorax nycticorax 81881 102090 thrrskiornitiiidafthreskiornit111dae faced white ibis x plegadis chiaichihichthi 82187 82391 280 GREAT BASIN naturalist volume 52

tahleTABLF 1 continued

month of year

J F M A M J J A S 0 N D specific dates

ANAIIDAF canada goose x 3787 branta canadensis green winged teal x 42781 anas crecca mallard x x x x x x x x x x x anas platyrhynchosplatyrhynchous northern pintail x x x anas abutaacuta cinnamon teal x x x x x x x anas cyanoptera american digeonwigeon x x anas americana canvasback x 32282 aythya dailuailvalivalisinenavalisineriadeitsvalissineriainena redhead X X X aythya amamericanaricana merganser x mergus sp ruddy duck x x x oxyurajainaicensisoxyura jarnaicensis caliiariiuarcjvri1a1r11dae turkey vulture cathartes aura accipitridafaccipiikidaf northern harrier x x x x x x x x x x x circus cyaneuscyaneous sharp shinnedchinned hawk accipiter stristnatwstriatusstriatumatus coopercooperss hawk accipiter coopedcooperiicoopem swainsonSwamswamsonssonss hawk buteo swainsonswainsoniswainsomswainsonisom red tailed hawk buteojamaicensisbutco jarnaicensis rough legged hawk x 3781 buteo lagopus golden eagle x 32090 aquilaagulia chrysaetos amerleanamericanAmencan kestrel x 9258292984 falcosparveriusfalco iparvenusiparvenus prairie falcon x x 4482 51188 falco ntexicanusmexicanusmexic anus RALLIDAERALLIDAF virginia railrallrali x x x rallus li nicola sora x x x porzana carolina american coot x x x x x x x fulica americana cliakauriidafcharadwidae killdeer x x x x charadrius vocivociferusvociferousferus recurvirostridae black necked stilt x 82187 himantopus mexicanusmexic anus 1992 NOTES 281

TABLFTABLE 1 continued

month of year

J F M A M J J A S 0 N D specific dates scolopacidae spotted sandpiper x 82187 actitisactipis maculariamacularia western sandpiper x 42086 caicalidnscalidrisCalidris maurimauffmaun dunlin x 42086 calidnscaicalidrisCalidris alpina common snipe X X X x x x x cautGaUigautgallinagonago gallinagosallinago LARIDAFLARIDAE gulls x lamslarus sp columbidae t mourning dove x x x x x zenaida macmacrouraroura STRIGIDAE burrowing owl athene cunicularcuniculariacumculanaia CAPRIMULGIcaprimulgidaedaeDAF common nighthawk chordeiles minor APODIDAE

white throated swift X X X aeronautesAeronautes saxasaxatahssaxatalistalistails PICIDAE northern flicker x x 102581 12681 colalptes auratusauranus tyrannidaeTYRAN nidalNIDAI western kingbird tyrannus vertiverticalisverticalismcalis ALAUDIDAEALAUDIDAF horned lark x x x x x x x x x x x efemEremeremophilaerenwphilaophila alpestnsalpealpestrisstris hirundinidae violet green swallow x x 8881 61382 tachycineta thalassinathalassinothalas sina barn swallow x 91981 hirundo rustica CORVIDAE common raven x x x x x x x x x x x corascorvus boraxcorax troglodyiidaftroglodytidaetroglodyridae marsh wren x x x x x x x x x x x cistothorusCistocistophorusthorus papalustnspalustrislustris muscicapidaeMUSC ICAPIDAE mountain bluebird x 82481 sialia currucoides MIMIDAFMIMIDAE northern mockingbird mimus polyglotpolyglottospolyglottoustos

sage thrasher X X X X oreoscoptes montanus motacillidae american pipit X X X X anthus rubescensrubescentrubescens LANIIDAE loggerhead shrike x x x laniuslamus ludovicianusludovicianus volume 52 282 GREAT BASIN naturalist

TABLE 1 continued

month of year

S N D specific dates J F M A M J J A 0

STURNIDAESTURNIDAK starling x x 221813781 stumusstamus vulvulgarisvulgurisguris emberizidae yellow warbler dendroica letechiapepetechiatechia yellow rumpedbumped warbler x x x x dendroica coronata palm warbler x 91981 dendroica paipalpalmarumpalnwrummarum common yellowthroat x x x x geothlypis trichasdrichas yellow breasted chat latelalctelaicteria birensvirens green tailed towhee pipilo chlomruschloncruschlochiochlonchionmruscrus x x 91680 12681 american tree sparrow spizella arborea brewer s sparrow spizella breweribrewere vesper sparrow x 92081 pooecetes gramineusgragraminousmineus 5287 lark sparrow x chondestes gramgrammacusgramnwcusmacus black throated sparrow x x x x amphispiza bilineatebilineatabilineata sage sparrow x x x x x x amphispiza belli lark bunting calanuspizacalanwspiza melanocorysffwlanocorysmelano corys savannah sparrow x x x x x x x x x x Passerpasserculusculus sandwichensis x 102090 fox sparrow passerella iliacailinca song sparrow x 12581 melospiza melodia 4481 lincoln s sparrow x melospiza lincolniilincollincolnianii white crowned sparrow x x zonotrichia leucophrysleueophrysleucleueophrys junco x illjuncosajuncospnco sp red winged blackbird x x x x x agelaius phoeniceusphoeni ceus western meadowlark x x x x x x stubellastumellaStumella negneglectaneglectslecta yellow headed blackbird x x x x x xanthocephalus xanthocephalus x brewer s blackbird x eupharuseuphagusEup hagus cyanocephalus brown headed cowbird x x x x x x molothrus ater fringillidae x 12681 american goldfinch carduelis bistis passeridaePASSER IDAE 102581 house sparrow x passer domesticusdomesticus

iclentifiedd tifatif 1 by fautin 1946 dates inin right column luieluteare toifottotfor two or fewer observations 199211992 NOTES 283

acknowledgments FAUTIN R W 1946 biotic communities of northern desert shrub biome in western utah ecological monographs 1625116 251310251.31025116251310310 I1 wishimshdimsh to thank david E joyner and clayton PETFRSONPETERSON R T 1990 A field guide to western birds M white for reviewing the manuscript and for houghton mifflin co boston 432 appp subsequent discussions US department orOF THE INTERIOR 1988 birds of the fish springs national wildlife refuge dugway utah rf6 65531 2 references WALTERS R E and E SORENSENSORFNSEN EDS 1983 utah bird distribution latilongLatilong study 1983 utah division of wildlife resources publication 831083 10 97 appp belilebeilleBEHLEBEIILE W HEff E D SORENSENSORFNSEN andandcanacC M WHITE 1985 utah birds a revised checklist occasional publication 4 utah museum of natural history salt lake city received 10 november 1991 108 appp accepted 22 june 1992 giedtgleatgreat basin naturalist 523 appp 284 287

WILDFIRE AND SOIL ORGANIC CARBON IN SAGEBSAGEBRUSHRU SH bunchgrassBUNCH GRASS vegetation

I1 steven A acker

big sagebrush artemisiaArternisianisla tntritrldentata wyorningensiswyomawyomm key words soil organic matter soil organic carbon wildfire degradation oregon Arterartemisianisianisla tntritrldentata tntritrldentata bunchbunchgrassgrass longtermlong term site

METHODS soil organic matter is an important compo- nent of the environment for plants one that of enhances availability of water and nutrients I1 studied soil organic matter at two pairs nelson and sommers 1982 contributes to a burned and adjacent unburned big sagebrush county suitable seedbed monsen and mcarthur 1985 bunchbunchgrassgrass stands in northern harney selected along and enhances seedling emergence wood et al oregon USA the stands were other for a study of post wild 1978 in the sagebrush region of the inter- with seven pairs sagebrush bunchbunchgrassgrass vegetation mountain west loss of organic matter due to fire big acker 1988 I1 selected as study recurring wildfire may be a mechanism of long- dynamics stands burned and adjacent unburned areas in term site degradation ultimately caused by which at least one of four climax bunchgrassbunchgrass excessive livestock grazing and the introduction species was present bluebluebunchbunch wheatgrasswheatgrass ofaggressive annual plants west 1988 loss of agropyron spicatumspicatum pursh scribn & smith matter or plant cover due to fire may organic indian ricegrassricegrass oryzopsis hymehynwnoideshymenoidesnoides R & erosion and decrease infiltration increase S ricker needleneedie and thread stipa comatacamata decreasing seedbed quality monsen thereby & duprrupr and thurber s needleneedlegrassgrass stipa of organic matter trin and mcarthur 1985 loss thurberthurbeiianathurbenthurberianahonanona piper hironaka et al 1983 the soils less friable and more likely iana may also render climate is semiarid 28928928.9 cm annual precipita- so increase the to form crusts upon drying and on average for burns oregon about 40 km overcome tion resistance emerging seedlings must north of the study area with hot dry summers is wood et al 1978 on the other hand it and cold winters franklin and dyrness 1973 conceivable that the increase of the introduced soils are stony and shallow over lava or welded annual cheatcheatgrassgrass bromus tectectoriumtectorumtorum L that ash deposits and are classified as lithic xerollicXerollic may follow wildfire west 1988 may increase haplargids mixed with lithic torriorthents soil organic matter over the long run due to lindsay et al 1969 within pairs the sites are litter accumulation documentation of the similar in elevation slope aspect and surface response of soil organic matter to wildfire in the soil texture table 1 other than incidental use sagebrush region is limited on relatively mesic none of the four stands was grazed by domestic several big sagebrush artemisia dentatatritrlttidentatatridentatatridentate nutt sites livestock during this study or over the occurrence of a single fire apparently does decades M armstrong personal communica- on all the not decrease organic matter in the surface soil tion shrub skeletons were present to the recent fires layers dimirnimir and payne 1978 humphrey burned stands thus prior had similar fire histories 1984 this study concerns the effect ofwildfire paired stands probably wildfire occurred in august 1981 on soil organic matter in relatively xeric big the initial sampled in the early summer sagebrush sites acker 1988 the stands were

science college of forestry madison wisconsin 53706 present addressaddi ess department of foiestforest idepartmentdep irtment ofif botany universityuniveisityofwisconsmof wisconsin madison oregon state university corvallis oregon 97331

284 199211992 NOTES 285

TABLE 1 environmental historical and vegetation data for burned odd numbers and adjacent unburned even numbers big sagebrush bunchgrassbunch grass stands harney county oregon USA soil texture determined by method of liegel etalet al 1980

stand elev aspect slope soil texture dominant plant 1 b number Wm category W top 10 cm species 1985

1 1325 9 17 sandy loam BRTE ERFI POSE PHHO ORHY 2 1325 8 12 sandy loam ARTRW PHHO aspiASFI 3 1360 3 19 loamy sand BRTE stc02 CHVI 4 1360 2 22 loamy sand ARTRT BRTE CHNA

al1 SSW 2 SSWS SW 3 SSEWSWSSE WSW 4 SEAsewSE W 5 ESEWNWESE WNW 6 ENWE NW 7 ENENNWENE NNW 8 NENNE N 9 NNE based on muirmun and lotan 1985 categoriesategoneantegone 1 4 are warm aspects categories 5 9 are cool aspects 1 plants with at least 3 covelcover in descending ordetorder ARTRT artemisiaArtemista tntritrldentata sspasp tridentata ARTRW artemisia tritrodentata sspasp iomingcnwwyoiningensisu ASHASFI astragalus filipes BRTE bromus tectorumtectoriumtectorum CHNA chrysothamnus nmiseosusnauseosus sspasp albialbicaulisalbtcauliscaulis CHVI chryiothamnnschnjsot1wmnus viscidiflonisviscicltflonis sspasp cisuchfloniviscidiflonis ERFI etigeronfilifoliusengeron filtfoltus ORHY oryzopsis hynaynhymenouiesoidesoldes PHHO phlox hoodiihowhihoodei POSE poa secandacundasecunda stc02 subastipaguba colatacoinatacomata voucher specimens on file at university of wisconsin madison heibherbherbariummum

TABLE 2 comparison of organic carbon in top 10 cm of soil in burned and adjacent unburned big sagebrush bunchgrassbunchgrass stands northern harney county oregon USA values are mean percentages of mass of oven dried soil standard errors in parentheses standard errors were computed using each stanstand s variance for 1987 and the number of subsamples for the year listed petersen and calvin 1986 the number of degrees of freedom for all tests is 30 E nordheim personal communication

result of two tailed t test year organic carbon N burned vs unburned

stands I1 and 2 1985 burned 1190241 19 0 24 3 4 P 2 NSnsfns1 unburned 0830230 83 0 23 3 1986 burned 1170211 17 0 21 4 P 55nsansNS unburned 1340201 34 0 20 4 1987 burned 1310101.31010131 olo010 16 4pap4 P 2nsans2 NS unburned 1150101 15 0 10 16 stands 3 and 4 1985 burned 063017 3 P 5 NS unburned 0680160 68 0 16 3 1986 burned 060015 4 P 5 NS unburned 0650.140 65 0 14 4 b 065014 1987198711 burned 083007 16 P 5 NS unburned 0840070840.070 84 0 07 16

not significant 1 both stands 3 and 4 burned between the 1986 and 1987 samplings

of 1985 1986 and 1987 stands 3 and 4 burned mate oxidation method of organic carbon deter- again in a wildfire september 1986 minationmination nelson and sommers 1982 I1 used I1 collected samples from the top 10 cm of the standard correction factor of 131.3 to adjust for soil 3 samples per stand in 1985 4 in 1986 and organic carbon not oxidized in the procedure 16 in 1987 in the first two years sampling loca- given the uncertain quantitative relationship tions were laid out in a systematic manner in between soil organic carbon and soil organic 1987 samples were collected in a stratified matter I1 report soil organic carbon as nelson random manner the randomization for the and sommers recommend 1982 only remaining unburned stand stand 2 was I1 used two tailed t tests to compare organic further restricted so that the area under shrub carbon between paired stands sokal and rohlf canopiescanopies was sampled roughly in proportion to 1981 for 1985 and 1986 I1 used the sample the cover of shrubs in the stand shrubs can variance from the 1987 observations and the influence spatial patterns of soil chemistry in big sample size from the year in question to deter- sagebrush vegetation doescher et al 1984 mine the denominator of the test statistic organic matter of the soil samples was petersen and calvin 1986 this was done due assessed using the walkley black rapid dicarodichro to the larger sample size and the stratified volume 52 286 GREAT BASIN naturalist professor bockheim and random arrangement of the 1987 samples fieldwork and jim soil analysis and access greig smith 1983 in the strictest sense these kurt schulz for advice on observations can only establish differences to laboratory facilities between adjacent stands applying these results sagebrush to burned and unburned big literature CITED bunchbuncligrassbunchgrassgrass stands more generally is tenuous due to the lack of replication hurlbert 1984 ackeraokerACKTR SAS A 1988 the effects ofwildfire on big sagebrush bunchgrassbunch grass vegetation in southeastern oregon theory RESULTS AND discussion and observations unpublished dissertation university of wisconsin madison 204 appp chemical changes in of stands there was no signif- BLANK R randarandjR and J A YOUNG 1990 for both pairs soil induced by fire in a community dominated by in the top 10 the icant difference in organic carbon shrub grass pages 256 259 in E D mcarthur E M cm of soil in any of the three years table 2 romney S D smith and P T tueller compilers none of the individual comparisons is sugges- proceedings symposium on cheatgrassCheatgrass invasion off and other aspects ofshrub biology and difference P 2020.20 in all cases shrub die tive of such a USDA forest service general techni- for a management although I1 did not test statistically tempo- cal report INT 276 intermountain research station ral trend soil organic carbon does not appear to ogden utah WINWARD have changed over the course ofthe study in any DOFSCIIIDOFSCIIIRDOESCIIFRR P S R F MILLERMILLFR and A H of fire at 1984 soil chemical patterns under eastern oregon of the stands thus the recurrence dominated by big sagebrush soil have altered plant communities stands 3 and 4 does not appear to science society of america journal 48 659 663 soil organic carbon FRANKLIN J fandcfandaF and C I1 DYRNESSDYRNFSS 1973 natural vegeta- changes in organic matter are by no means tion of oregon and washington USDA forest service general report PNW 8 pacific northwest only ecologically important soil changes technical the forest and range experiment station portland wildfire may cause in big sagebrush vegetatiovegetation oregon 417 appp soil ed eg increase of organic acids in burned soisol GREIGGKFIG SMITIISMIFII P 1983 quantitative plant ecology ard3rd blank and young 1990 furthermore the short university of california press berkeley 359 appp limit the gener- HIRONAKA M M A fosbergFOSBFRG and A 11 WINWARD duration and small sample size ofsouthern idaho stands are 1983 sagebrush grass habitat types ality of conclusions however these bulletin 35 university of idaho forest wildlife and not unlike others in the general vicinity where range experiment station moscow 44 appp of climax bunchbunchgrassesgrasses persist acker 1988 in HUMPIIRFY L D 1984 patterns and mechanisms plant grass sites in south- addition these stands offer a rare opportunity succession after fire on artemistaatteinisiaartemisha eastern idaho vegetatiovegetation 57 91 101 to observe big sagebrush bunchgrassbunchbuncligrassgrass vegeta- and the design of huHURLBFRI11 LB E RT S H 1984 pseudorephcationPseudopseudoreplicationreplication tion processes in the absence of livestock grazing ecological field experiments ecological monographs wildfire apparently has not decreased or 5418754 187 211 SCIIULIF 1980 increased soil organic matter on these stands LIFGFLLIEGFL E AACC R SIMON and E E wisconsin procedures for soil testing plant analysis 1 concluded that post fromfroin other studies I have and forage analysis no 6 soil fertility series stands and and feed wildfire vegetation dynamics in these department of soil science university of wisconsin similar ones nearby is dominated by cheatgrasscheatgrass extension madison 51 appp norgren G H increasing abundance of LINDSAY M G B B LOVFLL J A NORGRFN and does not feature D W ANDFRSONANDERSON 1988 to explain SIMONSON B R tilTiITHOMASOMAS and climax bunchbunchgrassesbuncligrassesgrasses acker 1969 malteurmalheur lake drainagedramagebramage basin general soil map something 1 s these trends may require invoking report with irrigable areas appendix 11212 of oregon oregon other than irreversible site degradation as indi- long range requirements for water state of 79 appp cated by loss of soil organic matter water resources board salem mc A RT I1 I1 U 11 1985 factorsfactorsinfluinfluinflux monsenMONSFNMO N S E N S bandeB and E D mcaiunur broadleaf herbs and encinaencing establishment of seeded acknowledgments shrubs following fire pages 112 124 in K sanders J durham et al eds rangeland fire effects a sympo- bureau of land fellow sium USU S department of interior research was supported by graduate management idaho state office boise and ships andor research grants from the national mumMUIR psandjP S andl E lolanLOIANLOTAN 1985 disturbance history ecol- science foundation the university of wiscon- serotinaserotinyseroserotmytiny of finusrinuspinus conconfortacontortatorta in western montana of the ogy 66 1658 1668 sin graduate school the davis fund carbon NFLSONNELSON D wandlW and L E SOMMERSSOMMFRS 1982 total wisconsin departments of botany pages 539 579 university of organic carbon and organic matter in and zoology and signiasigma xi I1 thank malteurmalheur A L page R H miller and D R keeney eds meth- field station and the bureau of land manage- ods of soil analysis part 2 chemical and microbiolog- society ofagronomy ment burns oregon for logistical support of ical properties ended2ndedand2nd ed american 1992 NOTES 287

inc and soil science society of america inc madi- weslWESIWEST N E 1988 intermountainintel mountain deserts shrub steppes son wisconsin and woodlands pages 209 230 in M G barbour and nimmNIMIR M B and G F PAYNFPAYNE 1978 effects of spring W D billings eds north american terrestrial vege- burning on a mountain range journal of range man- tation cambridge university press new york 31 agement 259 263 WOOD M KWK W H BLACKBURN R E ECKFRIECKERT jilJR and PFIFRSFNPETFIISFN R Ggandland L D CALVIN 1986 sampling pages F F PEIFRSONPETERSONPE lersonlensonIFRSON 1978 interrelations of the physical 30 51 in A klute ed methods of soil analysis part 1 properties of coppice dune and vesicular dune inter- physical and mineralogical methods 2ndand ed american space soils with grass seedling emergence journal of society of agronomy inc and soil science society of range management 31 189 192 america inc madison wisconsin SOKAL R R and F J ROIILF 1981 biometryBiornetry and2nd ed W H freeman and co new york 859 appp received 17 september 1991 accepted 22 june 1992 great basin naturalist 523 appp 288 289

STRUCTURE OF A WHITE TAILED PRAIRIE DOG BURROW

I1 lynn A cooke and steven R swiecki2swiecki 2

key words cynomys leucurusleucurus burrow structure hibernaculum nest

little published information is available on of an oval mound 151.5 m long 121.2 m wide and the structure of white tailed prairie dog cyn- 020.2 m high at an angle of 70 for approximately omys leucurusleucurus burrows clark 1971 1977 050.5os m and leveled off at a depth of 040.4 050.5 m described the structure of two partially exca- tunnels connecting entrances measured 80 vated burrows in wyoming and bubsbumsburns et al 220 mm high and 80 200 mm wide and were 1989 described structure and function of approximately circular in cross section these another burrow in montana neither of these connecting tunnels were all within 050.5os m of the studies reports finding either hibernating ani- surface A tunnel leading to the nest chamber mals or remains of known hibernators who died descended further turning bays as described over winter this note describes the structure of by scheffer 1937 for black tailed prairie dogs a burrow system in colorado that had a known were found near one entrance D fig 1 history of prairie dog use for two years prior to the nest chamber tunnel descended from an excavation burrow excavation was undertaken entrance without a mound D in fig 1 A side to establish fates of two juveniles who hiber- tunnel connected to the mound after branch- nated in the burrow in 1988 but were not ing the tunnel gradually descended to a maxi- resighted in 1989 mum depth of 1251.25 m another branch closer the excavated burrow is located on the to the nest appeared to rise and was not exca- arapaho national wildlife refuge walden vated due to time constraints the tunnel lead- colorado jackson county t8ntan r79w s5sa ing to the nest chamber was 115 150 mm wide dominant shrub species include greasewood and 105 225 mm high in front of the nest sarcobatus vermiculatusvenniculatus rabbitbrushrabbitbrush chryso- chamber were three small chambers 190 350 thamnus nauseosus and sagebrush artemisia mm long and 100 225 mm in diameter one of tritrldentata dominant grasses are wheatgrasseswheatgrasses these chambers 350 mm before the nest cham- agropyron sppapp the burrow system was exca- ber contained old fecal material whitehead vated by hand in june 1989 during excavation 1927 reported a feces filled chamber in a measurements were taken periodically of depth black tailed prairie dog burrow and suggested and dimensions of tunnels and chambers prairie dogs used it to avoid drowning the four entrances were located A B C and D present burrow system however had no provi- in fig 1 one of these entrances had an asso- sion to trap air if submerged foster 1924 ciated mound remaining entrances opened other chambers near bends in the tunnel may into semicircular pits approximately 060.6og m in have permitted animals to pass one another no diameter no material had been transported stored food was found in any chambers from below the surface or from the surrounding an enlarged chamber was located at the end surface to form a crater as constructed byblack of the burrow system this chamber had a tailed prairie dogs cynomys ludovicianusludovicianus domed ceiling a bowl shaped floor and mea- king 1955 cincotta 1989 all entrances sured 210 mm high by 210 mm wide by 250 mm except the mound were filled with loose soil long contained within the chamber was a mass the main entrance descended from one end of dry well chewed plant material primarily

departmentdepal tinenttenent offsystematicsandsystematics and ecology universityuniveiuniver sity of kansas lawrenceloilolLiiwrence kansas 66045210666045 2106 present address department of biology college of the holy cross woicesteiworcestcy massachusetts 016100 6102395610gioglo0161023952395 15410 I1 lelenhelenheien sontligatesontligiteSontsout ligate michigan 48195

288 199219921 NOTES 289

TOP VIEW acknowledgments

this research was supported in part by from the B grants university of kansas general research to K B the C fund armitage univer- sity of kansas department of systematics and ecology the theodore roosevelt memorial fund and sigmaxisigmaSigmaXixi the US fish and wildlife tb service tb tb N kindly permitted work on the arapaho

D national wildlife refuge we thank E C patten and J solberg for assistance in locating 1 suitable prairie dog study colonies the manu- script was improved by the comments of an anonymous reviewer

SIDE VIEW literature CITED C D B

M BURNS J ADA D L FLAIHFLATH andtwandewand T W CLARK 1989 on the structure and function of white tailed prairie dog bur- rows great basin naturalist 49 517 524 CLARK T W 1971 notes on white tailed prairie dog cyn- 1 structure of fig excavated white tailed prairie dog omys leucurusleucurus burrows great basin naturalist 31 burrow capital letters indicate entrances to the burrow 115 124 nest is system the chamber is indicated by a solid star the 1977 ecology and ethology of the white tailed location of feces filled is prai a chamber is indicated by a solid riene dog cynomys leucurusleucurus milwaukee public triangle bays are indicated by turning tb museum publications in biology and geolologyGeolology no 3 97 appp CINCOTTA R P 1989 note on grasses was nest chamber mound architecture of the this probably a and black tailed prairie dog great basin naturalist 49 not a food storage area because the plants found 621 623 were not preferred food plants kelso 1939 EGOSCUE H J and E S FRANK 1984 burrowing and personal observation several small denning habits of a captive colony of the utah prairie out dog great basin naturalist 44 495 498 pocketingspocketings were found off the nest chamber FOSTER B E 1924 provision of prairie dog to escape while the nest chamber and adjacent chambers drowning when town is submerged journal of mam- and outpocketings superficially a malogy 5 266 268 resembled KELSO 11 L H 1939 food habits of prairie dogs USDA maternity area as described by bums et al circular no 529 15 appp 1989 this burrow had no known use as a KING J A 1955 social behavior social organization and maternity burrow in three years to excava- population dynamics in a black tailed prairie dog town prior the tion did however in black hills of south dakota contributions from it resemble deep permanent the laboratory of vertebrate biology no 67 univer- systems described by egoscue and frank sity of michigan ann arbor 123 appp 1984 SCHEFFER TTHH 1937 study of a small prairie dog town transactions of the kansas academy of science 40 within the nest materials were skeletal 391 394 remains and an eartagcartag of a subadult female who VISHFR S S 1945 climatic maps of geologic interest hibernated in 1987 and was not resighted in bulletin of the geological society ofamencaofAof americameneamenca 56 713 1988 average frost depth in this area is between 736 WHITEHEAD L C 1927 notes on prairie dogs journal of 500 mm and I1 m visherwisher 1945 just above nest Marnmamammalogylogy 8 58 chamber depth juvenile males who used this burrow as a hibernaculum in 1988 were not received I1 may 1991 sightedreresighted nor were their remains found accepted 15 may 1992 gleatgreat basin naturalist 523 appp 290 292

HYBRIDS OF WHITE TAILED AND MULE DEER IN WESTERN WYOMING

charles E kay 1 2 and edward boe 131 3

key worcswordsporcs white tailed deer mulemulcmuic deer odocoileus virginianusvirginiadusvirgin ianuslanusiaDus odocoileus hemionushernionuslonus interspecific hybridization yoningwyoming14

though successful matings of captive mule observed and photographed three female deer odocoileus hemionus and white tailedtalled hybrid deer west of labarge wyoming in the deer 0 virginianusvirginianus have frequently been doc- green river basin the hybrids were always umented cowan 1962 whitehead 1972 day associated with female mule deer and fed with 1980 wishart 1980 interspecific hybridization the mule deer in sagebrush artemisa sppapp hab- in most natural populations appears to be rare itats the hybrids were often seen within a rel- kramer 1973 reported only 10 hybrids out of atively short distance 05 km of willow salix over 17000 deer killed in nebraska 2 out of983 sppapp communities and hayfields along labarge deer from kansas and only 6 out of several creek but we never observed the hybrids thousand observations in alberta in 34 years of keying on riparian areas as whitetailswhitetails commonly fieldwork in arizona knipe 1977 observed do in the andaridarndannd west wood et al 1989 instead only 8 definite hybrids the hybrids wintered in open sagebrush with the in recent years protein electrophoresis of mule deer where there was little hiding or ther- serum albumin and restrictive endonucleaseendonuclease mal cover even though temperatures of 45 C analysis of mitochondrial deoxyribonucleic acid or lower are common in this part of wyoming have been used to characterize gene flow during the winter and early spring of 1991 between mule and white tailed deer popula- 92 we made additional observations and photo- tions mcclymont et al 1982 based on protein graphs of hybrid deer in the green river basin electrophoresis of 201 deer from 31 localities on two separate occasions we saw a male hybrid mainly in the southwestern states derr 1991 8 kmkin south of big piney wyoming in an alfalfa found little evidence of nuclear gene introgres- medicago sativadativa fieldadthfield with approximately 100 sion between the two deer species cronin et al mule deer of both sexes we also made numer- 1988 reported that mitochondrial DNA and ous observations of hybrids along the section of serum albumin appeared to be distinct between labarge creek where we observed hybrids the mule deer and white tailed deer throughout previous year but in 1991 92 we saw more montana suggesting that interspecific gene hybrids including at least two males four flow was very low this was in contrast to data females and three fawns the three hybrid from texas that showed a 565.6sg hybridization fawns appeared to follow a single mule deer doe rate for 319 deer examined carr et al 1986 and may have been triplets these deer were stubbiefieldstubblefieldstubble field et al 1986 and alberta where usually observed with mule deer and occupied hybridization reportedly is increasing lingle primarily nonripariannonriparian areas as the hybrids had 1989 the previous year though whitetail mule deer hybrids have based on published characteristics and mea- been observed in eastern wyoming oceanak surementssurements cowan 1962 oceanak 1978 day 1978 they have not been previously reported 1980 wishart 1980 the deer that we observed from western wyoming on several occasions appeared to be first generation hybrids the during the winter and spring of 1990 91 we length of the ridge on their metatarsal glands

I1 dpartapartdcpaildhiitofrislieiiesaiklwilcllitetoff i d wildlife utah state university loginlogan utah 84322 2livsentaddresspic sent ultliess institute of politicalofpofifial fiecononoinyy utahut illtiltii state university logan utah 84322 piescntacklnss boxB 26 lalabaigelabaggebarp wyowyoiningininging 83123

290 1992 NOTES 291 was intermediate between typical whitewhitetailstails and 60 100 km to the west some reside yearyearlonglong in typical mule deer and the color of the metatar- riparian areas on labarge creek and the green sal tuft was primarily white their tails appeared river moreover by the november breeding to be slightly longer than normal whitetail tails season thousands of migrating mule deer have and were brown merging to black on the dorsal already returned to their lower elevation side and pure white on the underside when wintering areas and then commonly cross the frightened the hybrids used a bounding gait green river to winter in the breaks to the east with or without tail flagging typical ofwhitewhitetailsofwhitetailstailstalis so large numbers of mule deer occupy typical As reported by lingle 1989 the hybrids did whitetail riparian habitats during the rut with not appear to stott but used locomotion patterns the marked difference in their respective popu- intermediate between mule and white tailed lations it may be difficult for white tailed deer deer on all occasions female hybrids were dom- to find appropriate mates during the breeding inated by female mule deer they associateda4thassociated with season this may lead to a high hybridization and were frequently displaced from feeding rate relative to the whitetail population as sites by mule deer appears to be the case in western washington kramer 1973298 postulated that hybrid- where a remnant population of columbian ization between mule and white tailed deer may white tailed deer 0 v leucurusleucurus is surrounded be more frequent where whitetailswhitetails occur in very by a much larger population ofblack tailed deer small numbers this may be true in western 0 h columbianuscolumbianus and where 18 of the wyoming prior to european settlement white whitewhitetailstails tested possessed blacktail alleles at tails were apparently distributed throughout two of three diagnostic loci gavin and may wyoming but unrestricted yearyearlonglong meat 1988 hunting eliminated them from most of western wyoming by the turn of the century acknowledgments

whitetailsWhite tails have been in the process of either reoe cupyingcupping formerly occupied areas in western wyo- we thank V geist and an anonymous ming or rebuilding severely depressed populations reviewer for helpful comments for at least 30 years harry harju wyoming game and fish department personal communication 1991 literature CITED based on hunter surveys conducted through the CACARRR 11 S M S W BALLINGERBALLINGFR J N deliderrDFRR11 L H mail or over the telephone by the wyoming blankenship andland J W BICKIIAM 1986 mitochon- game and fish department 85 whitetailswhite tails were drial DNA analysis ofhybridization between sympatric white tailed deer and mule deer in west texas pro- killed in all of western wyoming in 1974 while ceedingsce ofthe national academy of science 83 9576 159 were killed in 1989 harju 1991 personal 9580 communication since few of these deer were CRONIN M A E RVYSER VYSFvyse and D G CAMERONCAMFRON 1988 genetic between mule deer and white checked by trained observers there is no way of relationships tailed deer in montana journal of wildlife manage- knowing how many deer reported by hunters as ment 52 320 328 whitewhitetailstails were actually hybrids COWAN imetI1 met 1962 hybridization between the blacktailblack tail in contrast the wyoming range mule deer deeranddebranddeer and the whitetailwhite tail deer journal of mammalogy 43 herd that between and 539 541 winters big piney DAY G I1 1980 characteristics and measurements of cap- fontenelle reservoir including labarge tive hybrid deer in arizona southwestensoutbwestenSouthsouthwesternwesten naturalist 25 creek numbered approximately 20000 aniani- 434 438 mals after the severe winter of 1983 84 since deimdelmDERR J N 1991 genetic interactions between white tailed and mule deer in the southwestern united then a series ofseven mild winters coupled with states journal of wildlife management 55 225 237 limited doe harvest allowed this herd to increase GAVIN T A and B MAY 1988 taxonomic status and to 55000 in 1990 harju 1991 personal com- genetic puritypunty of columbian white tailed deer an municationmunication in five years of observation we saw endangered subspecies journal of wildlife manage- ment 52 1 10 over 40000 deer in the big piney labarge KNIPE t1977T 1977 the arizona whitetail deer arizona game creek area and all but a few were mule deer and fish department special report 6 108 appp one was a typical male whitetail and the others kramellKRAMEKRAMERii A 1973 interspecific behavior and dispersion of were the hybrids described above two sympatric deer species journal of wildlife man- agement 37 288 300 though most of these mule deer summer in LINGLELIN cleGLE S 1989 limb coordination and body configuration the wyoming and salt river mountain ranges in the fast gaits of white tailed deer mule deer and 292 GREAT BASIN naturalist volume 52

their hybrids adaptive significant and management WHITEHEAD C JJRJ jit 1972 A preliminary report on white implications unpublished mastermasterss thesis university tailed and black tailed deer cross breeding studies in of calgary calgary alberta canada 289 appp tennessee proceedings of the annual conference of MCCLYMONTMcCLYM ONi R A M pentonFENTON and J R tiiTijTHOMPSONOMPSON southeastern association of game and fish commis- 1982 identification of cervid tissues and hybridization sions 25 65 69 by serum albumin journaljoul nal ofwildlife management 46 WISHARTWISHAKT W D 1980 hybrids of white tailed and mule 540 544 deer in alberta journal of mammalogy 61 716 720 01oceanakANA C 1978 two deer in one wyoming wildlife WOOD A KKRR J mackleMACKIEMACKIF and K L HAMLIN 1989 ecol- 423 24 27 ogy of sympatric populations of mule deer and white stubblefield S SRS R J warrenWARIUNWARRFN and B R MURPHY tailed deer in a prairieprame environment montana 1986 hybridization of free ranging white tailed and department of fish wildlife and parks helena 97 appp mule deer in texas journal of wildlifeVilalifeilfeiloe management 5068850 688 690 received 17 april 1991 accepted 17june17 lunejune 1992 information FOR AUTHORS the great basin naturalist welcomes previously VOUCHER SPECIMENS authors are encouraged unpublished manuscripts pertaining to the biologi- to designate properly prepare label and deposit cal natural history ofwestern north america pref- high quality voucher specimens and cultures docu- erence will be given to concise manuscripts ofup to menting their research in an established perma- 1200012 000ooo words nent collection and to cite the repository in publi- SUBMIT manuscripts to james R barnes editor cation great basin naturalist 290 MLBM brigham references IN THE TEXT are cited by author and young university provo utah 84602 A cover date eg martin 19198989 or martin 1989 multiple letter accompanying the manuscript must include citations should be separated by commas and listed phone numbersofnumbernumberssofof the author submitting the in chronological order use et al after name of manuscript it must also provide information de- first author for citations having more than two au- scribing the extent to which data text or illustra- thors tions have been used in other papers or books that acknowledgments under a centered main head- are published in press submitted or soon to be ingincludeing include special publication numbers when ap- submitted elsewhere authors should adhere to propriatepropriate the following guidelines manuscripts not so pre- literature CITED also under a centered main pared may be returned for revision heading lists references alphabetically in the fol- manuscript preparation consult vol 51 no 2 lowing formats of this journal for specific instructions on style and mack G D and L D flake 1980 habitat rela- format these instructions guidelines FOR manu- tiontionshipsships of waterfowl broods on south da- scripts SUBMITTED TO THE GREAT BASIN naturalist kota stock ponds journal of wildlife man- supply greater detail than those presented here agement 4444695695 700 in condensed form the guidelines are printed at sousa W P 1985 disturbance and patch dynam- the back ofthe issue additional copies are available ics on rocky intertidal shores pages 101124101 124 from the editor upon request also check the in S T A pickett and P S white eds the most recent issue of the great basin naturalist ecology of natural disturbance and patch dy- for changes and refer to the CBE style manual namics academicAcsacademicdemie press new york ath5th edition council of biology editors one coulson R N and J A witter 1984 forest illinois center suite 200 111 east wacker entomology ecology and management john goi PHONE drive chicago IL 60601429860601429860601601 4298 USA wiley and sons inc new york 669 appp 312 6160800616 0800 FAXVAXpax 312 6160226616 0226 24 TABLES are double spaced on separate sheets and TYPE AND DOUBLE SPACE all materials including designed to fit the width of either a single column literature cited table headings and figure legends or a page use lowercase letters to indicate foot- words at hand avoid hyphenated righthandright margins notes underline words to be printed in aliesailesititalics use stan- photocopies OF FIGURES are submitted initially dard bond 22 X 28 cm leaving cm margins on 25 with the editors may suggest changes altsidesallaltailaliait sides manuscript lettering onon figurefiguress should be large enough to SUBMIT 3 COPIES ofthe manuscriptandmanuscript indandand the originoriginal withstand reduction to one or two column width on a 5255.25 or 35 inch disk utilizing wordlewordpeWordwordperfectPerfectrfectact orlorioriginalsfinalsginals must be no larger than 22 X 28 cm 424.2 or above number all pages and assemble eachechpeheach NOTES would copy separately title page abstract and key words if the manuscript be more appro- as a short follow text acknowledgments literature cited appen- priate communication or note the above do dices tables figure legends figures instructions but not include an abstract A CHARGE of 45 per page is made for articles TITLE PAGE includes an informative title no longer published than 15 words names and addresses of authors a the rate for subscribers will be 40 per running head of fewer than 40 letters and spaces page however manuscripts with complex tables footnotes to indicate change of address and author andor numerous half tones will be assessed an I1 C additional charge reprints may be purchased at to whom correspondence should be addressed dif other than the first author the time of publication an order form is sent with the proofs ABSTRACT states the purpose methods results and conclusions oftheodtheof the research it is followed by FINAL CHECKCHECIC 6 12 key words listed in order of decreasing im- cover letter explains any duplication of infor- portance to be used for indexing mation and provides phone numbers tefftefateafTEXT has centered main headings printed in all 3 copies of the manuscript and WordwordperfectPerfect capital letters second level headingheadingsareheadingssareare centered disk in upper and lowercase letters third level head conformity with instructions ings begin paragraphs photocopies of illustrations ISSN 001736140017 3614 GREAT BASIN naturalistnaturalistvol volvoi5252.52 no 3.3 september 1992

CONTENTS articles plant adaptation in the great basin and colorado plateau jonathan PE comstock and james R ehleringer 195 life history abundance and distribution of moabamoapa dace moabamoapa coricorlcoriacorlacoriaceacoridceacoridceaacea G gary scoppettone howard L burge and peter L tuttle 216 condition models for wintering northern pintails in the southern high plains loren M smith douglas G sheeley and david B wester 226 evaluation of road track surveys for cougars felis concolor walter D van sickle and frederick CG lindzey 232 leaf area ratios for selected rangeland plant species mark A weltz wilbert H blackburn and J roger simanton 237 ecology and management of medusaheadmedusahead taeniataeniatherumtheniatherumTaeniaTheniatherum caput medusaemedusan sspasp esperumasperum sisimkmajmkj melderis james A young 245 roost sites used by sandhill crane staging along the platte river nebraska bradley S norling stanley H anderson and wayne A hubert 253 post pleistocene dispersal in hethe mexican vole microtus mexicanusmexic anus an exam- ple of an apparent trend in the distribution of southwestern mammals russell davis and J R callahan 262 can townsend s ground squirrels survisesurvive on a dietofdierofdiet of exotic annuals eric yensen and dana L quinney 269 notes avifauna of central tule valley western bonneville basin peter hovingh 278 wildfire and soil organic carbon in sagebrush bunchbunchgrassgrass vegetation steven A acker 284 structure of a white tailed prairie dog burrow lynn A cooke and steven R swiecki 288 hybrids of white tailed and mule deer in western wyoming charles E eayway and edward boe 290