ECOGRAPHY 27: 207- 217, 2004

Primary productivity and species richness: relationships among functional guilds, residency groups and vagility classes at multiple spatial scales

S.-A. Bailey, M. C. Horner-Devine, G. Luck , L. A. Moore, K. M . Carney, S. Anderson, C. Betrus and E. Fleishman

Ba iley. S.·A .. Horner-Devine. M. C. . Luck, G., Moo re, L. A. , Carney, K. M., Anderson . S .. Betrus. C. and Fleishman. E. 2004. Primary productivity and species richness: rela tionships among func tional guilds. residency groups and vagility classes at mu lt iple spat ial sca les. - Ecography 27 :207- 217.

One of the major de terminants of species richness is the amount of energy available, often measured as primary prod uct ivity. Heterogeneity of environmental variables has <dso been fo und to infl uence species richness. Predicting species distributions across landscapes a .1d iden tifying areas that have high species richness, or vulnerable groups of species. is usefu l for land management. Remotely sensed data may help identify such areas. with the Norma li zed Difference Vegetation Index (NOV!) providing an estimate of prima ry product ivity. We examined the re lationship between maximum productivity (NOV! ). heterogenei ty of prod uct ivity, and species richness of birds and butterflies at multiple spa tial scales. We al so explored relationships between productivity. functional guilds and residency groups of birds, and vagili ty classes of butterflies. Positive linear relationships between maximum NOV! and number of functional guilds of birds were found at two sp<~t i a l scales. We also found positive linear relationships between maximum N DVI and species ri chness of neotro pical migrant birds at two scales. Heterogeneity of NDVI. by contras t, was negatively associated with number of funct ional gui lds of birds and species ri chness of resident birds. Maximum NOV! was associated with species ri chness of all butterflies and of the most vagile butterflies. No associat ion was found be tween heterogeneity of NOV! and species richness of but te rfl ies. In the Grea t Basi n. where high greenness and availability of water correspond to areas of high species richness and maximum NDVI, our results suggest that NOV I can provide a rel iab le basis for stratifying surveys of biodiversity, by highlighting areas of po tent ially high biod iversity across large areas. Measures of heterogeneity of NOV I appear to be less useful in explaining species richness.

S. -A. Boilq ( sullic.bai/ey@fim' .\'lly.gsi.gov.uk ), Foreslry Commission, 231 Corstorphine Rd. Edinburg/; , UK. EH /2 7AT - M. C Hom a -Derine. L. A. Moore, S. Anderson and E. Fleislonan. Dep1 ofBio/ogiml Sciences~ 371 Serra !Vla/1. S tanford Univ., Stanford, CA 94305-5020. USA. - G. Luck. The JohnsTone Cm rre. School of Environmental and fn jimnatio11 Sciences. Charles Srurr Un it• .. Albury. NSW 2640, Australia. - K. M. Carney. Dept of Gcolo[;im ! a/1(1 Enoironmenw/ Scictrces. S tw?ford Unit•., Stanford, CA 94305. US , I. - C Betrus. Dept o{Zoology. Mia111i Unir .. Oxford, OH 45056, USA .

The amount of energy ava ilable in a system (often measured as primary prod uctivity) is tho ught to be one of the major determinan ts of species diversi ty. especially species richn ess (Currie 199 1. Ro:;enzweig

1995, Fraser and C urri e 1996, Hawkins and Porter 2003 ). Accordingly. the relationship between energy and diversity has received considerable attention (Latham a nd Rick lefs 1993. Abrams 1995, Leibold

Accepted 23 December 2003

Copyright © ECOGRA PHY 2004 ISSN 0906-7590

ECOGRAPHY 27:2 (2004) 207

1999, Waide et al. 1999, Lennon et a l. 2000, Mittlebach et al. 2001). The predom inan t pmductivity- richness relationship is unimodal (Rosenzweig and Abramsky 1993, Tilman and Paca la 1993, Abrams 1995, Le ibold 1999, Waide et al. 1999, Mittelbach et al. 200 1), but other relationships have a lso been o bserved (M ittelbach et al. 2001). Variability in the functional fo rm of the relationship between prod uctivity an d species richness is due to a number of f~tctors, including taxonomy (Horner-Devine et a l. 2003), community assemb ly se­quence (Fukami and Morin 2003) and both the spatial scale (e.g. local vs regional) and ecological scale (e.g. within vs among commun it ies) of the study (Waide et al. 1999, Grosset al. 2000, Willis and Whittaker 2002). For example, Chase and Leibold (2002) fo und that bo th producers and animal s exhibited scale dependent pro­ductivity-species richness patterns in pond s. At the local scale, both groups ex hi bited a hump-shaped relatio nship between productivity and species richness. wh ile at the regional scale, this relationship was positively linear. In addition, combining da ta for all species wi rhin a taxonomic group may mask pattern s at finer taxon omic levels. Thus, a critical iss ue that has largely been ignored is variability in the relati onship between producti vity and species richness th at may occur among func tional guilds or other eco logically-derived groups within a given taxon (but see Haddad et a l. 2000 and Horner­Devine et al. 2003).

Heterogeneity of environmental variables also can be a critical factor in determini ng the number of species in an area (Kolasa et al. 199 1, Rosenzweig 1995. Kerr a:·1 d Packer 1997, Hawk ins and Porter 2003) . Environ mental heterogeneity is posit ively co rrela tea with species rich­ness for a number of taxonomic groups. ac ross lr ulti plc spatial scales (Atauri and de Luci o 200 I. va n Rcnsberg et al. 2002). The relat ionship between heterogeneit y of productivity and species richness remains virt ually un­explored, yet may be si mila rly correlated. For example, Kerr et a!. (200 I) found that ITmotely-sensed hetero-

. geneity data helped to explain species richness or butterflies richness above and beyond the infl uence o f available energy.

Primary producti vity can be meas ured direct ly (e.g .. taking biomass sa mples in the fi eld ) or indi rect ly via extrapolation of val ues from fi eld reference si tes. 1 n addition, primary productivity can be est imated from remotely sensed data (Tucker et a!. 1985) Recent st ud ies have demonstrated that the remotely sensed or rnali zed Difference Vegetation Index (NDVI ) expla ined va- iabil ­ity in butterfly and bird species richness in select mountain ranges in the Great Basin in western No rth America (Seto et a l. in press). Thi :- suggests that remotely sensed data associated with productivity may be useful in predicting species rich ness over large area ~.

Predictions based on remotely sensed data could be especially useful tools in large ma naged landscapes

208

where it is not feasible to conduct thorough ground surveys and monitori ng programs.

Here, we investi gate the relationship between produc­tivity (es ti mated using N DVI), species richness of birds in different f'unctional guilds and residency groups, and species richness of butte rflies in different vagility classes in th e Great Basin. T hese analyses are intended to elucidate whether different subgroups of birds and butterflies respond simi larly to changes in productivity. We al so assessed relat ionships between the heterogeneity o f prod uctivity and species richness of taxonomic subgro ups of birds and butterflies to explore the potential of heterogeneity in production as a predictor o r species diversity.

T he extent of human land use, which typically lowers native biological d iversity, is continuing to increase in many geographic regions. As managers face increasing hum an lan d-use demands and reduction in funds for assessing patterns of species diversity, remotely-sensed measures of productivity may play a useful role in predicting patte rns of species richness. Moreover, in­fo rmation on the relationship between environmental heterogeneity and species richness, and how particular fu nctional and ecological groups respond to variable levels of productivity. would be of value to managers hoping to infl uence the species richness or abundance of particular taxonomic groups (e.g. neotropica l migrant birds)

Methods

Study system

The G reat Basin includes > 425 000 km 2 of western North America from the Sierra Nevada in the west to the Wasatch Ran ge in the east (Grayson 1993). The ecosystem is cha racteri zed by > 200 north-south or­iented mo untain ranges incised with canyons across their eastern and western slopes. The region is a cold desert; most precipi ta tion fall s as snow in the winter months, and to tal annua l precipitation is < 250 mm (Ricketts et al. 1999). With in the Great Basin there is an elevational gradient in vegetation composition. As elevation in­creases, the dominant vegetation generally changes from sagebush Artemisia tridentata spp. to pinon-juniper woodland Pinus monophy//a, Juniperus osteosperma to low brush (Tueller and Ec kert 1987). Canyons with permanen t o r ephemeral streams often contain willow Salix spp., rose Rosa JFoodsii , nettle Urtica dioica , and various grasses and forbs (Fleishman et al. 1997).

Water is a limiting factor in the Great Basin, with the highest producti vity centered near permanent and ephemeral st reams. Due to the aridity of the inter­moun tHin valleys (tens of kilometres wide) separating the mo unt11in t·a nges, the mountain ranges function as hab ita t islands for many species of animals (Fleishman

ECOGRAPHY 27:2 (2004)

and Murphy 1999, F leishman et a l. 2002). T he study landscape consists of three adjacent mountain ran ges in the central Great Basin: the Shoshone Mou ntains (SH ). Toiyabe Range (TY ) and Toqui ma Range (TQ). These ranges share regional climate. biogeographical past. ancestral biota and hum an lan d use hi stories (G n1yson 1993). We partitioned the la ndscape in to th ree nested spatial levels: mo un ta in ra nges, canyons wi thin moun ­tain ranges and segments of canyons (Fig. I). Canyon segments were delimi ted by elevati on; each segtY.ent was 100m wide and long enough to span a 100 rn change in elevation (Fleishman et a l. 1998. 200 I ) . Here we exam ine productivity-divers ity relationships fo r birds and butter­flies at the sampli ng gra in of canyon a nd ,canyon segment and the spatia l extent of an individua l mcun ta in range and landscape (all mountain ran ges com bined) (Gross et al. 2000).

Species data

Surveys of breeding birds were co ndu cted in five canyons in the Shoshone Mountains (25 canyon segments) and Toiyabe Range (31 segmen ts), a nd six canyo·.1s in the Toquima Range (28 segments). fo llowing standard methods (Bibby et a l. 2000, Betr us 2002) . Bird s were sampled three times at multiple points in each canyo n

Fig. I. The Great Basin ( oU!cr black li ne) . st udy Jn .1dscape (inner black lines), canyons (SH . Shoshone Mountains: TY. Toiyabe Range. TQ. T,1quima Ran ge). and ca;·yon sc :~net. c s (white lines on insert).

ECOGRAPHY 27;2 (2004 i

segment fo r fi ve minutes each during the breeding season (May a nd June) in 2001 using fixed-radius point counts (Betrus 2002) . Three surveys are co nsidered sufficient to dete rmine wh ich species of birds are present at point count locations (Siegel et a!. 200 I) . In addition, point coums have been shown to be an effective method of sampling birds in riparia n areas in the Great Basin (Dobkin a nd Rich 1998). Butterfly data were collected in 39 canyon segments in the Shoshone Mo untains, I 02 in the Toi yabe Range and 54 in the Toquima Range between 1994 and 200 I. fo llowing standard methods (Shapiro 1975. Hard ing et al. 1995. Fleishman et al. 1998). Surveys were cond ucted every two weeks through the fli ght season (app rox imately late May through 1\ugust).

NDVI and primary productivity

Primary prod uctio n can be inferred from remotely sensed data by using NDV I, an estimate of 'greenness' (Tucker et a l. 1985) NDVI is derived from the near in fra red (N I R) and visible red bands of a satellite image. It is computed by di vid in g the difference of the two bands by their sum (Wil kie and Finn 1996):

NDVl == (N IR - Red)/( NIR + Red).

The red a:1d N IR light reflected from pla nts is a function o f the photosyn thf tica ll y active compounds present, wh ich is related to overa ll plant biomass (Tucker et al. 1985). N DVI measures were generated from a single .:loud-Crcc geo-corrected Landsat Thematic Mapper image (WRS 4 1 -- 33) . T he image was acquired in June 2000 to co incide with the peak growing season and the most active period for b reeding birds and resident butterfl ies in the Great Basin. We calculated minimum, max im um. mean, ra nge a nd standa rd deviation of NDVI fo r each canyon segment and canyon based on va lues fo r a ll pixels (30 x 30 m resolution) within the ca nyon segment and canyon .

Bird residet<cy groups and NDVI

We classified bird species into three residency groups: residents ( 12 species). short-d istance migrants (28 spe­cies) and neotropical m igran ts (26 species). Residents remain in the Great Basin year-round. short-distance migrants mostly winter north of the tropics, and neot ropica l migrants win ter in the new world tropics (Go ugh ct al. 1998). A ll species breed in the Great Basin . We explored relation ships between species richness of each residency group a nd al l measures of NDVI. The st rongest relat ionships observed were for maximum N DVl. a nd these a re the o nly results presented . D ata fo .. 84 canycn segments were analyzed using linear and q uad ratic regressions at the canyon segment and canyon

209

grains across the landscape, and at the canyon s,' gmen t grain within each mo un ta in ra nge. We determ ined whether the minim um o r maxim urr value cf a curvi­linear relationship occu rred with in the bou nds o l· (lUr observed data using Mitchell-Oicls and Shaw 's ( 198 7) test. For all analyses, the distribution of variables was assessed prior to analysis and sca tter plots of residuals versus predicted va lues were exa mined for violati o11 o f regression assumptions. When necessary, data were log transformed.

Bird functional guilds and NDVl

Because NDVI is positively correlated wi; h' ~pcc , es richness of bird s in the study area (Seto eta!. in pt :ss ). we investigated whether the re was a similar rclati Jnsh ip between NDVI and num ber o f functiona l guilc! s. T he term " function a l guild " is taken from G itay and No ble (1997), who assignee! species to the same guild if tr e species used the same resource(s) in the same way. We assigned bird species to the same runctional gu lld if : he: species had the same food preference (e. g .. insects. fr c~i t s

or both), foraging substrate (e.g .. tree canopy, bark or ground) and foraging maneuver (e.g., glean in g or haw k­ing), using published data (Ehrlich et al 1988 . Alsop 2001).

We hypothesized that sites wit h high NDVl woLld have greater productwit y and vegcta :. i1·c complexi ty and thus support mo re gu ild s, asou ming that complex ity ccnd number of fun ctional guilds is related. Hen,;e, we assumed th at guild richness provides more info rmation on species diversity th an species rich ness a lone. w~ ;;lso determined if an increase in the num ber o f gui lds was primarily a res ult of the addi tion of gui ld s represen ted by only one species, and if the pro portion o r species occurring in single-species fun cti onal gui lds va ried with NDVI. The number of species in each guild provides c~n indication of the level of apparent ecological redund ancy (Walker 1992, 1995, Naeem 1998). T he fu nct ioning of

' ecosystems character ized by a high number of !'[JCc .cs

per guild (greater redundancy) may b<~ less suscept ible to disruption because the probabil ity <Jf los ing a n emire guild via random extinction oi species is lower ~haP if. guilds contain fewer species. We e.xplored rela t io nships between richness of g ui lds and all meas ures or 1'1 DV I. The strongest rel at ionships observed were fo r maximum NDVI, and these a re the o nly rcsul:s presen ted.

Butterfly vagility groups and NDVl

We grouped butte rny species into three vagility ::asses: low (an individual is li kely to move on the orlkr <>r tens of meters in its lifeti me; l1 '= 23 species), ink rmed ic.tC (a n individual may move hund reds or meter-;; : n = :!9 species) and high (an ind ividu al may move m01e tl·.a!l a

210

ki lometer; n = 34 species) (F leishman et a l. 1997). Low vagil ity btttt ·~rtl i es are unl ikely to move beyond one grid ce ll (30 x 30 m) at the resolution of data in this study. Inter media te and high vagili ty classes a re capable of movement between cel ls. We hypothesized that species with higher vagili ty wo uld have a stronger relationship with NDV I, assuming these species can and will move to prod uctive areas. We explored relatio nships between species richness of each vagility group a nd all measures of NDVI. The st rongest relationships were again re­co rded lo r maximum NDVI a nd therefore these are the only res uh s presented.

Heterogeneity of NDVJ and species richness of birds a nd butterflies

Calculation o/ heterogendly measures

Usin g ArcView (Anon. 1996a), we classified the NDVI image into five equally sized bins according to pixel NOV! va lues. We measu red heterogeneity of the NDV! classes us ing Sim pso n's dive rsity index (D'; Krebs 1999). Bins were treated as 'species' and pixels were treated as ' ind ivid uals.' To contro l for the effects of elevation, we selected a subset of canyon segmen ts that fell within as smal l a range of elevation as possible (range of starting ekvation = ::7 1 m). T hese analyses were thus done at the ca nyon segn 1ent grain a t the extent of the entire land­scape. Bird da ta were available for only half of the selected canyo n segments (birds, n = 15 segments; butterfl ies. n = 3 1 segments) .

For the subset of canyon segments for which we ana lyzed bird species data. we used multiple linear regressions with fo rwa rd stepwise selection of variables [stepping cr iteria used probability of F = 0.25 for entry and 0.1 lo1· removal (JMP 4.0; Anon. 1996b)] to assess the rel ct tionship betwee n different meas ures of richness o f bi rd~; and N DVI. O ur meas ures of richness were total species r ich cess. species richness within residency groups, and number of fo raging gu ilds. The two independent va riables us~d in the multip le linea r regressions were In dxin-; um N DV I and heterogeneity of NDVL Prior to regression analyses. we exam ined the two NDVI mea­sures Jo r colinearity [linear regression (JMP 4.0)]. We found weak negative cor n:lation between maximum "~DVJ and heterogeneity of NDYI (R2 = 0.237 p = 0 .056 ) for the subset of canyon segments included in the bird analyses. In co ntrast, the relationship between maximum N DVI and hete rogeneity of NDVI for the subse t or canyon segments used in the butterfly analyses was no1 stat ist ically sign ifica nt. Since these two measures o r NDY l co uld be considered independent for butter­llies. we co r d ucted separate linea r regressions of mea­su res o f bu ttcrny species richness using maximum NDVI aEd hete rogeneity of NDV I. For butterflies, total species rich11cc;s and species richness within vagility classes

ECOGR APHY 27:2 (2004)

were included as the dependent variable in di !Tercnt models.

Simpson's di versity was a •·csine transformed to no r­ma lize data fo r all a nalyses.

Results

Bird residency groups and NOVI

At the canyon gra in across the landscape the relationsh ip between species richness o r birds and max imum NOV ! was linear for neo tropical m igrants (R2 =0.7 l6. p < 0.001) and unimoda l for short-distance r,nigran ts (Mitchell-Olds and Shaw test: Prod'' < Prod max, t= - 1.611, p =O. l3l ; Procl* < f' rodmin, t=3 .567. p = 0.003) (Table I. Fig. 2).

At the canyon segment grai n across the l<1ndscape. species richness ol' neotmpica l migra nt birds increased linearly with increasing max imum N OVl (R 2 = 1}426. p < 0.001 ) (Table l ). At the can yon segment grain across each mounta in range, there was a positive li near relationship between ' pecies richness of bi rds a nd maximum N OV! fo r neot ropica l migrants in all three mounta in ranges. (S H R 2 = 0. 172, p < 0.0 i: T Y R 2 = 0.411 , p < 0.01: TQ R2 = 0.240, p < 0.01 ) for ,;hart­distance migrants in the Toiyabe Range (R 2

: .• D.269, p < 0.01), a nd for residents in the Toqu ima Ra nge (R 2 = 0.362, p < 0. 0 I ) (Table I) .

Bird functional guilds and NOVI

Bird species were classi fied into I 8 fun cti onal gui lds (Appendix I and 2). N umber of funct ion a l guil s increased significa ntly as maxim um NOV! increased at the canyon gra in across th e landscape (R2=0.53l. p < 0.01), a t the canyon segmen t grain across the lan d­scape (R 2 = 0.225. p < 0 .00 I ), a nd at the canyon s.:gmem

, grain in the To iyabe and Toquima ranges (TY R 2 = 0.322, p < 0.00 1; TQ R 2 = 0 395, p < 0.00 1) (Table 2). The num ber o i' gu ild s represented by only one species increased slightly as maximum NOV ! increased at the canyon grain across the land scape (R 2

= 0.208, p = 0.07), bu t the rela tionshi p between max imum NOV!

(f)

.2:! ( ) (!) c.. (f)

"E aJ

6,..

5 4

3

2

1

•

•

• •••

Residents 0 ..__,____......__ ... _ _... _ _.._ ..... _ _.

12

':r ~ 6 r :.<· 4L ~ Short-di stance Migrants

'16c--14 ~ '12 ~ 10

3 ,!) •

42 I • • r- Neotropical Migrants

80 90 100 110 120 130 140 150

lVI ax im urn N DVI f ig. 2. The relationshi p between species richness of birds and maxi mum NDYl a t the c;myon grai n across the landscape, fo r residen ts. short-distance migra nts. and neo tropical migrants. Solid lines deno te signi ficant relationships (see Table I) .

and the proportion of species in single-species guilds was not sta tistically significant.

Species riel·. ness of birds and heterogeneity of NDVI

Both maximum NOV! and heterogeneity of NOVI predicted tota l species rich ness of birds in multiple linea ;· regressions (R 1 = 0. 75, p < 0.05) (Table 3). Het­erogeneity of N OVJ expla ined ca IS'Yc, more variance in

Table I. Relationships between max imum NDV I and speci ~s richness of birds at ti1e c<:nyon segment and canyon gra ins across the landscape, and at the canyo n segment gra in across each mounta in range. NTM . neo tropical migrant s; SDM, sho rt-distance migrants; R ES, residen ts. SH. Shos ho 1e Mo untains: T Y. Toiyabe Range: TQ. Toq uima Range. Values are adjusted R1

; significance levels are *p < 0.05, **p < 0.01. ***p < 0.00 I. The rela tiono:h ip fo r SDM at the canyo n gra in across the landscape was quad ratic (ty = - 0.00473x1 + 1.1 82x - 63.184): a ll othe r relationshi1Js we re linea r nnd pos il ive. Colu mn head ings show the gra in; the extent is shown in pa ren theses.

Residency Group

Canyon (Landscape)

------~·

NTM SDM RES

0.7 16*** 0.629*«1 0.000

ECOGRAPH Y 27:2 (2004)

Canyon Segmen t (Landscape)

0.426* ** 0.035* 0.000

Canyon Segment (S H)

Canyon Segment (TY)

-- -.~~-~-~- .. ·- - --0.172* 0 000 0 000

0.411 *** 0.269'''* 0.036

Canyon Segment (TQ)

0.240** 0.040 0.362**

2 11

Table 2. Relati onships bct\\cen maximum NDVI and number of fun ctional guilds. number of single species functional guilds and the propo rtion of speci~s in single specie;, fun ct ional guilds for bi !·ds. SH, Shoshone Mountains: T Y. Toiyabe Range, TQ, Toquima Range. Values are R 2 significan~·e levels M C *p = 0.07. **p < 0.0 I. ***p < 0.00 I . Column headings show the grain ; the extent is shown in paren theses.

Canyon (Landscape)

Canyon ~!egmcnt ( 1_ar.dscape)

Canyo n Segment ( 51-!)

Canyo n Segment (TY)

Canyon Segment (TQ)

No. of guilds No. of single- species gu il ds Proportion of species in single

species guild

0.5] I** 0208* 1).104

0 . 225~ :~ *

0.()] 1 () 017

0.040 0.002 0.()7 1

0.322*** 0.0:18 0.00 1

0.395*** 0.015 0.011

---- - ----------------------------------

species richn ess of birds than did maximu m N O V! alone. The number of fo raging gu ilds decreased sign ifi­cantly as heterogenei ty o f N OVI increased (R 2 == 0.60, p < 0.01). When the bird spec ies we re categoh ze6 by residency, there was a negative linea r rt; la tionship between species richness o f resident bi rds and he tero­geneity of N OVI (R 2 = 0.64 , p < 0.00 I), but maxim um NOV! was the only predicto r (po, itive) o f SJecie; richness of short-distance mi grants and neot ropica l migrants (SOM R 2 = 038, p <fl .OO 1: NTM R 2 =' 0 8 !. p <0.001) (Table 3) . At the ca nyon segmen t level the relationship between max imum N OV! and bird species richness was stronger than at the ca nyon leve l.

Butterflies, butterfly vagility and NDV[

Significant positi ve rela ti onshi ps were fo und buwecn maximum NDVI and tola l species richness of butte rfies (R2 =0.196, p = 0.0086) and between max imum N OV! and species richness of the most vagilc class of but terf:ies (R2 = 0.3409, p = 0.0003 ). A weaker po:; iti ve rela ti c nship was found between max imum N OV! and species rich­ness of butternies with inte rmed iate vagi li ty (R2 = 0.! 0!. p = 0.0583); the rela tionshi p between maximum NOV I and species richness of the least vagik buttcdlico; was not statistically sign ificant. We did not find a significant relationship between heterogeneity o f N OV! and either total species richness of butternies or species rich ness o f any vagility class.

Discussion

We observed a general trend of increasi ng necic3 richness of birds with in creasing prod uctivity as mea-

sured by NOVl. Fo r exa mple, we fo und positive linear rel ati onshi ps between maxi m um NOV! and number of f"u nct iona l g·.t ilds of birds a t the canyon grain across the landscape, and at the canyo n segment grain in the Toiyabe and Toq uima ranges. We a lso observed positive relati onsh ips between maximum NOV! and species richness of neo tropical migrant bi rds at the canyon segment gra in across both extents, and at the canyon grain across the land scape. Maximum NOV! and species richness o f birds was stronger at the larger grain (canyon vs canyon segment) , possibly because the canyon grain integrates environmental values over a larger area. Becau se birds may move between canyon segments, the canyon segment gra in may be too small to detect ~ ~ relat ionship bet\.veen N OV! and species richness of birds.

Th ere was a sign ificant relat ionship between max­im um ND VJ and both species richness of all butterflies and speci ~s rich ness o f the most vagile class of butterflies at the canyon segment gra in at the extent of the entire landscape. No associat ion was foLmd between hetero­geneity of N OVI and to tal species richness of butterflies, or between heterogeneity o f N OV! and species richness o f bu tterfl ies in any vagil ity gro up.

Responses of subgroups of birds and butterflies to produdivit:·

Bird residem y groups

Species richness of res ident birds was not co rrelated with max imum NOVI except in the Toquima Range. How­ever, v;e cannot conclude that resident birds are un­afi'ected by p ri mary prod uctivity; across North America,

Table 3. Relationships between heterogeneity of NDVI. •mD, irnum NDV : <•nd measures of species richness of birds. All results are reported at the canyon segmen t ac ros:; t h ~ lat,dscape. ;\JTM . ncotropical migrants; SD M, shor t-d istance migrants; RES, residents. Rows marked with a ' - 'where no statist ically sign ifican t mode. produced. Values are R2

. signifi cance levels are *p < 0.05, **p < 0.01 , ***p <O.OO I.

Total species richness

Number of forag ing gui lds RES SDM NTM

212

N DVI variable

Max imu m NDVI 1-l ctcro>renei tv ol' l' DYI 1-ktetngeneit;, o r l\ DV I 1-lett:rogenci ty or N D\"1 rvLt ximum NDVI Maximum NDV I

Regression coellicien t

-- I 1.693 0.134

- 6.2b9 4.310 0.045 0.102

F ratio

8.06 10.25 5.53

22.97 1.998

35 .9 1

Model R2 (P)

0.75*

0.60** 0.64*** 0.38*** 0.8 1 ***

ECOGRAPHY 27:2 (2004)

species richness of resident bird s was co rrelated with minimum monthly NDVJ (H urlbert and Haskell 2003). Hurlbert and Haskell (2003) also fo und thal. in North America, the species richness of migrant birds was

related to seasonal fluxes in produ ctivity (NDV[) . This suggests that migrants arc able to use thei r mobility to track temporal and spa tial va riati on in prod uct ivity, while the richness or res iden t species (associated wit h minimum monthly NDYl ) is di ctated by the minimu n: productivity o f the env ironment.

It is interestin g th at the shape of th e relationsh ip between maximum NDY I and species rich ne:;s o!· bi rds differed between neotropical and short-d istan ce mi­grants, but it is not clear why specie; rich ness 9r short­distance migrants varied un imodal ly with ma\ imum NDYI.

The strong linear relationship between species richness of neotropical migrant birds and maximu m N DVI suggests either that neotropical migrants have more specific habitat requirements th an short-d istance mi­grants and year-roun d residents, or that ncotropica l migrants rely more strongly on structurally complex vegetation. Similar pattern s have been foun d elsewhere. For example, neotropica l migrant birds in the eastern US are more "sensitive' · w lan dsca pe structu re than otLcr residency gro ups (Fiather and Sauer 199[.), and in British Columbia , neo!ropical migrants are mo re closely tied to riparian areas than residenr species (Wiebe and Martin 1998). In the mountains of the Great Basin, it appears that neotropica l migran ts prefer areas with relatively lush vegetation and avai lable wate~, wh ich occur mostly in th e bottom of canyons.

The relationship between N D\' I and species richness of neotropical migrant s was posi tive at ail grains and extents, sugges tin g tha t N DVI may be a useful to•JI 1o r identifying locations with poten tially high spec i e~. rich­ness of neotropical migrant bi rds. Given i.he recent declines in neot ropi .~a l mig r<.nt populatic.ns across North America (Sauer et al. 2001 ), the abi li ty to qu ick ly identify areas that may ~; uppon spcc i ~s of comervation

' concern could be a va luable fo r la nd rnan:tgers. Our analyses suggest that, tor nee- tro pical migram birds, using NDVJ to est imate species richness may be most successful at coarse scales.

Bird functional groups

Relationships between prod uctivit.y and !'unct iona l Ji · versity have received litllc att ention in productivi ty­species richness studies. Some authors app,:-a •· to c. ss urne that species richness is an appropriate sut-rogare for richness of fun ctional groups (Lawt ::> n et a l. 1998 <1 ncl Tilman 1999), but thi s may not always be the case (see Diaz and Cabido 2001 and references th erein ). Thus, assessments of prod uctivi ty and the number of

ECOG RAPHY 27:2 120041

functiom!l groups in a system may improve our under­standing of the relationships between productivity and biological diversity sens u Jato. In our work, however, we have fo und posi tive relati onshi ps between NDVI and bot h species richness and number of functional guilds; in this system, species richness may indeed be a useful surrogate fo r num ber of fu ncti onal groups.

Sites with high NDVI had greater functional complex­ity, as measured by nu mber of functional guilds of birds, than sites with relat ively low NDVI. The guilds that tended to b<: added as NDVI increased included those th at probe bark tor insects (e.g. woodpeckers) or primar il y fo rage on trees (Appendix I and 2). This sugges ts that there is a positive relationship between NOV! and vegetat ion complex ity or, at the very least, that sites with high NDVI values support more trees than sites with lower values of NDYI. This concurs with the observation that si tes with high NDYI often occur in canyon bolloms and in close proximity to water. Howeve r, the presence of water can vary between years and thus mdy not be in dicative of productivity over many years. Remotely sensed data allows trends over time t.o be exami ned, and areas of consistent high product ive ,.reas to be identified .

Hete rogeneity of NDVI and richness of taxonomic subgroups

Environmental heterogeneity has been shown to have strong positi ve effects on species richness (Rosenzweig 1995). Ata uri and de Lucio (2001 ) found landscape heterogeneity to be a more effecti ve predictor of species richnesi of bi rds and lepidopterans than land cover type. In a revir:w of st udies on productivity and species rich ness, vVa ide et a!. ( 1999) found environmental heterogenei~y to be a confo und ing factor in many studies that revea lec' scale-dependent relationships. However, we fou nd tha t heterogenr: ity of NDYI only explained di,;ersity of some gui lds of bi rds (number of foraging gui lds and residents) and did not explain species richness of buttertl ies. Heterogenei ty of NDYI was negatively associated with num ber of runctional guilds of birds and species richness of resident bird s. Alth ough this may initia lly appear co unter- intui ti ve, it may reflect the fact that areas of unifo rmly hi gh NDYI represent sites with high structun l complex ity, whereas heterogeneous sites incl ude pa tches with li ttle vegetati on.

Parti: ularly fo r butterfli es. NDV I may not be the best meas ure or ~~nv iro n mental heterogeneity (Weibull et al. 2000) . Othe: measures, such as elevation, may be more appropriate (Hawkins and Porter 2003, Mac Nally et al. 2003). Di ffe ren t responses to meas ures of environmental heterogeneity by diiTerent groups of species illustrate the potential fo r mismatch between scales of measurement an d use of the land scape by species (Mazerolle and

213

Villard 1999, A tauri an d de Lucio 200 I). In ou r st udy. a n associa tion between heterogeneity of NDV J and both number of fo raging guilds ol· birds a nd species ri chness o f resident birds may indicate th at some birds use the landscape at a scale coincident with the grain at whic h heterogeneity o f NDV I was measured.

The application of NDVI for productivity and species richness assessment

In ecosystems simi la r to the G reat Basi n, when~ high greenness and wetness va lues correspon d to areas o f h igh species richness and maxi mum NDVI. N D)I l data provide a reliable bas is fo r strat ifying sur\ .:ys o f biodiversity accord ing to produ ctiv ity. In the G reat Basin, vegetatio n greenness and water availab ili ty may be easily observed o n the gro und. T hus. we recognize that in th is system. la nd ma nagers may be able to re ly on their famili arity wi tl1 the region to id entify «.rcas wi th potentially h igh species richness o f neot ropical migran t birds or short-d istance m igra nt b irds.

Whilst fie ld surveys mc.y have the advantage over NDVI at identi fying sites most likely to have high species richness, NDV J has the advan l<tge or high li gh ti ng areas o f potentia lly hi gh species richness across large land­scapes that wo uld req uire substanti a l investments ufti me and money to survey d irectly. In a rid systems. the presence of water can va ry between years and thus no t necessarily be ind icative of the overall produCLivit: ' over time. Remotely sensed data al lows trends to be exam in ed over time, and co nsisten tly high ly producti ve area s to be identified and ma naged appropri ately. NDVl may be most useful in a reas where environmct>ta l heteroge neity obscures di fferences in prod uctivi ty on the grou r,d .

Acknowledgeml!ll/s - This paper res ul ted from a sem:na r on Biocomplexity and Ecoinfonnat ics. sponsored by the Nat ional Center for Ecological Amilysis an d Synthes is (NC EAS). conducted by E. Fleishman at Stanl(ml Uni v. in 2002 M. Mayfield and C. Sekercioglu made vc. lua i> 'e contri btllions to the

· working gro up. We wou ld like to tha nK John Fay f.1r data transmission. Part ia l support for this work was prov ided by the Nevada Biod ive rsity Research and Conse rva tion In itia tive. by the Joint Fire Sciences Prog ram vi<l the Rocky Mo untai n Research Station, Forest Service, U.S. Dept of Agric ul cure, and by Miami Uni v.

References

Abrams, P. A. 1995. Monotonic ;md unimoda l d i•:ersi ty­productivity gradients: what does competition the•) i")' pre­dict?- Ecology 76: 20 19- 2027.

Alsop, F. J. 200 1. Bi rds of North .11 erica. - Sm it!-sonian Handbooks. New York.

Anon. 1996a. Using ArcView GiS. Er.\'ironrnent,d Research lnst. - Redl ands, Cali fornia, L SA.

Anon. 1996b. SAS lnsL - Cary, N. C.. USA. Atauri , J. A. a nd de Lucio. J. V. 2001. The role of h.n.Jscape

structure in species richness dist ributions or birds. ::mphi-·

214

bia ns. :·eptilcs and lepidoptcrans in Mediterranean land­scapes. - Landscape Eco l. 16: 14 7- 159.

Betrus. C. J. 2002. Refin ing the umbrella index complex: an app:ication to bird and butterfl y communit ies in montane canyo ns in the G rea t Basin. - M.S. thes is, Miami Univ. , Oxl'o>d. Ohio.

Bi bby, C. J. et a l. 2000. Bird census techniques. - Academic Pre,s.

Chase, J. M. and Leibold, M. A 2002. Spatial sca le dictates the product ivity-biod ive rsity relat ionship. - Natu re 41 6: 427 -430.

Curr ie. D . . 1. 1991 . Energy and large-scale patterns of animal­and plant -species richness. ·- Am. Nat. 137: 27 - 49.

Diaz. S. anc. Cabido. M. 200 1. Vive Ia di ffe rence: plant func itonal dive rsity matters to ecosystem processes. - Trends Ecol. Evo l. 16: 646- 655 .

Dobkir . D. S. and Rich . A. C. 1998. Comparison of line­trar,sect. spo t-map. a nd point-coun t surveys for birds in riparian areas of the Grea t Basin. - J. Field Ornitho l. 69: 4] 0 - 443

Ehrlich, P. R. , Dobkin, D. S. and Wheye, D . 1988. The Birder's handboo k: a field guide to the natu ral history of North American birds. - Si mon and Schuster, New York.

Flather. C. H. and Sauer. JR. 1996. Using landscape ecology to test hypo theses abou t large-scale abundance patterns in migra tory birds. - Ecology 77: 28- 35 .

Fleishmdn, E. and Murphy. D. D. 1999 . Patterns and processes of nestcdness in a G reat Basin butterfly community. - Oeco logia 11 9: 133 - 139.

Fleis hma n,[, Austin. G. T. and Murphy, D. D. 1997. Natural history ad biogeogra!Jhy of the butterflies of the Toiyabe Range, N~vada (Lepidoptera: Papilionoidea). - Holarct. Lepidopt. 4: I - 18 .

Fle ish:m111 E. , Austin, G. T. and Weiss, A. D. 1998. An empirical tes t of Rapopo rt 's ru le: elevational gradien ts in montane butterily communities. _, Ecology 97: 2482-2493.

Fleishman, E .. Austin. G. T. and Murphy, D. M. 200 1. Biogeography of Great Bas in butterfl ies: revisiting patterns, paradigms, and cli ma te cha nge scena ri os. - Bio i. J. Linn. So ~c . 74: 50 1 - 51 5.

Fleishman. E. et a l. 2002. Nes tcdness ana lysis and conservation pldn ning: the importa nce of place, enviro nment, and life history :tcross ta xonomic groups. - Oecologia 133: 78- 89.

Frase r, R. H. :mel Curr ie, D . .1. 1996. The species richness-energy hypothesis in a system -''here historica l factors are thought to preva il : coral reefs. - Am. Nat. 148 : 138 - 159.

Fukami. T. ;•nd Morin. P. .1. 2003. Productivity-biodiversity rehnionsh.ps depend on the history of community assembly. - N<tt ure •-f24: 423 - 426.

Gitay, 1-1 . and No ble, I. R. 1997. What are functional types and why should we seck them·; - In: Smith, T. M., Shugart, H. H . <:nd Woodwa rd , F. I. (eds) . Plant functional types. Their releva nce to ecosys tem pro perties and global change. Cam­bridge Univ. Press, pp. 3- 19.

Graysm~ . D. K. 1993 . The dese rt 's past: a natu ra l prehistory of the G reat Bas in. - Sm ithsonian lnst. Press, Washington DC.

Gress. K . L. et al. 2000. Patte rns of species density and producti vity at diffe ren t spatial scales in herbaceous plant com inur it ies. - Oikos 89:417 - 427.

Gough. G. A .. Sauer. J. R. and Ili ff, M. 1998. Patuxent bird identifica:ion infoccnte r. ver. 97.1. Patuxent Wildlife Re­search Cen ter, Laurel. Maryland, < ht tp://www.mbr-pwrc. ­usgs.gov/l •·focenter/infocen ter.html > .

H:tddad. N. 11.1 .. Haar:; tad. J. and Tilman, D. 2000. The effects of long- term ni t rogen load ing on grassla nd insect commu­ni ties. - Oecologia 124: n --84.

Harding, P. T. Asher. J. and Yate s. J. T. 1995. Butterfly mon ito ring I - recording the changes. - In: Pullin, A. S. (ed. ) Ecology and conservation of bu tterflies. Chapman and HalL pp. 3-22.

ECOGRAPHY 27:2 (2004)

Hawkins, B. A. and Porter, E. E. 2003. Does herbivore J i,er:;i tv depend on plan t diversi ty? The case of Californ ia buuerfl iei - Am. Nat. 161: 40 - 49.

Horner-Devine, M. C. et a l. 2003. Bacteria l di ve rsity patte rns along a grad ient of primary productivit y. - Ecol. Lett. 6: 613 - 622.

Hurlbert, A. H. and Haskell. .J. P. 200 1 . The effect o!· energy and seasona li ty on avian species richness and community composition. - Am. Nat. 161: 83 --97.

Kerr, J. T. and Packer. L. 1997. 1-l ab it .ll hcterogeneit _y as a determinant of mammal species richness in high-energy regions. Nature 385: 252 254.

Kerr, J. T. , Southwood . T. R. E. and Cihla r. J. 200 1. Remote!\ sensed habitat diversity predict> butterfly species r;chnes-s and community similari ty in Canada . - Proc. Nat! Acad. Sci . U.S.A. 98: I 1365 - 11 370.

Kolasa, 1.. Picke tt. S. T. and Al len .. T. F. H. 1991. ECJ log ical heterogeneity. - Spr'ngcr. 1

Krebs, C. J. 1999. Ecological methoclo logy. 2nd ed. ·· Addison­Wesley.

Latham, R. E. and Rick lefs. R. E. 1993. Globa l patterns of tree species ri chness in moist fo rests: e:1ergy-diversiry th2ory does not accou nt for varia tion in species richness. Oikos 67: 325 - 333.

Lawton, J. H . et al. 19'!8. Biod iversity and ecosyster,, funct ion : gett ing the Ecotron experimen t in its correct cc:1text. - Funct. Ecol. 12: 848- 852.

Leibold, M. A. 1999. Biodiversity and nutrient enrichment ir. pond plankton communities. - Evol. Ecol. Res. I: 73 - 95.

Lennon, J. J. , Greenwood . .1. J. D. and Turner, J. R. G. 2fl00. Bird diversity and environmen t<Ji gradients in Bri tain: a L's t of the species-energy hypo I hesis. - .1. An im . Eco l. 6S• : 58 1 -598.

Mac Nally, R . E. ct ai. 2003 . Modeling bu.t t rfly spe.:ies ricf.t;ess using n1esoscale environmen tal variables: mo<.kl con struc­tion and vali da tion. -- Bioi. Co r. serv. 11 0: 21 - 3!.

Mazerolle, M . .1. and Villard , M. A. 1999. Patch characteristics and landscape contex t as predic tors of species prcsen :e a t' d abundance: a review. - Ecoscience 6: I I 7- 124.

Mitchell-Olds, T. and Shaw. R. G. I n7. Regression -an<Jiysis of natural-select ion: stat istical-infcrnce and hi o l o~i ca l inter-pretation . - Evolu tion 4 1: 11 49- 1161 . ..

Mittlebach, G. G. et ai. 200 1. What is the observed reiat iPnship between species ri chness and produc tiv ity') - Ecology 82: 2381 - 2396.

Naeem, S. 1998. Species redundancy and ecosystec1 rel iabil ity. - Conserv. Bio i. 12: 39 --45.

Ricketts, T. H . et al. 15•99. Who"s where in North /1. ,ne rica'J - Bioscience 49: 309--38 1.

Rosenzweig. M. L. 1995. Speci~s diversi ty in space and time. - Cambridge Univ. Press.

• Rosenzweig, M. L. and AbrmnsKy. Z. 1993. How are <: ivers ity and product ivity re;ated0 · - ln: Rick lef. R. E. an t! Schulter. D. (eds), Species diversity in ecological collnm.ni ties:

ECOG RAPHY 27:2 t2004 i

hi stor ical an d geogra phical perspectives. Univ. of Chicago Press. pp. 52 - 65.

Van Rensburg, B . .1. , Chown . S. L. and Gaston, K. J. 2002. Species richness. environ mental correlates and spat ial scale: a tes t usi ng South African birds. - Am. Nat. 159: 566 - 577.

Sauer. J. R , Hines. J. E. and Fall on , J. 2001. The North American breeding bird survey, results and analysis 1966-.-:>000. ver. 200 1. 2. - USGS Patuxent Wildlife Res. Center, La ure l. MD.

Se to. K . C. . Fleishm an. E., Betrus, C. and Fay, J. in press. Predicting spatial pa tterns of butterfly and bird species richness 11 it h Landsat TM derived N DVI. - Int. J. Rem. Sens.

Shapiro, A. 11! . 1975. The temporal component of butterfly species diversi ty. - In Cody. M. L. and Diamond, J. M. (eel s) . [co logy and evo lu t i o1~ of communities. Belknap Press, Cambl"ldg ~ . pp. 181- 195.

Siegel. . . B .. Desan te, D. F. and Nott. M. P. 200 l. Using point counts to establish conservation priorities: how many visits are optima l'' ··· .J. Field Orn ithol. 72: 228 - 235.

Tilman. D. 1999 . The ecological consequences of changes in bi odiversi:y: a search for general principles. Ecology 80: 1455-- 1474.

T ilman. D. and Panda. S. 1993. The main tenance of species richness in plant commun ities. - In: Ricklef, R. E. and :>chulter. D. (eds) , Spec ies divers ity in ecological commu­nitie>: bistorical and geograph ica l perspectives. Univ. of Chicago Press, pp. 13 - 25.

Tucker. C. J. et al. I 985. Satellite remote sensing of total he rbaceo u,; biomass prod uc tion in the Senegalese Sahel: I 980 - 198' ' · - Rem. Sens. Environ. 17: 233 - 249.

Tt:el le r. P. T. and Eckhert. R. E. 1987. Big sagebrush (Artemisia­rr id,'11 /i'W r<1seymw ) and longleaf snowberry (Symphoricar­p oo: orcopJ.i/11s ) plan t associations in northeastern Nevada.

Creat Bas in Nat . 4: 117 131. Wa lker. B. H. 1992. Biodiversi ty and ecological redundancy.

- Conserv. Bioi. 6: 18- 23 . Wai kcr. B. H. I 995. Conserving biological diversity through

ecosys tem resilience. - Conserv. Bio i. 9: 747 - 752. Waide. R. B. et al. 1999. Th e relationship between produc­

tivi ty and species rich ness. - Annu. Rev. Ecol. Syst. 30: :!57 - 30(1.

Wiebe. K . L. and Martin. K. 1998. Seasonal use by birds of st ream-side rinar ian habitat in con iferous forest of north­cen tral Br it isJ{ Columbia. ·- Ecography 21: 124 - 134.

Wi lkie. D. S. o~ nd Finn , .1. T. 1996. Remote sensing imagery for nawral re .. our;:es monitoring. - Columbia Univ. Press.

Willi s .. K . J. and Whittaker. ~R. J. 2002. Ecology: species diversity: sca le mat ters. - Science 295: 1245- 1249.

Weibul l. A .. Bengtsson. J. and Noh lgren, E. 2000. Diversity of butterfJics in the agr;cultural landscape: the role of farming sy,;tem and landscape he te rogeneity. - Ecography 23: 743 -7:50.

215

Appendix I. Number of species of birds in eac:1 fu nctioml gui ld ancl total num ber of functional guilds in each canyon . Canyom are sorted (left to right) in ord er of increasing N DVl vaL1es. 2. Canyon codes are ava ilc1b le from the correspond ing author.

G uild MW :'-JE PT NW ME PW 13 E BA

AlP GGG G IGG GOG I HI G 2 4 IH IP IPR LI G LJ FG LI GG LOG N R s TGG T IFG TJGG

Total no. funct io na l guild ' 9 8 14 10 9 13

Appendix 2. The identity of species in each fun ctional guild.

Guild Code

Aeria l insect pursuers AlP

Ground gran ivore gleaners GGG

G round insect ivore/gnmiV\)1"\:! gk ancrs G IGG

Grou nd o m ni vore gleaners GOG

High insect glea ners HIG

Insect hawkers IH

21 6

Common name

Com mon nightha\vk

Vi nl t::: t -green swallow Wh ite-Jhroawd S\lifJ

Mo uming dove

Brown·hcHded cowbird Brewer's s JXl tTOW

G reen- ta iled towh~e Ho rned la rk L , rk spa rrow \Vcsran mead owiark

B!ack ·billed magpie Com mon raven \\le::> tern scrub·jay

A udubon's wa rbler Blue-grey gn et tca t c h~r

B!ack· throated gray wa rbler Con-:.m on yel!owthroa1 MacG ill ivray's warbler Mou ntain ch ickadee Oran t!c-crownt~d W(l rbleJ Plumbeous vireo \Va rh li ng vireo ··{ei lmv \\·arbler

Dusky flycatcher E:.rr:pidonax tlycatchcr Gray flycatcher Nestern wood ~ re\\ ·ee

G uil d codes are described in Appendix

R l

4

13

UN SH S.l WS

4 I

4

2

10

I

6

4

2 2

2

2

14 16 15

Scientific name

Chordei/es minor

Taclndne to tlwlassina Aeronawes saxaralis

Zenaida macroura

At/IJ!othrus ater

!)--"pi:t•/la hteli'CTi

Pipilo chlorurus

Crcmophi/a a/pcs/ris

Chondesres grammacus

Sturnclla neglec ta

Pica hud m nia

Con:us corax

Aphelocon1a cal{f'omica

Dewlroica cormuaa Polioptila l'Oerulea

/Jcli(/roica nigrcscens

Geoth~1pis rrichas Opomn1i.~· rolmiei

Puecile gan1be/i

Jl('rmil 'Ora cclata

Vireo plu111 beus I rirco gilrus Dcmlroin1 petechia

Einpidonax oberholseri E111pidonax sp. En1pidonax 11Tfghtii

Contopus sordidu/us

Kl BC Bl

2

2

4 I

2 4 I 2 4 6

2

13 12 14

ECOGRAPHY 27:2 (2004)

Appendix 2. (Con linu,·d)

Guild Code

J nsect pouncers II'

Insect probers IPR

Low insect gleaners LIG

Low in sectivore/frugi vores gleaners LIFG

Low insectivore/grani vore glt:aners LIGG

Low omnivore gleaners LOG

Nectarivores (h over) N

Raptors (pursue) R

Scavengers (pick) s

Tree granivore gleaners TGG

Tree in sectivore/fru givore gleaners T IFG

Tree insectivore/grani vure gleaners T IGG

ECOGRAPHY 27:2 (20041

Comm on name

\\'e,h.:rn kingbird

Moun ta in bluebird

L ;:: \~ ' :s· woodpecker

Downy woodpcL· k~~r

Hairy woodpecker

Ca n~ or, W l\.:n

Roc!\ wr~n

H l'~liSe wre n Rcd- ~harted !licker

Virginia's warbler

Sare tilra!:i he t·

Am erican robin He ··mit thrllsh

llrewer's b l::h.·kbird

Chi:Jping spa rrow

Lanll t bunt ing Red -w inged black Jird

Sage spa !Tow

S~otted towhee B i ack-thr~)<:ted sparrow Dark-eyed _iu nco Sons sp<:~ITO\\ '

Vl hit e-CHlwned sp:u :ow

America n crew.' Pi~:yon jay Clark ·s nu tcracker

Hroa d- tai lcd hummingbirJ

G l1 ld(:l1 ea gle i\:onhern goshawk J'< orthern harr ier Prair ie fa k <.nl Red-tni!ed ~Mwi-\

r\ merican ~est rd

rur!~ey \"l.lture

Cassin·s finch House fi nch

Pi11e siskin

Swai nson·s ti1rush

Weste rn wnager

Blaclvhea<kd t~rosbeak

8u:-:; :tt it Indi go bunt ing j un iper titmouse

Scient ific name

7_.1Tonnus cerricalis

Sialia Cl!ITliCOides

1\1l'lanerpes leu•is

Picoides pubescens

P1ct1ides l'i/los·11s

Catherpes 111exicanus Salp inctes ubsoletus

Ji·ogloc(Ftes aerlon Colapres auratus k(~rmivom l'irginiae

Oreoscoptes mont anus

llml11s 111 igrat(n·ius

Catlwrus gu({afus

Euphagus ( IW70ceplw/us

!Jjn·~ella passer ina

Passerilw an we1za Agclaius phoeniceus Amphispi:a belli

P1jJ ilo macula /us

Amphispi:a hilinea/a Junco hycnwlis .A.fcluspi: a melodia

Z onotrichia /eucophrys

Corrus brachyrhynchos

G:rn1Jun·hi11us (yanocepha/us

f'/uciji·aga columbiana

Sclusplwl"lls platycercus

.4tjuila chrysaetos

Accipii~'r gcutilis

Circus (\'aJu! us

Falco mexicanus

Bweo jmnaicensis

Falco span:erius

Ca 1/wrres aura

Corpodt1c·u.1 cassinii

CtlrpOLiacus nzexicaJzus

Carduehs pinus

Carlwrus ustulatus

Pircnga /udovicicma

PiU'udicus melanoceplwlus P.,;a/Jnjmrus minimus

Pasw rina cyanea

Baeoluphus ridgwayi

217