Diet Overlap of Mammalian Herbivores and Native Bees: Implications for Managing Co-occurring Grazers and Pollinators Author(s): Sandra J. DeBano, Samantha M. Roof, Mary M. Rowland and Lauren A. Smith Source: Natural Areas Journal, 36(4):458-477. Published By: Natural Areas Association DOI: http://dx.doi.org/10.3375/043.036.0412 URL: http://www.bioone.org/doi/full/10.3375/043.036.0412
BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published by nonprofit societies, associations, museums, institutions, and presses. Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use. Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder.
BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research. ABSTRACT: Many federal, state, and tribal agencies, as well as nonprofit organizations, have recently • increased efforts to understand how natural areas can be managed to enhance native pollinators and the ecosystem services they provide. However, managing this important group must be balanced with other services that natural areas provide including hunting, timber production, and livestock grazing. Significant knowledge gaps exist about how to effectively manage habitats used by large ungulates (e.g., Diet Overlap cattle (Bos taurus), elk (Cervus elaphus), mule deer (Odocoileus hemionus)) in ways that also enhance pollinators. One key gap is understanding the degree to which diets of mammalian herbivores overlap of Mammalian with floral resources used by bees, and how this overlap varies spatially and temporally. Invertebrate pollinators, including bees, rely on flowering forbs and shrubs for nectar and pollen. Ungulates also feed on flowering plants, although preferences vary by ungulate species, vegetation community, and Herbivores and season. Here we review existing literature on ungulate diets relative to flowering plants and compare this information with flower preferences of bees, drawing on studies of bee abundance and diversity at Native Bees: the Starkey Experimental Forest and Range in northeastern Oregon. Our review can inform managers about the potential dietary overlap between ungulates and native bees and aid planning efforts aimed at Implications for biodiversity conservation of pollinators. We discuss management implications relative to seasonal habitat use and dietary preferences of ungulates and variation in bee phenology, and conclude with guidance Managing Co- about timing and intensity of ungulate grazing when managing for multiple conservation objectives, especially in sensitive habitats like riparian areas. occurring Grazers and Index terms: dietary overlap, livestock, native bees, Starkey Experimental Forest and Range, wild ungulates Pollinators
INTRODUCTION growing emphasis on diverse services, as Sandra J. DeBano1,4 well as concerns over their vulnerability Humans enjoy a variety of benefits, or to various human activities, land managers 1Department of Fisheries and Wildlife services, from the structure and functions are increasingly seeking ways to manage Hermiston Agricultural Research associated with ecosystems (MEA 2005). multiple ecosystem services, often in the and Extension Center Historically, natural resource management absence of adequate data (Kremen and Oregon State University of public lands in the United States focused Ostfeld 2005; Mooney et al. 2009). Hermiston, OR 97838 primarily on producing food and fiber, so-called provisioning ecosystem services Pollination exemplifies a supporting eco- 2 Samantha M. Roof (Bengston 1994; MEA 2005; Chapin et al. system service of increasing concern to Mary M. Rowland3 2010). Sustained production of timber and managers, both because of the decline of Lauren A. Smith1 livestock in US forests and rangelands was some pollinator species (Potts et al. 2010; an overarching priority from the late 1800s Dumroese et al. this issue pps. 499-511); until the 1930s, when “multiple use forest- Hanula et al. this issue pps. 427-439) and 2 Bioresource Research Program ry” emerged in response to public interest the significant role pollinators play in Hermiston Agricultural Research in, and demand for, additional types of food production and supporting flowering and Extension Center ecosystem services (Bengston 1994). These plant diversity in natural areas (Ollerton Oregon State University Hermiston, OR 97838 included cultural services such as hunting, et al. 2011). Thus, a new challenge for fishing, and other recreational activities, resource managers is to continue providing 3USDA Forest Service and supporting services such as clean traditional services (e.g., timber, livestock Pacific Northwest Research Station water and soil retention (Bengston 1994; production, hunting, and fishing), but in La Grande Forestry and Range Sciences MEA 2005). In the last 30 years, the types ways that also conserve or benefit pollina- Laboratory of services considered in natural resource tors. Research aimed at understanding how La Grande, OR 97850 decisions have become even more diverse to effectively manage diverse ecosystem in response to paradigm shifts (e.g., the services in ways that maximize benefits “forest ecosystem management” approach), and minimize negative tradeoffs is scarce • and to the emergence of ecosystem services but growing (Bennett et al. 2009; Wainger 4 Corresponding author: sandy.debano@ management frameworks (Bengston 1994; et al. 2010). oregonstate.edu; 541-567-6337 ext.116 Chapin et al. 2010; Deal et al. 2012). As a result, managers now consider a wider In this paper, we examine one potential area variety of ecosystem services, many of of interaction between two key ecosystem which are supporting and regulating ser- service providers: vertebrate grazers and vices that are less directly connected to end invertebrate pollinators. For vertebrate Natural Areas Journal 36:458–477 products that humans value (MEA 2005; grazers, we focus on three ungulates: cattle Deal et al. 2012; Ringold et al. 2013). With (Bos taurus Linnaeus), elk (Cervus elaphus
458 Natural Areas Journal Volume 36 (4), 2016 Linnaeus), and mule deer (Odocoileus he- plants plays a role for some pollinators that as the availability of nutritious resources mionus Rafinesque). Cattle production is an build above-ground nests. Soil character- decrease (Findholt et al. 2005). As veg- important provisioning service supported istics, such as compaction, bare ground, etation senesces in late summer, cattle by public lands, with cattle making up the and stability, can affect ground nesting move closer to water and may consume largest proportion of livestock produced in bees (Cane 1991; Potts and Willmer 1997, relatively more forbs and browse if grass is the United States. Deer and elk, the most 1998). However, here we focus on poten- less available in those areas (Findholt et al. significant game animals in the United tial effects of ungulate grazing on floral 2005; Roever et al. 2015). However, cattle States, provide important cultural and resources used by native bees for food by consumption of forbs may actually decline provisioning services, such as hunting, examining the potential for dietary overlap as summer progresses and forbs senesce wildlife viewing and serving as tradition- amongst these groups. While some studies and become less palatable (Holechek et al foods for many Tribal nations (Bolon have examined ungulate grazing effects on al. 1982a). 1994, McCabe 2002). For pollinators, we floral resource availability (Carvell 2002; focus on native bees, the most diverse Hatfield and LeBuhn 2007; Sjödin 2007; Unlike mammalian herbivores, bees rely and abundant pollinators in natural areas, Sjödin et al. 2008; Xie et al. 2008; Kimoto almost entirely on forbs and flowering with approximately 4000 species in North et al. 2012b), none that we are aware of shrubs for pollen and nectar, with forbs America (NRC 2007). have explored how dietary preferences of being particularly important because this different ungulates, in combination with species-rich group is most likely to provide Several studies have examined effects of temporal and spatial patterns of use, may floral resources to the range of bees active ungulate herbivory on pollinators, includ- affect their dietary overlap with native bees. throughout the growing season (Dumroese ing native bees (see Black et al. 2011; et al. this issue pps. 499-511). Although Hanula et al. this issue pps. 427-439; for Ungulate diets have been an active area certain forb species may be particularly reviews). Results of these studies vary of research in range science and wildlife attractive to large numbers of pollinators depending on several factors, including biology for decades, with numerous studies (Dumroese et al. this issue pps. 499-511)), plant and bee community composition, across the western United States (Table bee species, like mammalian herbivores, intensity of grazing, type of grazer, timing 1), often in the context of understanding may vary strongly with respect to diet. and duration of grazing, and land use. Sev- dietary or niche overlap between wild and Most specialization occurs relative to eral studies of livestock grazing and native domestic ungulates. While deer, elk, and pollen, with bees typically ranging from bees have found significant effects, some cattle diets do overlap as they all feed on being broad generalists (polylectic) to negative (Kruess and Tscharntke 2002; graminoids, forbs, and browse (including relative specialists that focus on related Hatfield and LeBuhn 2007; Sjödin 2007; flowering shrubs), the relative proportions or morphologically similar plant taxa Xie et al. 2008; Kearns and Oliveras 2009) consumed by each can vary in time and (oligolectic) (Westerkamp 1996; Michener and some positive (Carvell 2002; Vulliamy space. Although elk and cattle generally 2007). Specialization is often associated et al. 2006; Yoshihara et al. 2008). Most prefer graminoids over other forage, their with morphological differences that pre- studies have been observational, but a large diets may consist of up to 51 and 20% sumably allow bees to more easily access scale manipulation in the Pacific Northwest forbs, respectively (Cook 2002; Stewart et certain flowers (e.g., long proboscises, showed cattle grazing altered native bee al. 2003; Findholt et al. 2005; Christianson specialized pollen collecting hairs, facial abundance, richness, diversity, and com- and Creel 2010). When preferred forages modifications) (Thorp 1979). Although munity composition, with taxa varying in are less abundant, browse is also consumed much is still unknown about floral spe- their sensitivity to grazing (Kimoto et al. by elk and cattle, and can comprise 18 to cialization and preferences of many bee 2012b). However, the mechanisms driving nearly 100% and 4−18% of diets, respec- species, numerous studies have examined observed responses in most of these studies tively (Kufeld 1973; Cook 2002; Findholt specific bee-flower relationships. This work are seldom directly examined. et al. 2005; Christianson and Creel 2010). has been supplemented with inferences Mule deer diets are typically more variable, based on matching morphological features Ungulate herbivory can impact native bees with forbs composing 14−56% of the diet and phenologies between bees and flowers, through various mechanisms, including and browse composing 16−48% (Hansen and assumptions of similarities among taxa. effects on plant growth, architecture, and Reid 1975; Stewart et al. 2003; Find- The result is that geographically-targeted diversity, and quality, as well as soil holt et al. 2005). technical guides that identify the genus characteristics and microhabitat conditions and species of flowering plants preferred (Kruess and Tscharntke 2002; DeBano Ungulate diets vary in time and space in by pollinators are becoming more common 2006a, b; Black et al. 2011; Schmalz response to a variety of factors including (Pendergrass et al. 2008; Ogle et al. 2011a, et al. 2013). These changes can impact changing plant diversity and phenology, b; Pollinator Partnership n.d. a, b; Tilley nesting resources and/or food that native general patterns of habitat use, and intra et al. 2013; Vaughan et al. 2015). bees depend on (Vázquez and Simberloff and interspecific competition for forage. 2004; Black et al. 2011). For example, For example, late summer and early fall Although much is known about diets of plant material is often necessary for nest can be a period of intense competition for ungulates and native bees, information construction, and the physical structure of forage, especially in drought-prone areas, from these disparate fields has not been
Volume 36 (4), 2016 Natural Areas Journal 459
T T Bee plant Bee PPB P Ob, T PPA, PPB, P, T Ob, T Ob, T Oa, Ob, T P, P, T, P, Oa, Ob, P, P, Oa, PPA, PPB, P, Oa, P, Oa, PPB, Ob, T
V P, Oa, Ob, Cattle diet Cattle D, V P, Ob, T V W, T Hb, Ha Hb. Ha D,
DE diet
D, Mule Deer Deer Mule DE Ha, DD T D D D Wi HR V Oa, Ob, T HR D, HR,MS, CO
BC, Elk diet Elk D K Wi CO K, CO GA, SK, YR, PI, K, D, Wi, BC HR, K, A, MA K, BL, KO, MC K, D, ROa, K, GA, SK, BY, PI, CO BR, CA, E, HS, KN, ROb, CO YR, K, ST, BY,
K, D, SE,
Sep-14
Jul/Aug 2014 Jul/Aug
Jun/Jul 2014 Jun/Jul May-14
x Native status Native I x III x x xI x x K, D, SE, BR, CA, E, HS, KN, ROb, CO K, D, SE, BR, CA, E, HS, KN, ROb, CO D, DE D, DE Ha, D, V Ha, D, V P, Oa, Ob, T I x I x Nx Lehm. N x x K, D, SE, BR, CA, E, HS, KN, ROb, CO D, DE Ha, D, V L. I x (A. Gray) Iltis N x Nutt. N x x Ledeb. N x Pursh N x (Benth.) Kuntze N x A. Gray N x ) Hook. N x x Richardson N x (A. Dietr.) E.P. L. (Hook.) Benth. N x L. L. Nutt. N x L.
L.
T
Bee plant Bee T Douglas N x
L. Bee plant Bee sp P P P Oa, P P PPA, Ob, T P Oa, PPA, Ob, T P, P, Oa, T (Continued P, P, Oa, T P, Oa, PPB, Ob, T P, Oa, P, Oa, PPA, Ob, T P, Oa, PPB, Ob, T P, Oa, P, Oa, PPA, Ob, T
Medicago lupulina Trifolium pratense/repens Vicia cracca Trifolium repens Trifolium wormskioldii Family Gentianaceae Gentianopsis simplex Family Grossulariaceae Ribes cereum Family Iridaceae Iris missouriensis Thermopsis montana Lotus unifoliolatus Trifolium pratense Ribes hudsonianum Hypericum scouleri Family Fabaceae Astragalus reventus Family Crassulaceae Sedum stenopetalum Olsynium douglasii Thlaspi arvense Table Family Caryophyllaceae Dianthus armeria Family Clusiaceae Hypericum perforatum Table 2 Barbarea orthoceras Descurainia Bicknell Family Lamiaceae Agastache urticifolia W
Cattle diet Cattle W Cattle diet Cattle D VP VP D Ha, Hb, Hc, D, W Hb, D,
Ha, Hb,
Hc, D, W Hb, D,
diet DE
diet DE
Mule Deer Deer Mule D D D, V DD Mule Deer Deer Mule DE D DD D D, V DE D, DE D, DE D MS, D D D, DE D, DE D MS, D D HR, D, DE D P, Oa, T HR, D, HR, D, DE D P, Oa, T HR, D, E, KR, SE, E, KR, SE,
D, GRa CO D, GRa CO
Elk diet Elk a Elk diet Elk CO a CO D D K K PR D, PR CO D, CO CO CO K, D K, BR, YR K, D K, BR, YR K, ROa K, ROa K, K, K, K, K, GA, BO, K, GA, BO, MS, K, D, ST, CO MS, K, D, ST, CO K, MS, D, E, KN, CO K, MS, D, E, KN, CO K, GO, MC, SC, K, GO, MC, SC, CL, GO, GRa, GRd, PE, SC, KN, CO CL, GO, GRa, GRd, PE, SC, KN, CO SI, SI,
K, D,
K, D, K, D, BL, BO, BR, KN, PR, SI, SK, K, D, BL, BO, BR, KN, PR, SI, SK,
Sep-14 Sep-14
Jul/Aug 2014 Jul/Aug Jul/Aug 2014 Jul/Aug
Jun/Jul 2014 Jun/Jul Jun/Jul 2014 Jun/Jul
May-14 May-14
Native status Native Native status Native I x N xx Nx N xx N x x Nx N x N x x K, D, BL, GA, N, ROa, SI, YR, PI, CO BA I x N xx Nx N xx N x x Nx N x N x x K, D, BL, GA, N, ROa, SI, YR, PI, CO BA (Lindl.) (Lindl.) Arnica Arnica Nutt. N x x Nutt. N x x Nutt. N x x MS, K, D, ST, CO (Hook. & Arn.) W. Bartram N x F.H. Wigg. N/I x x (Nutt.) J.M. (A. Gray) Greene N x Nutt. N x x MS, K, D, ST, CO (Hook. & Arn.) W. Bartram N x F.H. Wigg. N/I x x (Nutt.) J.M. (A. Gray) Greene N x L. var. Scop. I x x Less./ (Lindl.) DC. N x L. var. Scop. I x x Less./ (Lindl.) DC. N x Hook. N x (L.) Scop. I x (Pursh) Raf. N x x J.M. Coult. & Rose N x Hook. N x Link ex Roem. & Nutt. (L.) Scop. I x (Pursh) Raf. N x x J.M. Coult. & Rose N x Link ex Roem. & Nutt. Hook. Hook. DC. DC. Greene Greene ororia Agoseris glauca Perideridia gairdneri Mathias Family Asteraceae Achillea millefolium occidentalis Arnica chamissonis Heracleum maximum Family Apiaceae Angelica canbyi s G.L. Nesom Taraxacum officinale Schult. Lomatium ambiguum Coult. & Rose Table 1. Blooming plant species present during four sampling bouts at the USFS Starkey Experimental Forest and Range. “N” indicates native status, “I” indicates Table 1. Blooming plant species present during four sampling bouts at the USFS Starkey Experimental Forest and Range. “N” indicates the genus or species has been introduced, and “N/I” indicates that status depends on subspecies. Columns for elk, mule deer, cattle diets indicate whether to native bees. Bold-faced font detected in diets of those ungulates the literature examined. Bee plants indicate taxa identified as important indicates species-level matches and normal font genus-level matches. Plant nomenclature follows NRCS (2015). Madia glomerata Senecio serra Solidago missouriensis Stenotus lanuginosus Symphyotrichum spathulatum Tragopogon dubius Family Boraginaceae Myosotis stricta Family Brassicaceae Grindelia nana Erigeron speciosus Cirsium arvense Erigeron corymbosus ororia Agoseris glauca Table 1. Blooming plant species present during four sampling bouts at the USFS Starkey Experimental Forest and Range. “N” indicates native status, “I” indicates Table 1. Blooming plant species present during four sampling bouts at the USFS Starkey Experimental Forest and Range. “N” indicates the genus or species has been introduced, and “N/I” indicates that status depends on subspecies. Columns for elk, mule deer, cattle diets indicate whether to native bees. Bold-faced font detected in diets of those ungulates the literature examined. Bee plants indicate taxa identified as important indicates species-level matches and normal font genus-level matches. Plant nomenclature follows NRCS (2015). Perideridia gairdneri Mathias Family Asteraceae Achillea millefolium occidentalis Arnica chamissonis Heracleum maximum Family Apiaceae Angelica canbyi s G.L. Nesom Taraxacum officinale Schult. Coult. & Rose Lomatium ambiguum Erigeron corymbosus Cirsium arvense Erigeron speciosus Symphyotrichum spathulatum Grindelia nana Madia glomerata Senecio serra Solidago missouriensis Stenotus lanuginosus Tragopogon dubius Family Boraginaceae Myosotis stricta Family Brassicaceae
460 Natural Areas Journal Volume 36 (4), 2016
T T Bee plant Bee PPB P Ob, T PPA, PPB, P, T Ob, T Ob, T Oa, Ob, T P, P, T, P, Oa, Ob, P, P, Oa, PPA, PPB, P, Oa, P, Oa, PPB, Ob, T
V P, Oa, Ob, Cattle diet Cattle D, V P, Ob, T V W, T Hb, Ha Hb. Ha D,
DE diet
D, Mule Deer Deer Mule DE Ha, DD T D D D Wi HR V Oa, Ob, T HR D, HR,MS, CO
BC, Elk diet Elk D K Wi CO K, CO GA, SK, YR, PI, K, D, Wi, BC HR, K, A, MA K, BL, KO, MC K, D, ROa, K, GA, SK, BY, PI, CO BR, CA, E, HS, KN, ROb, CO YR, K, ST, BY,
K, D, SE,
Sep-14
Jul/Aug 2014 Jul/Aug
Jun/Jul 2014 Jun/Jul May-14
x Native status Native I x III x x xI x x K, D, SE, BR, CA, E, HS, KN, ROb, CO K, D, SE, BR, CA, E, HS, KN, ROb, CO D, DE D, DE Ha, D, V Ha, D, V P, Oa, Ob, T I x I x Nx Lehm. N x x K, D, SE, BR, CA, E, HS, KN, ROb, CO D, DE Ha, D, V L. I x (A. Gray) Iltis N x Nutt. N x x Ledeb. N x Pursh N x (Benth.) Kuntze N x A. Gray N x ) Hook. N x x Richardson N x (A. Dietr.) E.P. L. (Hook.) Benth. N x L. L. Nutt. N x L. L. T T
Douglas N x
L. Bee plant Bee sp PPB P Ob, T PPA, PPB, P, T Ob, T Ob, T Oa, Ob, T (Continued P, P, T, P, P, Oa, Ob, P, Oa, PPA, PPB, P, Oa, P, Oa, PPB, Ob, T
Medicago lupulina Trifolium pratense/repens Vicia cracca Trifolium repens Trifolium wormskioldii Family Gentianaceae Gentianopsis simplex Family Grossulariaceae Ribes cereum Family Iridaceae Iris missouriensis Thermopsis montana Lotus unifoliolatus Trifolium pratense Ribes hudsonianum Hypericum scouleri Family Fabaceae Astragalus reventus Family Crassulaceae Sedum stenopetalum Olsynium douglasii Thlaspi arvense Table Family Caryophyllaceae Dianthus armeria Family Clusiaceae Hypericum perforatum Table 2 Barbarea orthoceras Descurainia Bicknell Family Lamiaceae Agastache urticifolia
V P, Oa, Ob, Cattle diet Cattle D, V P, Ob, T V W, T Hb, Ha Hb. Ha D,
DE diet
D, Mule Deer Deer Mule DE Ha, D D D DD T Wi HR HR V Oa, Ob, T D, HR,MS, CO
BC, Elk diet Elk D K Wi CO K, CO GA, SK, YR, PI, K, D, Wi, BC HR, K, A, MA K, BL, KO, MC K, D, ROa, K, GA, SK, BY, PI, CO BR, CA, E, HS, KN, ROb, CO YR, K, ST, BY,
K, D, SE,
Sep-14
Jul/Aug 2014 Jul/Aug
Jun/Jul 2014 Jun/Jul May-14
x Native status Native I x x I x I x II x x K, D, SE, BR, CA, E, HS, KN, ROb, CO K, D, SE, BR, CA, E, HS, KN, ROb, CO D, DE D, DE Ha, D, V Ha, D, V P, Oa, Ob, T I x I x Nx Lehm. N x x K, D, SE, BR, CA, E, HS, KN, ROb, CO D, DE Ha, D, V L. I x (A. Gray) Iltis N x Nutt. N x x Ledeb. N x Pursh N x (Benth.) Kuntze N x A. Gray N x ) Hook. N x x Richardson N x (A. Dietr.) E.P. L. (Hook.) Benth. N x L. L. Nutt. N x L. L. ) Douglas N x L. sp (Continued Continued Family Gentianaceae Gentianopsis simplex Family Grossulariaceae Ribes cereum Family Iridaceae Iris missouriensis Trifolium wormskioldii Vicia cracca Ribes hudsonianum Family Fabaceae Astragalus reventus Lotus unifoliolatus Hypericum scouleri Medicago lupulina Trifolium pratense/repens Trifolium repens Olsynium douglasii Family Crassulaceae Sedum stenopetalum Thermopsis montana Trifolium pratense Table Family Caryophyllaceae Dianthus armeria Thlaspi arvense Family Clusiaceae Hypericum perforatum Bicknell Table 2 Barbarea orthoceras Descurainia Family Lamiaceae Agastache urticifolia Table 1. ( 1. Table
Volume 36 (4), 2016 Natural Areas Journal 461
T T Bee plant Bee PPB P Ob, T PPA, PPB, P, T Ob, T Ob, T Oa, Ob, T P, P, T, P, Oa, Ob, P, P, Oa, PPA, PPB, P, Oa, P, Oa, PPB, Ob, T
V P, Oa, Ob, Cattle diet Cattle D, V P, Ob, T V W, T Hb, Ha Hb. Ha D,
DE diet
D, Mule Deer Deer Mule DE Ha, DD T D D D Wi HR V Oa, Ob, T HR D, HR,MS, CO
BC, Elk diet Elk D K Wi CO K, CO GA, SK, YR, PI, K, D, Wi, BC HR, K, A, MA K, BL, KO, MC K, D, ROa, K, GA, SK, BY, PI, CO BR, CA, E, HS, KN, ROb, CO YR, K, ST, BY,
K, D, SE,
Sep-14
Jul/Aug 2014 Jul/Aug
Jun/Jul 2014 Jun/Jul May-14
x Native status Native I x III x x xI x x K, D, SE, BR, CA, E, HS, KN, ROb, CO K, D, SE, BR, CA, E, HS, KN, ROb, CO D, DE D, DE Ha, D, V Ha, D, V P, Oa, Ob, T I x I x Nx Lehm. N x x K, D, SE, BR, CA, E, HS, KN, ROb, CO D, DE Ha, D, V L. I x (A. Gray) Iltis N x Nutt. N x x Ledeb. N x Pursh N x (Benth.) Kuntze N x A. Gray N x ) Hook. N x x Richardson N x (A. Dietr.) E.P. L. (Hook.) Benth. N x L. L. Nutt. N x L. L. Ob, T
Ob, T Douglas N x
L. Bee plant Bee sp P T P PPA PPB, (Continued P, PPA P, PPA Oa, Ob, T Oa, PPA, Ob, T P, Oa, P, Oa, PPA,
Medicago lupulina Trifolium pratense/repens Vicia cracca Trifolium repens Trifolium wormskioldii Family Gentianaceae Gentianopsis simplex Family Grossulariaceae Ribes cereum Family Iridaceae Iris missouriensis Thermopsis montana Lotus unifoliolatus Trifolium pratense Ribes hudsonianum Hypericum scouleri Family Fabaceae Astragalus reventus Table 2 Barbarea orthoceras Descurainia Family Crassulaceae Sedum stenopetalum Olsynium douglasii Thlaspi arvense Table Family Caryophyllaceae Dianthus armeria Family Clusiaceae Hypericum perforatum Bicknell Family Lamiaceae Agastache urticifolia , Cattle diet Cattle Hc W , R V V DP V D
D,
Hb
diet
D Mule Deer Deer Mule DE D D Wi Ha BA D, Wi D, W T D, Wi MS, MS, D D CO CO D, , SE CO CO CO
CO Elk diet Elk CO K K, D, CO PI, CO CO K, YR D, CO Wi, CO D, Wi, BO, CA, E, GRd, KR, K, K, D, CO K, PI, YR K, BO, GA, GRd, KN, YR, K, KN, BR K, KN, BR MS, GA, GRc, MC, K, YR, HS, BO, CA, E, GRd, KN, ROa, SE, PR, PI, SI, YR, KN, ROa, SE, PR, K, KF, PI, K, D, KR, K, PI, SI, YR, HBE,
K, D, Wi, PR, SE,
HBE, K, Wi, GA, HS, SC, YR, BL, PT, KW, HBE, HBE,
Sep-14
Jul/Aug 2014 Jul/Aug
Jun/Jul 2014 Jun/Jul
May-14 Native status Native N x x N xx N x Hook. N x C. Presl N x Benth. N x Greene N x Nutt. N x x L. N x Pursh N x Durand N x D. Don ex G. Lindl. N x x x Lindl. N x x Lindl. N x Duchesne N x ) (Hook.) Greene N x (Pursh) Greene N x Raf. N x Hook. N x Douglas ex Hook. N x x x L. (Nutt. ex Torr. & A. S. Watson N x ) C. Presl N x y Continued Continued ) A. Gra y Allium madidum Camassia quamash Potentilla gracilis Family Liliaceae Allium acuminatum Triteleia grandiflora Rosa nutkana Veratrum californicum Prunella vulgaris Table 2 ( Monardella odoratissima Family Onagraceae Epilobium brachycarpum Fragaria virginiana Potentilla glandulosa Sanguisorba canadensis Family Ranunculaceae Ranunculus orthorhynchus Ranunculus uncinatus Don Family Rosaceae Crataegus douglasii Polygonum bistortoides Family Malvaceae Sidalcea oregana Gra Epilobium ciliatum Family Polemoniaceae Microsteris gracilis Polemonium occidentale Family Polygonaceae Eriogonum heracleoides Table 1. ( 1. Table
462 Natural Areas Journal Volume 36 (4), 2016
= Bee plant Bee CA ; PPB PPB PPB Ob, T 0)* 7
P, Oa, PPA, PPB, d (19 Cattle diet Cattle ; BY = Boy
9)*
5
diet (19 Mule Deer Deer Mule SV DV DV ates d C = Burt an C ; B 3)* 5 (19 a d
CO CO Elk diet Elk , KF, S K K, BR ; BR = Braz K, BR K, D, CO 4)* 7 K, D, KF, SC, K, D, SC (19 ne h K, D, BO, CL, GA, SK, YR, PI, CO D = Bo O
; B Sep-14
* ) Jul/Aug 2014 Jul/Aug d (1966)*
oo
1939 l
( Jun/Jul 2014 Jun/Jul May-14
x ; BL = B Native status Native Nxx N x N x x N x x N x N x (1983) and Robinette g = Bartmann YR = Youn A (Hook.) ); ; B y 2009 6)* ( 5 ) L. I x Howell N x Greene N x Greenm. N x (19 L. ffels . L. L. y l ) Pursh Sm. sp. Continued W = ); erson et a Continued d n A 1984 ( utt. ex Torr. & A. Gra = Plant species for this reference are those eaten equal to or in excess of their availability. These data are based on grazing trials of tame female elk collected April - October, 2004- Plant species for this reference are those eaten equal to or in excess of their availability. These data based on grazing trials Table 2 ( Family Rubiaceae Galium aparine N Penstemon Saxifraga nidifica Verbascum thapsus Family Valerianaceae Valeriana dioica Family Violaceae Viola adunca Viola nuttallii Galium boreale Saxifraga oregana Family Scrophulariaceae Castilleja cusickii Family Saxifragaceae Lithophragma parviflorum P = Pendergrass et al. (2008); Oa = Ogle et al.(2011a); Ob = Ogle et al. (2011b); PPA = Pollinator Partnership (n.d.a); PPB = Pollinator Partnership (n.d.b); T = Tilley et al. (2013). P = Pendergrass et al. (2008); Oa Ogle al.(2011a); Ob (2011b); PPA Pollinator Partnership (n.d.a); PPB Legend: Elk, Deer, and Cattle Literature Cited: A * = As reported in Cook (2002) Native Bee Literature Cited: Capp (1967)*; CL = Claar (1973)*; CO = Cook et al. (2014); D = Damiran (2006); DE = Deschamp et al. (1979); E =Eustace (1967)*; GA = Gaffney (1941)*; GO Gordon (1968)*; Capp (1967)*; CL = Claar (1973)*; CO Cook et al. (2014); D Damiran (2006); DE Deschamp (1979); E =Eustace = Holecheck et al. (1982a); Hb Holechek GRa = Greer (1959)*; GRc (1960)*; GRd et al. (1970)*; HR= Hansen and Reid (1975)*; HBE Hobbs (1981); Ha KO = Komberec (1976)*; KR Kirsch et al. (1982b); Hc = Holechek (1982c); HS Hash (1973)*; K Kufeld KF Korfhage (1974)*; KN Knight (1970)*; = Peek (1963)*; PI Pickford and (1963)*; KW = Knowles (1975)*; MA Martinka (1969)*; MC Mackie (1970)*; MS Mower and Smith (1989)*; N Nichols (1957)*; PE et al. (2005); SC = Schallenberger Reid (1943)*; PR = Probasco (1968)*; PT Picton (1960)*; R Rosiere (1975); ROa Rouse (1957)*; ROb (1958)*; S Sandoval V = Van Dyne and Heady (1965); Wi Wickstrom et (1965)*; SI = Singer (1975)*; SK Stark (1973)*; SE Stevens (1966)*; ST Stewart et al. (2011); T Thetford (1971); al. 2006 in the Sled Springs Game Management Unit Blue Mountains Ecoregion of Northeast Oregon. a Table 1. ( 1. Table
Volume 36 (4), 2016 Natural Areas Journal 463 synthesized, making it difficult for manag- current knowledge of dietary overlap to ten occurring in spring (Filip et al. 1989). ers to evaluate the magnitude of potential inform the timing and intensity of ungulate Dominant riparian shrubs include willow dietary overlap between these groups and grazing, especially in sensitive habitats like (Salix spp. L.), black hawthorn (Crataegus to mitigate negative effects resulting from riparian areas, with the goal of supporting douglasii Lindl.), thinleaf alder (Alnus in- overlap. Ultimately, the degree of realized multiple, compatible uses. cana (L.) Moench subsp. tenuifolia (Nutt.) dietary overlap is determined by three Breitung), black cottonwood (Populus bal- factors: spatial separation, temporal asyn- samifera L. subsp. trichocarpa (Torr. & A. THE STARKEY EXPERIMENTAL chronicity, and dietary differences (Stewart Gray ex Hook.) Brayshaw), and common FOREST AND RANGE AS A CASE et al. 2011). Understanding the first two snowberry (Symphoricarpos albus (L.) S.F. STUDY factors requires quantifying the spatial Blake). Scattered ponderosa pine, Douglas and temporal variability of native bees fir (Pseudotsuga menziesii (Mirb.) Franco), and blooming forbs and shrubs, which, in Study Area and western larch (Larix occidentalis Nutt.) temperate zones, often show strong season- are also found in the riparian corridor. al patterns related to phenology (Williams Starkey is located in the Blue Mountains of Herbaceous vegetation includes a variety of et al. 2001; Kimoto et al. 2012a). These northeastern Oregon (Rowland et al. 1997; forbs (described below) as well as sedges data can then be combined with results in Wisdom et al. 2005) (Figure 1). Elevations (Carex spp. L.), rushes (Juncus spp. L.), published studies of diets and habitat use range from 1130 m to 1500 m and annual common spikerush (Eleocharis palustris of ungulates and bees to identify when precipitation is approximately 50 cm, (L.) Roem. & Schult.), creeping bentgrass and where the most significant overlap with the majority falling as snow (Skovlin (Agrostis stolonifera L.), and fowl man- may occur, and to inform management of 1991). The forest and rangeland habitats nagrass (Glyceria striata (Lam.) Hitch.). sensitive areas. We illustrate this approach at Starkey were formally designated as a Upland habitat consists of mixed conif- with a case study of riparian areas along research site by the USFS in 1940 to eval- erous forest with Douglas-fir and grand Meadow Creek at the US Forest Service uate how management actions and natural fir Abies( grandis (Douglas ex D. Don) (USFS) Starkey Experimental Forest and disturbance affect multiple resources in this Lindl.), along with lodgepole (P. contorta Range (Starkey) in eastern Oregon. Ri- ponderosa pine (Pinus ponderosa Lawson Douglas ex Loudon) and ponderosa pine parian areas are of special concern, given & C. Lawson)/mixed conifer/native bunch- (Pinus ponderosa Lawson & C. Lawson). their key role in supporting high levels grass setting (Skovlin 1991). Beginning of biodiversity and ecological services, with the arrival of settlers on the Oregon Because of legacies associated with past their history of frequent disturbance and Trail, much of Starkey and the surrounding land uses (e.g., logging, livestock grazing, subsequent restoration, and their sensitivity landscape have been heavily degraded from planting exotic forage) in the riparian corri- to ungulate grazing (Case and Kauffman grazing and logging (Skovlin 1991). In the dor (Skovlin 1991; USFS 2012), the USFS 1997; Clary and Kruse 2003; DeBano et 1980s, approximately 10,000 ha at Starkey began actively restoring approximately 11 al. 2003; DeBano and Wooster 2003; Ha- were enclosed with game-proof fencing km of Meadow Creek in 2013, primarily nula and Horn 2011; Williams 2011). The so that closed populations of wild elk and to improve habitat for salmonids (USFS objectives of this paper are to (1) describe deer could be manipulated, and effects of 2012). Restoration efforts included planting the spatial and temporal variability of na- management evaluated, in an experimental approximately 40,000 native shrubs and tive bees and blooming forbs and shrubs setting. Research at Starkey since that time conifers along the stream, placing boulders in riparian areas along Meadow Creek at has focused on increased understanding of and large woody debris throughout the Starkey, (2) conduct a literature review of ungulate population dynamics, interactions reach, and developing upland water sources flowering plant composition of ungulate and grazing effects of wild and domestic and fencing to support a deferred rotation diets, (3) survey technical guides for lists ungulates, how common management ac- grazing system for cattle (USFS 2012). In of plants preferred by native bees, and (4) tivities affect ungulates and other resources, addition, research exclosures were built combine information from the literature and effects of natural disturbance regimes in three pastures along Meadow Creek reviews and technical guides with bee and on vegetation and wildlife (Rowland et al. to evaluate effects of ungulate herbivory plant data from Meadow Creek to deter- 1997; Wisdom et al. 2005). These efforts on restoration plantings, fish habitat, and mine the potential degree of dietary overlap have included numerous studies of deer, other variables within the riparian system. between mammalian herbivores and native elk, and cattle diets (Holechek et al. 1982a, To better understand effects of riparian bees at Starkey. Our experiences at Starkey b, c; Stewart et al. 2003, 2011). restoration on multiple ecosystem services, will illustrate one approach to identifying and to partition effects of wild vs. domestic and addressing factors posing the greatest Our study was conducted at 12 riparian ungulate grazing on floral resources, we management challenges to enhancing na- sites located along a 13-km reach of Mead- conducted a study documenting spatial tive pollinators given their dietary overlap ow Creek, a major tributary of the upper and temporal variation in native bees and with ungulates, and highlight management Grande Ronde River that flows through flowering plants of Meadow Creek. These implications relative to seasonal habitat use Starkey (Figure 1). Discharge in Meadow data not only provide a baseline to inform by ungulates and their dietary preferences. Creek varies within and across years, with planned future studies, but also to better This approach can aid managers in using severe scouring by ice and high flows of- understand potential overlap between diets
464 Natural Areas Journal Volume 36 (4), 2016 Figure 1. Location of the USFS Starkey Experimental Forest and Range in Oregon and the 12 study sites for sampling native bee and floral resources along Meadow Creek. Site 1 is the furthest upstream and Site 12 is the furthest downstream. of vertebrate grazers and native bees, and 400 cow-calf pairs graze at Starkey in a found that niche overlap between mule the implications of that overlap. deferred rotation grazing system, including deer, elk, and cattle varied by season and portions of Meadow Creek. Cattle grazing factor. For example, cattle used more gentle Use of Starkey by Deer, Elk, and in the new Meadow Creek pastures will slopes and lower elevations than did mule Cattle begin in summer 2016, with approximately deer and elk in spring and summer, whereas 120 cow-calf pairs in addition to the cattle distributions in relation to xeric grasslands Mule deer, elk, and cattle have been dom- that graze the remainder of Starkey. and the percentage of logged forest were inant herbivores in the Blue Mountains similar among the species in summer. ecoregion for well over a century. Tra- Long-term research on deer, elk, and ditionally deer were summer residents at cattle at Starkey has demonstrated strong Water is a consistent driver of ungulate Starkey, whereas elk were spring and fall separation among these species in space distributions in Starkey and elsewhere, migrants (Skovlin et al. 1968). Since the and time, including habitat use in relation although the magnitude of this effect differs initiation of the Starkey Project in 1989 to riparian areas and other water sources. across species and time. Cattle distribution and concomitant erection of the game-proof Research prior to the erection of the fence is often influenced by distance to water fence, deer and elk no longer migrate out of around Starkey revealed that elk and mule (Ganskopp 2001) and, in general, they Starkey but are fed on a winter feedground deer use was greater in pastures not grazed prefer areas with gentle slopes in close (Rowland et al. 1997). Current pre-parturi- by cattle and that the effect of cattle was proximity to water (Roever et al. 2015). tion estimates of deer and elk are 150 and greater on elk than deer (Skovlin et al. In evaluating patterns of use by mule deer 350, respectively, for the Main Study Area 1968). Johnson et al. (2000) explored dis- and elk at Starkey, Ager et al. (2003) found of 7760 ha, which includes Meadow Creek. tributions of mule deer and elk in spring that the average distance from water was Cattle have grazed the Starkey landscape and found strong separation between the 300 m for elk and approximately 280 m since the 1860s. Currently, approximately species. Similarly, Stewart et al. (2002) for deer. Stewart et al. (2002) examined
Volume 36 (4), 2016 Natural Areas Journal 465 distributional overlap of deer, elk, and cattle of a UV-reflective yellow plastic container collected with all methods were frozen, at Starkey and found that distance to water (15 cm diameter × 15 cm high) with a pinned, labeled, and identified to genus. was a strong predictor of resource use; all UV-reflective blue polypropylene screw Because only vane trapping was conducted species preferred areas closer to water. funnel with two 24 × 13-cm cross vanes at all sites during all bouts, we used these measuring 3-mm in thickness (SpringStar, data for all direct comparisons between Woodinville, WA, USA) (Stephen and Rao bouts and locations. Native Bees and Floral Resources in 2005). One vane trap was placed at each Riparian Areas at Starkey of the 12 sites, suspended approximately We collected data on the presence and 1.2 m from the ground with a wire hanger abundance of blooming plant species at Dietary competition not only depends on inserted into an aluminum pipe. No liquids each site during the same periods when the overlap in preferred plants of focal or other killing agents were used in traps. bees were sampled (Table 2) on 0.3-m species, but also on the degree of spatial Vane traps have been found to be effective wide belts along the same five parallel and temporal overlap of foraging activities. in sampling native bees in other studies transects described above. The species and Thus, managing potential overlap will in the region (e.g., Kimoto et al. 2012a, number of stems of each blooming plant be impacted by variability in both plant b; Gonzalez et al. 2013; Tubbesing et al. within the belt transect were recorded and availability and animal activity. Although 2014). Vane traps were open for approx- all plants were identified to species with habitat use patterns of deer, elk, and cattle imately 2 days (X̅ = 29.9 ± 0.6 daylight the following exceptions: Trifolium repens are relatively well studied, little work has hours). We conducted vane trapping at L. and T. pratense L. were combined into examined temporal and spatial variability all sites during all sampling bouts. We a single category, as were Arnica sororia of native bees and their floral resources in also used pan traps and hand netting, Greene and A. chamissonis Less. Two riparian areas. The objective of our field two methods commonly used to sample taxa were identified to genus level only study was to describe the spatial and tem- native bees (Westphal et al. 2008), during (Descurainia and Penstemon). poral patterns in native bee communities some of the sampling bouts (Table 2). Pan and floral resources at the 12 Meadow traps were 6 oz cups in fluorescent blue, To standardize effort across sites and times Creek sites. These data will help address florescent yellow, or white. We placed pan when making spatial and temporal com- several questions that have significant im- traps filled with a solution of water and parisons of bee abundance, we calculated a plications for ungulate management. For dishwashing soap at each of the 12 sites catch per vane trap per hour rate, account- example, are there “hotspots” of bee and/or along five parallel 20-m transects. The ing for the number of daylight hours each floral abundance and diversity in riparian transects were positioned perpendicular trap was open. We report plant richness areas, either in time or space? Are bee to the stream and each separated by 15 m. as number of species and bee richness and floral resource availability generally The array of transects was centered on the as number of genera. SYSTAT Version correlated through time and space? Does blue vane trap located in each study site. 12.0 (2007) was used for all Pearson’s the composition of the bee and flowering Three pan traps (one of each color) were correlations and means are reported ± one plant communities vary seasonally? placed in the center of each transect and standard error. left open for approximately 1.5 days (X̅ = Field Study Methods 23.5 ± 0.7 daylight hours). Pan trapping Bee and plant community composition was not conducted in May. Eight sites were were characterized with non-metric mul- To address these questions, we sampled sampled with the method in June, and all tidimensional scaling (NMS) ordination native bees and floral resources at the 12 sites in July and September. Hand-netting using the abundance of taxa and a relative Meadow Creek sites four times in 2014 was also conducted at each site for 15−20 Sorenson’s distance measure. All counts (Table 1; Figure 1). Native bees were sam- minutes for a total of 740 minutes in 2014 were transformed (log(X+1)) before anal- pled using three methods: vane traps, pan (eight sites in May and June, 11 sites in yses. For ordination analyses, rare taxa traps, and hand-netting. Vane traps consist July, and all sites in September). Bees (<20 individuals of a species for plants
Table 2. Dates of four bouts of native bee sampling and blooming flower surveys at 12 study sites in riparian areas of Meadow Creek, USFS Starkey Experimental Forest and Range, Oregon in 2014.
Bout Vane Traps Pan Traps Hand-Netting Blooming Flower Surveys I 24 − 26 May N/A 25 − 26 May 25 − 26 May II 29 June – July 1 30 June – 1 July 30 − June 1 − July III 29 July – 1 August 29 July – 1 August 31 July – 1 August 31 July – 1 August IV 19 – 22 September 19 – 22 September 20 – 22 September 20 − 21 September
466 Natural Areas Journal Volume 36 (4), 2016 and <2 individuals of a genus for bees) plants found on our transects were likely to stems counted, with Achillea millefolium were eliminated because rare taxa often be important as floral esourcesr to bees. We L., Potentilla gracilis Douglas ex Hook., mask general patterns in community com- focused our review on literature from the and Myosotis stricta Link ex Roem. & position in multivariate analyses (McCune western US, particularly studies from the Schult. being the most common species and Grace 2002); 34 plant species and 14 Northwest. There were several limitations on riparian transects (Figure 2b). bee genera were included in ordination to our approach: (1) our review was not analyses. The best solution was determined exhaustive, (2) more studies are available Spatial variation in bee abundance and through 250 runs of randomized data and on certain taxa than others (e.g., elk diets flowering plants was high; the mean num- dimensionality was determined by evalu- are more studied than deer), (3) we gen- ber of bees and number of genera collected ating the relationship between final stress erally did not account for preferences in per site in vane traps varied considerably and the number of dimensions. We used ungulate diets since those data were seldom (Figure 3a, b), as did the mean blooming Pearson’s correlation coefficients to quanti- available, (4) technical bulletins that iden- plant stem count and plant species richness fy relationships between bee and blooming tify plant species important to bees may (Figure 3c, d). The high variability asso- plant species abundance and ordination underrepresent nonnative species because ciated with mean abundance and richness axis scores (McCune and Mefford 2006). these publications are meant to encourage per site of bees and blooming plants was, best management practices, which typically in large part, due to marked seasonal We used multi-response permutation pro- exclude planting nonnative species, (5) we fluctuations in abundance and richness cedures (MRPP) to determine whether did not review the primary literature for (Figure 4). The number of bees collected nonanemophilous plant and bee community bee-plant association data given the mag- per site was lowest in May and highest in composition varied significantly through nitude of that effort, and (6) we focused the July/Aug sampling bout (Figure 4a). time. MRPP is a multivariate, nonparamet- on the 74 flowering species found on our Genus richness showed similar trends as ric procedure for testing the hypothesis of transects during our 2014 sampling bouts. bee abundance, except that richness was no difference between two or more groups. Although additional species occur in our high in both the June/July and July/August MRPP calculates the mean within group study area, the most common flowering bouts (Figure 4b). In contrast, blooming distance and generates an expected distance forbs and shrubs should be represented in plant abundance and richness was lowest through permutation (McCune and Mefford our review (with the exception of some in September, and relatively high for all 2006). The P value generated by the test mass blooming shrubs like willows, which other sampling bouts (Figure 4c, d). is the probability of observing a within bloomed before our first sampling bout). group distance smaller than the observed distance due to chance alone. MRPP tests While sites were quite variable through also provide a measure of the effect size RESULTS time and space, sites that, on average, (A), which is one minus the ratio of the had greater abundance in floral resources observed mean within group distance to the Field Studies of Native Bee and Plant throughout the season also had, on av- expected within group distance. An effect Communities in Starkey Riparian erage, more bees throughout the season. size of 1 indicates that all items within each Areas Mean bee abundance per site was highly group are identical (i.e., the within group correlated with mean number of blooming distances are zero), a value of 0 indicates During four sampling bouts in 2014, we stems per site (r = 0.69, P = 0.01, n = 12). that the heterogeneity within group is no collected 2259 bees in 29 genera using vane However, this pattern was not observed different from that expected by chance, and traps (760 individuals), pan traps (1245 in the relationship between mean rich- a negative effect size indicates there is more individuals), and hand-netting (254 indi- ness in bee genera and blooming plant heterogeneity within groups than expected viduals). The nine most common genera richness per site (r = -0.09, P = 0.78, n by chance (McCune and Grace 2002). We comprised over 94% of all bees sampled, = 12), suggesting that sites that had more used a relative Sorensen’s distance measure with Bombus, Lasioglossum, and Halictus species of blooming plants through the in all MRPP analyses. the most common (Figure 2a). We observed season did not necessarily have more bee and collected one species of concern, the genera, on average. In addition, there was Literature Review Methods western bumble bee (Bombus occidentalis little correspondence in site quality with Greene). Bombus occidentalis is listed regard to bee or blooming forb and shrub To determine the potential degree of as imperiled on the Xerces Society’s abundance and richness from one sampling overlap between ungulate diets and floral Red List of bees (http://www.xerces.org/ bout to another. Sites with high abundance resources used by bees, we compared pollinator-redlist), although populations and richness of bees or blooming plants blooming plant data collected on riparian have been detected in nearby regions of did not necessarily have high abundance transects at Meadow Creek during four eastern Oregon (Rao et al. 2011). During and richness during the next sampling sampling bouts in 2014 with published the same four sampling bouts, we counted bout (Table 3). The only exception was literature on flowering forbs and shrubs 7644 blooming stems of 74 species on belt for blooming plant richness between the in deer, elk, and cattle diets. We also used transects (Table 1). The ten most common second and third bouts. This suggests that several technical reports to identify which species accounted for more than 70% of all one cannot generalize the relative value of
Volume 36 (4), 2016 Natural Areas Journal 467 Figure 2. Relative abundance of (a) genera of 2269 bees collected using vane traps, pan traps, and hand-netting, and (b) blooming forb species identified in four bouts of sampling in 2014 at 12 riparian study sites on Meadow Creek at the USFS Starkey Experimental Forest and Range in Oregon.
Figure 3. Mean (± one SE) (a) bee abundance, (b) bee genus richness, (c) floral resource abundance, and (d) floral resource species richness at each riparian site on Meadow Creek at the USFS Starkey Experimental Forest and Range in Oregon, averaged over all four sampling bouts in 2014. Sites are arranged from upstream (Site 1) to downstream (Site12).
468 Natural Areas Journal Volume 36 (4), 2016 Figure 4. Mean (± one SE) (a) bee abundance, (b) bee genus richness, (c) floral resource abundance, and (d) floral resource species richness for each sampling bout, averaged over all 12 riparian sites on Meadow Creek at the USFS Starkey Experimental Forest and Range in Oregon in 2014. a site at one point in time with regard to its of that variation (Table 4). Three genera (Megachile, Halictus, and Anthidium) were ability to support bees or floral resources (Bombus, Melissodes, and Anthophora) more common late in the season. MRPP to future times. were significantly positively correlated analysis showed that the differences in with Axis 1 (and, thus, more common generic composition through the season Not only did bee and blooming plant com- in mid-season sampling bouts 2 and 3) was statistically significant (A = 0.22,P munities show strong seasonal changes in (Table 4). One genus (Lasioglossum) was < 0.001). abundance and richness, but the composi- negatively correlated with Axis 1 (and, tion of communities also changed across thus, more common early and late in the Trends in the species composition of the season. Ordination results revealed that season, during sampling bouts 1 and 4) blooming forb and shrub communities bee composition of sites varied seasonally (Table 4). The three genera positively likewise showed strong seasonal changes. (Figure 5a). A three dimensional solution associated with Axis 2 (Andrena, Lasio- Ordination results revealed that blooming explained 84% of the variation in bee glossum, and Osmia) were more abundant species composition of sites varied season- community composition at the genus level, early in the season (during bout 1); the three ally (Figure 5b), and a one dimensional with the first two axes explaining most genera negatively associated with Axis 2 solution explained 75% of the variation in
Table 3. Seasonal correlations across 12 sites for blooming plant and native bee richness and abundance in riparian areas of Meadow Creek, USFS Star- key Experimental Forest and Range, Oregon in 2014. Correlation coefficients are Pearson’s and statistical significance at P = 0.01 is indicated with “**”.
Bout I vs. II Bout II vs. III Bout III vs. IV Mean Bee Abundance -0.11 -0.01 -0.014 Mean Bee Genus Richness 0.06 -0.01 0.25 Blooming Plant Abundance 0.34 0.24 -0.13 Blooming Plant Richness 0.34 0.78** 0.17
Volume 36 (4), 2016 Natural Areas Journal 469 Figure 5. Non-metric multidimensional scaling ordination results for all sites from bouts in 2014 for (a) Axes 1 and 2 for the 14 most common bee genera and (b) Axis 1 for the 34 most common blooming plant species on Meadow Creek at the USFS Starkey Experimental Forest and Range in Oregon. Only three of the four sampling bouts were included in the ordination for plants because, in September, most sites had no blooming flowers on transects and so could not be used in community composition analyses. See Table 4 for genus/species correlations and percent variation explained for each axis. community composition. Six plant species of those species matched at the species spatially and temporally overlap in riparian were positively associated with Axis 1 (and, level (Achillea millefolium, Thermopsis areas are late summer and early fall, when thus, were species that bloomed relatively montana Nutt., and Fragaria virginiana water and the presence of green forage late in the season) and 14 species were neg- Duchesne) (Table 1). However, many ri- in riparian areas attract ungulates, while atively correlated with Axis 1 (and, thus, parian plant species were recorded in diets, forage quality and quantity decrease in the bloomed earlier in the season) (Figure 5b; or as preferred plants by only one or two uplands. During these times, high ungulate Table 4). MRPP analysis showed that the of the focal groups (i.e., deer, elk, cattle, densities and broadly overlapping niches differences in blooming species composi- and bees) and approximately 12% of plant can potentially have large effects on forbs tion through the season was statistically species belonged to genera not identified in riparian areas (Stewart et al. 2011) with significant (A = 0.32,P < 0.001). as food for any ungulate species (Table 1). cascading effects on native bees.
Literature Review of Ungulate and DISCUSSION AND CONCLUSIONS The degree of dietary overlap among Bee Diets ungulates at Starkey with regard to forbs We were able to take advantage of a that occur in riparian areas was less clear. We examined 43 diet studies of elk, well-studied system at Starkey with much General diet preferences of each species seven studies of deer, and eight studies known about the population dynamics, have been described, with elk and cattle of cattle and compiled data on flowering behavior, diet, and habitat use of its key typically eating more graminoids and plants preferred by native bees from six ungulate species (Skovlin 1991; Holechek deer consuming more browse. However, technical guides (Table 1). Although elk et al. 1982a, b, c; Rowland et al. 1997; all three are known to strongly select for had the highest number of diet records in Wisdom et al. 2005; Stewart et al. 2003, particular forbs in certain circumstances, the literature that matched plant species 2011). Mule deer, elk, and cattle are all key yet data on this potential diet overlap had found on riparian transects at Starkey, the players, with elk being the most abundant not been synthesized. In addition, there relative percentage of species in ungulate wild ungulate grazer in the system and the was no existing data on native bee com- diets that were also identified as species largest consumer of biomass per individual munities and flowering shrubs and forbs in important to bees was approximately the due to their large body mass and presence riparian areas at Starkey. Thus, there were same across the three ungulates (ranging from spring to fall in the study area. Habitat obvious gaps in our knowledge about the from 60−64% generic overlap and 50−57% use by the three species is generally spa- Starkey system that we addressed with our species overlap) (Table 5). We identified tially separated, with elk and deer avoiding empirical study and review. 17 species that occurred in both a genus areas grazed by cattle, and deer more com- identified in the diets of all three focal un- mon in areas where elk are less abundant Our work quantifying spatial and temporal gulates and in a genus of plants identified (Skovlin et al. 1968; Johnson et al. 2000). patterns of bee and flowering plant commu- as important to bees. A subset of three Key periods in which the three species may nities in riparian areas at Starkey revealed
470 Natural Areas Journal Volume 36 (4), 2016 Table 4. Pearson correlation coefficients of bee genera and plant species that were significantly correlated with non-metric multidimensional scaling -or dination axes during sampling bouts at 12 riparian sites of Meadow Creek, USFS Starkey Experimental Forest and Range, Oregon in 2014. Correlations statistically significant atP = 0.05 are indicated with “*” and those significant atP = 0.01 are indicated with “**”. Bee axis 3, not shown, explained only 12% of variation in generic composition of bee communities.
Axis Bee Axis 1 Bee Axis 2 Plant Axis 1 Variation 40% 32% 75% Explained: Taxa and r Positive: Positive: Positive: Bombus 0.74** Andrena 0.41** Achillea millefolium 0.71** Melissodes 0.36* Lasioglossum 0.34* Perideridia gairdneri 0.65** Anthophora 0.35* Osmia 0.33* Solidago missouriensis 0.56** Symphyotrichum spathulatum 0.53** Negative: Negative: Monardella odoratissima 0.46** Lasioglossum -0.68** Megachile -0.62** Hypericum perforatum 0.43* Halictus -0.53** Anthidium -0.40** Negative: Myosotis stricta -0.81** Microsteris gracilis -0.70** Agoseris glauca -0.60** Penstemon sp. -0.64** Camassia quamash -0.60** Taraxacum officinale -0.64** Ranunculus uncinatus -0.48** Fragaria virginiana -0.42* Lomatium ambiguum -0.40* Thlaspi arvense -0.40* Thermopsis montana -0.38* Viola adunca -0.37* Viola nuttallii -0.37* several important points. First, as noted by that average floral resource abundance and richness were still relatively high in Dumroese et al. (this issue pps. 499-511), throughout the season is the best predictor September, when almost no blooming forbs are indeed the most common and for average native bee abundance, although plants were found on transects. This diverse flowering plants in riparian areas the same relationship did not hold for floral asynchronicity may partially be driven by at Starkey. Second, bee and blooming plant and bee taxa richness. differential responses of plants and bees to abundance and richness in riparian areas weather conditions. May is often cold and at Starkey varied spatially and temporally, Our data also suggests that bee and floral wet, and bees will delay emergence and/ with strong interactions between time and resources in riparian areas at Starkey are or decrease activity in inclement weather, space. For example, spatial patterns in bee temporally decoupled. Like Williams et regardless of how many floral resources and flowering forb and shrub abundance al. (2001) and Kimoto et al. (2012a), we are available (Michener 2007). Forb se- and richness were generally not temporally found that native bee communities showed nescence in late summer and fall is also stable. This means that sites with relatively strong seasonal fluctuations that did not a common phenomenon, yet many bee high diversity and abundance at one-time correspond with cooccurring changes in species are active in late summer and fall, period may have relatively low diversity floral resource abundance and richness. mating and preparing for overwintering. and abundance at other times. This lack of a For example, the lowest bee abundance and “hot spot” effect makes management based richness occurred in May, when blooming Our data also showed that the community on “snapshots” of bee or floral resources flower abundance and richness were rel- composition of bees and plants varied sub- ineffective. Instead, our results suggest atively high. Conversely, bee abundance stantially through the season. For example,
Volume 36 (4), 2016 Natural Areas Journal 471 Table 5. Summary of number of plant genera and species on riparian transects sampled in 2014 at the Starkey Experimental Forest and Range, north- eastern Oregon, that were identified in ungulate diets and as significant floral resources for bees (see Table 1), and their potential overlap.
# of Generic Matches # of Species Matches Generic Bee Species Bee with Riparian with Riparian Plant Overlap Plant Overlap Focal Group # of Studiesa Transects (%)b Transects (%)c (%)d (%)e Elk 43 58 (78%) 32 (43%) 35 (60%) 17 (50%) Deer 7 35 (47%) 14 (19%) 22 (63%) 8 (57%) Cattle 8 33 (45%) 12 (16%) 21 (64%) 6 (50%) Bees 6 40 (50%) 22 (30%) -- -- a Number of studies reviewed; bee studies were technical reviews. b Number (and percentage) of flowering forb and shrub species found on riparian transects that matched the genus identified in the diets of the focal group from literature review (Table 1). c Number (and percentage) of flowering forb and shrub species found on riparian transects that matched species identified in the diets of the focal group from literature review. d Number of species of flowering forbs and shrubs found on riparian transects corresponding to a genus recorded in the diet of the focal ungulate and also identified as an important bee plant. Percentages reflect the percent of all plant species documented in the ungulate’s diet that occur in genera identified as important bee plants in the literature. e Number of species of flowering forbs and shrubs found on riparian transects that were both recorded in the diets of the focal ungulate and identified as an important bee plant. Percentages reflect the percent of all plants documented in the ungulate’s diet that were identified as important bee plants in the literature.
Andrena, Osmia, and some Lasioglossum frequently found in the diets of all three overlap and generated more specific hy- were dominant early in the season; Bombus, species of ungulates with elk reported to eat potheses to test, it has several limitations. Melissodes, and Anthophora were domi- the greatest variety. The relative breadth of First, as discussed above, relatively few nant midseason; and Megachile, Halictus, elk diets may be an artifact of the fact that studies have examined deer and cattle diets and Anthidium were dominant late in the our sample size of elk diet studies was much relative to elk; therefore, concluding that season. These patterns have significant larger than of deer or cattle studies. Regard- elk have the greatest diet breadth relative implications for managing dietary overlap less, elk diets do include many forbs, with a to forbs is uncertain. Second, our sampling as members of each genus have their own relatively high overlap with plants thought period did not coincide with several mass degree of specialization (e.g., Bombus and to be preferred by native bees. However, blooming events of shrubs even though Osmia are frequent visitors of Astragalus dietary intake of forbs important to native these species are common in the area, and Penstemon, Ogle et al. 2011a). In bees varied among ungulate species, and including willow (Salix spp.) and common fact, seasonal variation in community so must be considered in combination with snowberry (Symphoricarpos albus). Both composition of native bee communities the temporal habitat use by ungulates, and are valuable forage for cattle (Holechek is expected, not only because of variation the phenologies of bees and flowers. Our et al. 1982a, b), deer (Peek and Krausman among taxa in their ability to tolerate review allows managers to quickly assess 1996), and elk (Cook 2002), as well as inclement weather (Goulson 2010), but flowering species and genera that may be bees. Bee taxa that specialize on these also because of variation in the degree of of particular concern with regard to the diet resources may be more influenced by deer specialization and the phenology of plants overlap of particular ungulates and native herbivory given their preference for grazing upon which specialists depend (Michener bees. For Starkey, September may be the on browse. In addition, our review does not 2007). Indeed, our study suggests that time of greatest overlap among the four reflect the relative preferences of ungulates floral resources also show strong seasonal focal groups. In fall, few floral resources for particular species in most cases, it only changes in composition, which may be a are available, the abundance and nutritive indicates whether the species or genus has strong driver of bee responses. value of forbs are low, ungulates are more been recorded in a diet. For example, one likely to concentrate near water, potential- species reported as consumed by all three Finally, our literature review allowed us to ly resulting in high ungulate densities in ungulates that is also a preferred bee plant delve more deeply into specific areas of riparian areas, and bee abundance is still is Thermopsis montana or mountain gold- dietary overlap with regard to forbs and relatively high. enbanner. However, T. montana is highly flowering shrubs among the four focal toxic to elk and other livestock (Burrows groups. Our review showed that forbs are While our approach identified potential and Tyrl 2001) and, therefore, is proba-
472 Natural Areas Journal Volume 36 (4), 2016 bly seldom consumed. Likewise, species riparian restoration projects, may decrease nipulations of native ungulates and cattle listed as bee friendly plants in technical the overall suitability of the habitat for will allow us to directly test hypotheses guides, such as Achillea millefolium or native bees if precautions are not taken about dietary overlap, including the rel- common yarrow, may only be useful to a (Hanula et al. this issue pps. 427-439), ative importance of particular herbivores. fairly small subset of bees (Dumroese et as well as increase the probability of use Studies that address these uncertainties are al. this issue pps. 499-511). Finally, the by deer and elk attracted to the plantings. currently underway at Starkey and should role of nonnative species in supporting lead to significant advances in our current native bees is still unclear (Hanula et al. understanding of the interaction of ungulate Future Research Directions this issue pps. 427-439), yet these species grazers and native bees in the future. are almost certainly underrepresented in The work presented here is the first pub- the technical guides reviewed. However, lished attempt to synthesize results from a ACKNOWLEDGMENTS some nonnative species in the region have diverse set of sources concerning potential been found to be used by a large number overlap of native bee floral preferences and Funding for this project was provided by of native bees (McIver et al. 2009; McIver ungulate diets. We used the Starkey Exper- the USDA Forest Service, Pacific North- and Erickson 2012). imental Forest and Range as a case study west Research Station. S. Roof received to illustrate an approach that combined funding from Oregon State University Management Implications empirical data on temporal and spatial Provost’s Branch Experiment Station variability in bee communities and their Experiential Learning Program and a stu- Our work at Starkey suggests the following floral resources in a riparian area with a dent grant from the Northwest Scientific steps for managers investigating the degree literature review aimed at identifying the Association. L. Smith was supported by a of ungulate and bee diet overlap in their relative importance of riparian plants found USDA National Needs Fellowship from the own systems: at Starkey as food for ungulates and bees. National Institute of Food and Agriculture. This approach can be adapted to other We thank K. Coe for his help in identifying 1. Assess the current situation. This systems and will allow managers to begin plant species and S. Clark and S. Burrows process includes considering the type and to systematically evaluate the magnitude for help in identifying bees. B. Dick, R. abundance of grazers and their spatial of potential dietary overlap between these Kennedy, and D. Rea assisted with logistics and temporal use of habitat in the area of groups and design management plans that for housing and other field support. The interest in addition to understanding the mitigate negative effects resulting from comments of two anonymous reviewers basic characteristics and phenology of overlap. Realized dietary overlap at Starkey improved this manuscript. the native bee and the floral communities. was estimated by considering the degree While identifying the key players in the of separation of the focal groups in space, bee community should be a top priority, time, and dietary preferences. Sandra (Sandy) DeBano is an Associate identifying any sensitive species will also Professor in the Department of Fisheries be important. This case study illustrates how existing and Wildlife at Oregon State University’s 2. Identify key areas of potential overlap information can be used to guide ungulate Hermiston Agricultural Research and based on ungulate movement patterns and management and restoration to enhance Extension Center in northeastern Oregon. phenology of flowering plants that support multiple ecosystem services, including Sandy’s research interests focus on inver- native bees. biodiversity conservation of pollinators. tebrate ecology and invertebrate-mediated 3. Consider management options. Chang- However, there are several areas of research ecosystem services including pollination, es in cattle management may be the most that can enhance our knowledge of ungulate food web provisioning, and decomposition. readily implemented solution, including herbivory impacts on native bees, both She works primarily in grasslands, riparian avoiding grazing in areas during sensitive through diet overlap and nonconsumptive areas, and agroecosystems. periods or when high degrees of overlap effects. First, the refinement of current, with bees are indicated. However, managers coarse-scale relationships between ungu- Samantha (Sam) Roof is an undergradu- should also consider wild ungulate man- late diets and floral preferences of bees is ate at Oregon State University planning agement when feasible. Fencing could also greatly needed. Significant gaps remain to graduate with a BS in Bioresource be used to protect particularly valuable bee in our understanding of floral preferences Research. Her interests include ecology habitat, although the costs of building big of specific bee species as well as of deer, and conservation, particularly by looking game fences would not be cost effective elk, and cattle. Additional details about at pollination and the impacts of native for any but the smallest scales. Restoration ungulate landscape movement through and exotic plant species. plantings represent another opportunity to the season are also needed, especially decrease overlap by providing bees with as climate change potentially magnifies Mary Rowland is a research wildlife preferred forage plants (Dumroese et al. conditions that affect spatial and dietary biologist with the USDA Forest Service, this issue pps. 499-511). However, planting overlap, such as drought stress (Roever Pacific Northwest Research Station, in flowering shrubs, as is common in many et al. 2015). Second, experimental ma- La Grande, OR. Her research centers on
Volume 36 (4), 2016 Natural Areas Journal 473 understanding the role of ungulates as in the Blue Mountains of Oregon and Wash- western United States: Hydrology, Ecology, ecosystem drivers and how anthropogenic ington: Evidence from existing literature. and Management. Lewis Publishers, Boca disturbance impacts ungulate distribu- General Technical Report PNW-GTR-316, Raton, FL. USDA, Forest Service, Pacific Northwest tion and performance. Current research Cook, J.G. 2002. Nutrition and food. Pp. Research Station, Portland, OR. includes evaluating impacts of riparian 259–349 in D.E. Toweill and J.W. Thomas, Boyd, R.J. 1970. Elk of the White River Plateau, eds., North American Elk: Ecology and restoration on a variety of responses, Colorado. Colorado Division of Game, Fish Management. Smithsonian Institution, such as ungulate distributions, fish habitat and Parks, Denver, CO. Washington, DC. quality, and native pollinators, as well Brazda, A.R. 1953. Elk migration patterns and Cook, R.C., J.G. Cook, R.A. Riggs, and L.L. as the cascading effects of ungulates on some of the factors affecting movements in Irwin. 2014. Habitat-nutrition relations of restoration plantings. the Gallatin River Drainage, Montana. The elk during spring through autumn in the Journal of Wildlife Management 17:9-23. Blue Mountains of Eastern Oregon and Lauren Smith is a graduate student in the Burrows, G.E., and R.J. Tyrl. 2001. Toxic Plants their implications for forest landscape man- Department of Fisheries and Wildlife at of North America. Iowa State University agement. Final Report, National Council Oregon State University. Lauren received Press, Ames, IA. for Air and Stream Improvement. USDA, a BA in Environmental Science from Whea- Burt, T., Jr., and G.H. Cates. 1959. Pecos elk Forest Service, Forestry and Range Science Laboratory, La Grande, OR. ton College, Massachusetts, and a MS reproductive and food habits studies. Federal in Ecology from The Pennsylvania State Aid Completion Report, Project W-93-R-1, Damiran, D. 2006. Influence of previous cattle WP-3, J-6, New Mexico Department of and elk grazing on subsequent diet quality University. Her research interests are Game and Fish, Albuquerque, NM. and nutrient intake rate of cattle, deer, and focused in restoration ecology including Cane, J.H. 1991. Soils of ground-nesting bees elk grazing late-summer mixed-conifer the impacts of grassland restoration on (Hymenoptera: Apoidea): Texture, moisture, rangelands. PhD diss., Oregon State Uni- native plant and invertebrate communities. cell depth and climate. Journal of the Kansas versity, Corvallis. Entomological Society 64:406-413. Deal, R.L., B. Cochran, and G. LaRocco. 2012. LITERATURE CITED Capp, J. 1967. Competition among bighorn Bundling of ecosystem services to increase sheep, elk, and deer in Rocky Mountain Na- forestland value and enhance sustainable tional Park, Colorado. MS thesis, Colorado forest management. Forest Policy and Eco- Ager, A.A., B.K. Johnson, J.W. Kern, and J.G. nomics 17:69-76. Kie. 2003. Daily and seasonal movements State University, Fort Collins. DeBano, L.F., S.J. DeBano, D.E. Wooster, and habitat use by female Rocky Mountain Carvell, C. 2002. Habitat use and conservation and M.B. Baker. 2003. Linkages between elk and mule deer. Journal of Mammalogy Bombus of bumblebees ( spp.) under different surrounding watersheds and riparian areas. 84:1076-1088. grassland management regimes. Biological Pp. 77–97 in M.B. Baker, P.F. Ffolliott, L.F. Anderson, C.C., W.I. Crump, and T.C. Baker. Conservation 103:33-49. DeBano, and D.G. Neary, eds., Riparian 1956. Food habits of antelope, elk, deer, Case, R.L., and J.B. Kauffman. 1997. Wild Areas of the Southwestern United States: and moose. Federal Aid Completion Report, ungulate influences on the recovery of Hydrology, Ecology, and Management. Project W-27R-9, WP-1, J-1. Wyoming willows, black cottonwood and thin-leaf Lewis Publishers, Boca Raton, FL. Game and Fish Commission, Laramie, WY. alder following cessation of cattle grazing DeBano, S.J., and D.E. Wooster. 2003. Insects in northeastern Oregon. Northwest Science Bartmann, R.M. 1983. Composition and quality and other invertebrates: ecological roles and 71:115-126. of mule deer diets on pinyon-juniper winter indicators of riparian and stream health. Pp. range, Colorado. Journal of Range Manage- Chapin, F.S.I., S.R. Carpenter, G.P. Kofinas, 215–236 in M.B. Baker, P.F. Ffolliott, L.F. ment 36:534-541. C. Folke, N. Abel, W.C. Clark, P. Olsson, DeBano, and D.G. Neary, eds., Riparian D.M.S. Smith, B. Walker, O.R. Young, F. Bengston, D.N. 1994. Changing forest values Areas of the Southwestern United States: Berkes, R. Biggs, J.M. Grove, R.L. Naylor, and ecosystem management. Society and Hydrology, Ecology, and Management. E. Pinkerton, W. Steffen, and F.J. Swanson. Natural Resources 7:515-533. Lewis Publishers, Boca Raton, FL. 2010. Ecosystem stewardship: Sustain- DeBano, S.J. 2006a. Effects of livestock graz- Bennett, E.M., G.D. Peterson, and L.J. Gordon. ability strategies for a rapidly changing 2009. Understanding relationships among ing on aboveground insect communities planet. Trends in Ecology and Evolution in semi-arid grasslands of southeastern multiple ecosystem services. Ecology Let- 25:241-249. ters 12:1394-1404. Arizona. Biodiversity and Conservation Christianson, D.A., and S. Creel. 2010. A review 15:2547-2564. Black, S.H, M. Shepherd, and M. Vaughan. of environmental factors affecting elk winter DeBano, S.J. 2006b. The effect of livestock 2011. Rangeland management for pollina- diets. The Journal of Wildlife Management tors. Rangelands 33:9-13. grazing on the rainbow grasshopper: 71:164-176. Population differences and ecological cor- Blood, D.A. 1966. Range relationships of elk Claar, J.J. 1973. Correlations of ungulate food relates. Western North American Naturalist and cattle in Riding Mountain National Park, habits and winter range conditions in the 66:222-229. Manitoba. Canadian Wildlife Service, Otta- Idaho primitive area. MS thesis, University wa. Wildlife Management Bulletin 1:1-62. Deschamp, J.A., P.J. Urness, and D.D. Austin. of Idaho, Moscow. 1979. Diets of mule deer from lodgepole Bohne, J.R. 1974. Food habits, seasonal distri- Clary, W.P., and W.H. Kruse. 2003. Livestock pine habitats. The Journal of Wildlife Man- bution, and habitat utilization of elk in the grazing in riparian areas: Environmental agement 43:154-161. South Fork of Fish Creek, Lolo National impacts, management practices, and man- Dumroese, R.K., T. Luna, J.R. Pinto, and Forest, Montana. MS thesis, University of agement implications. Pp. 215–236 in M.B. T.D. Landis. 2016. Forbs: The foundation Montana, Missoula. Baker, P.F. Ffolliott, L.F. DeBano, and D.G. of pollinators, monarch butterflies, and Bolon, N.A. 1994. Estimates of the value of elk Neary, eds., Riparian Areas of the South- greater sage-grouse in the western United
474 Natural Areas Journal Volume 36 (4), 2016 States. Natural Areas Journal 36:499-511 ment of Fish and Game, Helena. grasslands. Ecosphere 3:88. Accessed on (This issue). Hansen, R.M., and L.D. Reid. 1975. Diet overlap 13 July 2016
Volume 36 (4), 2016 Natural Areas Journal 475 McIver, J., R. Thorp, and K. Erickson. 2009. of Agriculture, Natural Resources Conser- framework for identifying indicators of eco- Pollinators of the invasive plant, yellow vation Service, Portland, OR. system contributions to human well-being. starthistle (Centaurea solstitialis), in Pickford, G.D., and E.H. Reid. 1943. Competi- Frontiers in Ecology and the Environment north-eastern Oregon, USA. Weed Biology tion of elk and domestic livestock for sum- 11:98-105. and Management 9: 137-145. mer range forage. The Journal of Wildlife Rosiere, R.E., R.F. Beck, and J.D. Wallace. [MEA] Millennium Ecosystem Assessment. Management 7:328-332. 1975. Cattle diets on semidesert grassland: 2005. Ecosystems and Human Well-Being: Picton, H.D. 1960. A comparison of the re- Botanical composition. Journal of Range Synthesis. Island Press, Washington, DC. sults of feeding site examinations with the Management 28:89-93. Michener, C.D. 2007. The Bees of the World. vegetative composition of the sites in the Rouse, R.A. 1957. Elk food habits, range use Second ed. Johns Hopkins University Press, Sun River Game Range. Pp. 9 in Wildlife and movements. Gravelly Mountains, Mon- Baltimore, MD. Investigation (Dist. 4): Big Game Surveys tana. MS thesis, Montana State University, and Investigations. Federal Aid Completion Bozeman. Mooney, H., A. Larigauderie, M. Cesario, T. Report, Project W-74-R-5, J-A-1, Montana Elmquist, O. Hoegh-Guldberg, S. Lavorel, Department of Fish and Game, Helena. Rouse, R.A. 1958. Elk investigations (elk-live- G.M. Mace, M. Palmer, R. Scholes, and T. stock relationships). Pp. 1–3 in Wildlife Yahara. 2009. Biodiversity, climate change, Pollinator Partnership. n.d.a. Selecting plants Investigations (Dist. 3). Federal Aid Com- and ecosystem services. Current Opinion in for pollinators. A regional guide for pletion Report, Project W-73-R-3, JA-2, Environmental Sustainability 1:46-54. farmers, land managers, and gardeners in Montana Department of Fish and Game, the Nevada-Utah mountains, semidesert, Helena. Mower, K.J., and H.D. Smith. 1989. Diet sim- coniferous forest, alpine meadow prov- ilarity between elk and deer in Utah. The ince including parts of Nevada, Utah, and Roever, C.L., T. DelCurto, M. Rowland, M. Great Basin Naturalist 49:552-555. Colorado. The Pollinator Partnership and Vavra, and M. Wisdom. 2015. Cattle grazing Nichols, L., Jr. 1957. Forage utilization by the North American Pollinator Protection in semiarid forestlands: Habitat selection elk and domestic sheep in the White River Campaign, San Francisco, CA. Accessed during periods of drought. Journal of Animal National Forest. MS thesis, Colorado State 13 July 2016
476 Natural Areas Journal Volume 36 (4), 2016 Research Station, Portland, OR. 2013. Plant materials for pollinators and Westphal, C., R. Bommarco, G. Carré, E. Skovlin, J.M., P.J. Edgerton, and R.W. Harris. other beneficial insects in eastern Utah and Lamborn, N. Morison, T. Petanidou, S.G. 1968. The influence of cattle management western Colorado. Plant Materials Technical Potts, S.P.M. Roberts, H. Szentgyörgyi, T. on deer and elk. Transactions of the North Note 75, USDA, Natural Resources Conser- Tscheulin, B.E. Vaissière, M. Woyciechows- American Wildlife and Natural Resources vation Service, Boise, ID – Salt Lake City, ki, J.C. Biesmeijer, W.E. Kunin, J. Settele, Conference 33:169-181. UT – Lakewood/Denver, CO. and I. Steffan-Dewenter. 2008. Measuring bee diversity in different European habitats Stark, R.H., Jr. 1973. Elk-cattle interrelations on Tubbesing, C., C. Strohm, S.J. DeBano, N. and biogeographical regions. Ecological the Colockum Creek watershed. MS thesis. Gonzalez, C. Kimoto, and R.V. Taylor. 2014. Monographs 78:653-671. Washington State University, Pullman. Insect visitors and pollination ecology of Wickstrom, M.L., C.T. Robbins, T.A. Hanley, Stephen, W.P., and S. Rao. 2005. Unscented Spalding’s catchfly (Silene spaldingii) in D.E. Spalinger, and S.M. Parish. 1984. color traps for non-Apis bees (Hymenoptera: the Zumwalt Prairie of northeastern Oregon. Food intake and foraging energetics of elk Apiformes). Journal of the Kansas Entomo- Natural Areas Journal 34:200-211. and mule deer. The Journal of Wildlife logical Society 78:373-380. [USFS] US Department of Agriculture, Forest Management 48:1285-1301. Stevens, D.R. 1966. Range relationships of elk Service. 2012. Meadow Creek large woody Williams, N.M. 2011. Restoration of nontarget and livestock, Crow Creek Drainage, Mon- debris project Phase II proposal, USDA species: Bee communities and pollination tana. The Journal of Wildlife Management Forest Service, Wallowa-Whitman Nation- function in riparian forests. Restoration 30:349-363. Ecology 19:450-459. al Forest, La Grande Ranger District, La Stewart, K.M., R.T. Bowyer, B.L. Dick, and J.G. Grande, OR. Williams, N.M., R.L. Minckley, and F.A. Sil- Kie. 2011. Effects of density dependence veira. 2001. Variation in native bee faunas Van Dyne, G.M., and H.F. Heady. 1965. on diet composition of North American elk and its implications for detecting community Botanical composition of sheep and cattle Cervus elaphus and mule deer Odocoileus changes. Conservation Ecology 5:7. Ac- diets on a mature annual range. Journal of hemionus: An experimental manipulation. cessed 13 July 2016
Volume 36 (4), 2016 Natural Areas Journal 477