Skip common site navigation and headers
United States Environmental Protection Agency
Great Lakes Ecosystems
Begin Hierarchical Links EPA Home > Great Lakes EcosystemsUpland Ecosystems > 1993 Oak Savanna Conferences > John A. Shuey
Aquatic Ecosystems
EPA Region 5 Critical Ecosystems
Ecosystem Funding
Great Lakes Basin Ecosystem
Great Lakes Biological Diversity
Green Landscaping
Rivers and Streams
Upland Ecosystem



1993 Proceedings of the Midwest Oak Savanna Conferences1993 Proceedings of the Midwest Oak Savanna Conferences



John A. Shuey
The Nature Conservancy
1330 West 38th Street
Indianapolis, IN 46208
Tel: (317) 923-7547 

(Formerly with the Great Lakes 
Environmental Center
Traverse City, MI) 




Like most federally endangered insects, the Karner Blue (Lycaeides melissa samuelis Nabokov) has been negatively impacted by habitat loss. The greatest impact has been from the elimination of wildfire from oak barrens/savanna communities. The early successional habitats required by this butterfly (characterized by fire tolerant and shade intolerant plant species) have succumbed to successional communities which no longer meet Karner blue habitat requirements. Because most oak barrens/savanna communities are suffering from the effects of fire suppression, optimal Karner Blue habitats are generally limited in size and widely dispersed. This combination of reduced optimal habitat patch size, combined with increased distance between optimal habitat patches has destroyed the metapopulation dynamics of the Karner Blue. Loss of fire is currently causing the downward spiral of several metapopulations of the Karner Blue, even as regional attempts to restore these ecosystems proceed. Preserve managers and stewards must re-establish the ecosystem-level processes which created the barrens and savanna communities required by this butterfly. 


The recent listing of the Karner Blue Butterfly (Lycaeides melissa samuelis Nabokov) as an endangered species (Clough, 1992) has increased interest in managing and restoring populations of this charismatic invertebrate. The Karner Blue and other lepidopteran species are rapidly becoming symbols for restoring and conserving the savanna ecosystems which occur on well drained sands in the Great Lakes Region and New England. In the Midwest, the habitat types which support this butterfly are generally characterized as oak barrens/savanna communities. Each oak barrens/savanna ecosystem owes its existence to the unique interplay between past biogeographic events, edaphic conditions and past and present disturbance regimes. The same dynamic processes which produced the unique botanical communities also produced a highly specialized community of invertebrates adapted to this regime. Because of their general biological requirements, invertebrates are often closely linked to a few key ecological resources, such as specific soil types, edaphic conditions or to an individual hostplant species or genus. 

The importance of oak barrens/savanna habitats to invertebrates is illustrated by the Lepidoptera, a species rich group which includes the butterflies and moths. In Ohio (the only midwestern state with a completed state-wide survey of all Lepidoptera species), the Oak Openings, Ohio's only oak barrens/savanna community, supports the largest assemblage of imperilled butterflies and moths in the state. For example, five species of imperilled butterflies and 17 species of owlet moths (Noctuiidae) occur in the Oak Openings, representing approximately 4% and 3% respectively, of the resident species in Ohio (Shuey et al, 1987a,1987b; Metzler and Lucas 1990; Iftner et al, 1992; Rings et al, 1992). The maintenance of this ecosystem is vital for the preservation of lepidopteran biodiversity as well as for other lesser known plants and animals in Ohio. The importance of oak barrens/savanna communities to biodiversity maintenance in the other Great Lakes States is certainly similar to the situation in Ohio.

The decline of oak barrens/savanna lepidopteran communities can be attributed to several factors, but habitat loss, the disruption of ecosystem level processes and patch dynamics, as well as the collapse of metapopulation dynamics of many species are the primary contributors. The following sections discuss these intertangled processes, and the management implications and problems associated with each process as they relate to the Karner Blue. For ecological information regarding other imperiled midwestern lepidopteran species, see the species and habitat accounts in Rings et al (1992) and Iftner et al (1992). 


Habitat loss is the most easily implicated factor contributing to the decline of most imperiled invertebrate species (Hafernik, 1992); the Karner Blue is no exception. To be locally persistent, populations of the Karner Blue require relatively large stands of the hostplant, blue lupine (Lupinus perennis L.)(Opler and Krizek, 1984). Habitats which support the butterfly are generally open, sunny habitats with scattered trees and shrubs, but which are dominated by grasses and other herbaceous species growing in well drained, sandy soils: ie, healthy oak barrens/savanna communities. Oak barrens/savanna loss can be attributed to several factors, ranging from outright destruction to more subtle secondary impacts such as the encouragement of forest growth in areas of urbanencroachment. 

Oak barrens/savannas were subject to the same trends which altered almost every ecosystem in eastern North America. The expansion of agriculture was largely a process of trial and error: farming sand barrens was an error. In the trial process, many habitats were altered or destroyed and the local hydrology was often modified. This trial process came to a screeching halt during the prolonged drought of the 1930's, when it finally became apparent that the infertile soils of these communities could not support sustainable agricultural production.

The unfortunate location of many regional oak barrens/savanna communities also contributed to their destruction, especially in New England. For example; the Albany Pine Barrens sit adjacent to the city of Albany, New york, and the expansion of the city has, and still is contributing to the urbanization of this ecosystem; The Oak Openings ecosystem in Ohio is suffering the same fate at the hands of Toledo; And, the complex dune communities which once lined southern Lake Michigan have been almost eliminated by the industrialization and urbanization. Other examples are depressingly common. 

On a wider scale, the infertility of the sand soils themselves has led to the destruction of sand barrens communities. Many abandoned farms located in oak barrens/savanna ecosystems eventually reverted to federal and state ownership (via tax defaults), largely to become public forest land. Because the preservation of non-forest communities was not a high priority of national or state forests in the 1930's through the present, many oak barrens/savanna communities were converted into 'productive' use by conversion to pine plantations. These magnificent monocultures stand testament to the incomplete and short-sighted ecological planning of past eras. Degraded barren communities continue to be primary targets when siting new developments such as industrial parks and residential communities, possibly because the cost associated with acquiring barrens-land is less than the cost for purchasing productive agricultural lands. 


Closely related to the impact of habitat loss is the elimination of ecosystem level processes. Oak barrens/savanna communities are among the most dynamic in the Midwest. The open habitats which support the Karner Blue were originally maintained by a steady procession of wildfires, which killed woody invasive plants while favoring fire-adapted dune and savanna communities. Without fire, shade tolerant and fire sensitive species increase in density, and open barrens and savanna species decline. 

Functional oak barrens/savanna communities are in a constant but dynamic flux. Succession pushes the community towards an association characterized by fire intolerant woody and shade tolerant herbaceous species, while fire realigns the community towards one of fire tolerant and shade intolerant species. The original patch dynamics of these communities was in constant flux, and individual sites supported communities that reflected recent disturbance history. Although fire may have been a yearly occurrence within oak barrens/savanna ecosystems, the spacial distribution of the fire was less predictable. For example, in the Albany Pine Barrens the point fire frequency may have ranged between 6 to 18 years, with a likely average frequency of once every 10 years (Givnish et al, 1988). Thus, these communities were composed of a constantly changing patch-work of habitats, reflecting the hit or miss nature of recent wildfires. Interdispersed through this patch-work were the recently disturbed habitats which supported populations of the Karner Blue.

Modern culture has traditionally abhorred wildfire because of its perceived destructive nature. This viewpoint was promoted to 'cultural truth' by our government. Oak barrens/savanna ecosystems which are adjacent to urbanized areas are subject to routine fire suppression. Similarly, state and national forests routinely suppress wildfires occurring on their lands. With few positive attributes to associate with wildfire, active ecosystem management still remains controversial for the general public. Thus, society generally deprives these ecosystems of the very force that created them, a predictable and frequent fire disturbance regime.

Urban and agricultural encroachment eliminate habitat. Equally important they fragment oak barrens/savanna communities by inserting non or less-flammable landuses into these highly flammable ecosystems (Givnish et al, 1988). Such barriers limit the occasional wildfire to small land tracts, reducing the potential for naturally spreading wildfire to maintain the ecosystem in an early successional state. In addition, urban encroachment increases the difficulty of using controlled burns to manage oak barrens/savanna communities because of the liability and perceived danger and nuisance to residents.

Without the influence of a fire regime, oak barrens/savanna communities have succumbed to other community types. The impact of fire suppression on these communities has been as great or greater than outright habitat destruction. For example, these community types are critically endangered and the Karner Blue is extirpated from Ohio's Oak Openings, despite the 'preservation' of over 9,000 acres by state, local and private organizations. Most of the habitats in the Oak Openings which once may have supported oak barrens/savanna have converted to young oak forest. Similarly, what little remains of the Albany Pine Barrens in New York is now largely overgrown (Givnish et al, 1988). At its worst, land is dominated by black locust forest; at its best, dense scrub oak brushland is dominant. Both of these regional ecosystems will require massive management and restoration to become functional again. 


The plants and animals that together form oak barrens/savanna communities are adapted to the ecosystem level processes which originally structured these communities. To be regionally persistent in this dynamic ecosystem, invertebrates such as insects must cope with both the ecosystem patch dynamics as well as the factors that drive the patch dynamics. In simple terms, insect populations must shift locations as quality habitats disappear or become available and they must be able to survive wildfire, either directly or indirectly.

While healthy metapopulations of the Karner Blue may seem to occupy entire oak barrens/savanna ecosystems, individual sub-populations are usually highly localized and are often isolated from neighboring populations by 'barriers' of unsuitable habitat. These isolated sub-populations are extremely vulnerable to extinction from both community succession and ecosystem disturbance regimes. Optimal Karner Blue habitat is best characterized as open, oak barrens/savanna habitat. That is, Karner Blues use habitats recently disturbed by fire. In the absence of fire, these communities shift towards species adapted to more overgrown, shade tolerant conditions, and habitat quality for the Karner Blue decreases. Wildfire 'resets' habitat suitability to favorable conditions for the butterfly. To persist, populations of this butterfly must be adapted to the patch dynamics of the oak barrens/savanna community, which is very dynamic and based upon frequent and regular disturbance by wild fire. 

The Karner Blue is not adapted to directly survive fire. The very mechanism which is so critical for creating and maintaining habitat for this species, also kills all life stages of the butterfly (there is circumstantial evidence from Michigan that the Karner Blue may occasionally survive cool, low fuel-load fires, but this needs to be better documented). Recently burned habitats must be colonized or recolonized by individuals immigrating from nearby or adjacent habitats. Confounding this issue is the limited dispersal ability of the Karner Blue. Givnish et al (1988) estimate that maximum dispersal distance for colonization of unoccupied habitats is approximately 0.5 miles. This matches closely the results obtained with the ecologically similar and phylogenetically related butterfly, Plebejus argus in North Wales (Thomas and Harrison, 1992). Metapopulation dynamics over a seven year period indicated that the likelihood of colonization of suitable habitats decreases rapidly for habitats more than 1 km away from potential source populations. These authors concluded that if the continuity of suitable habitat is broken, entire metapopulations of P. argus are likely to collapse. 

Because most oak barrens/savanna communities are suffering from the effects of fire suppression, optimal Karner Blue habitats are generally limited in size and widely dispersed. This combination of reduced optimal habitat patch size combined with increased distance between optimal habitat patches has destroyed the metapopulation dynamics of the Karner Blue. Suitable but unoccupied habitats may not have a nearby Karner Blue source population from which colonization is possible. Likewise, occupied habitats must be recolonized following fire; recolonization becomes less likely as the distance separating occupied habitats increases. In effect, the rate of localized population extinction has been accelerated by declining habitat suitability and size, while the odds of new colonization events have declined as optimal habitats become increasingly fragmented. This disruption of metapopulation dynamics is currently causing the downward spiral of several metapopulations of the Karner Blue, even as regional attempts to restore these ecosystems proceed. 


Quantification of optimal Karner Blue habitats have not been adequately assessed (but see efforts in this publication). However, given our knowledge of the biology of this butterfly, it is possible to make broad generalizations which should be factored into long-term management strategies. Four habitat suitability componentsshould be considered during management plan development: hostplant abundance and distribution, ant/larva interactions, adult resources, and microenvironment diversity (i.e., habitat structure).

Hostplant abundance and distribution: 
Populations of the Karner Blue require fairly extensive and dense stands of lupine. In habitats which support Karner Blue populations, lupine is generally common, numbering from a few hundred to several thousand stems. Most sites supporting this butterfly are relatively free of shrubs, and have a continuous, low-growing ground cover dominated by native grasses. Lupine is generally distributed throughout the open habitats, but also into the edges of adjacent forests and shrub communities. However, ovipositing Karner Blues will apparently not utilize lupine that is densely shaded by shrubs or growing in dense herbaceous vegetation. 

The Karner Blue will utilize lupine growing in open forest/dense savanna (÷50% canopy cover) if the herbaceous layer is open. In these situations, the adults are generally found in light gaps, but how they exploit the lupine throughout this habitat type is poorly understood.

Ant/larva interactions:
The larvae of Lycaenidae butterflies often form mutualistic ecological associations with various ant species. Ants are attracted by a honey-like substance secreted by lycaenid larvae. Ants defend this 'resource', presumably increasing the survival of larvae by discouraging predators and parasitoids which might otherwise attack the larvae. The association between ants and Karner Blue larvae is apparently facultative, and recent studies have indicated that the benefit to some butterfly larvae, including the Karner Blue, may be subtle and difficult to document (Peterson 1993: Savignano 1987). However, as more data becomes available, the role of this mutualist relationship may become more apparent and management plans may need to be altered.

Adult nutrient resources: 
In the few butterfly species studied, the availability of adult nutrient resources can have a dramatic effect on reproductive success (eg, Lederhouse et al, 1990; Pivnick and McNeil, 1985). There is no reason to suspect that adult resources play a lesser role for the Karner Blue. The most obvious adult resource is the availability of flower nectar. For the spring brood, most Karner Blue habitats support a variety of wild flowers which are suitable resources, including lupine itself. However, the summer brood may have fewer resources available, especially in disturbed ecosystems which may have regenerated from abandoned agricultural land. Perhaps the most frequent and reliable native, non-weedy summer nectar sources in the Midwest are butterfly weed (Asclepias tuberosa L.) and New Jersey tea (Ceanothus americanus L.). If summer nectar sources are absent, butterfly weed or other appropriate native nectar producing forbes should be reestablished in Karner Blue habitats.

To become reproductively competent, adult males of some butterflies may also require salts and other substances which they obtain by imbibing moisture from damp soil or sand (Lederhouse et al, 1990). Male Karner Blues are often observed imbibing moisture from damp sand, forming groups called 'mud-puddle clubs'. The importance of this resource needs to be quantified for Karner Blues.

Microenvironment diversity: 
The important role of microclimate diversity in the long-term maintenance of endangered invertebrate populations has been documented by Dobkin et al (1987) and Weiss and Murphy (1990). These authors have demonstrated that habitats which provide a variety of microenvironments buffer invertebrate populations against the negative impacts of the extremes of seasonal weather by providing thermal or edaphic escapes from extreme conditions. 

For the Karner Blue, hostplant quality may become an issue in drought years as well as in abnormally wet years. By assuring that lupine is found in a variety of relatively sunny exposures, the effects of both extremes may be moderated. For example, in hot drought years, highly exposed lupine plants may become senescent before summer brood larvae have completed development or before summer brood adults can oviposit. Lupine in partially shaded situations persists longer during drought years, providing hostplant resources for both larvae and adults. Similarly, during cool wet summers, lupine is susceptible to a fungus (mildew) which can also result in early lupine senescence and lowered hostplant quality. Plants growing in more exposed areas may be less susceptible than plants which are partially shaded. 

Thus, populations with access to hostplants growing only in a limited variety of microenvironments are likely to be susceptible to stress from prolonged periods of environmental extremes. By developing and maintaining habitats which provide a variety of microenvironments, it may be possible to buffer Karner Blue populations against year-to-year environmental fluctuations. 


For the Karner Blue, the interplay between habitat suitability, habitat distribution and patch dynamics, metapopulation dynamics and metapopulation persistence is complex. This is best illustrated by the historical distribution of the butterfly itself. The ecosystems known to support metapopulations of this butterfly are generally large, measured in the tens-of-thousands of acres. Smaller sand barren/oak savanna complexes are less likely to have supported Karner Blues in historic times. This is probably a reflection of the element of chance involved with the interplay between the ecosystem processes which maintain habitat quality and the metapopulation dynamics of the species. The larger the ecosystem, the better are the odds that all the pieces fall together and populations persist. Smaller ecosystems may have provided fewer opportunities for population persistence, and Karner Blue populations did not persist to historic times. As Givnish et al (1988) poetically state, persistent populations exist as a "flickering mosaic of Karner Blue populations, with some going extinct in a given area as others are being founded on sites recently burnt by colonists from sites burnt a somewhat longer time ago." In smaller ecosystems, these populations may simply flicker out.

Given that almost every oak barrens/savanna community in the Midwest must now be actively managed to persist, much of the element of chance can be removed from Karner Blue management. With intensive management, carefully planned burn units and suitable habitats, Karner Blue populations should be manageable on preserves as small as 200 acres. At this scale, management would have to be almost mechanical, with approximately 10-15% of the land-base burned annually, and the burn units configured to provide adequate dispersal opportunities for Karner Blues. Larger management areas could be managed less mechanically, but would still require carefully planned management activities. Small management units could be used to establish 'core' populations within larger ecosystem management areas, from which dispersing butterflies could become more widely established. 

Finally, to protect against catastrophy, several independent sets of Karner Blue populations should be maintained in each oak barrens/savanna ecosystem. Because of the flammable nature of these ecosystems, true wildfires which consume thousands of acres at one time are a reality. Because individual Karner Blue populations may succumb to such an event, independent 'core' populations should be dispersed to ensure that single catastrophic events do not eliminate entire metapopulations. 

Preserve managers and stewards must struggle to re-establish the processes which created the barrens and savanna ecosystems they manage. If the Karner Blue is to survive, we must literally take it back to the big dance, where metapopulations swirled with patch dynamics to the music of fire. By managing remnant barrens and savanna communities in light of large-scale ecosystems processes, it should be possible to preserve not only the Karner Blue, but the untold other inconspicuous life-forms adapted to these ever changing ecosystems. 


I thank the following groups and organizations for providing the access, opportunity and funding for my forays into Karner Blue habitats: The Nature Conservancy (Ohio Chapter), Ohio Department of Natural Resources (Division of Natural Areas and Preserves and Division of Wildlife), Toledo Metro Park District, Michigan Department of Natural Resources, and most importantly, Save the Pine Bush, Incorporated (Albany, New York). Dr. Thomas Givnish, University of Wisconsin, provided valuable comments on an early draft of this manuscript. 


Clough, M. W. 1992. Endangered and threatened wildlife and plants: determination of endangered status for the Karner Blue butterfly. Federal Register 57 (240): 59236-59244.

Dobkin, D. S., I. Oliviera and P. R. Ehrlich. 1987. Rainfall and the interaction of microclimate with larval resources in the population dynamics of checkerspot butterflies (Euphydryas editha) inhabiting serpentine grassland. Oecologia 71: 161-166.

Givnish, T. J., E. S. Menges and D. Schweitzer. 1988. Minimum area requirements for long-term conservation of the Albany Pine Bush and the Karner Blue butterfly: an assessment. Report submitted to the City of Albany, New York. 

Hafernik, J. E. Jr. 1992. Threats to invertebrate biodiversity: implications for conservation strategies. in P. L. Fiedler and S. K. Jain, eds. Conservation Biology: the theory and practice of nature conservation, preservation, and management. Chapman and Hall, New York. 507 p.

Iftner, D. C., J. A. Shuey and J. V. Calhoun. 1992. Butterflies and skippers of Ohio. Ohio Biol. Surv. Bull. New Series Vol. 9, No. 1. 212 p.

Lederhouse, R. C., M. P. Ayres and J. M. Scriber. 1990. Adult nutrition affects male virility in Papilio glaucus L. Functional Ecol. 4: 743-751.

Metzler, E. H. and V. P. Lucas. 1990. An endangered moth in Ohio, with notes on other species of special concern (Lepidoptera: Saturniidae, Sphingiidae, Notodontidae, and Arctiidae). Ohio J. Sci. 90: 33-40.

Opler, P. A. and G. O. Krizek. 1984. Butterflies east of the Great Plains. Johns Hopkins Univ. Press. Baltimore, Maryland. 294p.

Peterson, M. A. 1993. The nature of ant attendance and the survival of larval Icaricia acmon (Lycaenidae). J. Lepid. Soc. 47: 8-16.

Pivnick, K. A. and J. N. McNeil. 1985. Effects of nectar concentration on butterfly feeding: measured feeding rates for Thymelicus lineola (Lepidoptera:Hesperiidae) and a general feeding model for adult Lepidoptera. Oecologia 66: 226-237.

Rings, R. W., E. H. Metzler, F. J. Arnold and D. H. Harris. 1992. The owlet moths of Ohio: Order Lepidoptera, Family Noctuiidae. Ohio Biol. Surv. Bull. New Series Vol. 9, No. 2. 219 p.

Savignano, D. A. 1987. The association of Lycaeides melissa samuelis (Lepidoptera:Lycaenidae), the Karner Blue butterfly, with attendant ants. Report to the New York Dept. of Environ. Conserv., Endangered Species Unit.

Shuey, J. A., J. V. Calhoun, and D. C. Iftner. 1987a. Butterflies that are endangered, threatened, and of special concern in Ohio. Ohio J. Sci. 87: 98-106.

Shuey, J. A., E. H. Metzler, D. C. Iftner, J. V. Calhoun, J. W. Peacock, R. A. Watkins, J. D. Hooper, and W. F. Babcock. 1987b. Status and habitats of potentially endangered Lepidoptera in Ohio. J. Lepid. Soc. 41: 1-12.

Thomas, C. D. and S. Harrison. 1992. Spacial dynamics of a patchily distributed butterfly species. J. Animal Ecol. 61: 437-446.

Weiss, S. B. and D. D. Murphy. 1990. Thermal microenvironments and the restoration of rare butterfly habitat. in Berger, J. J. ed. Environmental restoration: science and strategies for restoring the earth. Island Press, Wash. D.C. 398p. 


Begin Site Footer

EPA Home | Privacy and Security Notice | Contact Us