The Diana fritillary (Speyeria diana) and great spangled fritillary (S. cybele): dependence on fire in the Ouachita Mountains of Arkansas

The Diana fritillary (Speyerio dinna), a species of conservation concern throughout its range, and the great spangled fritillary (S. cybele) both occur in the Ouachita Mountains of west-central Arkansas and eastern Oklahoma. Both species depend on abundant, high quality nectar resources to support pupulations. Decades of intense fire suppression have greatly altered vegetation structure throughout the Ouachita Mountains resulting in profound ecalogical changes. In an effort to restore pre-European ecological relationships and biodiversity, managers are restoring shortleaf pine-bluestem cornrnunities with the use of frequent prescribed fire. Restored sites support greater abundances of nectar resources and Speyeda dtana and S. cybele than fire suppressed controls. Additional key words: Interior Highlands, surveys, restordtion The Diana fritillary, Speyeria diana (Cramer) is a species of conservation concern thropghout its range in the eastern United States. It is cun but habitat alteration including loss of forest habitat (Allen 1997), harvest of old-growth forests (Hammond and McCorkle 1983), and loss of nectar plants (Moran and Baldridge 2002) have been proposed. In the Interior Highlands, Carleton and Nobles (1996) reported records of S. diana from 11 counties in Arkansas, Missouri, and Oklahoma since 1980 and an additional 10 counties prior to 1980. Moran and Baldridge (2002) reported recent records for an additional seven counties in Arkansas, and our records add an additional three counties (reported in Moran and Baldridge 2002). Based on these records, S. dhna remains widespread in the Ouachita Mountains of Arkansas and Oklahoma and perhaps less so in the Ozark Mountains of Arkansas and Missouri. The great spangled fritillary, S. cybele (Fabricius), is widespread in the eastern United States and Canada (Opler and Krizek 1984). It remains relatively common in a variety of habitats throughout most of its extensive range and is common throughout the Ouachita Mountains of Arkansas. 'Research Associate. Florida State Collection of Arthropods, DPI, FDACS, Gainesville, Florida 32614, USA The Ouachita Mountains physiographic region encompasses an area of 3,237,600 ha of east-west oriented ridges and valleys with elevations ranging from 150 to 820 111 (Bukenhofer and Hedrick 1997). Historically, much of the region (especially the more xeric southand west-facing slopes) burned on a regular basis during fires ignited by lightning and Native Americans (Foti and Glenn 1991, Masters et al. 1995). These frequently burned forests consisted of open, pine-dominated canopies, sparse midstories, and a diverse understory of grasses and forbs (Featherstonhaugh 1884, du Pratz 1975, Nuttall 1980, Foti and Glenn 1991, Masters et al. 1995). In the absence of frequent fire, these forests tend to succeed to an oak (Quernrs spp.)-hickory (Carya spp.) community (Neal and Montague 1991). Managers of the Ouachita National Forest have been in the process, since 1979, of restoring the firemaintained shortleaf pine (Pinus echinata P. Mill.)bluestem grass (Schizachrium spp.) ecosystem on approximately 48,706 ha primarily on the Poteau Ranger District in west-central Arkansas (Bukenhofer and Hedrick 1997). The primary management tool used to restore shortleaf pine-bluestem habitat is prescribed fire. Restoration is being conducted to restore habitat for the federaIly endangered redcockaded woodpecker, Picoida borealis (Viellot), and to restore what is thought to be the original .forest structure (Foti and Glenn 1991, Bukenhofer and Hedrickl997). A number of authors have examined the effect of these restoration efforts on a variety of taxa (Masters et d. 1998,2002; Sparks et al. 1998,1999; Thill et al. 2004). As part of extensive studies on the effects of restoration of fire-maintained pine-bluestem habitats on lepidopteran communities, we conducted censuses of butterflies and nectar resources on (treatment) and unrestored (control) plots (Thill et al. 2004). Here we report results for S. diana and S, cybele and their nectar resources in relation to ecosystem restoration and the prescribed fire regime. STUDY AREA AND METHODS This research was conducted on the Poteau District (344S1N, 94O15'W) of the Ouachita National Forest in west-central Arkansas. The Ouachita Mountains, especially within the Ouachita National Forest, is predominately forested. North-facing slopes are dominated by oaks (Quercus spp.), hickories (Caya spp.), and a diverse array of additional species, southfacing slopes by shortleaf pine and a varying mixture of angiosperms. Landscape scale restoration was initiated in 1979, formally incorporated into the Ouachita National Forest Plan in 1996, and currently projects the restoration of 48,706 ha (7.3% of the Forest) to a shortleaf pinebluestem condition thorlght to mimic the pre-European forest structure (U. S. Forest Service 1996, Bukenhofer and Hedrick 1997). At the initiation of our studies, approximately 9,071 ha had been in a 42,148 ha landscape on the Poteau Ranger District (W. G. Montague pers, corn.). Restoration was accomplished by selective thinning of overstory trees, removal of most midstory vegetation, and implementation of a threeyear prescribed fire regime. Treatment blocks for restoration are typically 10-45 ha in size, however blocks are combined for purposes of prescribed burning, and individual prescribed fires ranged up to 1,300 ha. We use restored in a relative sense and recognize that stands are on a trajectory towards an ecological state thought to mimic pre-European conditions. We established nine study plots in stands and three in control stands. A11 stands had received initial thinning harvests and a minimum of four prescribed fires prior to the initiation of this study. Prescribed fire was applied to stands during March and April with few exceptions, all prior to the initiation of this study. During the course of this study prescribed fire was applied to three of the stands each year. Consequently, in any given year three stands represented first, second, and third growing seasons post-fire. For purposes of analysis, data were grouped by post-fire growing season. Adult butterflies (Papilionoidea and Hesperioidea) were censused using a time-constrained walking census along a 500-rn transect traversing a triangular course within each of the 12 study sites (Pollard 1977, Gall 1985). Individual transects were censused by slowly walking the length of the transect in approximately 20 min. Time required to count butterfly aggregations, net and identify individuals, and pauses to wait for appropriate wind or sun conditions (see below) was not included in the 20-min census time. All butterflies detected along the census transect during the 20-min census period were counted, regardless of &stance from the midline of the transect. Consequently, comparisons across species, especially for species of different sizes or behaviors, are not appropriate. Most individuals (87.2%) were identified to species. Census counts were conducted four times per year (first week of April, June, August, and October) during 2000 to 2002. Census counts on each site were replicated three times (a different observer on three separate days) during each month surveys were conducted. Individual censuses were conducted between 0900 and 1330 hrs CST on days when temperatures were between 18 and 36 C and wind velocity beneath the canopy was not strong enough to suppress butterfly flight (Beaufort Scale c 4). Census counts were further restricted to periods when sunlight was sufficient to cast discernable shadows. The response of butterflies to temperature, wind, and cloud cover varies seasonally and daily in complex ways. Consequently, observer judgment further constrained censusing to those periods when butterfly flight activity was judged to be substantial. Nectar resources were quantified during each butterfly sampling month within three 1x 100-m belt transects located paraIlel to each butterfly census transect. All nectar resources were counted and recorded by species. For most species, individual flowers or composite heads (capitula) were enumerated. Inflorescenses, or portions thereof, were counted for a few species with small andlor dense aggregations of flowers (e.g. Ceanothus amencanus L., Aabm sp., Solidago spp., Apiaceae). Enumeration decisions were based on the structure that most closely approximated a separate landing site for a typical butterfly. In addition, throughout our investigations (19992004) in the Ouachita Mountains, we recorded observations of nectaring and other feeding activities. Date, resource, and butterfly species were recorded. Voucher specimens of plant species were collected for later identification. Plant nomenclature follows Smith (1994). Plant vouchers will be deposited in the Stephen F. Austin State University Herbarium, Nacogdoches, TX. In the Ouachita Mountains, male S. diana emerge in late May to early June and S. cybele emerge in midto late May. Females of both species emerge approximately 7-10 days later in each species. Adult TABLE 1. Feeding records (1999-2005) of Diana fritillary (Spqeria &nu) maldfernale and great spangled fritilIary (Speyeria cybele) maldfemale/unknown in the Ouachita Mountains of western Arkansas by month. Source Species May m a Jd. Aug. Sept. Oct. Total Major Nectar Species A9cEeplns tuherosa L. S. diana

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.