Laserfiche WebLink
500 S.M. NELSON AND D.C. ANDERSEN <br />at revegetated sites may preclude site use by butterflies. Premature abscission of cottonwood leaves from <br />water stress at xerified sites could affect butterflies dependent upon these food sources. <br />Although the presence of another phreatophyte-dependent butterfly-N. antiopa-at the No Name <br />Lake and Havasu Refuge revegetation sites suggests a trend toward the `natural' Bill Williams River site, <br />this species could be a holdover using relictual willows. Unlike most other resident butterfly taxa, N. <br />antiopa can undergo seasonal migrations (Shapiro, 1986), a trait that would promote colonization of <br />disjunct revegetated sites. Despite differences in chronological age, the No Name Lake and Havasu <br />Refuge sites may both attract this butterfly because each offers moist soils and areas of deep shade from <br />closed canopies. Woodland butterflies often prefer shade (Warren, 1985) and Debinski and Brussard <br />(1994) found N. antiopa to prefer moist soils. Microclimate is important in butterfly habitat restoration <br />(Weiss and Murphy, 1990). <br />Impacts to assemblages <br />Butterflies associated with First Bottom vegetation have undergone a dramatic decline in numbers <br />because of the virtual elimination of First Bottom environment. The entire historic butterfly assemblage <br />may not be supported at revegetation sites because compositional (presence of larva] host plants), <br />structural (closed canopies), and functional (hydrology, flowering phenology) diversity is lacking. In some <br />cases the original intent of revegetation was to decrease water use by plant communities. This was done <br />by replacing dense stands of exotic tamarisk with native plants that would transpire less but provide <br />.increased wildlife values (Anderson and Ohmart, 1985). The depauperate butterfly assemblages at <br />revegetated sites, however, suggest that these few plantings at widely-spaced floodplain sites are insufli- <br />cient for restoration. <br />Potential butterfly source areas, along with being extremely rare, may be too far from revegetation sites. <br />The difficulties to butterflies of colonizing distant habitats are illustrated by Thomas et al. (1992) who <br />showed that some taxa.may take 2-3 years to colonize patches just 300-700 m from source populations. <br />The Bill Williams is one of the last regional areas containing the historic butterfly assemblage's major <br />elements. Butterflies dispersing from. BW -must travel large distances through, inhospitable habitat, a <br />consequence'of the isolation of the Bill Williams and Colorado River floodplains by the creation of Lake <br />Havasuy a reservoir with little native woody riparian vegetation along its shoreline. The time required for <br />colonists from BW to reach a revegetated site on the lower Colorado River probably exceeds the life span <br />of the site. Additionally, although areas of habitat needed by butterflies are largely unknown, the <br />relatively small size of revegetated sites may negatively affect the viability of butterfly populations. It is <br />possible that butterfly habitat was patchy and fragmented in the historic Lower Colorado River; however, <br />these patches were part of a heterogenous landscape, rather than surrounded by near monocultures of <br />tamarisk and agricultural crops as at present. <br />Assemblage restoration <br />The usefulness of created habitats to butterflies may not be fully discernable until appropriate taxa are <br />artificially introduced. Successful introductions could also protect remaining regional populations of some <br />butterfly species from chance extinctions. The introduction of phreatophytes such as willows and <br />cottonwoods at some revegetation sites suggests that reintroduction of phreatophyte dependent butterflies <br />(e.g., Calephelis nemesis, Limenitis archippus, Nymphalis antiopa) might be possible. The dependence of the <br />first two of these butterflies on nectar sources, however, argues that they cannot survive at any of the <br />xerified revegetation sites. Saline soils and lowered water tables may also preclude the introduction of C. <br />nemesis's host plant, B. salicifolia. Because the annual disturbance caused by flooding and seed sources <br />are now absent, recreating the original spatial mix of riparian vegetation patches along the lower <br />Colorado River is probably impossible. With canopy closure, however, Nymphalis-antiopa might be <br />reintroduced or repopulate additional revegetated sites. This butterfly differs from the other two in that <br />populations can apparently persist without extensive nectar plants and may use sap for nutrition (Scott, <br />1986). Autecological studies of these riparian butterflies are needed to determine whether successful <br />introductions are likely. <br />Copyright ~ 1999 John Wiley & Sons, Ltd. Regul. Riuers: Res. Mgml. 15: 485-504 (1999) <br />