9530 - Laserfiche WebLink

Home Browse Search

~, 9'~~0 del sore ~ ~nd~~o~~ REGULATED RIVERS: RESEARCH & MANAGEMENT Regul. Rivers: Res. Mgmt. 15: 485-504 (1999) BUTTERFLY (PAPILIONOIDEA AND HESPERIOIDEA) ASSEMBLAGES ASSOCIATED WITH NATURAL, EROTIC, AND RESTORED RIPARIAN HABITATS ALONG THE LOWER COLORADO RIVER, USA S.M. NELSONa•* AND D.C. ANDERSENb ' a US Department of the /nterior, Bureau of Reclamation, Technical Service Center, Denver, CO 80225, USA b US Geological Survey, Midcontinent Ecological Science Center, Denver, CO 80225, USA ABSTRACT Butterfly assemblages were used to compare revegetated and natural riparian areas along the lower Colorado River. Species richness and correspondence analyses of assemblages showed 'that revegetated sites had fewer biological elements than more natural sites along the Bill Williams River. Data suggest that revegetated sites do not provide resources needed by some members of the butterfly assemblage, especially those species historically associated with the cottonwood/willow ecosystem. Revegetated sites generally lacked nectar resources, larval host plants, and closed canopies. The riparian system along the regulated river segment that contains these small revegetated sites also appears to have diminished habitat heterogeneity and uncoupled riparian corridors. Revegetated sites were static environments without the successional stages caused by flooding disturbance found in more natural systems. We hypothesize that revegetation coupled with a more natural hydrology is important for restoration of butterfly assemblages along the lower Colorado River. Copyright ©1999 John Wiley & Sons, Ltd. xsv woR~s: butterflies; cottonwood; larval host; nectar source; restoration; revegetation; river regulation INTRODUCTION River regulation and.. development of water resources have had .deleterious effects on the riparian vegetation of the Colorado and other highly-managed rivers in the western US. Before construction of regulatory and diversion dams, the Colorado showed great irregularity in flow (Figure 1). Near its mouth, at Yuma, Arizona, discharge ranged from 0.5 m3/s to 7079 m3/s during the period 1904-1935 (US Geological Survey, 1954). This variation was probably similar at upriver sites above the confluence of tributaries such as the Gila and Bill Williams Rivers, which were also subject to large variation in peak discharges. For example, the 3-km wide by 10-km long Chemehuevis Valley that now contains Lake Havasu, the reservoir created by Parker Dam, was `all subject to overflow during normal high water' (Colorado River Commission of the State of California, 1931: 263, referencing Senate Document 142: 51-52). In the larger Parker Valley, immediately downstream, `the river is said to overflow the lower bottom lands with a flood exceeding 40000 cfs [ ~ 1130 m3/s]' (Colorado River Commission of the State of California, 1931: 265). Similarly, the 19-km long Palo Verde Valley `is little but above the bed of the stream, and the entire area is subject to overflow during normal high water' (Colorado River Commission of the State of California, 1931: 267). River management since the mid-1930s has resulted in a flat hydrograph and decreased river discharge (Figure 1). Historically, riparian areas along the lower Colorado River formed on two terrace-like landforms: a First Bottom that flooded annually or nearly so, and a higher Second Bottom that flooded less frequently (Grinnell, 1914). The First Bottom was marked by tall willows (Salix gooddingii) and Fremont cottonwoods (Populus fremonti) in largely continuous bands that, although typically -100-200 m in width, could be up to 11 km wide in valley areas (Grinnell, 1914). The dense foliage of these riparian areas was ' Correspondence to: Bureau of Reclamation, D-8220, P.O. Box 25007, Denver, CO, 80225-5007, USA. CCC 0886-9375/99/060485-20$17.50 Received 3 March 1998 Copyright ©1999 John Wiley & Sons, Ltd. Revised 25 September 1998 Accepted 3 December 1998 ~'S3o