9331 - Laserfiche WebLink

Home Browse Search

l~44 ~lkut ~6~~cl-t- ~'.~... . '~ 1~~t Channel narrowing by vertical accretion along the Green River near Green River, Utah Tyler M. AlIred* } John C. Schmidt Watershed Science Unit and Department of Geography and Earth Resources, Utah State University, Logan, Utah 84322-5240 ABSTRACT The Green River is the longest tributary of the Colorado River. Near the town of Green River, Utah, the Green River narrowed in two discrete phases during the twentieth century. The first phase of narrowing decreased aver- age width by about 5% and occurred between about 1930 and 1940, when the magnitude of 2-yr flood, mean annual discharge, and effec- tive discharge decreased by about 30%, 28%, and 37%, respectively. During this first phase of narrowing, saltcedar (Tamarisk spp.), an in- vading non-native tree, began to establish itself in the study area, but botanists of that time did not describe the tree as abundant. Channel width was stable in the 19408 and 19508 even though saltcedar were becoming already abundant on the river's banks. Further nar- rowing of an additional 14% occurred after 1959. This latest period of narrowing began foUowing three successive years when the mag- nitude of floods was less than the present l.5-yr recurrence flood and when saltcedar were al- ready abundant along the river. The deposits that comprise the banks of the narrowing Green River are composed of the suspended load of the river, and these aUuvial deposits are characterized by horizontal layers, which indi- cate that they formed by vertical accretion. A mechanism is proposed to explain the coarsen- ing-upward sequence of beds found in these verticaUy accreted deposits. These changes in the channel of the Green River are based on analysis of more than 2600 discharge measurements made by the U.s. Ge- ological Survey, resurvey of an abandoned measurement site, matches of historical ground-level photography, and analysis of his- torical aerial photography within a geographic information system. We have developed analy- ~ *Present address: 666 West 1725 North, Orem, Utah 84057; e-mail: tylerallred@earth1ink.net. tical techniques that permit analysis of width data from U.S. Geological Survey discharge measurements where gaging cross sections have adjustable beds and banks. These tech- niques aUow the spatiaUy rich but temporally poor data from aerial photographs to be sup- plemented with gaging station data, which add great detail about the timing and actual processes of channel narrowing that cannot be determined from aerial photographs alone. Such an analytical strategy provides a more complete record of historical channel adjust- ment than can be obtained by other means. INTRODUCTION 1\ventieth-century climate change, non-native vegetation invasion, construction of large dams, and transbasin diversions have altered the charac- ter and function of many riverine ecosystems in the western United States (Wtlliarns and Wohnan, 1984; Collier et al., 1996). The Colorado River basin has been affected by these climatic, biotic, and anthropogenic changes; dams are so large and numerous that the basin's ratio of total reservoir volume to mean annual discharge is the largest in North America (Hirsch et aI., 1990). In an era when reservoir operations are being revised to maintain or restore riverine ecological function downstream from dams, it is essential that we un- derstand the magnitude of recent channel change and the relative roles of climate, vegetation, and water development in causing those changes. Climatically driven changes in hydrology (Graf et aI., 1991) and non-native vegetation invasion (Graf, 1978) caused wrregulated alluvial streams to narrow throughout the southern Colorado Plateau between 1940 and 1980. Stratigraphy and historical data indicate that the primary mecha- nism by which these streams narrowed was depo- sition of an inset floodplain, as documented on Kanab Creek (Webb et al., 1991), Little Colorado River (Hereford, 1984), Paria River (Hereford, 1986; Topping, 1997), Vrrgin River (Hereford GSA Bulletin; December 1999; v. Ill; no. 12; p. 1757-1772; 15 figures; 5 tables. 1757 et al., 1996), and smaller streamS in southern Utah and northern Arizona (Graf, 1987; Hereford, 1987a, 1987b). Graf et al. (1991) and Hereford et al. (1996) showed that inset floodplain alluvia- tion began when the frequency of large floods de- creased, and these floodplains were constructed by moderate magnitude summer and fall floods trans- porting large suspended-sediment loads. There have been no studies of recent channel narrowing of undarnmed or undiverted streams in the north- ern part of the Colorado Plateau, where the inva- sion of non-native riparian vegetation is more re- cent (Graf, 1978) and the climate is different from that of the southern basin (Stockton and Jacoby, 1976). Wann season floods are relatively less im- portant in the northern part of the basin where snowmelt floods dominate the hydrologic regime. There have also been few studies of recent channel change of the Colorado River basin's large rivers, which Graf (1987) called inter-regional streams. The causes, rate, timing, and processes of narrowing of these large streamS are not well un- derstood. These streams are the Green River downstream from the Yampa River, the Colorado River downstream from the Gunnison River, the San Juan River downstream from Navajo Dam, and the Gila River downstream from the Mogollon Rim. The discharge of these rivers is primarily de- rived from snowmelt in the Rocky Mountains (Iorns et al., 1965), and each of these streams has been affected by dams. Channel change along these large streamS is a complex result of climatic fluctuations that affect runoff production in the Rocky Mountains, climate and land use changes that affect sediment yield and delivery in erodable, lower-elevation watersheds tributary to these streams, transbasin diversions in tributary headwa- ters, and/or mainstem dams. CHANNEL NARROWING OF THE GREEN RIVER IN UTAH Channel narrowing of the Green River has been more extensively studied (Goo, 1978; Andrews,