Suspended-load rivers are subject to rapid geomorphic changes. In particular during the Holocene Epoch, arroyos of the Colorado Plateau experienced several periods of rapid erosion and aggradation. The most recent period of entrenchment occurred around the turn of the 20th century. The mechanisms responsible for the modern period of aggradation that has followed the most recent period of entrenchment have not been well documented. The research presented in this thesis reveals the mechanisms responsible for modern alluviation of the San Rafael River, which drains the Colorado Plateau The lower 87 km of the San Rafael River, which enters the Green River south of the town of Green River, UT has experienced rapid geomorphic changes during the last 100 years. To quantify these changes, we used a complement of temporally precise and spatially robust methods. By understanding the rates, magnitudes and types of geomorphic changes, we could then identify the mechanisms of these channel changes. The San Rafael River narrowed by 83% between 1938 and 2009 and the floodplain aggraded 1.0 to 2.5 m. Channel narrowing was caused by a reduction in the transport capacity of the river, and was accelerated by the establishment of vegetation, including the non-native tamarisk shrub, on active channel surfaces and the floodplain. Significant water withdrawals during the 20th century have primarily been responsible for the reduction in transport capacity by decreasing the magnitude and duration of the annual snowmelt flood. During this time period, monsoon floods continued to deliver large quantities of fine sediment to the channel. During the 20th century, the channel bed incised in one segment and aggraded in five segments. The two periods of incision that we documented were related to human modifications of the channel and floodplain. With the knowledge of the physical processes that have been responsible for the channel changes in the San Rafael River, prediction of future channel conditions can then be made. The changes to the physical template of the San Rafael River have implications for the management of three endemic fish – the roundtail chub (Gila robusta robusta), the bluehead sucker (Catostomus discobolus), and the flannelmouth sucker (Catostomus latipinnis) – which currently utilize the study area.

The highly endemic native fish fauna has experienced dramatic reductions in abundance and range because of anthropogenic activity. In addition to a highly altered flow, temperature, and physical habitat template, many non-native fish species have established populations throughout the basin. The San Rafael River, a tributary of the Green River in southeastern Utah, is home to populations of flannelmouth sucker (Catostomus latipinnis), bluehead sucker (C. discobolus), and roundtail chub (Gila robusta), and has experienced degradations representative of many rivers throughout the Colorado River Basin. Using the San Rafael River as a template, I examined (1) the effect of the non-native fishes on the food web structure of a Colorado River tributary and (2) whether any changes to the food web structure by non-native fishes impact fitness-related vital rates of native fishes in the river. Stable isotope analysis revealed that the non-native fishes present novel predator archetypes to the “three species,” but size-at-age and body condition analyses did not reveal any reduction in fitness between native fishes captured in the presence of the non-native fishes. Additionally, I used site-specific biotic and abiotic measurements to develop a model predicting “three species” relative abundance. Non-native fish abundance and spatially auto-correlated measures of physical habitat were found to be particularly important predictors of “three species” relative abundance. The model was used in combination with a longitudinal habitat survey to predict the current continuous distribution along the length of the lower San Rafael River, as well as to simulate population-level effects of relevant restoration actions. The eradication of non-native fishes resulted in significant population increases for each of the ‘three species,’ and physical habitat restoration resulted in significant population increases when executed in certain reaches, but significant population reductions if executed in other reaches. These results suggest that the restoration of physical habitat without addressing populations of non-native fishes will likely result in a limited response by the “three species,” and that the choice of restoration location is critical to its success.