| This report examines the relationships among river energy expenditure, channel forms, and models of river change. A principal indicator of river energy expenditure, stream power, has a significant influence on many forms and process attributes of the fluvial system, yet few basin-wide analyses of stream power variations have ever been conducted. These variations are examined in two high mountain stream basins in the central Sangre de Cristo Mountains in northern New Mexico and southern Colorado. There are two principal research questions of this dissertation. First, how does stream power vary within a high mountain basin both geographically and as a response to different magnitude hydrologic events? Second, what are the most important variables that affect both stream power distribution and the geomorphic forms that result from the variation in stream power?; During the summers of 1998 and 1999, over 170 sites throughout the Costilla basin and the Canadian basin headwaters yielded evidence of abundant local control over fluvial processes and forms. In Costilla Basin, hypothetical patterns derived from hydraulic geometry overpredict stream power distributions in areas of glacial and glacio-fluvial materials, and underpredict stream power values in steep, bedrock-controlled channels and in gentle-slope areas with cohesive bank materials. Power variations in these discrete areas deviate from the hypothetical power curves by as much as 100%.; In the Canadian basin headwaters, a fortuitous large flood in 1999 allowed direct examination between flood processes and the evidence left by the flood. Because many researchers use this remaining evidence to piece together flood history in a basin, the correlation between the evidence and the actual flood processes is extremely important. Significantly, there were major differences between the flood processes in the Canadian headwaters and the evidence the flood left. Shear stress, stream power, and discharge reconstructed from the flood deposits using tractive-force methods often underpredicted these observed measurements of these variables. For example, the median discharge observed in the flood was 7.23 cms, whereas that constructed through sediment sizes was 2.92 cms. If paleoflood reconstruction is to be useful, researchers must come to grips with the enormous heterogeneity of flood processes across the landscape.; Simultaneous-equation models of the adjustment of the forms in Costilla basin have similar directions of response as those proposed by Rhoads in his analysis of ephemeral semi-arid mountain streams. A large amount of the variance in the adjustments in fluvial forms and processes is explained by the system of equations, with only sediment size characteristics poorly explained. This type of model system exhibits a dampened response when converging on a new equilibrium state. This damped response is a numerical analogue to Graf's rate law in geomorphic systems. Improvements in this class of river models will require the inclusion of known extrinsic and intrinsic thresholds through logical hierarchies of discriminant equations.; The analysis of data in Costilla, Canadian, and Pecos basins in northern New Mexico does not support the generally-held assumption that the “bankfull” stage of a river is equivalent to the “dominant discharge” of that river. Instead, the bankfull level is the product of a combination of water forces and local sediment and bank processes. This is a major assumption in recent channel restoration schemes, and requires a critical assessment of the idea in general. |
