Remnants of river terraces and alluvial fans preserved in the deeply incised canyons of the Colorado Front Range reflect a history of long-term river incision, large scale Pleistocene sediment budget perturbations, and small scale Holocene perturbations. Relative dating techniques including the degree of soil development, landform shape, degree of rock weathering, and stratigraphic position have been used to draw broad correlations between the timing of terrace formation and glacial events and to infer long-term evolution of longitudinal river profiles. In my thesis I explore the relationship between glacial chronology of the Front Range and the fluvial responses to glacial events as preserved in Boulder Canyon through the use of 26Al and 10Be cosmogenic exposure age dating techniques. I also apply 14C dating methods to determine ages of Holocene alluvial fans and terraces which indicate small-scale Holocene sediment budget perturbations within side channels of the canyon. The application of these field-based isotope dating techniques in Boulder Canyon represents the first attempt to quantitatively constrain the timing of multiple depositional and erosional events in a canyon draining the Colorado Front Range, and to correlate these events with climate change. Cosmogenic exposure age dating constrains the timing of middle and late Pleistocene glacial events in the region and provides a reasonable estimate of the incision rates of Middle Boulder Creek. 14C dating provides new information concerning the Holocene history of Boulder Canyon, and the character of erosive processes in the canyon.
In this thesis, I explore the timing of two late Pleistocene glacial advances, and their effects on the fluvial dynamics of Middle Boulder Creek. The Bull Lake glaciation and the Pinedale glaciation brought lobes of ice out as far east as Barker Reservoir, Glacial Lake Devlin, and Twin Lakes Reservoir (see location map). Middle Boulder Creek flows through the steep, narrow Boulder Canyon en route to the western High Plains, where numerous workers have correlated stranded alluvial surfaces with particular glacial episodes. The fluvial fill terraces flanking the valley walls of Boulder Canyon have not until now been reliably correlated with High Plains surfaces, nor been explained in terms of the system response to sediment budget perturbations.
Glacial moraines near Nederland in the Front Range and near Twin Lakes Reservoir in the eastern Sawatch Range, Colorado preserve evidence for several late Pleistocene glaciations. Near Nederland, moraines have been mapped as pre-Bull Lake (middle Pleistocene), Bull Lake (late Pleistocene), and Pinedale (latest Pleistocene), based on soil development, extent of boulder weathering, and moraine morphology. Bull Lake and Pinedale-age deposits dominate the glacial record, but pre-Bull Lake deposits are present locally. In most of the upland Front Range, Pinedale till and outwash fill broad valleys whereas Bull Lake till crops out in thin zones, between Pinedale moraines and high bedrock ridges. The upper reaches of alpine cirques preserve post-Pinedale glacial and glaciofluvial deposits locally, but far upvalley of pre-Bull Lake, Bull Lake, and Pinedale moraines. Near Twin Lakes, in the Sawatch Range, there is some uncertainty about assignment to the Bull Lake or Pinedale glaciations. Cosmogenic exposure ages of moraines near Nederland, Colorado and near Twin Lakes Reservoir should indicate the timing of maximum glaciation since the moraines in these locations are in close proximity to the farthest extent of glaciers in the region.
If fill terrace formation is related to glacial events, then exposure ages of fluvial fill terraces in Boulder Canyon should also provide constraints on the timing of glacial events. Fluvial terraces, with upper strath surfaces 1 to12 m above the present level of Middle Boulder Creek, are common in the inner canyon, along with extensive areas of colluvial debris and localized alluvial fans. Fill terraces form by valley aggradation of alluvial material above a bedrock channel, followed by channel incision through the alluvium. A fill terrace may record aggradation in response to an increased sediment load followed by incision when the sediment load is removed. In glaciated regions, widespread exposure of unconsolidated sediment during and immediately after glacial retreat led to high sediment loads transported by glacial meltwater and aggradation for some distance downstream. When the sediment supply diminished, meltwater cut down through the alluvium, stranding terraces. By comparing exposure ages of the glacial moraines west of Boulder Canyon and exposure ages of fluvial fill terraces within Boulder Canyon with the established regional glacial chronology, I correlate glacial events in the uplands with river responses in Boulder Canyon. Mass balance calculations on the Middle Boulder Creek catchment help to reveal whether or not my inferences about fluvial responses to sediment budget perturbations are reasonable.
Holocene alluvial deposits in Boulder Canyon provide perspective on relatively recent events in the canyon. Alluvial deposits are preserved in several side drainages within the canyon. Within the drainages, small streams cut through extensive terraced alluvial fans that grade into low (1 to 3 m above Boulder Creek) flat surfaces. The fans did not receive sediment from glacial ice, since no ice advance ever came close to their drainage basins. The fans most likely record sudden changes in sediment budgets, such as the loss in slope stability that results from forest fires. Charcoal collected from the base of fans and from fill terraces located downstream of the fans provides information concerning the timing of fan formation, and concerning the erosive character of the event(s) that led to fan development. Comparison of the timing of fan and low terrace formation with the Holocene climate record contributes to discussion about links between characteristic climate regimes and alluvial fan/fluvial terrace formation.
Incision rates and evolution of the river profile
The exposure ages of preserved features in Boulder Canyon reflect the history of Boulder Canyon incision. Strath terraces form when rivers cut down through bedrock channels, stranding the previous valley floor as terraces on either side. The age of a strath terrace represents the time that the river incised to a specific depth in the canyon. Isotopic exposure ages of strath terraces are potentially less ambiguous than ages of fill terraces in determining rates of bedrock incision, since fill terraces were deposited some time after incision down to the bedrock straths that they bury.
Incision rates provide important information for models that attempt to describe the roles of climatic and orogenic processes that produced the deeply incised canyons of the Colorado Front Range. The relative importance of each of these factors is still indeterminate after a century of geologic discussion. Exposure ages of terraces in Boulder Canyon can help determine a rate of incision for the most recent river incision. However, poor preservation of deposits and of polished bedrock above a height of 20 m above the present valley floor hampers measuring incision rates for most of the 200 to 300 meter depth of the canyon. If I extrapolate my measured rate of incision through the rest of the canyon, I can approximate a time of the initiation of downcutting. How well this age correlates with the presumed initiation of downcutting indicates whether or not the Pleistocene/Holocene rate of incision is faster or slower than incision rates of the past.
Finally, there is much to be learned from analyzing present and past geometrical characteristics of the river channel and valley. Longitudinal river profiles ("long profiles") are plots of river length against elevation. Terraces preserved along valley walls above the active valley bottom record a history of past long profiles. Long profiles can reflect characteristics of a drainage basin including variations in bedrock lithology, variations in resistance due to weathering or fracturing/faulting, and/or show a snapshot of one stage of a river's response to a change in base level. The shape of a river valley also reflects rates of incision and bedrock resistance. Rapid incision and resistant bedrock favor the formation of steep, narrow valleys, while slow incision and erodable rock types favor wide valleys. In a single long profile, regions of more resistant bedrock often exhibit steeper gradients. In studying coastal rivers, Pazzaglia et al. (1998) found that when incision rates are controlled by tectonic uplift, long profiles exhibit a strongly concave-upward shape due to the river's rapid response to changing base level. Tectonically active settings are characterized by high stream powers that are able to keep up with or exceed the rapid rates of rock uplift, producing relatively smooth concave-upward profiles. However, in tectonically stable regions, stream power tends to be limited. As a result, climate and bedrock characteristics are the dominant factors in controlling long profile shape after a change in base level. Profiles from stable regions tend to exhibit knickpoints, especially in regions of resistant bedrock. Knickpoints are localized convexities in an otherwise concave long profile. The longitudinal profile of Middle Boulder Creek preserves information about the tectonic, lithologic, and incision history of the Front Range region.