70 Sediment Composition and Texture of Reservoir Deposits in Lake Powell Tributary Canyons, Utah

Faculty Mentor: Brenda Bowen (Geology and Geophysics, University of Utah)

Abstract

Since the impoundment of the Colorado River by Glen Canyon Dam in the 1960s, sediment delivered by the river and tributary streams has accumulated within Lake Powell, forming extensive deltaic and lacustrine deposits. Prolonged reservoir drawdowns has now exposed large tracts of these fine-grained sediments, offering an opportunity to evaluate their composition, texture and post depositional modification. This study characterized exposed reservoir deposits in three tributary canyons- Clearwater, Dark and Gypsum canyon in southeastern Utah using integrated sedimentological and geochemical methods. Sixty-three samples were analyzed for grain size distribution with Laser Grain Size Analysis (LGSA); twenty representative samples of the most fine-grained deposits, interpreted to be deposited by reservoir rather than river processes, were further examined using X-ray Fluorescence (XRF), X-ray Diffraction (XRD) and Scanning Electron Microscopy (SEM). LGSA results define eight distinct sediment types ranging from clay to sand dominated textures. XRF analyses indicate relatively homogenous major element compositions dominated by Si, Al and Ca, whereas XRD identifies quartz, calcite and illite as prevalent crystalline phases and up to 55% phyllosilicate content (an average of 34%). SEM observations reveal pronounced variability in grain morphology, including platy clay aggregates, etched quartz grains and carbonate and iron oxide coatings that record diagenetic alteration. Collectively, these data demonstrate that while geochemical composition is broadly uniform across canyons, mineralogical and textural heterogeneity reflects localized depositional energy and subsequent diagenesis. These findings prove a foundation for assessing sediment cohesion, failure potential and ongoing geomorphic adjustment within the desiccated reservoir floor.

Introduction

Reservoirs profoundly modify natural sediment transport processes by trapping fluvial material, altering downstream sediment budgets and creating artificial basins that record both natural and anthropogenic environmental change (Graf et al., 2010; Syvitski & Kettner, 2011). Lake Powell, formed by the construction of Glen Canyon Dam in 1963, has stored vast quantities of sediment delivered by the Colorado River and its tributaries for more than six decades (Root & Jones, 2022). These deposits, which include both deltaic and lacustrine facies, provide a detailed archive of sediment provenance, hydrologic variability and geochemical evolution within the Colorado Plateau watershed (Johnson et al., 2021). Recent megadrought conditions in the southwestern United States have caused historically low reservoir levels, exposing extensive sediment accumulations and re activating sedimentary processes along the former reservoir margins (Williams et al., 2020; Udall & Overpeck, 2017).

Exposed reservoir sediments represent a unique geomorphic setting in which previously submerged materials are subjected to subaerial weathering, desiccation and renewed erosion. The rapid exposure of reservoir sediment presents new environmental and geotechnical challenges, including slope instability, dust generation and potential remobilization of contaminants previously sequestered under anoxic condition (Johnson et al., 2021; Draut & Rubin, 2008). Understanding the physical and geochemical characteristics of these sediments is therefore critical for predicting their stability and behavior during post drawdown landscape adjustment.

The composition of reservoir sediment is influenced by multiple factors, including source lithology, hydrodynamic sorting and post depositional diagenesis (Bouchez et al., 2011; Hartley, 2022). Elemental and mineralogical analyses of such deposits can reveal signatures of watershed geology and sediment provenance while also identifying chemical transformations that occur during submersion and subsequent exposure (Shotbolt et al., 2005; Wildman et al., 2011). For Lake Powell and similar large reservoirs, distinguishing between primary depositional characteristics and secondary diagenetic overprints is essential to evaluate long term sediment stability and environmental risk.

This study focuses on sediment samples from three tributary canyons – Clearwater, Gypsum and Dark Canyons, located along the former margins of Lake Powell in southeastern Utah. Each canyon drains distinct bedrock units that contribute unique sedimentary inputs, yet all experienced similar phases of reservoir inundation and subsequent drawdown. The study area and sample locations are shown in Sam Bagge’s Sedimentation and Remobilization of Exposed Lake Powell Reservoir Deposits in Cataract Canyon along the Colorado River (2025) thesis map (Figure 1), which provides a detailed overview of the geomorphic setting and sampling sites. Field observations from this project reveal textural contrasts among canyons, ranging from coarser, sand-dominated deposits in Clearwater Canyon to fine, clay-rich sediments in Gypsum Canyon, with Dark Canyon showing intermediate textures. These differences likely reflect variations in depositional energy, sediment source and post-depositional processes that occurred during and after reservoir infilling.

The objective of this study is to characterize the grain size distribution, mineralogical composition and geochemical variability of these exposed reservoir sediments using integrated Laser Grain Size Analysis (LGSA), X-Ray Fluorescence (XRF), X-Ray Diffraction (XRD) and Scanning Electron Microscopy (SEM). By combining these techniques, we evaluate whether compositional similarities across canyons reflect homogenization during reservoir deposition or if textural and mineralogical distinctions preserve evidence of localized depositional conditions. This integrated approach provides insight into how fine-grained reservoir sediments respond to prolonged submersion and subsequent re-exposure, contributing to broader understanding of sediment dynamics and landscape evolution in large reservoir systems of the arid Southwest.

Methods

Sample Collection and Preparation

A total of 63 sediment samples were collected from Clearwater, Gypsum and Dark Canyons-tributaries along the former margins of Lake Powell, Utah. All samples were collected with permission from Glen Canyon National Recreation Area under an approved research permit in coordination with the Returning Rapids Project. Field sites were selected to target fine-grained, stratified deposits interpreted as exposed reservoir muds. Each sample was collected using sterile scoops, stored in labeled polyethylene bags and transported to the B. Bowen Lab (Department of Geology & Geophysics, University of Utah) for processing.

Samples were air-dried and prepared following the standard procedure for Laser Grain Size Analysis (LGSA). Each sample was sieved to <2 mm to remove coarse fragments and homogenized thoroughly to ensure representative subsampling. Approximately 10 grams of material from each sample were weighed for analysis. Cohesive or aggregated muds were gently disaggregated using a ceramic mortar and pestle to minimize particle breakage and preserve natural grain structure. All equipment was cleaned between samples to prevent cross-contamination.

Laser Grain Size Analysis (LGSA)

Grain size distribution was measured using a QICPIC particle size analyzer with HELOS sensor control software, housed in the Department of Geology and Geophysics at the University of Utah. Each sample was analyzed at a dispersion pressure of 4 bar to ensure complete particle disaggregation. Measurements were conducted using R4 and R6 lenses, with three replicate runs per sample to verify precision and reproducibility.

Raw LGSA data consisted of particle density value across 40 grain size bins (0.95-320 µm). Distributions were averaged and reshaped into long format for analysis. Samples were first grouped into three broad textural class – clay (<4 µm), silt (4-63 µm) and sand (>63 µm) – based on modal grain size. Within each class, clustering was applied to define eight distinct sediment types that represent variations in depositional energy and sorting across the three canyons systems.

All data cleaning, clustering and visualization were performed using Python (pandas, NumPy, matplotlib and scikit-learn) and R (tidyverse, ggplot2). Modal peaks were extracted from raw distributions, log transformed and classified using k-means clustering, with silhouette scoring used to validate cluster number and consistency. These eight sediment types were standardized and color coded to remain consistent across Clearwater, Gypsum and Dark Canyons, forming the textural framework for subsequent XRF, XRD and SEM analyses.

Grain size distributions for each canyon were plotted as stacked panels, with shaded background zones marking clay, silt and sand fields. The resulting plots highlight textural contrasts and allow direct comparison across canyons.

X-Ray Fluorescence (XRF) Analysis

Elemental composition was measured on the Eagle III Microspot EDS-XRF in the University of Utah NanoFab Lab. The twenty finest LGSA-identified samples were lightly homogenized and pressed into pellets using a manual press to create smooth surface, Pellets were loaded into the instrument’s vacuum chamber and scanned under identical conditions.

Major elements (Si, Al, Ca, Fe, K, Ti) were reported as atomic percent (At%). Each sample was run in triplicate for consistency. Data were cleaned and averaged by canyon in Python (pandas, NumPy, matplotlib) and results were visualized as pie charts to compare relative elemental abundances across Clearwater, Gypsum and Dark Canyons.

X-Ray Diffraction (XRD) Analysis

Mineralogical composition was determined using the Bruker D8 DISCOVER High-Resolution X-ray diffractometer (Bruker AXS, Germany) housed in the Utah NanoFab Laboratory at the University of Utah. The twenty fine-grained samples selected from the LGSA dataset and analyzed by XRF were also used for XRD analysis. Each sample was pressed into pellet form and analyzed under a vacuum chamber to minimize background interference and improve signal quality.

Coupled 2θ measurements were performed at a step size of 0.05° using Cu Kα radiation (40 kV, 40 mA) with a Göbel mirror and an Eiger 2R 250K detector. The Crystallography Open Database (COD) was used to match crystal structures to the resulting diffraction patterns.

Each sample was scanned from 5° to 70° 2θ and the resulting diffraction data were processed to identify mineral phases and estimate their relative abundances. Major crystalline minerals identified across samples include quartz, albite, illite, muscovite, kaolinite, calcite, dolomite, rutile and hematite.

Data were compiled and visualized in Python using pandas, NumPy and matplotlib libraries. Average mineral abundances were calculated for each canyon (Clearwater, Gypsum and Dark) and displayed as grouped bar charts to compare relative mineralogy across sites.

Scanning Electron Microscopy (SEM/EDS)

Surface morphology and micro textural features were examined using the FEI Teneo™ Scanning Electron Microscopy (SEM) equipped with EDAX Octane Energy Dispersive Spectroscopy (EDS) detector at the Utah NanoFab Lab at the University of Utah.

The twenty fine grained samples selected from the LGSA dataset were used for SEM/EDS imaging. Each sample was mounted on aluminum stubs using carbon adhesive tabs and lightly coated with gold to improve conductivity. Samples were analyzed under high vacuum mode at varying magnification to observe grain shape, angularity and surface texture.

EDS point analyses were performed on representative grains to determine elemental composition and verify mineral phases observed in XRD results. SEM images were compiled for each canyon to document textural variability and diagenetic features such as clay aggregates, etched quartz surfaces and iron oxide coatings.

Results

Laser Grain Size Analysis (LGSA)

Grain size distribution across sixty-three samples shows textural variability among the three tributary canyons (Figure 3). Eight sediment types were identified from clustering, ranging from clay dominated to sand dominated textures. Clearwater canyon samples generally exhibit coarser-sand rich distributions with multiple peaks. Gypsum Canyon samples are predominantly fine grained, silt and clay rich. Dark Canyon samples display intermediate, mixed mode – distributions.

The fine-grained samples chose for further analyses correspond to the lower end of the grain size spectrum (2-20 µm), providing a representative dataset for comparison of chemical and mineralogical variability.

X-Ray Fluorescence (XRF)

Elemental compositions (Figure 4) show broadly similar chemical trends across all three canyons. Sediments are dominated by silicon, aluminum and calcium, with minor contributions from iron, potassium and titanium. These elements correspond to oxide, silicate and carbonate phases common to all sites.

Subtle differences in geochemistry occur between canyons: Clearwater shows slightly higher calcium content, Gypsum Canyon contains more aluminum and potassium and Dark Canyon exhibits intermediate values across all elements.

X-Ray Diffraction (XRD)

Mineralogical results (Figure 5) show consistent assemblages dominated by quartz, feldspar, and clay minerals, with variable amounts of calcite and dolomite. Minor phases include rutile. Quartz constitutes the largest fraction in all samples (30–45 wt %), while phyllosilicate clay minerals (illite, muscovite, and kaolinite) collectively make up 25–40 wt %.

Crystalline results indicate that Clearwater Canyon samples average 67 % crystalline and 33 % non-crystalline, Dark Canyon samples average 64 % crystalline and 36 % non-crystalline, and Gypsum Canyon samples average 68 % crystalline and 32 % non-crystalline. Gypsum Canyon has the highest clay content, whereas Clearwater has higher carbonate proportions.

Scanning Electron Microscopy (SEM/EDS)

SEM observations (Figure 6) reveal distinct surface textures between canyons. Clearwater samples contain angular to sub-angular quartz grains with carbonate cement and occasional etched surfaces. Gypsum Canyon samples are fine-grained and compact, dominated by platy clay aggregates and iron-oxide films. Dark Canyon samples show mixed grain sizes and variable surface coatings.

EDS point and map analyses confirm the presence of silicon, aluminum and calcium-rich phases, consistent with the elemental and mineralogical data. Elemental maps highlight fine-scale heterogeneity, with silicate and carbonate domains occurring within the same samples.

Integrated Results

Across all datasets (Figure3-6), sediments display chemical uniformity but textural and mineralogical variability at the canyon scale. All sites are enriched in quartz and aluminosilicate minerals but differ in clay and carbonate proportions and in grain morphology.

Discussion

Depositional Variability and Energy Conditions

Grain-size data indicate that sediment texture varies strongly between tributary canyons even though bulk geochemical composition remains similar (Figures 3–6). Clearwater Canyon’s sand-rich distributions reflect higher-energy inflows and more direct fluvial input during reservoir filling. Gypsum Canyon’s clay and silt dominated samples suggest low-energy deposition under lacustrine or suspension settling conditions, while Dark Canyon’s mixed distributions indicate variable inflow velocity and intermittent reworking. These contrasts imply that local hydrodynamic conditions within the former reservoir basins, rather than differences in sediment source, primarily governed depositional processes.

Mineralogical and Diagenetic Differences

Although major-element chemistry is relatively uniform among canyons, mineralogical and microscopic evidence indicate localized post-depositional modification. Carbonate minerals such as calcite and dolomite are more abundant in Clearwater Canyon, consistent with cementation under partially evaporative or groundwater-influenced conditions. In contrast, Gypsum Canyon’s higher clay content and pervasive iron-oxide coatings reflect finer-grained accumulation and early diagenetic oxidation within saturated muds. The coexistence of quartz-rich and clay-rich textures across short spatial scales suggests alternating inflow and still-water phases during reservoir operation.

Textural Controls on Sediment Stability

SEM observations show that particle morphology varies with depositional environment, influencing the behavior of exposed sediments. Coarser, carbonate-cemented units in Clearwater Canyon appear more indurated, whereas platy clays in Gypsum Canyon are compact but weakly bonded, becoming cohesive when wet and friable when dry. These contrasts in texture and bonding are expected to affect slope stability, dust generation and erosion as these surfaces continue to weather.

Implications for Exposed Reservoir Landscapes

The combined dataset shows that compositional uniformity does not equate to uniform environmental behavior. Subtle mineralogical and textural variations control how rapidly different deposits erode once exposed. Fine-grained, clay-rich materials in Gypsum Canyon may retain moisture and resist wind erosion but can fail when saturated, while carbonate-rich deposits in Clearwater Canyon may fracture along cemented zones. Understanding these textural dependencies is critical for predicting sediment mobility and assessing erosion risk as reservoir levels continue to decline throughout the Lake Powell region.

Conclusion

This study characterizes the composition and texture of exposed reservoir sediments from Clearwater, Gypsum and Dark Canyons, providing insight into how depositional energy and diagenetic processes influence sediment properties following reservoir drawdown. Results from integrated LGSA, XRF, XRD and SEM/EDS analyses show that, while the sediments share a consistent geochemical composition dominated by silicates and carbonates, they differ notably in grain size, mineralogy and micro texture.

Clearwater Canyon contains coarse; carbonate rich deposits associated with higher energy inflow conditions. Gypsum canyon is dominated by fine grained; clay rich muds formed under low energy settings. Dark Canyons represents a transitional environment with mixed textures. These variations reflect localized depositional environments within the Lake Powell reservoir during “full pool” conditions and impact the present-day stability and erosion potential of the exposed surfaces.

Overall, the findings demonstrate that textural and mineralogical differences, rather than bulk chemistry, govern how exposed reservoir sediments respond to re exposure and weathering. As reservoir levels continue to decline, understanding these sediment properties provides a foundation for assessing erosion risk, dust generation and long-term landscape evolution in the Colorado River basin.

Acknowledgements

This research was supported by the University of Utah Office of Undergraduate Research (UROP), the Wilkes Center for Climate Science & Policy, the Society, Water and Climate Program and the Department of Geology & Geophysics. Analytical work was conducted at the B. Bowen Lab and the University of Utah NanoFab Laboratory. This work made use of University of Utah shared facilities of the Micron Technology Foundation Inc. Microscopy Suite, sponsored by the John and Marcia Price College of Engineering, Health Sciences Center and the Office of the Vice President for Research. Acquisition of the Bruker D8 DISCOVER system was made possible by the Air Force Office of Scientific Research under project number FA9550-21-1-0293.

Field access and research permits were provided through the Returning Rapids Project in collaboration with the National Park Service, including Canyonlands National Park and Glen Canyon National Recreation Area. The author thanks Dr. Brenda Bowen for mentorship and guidance and Sam Bagge for assistance with data analysis and field logistics.

Bibliography