Lateral response of alluvial rivers to the downstream passage of episodic sediment pulses

Melendez Bernardo, Mishel Milagros

Permalink

https://hdl.handle.net/2142/127456 Copy

Description

Title

Lateral response of alluvial rivers to the downstream passage of episodic sediment pulses

Author(s)

Melendez Bernardo, Mishel Milagros

Issue Date

2024-11-19

Director of Research (if dissertation) or Advisor (if thesis)

• Best, James L
• Cienciala, Piotr

Doctoral Committee Chair(s)

• Best, James L
• Cienciala, Piotr

Committee Member(s)

• Garcia, Marcelo H
• Lara, Mark J
• Nicholas, Andrew

Department of Study

Geography & GIS

Discipline

Geography

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

sediment pulse, lateral response, planform morphology, remote sensing, fluvial geomorphology

Abstract

River channels are dynamic systems that continuously adjust their morphology in response to varying inputs of sediment and water over a range of timescales. These adjustments profoundly influence ecological processes, hydrological dynamics, and geomorphological stability. Understanding the lateral response of alluvial rivers to sediment pulses—transient increases in sediment supply due to natural events like landslides or anthropogenic activities such as mining—is crucial for assessing river channel resilience and managing associated hazards. This thesis investigates the morphological changes in river channel planform induced by sediment pulses. Specifically, the study focuses on three key metrics: active channel width, braiding intensity, and area of bars. These metrics serve as proxies for understanding how rivers accommodate and adjust to increased sediment loads from their watersheds, which can lead to shifts in sediment distribution, channel avulsions, and alterations in channel morphology. Methodologically, the thesis utilizes high-to-medium resolution imagery and historical aerial photographs spanning multiple decades to track temporal changes in channel morphology. Statistical analyses, including linear regressions, are employed to identify the dominant factors driving morphological adjustments in each river system. Factors such as valley confinement, hydrological regime, and the nature of sediment supply are assessed to understand their respective roles in shaping river channel evolution. Case studies are conducted on two contrasting river systems to apply this methodology: Lillooet River, British Columbia, Canada, which was impacted by a large sediment pulse from a major landslide in 2010, and the Puquiri-Colorado and Inambari rivers in Madre de Dios, Peru, affected by intensive gold mining and deforestation since the early 1990s. These case studies allow an analysis of how different natural and anthropogenic sediment sources and landscape characteristics influence river channel responses. The 2010 sediment pulse caused significant changes in the planform morphology of Lillooet River, increasing its active channel width, braiding intensity, and area of bars. The river reverted to its pre-disturbance braiding intensity 3 to 5 years after the landslide. However, the increased active channel width and bar area persisted for longer, with higher magnitudes observed in the final year of analysis (2022) compared to the period before the landslide (2009-2010). This implies that a substantial portion of the sediment delivered to the valley floor remains in temporary storage. Factors such as time since the landslide (a proxy for propagation of the sediment pulse downstream), water discharge, and valley morphology influenced the lateral response of Lillooet River. The presence of alluvial fans, fluctuating discharge, and different sources of sediment supply, such as the landslide deposit itself and avulsions/cutoffs, make observation of a single and unique dispersion process impossible and instead point to the likely presence of a sequence of different magnitude dispersive waves being transported at different rates downstream. Overall, the present study indicates Lillooet River is likely to continue changing over periods of several years, especially in active channel width and area of bars, due to a range of processes such as remobilization of the sediment stored within the river corridor and potential future flood events. This research also examined the evolution of Lillooet River from 1947-2010, in order to place channel change associated with the 2010 landslide-generated sediment pulse in a longer temporal context. The cumulative volume of sediment supplied from landslides in the Meager Creek Volcanic Complex, a significant source of sediment in the Lillooet River catchment, is found to be one of the principal factors controlling the evolution of active channel width. In addition, the increase in spring temperatures early in the 20th century, which led to a greater glacial-melt-driven sediment supply, as well as the following rapid retreat of Lillooet Glacier that enabled the integration of moraine sediment into the river system, positively influenced the increases in active channel width until 1948. However, sediment delivery from glacial sources may have become limited after this year due to cooler temperatures, resulting in a decrease in active channel width. In addition, decreasing channel width after 1948 was also likely influenced by reduced availability of proglacial sediment caused by the formation of Silt Lake between 1947 and 1962. Both of these factors caused a decreased supply of coarse sediment to Lillooet River, potentially causing the narrowing of the active channel width from 1948 to the early 1980s. Additionally, the influence of the hydrological regime (cumulative discharge above an assumed threshold for significant alluvial erosion) was significant in determining the active channel width. The Lillooet River valley has been facing an increase in peak flows, and the magnitude and frequency of floods have risen over time. Furthermore, the coupling of readily available sediment stored in the channel, and the occurrence of a large peak flow or flood event that has a long duration, has been important to reworking sediment and, consequently, increased rates of active channel width. Since a rise in air temperatures and precipitation events can destabilize the hillslopes that, in the past, used to be covered by glaciers, these hillslopes are more prone to landslide events, as observed between 1984 and 2010 with the consequent rise of active channel width. In this context, the substantial increase in active channel width following the 2010 landslide was still insufficient to reach the values observed in 1947 and 1948. In relation to the Puquiri-Colorado River, Peru, the current research showed that active channel width increased between 1968 and 2010 and between 2018 and 2022, with occasional decreases in between. Meanwhile, the Inambari River displayed a mixed pattern of behavior, with a declining active channel width during most periods, but a general trend of increase since 1968. The impact of hydrological factors, such as peak flows magnitudes and water yield, increased over time, likely due to deforestation and other changes in land use associated with gold mining. The correlation between the cumulative area of gold mining and active channel width was found to be stronger in the Puquiri-Colorado River compared to the Inambari River, indicating a more pronounced impact of gold mining on the former. Finally, multiple linear regression analysis confirmed the role of hydrological and land-use change as drivers of active channel width. Overall, this study highlights the substantial geomorphic impact of artisanal gold mining operations on the Puquiri-Colorado River. Taken together, the findings of this research contribute to a deeper understanding of the geomorphic processes governing alluvial rivers under varying sediment supply regimes, by elucidating the complex interactions between sediment pulses, hydrological processes, and landscape characteristics. In a broader context, these findings also highlight important watershed management challenges related to ongoing and future environmental change. In Lillooet River, well-defined changes in planform morphology due to the 2010 landslide were mostly detected in the initial five years, reversing the long-term trend of active channel narrowing, and were localized upstream of the confined region of alluvial fans. However, channel response is expected to move progressively downstream, causing concerns about natural hazards to the downstream communities. Rising temperatures and continuing deglaciation provide new sources of sediment to the channel, which, coupled with more frequent flood events, can facilitate the transport of coarse sediment and enhance the lateral activity of the river, as indicated by increases in active channel width. Changing climatic conditions are also likely to result in an elevated risk of further catastrophic landslides. In the Puquiri-Colorado River system, ongoing alluvial gold mining is altering the original valley bottom, introducing vast quantities of sediment into the river and facilitating storage of this material as “legacy sediment” within its floodplain, with associated contaminant storage. Recovery times in this river are difficult to estimate, particularly since mining activities are still active partially, and promoting continued widening of the river. Thus, continued tracking of the response of these rivers is required in order to understand fully the long-term response of these rivers. Finally, the methodologies and results presented herein can serve as a framework for future studies aimed at assessing and managing riverine ecosystems amidst increasing anthropogenic pressures and climate uncertainties.

Graduation Semester

2024-12

Type of Resource

Thesis

Handle URL

https://hdl.handle.net/2142/127456

Copyright and License Information

Copyright 2024 Mishel Milagros Melendez Bernardo

Owning Collections