Constraining landscape sensitivity to climate change using geomorphological and sedimentological approaches

Abstract

Climate exerts a profound control on the processes that shape landscapes and produce the sedimentary deposits with which we can interpret the Earth’s history. However, we lack a complete understanding of how sensitive tectonically-active, eroding landscapes are to climate and climate change. How does a simple sediment routing system react to a change in rainfall rate? Can mountainous landscapes respond quickly enough to preserve a record of high-frequency climate changes, e.g., glacial-interglacial cycles? What effect does headwater glaciation have on downstream sediment characteristics? Can we quantify past climate changes using the sedimentological properties of terrestrial stratigraphy? Geologists lack complete answers to these questions, among many others. Theoretical work, using physical first principles and numerical models, has produced a range of hypotheses about landscape sensitivity to climate, but we now need empirical data to test and make sense of these ideas. This thesis therefore explores empirically how geomorphological and sedimentological records have responded to climatic gradients across time and space. In the first part of this thesis, the extent to which spatial climate gradients are recorded by the longitudinal geometry of river channels is investigated. I use a simple stream power erosion law to predict an inverse relationship between channel steepness and average precipitation rate, and then test this theory using data from a variety of study areas and two complementary analytical approaches. Climate is found to be an important control on river longitudinal geometry across a range of climatic and tectonic conditions, in a way that conforms to existing theoretical knowledge and also allows the climatic signal to be discriminated from tectonics. This work therefore demonstrates that a widely-used geomorphological measurement—the channel steepness index—is quantifiably sensitive to climate in tectonically-active areas, and these findings offer a new explanation for geographic variations in channel steepness that cannot be explained by tectonics alone. The second part of this thesis focuses on the sensitivity of simple mountain catchment-alluvial fan systems to climate changes associated with the last glacial-interglacial cycle, as expressed in the south-western United States. First, eight debris flow-dominated systems located in the south-eastern Sierra Nevada, California are examined. I establish a detailed chronostratigraphic model for these fan systems by building upon and integrating existing exposure age constraints reported by others, and additionally developing a new technique for estimating the ages of these fan deposits. This technique is based on calibrating the rate of enlargement of common weathering fractures observed in exposed surface boulders, which are shown to widen at a steady and predictable rate post-deposition, and can be used as reliable age indicators for > 100 ka at this location. Using the detailed temporal record of deposition established for these fan systems, a large (> 30,000 particle) grain size data set that spans the last full glacial-interglacial cycle is examined. I demonstrate that debris flow grain size is a highly sensitive recorder of past climate changes, capturing the glacial-interglacial cycle as a sustained and high-amplitude time series with a rapid response timescale of < 10 ka. These debris flow deposits become significantly coarser-grained with warming and overall drying of the climate, and this thesis outlines quantitative reasons why this signal can be attributed to increasing storm intensity with warming. Finally, these debris flow-dominated systems are contrasted with two carefully-selected stream flow-dominated fan systems in Death Valley, California. Using measures of down-system grain size fining and a self-similarity model of sediment calibre, sediment flux estimates during arid interglacial and wetter glacial climate conditions are derived and compared. This study shows that a decrease in average rainfall rate of ~ 30 % produced a corresponding decrease in sediment flux of ~ 20 %. However, I also demonstrate the circumstances in which signal buffering due to incision and sediment recycling destroys this climate signal. Consequently, this thesis demonstrates both the causes and results of complexity in the relationship between climate change, geomorphology, and well-dated terrestrial sedimentary records. Ultimately, this is an expression of how sediment transport processes, tectonics, the magnitude-frequency distribution of rainfall, and other factors interact to generate different climate responses in different systems. Nevertheless, for both geomorphic and sedimentological records examined here, I demonstrate that the effects of climate can be quantified clearly: channel steepness can be quantified as a function of rainfall rate; debris flow sedimentology can be quantified as a function of storm intensity; and alluvial fan sedimentology in Death Valley can be quantified as a function of glacial-interglacial climate changes. Essentially, this thesis finds that terrestrial landscapes are sensitive to known climate changes in the recent geological past, and this result is profoundly important for improving our ability to decode geomorphic and stratigraphic archives effectively. The data and ideas within this Ph.D. research provide useful opportunities for (i) testing and updating our models of how sediment routing systems respond to climate, (ii) extracting quantitative information about past climates from the sedimentary record, and (iii) predicting the effects of future climate changes on the landscape.