River response and climate change in the Wet Tropics, northeast Queensland Australia

Abstract

An improved understanding of the relationship between climate change and river response is critical for predicting future trajectories of fluvial landforms. This is particularly important with climate change projections showing an increase in extreme events such as floods and cyclones. In Australia and internationally, a valuable record of river response has been developed from the morphologic and sedimentologic archives of river terraces and floodplains, mainly in temperate regions. These studies reveal the key role of climate over timescales ranging from 100 kyr glacial/interglacial cycles through to decadal cycles driven by the El Nino Southern Oscillation (ENSO). However, the understanding of fluvial records in the humid (wet) tropics is comparatively poor despite these regions being geomorphic hotspots for catastrophic change. The relevance of key geomorphic concepts to wet tropical rivers is largely unknown and therefore the knowledge gained from studying these settings is invaluable. The overarching aim of this thesis is to characterise the nature of regional river response and its relationship to climatic change in the Wet Tropics of tropical northeast Queensland, Australia. Chapters 2 to 6 explore the various aspects of this research question and an overall synthesis of the thesis key findings and future research priorities is presented in Chapter 7. Chapter 2 presents the research on terrace distribution and chronostratigraphic characteristics to provide the overall spatial context of river landforms across five catchments. Two terraces are correlated and mapped across the study area, while the terrace chronology details two periods of aggradation-incision that post-date the Last Glacial Maximum. Explanations for spatial differences in terrace preservation are explored using a range of catchment force-resistance variables with the results underpinning a new conceptual model of terrace types and preservation processes. Chapter 3 details the numerous morphological and hydrological characteristics of Wet Tropics floodplains. An array of data and techniques is applied across four catchments and yields the first comprehensive geomorphic dataset for floodplains in the region. Floodplains exhibit complex surfaces dominated by erosional forms and are inundated by up to 6 m during annual wet season floods. Floodplain morphology is found to reflect the complexity of the tropical hydrological regime, rather than simple estimates of channel hydraulics. In Chapter 4, presentation of floodplain chronostratigraphic data provides the longer-term context of floodplain formation and is used in conjunction with the results from Chapter 3 to develop a conceptual model of floodplain formation for the Wet Tropics. Floodplain formation occurs in five phases that span the last 1.5 kyr. Individual phases alternate between domination by erosion or deposition processes. Floodplain erosion occurs through wholescale and partial stripping most likely during wetter climatic periods, while depositional phases appear to be associated with both wetter and drier climates. The new conceptual model advances some earlier models of floodplain construction and destruction by acknowledging: (i) the time frames of aggradation and incision, (ii) the roles of different destruction processes, (iii) the integration of detailed sedimentology and stratigraphy, and (iv) an understanding of climatic variability. Chapter 5 enhances understanding of hydroclimate variability in the Wet Tropics by examining the potential of high-resolution coral proxies from the Great Barrier Reef (GBR) to reconstruct river discharges. A good relationship is found between annual coral luminescence and wet season discharge from the North Johnstone River, which is used to develop the first centennial-scale reconstruction of hydroclimate for the last 400 years. The reconstruction highlights the existence of several cyclical periods of enhanced and reduced flooding that were not previously identified in broader reconstructions of GBR hydroclimate. Analyses of instrumental records reveal that large floods are often associated with cyclones. Based on this association, an existing, published reconstruction of cyclone activity is used to infer on periods of flooding and thus provide an independent line of evidence for long term hydroclimatic conditions in the Wet Tropics. Chapter 6 provides an overall synthesis of the findings presented in Chapters 2 to 5. This chapter collates the record of fluvial response for the past 30 kyr from the nature and timing of alluvial aggradation and incision as recorded in terraces and floodplains. A new 30-kyr record of climate change is also synthesised from existing palaeoclimate proxy records for the region. Over the past 30 kyr, fluvial response appears relatively synchronous across the region with four phases of aggradation and incision. Comparison of the fluvial and palaeoclimate records reveals the relationship between river response and climatic changes is not straightforward and likely reflects important catchment differences in response and recovery times. This record presented for the Wet Tropics also illustrates that there are important similarities with the alluvial records from southeastern Australia.