Landscape evolution of the southeast Queensland dune fields.

Abstract

The sub-tropical coastal dune fields of southeast Queensland, Australia are recognised for their outstanding beauty, cultural importance, and ecological diversity. Their soil and vegetation development have been intensively investigated and more recently, the geochronology of the dune fields has been expanded. However, there has been little focus on the evolution of the dune fields after the dunes stabilise. The aim of this thesis is to enhance our fundamental understanding of dune field evolution by evaluating the complete topographic development of the dunes from their emplacement (stabilisation) to maturity (denuded topography). Principals and concepts derived from hillslope geomorphology were used to determine styles and rates of landscape change. The Holocene section of the Cooloola Sand Mass (CSM) is the primary focus of this study, while Holocene dunes on K’gari (Fraser Island) were also investigated. The dune fields were selected because most of the major environmental factors contributing to landscape development in the Holocene can be constrained, and they contain one of the most complete coastal dunes sequences in the world.

Quantitative topographic analyses from high-resolution digital elevation models with landscape evolution theories (linear and nonlinear sediment transport) were used to better describe and understand dune fields and dune landforms. Principally, the foundational idea that gravitationally driven transport processes smooth dune landforms to their base-levels thereby reducing mean local relief, was used. The concept that landscape smooths with time provides the framework to establish morphostratigraphical mapping, 2-D numerical modelling, and roughness-age modelling.

From the geomorphological mapping it is observed that the SE Queensland dune fields are constructed of five Holocene (including active dunes) and four Pleistocene dune morphosequence units. Dunes and their units systematically smooth with time and this evolution is well explained using surface roughness (σC). It is demonstrated that Holocene dune σC-age relationships evolve in two distinct phases. The first phased is described well using nonlinear sediment transport with a soil transport coefficient (K) value of 0.06 mm² yr⁻¹ and a critical gradient of 0.65 m m⁻¹, which is the angle of repose. The dune evolution switches to a K value of 0.002 m² yr⁻¹ after ca. 1 ka that can be modelled either using nonlinear or linear sediment transport.

The evolution of the whole landscape can be empirically described using an exponential function ((∂σC)/∂t ∝ σC). The predictable changes in dune topography permits a σC-age relationship to be calibrated on the CSM and tested against an independent OSL chronology from K’gari. The model generates age estimates for every dune thereby producing the first complete Holocene chronology in the dune field. This procedure can be easily expanded to dune fields globally to fill in chronological gaps using only high-resolution elevation data and a small number of absolutely dated dunes. The age estimates support the morphostratigraphical mapping and demonstrate dune emplacement peaking at ca. <0.5, 1.5, 4, and 8.5 ka. These ages are similar, but not identical, to the dune emplacement timings in the published literature but they tie closely to sea-level variability, which is the inferred primary cause of dune field activation.

The σC-age relationship was evaluated further by placing the modelled outputs into the context of sedimentary records from dune foot slope positions. The first sediment transport phase corresponds with the period of rapid lowering of relief and elevated erosion/sedimentation rates (0.47 ± 0.08 cm yr⁻¹) associated with the dominance of episodic sediment transport (i.e., dry-ravel and sheetwash). These transport styles are the consequence of fires on steep hillslope gradients. These events deposit charcoal as layers in foot-slope positions. This phase occurs for the first ca. 1-1.5 ka after dune emplacement until hillslopes are lowered below their angle of repose (0.65 m m⁻¹ or 33°). In the second phase, erosion/sedimentation rates decrease by an order of magnitude (0.08 ± 0.05 cm yr⁻¹) due to the dominance of slow and continuous sediment transport processes (i.e., biogenic soil creep and granular relaxation). Although fires are present, the absence of episodic sediment transport results in disseminated charcoal rather than charcoal layers in foot-slope positions. Nevertheless, fire frequency and intensity can be inferred from these records and the thesis highlights and develops the idea of utilising dune depositional records for fire histories. These deposits produce a ca. 7 cal ka BP fire record that identifies increased fire activity at ca. <0.1, 1.1-0.3, 2.2-1.8, 3.6-2.6, and 6.7-5.3 cal ka BP. These periods are consistent with local and regional fire histories from traditional charcoal records.

In summary, this thesis contributes new insights into landscape evolution using a dune field as a natural sandbox laboratory. It offers a novel perspective on aeolian systems and provides new lines of research into a variety of environmental processes from dunes.