NG43A-01
Physically Based Numerical Model of Wind-Blown Sand Suggests Deficiencies in Classical Saltation Theory
Wind-blown sand, or ‘saltation,' is an important geological process, as well as the primary source of atmospheric dust aerosols. Despite its importance in geological and atmospheric processes, saltation is not well understood yet. We present here the first physically based model of saltation that is rigorously tested with data from field measurements. The model explicitly simulates the trajectories of sand grains moved by wind, calculates the wind profile resulting from the transfer of wind momentum to saltating grains, and accounts for the impact of saltating grains with the soil bed, including the ejection of surface grains. The results of the model, verified by field measurements of saltation mass flux profiles, reveal some surprising deficiencies in the classical understanding of saltation. Model results strongly suggest that the common assumption that particle speed scales linearly with the wind shear velocity (Bagnold, 1941) is incorrect. Moreover, the common hypothesis that the fluid shear stress at the surface remains at the threshold value for particle entrainment results in large discrepancies between model results and experiments. This classical hypothesis, proposed by Owen (1964), is thus probably incorrect. Our model is general and can be adapted to study saltation under a variety of physical conditions, including on sloped terrain such as sand dunes, and on other planetary bodies such as Mars, Venus, and Titan.
NG43A-02
Process-scaling Issues of Saltation Dynamics in Wind Tunnel and Field Environments
This paper examines the scaling up process of aeolian sand transport models from wind tunnels to prototype beach and desert environments. Analyses of enhanced roughness lengths derived from wind velocity profiles and particle trajectory characteristics demonstrate that airflow - surface relationships and subsequent transport models derived from wind tunnel studies should not be applied to natural systems without scaling corrections. Specifically, the scaling constraints of wind tunnels impact aeolian processes in two ways. Firstly, wind tunnel geometry inhibits the development of turbulent coherent flow structures and, thus, particle trajectories are reduced in height and length, and secondly, the saltation enhanced roughness is much smaller in wind tunnels; probably for the same reason. Until these process scaling issues are explicitly addressed in wind blown sand studies, the accuracy of transport models will continue to be poor except by accident.
NG43A-03
Turbulent Flow and Sand Dune Dynamics: Identifying Controls on Aeolian Sediment Transport
Sediment transport models are founded on cubic power relationships between the transport rate and time averaged flow parameters. These models have achieved limited success and recent aeolian and fluvial research has focused on the modelling and measurement of sediment transport by temporally varying flow conditions. Studies have recognised turbulence as a driving force in sediment transport and have highlighted the importance of coherent flow structures in sediment transport systems. However, the exact mechanisms are still unclear. Furthermore, research in the fluvial environment has identified the significance of turbulent structures for bedform morphology and spacing. However, equivalent research in the aeolian domain is absent. This paper reports the findings of research carried out to characterise the importance of turbulent flow parameters in aeolian sediment transport and determine how turbulent energy and turbulent structures change in response to dune morphology. The relative importance of mean and turbulent wind parameters on aeolian sediment flux was examined in the Skeleton Coast, Namibia. Measurements of wind velocity (using sonic anemometers) and sand transport (using grain impact sensors) at a sampling frequency of 10 Hz were made across a flat surface and along transects on a 9 m high barchan dune. Mean wind parameters and mass sand flux were measured using cup anemometers and wedge-shaped sand traps respectively. Vertical profile data from the sonic anemometers were used to compute turbulence and turbulent stress (Reynolds stress; instantaneous horizontal and vertical fluctuations; coherent flow structures) and their relationship with respect to sand transport and evolving dune morphology. On the flat surface time-averaged parameters generally fail to characterise sand transport dynamics, particularly as the averaging interval is reduced. However, horizontal wind speed correlates well with sand transport even with short averaging times. Quadrant analysis revealed that turbulent events with a positive horizontal component, such as sweeps and outward interactions, were responsible for the majority of sand transport. On the dune surface results demonstrate the development and modification of turbulence and sediment flux in key regions: toe, crest and brink. Analysis suggests that these modifications are directly controlled by streamline curvature and flow acceleration. Conflicting models of dune development, morphology and stability arise when based upon either the dynamics of measured turbulent flow or mean flow.
NG43A-04
Modeling Aeolian Sediment Transport Thresholds on Mars: A Shear Stress Partitioning Approach
This presentation explores the effect that larger roughness elements may have on entrainment of sediment by Martian winds using a shear stress partitioning approach based on a model developed by Raupach et al. (1993). This model predicts the shear stress partitioning ratio, i.e., the percent reduction in shear stress on the intervening surface between the roughness elements as compared to the surface in the absence of those elements, as a function of the geometric properties of the roughness elements, the characteristic drag coefficients of the elements and the surface, and the assumed effect these elements have on the spatial distribution of the mean and maximum shear stresses. Recent work by King et al. (2005) and field testing by Gillies et al. (2006, 2007) have demonstrated that this model can effectively predict roughness effects on entrainment threshold for terrestrial atmospheric conditions. As the model can be used for different fluid properties it makes it applicable to evaluating how the atmospheric conditions and winds of Mars interact with the roughness to predict the effect that roughness will have on sediment entrainment there. Utilizing the results of Gillies et al. (2007), which provides data on the effect large solid roughness elements of varying roughness density have on shear stress partitioning and particle threshold and aerodynamic parameters, this paper will use the shear stress partitioning model to evaluate how Martian atmospheric conditions will affect threshold on Martian surfaces with the same range of roughness evaluated by Gillies et al. (2006, 2007). Finally, based on available estimates of roughness density for Mars drawn from the literature, estimates of threshold wind speeds for these rough surfaces will be presented.
NG43A-05
Sediment Transport and Dust Flux in Disturbed and Undisturbed Dryland Ecosystems: From Site Specific Estimates to Trends Across Gradients of Woody Plant Cover
Aeolian sediment transport and associated dust flux are important processes in dryland ecosystems where vegetation cover is inherently sparse relative to more mesic ecosystems. Aeolian processes in dryland ecosystems are strongly influenced by the spatial density of roughness elements, which is largely determined by woody plant height and spacing. Despite the global extent of dryland ecosystems, relatively few measurements of aeolian sediment transport have been made within these systems, and these few existing measurements have not been systematically evaluated with respect to gradients of woody plant cover. We report measured aeolian sediment transport in an undisturbed and disturbed semiarid grasslands in southern Arizona. To place our estimate in a broader context, we compared our site-specific findings to other recently published measurements of aeolian sediment transport in disturbed and undisturbed dryland ecosystems. We propose a new conceptual framework for dryland aeolian sediment transport and dust flux as a function of woody plant cover that integrates our site-specific data with the broader literature base. Our findings suggest that for relatively undisturbed ecosystems, shrublands have inherently greater aeolian sediment transport and associated dust flux than grasslands, woodlands and forests due to wake interference flow associated with the height and spacing of woody roughness elements. Furthermore, the proposed framework suggests that for disturbed ecosystems, the upper bound for aeolian sediment transport increases as a function of decreasing woody plant cover. As a result, aeolian sediment transport spans a relatively small range in woodlands and forests, an intermediate range in shrublands, and the largest range in grasslands. Our framework is applicable both within locations and across broad gradients
NG43A-06
Influence of Temporal Scales and Soil Properties on Dust Emissions: a Measurement and Modeling Effort
Dust emission events in deserts are usually modeled as large magnitude annual or biannual events driven by larger-scale meteorological processes (e.g., frontal passages, thunder cell downdrafts), partly owing to the relatively coarse temporal and spatial scales of observation. However, there is evidence that smaller, more frequent wind events (either through sandblasting or direct aerodynamic entrainment), may play a previously unknown, but important role in dust emissions, suggesting that current models do not account for all of the dust that is emitted into the atmosphere. This has important implications from the perspective of landscape dynamics for the rates of replenishment of surface dust, principally through processes of deposition or production of silt- and clay-sized particles. Three fundamental research questions remain largely unanswered. The first is: Do large-scale dust storms contribute the majority of fugitive dust to the atmosphere? The spatial and temporal complexity of wind erosion leads to dust emissions that are often too small to be detected by remote sensing and we can ask "do smaller and likely more frequent events (in comparison to annual major dust storm events) result in a large proportion of dust emitted from arid environments? The second question to be addressed is: What is the role of direct entrainment of desert dust in contributing to regional dust emissions? Only recently has research suggested that direct entrainment is a process that can emit a significant amount of dust into the atmosphere. We are investigating whether there is a temporal limitation (associated with deposition or more generally, replenishment) that bounds the magnitude of these types of emissions. The last question to be addressed is: How do soil surficial characteristics affect the temporal aspects of dust emissions. Exposed surfaces in arid climates are often covered with a variable thickness of salt crust. These crusts can vary spatially and temporally between loose individual silt to sand-sized crystals to hard crusts composed of interlocking crystals. In this aspect of the research we aim to examine how surficial salt crust morphology and mineralogy affect dust emissions on various spatial and temporal scales. This paper attempts to provide preliminary answers to these questions by presenting data which: (1) characterize the potential of different soil types to emit dust, based on measurements made using new wind tunnel technology, (2) quantify the vertical dust and horizontal sediment fluxes in conjunction with the physical, chemical, and mineralogical (salt) soil characteristics obtained with new analytical techniques (e.g., XRD, SEM) and, (3) define a temporal scale and associated time-lag for soil surfaces to emit fugitive dust using a new model based on soil profile characteristics (i.e., resistance, genesis and seasonality). Results of this work have a direct impact on the understanding of aeolian and soil surface dynamics and will result in improved accuracy and temporal resolution of dust emission models.
NG43A-07
Dynamics of a Barchan Dune Field: a Discrete Numerical Model
Barchans are crescent-shaped dunes that form on solid ground in areas with a relatively low sand supply and a unidirectional wind regime. Isolated barchans have been successfully modeled with regard to their shape and propagation velocity. However, emergent effects that arise for the case of a field of dunes have proven difficult to capture. These behaviors include selection of a preferred size and spacing within a patch of dunes and additionally the presence within a dune field of multiple patches, greatly extended in the downwind direction, each exhibiting a different dominant size. It is suspected that these sorting inhomogeneities in the dune field are self- organized and not the result of external forcing. Here, we present the results of modeling efforts using a discrete numerical model representing a field of barchan dunes. We use simplified equations for dune shape, mass balance, and propagation. Dunes interact by merging and by means of the downwind sand flux. Additionally, we include a simplified treatment of dune calving. Tentative conclusions can be drawn from the rich behavior of the model. In it, spatial inhomogeneities can arise due to feedbacks triggered by stochastic fluctuations about critical values of the input parameters. Isolated groups propagate at velocities independent of those of their constituent dunes. Size selection occurs to a limited extent due to the onset of calving at a critical size. In sum, the model displays some of the emergent dune field characteristics that have not previously been replicated.
NG43A-08
Cellular Automaton Simulation of Vegetated Dune Field Dynamics
Vegetated aeolian dune fields develop through non-linear interactions between physical geomorphic processes and ecological vegetation growth and response into complex ecogeomorphic systems that are sensitive to both climatic and environmental variations. We present a Discrete Ecogeomorphic Aeolian Landscape (DECAL) cellular automaton model that replicates the self-organisation of vegetated dune systems and enables the investigation of conditions necessary for long-walled (hairpin) parabolic dune and nebkha formation in coastal and semi-arid environments over various temporal and spatial scales. The algorithm utilises simple transport rules and mutual feedback between geomorphic and ecological components to investigate vegetation pattern formation and how and why this influences dune dynamics. We examine ecogeomorphic interactions both by exploring system mechanics via dune mobility and by more descriptive numerical state variables, facilitating the investigation of trajectories and potential attractors as a function of environmental parameters and system perturbations and leading to the identification of possible system sensitivities and thresholds. The model simulations elucidate possible dune field responses to anthropogenic impacts and palaeo and future climate variations and highlight the ability of vegetation to impart a characteristic length-scale on a landscape. This simple vegetated dune model illustrates the power and versatility of a cellular automaton approach for exploring ecological and geomorphic interactions in complex earth surface systems.