Characterizing Subsurface Complexity of Aeolian Morphotypes with Georadar

Abstract

Aeolian landforms are classified based on their plan morphology, which is a function of sediment transport volume, wind direction, and vegetation. In the case of compound landforms or two-dimensional exposures (outcrops), there is insufficient information for discriminating between 3D morphotypes (e.g., barchans vs. parabolic dunes). To characterize the dip-section architecture of near end-member morphologies (interacting barchans and sparsely vegetated parabolics), a series of axial transects were selected from >25 km of high-resolution (500 MHz) ground-penetrating radar (GPR) data from the gypsum dune field of White Sands National Monument, New Mexico. For dunes of comparable size (6-7 m high), a series of attributes were analyzed for unsaturated portions along the thickest (axial) radargram sections. Given the limitations in vertical resolution (7 cm in dry sand), the average measureable slipface thickness in barchans ranged between 10-22 cm, whereas parabolic slipfaces were thinner at 10-14 cm. High-amplitude diffractions produced by buried vegetation, semi-lithified pedestals, and bioturbation structures were rare within barchans (point-source diffraction density = 0.03/m2; hyperbolics per 1-m-wide cross-sectional area of the image), in contrast to a point-source density of 0.07/m2 in parabolics. An aeolian internal complexity threshold (ϖ) is proposed, which incorporates standardized scores of slipface thickness, point-source diffraction density, and continuity of major bounding surfaces at mesoscale range determined through semivariogram analysis. For the study region, these variables were sufficient for discriminating barchans (ϖ = -2.39 to -0.25; ϖ ̅b= -1.65) from parabolic (ϖ = 0.13 to 2.87; ϖ ̅p= 1.65) dunes. This threshold has the potential for differentiating dune morphotypes in areas where surface morphology is masked and for identifying compound landforms (e.g., a re-activated parabolic dune converted into a barchan in situ). Ultimately, characterization of bedding complexity in ancient aeolian sequences will provide useful information about key paleoenvironmental variables.