Controls on Yardang Development and Morphology: 1. Field Observations and Measurements at Ocotillo Wells, California

Abstract

Yardangs are streamlined hills formed in part by the erosive action of wind and wind‐blown sediments. Here we examine the controls on yardang development and morphology using the Ocotillo Wells State Vehicular Recreation Area (OWSVRA), California, as a study site. We measured the compressive strengths, strikes, and dips of bedrock strata, eolian sediment fluxes (including their vertical profiles and spatial variations around yardangs), and erosion rates derived from geologic constraints and multitemporal Terrestrial Laser Scanning (TLS). We used a combination of TLS‐based and airborne lidar‐based Digital Elevation Models (DEMs) to test the applicability of an asymmetric Gaussian function for characterizing yardang form and quantify the relationships among yardang lengths, widths, heights, spacings, and their controlling factors. Yardang aspect ratios are controlled by bedrock structural attributes, specifically by the tangent of the dip and the angle between the strike and the prevailing wind direction. Yardang spacings scale linearly with yardang width. Yardang heights increase as the square root of width such that larger yardangs tend to have gentler side slopes. Sediment fluxes reach a maximum in the troughs among yardangs, consistent with the hypothesis that yardang development involves the focusing of wind and wind‐blown sediments into troughs. The vertical distribution of eolian sediment flux follows a power law with an exponent of −2.5, a result consistent with an advection‐diffusion‐settling model of transport near the saltation‐suspension transition. Erosion rates are several mm/yr over time scales of ~100 and ~106 years.

Controls on Yardang Development and Morphology: 2. Numerical Modeling

“…The breakdown of this assumption is one reason why Anderson () developed a separate set of equations for abrasion by saltating particles and applied equation only to suspended particles. However, in the companion paper (Pelletier et al, ), we showed that the concentration and flux profiles obtained in OWSVRA are well represented by an advection‐diffusion‐settling model at the saltation‐suspension transition. Moreover, in section 3.3 of this paper, I demonstrate that the advection‐diffusion‐settling model predicts abrasion profiles consistent with those measured by Sharp () in the Coachella Valley.…”

“…Inputs to the PHOENICS model runs described in this section include terrain models representing idealized yardangs with a wide range of aspect ratios, an aerodynamic roughness length ( z 0 ), and a prescribed horizontal velocity at a reference height far from the ground ( u r = 10 m/s at z r = 100 m was used for all of the model runs of this section). The ground surface is prescribed to be a fully rough boundary, that is, one that results in a law‐of‐the‐wall velocity profile, that is, where u ( z ) is the wind velocity at a height z above the bed, u * is the shear velocity constrained using the reference velocity u r , κ is the von Kármán constant (0.41), and z 0 is the aerodynamic roughness length (assumed to be 1 mm for all of the model runs associated with this section, i.e., at the upper end of values measured in OWSVRA; Pelletier et al, ). At the upwind boundary of the model domain an inlet law‐of‐the‐wall velocity profile is prescribed with u * and z 0 values consistent with the reference velocity, reference height, and the z 0 value prescribed for the ground‐surface boundary condition within the model domain.…”

“…For the purposes of computing the dependence of the yardang drag coefficient on the aspect ratio, yardangs were modeled as asymmetric Gaussians, that is, where z is elevation above the base of the yardang, z max is the height of the yardang crest (assumed to be 3 m for all of the cases associated with this section), x is the distance downwind from the yardang crest along the yardang axis, y is the distance from the yardang axis along a direction perpendicular to the axis, and x l and x w are characteristic dimensions of the yardang in the directions parallel and perpendicular to the wind, respectively. The companion paper (Pelletier et al, ) demonstrated the adequacy of equation for representing the first‐order morphology of yardangs in OWSVRA. The yardang width scales with 2 x w , while the length scales with x w + x l .…”

“…The concentration profile over a horizontal surface predicted by the advection‐diffusion‐settling model of eolian sediment transport is (Anderson & Hallet, ) where R # is the Rouse number, defined as w s /κ u * where w s is the particle settling velocity. The companion paper (Pelletier et al, ) demonstrated the efficacy of equation for characterizing measured sediment flux data in the vicinity of yardangs at OWSVRA.…”

“…The literature supports the utility of the diffusion equation for quantifying hillslope evolution by freeze‐thaw‐driven creep and rain splash, provided that the slope is transport‐limited (i.e., soil is present) and slope gradients are modest (Dunne et al, ; Furbish et al, ; McKean et al, ). Pelletier et al () documented how empirical models for interrill and rill erosion may also be approximated by the diffusion equation. The topographic diffusivity is the fundamental parameter of the diffusion equation, representing the efficiency of water‐driven sediment flux for a given slope.…”