Greedy parabolics: Wind flow direction within the deflation basin of parabolic dunes is governed by deflation basin width and depth

Abstract

Parabolic dunes are ‘U’ or ‘V’-shaped aeolian landforms that form on pre-existing sand deposits. Their morphology consists of an upwind deflation basin, bordered by often vegetated trailing arms and a downwind depositional lobe. The orientation of parabolic dunes is commonly attributed to the prevailing or resultant wind direction. Consequently, the orientation of parabolic dunes stabilised by vegetation growth has been used as a proxy for wind direction during past climates in several studies. However, the ability or extent of parabolic dune morphology to steer incident wind flow parallel to the orientation of the landform, and thus migrate in the direction of the current landform orientation rather than prevailing wind direction, is unknown. By numerically modelling wind flow within the deflation basin of eight parabolic dunes, we demonstrate for the first time that wind flow direction within the deflation basin of a parabolic dune is highly controlled by the depth and width of the deflation basin. The greater the depth–width ratio of the landform (i.e. the deeper and narrower the deflation basin), the greater the degree of flow steering relative to the axis orientation of the landform. These results demonstrate that future studies must exercise caution when using parabolic dune orientation as a direct proxy for prevailing wind direction, especially where parabolic dunes have a relatively high deflation basin depth–width ratio, as the deflation basin of these landforms may continue to migrate in an antecedent wind direction.

Aeolian sand transport thresholds in excavated foredune notches

“…The normalised wind speed profiles at lines A, B and D inside the notch are compared with field observations at masts A, B and D, respectively. This validation method is commonly used in CFD studies on coastal dunes (Smyth et al, 2020; Wakes et al, 2010).…”

“…During the last decade, flow–form–sand transport interactions have been studied on various coastal dune landforms, including foredunes (de Winter et al, 2020; Delgado‐Fernandez et al, 2013; Hesp & Smyth, 2016; Hilton et al, 2016; Jackson et al, 2013, 2011; Lynch, Jackson, & Cooper, 2009; Lynch, Jackson, & Cooper, 2010; Petersen, Hilton, & Wakes, 2011; Schwarz et al, 2021; Wakes, Bauer, & Mayo, 2021; Wakes, Hilton, & Konlechner, 2016; Wakes et al, 2010; Walker et al, 2017), foredune scarps (Hesp & Smyth, 2019; Piscioneri, Smyth, & Hesp, 2019), blowouts (Garès & Pease, 2015; Hesp & Walker, 2012; Hugenholtz & Wolfe, 2009; Pease & Gares, 2013; Smyth, Jackson, & Cooper, 2012; Smyth, Jackson, & Cooper, 2013; Smyth et al, 2019; Smyth, Jackson, & Cooper, 2014), parabolic dunes (Anderson & Walker, 2006; Delgado‐Fernandez et al, 2018; Hansen et al, 2009; Smyth et al, 2020; Wakes, 2013), nebkha (Hesp & Smyth, 2017; Zhao et al, 2019, 2020, 2021), sand cays (Hilton et al, 2019), beaches with large woody debris (Grilliot, Walker, & Bauer, 2018; Grilliot, Walker, & Bauer, 2019), and beach scarped ridges (Smyth & Hesp, 2015). These studies have contributed to our understanding of flow–form–sand transport relationships in a wide range of aeolian situations.…”

“…The normalised wind speed profiles at lines A, B and D inside the notch are compared with field observations at masts A, B and D, respectively. This validation method is commonly used in CFD studies on coastal dunes(Smyth et al, 2020;Wakes et al, 2010).The CFD profiles substantially overestimate windspeed compared with the observed measurements. The deviation between the CFD and observed speeds depend on location and appears to be approximately 25% at 2.5 m, 15% at 1.05 m, 10% at 0.46 m and 5-10% atF I G U R E 3 Computational domain: (a) domain orientations, dimensions and boundary definition used in the CFD modelling.…”

Wind flow dynamics and sand deposition behind excavated foredune notches on developed coasts

“…One of the key advantages of CFD is that it is easily adaptable once a simulation is constructed, and boundary conditions such as wind speed and wind direction can be quickly altered to test a range of hypotheses (Araújo et al, 2013;Hesp et al, 2015;Jackson et al, 2013Jackson et al, , 2020Nguyen et al, 2021b;Wakes et al, 2016Wakes et al, , 2021. Simulations can also provide flow information at a landscape scale, such as a foredune system (Hesp et al, 2015;Pattanapol et al, 2011), parabolic dunes (Delgado-Fernandez et al, 2018;Smyth et al, 2020) and a dunefield (Smith et al, 2017(Smith et al, , 2021 where it is expensive to instrument and conduct effective field experiments. The current study applies CFD for two purposes: (i) to examine the pattern of flow in the lee of the notch during shorenormal incident wind conditions which were not observed during the field experiment (research question 1); and (ii) to examine the pattern of flow behind the notched foredune system (research question 2).…”

“…The renormalization process enables the Navier-Stokes equations to account for smaller scales of motion than the standard k-ε model, which employs a single turbulence length scale (Smyth, 2016). Previous work has shown that the k-ε RNG turbulence model modelled wind flow well, compared with field observations, over complex dune landforms such as a foredune (Hesp et al, 2015;Pattanapol et al, 2011;Wakes et al, 2010), blowouts (Smyth et al, 2012(Smyth et al, , 2019 and parabolic dunes (Delgado-Fernandez et al, 2018;Smyth et al, 2020;Wakes et al, 2016). A second-order spatial discretization scheme was employed to interpolate values between cell centres, and calculations were considered complete once the initial residual of each iteration was <0.0001 m s À1 for U X , U Y and U Z .…”