Quantifying vegetation and its effect on aeolian sediment transport: A UAS investigation on longitudinal dunes

Studying the role of vegetation in regulating aeolian sediment transport is complicated by the diversity of plant geometry and spatial distribution. Using Unmanned Aerial Systems (UAS) surveys of four partially vegetated sand dunes in the Simpson Desert, this study explored statistical associations between vegetation and the location and quantity of aeolian ripples. Employing mosaic image classifications, Digital Surface Models (DSM), and Canopy Height Models (CHM), four core independent metrics were computed: The fractional cover (f_c); frontal area index (λ), mean gap length (L), and shadow casting or Shadow Area Ratio (SAR). The strongest predictor of aeolian ripple abundance was the mean scaled gap length (individually scaled by the lesser of an adjacent plant’s width or height) ($\stackrel{-}{L_{\mathrm{sf}}}$) (R² = 0.83). $\stackrel{-}{L_{\mathrm{sf}}}$ (and $\stackrel{-}{L_h}$, which only used plant height) effectively resolved the spatial and structural distribution of vegetation, which was partially governed by the composition of functional plant types. f_c was also strongly associated with ripple abundance (R² = 0.81). Ripple cover varied continuously with f_c without a clear threshold for the onset of sand transport, though the curve flattened above f_c ≈ 25–30%. Moderate associations were found for SAR (R² ≤ 0.57) and λ (R² = 0.63). Shadow lengths (in units of plant height) of 1–3 best explained the location of ripples. The efficacy of shadow casting was affected by the signal to noise ratio in the DSMs at the scale of very small plants. UAS data nevertheless displayed strong potential for advancing the study of vegetation and aeolian activity.