A rain splash transport equation assimilating field and laboratory measurements

Abstract

[1] Process-based models of hillslope evolution require transport equations relating sediment flux to its major controls. An equation for rain splash transport in the absence of overland flow was constructed by modifying an approach developed by Reeve (1982) and parameterizing it with measurements from single-drop laboratory experiments and simulated rainfall on a grassland in East Africa. The equation relates rain splash to hillslope gradient, the median raindrop diameter of a storm, and ground cover density; the effect of soil texture on detachability can be incorporated from other published results. The spatial and temporal applicability of such an equation for rain splash transport in the absence of overland flow on uncultivated hillslopes can be estimated from hydrological calculations. The predicted transport is lower than landscape-averaged geologic erosion rates from Kenya but is large enough to modify short, slowly eroding natural hillslopes as well as microtopographic interrill surfaces between which overland flow transports the mobilized sediment. Models of Hillslope Evolution[2] Hillslope evolution under transport-limited conditions is usually modeled as the result of one-dimensional sediment transport, represented by the mass balance equationwhere @z/@t is the local rate of elevation change (m À1 a À1 ), x is horizontal distance, r b is the sediment bulk density (kg m À3 ), and Q s is the mass transport rate of sediment per unit width of hillslope (kg m À1 a À1 ). Much of the work using equation (1) has been concerned with possible forms of equations for Q s , and with the hillslope profiles such equations predict when subject to chosen boundary conditions [Culling, 1960;Hirano, 1969;Kirkby, 1971]. Ahnert [1976, 1987], Kirkby [1985Kirkby [ , 1989, and Willgoose et al. [1991] incorporated the mass balance equation into numerical simulations of landscape evolution for a range of processes and boundary conditions such as base-level control.[3] The theoretical work challenged geomorphologists to develop formulae relating sediment transport to its controls in order to quantify analyses of the geomorphic effects of climate, hydrology, vegetation, material properties, tectonism, and time. Field and laboratory measurements of the sediment transport processes responsible for landform evolution have accumulated more slowly than the theoretical developments. They are necessary for understanding actual magnitudes and rates of geomorphic change, the relative roles and interactions of various processes, and the influence of environmental factors on rates of landform change and sediment production. Yet Dietrich et al. [2003] concluded that geomorphology has made little progress in developing processbased sediment transport equations for modeling landform evolution.[4] Here we construct a transport equation for aerial rain splash, including particle creep along the surface, in the absence of overland flow. It is based on field experiments on grass-covered hillslopes in Kenya, laboratory experiments by us and o...

Floor Cover Percentage Determines Splash Erosion inChamaecyparis obtusaForests

“…Thus, we used erosion rate units of grams per meter per year. The dimensions of the units are compatible with those of the splash cup experiment of Dunne et al (2010). We installed fences before commencing floor cover removal experiments at the beginning of WY1996 to avoid sediment inflow from the upper slope outside of the study plots.…”

“…We defined these five size classes of sediment as fine earth (£2.0 mm in diameter), fine gravel (2.0-6.7 mm), small gravel (6.7-22.4 mm), medium gravel (22.4-63.0 mm), and large gravel (>63.0 mm). We weighed each sediment size class after oven drying at 70°C for 48 h. The sediments were collected from the 25-cm width of the trap, so sediment weights were converted to 1 m of contour width (g m −1 ) to be comparable to data from Ellison splash samplers (Dunne et al, 2010). We also measured FCPs within a 0.25-m 2 area just upslope of each sediment trap by counting the number of 5-cm-mesh squares covered by vegetation, litter, gravel, or soil (Fig.…”

“…3). Dunne et al (2010) demonstrated by citing several previous studies that most detachable soils with a median grain size of 0.125 to 0.21 mm and particles ³4.0 mm in diameter are not detached by raindrop impacts. Particles ³4.0 mm in diameter might be transported in C. obtusa stands partly because raindrops are larger in throughfall than in incident precipitation (Brandt, 1990;Nanko et al, 2004).…”

“…To establish practical management guidelines, the relationship between FCP and erosion rate must first be quantified. Quantitative relationships between FCP and erosion rate have been clarified for a grassland in Kenya (Dunne et al, 2010) and a beech (Fagus crenata Blume) forest in Japan (Chu et al, 2010), but no such data exist for C. obtusa forests.…”

Constraining the relative importance of raindrop‐ and flow‐driven sediment transport mechanisms in postwildfire environments and implications for recovery time scales

“…Since the influence of raindrop impact on the bed is known to vary with overland flow depth [Proffitt et al, 1991;Gabet and Dunne, 2003;Dunne et al, 2010], many modeling studies [e.g., Heng et al, 2009Heng et al, , 2011Kim et al, 2013] prescribe the raindrop detachment and redetachment coefficients, a and a d , respectively, as decreasing functions of flow depth. In this study, we let…”