Roughness indices for estimation of depression storage capacity of tilled soil surfaces

Information on the depression storage capacity (DSC) of soil surfaces is needed for modelling surface runoff and erosion. However, time-consuming measurements of the soil surface microrelief in a two-dimensional grid are necessary to obtain accurate estimates. Therefore, estimation of DSC from roughness indices requiring fewer measurements is desirable. In this study a new simple and physically based roughness index was developed and tested. Thirty-two soil surface microreliefs were determined in Danish erosion study plots on two sandy loam soil types, Humic Acrosol and Haplic Luvisol according to the FAO soil classification. The soil treatments were: cultivated up-and-down the slope (black fallow), mouldboard ploughed up-and-down the slope, winter wheat (Triticum aestivum L.) drilled up-and-down the slope and winter wheat drilled across the slope. The slope was ca. 10% for all plots. Data were treated initially by removing the slope after which 20 artificial slopes (1-20%) were introduced producing 640 data sets. The DSC was calculated for each data set and a number of roughness indices were related to the calculated DSC. The best correlation to DSC was obtained by the newly developed index, mean upslope depression, MUD (R-2 = 0.86), followed by the geostatistically based limiting elevation difference, LD, and limiting slope, LS (R-2 = 0.76), and some statistically based parameters as standard deviation adjusted for oriented roughness up-and-down the slope (R-2 = 0.76), random roughness, RR (R-2 = 0.67), and standard deviation adjusted for oriented roughness both up-and-down and across the slope (R-2 = 0.62). Furthermore, MUD appeared to give less biased estimates than other indices, which overestimated small DSC-values and underestimated larger DSC-values. Calculations showed that the MUD index required relatively few measurements. MUD based on less than 200 surface elevation readings gave a good prediction (R-2 > 0.7) of DSC of tilled soil surfaces. (C) 1999 Elsevier Science B.V. All rights reserved.