No abstract

This paper discusses aspects of grass vegetation in relation to soil erosion control. By means of a literature research, four options for using grass vegetation were recognized, each having its own requirements concerning maintenance, vegetation characteristics and field layout. The main filter mechanisms, application in the field and effects on runoff and soil loss are discussed. Field experiments on filter strips were carried out to determine whether literature data for water and sediment retention by vegetation can be applied to sloping loess soils in South Limburg (The Netherlands). The field experiments simulated a situation in which surface runoff carrying loess sediment from an upslope field enters a grass strip. The retention of water and sediment by grass strips was determined by measuring runoff discharge and the sediment concentration at the inflow and outflow points from bordered plots. Two locations with different grass age and agricultural management were studied.Results show that grass strips are effective in filtering sediment from surface runoff as long as concentrated flow is absent. Outflow sediment concentrations could be described as a function of inflow concentrations and strip width. Reductions of sediment discharge vaned between 50-60, 60-90 and 90-99% for strips of 1, 4-5 and 10m width, respectively. Old grass, extensively used as pasture, is more effective in reducing erosion than the younger grass which was often accessed by tractors for mowing. Differences in water retention between both grass locations appear to be caused mainly by differences in grass density.

Monthly runoff and soil loss data of three fallow experimental plots are presented, comprising a summer and following winter season. The fallow plots were only tilled once, at the end of April. Summer runoff appeared to be controlled by rainfall intensity and conforms to the Horton model of overland flow generation. Winter runoff was primarily controlled by rainfall amount and conforms to the saturation or storage control model of runoff generation. Summer runoff volume was one fourth of winter runoff volume. Summer soil loss was twice as high as winter soil loss and was caused by high intensity, high energy rainfall. Winter soil loss was due to detachment limited erosion, caused by low intensity, low energy rainfall. Mean sediment concentration of winter runoff was one seventh of that of summer runoff. Implications for runoff and erosion of climatic change, involving increased rainfall amounts or intensities in summer or winter, are given.

SUMMARYIn wooded dry valleys in the Luxembourg Ardennes a gravelly and slightly organic colluvial deposit occurs. The depositionary nature of the material is ascertained by its micromorphological characteristics and its position on top of a truncated argillic horizon of an older soil. Pollen analysis revealed the colluvium to have been formed in the last 500 years, the lower 37 cm in the Haarts valley under non-forest (presumably arable land) conditions and the upper lOcm under today's oak woodland. A formation of the colluvium by Horton overland flow is ruled out by permeability and infiltration capacity measurements of the forest soil. The importance of rainsplash erosion for the formation of the upper part of the colluvium is demonstrated with simplified Ellison type splash boards. Two modes of transport, induced by raindrop impact, may be distinguished, one through the air (particles up to 6.3mm) and ' one across the soil surface (particles up to and above 1 cm). The overall grain size distribution of the trapped soil strongly resembles that of the colluvium, and the gravel sized rock fragments in the colluvium certainly do not exclude an origin by rainsplash erosion.

No abstract