Dino Torri

The European Soil Erosion Model (EUROSEM): a dynamic approach for predicting sediment transport from fields and small catchments

The European Soil Erosion Model (EUROSEM) is a dynamic distributed model, able to simulate sediment transport, erosion and deposition over the land surface by rill and interill processes in single storms for both individual fields and small catchments. Model output includes total runoff, total soil loss, the storm hydrograph and storm sediment graph. Compared with other erosion models, EUROSEM has explicit simulation of interill and rill flow; plant cover effects on interception and rainfall energy; rock fragment (stoniness) effects on infiltration, flow velocity and splash erosion; and changes in the shape and size of rill channels as a result of erosion and deposition. The transport capacity of runoff is modelled using relationships based on over 500 experimental observations of shallow surface flows. EUROSEM can be applied to smooth slope planes without rills, rilled surfaces and surfaces with furrows. Examples are given of model output and of the unique capabilities of dynamic erosion modelling in general.

Effects of rock fragments on soil erosion by water at different spatial scales: a review

This paper reviews the various effects of rock fragments on soil erosion by water. Since these effects are scale dependent, they are investigated at three different nested spatial scales: the microplot (4 × 10−6–100m2), the mesoplot (10−2–102 m2) and the macroplot (101–104 m2). For each scale the corresponding process mechanisms are discussed. Particular attention is paid to the effects of rock fragment cover on the intensity of soil erosion processes. At the mesoplot scale, i.e. on interrill areas, rock fragments at the soil surface can have negative as well as positive effects on sediment yield. These ambivalent effects are conditioned by the type of fine earth porosity, soil surface slope, vertical position and size of rock fragments and by the occurrence of horseshoe vortex erosion. At the microplot scale, i.e. the soil surface area which is covered by a single rock fragment, and at the macroplot scale, i.e. upland areas where both interrill and rill erosion takes place, rock fragments at the soil surface have a negative effect on sediment yield. In these two scales rock fragments can thus be considered as natural soil surface stabilizers. At the macroplot scale the mean decrease of relative interrill and rill sediment yield with rock fragment cover can be expressed by an exponential decay function. The scatter of the data indicates that a given rock fragment cover can have different efficiencies in reducing interrill and rill sediment yield depending on the varying intensities of the hydrological and erosion subprocesses. These findings have implications for erosion modelling and soil conservation.

Splash detachment: Runoff depth and soil cohesion

Summary A set of laboratory experiments has been carried out in order to investigate the importance of soil cohesion and the interference due to the depth of the surface runoff on the detachment rate of soil particles and aggregates. Experimental data show that the soil detachment rate decreases as the runoff depth increases. This indicates that the detachment power of the raindrops is partially dispersed by the water layer. An inverse relationship between soil detachment and soil cohesion has been confirmed. Nevertheless the grain-size distribution must also be used in order to define soil detachability.

Predictability and uncertainty of the soil erodibility factor using a global dataset

Abstract This paper investigates the predictability of the soil erodibility factor (K) in the revised universal soil loss equation from soil clay content (C), the Naperian logarithm of the geometric mean particle size (D G ) and organic matter content (OM), using a global dataset compiled from published studies. A multiple regression equation accounted for only 41% of the observed variance. Because of the large unexplained variance, an alternative procedure was explored to describe theK data in terms of the lower and upper bounds of the range of variation and the most probable value in this range. However, this approach did not provide sufficient information about the distribution of the observedK values. A procedure based on fuzzy logic and fuzzy mathematical theories was then developed, using the program FUZKBAS, which describes the frequency distribution of observedK values for a given soil, characterised byD G ,C and OM, in terms of membership functions.

Preliminary study of the erosion mechanisms in a biancana badland (Tuscany, Italy)

Abstract Italian badlands are present along the Appennini mountains, with an annual precipitation which ranges from 400 to 1200 mm. The genesis of these badlands has often been studied, whereas current processes and erosion rates have only rarely been examined and measured. This paper describes a preliminary study on a badland developed in temperate mediterranean climate and located in SW Tuscany (Italy). The selected site is characterized by biancane, dome-shaped features a few meters in height, generally having a vegetated NNE aspect and severely eroded SSW slope, with a sub-horizontal basal pediment. Field rainfall and runoff experiments were carried out. Field results were substantiated by data collected in specially devised laboratory tests. It was concluded that near-surface pipes — which develop at the point of contact between a surficial layer of weathered and severely fractured regolith and the less fractured and weathered regolith underneath — are the main drainage channels, while rills play a secondary role. Soil detachment is largely due to mechanical and chemical slaking of the regolith during infiltration. The role of runoff is almost solely transport. In particular, it was found that the runoff detachment rate is limited by the rate at which the water front penetrates into the regolith.

Within-storm soil surface dynamics and erosive effects of rainstorms

Abstract This study deals with the characteristics which make a rainstorm an event that can produce intense erosion and even trigger the formation of a badlands site. In order to keep the presentation closely linked to a real situation, a rainstorm which took place on an experimental farm equipped for soil erosion studies was selected. The effects of the erosive rainstorm, which fell on dry antecedent moisture conditions, are given in terms of total rill erosion and rill cross-section along the slope. Unfortunately, the data collected did not answer the basic question, i.e., what combination of factors makes a rainstorm critical? A set of rainfall simulation experiments was therefore carried out, in the field and in the laboratory, in order to evaluate the soil surface variations caused by the rainstorm. All the experiments were performed on dry antecedent soil moisture conditions. It was confirmed that the characteristics of the infiltration curve are modified considerably during such rain events. The saturated conductivity of the first thin top-layer is also modified and it can easily decrease by a factor of 10 due to drop impact forces. The runoff coefficient is also influenced by the raindrop impacting energy and it increases sharply with cumulate energy until a maximum value is reached. The surface micro-relief dynamics was also studied. It was very clearly shown that impacting drop kinetic energy is the rainfall characteristic which is linked to random roughness decay. Cumulative rainfall was not able to align all the data in a single trend. The effect of surface micro-relief decay on the rainstorm erosive power was examined using two equations, thus linking Mannings hydraulic roughness to random roughness. Using a simulated runoff over the field plots that were particularly eroded by the rainstorm, it was possible to observe that the runoff drag forces reached values of between 3 and 100 times the ones which would have been calculated if the random roughness had been constant during the same event. Many of the soil surface characteristics that are modified interact with one another and with erosion. Examining each of them in isolation cannot explain the drastic increase in erosivity of a rainstorm, as the latter is the result of the combined effects of all the surface modifications.

Effects of water quality on infiltration, runoff and interrill erosion processes during simulated rainfall

This paper discusses the effects of water quality on the hydrological and erosion response of non-saline, non-sodic soils during simulated rain experiments. It is well known that rain water quality affects the behaviour of saline soils. In particular, rain simulation experiments cannot be run using tap water if realistic values of infiltration rates and soil erosion are to be found. This paper reports on similar effects for non-saline, non-sodic soils. Two soils – a well-aggregated clay-rich soil developed on marine silty clay deposits and a soil developed on silt loam – were selected and subjected to a series of simulated rainstorms using demineralized water and tap water. The experiments were conducted in two different laboratories in order to obtain results independent of the tap water quality or the rainfall simulator characteristics. The results indicate that time-to-ponding is largely delayed by solute-rich water (tap water). When tap water is used, infiltration rates are significantly overestimated, i.e. by more than 100 per cent. Interrill erosion rates increase by a factor of 2·5–3 when demineralized water is used. The silty clay soil was more affected by the water quality than the silt loam soil, with respect to infiltration and runoff production. Regarding interrill erosion rates, the two tested soils were similarly affected by the water quality. Therefore, it can be concluded that rainfall simulation experiments with non-dispersive soils (e.g. non-saline, non-sodic) must also be conducted using water with very low electrical conductivity (i.e. less than 30–50 µS cm−1), close to that of distilled water. The use of tap water certainly hampers comparisons and the relative ranking of the hydrological and erosion response of different soils, while parameter values, such as final infiltration rate or time-to-ponding, cannot be extrapolated and extended to natural situations. Therefore, the majority of hydrological and erosion models and parameter values measured during rainfall simulations in the past should be used with caution for all types of soils. Copyright

Soil shear strength: Its measurement and soil detachability

Abstract Recent research on soil erosion mechanics has pointed out that the condition at which rill flow becomes erosive is controlled by soil surface shear strength ( TORRI et al. 1987a , RAUWS & GOVERS 1988) . Also splash detachment processes are strictly linked to soil shear strength as shown by NEARING & BRADFORD (1985) , TORRI et al. (1987b) . A comparison of the proposed threshold conditions for incipient rilling immediately shows that they differ in the two quoted papers. The same situation characterizes the relations proposed between splash detachment and soil shear strength. Those differences may depend on the different instruments used for measuring soil shear strength: a Geonor model g-200 laboratory cone penetration apparatus, a laboratory vane tester and a pocket vane tester. Those instruments differ in shape (pocket and laboratory vane testers) or in the physics involved in their functioning (vane tester and drop-cone penetrometer). The experimental results show that the drop-cone penetrometer is characterized by a low repeatability. Moreover, it is sensitive to small variations in soil texture. The laboratory vane apparatus seems to determine a surface of shear close to that of a sphere circumscribed to the vane. The pocket vane apparatus, instead, is characterized by a cylindrical surface of shear. Those results seem to explain the main differences in soil detachment. In particular, incipient rilling conditions are clearly reduced to a single relation between critical shear velocity and soil shear strength. It can consequently be concluded that soil detachability is strictly linked to soil shear strength.

The effect of soil surface slope on raindrop detachment

Abstract The effect of slope angle on splash detachment has been a matter of discussion for long because it has not always been clearly observed in experiments. In this paper a model, based on the physics of the drop impact, is developed. The theoretical model indicates that a wide slope effect exists. Data collected by different authors were compared with the theoretical trends showing a satisfactory agreement. Some experiments showed that small changes in soil surface characteristics may hide the slope effect.

The EUROSEM model.

The European Soil Erosion Model (EUROSEM) is a dynamic distributed model for simulating erosion, transport and deposition of sediment over the land surface by interrill and rill processes. It is designed as an event-based model for both individual fields and small catchments. Model outputs include total runoff, total soil loss, the storm hydrograph and the storm sediment graph. EUROSEM provides for explicit simulation of interrill and rill flow; the effects of plant cover on rainfall interception, infiltration, rainfall energy and flow velocity; and the effects of rock fragment cover on infiltration, flow velocity and splash erosion. Catchments are represented as a simplified cascading network of elements, which may be either planes (for hillslope segments) or channels. Each plane is considered uniform in its soil, slope, surface microtopography and land cover.