Search

This note describes the technical characteristics and the preliminary setup and calibration of a nozzle-type rainfall simulator installed at the University of Perugia's Masse experimental station. This simulator includes some innovative features, usually not present in similar low-cost instruments employed in field experiments: 1) both single and multi-nozzle operation are feasible, thus enabling the possibility to simulate variable-intensity rainfall events; 2) events are simultaneously reproduced in two identical plots (width 1 m, length 0.92 m, and slope 16%), thus obtaining two replications for each experiment; 3) the runoff volume coming from the micro-plots and the water volume falling outside the plots are both collected and conveyed to separate outlets, thus allowing an easy calculation of the infiltrated volumes and of the system efficiency. The first step of the calibration regarded the spatial distribution of rainfall, the stability of the rainfall intensity over time (within the experiment), and the reproducibility of the rainfall intensities both in space (between the two plots) and over time (among successive experiments). Next, the drop size distribution (DSD) and the related rainfall characteristics (median volumetric drop diameter D50 and mean kinetic energy per unit area and unit depth) were evaluated by the oil method for the single and some multi-nozzle applications. An effective automatic drop recognition procedure by the Fiji open-source software is proposed and illustrated. Results indicate a high uniformity of rainfall (Christiansen uniformity coefficient > 92%) in both single and multi-nozzle operation for both plots. The reproducibility of experiments over short time intervals and under similar environmental conditions is satisfactory (coefficient of variation of mean intensity < 3%). Comparing initial and final rainfall intensity values, a slight decrease was observed within each experiment (about 2% on average). Some systematic differences in rainfall intensity were also detected between the plots. The analysis of both the intensity and the DSD in the multi-nozzle operation indicated the presence of some interference among the individual sprays when larger nozzles are simultaneously activated. Moreover, in agreement with the literature, it was confirmed that this type of rainfall simulator always produces kinetic energy values lower than those associated with natural rainfall of similar intensity. [ABSTRACT FROM AUTHOR]

• Rainfall simulation allows testing soil changes after tillage. • Sorptivity decreased abruptly during post-tillage wetting and drying. • The decrease of saturated soil hydraulic conductivity was gradual. • Soil hydraulic characterization should not be limited to shortly after tillage. • The applied methodology seems appropriate to obtain hydrologically useful soil data. The changes in the soil hydrodynamic properties following soil tillage were investigated in rainfall simulation trials of intermittent rain at the Masse experimental station (Soil Erosion LABoratory, SERLAB) in central Italy. The experiments were designed to build a database as representative as possible of situations that may occur in nature. The data collected during the experiments were used to determine the saturated soil hydraulic conductivity, K s , the soil sorptivity at the antecedent soil–water matric potential Ψ i , S , and the flow-weighted mean pore size at Ψ i , λ m. It was also verified if the energy content of total rainfall after tillage explained the short-term temporal variability of K s. The results showed that during a sequence of rainfalls with wetting and drying cycles, there was a reduction of both S and K s by 2.9–3.1 and 1.4–2.2 times, respectively, depending on the plot. This decrease was abrupt for S and more gradual for K s. The analysis confirmed that K s decreased as the overall energy dissipated at the soil surface, E , increased. The range of possible K s values should be expected to become smaller as the dissipated rainfall energy after tillage increases (< 30 mm/h for E = 2 kJ/m2 and < 5 mm/h for E = 8 kJ/m2). For this reason, for the prediction and mathematical simulation of the rainfall-runoff transformation process, it is not advisable to limit the investigation to a single hydraulic characterization carried out immediately or shortly after tillage but data should be collected in a relatively long time span after tillage to properly characterize the soil in a condition favourable to surface runoff occurrence. For very high energy values, a recovery mechanism of the hydraulic properties of the altered/compacted layer was observed, but this behaviour should be confirmed by further investigations. [ABSTRACT FROM AUTHOR]

The suitability of a smartphone camera for the structure from motion (SfM) reconstruction for monitoring variations in soil surface characteristics and soil loss originated by a low intensity erosive event was evaluated. Terrestrial laser scanning (TLS) was used to validate the SfM model. Two surveys of the soil surface, one before and one after the rainfall event, were carried out for SfM and TLS. The point clouds obtained by the SfM were compared to the TLS point clouds (used as reference). From the point clouds, digital elevation models (DEMs) (0.01 m× 0.01 m) were obtained. The differences of the DEMs (DoDs) obtained from the two surveys for SfM and TLS were compared. To assess the uncertainty of the DEMs, from the DoDs the minimum level of detection was derived. The soil loss was evaluated from DoDs (for SfM and TLS, respectively) considering negative values as erosion and positive values as deposition. The SfM appears appropriate and sensitive for detecting small soil surface variations induced by low erosive events. The SfM estimated correctly the measured soil loss, while TLS underestimated 26%. Further studies could be carried out to consolidate these first results. [ABSTRACT FROM AUTHOR]

The potential of coupling soil moisture and a Universal Soil Loss Equation-based (USLE-based) model for event soil loss estimation at plot scale is carefully investigated at the Masse area, in central Italy. The derived model, named Soil Moisture for Erosion (SM4E), is applied by considering the unavailability of in situ soil moisture measurements, by using the data predicted by a soil water balance model (SWBM) and derived from satellite sensors, i.e., the Advanced SCATterometer (ASCAT). The soil loss estimation accuracy is validated using in situ measurements in which event observations at plot scale are available for the period 2008-2013. The results showed that including soil moisture observations in the event rainfall-runoff erosivity factor of the USLE enhances the capability of the model to account for variations in event soil losses, the soil moisture being an effective alternative to the estimated runoff, in the prediction of the event soil loss at Masse. The agreement between observed and estimated soil losses (through SM4E) is fairly satisfactory with a determination coefficient (log-scale) equal to ~0.35 and a root mean square error (RMSE) of ~2.8Mgha-1. These results are particularly significant for the operational estimation of soil losses. Indeed, currently, soil moisture is a relatively simple measurement at the field scale and remote sensing data are also widely available on a global scale. Through satellite data, there is the potential of applying the SM4E model for large-scale monitoring and quantification of the soil erosion process. [ABSTRACT FROM AUTHOR]

Summary The paper presents an analysis of the rainfall structure for 228 events defined using a minimum interevent time MIT = 6 h recorded in the experimental station of Masse (Central Italy). The events were further sub-sampled to define a set of storms , these being events in which at least one 5 min interval had a rainfall rate of at least 9.6 mm/h. Some storms were erosive and some non-erosive respectively if the corresponding runoff volume and soil loss measured at Masse were greater than zero or not. The selected rain rate threshold is not able, alone, to sample all and only the erosive events recorded at Masse. The properties of the storms were further analyzed in terms of the properties of blocks (or runs) of 5 min periods that had rain above the nominated threshold rainfall rate and of the corresponding antecedent and subsequent rainfall. The goal was to develop ways of representing the time distribution of rainfall that might have explanatory power in terms of soil loss and that might be useful to select compound criterion (overall and profile characteristics) for the identification of erosive events from pluviograph records and to simulate rainfall corresponding with those of natural erosive events in Central Italy. Some statistical erosive storm hyetographs have been simulated according to the results of the analysis. [ABSTRACT FROM AUTHOR]

Abstract: The paper introduces the results of the calibration of the runoff and sediment concentration measurement system adopted at the Masse experimental station recently set up by the Department of Civil and Environmental Engineering to study erosion processes at the plot scale in Central Italy. The collected runoff is stored in cubic tanks equipped with ten taps for sampling the suspension. When an erosive event occurs, the plot runoff volume is measured by means of the water level in the tank, and the corresponding mean measured concentration in the suspension is quantified. The calibration curve of the specific storage system and of the measurement procedure used makes it possible to derive the actual concentration. The event plot soil loss is quantified as the product of the actual concentration and the runoff volume. The calibration curve was determined for two soil types. In any case a linear relationship was found between an actual and measured concentration. Slopes of regression vary with water depths in the tank. The calibration model does not significantly vary with the soil type considered due to the low difference in the coarse sand percentage in the two soils (according to Stokes'' law). The measurement devices and the sampling procedure being equal, the calibration curve determined can be used to correct the measured concentration in the actual concentration for any soils that are characterized by a coarse sand percentage similar to that of the soil used to derive it. [Copyright &y& Elsevier]

• 5 types of indices were used to describe the roughness dynamic at interrill scale. • Traditional contact methods (pin-meter and chain) were derived from DEMs. • A new soil roughness index computed from the point cloud was proposed and tested. • It was emphasized the influence of the resolution in roughness quantification. • Chain method, point cloud and moving window-DEM indices denoted the best performance. The principal aim of the paper is to determine which index of random roughness is most appropriate for the description of the microtopographic variations within a rainfall series at the interrill scale. The study was conducted with rainfall simulation on interrill-sized plots. The roughness indices were computed from the point cloud or from the Digital Elevation Model (DEM) of the soil surface, obtained from photogrammetric surveys. The indices evaluated can be grouped into five categories: 'contact method – pin' and 'contact method – chain' where the indices are obtained from the DEM simulating manual techniques (pin meter and chain respectively); 'fitting plane/surface', where the indices are the standard deviation of the height differences from a best fitting plane/surface of the DEM or of the cloud; 'moving window DEM', where the index is the average standard deviation of the residual topography; and 'kernel size cloud', a new index given by the standard deviation of the distances between each point of the cloud and a best fitting plane computed from its nearest neighbors (found in a radius called "kernel size"). The performance of the indices was assessed based on their relationship with the measures of cumulative rainfall and kinetic energy, depression storage, cumulative runoff and soil loss. The best index to represent random roughness variations is that of the 'moving window DEM' category with a resolution of 0.0025 m2, RI 25 , for which the relationships with the cumulative rainfall, the cumulative rainfall kinetic energy and the depression storage, have coefficients of determination of 0.86, 0.85 and 0.80 respectively. For the same relationships, a good but slightly lower performance is provided by the indices of the categories 'contact method – chain' and 'kernel size cloud'. These three categories gave equivalent results regarding the correlation with cumulative runoff and soil loss (mean Pearson correlation coefficient −0.95, CV = 0.025). Instead, the indices of the categories "fitting plane/surface" and "contact method – pin" resulted poorly correlated with all the measured variables. [ABSTRACT FROM AUTHOR]

This paper reports the results of a field investigation on the rill morphology and the corresponding soil loss, using a Terrestrial Laser Scanner (TLS), at the Masse experimental station (Umbria, Central Italy). Laser scanning is a survey method that allows determining the three-dimensional position of a large number of points (point cloud) through the measurement of angles (azimuth and zenith) and distances. For this paper five scans were made of some rills, which formed during particularly erosive rain in the plots of the Masse station. The point cloud was first interpolated by the natural neighbor method to create a discrete 0.02 m × 0.02 m square cell grid Digital Elevation Model (DEM) used for mapping the flow network. The same cloud was also used to generate a continuous modeling of the surface (by the Triangulated Irregular Network model, TIN) to quantify the total eroded volume and the morphological characteristics of the rill formations. Three methods were applied to the DEM for detecting the channel network: the slope method (Horn, 1981; Wood, 1996), the constant drop method, described by Broscoe (1959) and the method of landform curvature (Tarolli et al., 2012). The comparison has shown a good agreement between the three methods, but Broscoe's method seems to be more accurate for rill recognition. The morphological characteristics of the rill formations derived by the TIN model (i.e. length, width, depth and volume) were compared with the corresponding characteristics obtained manually using a profilometer. The analysis showed a good agreement between the width at the top measured by the two methods, a general overestimation of the maximum depth and of the cross section areas and an underestimation of the rill length when the manual method is used. Lastly, the comparison between the volumes obtained by the TLS survey and by the manual method showed that the total volume calculated by the manual measurements overestimates that evaluated by the TLS, with a good agreement between these variables. [ABSTRACT FROM AUTHOR]

In this paper, the physical model concept by Nearing (1998. Catena 32: 15–22) was assessed. Soil loss data collected on plots of different widths (2–8 m), lengths (11–44 m) and steepnesses (14.9–26.0%), equipped in south and central Italy, were used. Differences in width between plots of given length and steepness determined a lower data correlation and more deviation of the fitted regression line from the identity one. A coefficient of determination between measured, M , and predicted, P , soil losses of 0.77 was representative of the best-case prediction scenario, according to Nearing (1998). The relative differences, Rdiff = ( P − M ) / ( P + M ), decreased in absolute value as M increased only for erosion rates approximately > 1 kg m − 2 . An alternative applicative criterion of the physical model concept, based on the | P − M | difference, was valid for the entire range of measured soil losses. In conclusion, the physical model should be defined in terms of perfect planimetrical equivalence. The best applicative criterion of the physical model concept may vary with the considered dataset, which practically implies the need to further test this concept with other datasets. [ABSTRACT FROM AUTHOR]

Abstract: Including runoff in USLE-type empirical models is expected to improve plot soil loss prediction at the event temporal scale and literature yields encouraging signs of the possibility to simply estimate runoff at these spatial and temporal scales. The objective of this paper was to develop an estimating procedure of event soil loss from bare plots (length=11–44m, slope steepness=14.9–16.0%) at two Italian sites, i.e. Masse, in Umbria, and Sparacia, in Sicily, having a similar sand content (5–7%) but different silt (33% at Sparacia, 59% at Masse) and clay (62% and 34%, respectively) contents. A test of alternative erosivity indices for the Masse station showed that the best performances were obtained by the USLE-MM, originally developed at the Sparacia station. This model includes an erosivity index equal to a power (exponent>1) of the runoff ratio, QR , times the single-storm erosion index, EI 30. The fitted models at the two stations were found to coincide for highly erosive events (i.e., QREI 30 >6MJmmha−1 h−1). A parallelism of the two models was detected for low erosivity events. In conclusion, runoff was an important predictor of soil erosion, and local soil characteristics had a more noticeable effect on soil loss for events with a relatively low soil detachment and transport ability. A wider applicability of the USLE-MM scheme can be suggested than for the Sparacia station alone. Additional testing of the model in other environments and development of accurate procedures to estimate plot runoff coefficient at the event temporal scale are required. [Copyright &y& Elsevier]

Summary: The objective of this investigation was to estimate event soil loss per unit area from bare plots in central and southern Italy using an empirical model for sediment concentration. The analysis was developed using data collected on bare plots differing in length (11–44m) and slope (10–26%) at three Italian stations (Masse, Umbria; Caratozzolo, Calabria; Sparacia, Sicily). At first, an analysis was carried out, using the experimental data collected at Sparacia, to establish a relationship between sediment concentration and hydrological variables, such as runoff, rainfall amount and single storm erosion index. Then, an empirical model to estimate plot soil loss as a function of rainfall amount, single storm erosion index, runoff amount, and slope length was developed. Finally, using the complete data set (Masse, Caratozzolo, Sparacia), the model was extended to include soil erodibility and slope steepness effects on plot soil loss. [ABSTRACT FROM AUTHOR]

The objective of this investigation was to determine empirically the plot width and length effects on runoff volume, Ve , soil loss, Ae , and sediment concentration, Ce , by using data collected, at the temporal scale of the erosive event, on bare plots differing in width (2–8 m) and length (11–22 m) for two Italian stations (Masse, Umbria; Sparacia, Sicily). Mean results differed by a maximum factor of 1.6 for Ve , 1.8 for Ae and 1.2 for Ce when plots differing in width were compared and by a maximum factor of 1.4 for Ve , 1.2 for Ae and 1.3 for Ce when comparison between plot lengths was conducted. Differences between two plot widths or two plot lengths were not statistically significant (P = 0.01). Soil loss was the variable most sensitive to plot size (maximum factor of discrepancy equal to 11.6 for plot width and 25.0 for plot length). Plot width and length effects varied with event erosivity, being low for the most erosive events. Relatively short and narrow plots may give long term results representative of longer and wider plots within the sampled sizes. An appropriate sampling scheme may be based on relatively narrow and short plots instead of wide and long plots, with a reduction of experimental efforts, at least when the events of interest are the most erosive ones. Reducing plot length from 22 m to 11 m is not expected to reduce soil loss per unit area at the two considered locations. [Copyright &y& Elsevier]