Francesco Colleselli

Specific Surface and Hydraulic Conductivity of Fine-Grained Soils

AbstractThe Kozeny-Carman (K-C) equation can be used to predict the saturated hydraulic conductivity of a soil, provided enough information is given regarding the porosity and the specific surface. The main limitation of the K-C equation lays in the determination of the specific surface of the soil, a parameter that is not regularly measured but can be estimated with analytical or empirical relations. A method to evaluate the soil’s specific surface is proposed, which combines the contribution of the plastic clayey fraction and that of the granular silty-sandy particles. The applicability of the proposed method and its effect on the performance of the K-C equation have been verified by comparing the K-C estimated hydraulic conductivity with the result of approximately 130 hydraulic conductivity tests on saturated soil specimens in the triaxial apparatus. The results show that the K-C equation and the method for estimating the specific surface predict within an order of magnitude the saturated hydraulic co...

Compression and Creep of Venice Lagoon Sands

AbstractA laboratory test program was conducted to evaluate the one-dimensional (1D) compression and creep properties of intact sand (and silty-sand) samples from a deep borehole at the Malamocco Inlet to the Venice Lagoon. The tests were performed with a constant rate of strain consolidometer and included special procedures for trimming the frozen samples and measuring strains during thawing and backpressure saturation. The specimens had variable fine fractions ranging from 6 to 21% and mica contents ranging from 1 to 10%. The results confirmed that there is a strong correlation between the creep rate coefficient and the compressibility index and between the swelling index and mica content. The compression behavior in all tests is well described by a nonlinear compression model with a unique limiting compression curve and a variable transition parameter that reflects the fines and mica content. Creep tests performed at different confining pressures are also well represented by a simple two-parameter model.

Back-Analysis of an Artificially Triggered Landslide: A Case Study in Northern Italy

Open image in new window The paper presents a case study of a landslide event, artificially triggered by an exceptional infiltration in an otherwise stable slope. The work aims at investigating the slope failure mechanism within a simplified two-dimensional conceptual framework based on the formation of a perched water table. The landslide occurred in Northern Italy in April 2010, on a hillside with average slope angle 36°–37°; the event affected an area of about 200 m2, the slip surface was located approximately 1 m below the slope profile, in the uppermost layers of a predominantly coarse, well graded soil. A series of numerical simulations were performed to back-analyze the event, using a commercial computer program. The artificial water infiltration and water content evolution were simulated with a two-dimensional finite element (FE) model of the unsaturated-saturated domain with appropriate infiltration boundary conditions. The slope stability analyses were conducted with classic limit equilibrium (LE) methods and were performed at different time instants during the infiltration process. The soil-water retention curves and conductivity functions were defined according to the van Genuchten-Mualem model, with parameters estimated by means of the software Rosetta (United States Department of Agriculture). The combined FE and LE simulations showed the gradual formation of a perched water table, whose associated localized pore pressure distribution results in the loss of the suction stabilizing effect and thus in the reduction of the safety factor. Although supported by basic soil mechanical and hydraulic characterization, the numerical simulations allowed to perform a back-analysis which effectively captured the timing of the event, the location and depth of the slip surface along the slope.