Predicting the Coefficient of Permeability of Non-Plastic Soils

Abstract

The factors that affect the coefficient of permeability for a given soil are particle size distribution (grading curve), void ratio, level of saturation, soil structure, and soil imperfections or discontinuities [1,2,3,4]. The coefficient of permeability increases significantly with increase in the void ratio. Uniformly graded soil has a higher coefficient of permeability than well-graded soil. Natural plastic soils are often stratified and include lenses of nonplastic permeable soils, resulting in much higher horizontal than vertical permeability. Imperfections such as root cavities, fissures, and cracks significantly increase the soil permeability.For these reasons many engineers prefer field test over laboratory tests to obtain the permeability of soils. The main disadvantage of laboratory permeability tests is the small size of the usually reconstructed sample. The errors of different laboratory permeability tests are widely discussed by Chapuis [4].There are many different formulae to predict the coefficient of permeability of soils [2, 3, 4], particularly nonplastic soils [3,4,6,7,10]. A century ago Hazen [3,4] developed an empirical formula for predicting the coefficient of permeability of loose saturated sand in the simple formwhere k is the coefficient of permeability (cm/sec), C H is the Hazen empirical coefficient, and d 10 is the particle size for which 10% of the soil mass is finer (cm). This formula is frequently used in engineering practice. The published values C H range from 1 to 1,000 [5]. The Hazen formula (1) applicability is limited to 0.01 cm < d 10 < 0.3 cm [5]. The Hazen empirical coefficient C H expresses the influence of the void ratio, uniformity, and other factors on perThe Hazen formula, Kozeny-Carman formula, Pavchich formula, and Shahabi et al. formulas were used to predict the coefficient of permeability of nonplastic soils and compared with laboratory tests performed for eight soils of different porosities. The best correlation was obtained when using the Shahabi et al. formula. Knowing the velocity of water flow in soils is very important for the verification of stability against hydraulic hive, internal erosion, and in the design of drains and walls. The flow of water in soils is almost always laminar, and the water velocity is proportional to the coefficient of permeability.

Laboratory Flume and Numerical Modeling Experiments Show Log Jams and Branching Channels Increase Hyporheic Exchange

“…The median grain sizes of the deep layer and surface layer are 2.83 and 12.7 mm, respectively. According to the Shahabi empirical method (Dolzyk & Chmielewska, 2014;Shahabi et al, 1984), hydraulic conductivities are estimated as 8.9 × 10 −5 m/s for the deep layer and 2.5 × 10 −3 m/s for the armored surface layer, but model-data comparison (Supporting Information S1) and typical hydraulic characteristics of unconsolidated gravel and sand sediments (Freeze & Cherry, 1979) suggest that hydraulic conductivities are likely an order of magnitude greater. The two layers' porosities are estimated at 0.3, typical of unconsolidated fluvial sediments (Freeze & Cherry, 1979).…”

Suitability of Engineering-Geological Environment on the Basis of Its Permeability Coefficient: Four Case Studies of Fine-Grained Soils

“…To date, numerous studies have been carried out showing significant differences between results obtained with different empirical formulae, as well as empirical formulae and laboratory and field studies [16,17,19,22,41,[46][47][48] There have been few studies comparing different methods of determining permeability coefficient with consideration of particle shape characteristic and porosity. In their paper, Cabalar and Akbulut compared the values of permeability coefficient for sands with different gradation and shape, with the use of SEM images and simple indices classifying particle shape [18].…”

Predicting the impact of biochar on the saturated hydraulic conductivity of natural and engineered media

“…is the specific surface area (m 2 /kg), and C is a dimensionless factor that accounts for shape and tortuosity of flow channels. The Kozeny-Carmen (K-C) model was developed considering a porous material as an assembly of capillary tubes for which Navier-Stokes equations describe flow (Dolzyk and Chmielewska 2014). The value of C = 0.20 includes simultaneously the notion of equivalent capillary channel cross-section and tortuosity (Chapuis andAubertin 2003, Eibisch, Durner et al 2015).…”