Regression models for hydraulic conductivity and field test of the borehole permeameter

Abstract

The saturated hydraulic conductivity Ks of sediments in the vadose zone is an important parameter in predicting the seepage rates of water and contaminants. The borehole permeameter is an in situ technique to test a relatively large sample size at any depth. Solutions are presented which account for the effects of unsaturated flow. These solutions are derived from a regression analysis of results of numerical simulations in which unsaturated hydraulic conductivity is represented by one or two parameters. The results of a borehole permeameter test in a uniform sand are compared with field ponding tests and air entry permeameter tests. The regression‐based solutions for the borehole permeameter which account for capillarity provide very good agreement with other permeameter results. Depending upon the approach used to solve the borehole problem Ks values determined by methods which neglected and included capillary effects varied by a factor of only about two for the soil tested.

Comparison of methodological uncertainties within permeability measurements

“…This over-estimation is expected as the solution neglects the effect of capillarity on the flow from boreholes whilst the other three solutions take account of the effects of capillarity. The Stephens et al, (1987) solution mostly over-estimates K s by a factor of 1Ð7 to 3, while the Philip (1985) solution is most similar, underestimating K s by a factor of 0Ð05 to 0Ð6 (Figure 4b).…”

“…As many of our piezometers have a screen length and effective radius of similar size, the CHBP equations of Talsma and Hallam (1980); Philip (1985) and Stephens et al, (1987), are likely to produce less accurate permeability values in comparison to the solution of . Differences in the K s values between these analytical methods, therefore, are shown with reference to the solution (Figure 4b).…”

“…The difference in estimated K s values is not, however, due to an incorrect assumption that the piezometer base rests on an impermeable layer used in the Bouwer and Rice (1976) analysis, as this would lead to an over-estimation of K s . (1951) and Bouwer and Rice (1976) analytical solutions to slug test data (including a 1 : 1 line), and (b) the uncertainty in estimates of K s using the Talsma and Hallam (1980); Stephens et al, (1987) and Philip (1985) solutions to CHBP results in comparison to those produced by the solution…”

“…The original solution of Talsma and Hallam (1980) was used to analyse the test data. Because this solution does not include the possible sorptivity effects of the unsaturated zone away from the borehole (Stephens and Neuman, 1982a,b;Philip, 1985;Bell and Schofield, 1990), alternative solutions given by: (i) , (ii) Philip (1985) and rearranged by Bell and Schofield (1990), and (iii) Stephens et al, (1987), were also applied.…”

Variabilidade da condutividade hidráulica do solo em valores fixos de umidade e de potencial matricial.

“…Recentemente, algumas modificações ou novos instrumentos e métodos têm sido desenvolvidos para determinar a condutividade hidráulica in situ, como por exemplo, o infiltrômetro de Guelph (Reynolds et al, 1985), o permeâmetro de furo de trado (Stephens et al, 1987), o infiltrômetro de disco (Perroux & White, 1988), o infiltrômetro de carga constante (Amoozegar, 1989), o infiltrômetro de anel e disco (Angulo-Jaramillo et al, 1998), o infiltrômetro de duplo anel (Maheshwari, 1997), e o permeâmetro de disco (Perroux & White, 1988). Por esses métodos medem-se as taxas de infiltração de água no solo sob pequenas sucções, o que permite estimar a condutividade hidráulica do solo para uma pequena faixa de umidade próxima à saturação.…”