Comparison of Procedures to estimate Steady Flow Rate in Field Measurement of Saturated Hydraulic Conductivity by the Guelph Permeameter Method

Abstract Different procedures for estimating steady flow rate during field application of the Guelph permeameter method were compared in terms of field-saturated hydraulic conductivity K fs and equilibration time t s estimates. The steady flow rate was evaluated by the slope of the linear portion of the cumulative drop in water level in the permeameter reservoir versus time plot (reference procedure) and by approximate procedures that use the first three or four consecutive values of the rate of fall of the water level equal or differing by less than a given percentage. The influence of the selected time interval between two successive readings ( Δt =2 or 10 min) at the permeameter was also investigated. On average, the approximate procedures were practically equivalent and produced a K fs overestimation for the selected experiments (sample size N =81; Δt =2 min; 0·7 mm/h⩽ K fs ⩽121·5 mm/h) which was not, however, consistent with the observed, substantial t s underestimation. The K fs prediction accuracy varied with soil permeability: in low permeability soil, the K fs overestimation was substantial, being also equal to more than an order of magnitude; in high or relatively high permeability soil ( i.e. for reference K fs values greater than about 25 mm/h), the approximate and the reference procedures produced similar results, i.e. differing by a maximum factor of 1·5. The increase in the time interval between two readings of the permeameter gave an improved accuracy of the t s and K fs estimates but also a decrease of the number of successful experiments ( i.e. experiments producing estimates of K fs ) within each approximate procedure. An empirical relationship between the minimum equilibration time required to obtain accurate K fs values and the early-time infiltration amount was derived. Using this relationship effectively allowed improvement in K fs estimation accuracy compared to the considered approximate procedures.