Estimation Of Permeability Coefficient Of Saturated Belt Based On Soft - Hard Data

Hydraulic conductivity is an important hydrological parameter to characterize the structureand heterogeneity of the aquifer. Its complex spatial variability property makes the study onestimating hydraulic conductivity hot in groundwater research area. Borehole data belongs to thehard data, which is measured and observed directly. It can provide aquifer lithologic and spatialdistribution information. However, it is time-consuming, relatively expensive, data volumelimited and hard to describe the aquifer spatial variability. Vertical electrical sounding (VES)survey obtains the soft data, which is indirect observed. It reflects variation law of the aquiferhydrologic parameter based on the electric property about stratum. Stratum resistivity of varyingdepth are recorded through adjusting the current electrode separation (AB/2), then with theassistance of semi-empirical formula, hydraulic conductivity of saturated zone can be calculated.Soft data acquisition method is nondestructive and convenient. Moreover abundant data cancontribute to describe the aquifer spatial variability. But, parameter selection in semi-empiricalformula is a problem in study. The idealized single assignment will difficult to reflect the aquiferspatial variability.In this dissertation, hard data (borehole data) and soft data (VES data) are integrated tocalculate hydraulic conductivity. Considering the spatial variability of aquifer hydraulicparameter, three-dimensional geologic stochastic model is utilized to dig detailed aquiferlithologic information from―hard data”. Model result provides the basis for parameter selectionin semi-empirical formula (Archie and K-C formula) to help―soft data” calculate hydraulicconductivity and characterize the spatial variability. As an example, northeast of Bejing Plainarea is chosen as study area because of abundant boreholes data and surface resistivity datasupporting. The calculated results are analyzed and compared with that gained by the pumpingtest to verify the feasibility of this method. The results show that it is effective to combine―softdata” and―hard data” to reflect spatial variability, and the estimated hydraulic conductivity isreasonable. Comparing with the result calculated by one parameter value, estimated results canreflect abrupt in spatial distribution, which is more conforming to the actual situations. By usingsemi-empirical formulas, the estimated results have significant correlation with the particle size,which means lithology has important effect on hydraulic conductivity.