Premium
Using Direct Current Resistivity Sounding and Geostatistics to Aid in Hydrogeological Studies in the Choshuichi Alluvial Fan, Taiwan
Author(s) -
Yang ChiehHou,
Lee WeiFeng
Publication year - 2002
Publication title -
groundwater
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.84
H-Index - 94
eISSN - 1745-6584
pISSN - 0017-467X
DOI - 10.1111/j.1745-6584.2002.tb02501.x
Subject(s) - hydrogeology , aquifer , geostatistics , depth sounding , hydraulic conductivity , vertical electrical sounding , geology , electrical resistivity and conductivity , aquifer properties , spatial variability , alluvium , hydrology (agriculture) , soil science , geotechnical engineering , groundwater , geomorphology , groundwater recharge , mathematics , soil water , engineering , statistics , oceanography , electrical engineering
Ground water reservoirs in the Choshuichi alluvial fan, central western Taiwan, were investigated using direct‐current (DC) resistivity soundings at 190 locations, combined with hydrogeological measurements from 37 wells. In addition, attempts were made to calculate aquifer transmissivity from both surface DC resistivity measurements and geostatistically derived predictions of aquifer properties. DC resistivity sounding data are highly correlated to the hydraulic parameters in the Choshuichi alluvial fan. By estimating the spatial distribution of hydraulic conductivity from the kriged well data and the cokriged thickness of the correlative aquifer from both resistivity sounding data and well information, the transmissivity of the aquifer at each location can be obtained from the product of kriged hydraulic conductivity and computed thickness of the geoelectric layer. Thus, the spatial variation of the transmissivities in the study area is obtained. Our work is more comparable to Ahmed et al. (1988) than to the work of Niwas and Singhal (1981). The first “constraint” from Niwas and Singhal's work is a result of their use of linear regression. The geostatistical approach taken here (and by Ahmed et al. [1988]) is a natural improvement on the linear regression approach.