Geostatistical analysis of hydraulic head spatial variability and its application to piezometer network design in Harat Aquifer, Yazd,Iran

摘要

The objective of this research is to use a Krigingbased approach to design a network of piezometers in Harat Aquifer for groundwater monitoring. This aquifer is located in central part of Iran, a region affected by lack of irrigatable and potable water. The observed water level data at 20 piezometers were used for spatial analysis of groundwater level using geostatistics. These data have been collected for a period of 14 years on a monthly basis. The experimental variograms of water level are constructed for each month and the effect of time on the shape of variograms was investigated in some detail. The spatial distribution of the error variance was computed and plotted for the aquifer based on Kriging computations. This variance field was used as an indicator of uncertainty by which one can determine the candidate locations for data augmentation. By an iterative procedure, the new set of piezometers was added over the regions of high variances and the variance field was replotted. The piezometers are added where a significant improvement on the error variance field was observed. Using this procedure combined with some practical field observations, 11 additional locations for installation of new piezometers are suggested. The results of this study show that geostatistics can be considered as a powerful tool for groundwater data analysis and network design.