Resumo

DIAZ-VIERA, Martín A.  e  CANUL-PECH, Félix. A Geostatistical Approach to the Optimal Design of the "Saltillo-Ramos Arizpe" Aquifer Monitoring Network for Proper Water Resources Management. Tecnol. cienc. agua [online]. 2014, vol.5, n.5, pp.141-159. ISSN 2007-2422.

The purpose of this study is to determine the optimal design of the piezometric level monitoring network in the "Saltillo-Ramos Arizpe" aquifer using a geostatistical approach. To identify the best wells for monitoring, two options were considered¾ one fixed the level of variance in the estimation error and the other fixed the number of wells regardless of the resulting estimation error. Both options applied the optimal successive inclusions technique in combination with the ordinary kriging method as a spatial estimator. The data used were taken from 750 hydraulic installations (wells and treadmills) in a geohydrological study conducted in the aquifer in 2007. Groundwater levels showed spatial trends i.e., lack of stationarity which was estimated by a first degree polynomial fit and then subtracted from the original data, thereby obtaining residuals. With these residuals, semivariograms were calculated and an isotropic spherical model was fitted. The resulting optimal network consists of 144 wells, with a standard deviation of error of 21 meters determined by ordinary block kriging. This represented 19.2% of the 750 existing hydraulic installations in the study aquifer, and implies a 80.8 % savings in the cost of monitoring the wells. The best 50, 100, 200, 300, 400 and 500 monitoring wells were also determined, and represent options which can be considered depending on the material, financial and human resources available for this activity.

Palavras-chave : Estimation; kriging; optimization; successive inclusions; geostatistics; aquifer; monitoring network; optimal network design.

        · resumo em Espanhol     · texto em Espanhol     · Espanhol ( pdf )