MULTIVARIATE GEOSTATISTICAL DESIGN OF GROUNDWATER MONITORING NETWORKS

A multivariate approach for the design of ground-water monitoring netw orks is presented. The proposed technique is based on the geostatistic al method of cokriging. The network design problem is posed as an opti mization model in which the variance of estimation is minimized. The m ultivariate approach used herein accounts not only for autocorrelation s but also for variable cross correlations. By considering a network d esign in which several variables are jointly monitored, lower monitori ng costs are achieved; also, and more importantly, a better estimation of the monitored parameters is yielded given the fact that lower esti mation variances are reached. The method used for selecting the optima l monitoring sites is based on a simultaneous search technique using a branch and bound algorithm guaranteeing optimality of the solution. T he methodology is applied for the design of a monitoring network to ob serve aquifer transmissivity (T) and the specific capacity (SC) in the Yolo County Basin, Calif. The results show the superiority of the mul tivariate geostatistical approach and how it provides a better and mor e economical network design than the univariate approach.