Geostatistical Analysis of Groundwater Quality Data from the Madrid Basin Using Adjoint State Maximum Likelihood Cross-Validation

摘要: Mathematical models of groundwater quality require that information about chemical and isotopic variables be quantified in space time by appropriate statistical methods. We present some results a geostatistical analysis applied to 13 hydrochemical from the large sedimentary Madrid basin central Spain. The is performed an adjoint state maximum likelihood cross-validation (ASMLCV) method recently developed authors. can used estimate spatial covariance structure intrinsic nonintrinsic functions point or spatially averaged data may corrupted noise. It also allows use model identification criteria select best among set alternatives compute lower bound for matrix parameter estimates.