Identification of spatial distributions and uncertainties of multiple heavy metal concentrations by using spatial conditioned Latin Hypercube sampling

摘要: Abstract This work develops spatial conditioned Latin hypercube sampling (scLHS), a novel advance on (cLHS) [Minasny and McBratney, Computers & Geosciences (2006) 1378–1388]. The difference between cLHS scLHS is that the latter introduces variograms of ancillary variables into objective function optimization procedure selects locations. improvement was evaluated by applying both to simulated samples multiple heavy metals (Cr, Cu, Ni, Zn) in soil Changhua County Taiwan. Simulation done using sequential indicator simulation (SIS), which generated 1000 realizations distributions metals, based existing sample data, represent real metals. results Ripley K analysis show locations obtained approaches were not significantly spatially segregated. declustering may dominate process scLHS. Statistical indicated compared with cLHS, means, standard deviations contamination proportions captured more efficiently variability SIS realizations. Moreover, experimental samples, especially for small sizes, efficiently. Therefore, use approach recommended as alternative without need reconnaissance survey increase efficiency capturing structures delineating contaminated sites.