Comparison of methods to assess the accuracy of the incorporation of censored chemical data in descriptive statistical analysis of contaminated groundwate

Chemical analyses of groundwater often present data sets with censored values, i.e., below the detection limit (LOD). When proportion values is significant, descriptive (mean, median and standard deviation) or exploratory geochemical analysis may be impaired. Ignoring such replacing them some predetermined value not always recommended alternative. Thus, objective this research to investigate applicability four methods in estimating chemical from an area contaminated groundwater. Three statistical were used: parametric (Maximum Likelihood Estimation, MLE), non-parametric (Kaplan-Meier, KM) robust (Order Regression Methods, ROS), addition traditional method direct replacement data, using LOD/2. The MLE, assuming a Gaussian distribution (MLE-no), yielded allowable substitution factors, close 0.5, similarly (LOD/2). Validation complete datasets same estimation considering three artificial LOD attested good results MLE-no ROS 25% 50% respectively, as well first two are preferable LOD/2 they statistically based. It future studies that combined other geostatistical treatments improve spatial hydrochemical datasets.