Various statistical tests on concentration data serve to support decision-making regarding characterization and monitoring of contaminated media, assessing exposure to a chemical, and quantifying the associated risks. However, the routine statistical protocols cannot be directly applied because of challenges arising from nondetects or left-censored observations, which are concentration measurements below the detection limit of measuring instruments. Despite the existence of techniques based on survival analysis that can adjust for nondetects, these are seldom taken into account properly. A comprehensive review of the literature showed that managing policies regarding analysis of censored data do not always agree and that guidance from regulatory agencies may be outdated. Therefore, researchers and practitioners commonly resort to the most convenient way of tackling the censored data problem by substituting nondetects with arbitrary constants prior to data analysis, although this is generally regarded as a bias-prone approach. Hoping to improve the interpretation of concentration data, the present article aims to familiarize researchers in different disciplines with the significance of left-censored observations and provides theoretical and computational recommendations (under both frequentist and Bayesian frameworks) for adequate analysis of censored data. In particular, the present article synthesizes key findings from previous research with respect to 3 noteworthy aspects of inferential statistics: estimation of descriptive statistics, hypothesis testing, and regression analysis.

International audienceAn understanding of the heterogeneity of quaternary gravelly deposits is required to predict flow and contaminant transfer through these formations. In such deposits, preferential flow paths can lead to contamination at depths greater than predicted under the assumption of a homogeneous medium. The difficulties in characterizing their complex structure with conventional methods represent an obstacle for this prediction. In this study, we developed an approach relying on the use of ground penetrating radar (GPR) for the detection of sedimentary depositional units. A genetic interpretation of the radar stratigraphy allowed us to construct a distribution model of lithofacies. The study was conducted on glaciofluvial deposits underlying a stormwater infiltration basin. Two main system tracts were characterized: a top stratum (50–80 cm deep) corresponding to massive gravel and open-framework gravel, and a base stratum corresponding to trough-fill structures with associated sandy, open-framework, massive, and matrix-rich gravelly lithofacies. The knowledge of the hydraulic properties linked to each lithofacies led us to propose a hydrostratigraphic model. Based on this model, we formulated a hypothesis about the hydraulic behavior of the deposit during stormwater infiltration. Open-framework gravels can act, during complete saturation, as preferential flow paths, and capillary barrier effects may occur under variably saturated conditions. These hypotheses were tested by measuring water content variations (using time domain reflectometry probes) at three depths (0, –0.5, and –1.15 m). Experimental data show infiltration behavior that can be explained by a capillary barrier effect between the two lower probes. These results suggest that our hypothesis about hydraulic behavior is reasonable

No abstract

No abstract