WORK PLANS - Laserfiche WebLink

Home Browse Search

DEPARTMENT OF HEALTH SERVI% r;S TOXIC SUBSTANCES CONTROL PROGRAM TOXICOLOGY AND RISK ASSESSMENT UNIT SCIENTIFIC GUIDANCE MEMO 89-2 August 4, 1989 Page 3 of 4 Use of Non-Detect and Limit of Detection Values in Exposure Assessments 5. If the data are shown to fit a normal distribution without transformation then one should use the LOD/2 for a ND value. In calculating the average concentration of the chemical in a given medium for this case, one should use the arithmetic mean. Arithmetic mean = (Cl + C2 + C3 + . + C„) n C1 = Concentration of a specific chemical in sample 1 C2 = Concentration of a specific chemical in sample 2 C3 = Concentration of a specific chemical in sample 3 n = number of samples 6. The distribution pattern of the concentrations for a chemical contaminant in a given medium across the entire site, may indicate "hot spots" of contamination. Such "hot spots" are illustrated by a low proportion of the values exhibiting relatively high concentrations and a concomitant high proportion of values near or below the detection limit. If this is the case, professional judgement is required; consider requesting additional sampling and analysis to define more stringently the areas of contamination. You may consider dividing the site into_. multiple area sources instead of considering the entire site as a single source. 7 . With modeling one may also justify using ND values less than LOD/1. 414 or LOD/2 where the spatial variation of concentrations or the variation of concentrations over time clearly predicts lower concentrations. For instance, if serial samples over time showed values above LOD early on, but ND values have been reported for a relatively long period up to the present, then it may be appropriate to divide the LOD by a number >2 in assigning a non-zero concentration to a recent sample reported as ND. 8 . As was mentioned above, the limit of detection of a chemical will vary in different samples from the same medium if one chemical contaminant interferes with the analysis of another chemical. For example, in "clean" or relatively uncontaminated soil, the detection limit of xylene might be 10 ppb, whereas in a soil sample containing 20, 000 ppb toluene, the detection limit of xylene may be 200 ppb. In this case, include the higher LOD for estimating chemical concentration in such samples.