ARCHIVED REPORTS_XR0005378 - Laserfiche WebLink

Home Browse Search

i ' 1444 MARIPOSA ROAD WKA No 2938 06 September 6, 2000 ' Page 15 of contaminants in soil then the contaminant volume in groundwater Discussions of our ' approach to each area, along with assumptions, are listed below. ' Soil Contamination ' Based on all available data, we assumed that soil contamination resulting from the release was largely limited to the immediate vicinity of the former UST excavation. This area was measured ' to be approximately 4,650 square feet(Plate No 3) Concentrations of TPH as gasoline and BTEX were averaged over the entire impacted area using ' a method of log normal distributions (Gilbert, 1987) Where sample concentrations were reported below detection limits, a value equaling half the detection limit was used in our calculation The lognormal distribution was determined by taking the sum of the log for each contaminant concentration in a given data set The anti-log of the sum was then taken to determine lognormal distribution of the average concentration for the given compound within the data set The lognormal distribution method is typically used for analytical data that vanes by several ' orders of magnitude The log normal distribution method normalizes the varying orders of magnitude to a standard bell-shaped distribution For example suppose we have three samples ' with concentrations of 1, 100, and 1000 mg/kg Sample Concentration Log 1 log (1)=0 100 log (100) =2 1000 log(1000) =3 E= 1101 E= 5 1101/3 =367 5/3 = 167 log-1 167=46.8 Using a straight mathematical average, the average concentration of the three samples,would be ' 367 mg/kg Using the log normal distribution method,the average concentration of the three samples would be 46 8 mg/kg Because only a very small number of the previous samples t i o WALLACE KUHL Regcled PaNt EF ASSOCIATES INC