ESTCP Project #1011, Rhizosphere
Final Report
By expressing changes in the composition of petroleum relative to the recalcitrant biomarker
decalin, we normalized degradation rates with respect to concentration differences and thereby
reduced concentration variability at each site. However, each site was treated for different
lengths of time and at different conditions. To account for this, we normalized the treatment time
based on temperature at the site. Due to issues common at remote field demonstration sites, we
were unable to collect reliable soil field temperature data. As an alternative, we used air
temperature data available from the National Oceanographic and Atmospheric Administration
(www.cdc.noaa.gov). Barrow and Annette Island data were obtained from this database, but
Galena data were not available. To substitute for Galena data, we used Fairbanks temperature
data for the Galena site. The latitude and air temperatures at Galena and Fairbanks are similar.
Data not normalized to GDD, but with the GDD given, are shown in Appendix E.
Growing degree days (GDD) were calculated as below, using 0C as the base temperature:
{GGD = ∑((average high + average low)/2)-0}
Using the above normalization techniques, decalin, GDD, normalized data for the dependent
variables listed below were calculated:
TPH
Summed PAHs
Aliphatic fractions
o C8-C10
o >C10-C12
o >C12-C16
o >C16-C35
o C8-C35 (the sum of the aliphatic fractions)
Aromatic fractions
o C8-C10
o >C10-C12
o >C12-C16
o >C16-C21
o >C21-C35
o C8-C35 (the sum of the aromatic fractions)
4.3.2 Means and 95% Confidence Intervals - Comparison of Plot Treatments. Means
and 95% confidence intervals tabulated for each variable and expressed as (% Depletion / GDD-
C), by site and by treatment, are shown in Tables 4. Negative values resulting from the
calculations are shown for completion. The magnitude of the depletion values ranges up to 0.123
%/GDD-C. Variability is high. The magnitude of the derived GDD normalized depletion values
is similar for all three sites.
tested for distribution. All data were relatively normally distributed. Frequency distributions for
27
Previous Page
