Performance Assessment - LR-04-190026

LR-04-19
Page Navigation
20

variables listed above. Probability values are listed in Table 2. In the table, P values less than .20

are noted via bold type. Due to the variable nature of field data, probabilities less than< 20% are

often considered to have practical significance and we have done so in these analyses. The

implication of these findings is that a one-way ANOVA comparison of treatment effects is

reasonably representative of the approach likely to be used in typical field demonstrations--three

to four replications of two to several treatments. This ESTCP project provides data comparing

two levels of two treatments, replicated four times at each of three locations, and normalized for

concentration differences and the temperature of the locations; and the data did not uncover

significant effects P<.05 for any of the treatments. Using a one-way ANOVA, only one fraction,

the aromatic C>10-12, showed a significant treatment at P=0.146, and this was a reduction in

treatment efficacy for the fertilizer treatment relative to the control or other treatments (Figure

9). Our data from similar studies conducted at two locations in Korea showed an apparent

reduction in treatment efficacy, relative to both the control and planted treatments, when

fertilizer alone was used (Reynolds et al., 2001). These data suggest that "standard" monitoring

approaches for "typical" treatment durations are unlikely to detect a rhizosphere treatment effect,

and suggest that the greatest effect relative to a control treatment is in specific petroleum

fractions.