BACKGROUND: Two bio-pile studies undertaken in a controlled laboratory were aimed at deciding optimal strategies to remediate two different artificial diesel-contaminated soils. Bioaugmentation with various inoculations, biostimulation with various levels of biosurfactant rhamnolipid, and nutrient enhancement were the proposed remediation courses. RESULTS: In Case I, the average of the first-order kinetic degradation rate constants during the first degradation stage for the bioaugmentation/biostimulation treatments was 0.0195 +/- 0.0056 d(-1), which was about 2.6-fold higher than that of the control batch (0.0075 d(-1)). Conversely, in Case II, the rate constants for treatments with amendments and those for the control batch were found to be comparable, 0.0172 +/- 0.0015 d(-1) and 0.0158 d(-1), respectively. Microarray results indicated a less diverse indigenous bacterial community in Case I and an abundant indigenous community in Case II both from the control batches on Day 0. The dynamics of the two microbial communities, revealed by NMS plots, emphasized the similarity among the different treatments during the first degradation stage. CONCLUSIONS: Prior to a remediation project, the usefulness of a bioaugmentation approach can be investigated using an ITS oligonucliotide microarray. Results from the microarray answered why the bioaugmentation approach was useful in Case I, but not in Case II. The abundance of the diesel-degrading community determined the usefulness of bioaugmentation. Relatively quantified TRFLP results analyzed via the NMS plots demonstrated comparable microbial communities during the first degradation stage, regardless of differences between the two batches. The bacterial community structure might shift with the availability of hydrocarbons. (C) 2009 Society of Chemical Industry