University of New Hampshire Wash off Study: Final Report

The University of New Hampshire (UNH) Storm Water Center attempted to conduct a two-year run off study of RTS-sealed, AS-sealed, and unsealed parking lots at their state-of-the-art parking lot run off testing facility. In autumn 2007, two segments of the study parking lot were sealed with was believed to be RTS and AS. PCTC representatives visited the site about a year into the study, in November, 2008. Experienced sealcoaters immediately recognized that the 2007 RTS application had failed. As it turns out, the RTS was applied under conditions of low temperature and high humidity (it rained shortly after application), resulting in a wash off event. The photograph at the top of this post shows samples collected during that first rain, and provides all the evidence an experienced applicator needs to conclude that a wash off event occurred. The UNH investigators did not seem convinced that there was a problem and continued with the study. After completion of the study, the investigators noted that some of the samples collected from the AS-sealed run off contained high PAHs. The high concentrations seemed to convince the investigators that the lot they thought was sealed with an asphalt-based sealant had really been sealed with RTS. In the paper describing the study published in the journal Environmental Science & Technology, the authors claimed to have called the applicator, who admitted that a mistake might have been made. On learning of this development (which was when the paper was published), PCTC spoke with the applicator who stated that he had not spoken with anyone from UNH since he had been paid for the job in late 2007 or early 2008. The applicator also provided equipment trip reports from the date of the 2007 application which showed the before & after weights of two trucks loaded with two different products (one RTS, one AS) that were deployed to UNH. PCTC is unable to resolve questions about the composition of the material in the two trucks. That said, the published UNH paper did not reveal that a wash off had occurred, and presented two years of data for a parking lot from which the RTS had been washed off on the first day. And the paper stated as a fact that the applicator made a mistake that the applicator says he did not make.

PCTC first asked the authors to withdraw the paper published in Environmental Science & Technology pending correction, and then wrote to the journal requesting that the paper be withdrawn or retracted. The letter is available here. The editor of the journal responded as follows:

Thank you for your recent letter (attached). I am sorry that you believe the (above referenced) paper to be in error. However, we do not withdraw or retract papers except in highly unusual circumstances of scientific fraud or the like. When one believes a paper to be in error, the normal procedure is to write a Correspondence (Comment/Rebuttal) to the journal which will be reviewed by me and a decision will be made to publish it (or not).

PCTC disagrees – the authors did not make errors, they failed to report the conditions under which the study was conducted. The journal’s peer reviewers were thus not given the opportunity to assess the potential impacts of those conditions on the study, and anyone who wished to reproduce the study would not have the information needed to reproduce study parameters.

The final report submitted by UNH to the funding agency (EPA’s Great Lakes Program) is available here. Documentation of the job from the applicator is available here. And finally, the image below shows runoff collected from a storm drain during the first rain event after sealcoating with RTS. The rain event occurred about a week after application of the sealcoat, and runoff would also be expected to contain particles and petroleum products that may have accumulated on the road surface after sealant application. Runoff in the tall jar on the right was collected within the first 5 minutes after the rain began. Runoff in the short jar on the left was collected about 5 minutes after the first sample.



Leave a Response