A couple of weeks ago (29 March) there was a significant failure of an ExxonMobil oil pipeline in Mayflower, Arkansas. It attracted a lot of attention because it occurred in a suburban area – you are guaranteed to get lots of coverage when there are stories of numerous residents evacuated for a fortnight and pictures of thick black tar-sands crude pouring down kerbed streets. A total of around 800 kL was spilled, with implications for the neighbourhood and a nearby artificial lake but fortunately there was no fire.
Investigations into the failure are proceeding relatively slowly; most of the effort has been directed to the clean-up. The damaged pipe section was cut out for testing only yesterday. Although there are no official reports this picture tells at least part of the story:
Clearly there has been a failure of the longitudinal seam in a full joint of pipe, from girth weld to girth weld. The split is described as being 22 ft long (6.7 m). The pipeline is apparently 65-70 years old, probably made from old low-frequency ERW pipe (quite different to the modern high-frequency ERW pipe used now). Exactly why it failed now is not yet known; operating pressure at the time of failure was well below MAOP and there has been mention of hydrostatic testing not too many years ago.
Investigation seems to be in the hands of the Arkansas Attorney General’s department. The NTSB appears not to be involved, and I can’t find out anything about PHMSA involvement because their web site is down as I write this. I have copied one of the more interesting commentaries here.
In Australia we are fortunate that we have very few “legacy” pipelines, and certainly none as old as this. We seem unlikely to see similar failures here.