I was listening to the radio the other day to a discussion about fracking, that contentious method for extracting natural gas and oil, and one of the callers expressed concern with the fact that natural gas has a tendency to go boom, with disastrous consequences. Now, here in my corner of the world most people heat with fuel oil. Some use propane, and for that they have propane tanks. City dwellers may have gas lines, but those of us living out in the woods tend not to.
Natural gas forms explosive mixtures with air for air/gas concentrations from 5%–15% by volume. Above and below those limits, it won't explode; within those limits, it will. In case you're curious, the U.S. Dept. of Transportation's Pipeline & Hazardous Materials Safety Administration keeps records of what they call serious pipeline incidents, accidents that result in either deaths or injuries sufficiently severe as to require hospitalization. Here's a link to their listing of serious incidents for all pipeline systems for the period from 1992 to 2011, and here are the numbers for gas distribution incidents for the same period. Pretty dry, right? Not quite as attention grabbing as the tragic 2010 gas explosion in San Francisco's San Bruno suburb that killed eight people and sparked a flurry of investigations into how the safety of PG&E's natural gas pipeline systems was monitored.
The fact is, there are many, many pipelines in our lives—water, sewer, and gas to name just three. The process industries use pipelines all the time and have devoted a great deal of time, money, and ingenuity into finding technologies that help them monitor pipes to spot problems or leaks as early as possible. Water leaks are wasteful. Gas, oil, or chemical leaks are wasteful and dangerous. It should come as no surprise, then, that concerned parties have been doing some research into better sensor networks for pipeline integrity monitoring and in late 2010, the final report (PDF) on the Instrumented Pipeline Initiative was published to develop and evaluate a sensor system for low-cost monitoring and inspection.
The Instrumented Pipeline Initiative addresses the health of the pipelines, but how about checking for leaks? I've got two interesting approaches to mention, and this is by no means a complete list. The August 2010 issue of Pipeline & Gas Journal contains the article "Airborne Leak Detector Certified In Germany For Urban Gas Grids" that discusses a German project to use an airborne remote methane detector (housed in a helicopter) to spot methane leaks. Or, the project out of Boston University that involved a car-mounted sampling system and a cavity ringdown spectrometer (a very sensitive gas analyzer). By detecting the natural gas concentrations as the researchers drove around the city, they generated maps of gas leaks.
Even with systems in place leaks are still going to happen. But maybe by using combinations of systems such as the ones I've mentioned, the incidence of dangerous leaks can be decreased.