In July 2007, the lives of residents of Burnaby, BC were turned upside down when Kinder Morgan's TransMountain Pipeline was ruptured in their neighbourhood. A road crew's excavator struck the pipeline. The result was disastrous.
Regulators ruled that the pipeline was improperly marked. To their end, Kinder Morgan has rejected and challenged that finding. They feel that the construction crew did not follow proper procedures, and that the road crew was at fault for the rupture. To this day, residents remain deeply concerned that history could repeat itself -- except perhaps worse -- if the expansion of the TransMountain pipeline is approved.
As it was their properties, their neighbourhoods, covered by oil from that rupture it's hard to begrudge them those worries. They shouldn't be begrudged their concerns.
Of course, accidents like the one in Burnaby can be prevented if everyone properly follows the necessary procedures. But, as the disagreement between Kinder Morgan and provincial regulators shows, that doesn't always happen. Regardless of who is right, someone was wrong. That may seem like a tautology, but it's a deeply relevant one.
If Kinder Morgan had at their disposal some means of monitoring the pipeline route, perhaps the accident could have been avoided.
Enter German firm Definiens, French firm GDF Suez, and Spanish firm Enagas. Their iNTeg-Risk joint venture may provide just the tool for such a job.
Using unmanned aerial vehicles -- more specifically the Gatewing X100 UAV platform -- iNTeg-Risk would take flyover video footage of the pipeline route and analyze it in order to detect any developments that may pose a risk to the pipeline. This would include not only the work of nearby construction crews, but potentially even geological developments.
The entire process would be automated. The UAVs would fly on a precisely-programmed route, navigating according to GPS coordinates. The goal is to capture footage as perfectly comparable as possible.
In the second stage, a computer program named eCognition compares the image to previous images to detect any changes. This process can be extremely complex -- comparing thousands of what the iNTeg-Risk partners call a "geo-referenced mosaic of images."
In the third stage, these images are processed to detect construction work, the staging of construction work, soil upheavals, soil erosion, the planting of new trees, and other potential threats. The eCognition software uses specialized algorithms to identify objects according to their shape, and according to the shadow they cast.
The iNTeg-Risk partners have approached the issue of pipeline safety with what some people might consider to be rather novel: that a false alarm is better than no alarm at all. Of course, the system can't work on its own. It relies on human decision-makers to take these alarms -- false or not -- seriously. While iNTeg-Risk can improve the abundance of information regarding threats to a pipeline, and increase the ease with which it's collected, it cannot in and of itself replace often-flawed human decision making.
The recommended interval between monitoring -- two weeks at a time -- could also leave something to be desired considering the pace at which modern construction can sometimes proceed.
This is all before one considers that monitoring a pipeline such as the TransMountain pipeline with unmanned aerial vehicles could be perceived as threatening to the privacy of residents in the area. That's worthy of consideration as well.
Then again, if such monitoring can prevent another accident like the one in Burnaby, perhaps local residents may be comforted by it. Preventing such accidents is easily worth it.