T he year 2009 heralded the beginning of a new crude oil production boom in North America. The discovery of massive shale plays in the Bakken region around North Dakota (and other neighboring states) coupled with developments in hydraulic fracturing among other extraction technologies were key drivers of this growth in the United States. The Permian Basin in Texas and Colorado also entered a new phase of exploration and development, leading to new investments. At the same time, Canada was also experiencing an expansion in its crude oil production, largely from the oil sands of Alberta. In the six-year period from 2009 to 2014, Canada’s production grew by 40% from 2.58 million barrels per day (mbpd) to 3.6 mbpd. During the same period, US crude oil production grew at an even faster rate of 57%—5.35 mbpd to 8.68 mbpd. The American trend is all the more striking given that domestic production had been on the decline at an average rate of 40% in the preceding 24 years. Production in Mexico has been contracting since the early 2000s, partly due to a lack of investment in capacity. However, from 2009, crude oil production stabilized as Mexico posted a growth rate of -24% from 2004-09 compared to -7% from 2009-14.

Crude oil production in North America (Canada, Mexico and USA) from 2009 through 2014. (Data source: EIA)

This extraordinary growth in American oil production took the transportation sector by surprise. The relatively short span of time was inadequate for large-scale and potentially risky investments in transport infrastructure. Mexico largely ships its unrefined crude to the US and other markets around the globe. It has no land links for international transfer within the continent and its inland refining capacity is limited. Canada does not have transnational crude oil transport capacity and has relied on the US to market its oil over the past several decades. Thus, there are several oil pipelines that transfer heavy crude from Western Canada to US refineries. These, however, have not been adequate for increased production. In response, crude-by-rail movements have skyrocketed both across Canada (from West to East) and from Canada to the US. More significantly, crude-by-rail movements increased by more than one order of magnitude from 55 thousand barrels per day (kbpd) in 2010 to 875 kbpd in 2014. Consequently, the number of rail spills and fires have risen in the same period.

Volume of crude oil transported across the United States and between the US-Canada border from 2010 to 2014 (Data source: EIA/NEB)

A major incident was the train derailment and subsequent explosion in Lac-Mégantic, Quebec, which destroyed half the town and claimed 47 lives in Canada. Other major incidents in the US include the following:

  • 14 barrels (bbl) of crude oil spilled in 94-car train derailment, Parkers Prairie, MN (2013)
  • 3 gal of crude oil spilled in 15-car train derailment, Penobscot, ME (2013)
  • 25-car derailment (train below 40mph limit) spilled oil into marsh, Aliceville, AL (2013)
  • 13-car derailment; 595 bbl spilled into James River; 78000 evacuated; Lynchburg, VA (2014)
  • 786 bbl of crude oil spilled in 22-car train (106 cars) derailment, Culbertson, MT (2015)

The map below (source: Earthjustice) details the major crude-by-rail incidents that have occurred across North America in recent years.

While pipelines spill more oil per incident, rail spills tend to be devastating, as the lines run by rivers or water sources and populated areas. Crude-by-rail spills therefore have associated public-safety and environmental risks. Furthermore, the crude-by-rail spike has also disrupted the agricultural sector, as operators have prioritized oil cargo at the expense of grain shipments for reasons of profitability.

Given these pressing concerns, urgent steps need to be taken to address the situation. As researchers, we can use equilibrium models to analyze the oil market in North America and perform scenario analyses to provide best solutions for capacity investments to address the crude-by-rail issue. There is a void in the academic literature in this regard. Our proposed tool—the North American Crude Oil Model (NACOM)—is currently in development.