Wastewater Spills In North Dakota: What The Data Tell Us

Print More

Jordan Wirfs-Brock/Inside Energy

Each dot represents a reported spill, plotted by date and size – the bigger the dot, the bigger the spill. (The vertical axis, displaying size, is shown on a logarithmic scale.) As you move to the right, spills toward the top get denser, which means large spills are becoming more common.

Wastewater, also known as brine, is a by-product of oil and gas drilling. When it spills on soil, it can leave the land barren for years. When it spills in waterways, it can be toxic to plant and animal life. This is an ongoing problem anywhere drilling occurs – earlier this month, a wastewater pipe in North Dakota leaked 2.94 million gallons into Blacktail Creek, the worst brine spill since that state’s oil boom began.

When Inside Energy first reported on this topic in August of 2014, we found that the rate of spills per well in North Dakota has increased since the start of the oil boom. In the intervening months, the New York Times released several major investigative stories on North Dakota’s oil industry, including a comprehensive database of reported spills (available for anyone to download) which opened up information that was difficult to extract from the North Dakota spills website.

So we decided to revisit North Dakota spill data, focusing on wastewater (referred to as salt water in the North Dakota Department of Mineral Resources database and as brine in the New York times spills database). Here are three of the most important things we learned:

1) By volume, roughly twice as much brine is spilled as oil.

More than 3 million gallons of brine were spilled in 2014 alone, compared to 1.3 million in 2010. View a table of yearly brine and oil volume totals.

Jordan Wirfs-Brock / Inside Energy

More than 3 million gallons of brine were spilled in 2014 alone, compared to 1.3 million in 2010. View a table of yearly brine and oil volume totals.

What we know:
Based on what is reported, from 2006 to the end of 2014, there were more spills involving oil than brine – 5,216 oil spills versus 3,650 brine spills. But size matters. In terms of volume:

  • 5.9 million gallons of oil were spilled in North Dakota, an average of 1.4 gallons per minute.
  • 11.8 million gallons of brine were spilled in North Dakota, an average rate of of 2.8 gallons per minute.

output_BffNiI
What we don’t know:
Reporting on spill volume is iffy at best, especially when it comes to brine. Volumes are often rough estimates, or simply omitted from the official North Dakota spill report. For roughly five percent of spill reports submitted to the state since 2006, the volume section of the form was left blank.

2) The average size of brine spills is getting larger, and large brine spills are getting more common.

Each dot represents a reported spill, plotted by date and size – the bigger the dot, the bigger the spill. (The vertical axis, displaying size, is shown on a logarithmic scale.) As you move to the right, spills toward the top get denser, which means large spills are becoming more common.

Jordan Wirfs-Brock / Inside Energy

Each dot represents a reported spill, plotted by date and size – the bigger the dot, the bigger the spill. The vertical axis, displaying size, is shown on a logarithmic scale. As you move to the right, spills toward the top get denser, which means large spills are becoming more common.

What we know:
We know that the number of spills per well is increasing, and that this is true for both brine and overall spills. Those spills are also getting bigger. With the exception of 2006, when a single brine spill released one million gallons, the average brine spill size has grown from 2,811 gallons in 2007 to 3,524 in 2014 – a 25 percent increase. The number of large spills has also grown over time. For example, in 2006, there were two brine spills larger than 50,000 gallons; in 2014, there were seven. View a table showing the annual counts of large spill sizes.

What we don’t know:
Because of the problems with reporting volume mentioned above, analyzing spill size over time is tricky. We also have very little information about how much area each spill covered, so it’s impossible to know how much land or water has been touched by brine.

3) Some companies are serious repeat offenders when it comes to brine spills.

Search for spills by company in the New York Times interactive database. View a summary table of company ranks by active wells, all spills and brine spills.

Search for spills by company in the New York Times interactive database. View a summary table of company rankings by active wells, all spills and brine spills.

What we know:
Some companies spill more than their fair share, as the New York Times reported, calling out Continental Resources as North Dakota’s top spiller.

As of January 27, 2014, 188 companies were operating active wells based on North Dakota’s wells database. The chart above visualizes where the top 20 oil companies operating in the Bakken ranked – as determined by number of active wells –  based on overall number of spills (including oil, brine and other chemicals) and brine spills alone since 2006. It’s not a perfect comparison, but it does provide a rough idea for how companies measure up.

Continental Resources leads North Dakota in active wells, spills of all kinds, and wastewater or brine spills. Some companies spill less than expected, like Fidelity, Enduro and QEP, which are among the state’s top 20 producers in terms of number of active wells, but rank in the 30s and 40s for spills. Others, like SM Energy and Oxy, spill more. SM Energy is eleventh based on number of wells, but has the second highest number of brine spills.

Looking at spills by company since 2006:

  • 21 companies reported more than 100 spills
  • 17 companies reported more than 50 brine spills.

What we don’t know:
Wells are always coming on and offline, so to get an accurate picture of which companies are spilling disproportionately, we would need to look at how many wells they have been operating over the entire period from 2006 to 2014. Also, not every spill gets reported. Companies that meticulously report spills may appear to have worse records than those who under-report. We have no way of knowing how many spills never make it to the state’s database.

What’s next?

  • Check out our recent coverage of the latest spill in this story.
  • Accurate and transparent information is essential for understanding a problem as serious and widespread as wastewater spills. Stay tuned for reporter Emily Guerin’s examination of how state officials have been addressing this issue.

  • Anonymous

    Why aren’t the figures compared to volumes? In a state that has increased it’s production 25x, a 2.5x increase in incident volumes is a success.

    • Jordan Wirfs-Brock

      Hi,

      My name is Jordan Wirfs-Brock, Inside Energy’s data journalist. That’s a very good point about volume being an essential part of the analysis required to really understand what’s going on with spills. There are three main reasons that we’ve focused on the spill number when looking at the spill rate over time:

      1) The spill number to well number ratio is what North Dakota’s Department of Mineral Resources looks at (as well as containment rate). When we asked them specifically if they tracked spills size and volume trends over time, they told us they did not. From an email with Alison Ritter of the DMR: “No, we do not track size over time. Again, our priority when looking at spills is was the spill contained or did it leave location?”

      2) Volume reporting is inconsistent. As mentioned in the post, sometimes it is not given at all, sometimes the value listed a very rough estimate, so volume data should be taken with a grain of salt – sorry, pun intended! However, at Inside Energy we felt it was better to provide imperfect data, along with an explanation of its limitations, than to not look at it at all.

      3) We tried to get produced water/brine/salt water volume that came out of the ground, over time, so we could compare the spill rate to the volume of waste water produced. The waste water production numbers we’ve received from the state are, at this point, contradictory, so we are still working on getting good figures. Once we do, we are hoping to follow up with a production volume to spill volume rate analysis.

      I appreciate your comments, and please let me know if there is other analysis you would like to see. Data only tells part of the story. Make sure you take a look at the on-the-ground reporting Emily Guerin has been doing in North Dakota as well: http://insideenergy.org/2015/01/28/in-north-dakota-oilfield-spill-problems-worsen/

      Best,
      Jordan

  • Pingback: North Dakota Plagued By Oil Spills: 745 in One Year – Counter Current News()

  • Pingback: North Dakota Plagued By Oil Spills: 745 in One Year | Junk Host()