Spill Analysis by Year

In addition to persistently-poor air quality, the immense amount of oil & gas infrastructure present throughout Colorado poses another risk in the form of spills—of oil, methane, and produced water. While the amount of “spilled” methane gas is more difficult to quantify, the Colorado Energy & Carbon Management Commission (ECMC) has done a better job of capturing the quantities of oil and produced water spilled at oil & gas facilities in Colorado. This data is available to download from the ECMC web site, but is poorly presented as a data table PDF that cannot be machine read into Microsoft Excel or other tools for easy analysis.

Can you spot the “analysis?” The amount of oil and water spilled has been expressed as a percent of the total volume of oil and water produced, respectively. At best, it seems that the volume of oil spilled has declined relative to production, but the number of spills have quadrupled over the reporting period! What’s happening?

Qualitative Charts to the Rescue

Let’s look at some qualitative charts of the data, meaning we’re going to look at changes within data series and loosely relate them to other data without relying too much (if at all) on the absolute numbers.

First, let’s look at the number of spills relative to the number of active wells. Spills are on the rise (bad), but the number of active wells is falling (good). We’re plugging and abandoning more wells than we’re drilling (as has been the case since 2016).

Are the spills more related to the process of drilling/fracking or plugging and abandoning a well? Note that we only have “fracked” numbers from 2012 onwards from FracFocus.

Hypothesis: The increase in spills lag behind, but may be correlated to the plug and abandonment process.

How does having fewer active wells affect production? After peaking in 2019, production has been on the decline as well. This makes sense given horizontal drilling became the majority drilling method in 2012, and horizontal wells produce the majority of their lifetime oil in the first three years.

Trivia: Production is measured in 42 gallon “blue barrels” (BBL), which has been the industry standard since 1872.

Let’s look at the quantities of oil and produced water spilled—with some peaks in the mid 2000s, the volumes of spilled oil and produced water have held fairly constant over the last 10 years.

Hypothesis: Spills happen during the normal lifetime of a well, not as it’s being drilled.

The volume of oil spilled has decreased in recent years despite production increasing (good).

Hypothesis: With the dearth of inspections of existing oil & gas infrastructure, spills are generally only discovered as wells are being plugged and abandoned.

Looking at the volume of produced water spilled vs produced, it’s more difficult to draw any conclusions.

Hypothesis: The rate of spilled water is relatively independent of the amount of water produced.

Finally, we show the ECMC “analysis” in graphical form. As a fraction of the amount of oil and water produced, spills have relatively decreased. Oil production is 6 times higher today than it was in 1999, thus artificially suppressing the amount of oil spilled.

Hypothesis: The amount of spilled oil has relatively little to do with the amount of oil produced.

Hypothesis: The decrease in spilled oil has more to do with the increasing numbers of plugged and abandoned wells.

Thank you for taking the time to read through this analysis! It’s our pleasure to turn the data provided by the ECMC and others into more useful information. We’d love to hear from you in the comments below on your interpretation of the data.

Source Data

The source data for this analysis is available on our Github repository with a Creative Commons Zero v1.0 Universal license.

Leave a Reply

Your email address will not be published. Required fields are marked *