Even oilfields aren’t immune to the ravages of time: A new study finds that as some of the world’s largest oilfields age, the energy required to keep them operating can rise dramatically even as the amount of petroleum they produce drops.
Failing to take the changing energy requirements of oilfields into account can cause oilfield managers or policymakers to underestimate the true climate impacts, Stanford scientists warn.
The new findings, published in the journal Nature Climate Change, have implications for long-term emissions and climate modeling, as well as climate policy. “Current climate and energy system models typically don’t explore the impacts of oil reservoir depletion in any detail,” said study co-author Adam Brandt, an assistant professor of energy resources engineering at Stanford’s School of Earth, Energy and Environmental Sciences. “As oilfields run low, emissions per unit of oil increase. This should be accounted for in future modeling efforts.”
An accurate estimate
In the new study, Stanford postdoctoral researcher Mohammad Masnadi worked with Brandt to apply a new software tool developed at Stanford for calculating greenhouse gas emissions to oilfields around the world that have produced more than 1 billion barrels of oil over their lifetimes, sometimes called “super-giant” oilfields.
Conventional greenhouse gas estimates calculate emissions through a kind of economic reverse engineering, whereby an economic index is used to convert the monetary value of an oilfield’s final products – whether it be processed oil, natural gas or petroleum-based products – into greenhouse gas emissions. “This top-down approach for converting economic values into environmental and energetic costs misses a lot of underlying information,” Masnadi said.
What’s more, many studies look at data from only a single point in time, and as a result capture only a snapshot of an oilfield’s greenhouse gas emissions. But the Stanford scientists argue that in order to paint the most accurate picture of an oilfield’s true climate impacts – and also have the best chance of reducing those impacts – it’s necessary to assess the energy costs associated with every stage of the petroleum production process, and to do so for the oilfield’s entire lifetime.
Developed in Brandt’s lab at Stanford, a software tool called the Oil Production Greenhouse gas Emissions Estimator (OPGEE) is designed to do just that. For any given oilfield, OPGEE performs what’s known as a lifecycle assessment, analyzing each phase of the oil production process – extraction, refinement and transportation. It then uses computer models to calculate how much energy is consumed during each step. From this, scientists can calculate precisely how much greenhouse gas each oilfield emits.
“This bottom-up type of analysis hasn’t been done before because it’s difficult,” Masnadi said. “For this study, we needed over 50 different pieces of data for each oilfield for each year. When you’re trying to analyze an oilfield across decades, that’s a lot of data.”
Unfortunately, most oil companies are reluctant to release this type of temporal data about their oilfields. The Stanford researchers developed two workarounds to this problem. First, they gathered data from places where transparency laws require oil production data be made publically available. These included Canada, Norway and the U.K., and the state of California in the U.S. Secondly, the pair conducted a deep data mine of the scientific literature to seek out clues about oilfield production levels in published studies.
A second-year PhD student in Earth and planetary sciences and bestselling science fiction author, Ashing-Giwa never misses a chance to blend lab and lit.
A Stanford dune expert discusses watching desert-based movies from the perspective of a geoscientist, the realities of otherworldly dunes, and what his research can tell us about the ancient environment of Earth and other planets.
Difficulties in connecting charging sites to the grid pose the biggest delays in bringing publicly accessible EV charging stations online.