Perspectives on U.S. Natural Gas Resources

Author: Zachary Ming

Only six years ago, analysts projected that by 2012 natural gas in the U.S. would cost around $10/million Btu and as much as 15 percent of daily consumption would come from imported liquefied natural gas (LNG).  Today, gas actually costs around $3/million Btu, and policy makers are contemplating whether the U.S. should export gas.  This complete reversal can be attributed to the ‘shale gale’ – the unexpected breakthrough in producing large volumes of gas from organic rich shale rocks previously thought to be unproducible.  This is due in large part to the application of horizontal drilling techniques and multi-staged hydraulic fracturing.  Since 2006, U.S. gas production has increased by almost 32 percent, and prices are down 61 percent.  This is the new landscape of natural gas.

U.S. natural gas production has exceeded expectations.

If it is clear that new technology is opening up vast amounts of new production potential, the question then becomes, exactly how much gas do we have?  It is an important question for policymakers and businesses as they decide whether to invest in capital-intensive projects over the coming years. For example, about $70 billion of planned industrial investment and 41 gigawatts of coal-fired power plant retirements depend on the long-term availability of low-priced gas. Considerations of natural gas as a transportation fuel and LNG exports also depend on growing supplies. Decision makers are cautious after the volatility and price spikes of the last decade.

In their 2011 Annual Energy Outlook, the Energy Information Agency (EIA) estimated U.S. recoverable natural gas resources to be 2,543 trillion cubic feet (TCF).  This represents 100 years of supply at current natural gas consumption rates.  The EIA also ascribed 410 TCF of this technically recoverable resource to the Marcellus shale, a gas field in the northeastern United States that has seen its share of controversy.

A few months later, the United States Geological Survey (USGS) reported that the Marcellus shale contained 84 TCF of undiscovered technically recoverable reserves.  Natural gas critics, such as reporter Ian Urbina from the New York Times, assumed the USGS number represented a huge downward adjustment of the EIA estimate and cited this as proof that there is far more uncertainty surrounding natural gas resources than previously purported.

So, what is the right number? The italicized descriptions from the two reports are critical. To understand the descriptions, one must understand how reserves are classified in the United States.

The most basic and fundamental category is simply, reserves.  According to the broadly accepted Society of Petroleum Engineers (SPE) Petroleum Resource Management System standard (Figure 1), reserves are commercially recoverable with today’s prices and today’s technology.  There are three subcategories within reserves – proved, probable, and possible.  Proved reserves have the highest degree of certainty.

Next, contingent resources are recoverable with today’s technology, but they may not be economically recoverable at today’s prices, or they may have other obstacles such as government regulations.  Finally, prospective resources are an estimate of what is yet-to-be discovered and are based on indirect evidence, not production data.

A breakdown of resource definitions

The use of standard definitions might have spared many anxieties over the meaning of the EIA and USGS resource estimates. In retrospect, the EIA technically recoverable resource estimate would match the contingent resources category. Classification of the USGS undiscovered technically recoverable reserves is less certain. Undiscovered would imply a match with prospective resources but technically recoverable would fit with contingent resources.

Adding to the confusion surrounding the size of the Marcellus gas resource, in the 2012 Annual Energy Outlook, the EIA reported 141 TCF of unproved technically recoverable reserves. Summing previously published proved reserves figures with this leads to a total estimate of 154 TCF, but it is not clear if these numbers incorporate estimates from all categories of the SPE classification. Independent research firm IHS reports that its updated Marcellus resource estimate ranges from 267 to 534 TCF depending on gas prices and the ultimate drilling density. This range of estimates spans the three prime SPE categories and also embraces the original EIA estimate of 410 TCF.

Confusion regarding the EIA and USGS estimates might have been avoided had they tuned their methodology and the terminology used to describe their estimates to align with a standard classification. Reporting a range estimation rather than trying to compress the issue into a single value might have also helped to minimize the conflict. One can see how different approaches to reach a single number from the plethora of data might result in misunderstandings.

Definitions bring both clarity and complexity to the issue.  Surely, it is difficult to ascribe a single number to total recoverable resources when this is an aggregation of multiple categories.  How much weight should be applied to what we know (proved reserves) vs. what we don’t know (prospective resources)?  To answer this question, it is useful to examine another popular metric.

For years, analysts have used the reserves-to-production (R/P) ratio (total proved reserves divided by current annual production) as an indicator of reserve life.  Theoretically, this should provide with certainty the remaining number of years of supply of natural gas.  Nevertheless, the U.S. practice of using only proved reserves in this calculation has limited the value of this metric. From 1977 through 1999 the R/P ratio languished between 8.3 to 8.9 years in spite of the passage of over 20 years. The R/P ratio did not exceed 10 until 2009 even though the estimated U.S. technically recoverable gas resources reached almost 2,100 TCF – more than eight times the proved reserves. The 2009 resource-to-production ratio of 82 provides much more insight into the relative health of U.S. gas resources and supplies than the traditional proved reserves R/P ratio.

From this example, it is clear that it is critical to understand the size and distribution of contingent and prospective resources to properly understand the relative risks of transforming resources to supplies.  For this reason, the SPE and others in the petroleum industry are pushing government agencies and regulators to adopt more transparent reporting of all components in official reserves and resources data compilations.

In the end, the EIA updated and downgraded their total U.S. natural gas resources estimates in the 2012 Annual Energy Outlook. Their ultimate revised results don’t differ at all from the notion of plentiful gas.  Their new estimate of over 2200 TCF for the total recoverable resource represents almost 90 years of natural gas supply at current consumption rates.  And this number may very well increase; the industry has a history of revising this number upward, not downward.

The Stone Age did not end for a lack of stones and the current age of fossil fuels will not end for a lack of hydrocarbons.  Distinguished author Thomas Friedman acknowledges that natural gas is a game changer for the economy, environment, and national security, but only if it is done right.  Natural gas presents challenges; An adequate resource is not one of them.

Image Source:


Zachary Ming is a graduate student in Management Science and Engineering with a focus on energy, the environment, and economics.  He has experience in natural gas trading and resource evaluations.

Pete Stark is Senior Research Director and Advisor for IHS CERA in Englewood, Colorado. He has analyzed gas and oil resources and supplies for more than 20 years.