Haynesville Forecast Study: Significant Contributor of Nat Gas for 30 Years (UT Austin)

Study forecasts gradual Haynesville production recovery before final decline

John Browning, Svetlana Ikonnikova, Frank Male, Gürcan Gülen, Katie Smye, Susan Horvath, Carl Grote, Tad Patzer, Eric Potter, Scott W. Tinker

Bureau of Economic Geology, University of Texas Austin 

Low natural gas prices have slowed development of the Haynesville shale in east Texas and northwest Louisiana. Despite this, the play remains promising under the right economic conditions. This article summarizes a study by the Bureau of Economic Geology (BEG) of the Haynesville’s resource and production potential. It integrates engineering, geology, and economics into a simulation model capable of forecasting drilling and production scenarios based on several technical and economic parameters.

The article determines base cases for the remaining technically recoverable gas resources, total-field estimated ultimate recovery, potential future drilling activity, and the play’s production peak. All of these are dependent on natural gas prices.

In a base-case scenario using $4/MMbtu Henry Hub pricing and other conservative parameters, we estimate 46 tcf of cumulative Haynesville production by existing and new wells, to be drilled through 2045 and producing through 2064.

The Haynesville’s annual production has declined from a 2012 plateau of about 6 bcfd to roughly 4 bcfd in 2015. Production will recover slowly to 5 bcfd in the early 2020s, before starting its permanent decline to 1.7 bcfd by 2045.

Even using a constant $4/MMbtu Henry Hub price assumption, the formation will continue to be a significant contributor to US natural gas production for at least 30 years.  

Parameters & Methods 

The study assesses production potential in six geographic tiers and estimates future production scenarios according to these tiers.

Well economics vary across the basin because of productivity and cost differences caused by geology and other factors. The article accounts for these variances, as well as for distributions around natural gas price, drilling cost, economic limit of each well, advances in technology, and many other geologic, engineering, and economic parameters. Including these variables allows determination of how much gas can be extracted from future wells under different economic and technical conditions.

The study includes a method of estimating ultimate production for each well based on the physics of the system, rather than using just the mathematical decline curve. This method has successfully predicted shale-well production declines in other basins and was used by BEG in previous studies of the Barnett and Fayetteville shales.

Production Decline Analysis

The study analyzed the decline of all 2,527 wells drilled through 2012, determining their individual expected ultimate recoveries (EURs).9 Key input variables in the study included base-well declines; the effects of late-life deterioration from interfracture interference within the well’s drainage area; and an assumed maximum 25-year life. Decline analysis predicted an EUR of 10.3 tcf for the 2,527 wells drilled through 2012.

The study used a well-production decline method based on linear-transient flow in the reservoir.1 Per-well production decline was inversely proportional to the square root of total time over the first 1 to 2 years of well lifes, depending on reservoir properties and completions. An exponential decline followed, as interfracture interference affected production. High reservoir pressures prevented absorbed gas from contributing to production.

A theoretical linear flow solution yields a straight-line increase of cumulative production versus log time until interfracture-boundary conditions are reached within the well-fracture pattern, resulting in the predicted decline.

In addition to this theoretical model, BEG divided production for each well by the fitted value of gas within the stimulated reservoir volume, then plotted cumulative production against the square root of time divided by the time to interfracture interference (Fig. 4). Wells that had not yet experienced interfracture interference had their time-to- interference estimated from reservoir properties, and then used it to forecast production. 

Reservoir-quality tiers

BEG found the EUR of Haynesville wells increasing with lateral well length, but the average incremental EUR per unit of lateral length slightly decreasing with length. The study normalized EURs as if all wells had been drilled to a uniform of 4,800 lateral ft, reflecting common drilling practices at the time.

Length-normalized EUR/ft values were mapped along well-drilling paths, using directional surveys for all wells. Each 1-sq mile block was assigned a weighted average EUR/ ft, based on the well segments penetrating the block.

Well Economics

The study looked at average EUR/well/ tier, assuming a 20-year well life (Fig. 7). More than 80% of EUR is recovered in the first 5 years, except for Tier 1, where an average well recovers 78% of EUR during that period. Most wells will be nearly depleted by year 10, with more than 90% of EUR produced. The average EUR for all wells is likely to be lower because of attrition and economic limits.

The study’s production model includes historical attrition rates, which increase as rock-quality tier decreases.

BEG applied the average well profile in each tier to estimate average well economics. Input from operators in the Haynesville validated a representative set of well-economic parameters. 

 

Production Outlook

The study modeled the pace of future Haynesville development using the productivity-tier map, inventory of future well locations available in each tier, and an understanding of the economics of an average new well in each tier.

An activity-based model predicts new drilling based on available-location inventory and well economics. The pace of activity is adjusted annually in the model, driven by the economics of the average well in a given tier. The model distinguishes six productivity tiers based on economic incentives to drill. The historical pace of drilling is used to help scale the model’s reaction to future prices.

BEG’s model tracks the number of wells/year drilled in each tier and totals the production effect using average well profiles by tier. 

 

 

 

Views: 1902

Reply to This

Replies to This Discussion

In regards ESTIMATED ORIGINAL FREE GAS IN PLACE pic, why is it that BCF recovery does not correlate in the HAYNESVILLE 20-YEAR GAS PRODUCTIVITY pic?

Im specifically looking at my area which is Chemed Lake field but other field areas apply.  Based off the Free gas in place you would think that would equate to better wells but not the case.

 

There is a lot of meat in this that ought to keep shalers thinking into the new year.  First comments:

The maps are probably the best available, outside of subscription services.  But some areas with decent potential (e.g. southern Angelina county) aren't in the analysis.  

The study does not include well drilling and completion methods that have developed since 2012.  If I understand right, both EUR per well is increasing, and the cost is declining.  so field wide recovery may be higher.  The model also would suggest a number of wells, based on EUR, but increased lateral length means that their modeled value is likely to be reached with fewer wells.

The modeled EUR based on the wells in the study is about 4 BCF/well.  I would expect that average EUR for wells completed in 2014 and 2015 will be much higher due to high grading.

I also suspect that the lower tier acreage will be back-end loaded during development.  

Most excellent post, Keith.  I think you hit on the most vip study (until Dr. Foss's paper comes out early next year) on the importance of Haynes in setting prices at HH in the years ahead.  Important to your readers for obvious reasons and imp to me as an investor in 2019 natgas futures (now at $3 and having turned a month ago...even in the midst of the current carnage)

dbod - best data "outside of subscription services"?  Please. Which services would those be, and how please tell me us the financial and personnel resources used/their agendas vs. the 3 year study the BEG/UT with the backing of several million from the Sloan Foundation.  

The data has 20x the granularity of the data collected by the EIA which is at the county level vs. the BEG data at the square mile level.  

before addressing the rest of your comments - the "new technology" which your are inferring the BEG missed (read the study plz) and cost changes resulting from the the Saudis pulling the plug on the $90+ moat, and the high grading (neither of which are sustainable in a re-ramp given the service providers are using the same go-forward pricing the E&Ps used to use before something called "debt covenants" curtailed that behavior)....also take a look at the supply and demand projections from Bernstein.

http://www.benzinga.com/analyst-ratings/analyst-color/15/07/5703298...

from what i read - 5 more BCF/day outta Haynes before 2020.  Thing is, $4 don't pay the rent for 9 bcf/day at Haynes.  

Something's gotta give.

...and when Bern updates their data with the additional demand out of Mexico (8 bcf/day by 2020?...per Platts)...and the additional demand caused by the CPP and mercury regs...and community pushback

...and the reduced supply from Associated Gas (note that the Saudi decision was not factored in at the time of that 2014 demand/supply forecast by Bern which had 3 BCF/day incremental from Associated - and looks like we may need to reverse that and more)

maybe bern has updated - takes a lot to break away from the gravitational pull of a whole sector in liquidation - including the commodity funds that go long on long term futures.  Imagine when the firesales and the pressure from money hoping to suffocate current owners disappates

http://www.investing.com/commodities/natural-gas-contracts?page=cha...

The relevance of any analysis of the Haynesville Shale prospective area is severely limited by the use of data collected from 2008 to 2012.  The evolution of the design of Haynesville Shale wells generally falls into three categories to date:  The early wells that remained largely unchanged excepting the change in surface locations from within the original units to adjacent sections, call them HA 1.0, the slightly longer laterals with significant changes in stage lengths, perf clusters and amount of proppant, call them HA 2.0, and the current Cross Unit Lateral (HC wells) which have continued and advanced the changes begun in the HA 2.0 wells.  The HA 1.0 wells would be the ones drilled from 2008 to 2012.  The 2.0 wells from 2012 to 2014.  And the 3.0 wells from the end of 2014 to the present.

Skip,

Did you check out the industry and educational backgrounds of the 10 authors that put this study together?  

So you are suggesting they just used old/obsolete data and projected it out for years...thereby making the report that is absolutely worthless.  Ok.

They do mention in a couple places they take into account tech advances (which, of course, are fighting the move to inferior rock):

"The study underlying this article used production data from all individual Haynesville wells drilled 2008-12, starting with the production history of all wells and then determining what remains to be drilled under various economic, geologic, and technologic scenarios."

"...The article accounts for these variances, as well as for distributions around natural gas price, drilling cost, economic limit of each well, advances in technology, and many other geologic, engineering, and economic parameters."

----- 

Regarding the 1.0, 2.0, 3.0 wells...look at these numbers from a recent EIA study. Look at the costs from Dec. 2012 to Aug. 2014. Where are all the cost savings for 2.0?

No doubt "3.0" has some savings - as the cost cratered after the Saudi decision and service providers reduced costs 30 or even 40%, high grading, layoffs, etc...but much of that will be reversed in a re-ramp.

http://www.eia.gov/todayinenergy/detail.cfm?id=21712

------ 

BEG quotes a lot of refs in their paper.  Interesting no mention of the four studies by Kaiser & Yu (from LSU).  Their 3rd and 4th (last) installments were published Feb and March/2014 – the 3rd one is on economics. At $4 per MCF, the cumulative value of Haynesville wells have an NPV of NEGATIVE $2.2 Billion.

http://www.ogj.com/articles/print/volume-112/issue-2/exploration-de...

When I can find the time around holiday commitments I can supply you with some specifics.  I'll try to get to that in the next couple of weeks.  I'm limiting my comment at this time to what I know about well design and performance.  I'm not criticizing or calling into question the expertise of the authors.  I do however run across more than a few scholarly papers and expert analysis that use data sets that don't tell the full story, IMO.  The Haynesville Shale is one of those that crops up somewhat frequently, particularly when Art Berman is on the prowl, and that I have a deep and broad knowledge of from following it on a daily basis from its inception. 

@ Arrest ID

I am not a geologist or an economist (thankfully).  But I am a scientist.  Crappy well economics is crappy well economics.  And averages are misleading.  The economics link you sent is behind a pay wall.  I have worked for some smart people who think there is money to be made when gas is below $4 in selected parts of the Haynesville, although I suspect $2 gas is a money loser in the near term. 

I stand by my earlier comments - EUR for 2014 and 2015 wells will be higher than that described in the analysis, for a given rock type/tier, based on changes in well design/completion, as well as a flight to the best rock.  

Regarding data, there are aggregators who can give you cumulative production and decline curves for any of the wells with public reports, and you can get well by well completion information as well, at least everything that is publicly available.  If you participate in a log library, you can get tons of well logging data too.  As i said, I think this is probably the best available to the public, but it is incomplete on the edges.  

Economics dictate that the poorest rock will only get drilled when the price gets back up, so it will back end load the development in the fringes.  

ultimately, the well count per unit area of the play will be lower than based on the study criteria, because changes in completion and well design mean a single well covers a lot more rock than a 2012 well did.  The marginal cost to drill and complete the next 2500' is a lot lower than the first 5000' of the lateral.  

This "study" is more "speculation" or "wild guess" based on theory of small amount of scientific data.  I wonder if anyone could determine the longevity and total reserves of natural gas originally in place contained in the "Monroe Gas Rock" formation?  Assuming the original reserves were still in place due yet to be exploited by drilling operations of course.  

I found a lease dated around 1913 while working a project in Monroe some years ago.  I am not sure exactly when the discovery date of the shallow gas and development began but the longevity astounded me.  Even more astounding was the lady who executed the lease in 1913 resided in New Orleans and included a rider in the lease with worded in style of a pugh clause or "Freestone rider" and she should have been given credit for such vision or whoever advised her on the lease contract.  I wish I still had a copy I could share for all site members.  Gas was still being produced from the lease when I was working the project in 2001.

1. here’s another link to the study (no paywall)

http://www.beg.utexas.edu/shale/docs/Haynesville%20Shale%20Gas%20Pl... 

2. yes, marginal cost per additional foot goes down, but the “BEG found the EUR of Haynesville wells increasing with lateral well length, but the average incremental EUR per unit of lateral length slightly decreasing with length. The study normalized EURs as if all wells had been drilled to a uniform of 4,800 lateral ft, reflecting common drilling practices at the time.”

One of the key diffs. between this study and the EIA/industry is the resolution.  The BEG data has 20x the granularity.  The EIA takes the wells drilled in a county to date, and applies that avg. to ALL the wells in a county.  Given the focus on sweetspots and their size relative to the size of the county – and this is KEY (see figure 5 for an eg), the EIA overestimates the amount of natgas recoverable at a given price. 

The BEG analysis is done at the square-mile level…and it was not done by a team of enviro/academics.  Again, would suggest reviewing the background of the authors and their professional backgrounds and industry connections.  This is not one person; nor is it 1913.

I like the BEG, I think they generally do good work.  I am not a geologist or economist, so I make no claims regarding their reserve estimates.

Agreed that the EIA maps are not particularly compelling, nor are most USGS maps.

Regarding industry maps, there is one set of maps at any company for public consumption, then often a much more detailed set of maps for their geologists, based in software like Petra.  The economics are obviously a matter of debate within the Industry, with some folks abandoning acreage, some some continuing to drill.  Obviously those leaving didn't think the economics make sense, but those staying do.  

The BEG map is not defined to an end point around its edges.  Notably, the very successful EOG Sarge well in Angelina County, and the less successful vertical test of the Bossier in Trinity County.  I assume there are other missing edges in Louisiana.  

I haven't looked in detail enough, but I presume this map is Haynesville only and does not co-mingle Bossier into the data.  

Regarding the slight decrease in incremental EUR with lateral length, I would argue this is due to the dates of the well in the study (which was necessary - you can't delay the study until the last well is drilled).  Spend some time on a frac site and you'll find out we've gotten a lot better at stimulating longer laterals.  This influences the economics a lot.

I think its a neat study and pretty useful.  But I stand by my earlier comments, in particular this one " - EUR for 2014 and 2015 wells will be higher than that described in the analysis, for a given rock type/tier, based on changes in well design/completion, as well as a flight to the best rock. "

The BEG study estimates of IRR would suggest that no one should be drilling in the Hanyesville at the current gas price. 

My life has been both of joy and sorrow and some of my darkest days were spent during the Haynesville play back in 2009.  I would not wish my life on anyone but would not trade a single day of my ongoing life adventure.

Seem those who judge other do not follow the teaching of Jesus but think they are of equal virtue which is false.  Do not judge me on my past as I no longer reside there.

RSS

Support GoHaynesvilleShale.com

Not a member? Get our email.

Groups



© 2024   Created by Keith Mauck (Site Publisher).   Powered by

Badges  |  Report an Issue  |  Terms of Service