Too Much Marcellus/Utica NGL Pipeline Capacity? Or not enough?

Three years ago the predicted onslaught of Marcellus NGL production kicked off a horse race to build new ethane pipelines out of the region.  At one time, at least seven different projects were being promoted. Since then most of the projects have dropped by the wayside, yielding to Mariner West (MarkWest/Sunoco) and ATEX (Enterprise).  Do those two projects provide enough capacity?  Perhaps too much?  What about possible competition from a new Marcellus/Utica NGL pipeline project that hit the radar screen last week?  Could that be a signal that a lot more liquids are on the way?  

Last week I spoke at the Infocast Marcellus & Utica Infrastructure Summit in Pittsburgh.  Capacity issues out of the region for both ethane and propane were the topics of my presentation, titled ‘Marcellus Infrastructure: Getting Ethane and Propane to Market’.  We can’t cover everything in that presentation here, but today I’ll hit the high points of the ethane situation and we’ll pick up the propane issues later in the week.    

Before we get started, we need to recount a semi-snafuish interchange from the Infocast Summit.  We mention it here only because it is directly relevant to our topic.  It started when Caiman Energy CEO Jack Lafield mentioned in a panel presentation that Williams Partners (recent purchaser of Caiman’s midstream business and major investor in Caiman Energy II) is planning to build a new 150-200 Mb/d pipeline to move y-grade from MarkWest’s Houston plant in Pennsylvania to the Gulf Coast.  There was even a name for the new pipeline – Bluegrass.

The story was picked up the next day by the energy press.  Well….the announcement must not have been ready for prime time, because the next day Williams tried to put the quietus (East Texas word) on the whole thing. For example, Argus Natural Gas Americas noted in their next edition that Williams has “no immediate plans to build NGL takeaway capacity out of the Marcellus or Utica” and indicated that Mr. Lafield clarified his statement to say that Williams is “studying such a project but no decision has been made.” 

Regardless how all this plays out, it certainly sounds like serious consideration is being given to yet another route for NGLs to exit the region.  How does that jibe with current forecasts for NGL/ethane outbound flows?  Well, if a new pipeline out of the region is being considered, it‘s because either (a) the new alternative is expected to significantly out-compete ATEX in the market (unlikely since ATEX has firm throughput commitments from Chesapeake and Range that grow from 65 Mb/d in 2014 to 105 Mb/d in 2017), or (b) A LOT more liquids are expected to be produced in this region than most forecasts are predicting.  Contemplate that concept for a couple of minutes before we get into the nitty-gritty. 

NGL Production Outlook

Total NGL production out of the Northeast today (all from EIA Appalachian No. 1 Region) is 40 Mb/d.  That’s it.  Half of that total is propane, the other half butanes and natural gasoline.  EIA reports a miniscule 1 Mb/d of ethane which is probably being moved by railcar as a mixed stream.  Compare 40 Mb/d to total U.S. gas plant production of NGLs – 2,300 Mb/d.  Today the entire Northeast makes up less than 2% of the nation’s total NGL production. 

All that is about to change.  According to Bentek numbers, wet gas production in the Marcellus has grown from 0.3 Bcf/d in 2010 to 1.3 Bcf/d today.  That level of production will more than double in five years to 2.7 Bcf/d.  And then there is the Utica – the broader shale formation that lies underneath the Marcellus.  Utica data is pretty sparse so far, but Bentek has production getting to 1.6 Bcf/d in five years based on current evidence.  That’s 4.3 Bcf/d of wet gas production where the liquids content is quite high - call it 4 to 9 gallons per Mcf (GPM) (we explained GPM in Golden Age of Gas Processing - Party on Dudes). 

How do we translate that into an NGL production forecast?  Taking a conservative average GPM of 4.2., with production at 4.3 Bcf * GPM of 4.2, that makes 18 M gallons (times 1,000 = 18,000 gallons).  Divide by 42 gallons per barrel = 18000/42 = 430 Mb/d.  I.E. an NGL production forecast of 430 Mb/d.  Marcellus NGLs are rich in ethane, so at least half that forecast number is ethane.    Let’s call it exactly half, or 215 Mb/d ethane. 

Processing and Fractionation

If 4.3 Bcf of wet gas production is expected from the region in the next five years then midstreamers will need to process that amount.  Looking at the projects announced so far, that won’t be a problem.  Eighteen new or expanded processing plants in Pennsylvania and West Virginia will crank up natural gas processing capacity from about 1.7 Bcf/d today to more than 5.0 Bcf/d in the next 18 months.  Half the new capacity is being developed by Mark West and the rest is split between Caiman, Dominion, NiSource and the joint venture plant announced in March between Chesapeake Midstream (new name coming soon)/M3 Midstream LLC (Momentum) and EV Energy Partners.. According to Bentek’s NGL Facilities Databank, all 18 projects are scheduled to be completed by year-end 2013.

It is a similar story for fractionation.  Eight new fractionation projects will increase capacity from about 120 Mb/d today to 370 Mb/d, with all but 40 Mb/d scheduled for completion in 2012 and 2013 (also from Bentek’s NGL Facilities Databank).  That is a little short of the 430 Mb/d of total NGL production we calculated above.  But not to worry.  Before the pixels were dry on my presentation’s powerpoint slides, NiSource announced a joint venture with Hilcorp to build a new processing plant in Ohio, and the partners are considering the development of their own fractionation facility.  Then Kinder Morgan announced at the Infocast Summit that they will be building a 300 MMcf/d gas processing plant and a 48 Mb/d (expandable to 192 Mb/d) fractionation facility in Tuscarawas County, Ohio.  They keep coming fast and furious.  Neither of these projects are in the capacity numbers above.

Transportation

With all of these liquids being processed and fractionated, the resulting ethane output cannot be simply blended off into the residue gas stream, due to BTU restrictions on downstream natural gas pipelines.  Two years ago at least seven projects were being promoted to handle the ethane excess, including El Paso MEPS, Kinder Morgan Cochin, Buckeye Union, MarkWest/Sunoco Mariner East and the Williams Confluence project.  Only two of these projects are left standing today:

• Mariner West, the MarkWest/ Sunoco project to convert 350 miles of an existing refined products line to transport ethane from the MarkWest Houston Complex to Sarnia, Ontario and the Nova petrochemical plant there.  The pipe will have an initial capacity of 50 Mb/d, but will be expandable to 65 Mb/d.  The project is expected to be operational in the next few months.
• Enterprise Appalachia to Texas (ATEX) Express pipeline, a three part 1,230-mile system.   The first part is a 370 mile new build from Pennsylvania to Indiana. The second part  reverses one of the pipelines in the Enterprise TEPPCO system – a 16” pipeline that runs from Indiana to Beaumont.  The third part runs the 55 miles from Beaumont to Mont Belvieu. The ATEX pipeline will initially have 190 Mb/d capcity and is expected to be in service in the first quarter of 2014.

If we add the initial ATEX volume (190) to the initial Mariner West volume (50) that gets us to 240 Mb/d.  That sounds like more than enough capacity to handle our ethane production forecast of 215 Mb/d, with room to grow.

So why might Williams be contemplating an additional new 150-200 Mb/d y-grade pipeline from the Marcellus to the Gulf Coast?  Perhaps the price paid for three year old Caiman is a clue?  That would be $2.5 Billion.  That number would seem to place a big value on the upside of Marcellus midstream.  Should we consider the pipeline “prenouncement” and the rich price paid for Caiman as evidence that our estimates of Marcellus and Utica wet gas production (and the GPM assumptions that we calculated above) are vastly understated?   If that is true then our estimates of too much ethane pipeline capacity are about to flip – and considerably more capacity will be needed in the very near future.  Stay tuned, because what happens here might signal that we’re only beginning to understand the scope of the combined Marcellus/Utica play.

Thus far we’ve focused on ethane.  But ethane is not the only NGL with the potential for capacity constraints out of the Marcellus/Utica.  Propane is looking pretty long too.  Remember, the hypothetical Williams pipe is supposed to move y-grade, not ethane-only.  That is a solution that addresses both ethane and propane – a good thing, and it has a variety of interesting implications for Marcellus/Utica processors and producers.   In the next installment of this series we’ll look at the propane situation, and build on the scenario that we laid out in Changes in Latitude - Changes in % of Crude: Propane will not be the same.