The Marcellus Shale Production Update

The PA DEP finally released Marcellus shale production data for second half of 2012. It shows significant production growth from the first half to second half. So I want to study the data closer to get an idea what's going on.

Looking at the data I note the following summary:

• In H1-2012, there were 2835 wells produced a total of 895.555 BCF of shale gas in 442194 well production days. That figures to an averages of 4.92 BCF/day total production, and an average of 2.0246 MMCF/day per well.
• In H2-2012, there were 3551 wells produced a total of 1146.8 BCF of shale gas in 558212 well production days. The average is 6.233 BCF/day total production, 2054.4 MCF/day per well.

There are two surprises to me. One is the data suggests that there seems to be 716 new wells added in H2-2012, averaging 3.9 new wells per day. I knew from my previous study that 4 wells per day in average is needed to maintain flat Marcellus production, and the well spud rate dropped to roughly 2.5 new wells per day starting in July 2012. Obviously the drop of spud rate has not translated into the drop of well completion rate so far yet. It will take a bit more time for the slow down of well drilling to lead to slow down in well completion.

Another surprise to me is that the data seems to suggest that production from existing wells some how are not dropping as fast as I believe they should. Thus I looked at the data closer.

I screwinized the production data and removed a few wells that were producing in H1-2012 but not in H2-2012, as they may distort the data. The remaining wells are in two categories:

• Existing wells that produced in both H1 and H2 of 2012. There are 2823 of them. They produced 892.7364 BCF in H1 in 436366 well production days. They averaged 4.905 BCF/day in total production and 2045.843 MCF/day per well. The average days of production is 154.6 days in the 182 days H1 period.
• These same 2823 existing wells produced 881.8125 BCF in H2 in 490046 well production days. That averages 4.79 BCF/day in total production, and 1799.45 MCF/day per well. The average days of production is 173.6 days, in the 184 days H2 period.
• There are 727 new wells in H2-2012. They produced 264.9895 BCF in 68165 well production days, averaging 1.44 BCF/day in total production and 3887.5 MCF/day per well. The average days of production of these wells is 93.8 days in H2-2012.

The data suggests that productivity of existing wells declined from 2045.843 MMCF/day to 1799.45 MCF/day in 182 days. The average decline is losing -0.07%/day. That's a far slower decline than my previous rule of thumb of losing -0.2%/day. What's going on? I will come back to look at this decline closer.

If existing production rate of 5 BCF/day is losing 0.07% per day. That's losing 3.5 MMCF/day. Since each new well averaged 3.8875 MMCF/day, the data suggests that 0.9 new wells per day is the threshold needed to maintain flat production. My previous estimate was 4 new wells per day.

Why the existing wells are declining at only -0.07%/day?

I examined the individual wells closer.

I find that MOST of wells indeed decline much faster than that. For example well 005-30503 declined 25% from H1 to H2. Well 015-20258 declined 43%!!! Well 015-20262 declined 40%!!!

Nowever there are SOME exceptionally good wells from the seeming sweet spots. These wells have extremely high production rates, and they don't seem to decline much so far.

For example well 015-20444 and 015-20488. The well 015-20488 produced at 7.848 MMCF/day in H1 and 7.554 MMCF/day in H2. That's hardly a decline of -3.75% in half a year. That well, owned by CHK, has produced 6.64 BCF so far. It has no sign of slowing down any time soon. That is really an amazingly high production well.

Are these exceptionally good sweet spot wells the exception, or the norm? I do not know. I have to scrutinize the data further.

But for H1 to H2 of 2012, lack of decline of existing wells supported the total production and pushed it higher. This trend may continue for Marcellus for a little while longer.

But at least by now I know the EIA data on Marcellus production is off. According to EIA, who ontained the data from Bentek, the Marcellus production averaged 6.873 BCF in H2-2012.

The PA DEP data, as discussed above, indicated an average production of 6.233 BCF/day in H2-2012. So the EIA data was off by 0.64 BCF.day. That is a pretty significant discrepancy.

[UPDATE]

I noticed one curious and interesting FACT.




When looking at the wells individually, most of wells decline at rates where I expect them to be, losing 25% to 30% in half a year.



However for those few wells that are exceptionally highly production, and also declines at incredibly low rate, there is NO EXCEPTION. Each and every one of such wells belongs to CHK. I have not find one such high productive and low decline well belong to another company.

Here are just some of the well numbers

131-20062
115-20341
115-20271
015-20482
115-20458
115-20279
115-20257
115-20393
115-20325
115-20324
115-20256
115-20269
115-20244
115-20303
115-20243
015-20673
131-20058
015-20471
015-21224
015-20242
015-20445
015-20488
015-20600
015-20458
115-20268
015-20457
115-20296
115-20351
115-20272
015-20866
015-20452
015-21656
015-20403
015-21655
015-20991
115-20212
015-21609
015-20646
015-20423
115-20424


Very interesting.


Is CHK, Chesapeake Energy, pulling a DATA SCAM on the records?


I don't know how they are going to explain who they have such high production wells that simply will not decline like most shale wells do.