Any financial advisor worth their salt will caution you that past performance is not a guarantee of the same in the future. And so it is with US shale oil. Recent EIA estimates claim potential production of up to 10 million barrels per day (BPD). That projection is pure fiction and in this article I will explain why.
Right off the bat, assuming annual declines of 2.0 Million BPD, wells having a 365-day average annual rate of 400 BPD requiring 5,000 wells and utilizing 330 rigs (with each well drilling 15 years per annum) would be needed just to offset declines. Should annual declines exceed 20% per new and vintage production, the number of rigs goes up proportionally.
Productivity of shale basins (a misnomer as the dominant facies consist of mixtures of sandstone, dolomite, limestone, siltstone, and clays) is determined by a number factors.
The first is the degree of catagenesis (i.e. the degree of cooking of the kerogens in the source rock). This determines the type of product at a given reservoir depth. Increasing depth usually grades from oil to wet gas, and then dry gas. The wet gas and dry gas regions have increasing pressures with dry gas reservoirs having pressure gradients up to of 1.6 times normal pressure. Pressure gradients in the wet gas window range form slightly over-pressured to 1.4 times normal pressure. The second is the rock facies and brittleness of the rock. Areas with high clay content are not frackable. Finally, porosity, permeability and product viscosity determine producibility. Wet gas regions are far and way the most profitable as they are capable of prolific production. They produce at higher values per BOE than dry gas or oil. Simply put, they have the best netbacks per BOE. This is the reason that wet gas areas see the most activity no matter the basin, be it Permian, Eagleford in the US or the Montney in Canada.
As we know, most plays (except the Bakken which is primarily an oil play) contain a mixture of all three products depending on depth. The Eagleford comprises of 22,700 square miles in twenty-three counties with 58%, 23% and 16% respectfully in the oil, wet gas, and dry gas regions. Twenty five percent of all Eagleford production is produced from only three counties; Karnes, Lasalle and Gonzales. To put this into perspective, Karnes county with only 754 square miles (3.3 % of the entire Eagleford foot print) produces 146,000 BPD. The total area in the wet gas window is in these three counties is about 1,400 square miles or 6% of the total foot print of the Eagleford. Both Gonzales (which adjacent to Karnes) and Lasalle counties are on depth trend with Karnes county.
At 75,000 square miles, the Permian is roughly three and a half times as large as the Eagleford. It has two main producing horizons; the Sprayberry (600 feet thick in several intervals) and the Wolfcamp (three hundred feet thick). The Permian has not been mapped in terms of product type, although we can assume that the same ratios apply, thus there could possibly be up two 15-20,000 square miles in the wet gas window. The caveat however is that 23% of all Eagleford oil production is produced from only 9% of the Eagleford footprint.
Much talk has been given towards the Permian and the plethora of intervals in each well. The geology of the Permian Basin is rich and complex, both horizontally and vertically. The basin has commercial accumulations of oil and gas in stacked layers, at depths ranging from 1,000 feet to more than 25,000 feet. At this point mapping of each product type (as is the case in the Eagleford) in the Permian has not been completed to my knowledge. A further complicating factor is the basement heat gradient, which is simply the temperature rise per unit of depth. In the case of the Montney, the producers who figured this out have the best (by far) land base, purchased at the lowest costs, before the herd mentality started. Should the basement heat gradient vary across the basin each product type could occur at different depths. Because of its large foot print, it is highly probable that the basement heat gradient varies across the Permian as it does in both the Eagleford and Montney.
Looking at the stacked intervals in the Permian, tracer logs that I am aware of, show that frack heights vary between 75 and 150 feet. This means that up to three wells could be needed to tap all intervals assuming a total thickness of 600 feet. A complicating factor arises if all the intervals do not have the same facies, product type and or are thick enough to be economic.
The EIA has estimated Ultimate Economical Recoverable Reserves of 11 to 33 Billion barrel in the Permian. Assuming an average type well across the basin has reserves of 600,000 barrels of oil per well, a total of 18,300 wells will be needed to recover 11 billion barrels. Assuming each rig drills only one interval, a total of 4,600 square miles will be consumed. Likewise, in the P50 case of 22 billion barrels, 9,200 square miles will be consumed. Based on the example of the Eagleford where production is centred on a few counties, the areal extent is likely between 4600 (6% of 75,000) and 9,200 (12.3%) square miles.
So, how high can the Permian go? Assuming net annual growth of 500,000 BPD with 375 rigs (5625 wells per year), the play would be exhausted in 3.25 years with production reaching 4 million barrels per day. This assumption is not plausible as the number of rigs will be based on updated play knowledge. Therefore, as the extent of the sweet spots become more defined the member of rigs will slowly decline should the lower case become evident or hold steady for a longer period. And the pace of development of course will depend on the price of oil. It’s likely that the pace of development will adjust as does the price of oil gas.
Weekly production estimates are published in the petroleum inventory report (which has recently been reduced as actuals provided by producers are obtained). Using actuals, the EIA reports have been over estimated by as much as 300,000 barrels per day recently. Month over month production saw June slightly lower than May.
In 2016, the EIA in a report to Congress estimated that by 2022 US shale oil production would reach 6 million barrels assuming a $50 WTI price and 10 million barrels per day assuming a price deck of $80 per barrel.
Estimating future production should be based on a play by play basis. The Bakken was initially thought to have the capacity to reach 2 million BPD. We now know that the Bakken is having difficulties achieving its peak production of 1.3 million BPD. Eagleford production peaked in 2015 at 1.65 million BPD. Many forecasts indicate that at best, production will climb slowly but may not reach the peak production levels. The Permian obviously is the wild card in the equation. Since 2011, with the advent of horizontal drilling, production in the Permian has grown by 46% per year, from 300,000 BPD to 2,300,00 BPD today. Using “Straitlineology” the Permian should be producing over seven million BPD by the end of 2020. This continuance of this level of increase growth is impossible to determine. But at some time the production will peak and then start a terminal decline as productivity slowly falls. Hint: Producers always drill the best first.
My best guess for peak shale oil production is 5.8 million BPD coinciding with peak Permian production of 3.5 Million BPD in 2020. During this timeframe, the Bakken should produce at a rate 1.0 million BPD, and the Eagleford at 1.3 million BPD. Factors that could derail this estimate are fewer wet gas areas in the Permian, lack of success in different intervals, increased service costs and of course lower than expected production from the Bakken and Eagleford.
On the other hand, should producers “discover” new wet gas areas in the Permian production could rise as high as 4.5 million BPD. This would raise total shale production to 6.8 million BPD. With higher prices (WTI at $50/barrel), the pace of development will increase and so will production rates. But as happened in 2014, price declines lead to a drop in activity and therefore production.
There is almost zero chance of US shale oil production of ten million BPD. The most likely range is between 5.8 and 6.8 million BPD if the Permian lives up to its initial promise. A further complication is that unconventional plays are not consistent through the full footprint of the basin. Only a fraction (10 to 15 % or less) should be considered as prime wet gas areas. The winners in this game will be those companies who have figured out the geology first and hold the best land at first mover prices. The losers will be those who believe that shale plays are homogenous throughout the basins and move into the plays later and pay to much for land that is not always the best.