During 2011, 5,000 horizontal wells will be drilled and fractured within domestic shale formations, which will contribute to the steepest growth ever of gas production in the US since the late 1960s. Petroleum reserve additions also will see a remarkable rise due to the tremendous technological progression that has allowed more hydrocarbons to be derived from these source rocks.

If activity continues at this pace, shale gas production could reach a level of 22 Bcf/d within a decade, or close to 40% of the current demand for gas (~66 Bcf/d) in the US. Currently, the US lower 48 produces about 60 Bcf/d of marketed gas and imports the rest mainly via pipeline from Canada (about 6 Bcf/d) and almost 1 Bcf/d from LNG imports.

Shale gas represents about 24% (15 Bcf/d) of domestic production. Tight gas production is currently around 19 Bcf/d, while coalbed methane is a little more than 5 Bcf/d. Onshore and offshore conventional production is expected to taper off.

By 2020, shale gas will contribute more than any other source to total US production. (Source, Hart Energy

Determining decline

Shale gas wells decline rapidly after initial high production flow rates. This has two implications for shale gas development – it takes several years to ramp up production from a shale gas play because so many wells must be drilled, and it is necessary to drill hundreds of wells each year to maintain production.

Observations of thousands of shale gas wells drilled to date indicate the wells exhibit similar decline trends following a power law decline function. If well data can be fitted to a power law function, it is possible to determine EUR for the well.

The power law decline function is as follows:

q = qi * exp(-Doo*t – Di tn) [in volume per unit time, usually days]

where:

qi = rate “intercept” (i.e. q at t = 0)

Di = decline constant

n = time “exponent” defined by D = Doo + D1 * t-(1-n)

Doo = decline constant at infinite time

Barnett wells indicate higher time exponents and lower decline constants than horizontal wells. (Source: Texas Railroad Commission, Hart Energy Global Shale Gas Study)

The process of fitting well production data to the power law function requires estimating the parameters n, Di and qi. Hart Energy has analyzed numerous wells in the Barnett shale and the Haynesville shale, and oil wells in the Bakken shale using the power law decline function. This method has been found to accurately match actual well production data in all cases except where production is erratic due to well problems.

The power law parameters, n and Di, do not vary much within the same shale play for the same type of well (horizontal or vertical). The Barnett horizontal wells tend to exhibit time exponents, n, between 0.20 and 0.25 and decline constants, Di, between -0.80 and -0.85. Vertical Barnett wells indicate higher time exponents and lower decline constants than horizontal wells. Haynesville wells, meanwhile, exhibit on average higher initial production (IP) rates and slightly higher time exponents than Barnett wells. Bakken oil wells have lower time exponents than the Haynesville with similar Di values.

The power law equation parameter that does vary significantly is qi, the instantaneous rate at time zero. This is a function of the IP rate. Wells with higher IP rates have higher rates over time as long as the other power law parameters are similar. A key conclusion from the power law decline analysis is that the ultimate recovery of an individual shale well is highly dependent on IP rate.

On average, Haynesville wells exhibit higher IP rates (Source: Louisiana Department of Natural Resources, Hart Energy Global Shale Gas Study)

Fracing

Hydraulic fracturing technology continues to improve as operators learn from experience which techniques work best within a given play. This has resulted in higher IP rates and in many cases lower decline rates. The power law parameters may ”improve” over time within a play, but the function still fits production data and provides an accurate means of estimating ultimate well recovery.

During the fracing process, water pressure induces shear-slip, which are micro-seismic events that generally have magnitudes of less than -1.5 on the Richter scale ? about as much energy as is released by a gallon of water dropped from a height of 1.5 m (5 ft) to the floor. It is possible to determine by listening at the surface and in neighboring wells how far, how extensively, and in what directions the shale has cracked from the induced pressure.

Microseismic can be used to monitor where the hydraulic fractures propagate, helping to define the success and orientation of the fractures created. This information provides additional insight into the characteristics of the formation and informs decisions on future wells, such as strategic placement and better fracture designs. As more formation-specific data are gathered, service companies and operators can optimize fracture patterns.

By 2020, tight oil production in the US is expected to be comparable in volume to conventional onshore and offshore production. (Source, Hart Energy)

Competitive pricing

The continuous increasing cost competitiveness of shale gas drilling has brought regional gas prices down to one-quarter of the price of oil measured in dollars per burning equivalent. Companies continued to drill into shale formations in 2009 and 2010 despite the fact that a large share of the new wells had breakeven prices above Henry Hub forward prices.

Average half-cycle breakeven gas prices for major North American shale plays

The rationale for this apparently uneconomic behavior was that hedged prices were still relatively high and that inactive leases would expire. There was also the matter of cost obligations associated with joint ventures within the unconventional arena. The effect has been a further reduction in spot prices as well as price volatility, a trend that has further reinforced a relatively more full underground storage.

As a consequence of a much lower gas price in comparison to oil on a Btu (British thermal unit) basis, companies have shifted focus from dry gas acreage to wet gas areas or pure liquid plays such as the Bakken, the Eagle Ford, and the Niobrara plays.

In the tight granitic sands of the Granite Wash play in the Texas panhandle and Oklahoma, such wells have recently reported up to 10,000 boe/d of initial production, an instantaneous production rate matching steady-state rates at world-class giants like the Troll field off Norway. This hunt for liquids, however, does not necessarily help increase gas prices because two-thirds of the hydrocarbons from these new “liquids” wells is actually associated dry gas, bringing even more unwanted gas to the market. With a potential of supplying 2 MMbbl/d to the US market, the macro effect of the emerging tight oil plays may become significant.