If offshore hydrocarbon deposits are to be tapped without surface facilities, then operators need a subsea production system that sometimes includes a long-distance subsea tieback. This article outlines some of the problems associated with designing such a system.

Dr. Bill Loth, in his President's Lecture to the Institute of Marine Engineering, Science and Technology, described tieback design problems by first outlining Exxon's prototype subsea production system (SPS) which featured subsea boosting and separation systems plus closed-loop hydraulics, conceived in the early 1970s.

The SPS was designed in the belief that it would be possible to build a system capable of operating in a water depth of 3,000 ft (1,640 m) which could produce up to 100 miles (160 km) back to a shore-based facility - with the objective of a major cost reduction through elimination of surface facilities. The SPS was a step along the way, and Loth later helped design and install the first multiwell-template SPS

Loth's IMarEST paper concentrated on a hypothetical scenario for designing a subsea production system to exploit a wet gas field using a long-distance tieback after indicating dry gas "was almost too easy," and black oil "too hard."

He suggested a long-distance tieback would involve a series of subsystem trade-offs, and with wet gas he said the majority of the problems would involve dealing with a minimally feasible production rate in excess of 300 MMsf/d and liquid handling above 5,000 b/d - much of it water. "It is these liquids, and especially the water, which will cause the majority of problems."

Loth suggested that only 10 years ago, a 18.75-mile (30-km) tieback would have been a headline-grabbing achievement. Today it is run-of-the-mill, as demonstrated by recent tieback distances (see Table 1).

Loth highlighted both water depth and bathymetry as significant design factors for a long-distance production system: Where a deposit is located 112 miles (180 km) offshore it is commonly beyond the transition zone, where depth might extend to 1,140 ft (350 m) to beyond the end of the continental shelf in deep water ranging from 4,920 ft to 6,560 ft (1,500 m to 2,000 m) or more.

Commonly these deepwater reservoirs are relatively shallow. "Reservoirs located between 6,560 ft and 9,840 ft (2,000 m and 3,000 m) beneath the seafloor would not be uncommon."

Complicating the picture is the presence of subsurface gas hydrate deposits - solid gas molecules surrounded by a cage of water molecules. This can cement loose surface layer sediment to several hundred meters thick. Drilling through it is well understood, Loth said, and disassociation of the hydrate - melting - will leave a subsurface soil of unknown strength, which will have implications for the foundation of subsurface structures.

One of the other issues is draining the reservoir, either via natural pressure depletion or artificially with water injection.

"As more gas is produced, the reservoir pressure, the driving mechanism for production, is reduced. It will be obvious that there is an incentive to reduce the pressure against which the reservoir must produce," Loth said.

Simply put, economic oilfield developments depend largely on a calculation of world oil prices and transport costs, while an economic gas project presents a more complex scenario, entailing risks which need to be evaluated. "But the inevitable conclusion of these evaluations is that some flexibility in system design is required to manage both the up and down side of the risk."

Loth suggested a development using surface trees offers the best prospect of ultimate recovery because of the lower cost of well workovers. But the downside is relatively shallow well targets difficult to reach from a single location, plus the threat to a surface structure posed by the melting of gas hydrates in the sub-strata, which can happen in areas where there are a large number of wells in close proximity.

Assuming a long-distance tieback is preferred to a surface facility and dry trees, then Loth suggested a primary issue is the back pressure against which a reservoir has to produce. There are a number of factors which might be adjusted to define this back pressure, including tubing size, the diameter of infield flowlines and the diameter of the export flowline. "In deep water, the inescapable impact of hydrostatic head may dominate, but there is still considerable latitude for adjustment," Loth said.

But the aim is not simply to reduce the back pressure to the lowest possible value since productivity issues in gas fields are more complicated than for oil wells: Loth explained that in gas reservoirs, as reservoir pressure decreases, gas density falls, and therefore friction losses for an equivalent production rate increase. Also, if reservoir pressure falls bellow the dew point, more liquid forms, so additional liquids are produced. This is where export pipeline sizing becomes a key element in a subsea production system.

Sizing

Export pipeline sizing can be based on steady state conditions, projecting total dynamic losses and slugging conditions. Loth advised that export lines should avoid steep inclines where possible as they traverse from deep to shallow water, but this is where most problems are likely. "If the line is too large, velocities will decrease and liquids will fall out and collect until the pressure behind them increases sufficiently to displace them up the slope," he said. But he noted that resulting pressure changes could adversely affect a reservoir along with the necessary separator and slug-catcher at the receiving end. Conversely, if the export line is too small, dynamic losses will adversely impact production.

Then comes the question of varying production rates. Consequently, it is customary to investigate outcomes with a 50% turndown rate - i.e. 50% less than design conditions. Furthermore, high initial flow rates might damage any completion such as gravel packs installed to prevent the influx of sand into a production well.

Pipewall thickness is a further issue for the subsea production system designer, where deepwater pressure can cause the collapse of an empty flowline.

Loth suggested it may be necessary to design an export flowline with varying wall thickness along its length, depending on its location within the subsea production system. However, there is another snag: "Beyond some point, perhaps 1.25 in. [of wall thickness], the number of mills that can produce substantial lengths of this heavy wall pipe becomes limited, as does the number of vessels capable of laying this heavy pipe in deep water."

Loth noted that at the time his paper was prepared, October 2004, that international procurement of steel was restricted due to demand from China.

Thickness

Looking at various pipewall diameters and wall thickness, together with varying flow rates from 150 MMcf/d to 300 MMcf/d, Loth pointed out that while installation and production costs can vary depending on which flow regime is selected, a direct subsea tieback might only provide a partial development solution. Alternatively, other schemes, including floating production, tension leg platforms or spars, may also need to be considered.

He went on: "Subsea processing is slowly evolving to a useful technology, but current applications are small-scale." Equipment on the seafloor at present consists mainly of multi-phase pumps - of little use to wet gas fields, he suggested - plus units for removal and re-injection of produced water to reduce topsides processing requirements.

"Application of this technology might enable some extension of production, but the potential is limited at the current state of technology by several factors. Subsea separation at reasonable pressure may remove free liquids, but it will not remove the several thousands of barrels per million of liquid that could condense as the saturated gas moves towards the shore and pressure decreases while temperature increases."

Additionally, "Some of the hydraulic problems would be diminished, but the major threat of hydrate formation remains."

Pipeline designers are then encumbered by considerations concerning high discharge pressure at the landfall end of a pipeline, and with even relatively low decreases in arrival pressure, "the hydrodynamic losses increase to an extent that the net back pressure on the reservoir decreases very little."

And delivery of a lower-pressure gas from a subsea separator is not as promising, he said. "The final problem is power," Loth suggested. "The state of the art is such that subsea power consumers greater than a few megawatts, and the distribution and switching systems associated with them, have yet to be proven."

Subsea electrical equipment to produce this kind of power for subsea compression has yet to be developed, and therefore Loth said it would be prohibitively expensive. Surface facilities provide a much better development scenario.

A two-phased approach for a wet gas field, with a long distance pipeline in the initial stage of development and then a later surface facility, provides better overall recovery and economics.

"This has the immense advantage of creating an early revenue stream which supports the more costly remote surface facility and, not inconsequentially, provides early returns from the pipeline investment," he said.

But this isn't the final answer either, and all the factors from the reservoir reserves and pressure through to the landfall pipeline discharge pressure have to be taken into account when designing a development solution. And that depends, finally, on selling the production at an economic cost. "If the additional production cannot be sold, and sold in a timely manner, the most elegant engineering will not salvage the development," he noted.