Macondo – Gas Hydrates, Shallow Water Flows and Burst Disks

Ignore the geology at your peril. As news comes through of probable seabed seepages of oil being observed again at Macondo, the only question will be whether or not these are reactivated “residual” seeps or the nightmare scenario of new flow paths through the shallow strata. Whichever it is, the latest events are clear evidence that the Macondo well casing was breached, a belief held by several commentators in the past.

Macondo MC 252 is in an area of well known severe geological hazards, including gas hydrates and weak shallow (< 5,000 ft) artesian pressurised sand layers which can cause casing distortion and even buckling collapse, high inflow of drilling fluids, washout and partial collapse of those layers due to rapid drilling disturbance.

This “geohazard” is known as “shallow water flow” [SWF] and the associated risks have been extensively reported in the technical literature, and were ranked and mapped by the MMS. [Mississippi Canyon SWFs]. They were well known by individuals within BP America. In addition, the potential presence of extensive layers of in-situ frozen gas hydrates [GH] along the Mississippi Canyon continental shelf edge is understood and documented. Furthermore, this well location was selected and drilled very close to a gas hydrate bearing mud volcano feature in order to maximise the potential for a positive result at Target Depth. A safer method, known as “Riserless Drilling” does not appear to have been considered and was probably seen as too costly.

Despite all the above, BP took the decision to continue to attempt to drill the one of the world’s deepest wells to date in a slapdash, corner-cutting, driving down cost, “maximise added” value fashion. These cut corners and lack of management control have been well documented to date.

With proper drilling design, control and cementing, wells have been completed under these conditions in the past. However the risks should always be thoroughly assessed, with appropriate prevention and mitigation techniques and controls in place. Such measures are also well documented based upon the past experiences of GoM operators, including BP. However, as well as designing a high risk “long string” well, the MC252 Exploration Well was designed with safety valves (known as “Burst Disks”) at shallow depths which were close to the levels of three known and mapped SWF sands, and in a GH prone area. This was part of the plan to convert the exploration well to production at a later stage. More savings.

Indications are that the drillstring at Macondo was fractured/damaged as a result of SWF drilling fluid fracture, washout and collapse over a certain depth interval. Severe problems and drilling mud losses were recorded at the time. The expanded pressurised melted hydrate gas may have blown through the damaged casing and weak pathways in the bad cement job following the final negative leak off test. Drilling mud was displaced by seawater within the drillstring over too great a depth, leading to reduced internal hydrostatic pressure and a sudden imbalance between the internal (fluid) and external (expanding gas in sand) pressures. The hydrate almost certainly would have been steadily melting around the casing due to heat given off by the curing cement, a problem well understood by Haliburton, explaining their concerns over the use of a nitrogen foam cement.

The recent very detailed DNV forensic report on the BOP failure as well as an internal highly detailed Transocean Report leads to the conclusion that any BOP would have failed even if it had been in perfect condition, due to the condition of the well, it’s out of vertical alignment and the sudden immense force of the gas and fluid flow.

The question that should have been asked to date is: where did that huge quantity of gas really come from? Calculations may show that the valve system at the bottom of the well is unlikely to have “somehow” failed as a result of pressure changes far up the drillstring, causing a sudden influx of a vast quantity of gas from the hydrocarbon reservoir some two and half miles below seabed to burst upwards through fairly dense drilling mud at such a high velocity.

Consideration of the temperature and pressure regime in the shallow hole section below seabed suggests that natural in-situ hydrates are very likely to have been present at the Macondo location over a depth interval of a few hundred metres below mudline. The heat generated as a result of cement curing is likely to have led to melting of section of this natural hydrate bearing zone some distance radially from the cement and a subsequent increase in pressure as the gas tried to expand within one or more of the known SWF sand
layers. The presence of channels at certain levels within the cement is likely to have permitted a pathway(s) to form along part(s) of the 16” casing. Due to possible earlier drilling disturbance of the known layers of SWF sands prior to the melting of the hydrates, the 16” casing may have been out-of-straight or even slightly buckled as a result of partial liquefaction and softening of the SWF sands (similar to that observed at the BP/Shell URSA in 1999 in the GoM). This loss of lateral support may have caused a crack or breach in the casing, or even a loosening or damage at the casing joint(s). This casing is suspected to have been of too low a yield strength for the well design. At the point during the negative leak-off test when the pressure differential became sufficiently high, it is now well documented and accepted that the three “burst disks” (essentially pressure relief valves) placed at certain points on the casing joints down the casing string probably blew out at their inwards blowing rated burst pressure of 1600 psi. At this point the large pressure drop occurring within the mud fluids in the annulus between the production casing and the 16” casing might have been sufficiently large to allow a rapid influx of trapped pressurised gas lying within the SWF sand(s) and in the pathways which had formed in the bad exterior cement job. Melted hydrate expands to approximately 64 times its original frozen volume. This gas build up may have caused a very high pressure jet to blow out one of the rupture disks in the 16” casing, if the pressure differential between the seawater filled production casing and the annulus on the other side were sufficiently high. This immensely powerful gas jet stream would have travelled very quickly up the production casing.

This subsequent upwards rush of gas would have caused a siphoning of seawater, mud and subsequently oil with it as the production casing shoe was blown due to the very high suction force exerted. Once the initial shallow blast of GH sourced gas blew the BOP, reservir pressures would have been sufficient to allow the flow of oil to be maintained.

Much detail has been written in formal reports, books and publications to date about the mechanics of what happened. However virtually nothing has been stated about the root “geohazard” causes – shallow water flows and gas hydrates, a horrible but all too feared combination.


About thegallowglaich

Offshore Engineer and Researcher
This entry was posted in Uncategorized. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s