(25 July 2010, firstname.lastname@example.org).
There has been so much information (or mis-information) on the disaster it is difficult to separate the facts from the myths, let alone decide who is or are to be held responsible for the oil spill disaster. There is a need for a working geological model to integrate all the scattered pieces of information and evidence together, so that law makers can zoom into areas where data had been lacking (or withheld) and the wrongs be corrected in order for the industry to move forward. The fact that so many wells (even in deeper waters) had been drilled successfully in the past in the same Gulf region suggests that there may be more “hidden” factors that caused this blowout to be so disastrous.
The geological model presented here is based on facts derived from past blowout investigations that had been equally puzzling. It provides a fresh perspective into the blowout investigation which until now had been overly focused on the drilling itself. If the well blowout was already a disaster in waiting, there is absolutely nothing the drilling crew could do to prevent the blowout, short of abandoning the well prior to reaching the reservoir. The fact that this geological model had been independently generalized from data and information available on the public domain means that there is room for more detailed infill and ample opportunities for BP’s technical experts to prove the model wrong. On the other hand, if subsequent revelations (from yet to be published data or information) substantiate or improve on the accuracy of the model, then this geological modeling effort, is heading the right direction in providing a more sound basis for corrective measures towards making the oil industry safer from such future disasters.
1 Key components of the qualitative geological model.
It is reasonable to assume that BP was targeting a structural reservoir in the vicinity of a salt dome. In BP's bathymetric chart, both Macondo’s wells (A & B) were located on an escarpment discernible on satellite images of the seafloor obtained from Google Earth. Texaco Rigel well which is about 2.43 km from BP Macondo A, is about 1 km away from the edge of the escarpment. Thus, while a salt dome is selected for the model, any vertical geological structure like an intrusive dyke or a vertically inclined fault zone (lateral fault), would essentially produce the same effects. The present qualitative geological model can be converted to a quantitative one when sufficient quantitative data is available. For now this qualitative model is sufficient for us to understand how the blowout occurred, why it occurred, what should have been done to remedy a bad situation from getting worse and how it could have been prevented in the future.
2 Information substantiating the qualitative geological model
There have been “unconfirmed” reports that Macondo Well A which was first drilled by TransOcean Marianas and aborted on 9th Nov 2009 after reaching a depth of 4023 feet (1226 m) below seabed, was re-entered by TransOcean Deepwater Horizon on 13 or 15 Feb 2010. Thus the present blown out well is Macondo B. There were also unconfirmed reports that Macondo B was so badly blown, that the well which is been shown to the worldwide audience is the first Macondo A well which blew earlier in early March (??), before the 20 April blowout. While such “unconfirmed” information would fit in quite nicely with the geological model, it does not affect its validity even if they are not true.
On 13 Feb BP told MMS they were trying to seal cracks in the well. It took 10 days to plug the first cracks. In early March , BP told MMS they were having trouble maintaining control of surging natural gas (according to emails).
A March 10 e-mail to Frank Patton, the U.S. Minerals Management Service’s drilling engineer for the New Orleans district, from BP executive Scherie Douglas said BP planned to sever the pipe connecting the well to the rig and plug the hole. “We are in the midst of a well control situation on MC 252 #001 and have stuck pipe,” Douglas wrote, referring to the subsea block, Mississippi Canyon 252, of the stricken well. “We are bringing out equipment to begin operations to sever the drillpipe, plugback the well and bypass.” Bloomberg News (31 May 2010).
According to Bloomberg news, Douglas or BP received verbal approval at 11pm on 11 March to insert the cement plug about 750feet (229m) above the bottom of the hole. The Federal regulators gave BP permission to cement the well at a shallower depth than normally would have been required after the hole caved in on drilling equipment.
In the congressional hearing on 15 June 2010, BP Chief Executive Officer Tony Hayward and other top executives gave the impression they were ignorant of the difficulties the company’s engineers were grappling with in the well before the explosion… according to U.S. Representative Henry Waxman, chairman of the House Energy and Commerce Committee. “We could find no evidence that you paid any attention to the tremendous risk BP was taking,” Waxman said as Hayward waited to testify. “There is not a single email or document that you paid the slightest attention to the dangers at this well.”
BP Chief Operating Officer Doug Suttles and exploration chief Andy Inglis “were apparently oblivious to what was happening,” said Waxman, a California Democrat. “BP’s corporate complacency is astonishing.”
Perhaps Henry Waxman was not aware that there was a massive share sell-off (531,461 shares in total) by 4 BP directors just days after the 11 March incident. Tony Hayward sold 223,288 shares (a third of his total holding) on 17 March. This was followed by Byron E Grote on 18 March (58,536 shares), Andy Iglis on 23 March (219,500 shares) and Ian C Conn on 30 March (13,073 shares). And that were only BP’s directors. What about the shares sell off by BP’s executives? See Massive Shares sell off prior to expected disaster.
It is not that BP directors and executives were ignorant to the problems on the Macondo wells. Their personal fortune mattered more. It is not that they do not know a blowout was inevitable. They were only wrong in thinking that the blowout could be controlled. They had not expected the blowout to spin so badly out of control.
It did not matter whether Macondo A or Macondo B was eventually drilled to reservoir level since both wells were located right on top of the seabed escarpment which is clearly an indication of some massive geological structure beneath.
Texaco’s Rigel well 2 km from BP’s Macondo wells (but 1 km from the edge of the escarpment), was drilled safely in stark contrast to BP’s ill fated wells. Why? The reason is obvious on Figure 1a.
The Rigel exploration well, the Texaco OCS-G-18207 #1, was drilled in 1999 in Gulf of Mexico block MC 252 in 5200’ water depth. The well targeted a Miocene age, low-relief downthrown closure/stratigraphic trap that was supported by a strong amplitude response on the 3D seismic data. The results from the Rigel exploration well were disappointing. The well encountered what was interpreted to be a 176’ thick gas-charged, low-permeability siltstone in the Rob E-age target. This reservoir was believed to be uneconomic at that time. This presentation focuses on a few stalwart individuals’ efforts to continue to pursue appraisal of this marginal discovery. These efforts included pre-appraisal geologic modeling, reservoir modeling, and analog work. (Westside - Rigel Deepwater Field Appraisal and Development 16 Nov 2005.)
3 What possibly happened?
Figures 1a shows the geological setting just prior to drilling BP’s Macondo well. Problems started as soon as the drilling entered the GWSF hazardous zone. The top hole condition would have deteriorated as escaping gas swirled outside the well casing, enlarging the well bore. With heavy circulation losses, the drillers would have reduced ECD (equivalent circulation density) to limit mud losses and minimize damage to the pervious (weak) rock formation. Unfortunately, each time the ECD dipped below the previous charged pressure, gas influx would kick in. Thus the drillers would have no choice but to keep ECD high enough to keep the gas out. Cementation to isolate the hydraulic connection between layers would be futile as the cement would not remain static long enough to set. This was partly due to pressurized gas and cavitations in the GWSF zone caused earlier, by drilling in an open hole. The dynamic movement of fluids in the GWSF zone gradually increased the fractures and permeability in the vicinity of the poorly cemented well bore as the drilling continued deeper.
The presence of gas-saturated weak rock formation immediately underlying the non-lithified sediment is a slow acting hazardous condition (GWSF hazards) not readily recognized or understood by the industry despite being the common factor in most blowouts. Although GWSF hazardous conditions do not immediately caused a blowout, the seeds of destruction are sown at this shallow sub-formation depth. The deterioration of the well bore outside the casing and damage to the rock sub-formation is beyond the control of any drillers. Pumping in cement to seal the cracks would not work under gas-charged conditions.
The drilling problems were further compounded when up-dipping beds were encountered with sudden loss of circulation. To cut mud loss, ECD had to be reduced. But when pressure in the well dipped, gas influx kicked in as the Extended Gas Charged Pressure (EGCP) zone had previously been charged to a higher mud weight. See illustrations in figures 1d &1e.
The permeable contact aureole of the salt dome or an intrusive dyke, obviously added to the problem. It is like having a “U-tube” counterbalancing the mud column inside the well. No wonder the drillers described the Macondo well as a “Hell Well”. Compare this nightmare scenario with the Texaco Rigel well which was drilled safely just a km away from the salt dome. BP’s management should have correlated the drilling problems with the geological structure. If they had done that (which is the gist of this article), they would have realised that the Macondo well was just a disaster waiting to happen. They should have taken the responsible way out by abandoning the well before reaching the reservoir.
By failing to do that, they were just postponing the inevitable. The “giant aquifer system” was fully charged and just waiting for any mistake to trigger the blowout. No wonder the directors and top executives were rushing to sell off their shares after the 11 March incident, in anticipation of the worse to come. Perhaps BP should stand for “Before Public-interest” for the blatant manner in which personal profits come before the welfare of the environment and public.
As soon as the pressure in the well dipped below the EGCP (replacing the drilling mud with seawater) gas influx kicked in, at the leaking points in the well from pressurised gas stored at the GSWF zone. When the gas bubble in the well started to rise and expand with lower pressure, it rapidly displaced the seawater column (>5,000 ft) in the riser. This is like sucking liquid out of a glass with a straw. The tremendous suction and static pressure exerted by the reservoir created a sudden jump in differential force, resulting in the breach of the bottom cement plug 2 days later. This triggered the uncontrollable continuous gushing of oil and gas out of the reservoir through the blown well. See figure 1f.
The futile attempts to “Top Kill” or “Top Cap” the gushing well only made the bad situation worse by increasing the damage to GWSF zone and increasing the EGCP size. See previous article; The high risk of top capping the gushing well.
After quickly reaching 6,400 psi in the pressure test using the TOP CAP, the increase in the well pressure slowed down to 10, then 2 to less than 1 psi per hour. Oil and gas are obviously being forced into the “giant aquifer” which kept expanding and finding new pathways in the rock formation. That is why the initial 8,000 to 9,000 psi passing mark would never be reached. After 41 hours, the pressure inside the top capped well was 6,745 psi and still rising very slowly. Of course, the pressure inside the capped well would never decrease (until the reservoir is depleted) even as oil and gas are being forced further into the EGCP zone and into the giant aquifer.
As only the light hydrocarbons (methane) filter or seep through the Quaternary Sediment layers, no oil seeps would be evident at the sea floor yet. The oil would remain buried beneath the sea floor until weaknesses in the sediment developed into cracks big enough to result in active oil seeps (which would also mean a near calamity). By then the hot oil and gases from the reservoir may have tilted the world into an irreversible ecological disaster, by warming up and vaporising strata of methane hydrates into gas. The result would be an exponential increase in dissolved methane in the deep waters of the Gulf and eventually into our atmosphere. No one knows how much methane hydrates lay beneath the Gulf sea floor.
But one thing is for sure. The longer the gushing well stays “top capped”, the more severe is the environmental damage. There is no logical reason why the gushing oil could not be tapped through the LMRP TOP CAP with a floating platform or subsea facilities; rather shutting it off completely to cause further damage to the fragile sub-seabed structure and sediment.
4 What you don’t see can be covered up.
Perhaps the botched-up “photochop-chop” photos put up by BP was just a test. To see how keen the public eyes were in following BP’s clean up efforts. It would be hard to believe BP paid professionals for such a shoddy job. We should give BP more credit than that (remember the shares issues)? Let’s play dumb and the problems will go away.
Many experts in the oil industry were surprised and questioned the rationality of capping the well when the relief wells were so close to achieving their “bottom kill” objectives. They could have installed the TOP CAP much earlier. This means that BP knew if the gushing well was completely shut at the top, the oil and gas would spread beneath the sea floor and gas seeps would start appearing. So the TOP CAP had to be placed just before the relief well was ready for the “magic show”. Hurricane Bonnie spoilt the show and the delay is already showing signs of stress (gas seeps).
This could also mean that BP was getting less and less confident that the relief wells would work. The relief wells were held up as the last Trump card. If it fails in full (ROV) view of the concerned public throughout the world, BP’s shares would drop like a stone. There are good geological reasons why the chances of the relief wells’ success are less than 30%. But that would be in the next posting.
So instead “of going on a public stage with a final trump card of 30% chance of success” and risking everything BP stands for, a magic show will be set up so that what ever happens, it will be a success. How?
With a gushing well in full view, a successful bottom kill would show oil slowing down to eventually a tickle. With the cap on, it would be easier to manipulate the data. Thus botched-up photos were a test to check the keenness of the public eye. If the bottom kill fails, there is no independent monitor to prove it. BP could quickly pack and leave the site. Without ROVs’ video, the world is blind. Independent scientific researches later on could be disputed or controlled in post-recovery mopped up battle plan.
The TOP CAP had to be installed and the integrity pressure tests used as an excuse to completely shut down the flow. There is no need to prove the well is leaking. It is already a fact. David Copperfield could not have performed better.
For complete appendix to article see Diagrammatic Illustration of blowout