Backflow from the drill pipe is a normal occurrence when a connection is broken at the rig floor. Backflow can result from the density differential of warm (low density) surface water pumped down the pipe against cold (denser) water in the ocean, air that has been trapped during connections and pumped down the pipe, dense cuttings or mud in the annulus flowing back ("U-tubing") to equalize hydrostatic pressure, and so on. Backflow into the pipe is usually reduced by the closure of the downhole float valve but also occurs while retrieving core barrels and through the bit nozzles. Hydrocarbons, hot acidic fluids, H2S, and/or cuttings and debris from the hole may backflow into the pipe and plug the pipe or bit nozzles or jam the downhole float valve open. Backflow will usually gradually decrease within a short time as the pressure differential is equalized.
In deep water, an uncontrolled flow (or "kick") of hydrocarbon gases or fluids exiting from a drilled hole at the seafloor probably would be diluted by mixing with the seawater column and dispersed by currents so that the flow might not be visibly evident on the ship. Fluctuating pump pressures, packing off in the annulus, decreasing string weight, and hole problems may indicate that a kick is in progress. The precision depth recorder (PDR) could be used to look for suspicious plumes in the water column if a gas flow is suspected. The VIT televiewer could be used to check the hole at the seafloor for flow (i.e., an unusual debris cloud or turbidity). If a hydrocarbon kick is suspected, a kill procedure should be started immediately.
A kick up the pipe is most likely to occur when the annulus is packed off, the pipe is open-ended (i.e., no float valve), or the float valve is held open by a core barrel, debris, or malfunction. A kick inside the drill pipe might be differentiated from normal flow-back events because the flow-back rate from the pipe becomes progressively stronger with time.
Note: As the pressure is reduced when gas rises, gas expands in inverse proportion (Boyle's law: P1V1 = P2V2). In the event of heavy and increasing flow from the drill string, circulation should be reestablished as quickly as possible to pump intruding fluid out of the pipe. If the top drive is in use, it should be made back up to the drill string immediately. If the top drive has been racked, it will be faster to install the rig-floor safety valve and close the valve to stop backflow. The top drive or a circulating head can then be used to circulate down the drill string.
It is more difficult to kill a flow if the bottom of the pipe is not below the flow. If the pipe is off-bottom and the ODP Operations Manager, Transocean OIM, and Master agree that an attempt to kill the flow does not pose a risk to the ship and personnel, an attempt may be made to run pipe back in to bottom. If a drill string safety valve has been installed, it may be necessary to install a sub with a Baker model G (5f6R) float valve above the safety valve so the safety valve can be opened at the rig floor. A rig-floor safety sub with a Baker float valve is on the rig floor at all times to act as a check valve, permitting fluid to be pumped down the pipe but preventing backflow on connections. The Baker float valve can be used in instances when the top drive is set back and/or a float valve is not in the string, such as when using a logging bit or after dropping a bit.
Most often, the pipe can be run back down into the good open-hole section using the top drive to fill the pipe (to ensure gas is not moving up the pipe). The drill string should not be forced down into bad hole conditions because stuck-pipe severing operations would not be possible through a drill string float valve. Bad hole conditions probably indicate that the hole is collapsing and the flow will kill itself. The crew should attempt to pump kill mud as deep as possible under good hole conditions.
The record of DSDP/ODP remains unblemished with regard to hydrocarbon pollution from scientific boreholes. That is a tribute to the careful screening procedures of scientific planning and safety panels, adherence to shipboard monitoring procedures, and application of proper abandonment procedures by shipboard personnel. The possibility remains that an uncontrolled flow of gas or petroleum could occur despite all the safety precautions. In case a kick should occur, the Operations Manager must be prepared to take immediate and appropriate action in concert with the Transocean OIM to kill the flow if possible.
The JOIDES Resolution has no riser, recirculating mud system, BOP, or choke and kill lines to control hydrocarbon or water kicks in the normal oilfield manner (i.e., circulating heavy mud through a choke with backpressure). Penetrating a significant hydrocarbon reservoir is unlikely because potential traps for significant hydrocarbon accumulations are strictly avoided. In ODP's scientific operations, open (uncased) holes are cored to relatively shallow penetration depths in soft to semi-indurated sediments in deep water; therefore, the formations could not withstand the pressure of a heavy-mud hydrostatic column.
The objective in killing a flow is to quickly fill the hole with a mud column that has enough hydrostatic pressure to slightly exceed the formation pore pressure. However, the kill-mud weight must not exceed the formation fracture pressure, which would cause the mud to flow laterally, reducing the effective height and hydrostatic pressure of the kill mud column.
It may be prudent to advance the bit on a core-by-core basis if there is an increasing indication of migrated (but not liquid) hydrocarbons. In most circumstances, the detection of migrated and more thermally mature or liquid hydrocarbons requires suspension of drilling operations. Some areas with known gas seeps or dead hydrocarbon stains have been cored successfully using data from offset holes and a series of pilot test holes downdip from the primary site.
Any flow or kick is likely to be from flow along a fault or a flow of the low-pressure and low-volume shallow gas pocket or salt water variety. Without casing for hydrostatic pressure containment, circulating dense (heavy) mud weights exceeding 10.5 ppg (1.26 gm/cm3) might fracture soft sediments.
The fracture gradient at the weakest point in the hole (usually the casing shoe) is the effective limit on the imposition of additional hydrostatic kill pressure. A standard Gulf of Mexico pore pressure/fracture gradient/mud weight graph for riserless drilling can be used to predict formation pore pressures. For example, in 915 m (3000 ft) water depth and 915 mbsf (3000 ft) of penetration, the predicted formation pore pressure is 10.1 ppg (2925 psi). If the hole was loaded with 10.1-ppg kill mud, the formation fracture gradient would be exceeded at ~150 m (500 ft) with normal trip (surge) and circulation pressures. Therefore, 10.1-ppg mud would probably fracture (i.e., break down) the formation and the mud would flow out into the formation at that point (i.e., more or heavier mud would not increase hydrostatic pressure control).
At 1500 mbsf penetration, the pore pressure is 10.5 ppg and the fracture gradient would be exceeded above 450 m (1500 ft). Therefore, overall considerations indicate that a 10.5-ppg kill mud is probably the heaviest practical kill mud for holes with <1500 mbsf penetration under normal circumstances. A volume of heavier kill mud (perhaps 100 bbl of 12.5 ppg) could be placed on bottom (i.e., below 10.5-ppg mud) in deeper holes if fracture gradient conditions permit. Note that cement does not set in the presence of a gas flow; therefore, mud must be used to kill a gas flow before the hole is plugged with cement.
If a kick occurs, an attempt should be made if practical (and safe) to run pipe to total depth and fill the hole with premixed kill mud and/or cement slurry. As in all well-control situations, judgment and rapid response are critical. It is probable that regardless of any attempt at human intervention, the turbulence from flowing fluids during the kick would destabilize the soft sediments in the borehole wall and the hole would load up with debris and/or collapse and reseal itself (which is what happens in natural flow events).
A relatively minor or weak flow of gas or liquid hydrocarbons could seep into the hole from a formation that has been penetrated and go completely undetected for the duration of drilling operations in deep water. A minor flow could manifest itself in unstable hole conditions and "packing off" around the drill string. If a flow is suspected, the PDR could be used to look for suspicious plumes in the water column. it might be possible to run the VIT televiewer to look for gas bubbles or liquids escaping from the hole, which might be detectable as white hotspots on the Mesotech sonar. An attempt should be made to kill such a suspected flow if it appears to be a safe operation.
ODP policy requires that sufficient 10.5-lb/gal kill mud should be premixed and in the reserve pit at all times to completely fill the hole being drilled (usually ~250–350 bbl). If the pipe is open-ended or the downhole float valve is malfunctioning, the drill string safety valve and drill string float valve should be put into the drill string below the top drive before the pipe is run to total depth to displace the kill mud (in case the annulus packs off during pumping operations and flow is diverted up the pipe). While the kill mud is being displaced, preparations should be made to follow it with heavier mud or cement, if required. If the flow can be stopped, the hole should be plugged with cement in accordance with PPSP guidelines.
In the event that a hydrocarbon flow is detected, coring or drilling operations will be terminated immediately. The Operations Manager, Transocean OIM, Staff Scientist, and Master, in dialog with the Co-Chief Scientists, should review the situation and agree on a plan of action. ODP is a self-regulating program with a long history of pollution-free scientific ocean drilling and is committed to maintaining an environmentally sound, pollution-free operation. However, if the Operations Manager, Transocean OIM, or Master feel that a kill attempt is too risky to the ship or personnel, the bit should be pulled above the seafloor and the ship should be moved off location upwind in dynamic positioning (DP) mode before the remainder of the drill string is recovered. On the positive side, environmental damage from shallow gas blowouts is usually limited because the soft sediments in shallow holes tend to collapse and kill the flow after a relatively short time. Activities depend on water depth as follows:
The kill mud should be followed by heavier kill mud (if required to control the flow) and cement to permanently plug the hole. A flowing open hole is often unstable, and the chances of getting the pipe stuck are significant. If the drill string becomes stuck, the normal through-the-drill-string severing procedures might be impossible or too hazardous. In an emergency situation that requires moving the ship immediately away from hydrocarbons, the options would be to intentionally offset or drive-off or drop the drill string. However, the danger to the ship and personnel from a hydrocarbon flow in deep water (with riserless operations) would be small under normal conditions. Hasty actions such as offsetting the ship before the pipe is clear of the seafloor or dropping the drill string might aggravate the situation, endanger personnel, or lead to the unnecessary loss of expensive hardware, if not done properly.
Rapid pipe or tool movements that may swab fluid into the hole or fracture formations should be avoided. If hydrocarbons are detected or anticipated in substantial quantities, drilling will be stopped and the hole plugged.
The hole should be filled with viscous gel barite mud of 10.5 ppg (78 lb/ft3) weight, allowing extra volume for hole enlargement and loss by displacement. The hole should be filled to the uppermost competent layer and a cement plug spotted. A minimum-sized plug should be 200 sacks of 12–15 ppg. Where possible, a plug catcher or calibrated displacement tanks should be used in placing the cement.
If hydrocarbons are indicated and the hole has penetrated semiconsolidated or consolidated rocks, proper placement of cement should be confirmed by probing with the drill string or sampling the cement with the core barrel. The cement plug should be calculated to be at least 15 m and preferably 30 m in length.
The hole should be filled with viscous gel barite mud of 10.5 ppg (78 lb/ft3) weight, allowing extra volume for hole enlargement and loss during displacement.
Holes drilled in consolidated or semiconsolidated rocks on the continental shelf, slope, or rise should be plugged with cement. Holes drilled in unconsolidated sediments in which shows of oil or gas occur should be filled with mud. Holes on the deep ocean floor in which no shows are encountered or holes in igneous rocks may be abandoned without plugging.