Oil! A Primer on Petroleum Exploration & Production

I have been following Jayman's interesting and detailed description of the workings of a refinery in http://priuschat.com/forums/freds-house-pancakes/47896-pickup-trucks-oil-refinery-discussion-jay-jimmie-production.html. This topic has generated much interest, and I thought that a similar thread about the upstream side of the petroleum industry is in order. As such, I will begin posting installments discussing petroleum exploration and production (E&P). I will focus primarily on land-based oilfields but could certainly address any questions regarding gas fields or offshore E&P, though they are all similar in principle.

A little background on myself:
I have a BS in Geological Engineering from Montana Tech. My primary focus was reservoir characterization and simulation. I have a fairly wide breadth of experience in the industry. I have worked for one of the large service companies in the deepwater Gulf of Mexico building, testing, deploying, and operating sand and gravel frac/pack and tubing conveyed perforating systems (more detail on these topics to come). I have also worked for a large independent as a reservoir engineer in one of the nation's biggest gas fields. I am currently a reservoir engineer for a major (integrated multinational oil co.) in Alaska, where I perform a variety of engineering tasks for fields on the North Slope.

I envision this primer as roughly following the following outline:
-Prospecting (subset of exploration)
-----Basic Petroleum Geology
-Exploration
-----Seismic
-----Drilling
-----Appraisal
-Field Design
-----Characterization & Simulation
-----Subsurface and Infrastructure
-Execution
-----Phased Development
-----Drilling
-----Completions
-Field Life
-----Maintenance
-----Expansion
-Decommissioning
-----Blowdown
-----Plug and Abandon

These topics can be as basic or detailed as the audience demands, and I will try to answer all questions that are posed. A good resource for oilfield terminology can be found at Schlumberger Oilfield Glossary Search Results

 

Okay, sorry it has taken me so long to get rolling on this, but here goes...

Chapter 1 - Petroleum Geology and Prospecting
Since I intend to focus more on the engineering aspects of the industry, I will spare you all the gory details about the deposition and burial of organisms which ultimately leads to the generation of hydrocarbons and focus more on what constitutes a petroleum reservoir. From a geological standpoint all* of the following seven conditions must be in place to constitute a reservoir:
1. Source** - aforementioned burial of organic rich sediments
2. Reservoir - a porous and permeable*** layer of rock (e.g. sandstone)
3. Seal - a relatively impermeable layer of rock overlaying the reservoir (e.g. shale)
4. Trap - a physical barrier either stratigraphic (seal rock) or structural (faulting and folding), though the trap is usually a combination of the two
5. Timing - if the oil is generated in the source rock and migrates into the reservoir rock prior to a seal being deposited or a trap forming, it will not accumulate
6. Maturation - exposure of organic rich sediments to temperature and pressure for a significant duration to generate hydrocarbons
7. Migration - movement of oil from source rock to reservoir rock

From an engineering/economic standpoint one more very important condition must be met:
8. Oil must be technically and economically recoverable

Now, assuming we have met all the conditions necessary for a petroleum reservoir, we must go about finding it. In the early days of petroleum exploration, most oil deposits were discovered by obvious oil seeps on the surface or by telltale surface expressions of a subsurface structure that is indicative of an oil trap (e.g. anticlines surface/subsurface cross-section). The rest were found by dumb luck. Fast forward to today, and all the "easy" fields have been found. Most exploration prospects today are identified by their proximity to and similar structure to existing fields (e.g. ANWR/Prudhoe). Alternatively, some governments will hire geophysical firms to shoot seismic surveys in potential oil producing areas. If the results are promising, they could potentially persuade an oil company to seek exploration rights. After an area has been identified as a reasonable candidate for exploration, the way forward depends largely on uncertainty and risk tolerance. We will work under the assumption that the field we wish to explore is quite large, and, as such, is likely only to be developed by a large, multinational oil company. Companies of the scale of ExxonMobil or Shell have the assets to be rather risk tolerant; indeed, a robust exploration program is an absolute requirement. If new resources are not booked to replace those that are produced and sold, the company's potential to generate revenue literally shrinks.

So, at this point, we have identified a prospect for exploration, but we are not ready to drill yet. First, we must at the least perform a 2D seismic survey****. This will give us two cross-sections (note fault and folding of strata) of the subsurface structure, but certainly not enough to piece together a comprehensive picture of it, let alone any indication of the presence of hydrocarbons. If a compelling case for drilling can be made with the 2D seismic survey, the decision to go ahead with the drilling can be made. If the survey is inconclusive but indicates potential, a 3D seismic survey can be requested. A 3D seismic survey is a tremendously expensive and time consuming process. Even after the physical work required to conduct the survey, be it with thump-trucks or explosive charges, the data recovered must be filtered, processed, and interpreted (a process called inversion). Depending on the amount of processing desired, the inversion can take a few to several months to complete. Much information can be gleaned from the data depending on how much you want to spend, though there is a possibility that the data is garbage, and all the money spent on the survey was a waste. Assuming the data is not corrupt, it can provide a detailed picture of the subsurface structure and faulting and, sometimes, indications of porosity, mineralogy, and other attributes. Although, a great deal of uncertainty can be eliminated with a 3D seismic survey, ultimately, it still cannot answer the most important question: Are hydrocarbons present? The only way we can answer that question is to drill an exploration well, which, assuming our information gathered thus far indicates potential, is our next step in the exploration process and the focus of my next installment.

Notes:
*It is extremely rare that all of these conditions are met - on the order of 2-3% of potential reservoirs ever come to be.
**The organic rich components in the sediment comprising source rocks are initially transformed into a collection of organic compounds referred to as kerogen. If kerogen is heated to the right temperature, it is converted into crude oil and/or natural gas. Many potential source rocks are never buried deeply enough and, thus, not heated enough to be converted into crude oil. A great example of this is the Green River formation in Wyoming, which is one of the "oil shales" (bit of a misnomer: actually an organic rich marl - geology is 90% semantics) that have recently gained fame as our saving grace from the "energy crisis." The only way to extract oil from these formations is to heat it to convert the kerogen to oil and retort it from the rock, thereby essentially "speeding up" geologic time. So, the vast "oil shales" in the US are incorrectly referred to as reserves, when in fact they could likely not even be booked as resources.
***Assuming we are discussing a conventional reservoir. An unconventional reservoir such as the Bakken Formation in the ND is very much lacking permeability, though it is most certainly a legitimate reservoir.
****The cost and level of complexity of seismic surveys vary widely depending on where they are performed. Land based surveys in areas with established roads, such as fields in WY, can be performed relatively easily with vibroseis trucks. Land based surveys in areas like the North Slope of AK are quite labor intensive and expensive, as the energy source for the survey is typically in the form of explosive charges and all equipment must be deployed in sensitive areas with special equipment. Offshore surveys are the relatively easy, quick, and inexpensive to complete, as all the equipment is towed behind a boat with a self contained energy source.

Lastly, you may have noticed the conspicuous lack of mentioning the permitting that must be completed before commencement of each part of the process. Permitting is an extremely important aspect of the business, but it is well beyond the scope of this discussion. A veritable army of people dedicate their entire careers to the task of permitting.

Well, hopefully I haven't scared everyone off with all this rambling. If you have any questions, feel free to ask, and I will answer them to the best of my ability.

 

With regards to the seismic work, I should have made that more clear. The oil company conducting the exploration will decide if they want to conduct a seismic survey and will likewise pay for it. Permitting requirements can vary greatly depending on where the prospect is located. For the sake of this discussion, let's assume that the acreage is located on either state or federal land. Prior to the exploration work that I described, the acreage in question must be leased. If the acreage is federal land, the mineral rights can be leased from either the BLM (onshore) or the MMS (offshore) by any company wishing to conduct petroleum exploration. Acreage with suspected or demonstrated potential are typically auctioned, and the rights go to the highest bidder. A parallel process is conducted by equivalent state land management agencies in the case of state lands. The leasing process for public lands is fairly straight forward. The permitting required for actually drilling on said land is a lengthy, complicated, and expensive process. A few of the organizations/agencies involved in the permitting process are the EPA, DEQ, their state equivalents, various NGO's, and private interests. The issuance of the necessary permits takes on the order of years, and, especially in sensitive areas, can be tied up in the court system for quite a long time. Another interesting note about the lease agreement is that it typically has stipulations. These stipulations can vary depending on who the land was leased from, but in general the lease holder is required to drill x number of wells within the next x number of years or they forfeit the lease. This prevents companies from buying and shelving a lot of acreage that would otherwise generate federal and/or state revenue were it developed.

The area of ANWR which is under the most consideration for petroleum exploration is the 1002 Area. This is the location of the only well ever drilled in ANWR, the KIC#1 well. This well was a joint venture by Chevron and BP in 1980, and all information pertaining to this well remains a closely guarded secret. The KIC#1 well was located on native lands, which further complicates matters from a permitting standpoint, but I am not going to go into detail about this. ANWR contains acreage that is federal, state, and native held. For the sake of simplifying the discussion, I will address the process as though it were a federally conducted lease. So, if ANWR were opened up tomorrow, the BLM would auction the mineral rights to the highest bidder. Any company that wanted to conduct exploration would bid on acreage, and try to position themselves in the best parts of the acreage based on current knowledge. After the lease transaction is completed and stipulations put in place, the lease holder is then able to apply for permits to begin exploration work. My discussion of prospecting and exploration worked under the assumption that the leasing process was complete and all permits were in place. Typically, no one company attains the rights to an entire area, but rather the large oil companies usually work as partners from exploration and throughout the life of the field. This is a means of spreading the extreme risk associated with petroleum exploration.

 

Thanks for all the feedback so far, and I am glad that you are all enjoying the read. I will try to get the next installment posted sometime tonight.

Evan, I had not considered turning this into a wiki, but it is certainly a possibility.

Hill, I will try to address some of the extraordinary costs associated with exploration drilling in the next episode. In the meantime, consider that deepwater offshore daily rig rates are on the order of $250,000+/day, and that just scratches the surface of total expenditures.

And now, back to work...

 

Sorry it is taking so long to continue this story. Work and home have been busy lately. Anyhow, last time we left off, we had completed our 3D seismic survey. For the sake of discussion, let's assume we found promising structural features and we have all permitting in place to proceed with exploration drilling. I would also like to add that the process described in my first installment is typically years in duration, but all oil companies keep a sizable stable of identified prospects so that they can be "high-graded" when exploration money is budgeted. So, enough of that, let's drill a well.

Chapter 2 - Exploration Drilling
So, at this point we have completed our seismic survey, and we have found what looks to be a promising structure. For instance, let's say that we found an anticlinal/domal structure like the cross section shown in Chapter 1. Virtually all petroleum reservoirs contain both oil and water (almost always saline), and quite often they contain gas. These fluids have migrated upslope (updip in geological parlance) into the reservoir and separated out by density. We will assume that our reservoir contains oil, water, and gas. The fluids (gas is a fluid [note: not liquid]) are stratified, that is, water is on the bottom of the structure, oil in the middle further up the structure, and gas is at the crest of the anticline (i.e. a "gas cap"). We have a contour map of the structure from our seismic survey but we have no idea where the oil/water contact (OWC) is or where the gas cap begins or, indeed, whether there is oil present at all. The only way we can find out is to actually drill a well. But where do we drill? We don't want to drill so far down that we penetrate the reservoir below the OWC and find only water, as this could end our exploration. On the other hand, we don't want to drill right at the crest of the structure because, while this may reveal the presence of a gas cap to us, it will not tell us whether or not oil is present in the reservoir. We will choose a location, say halfway up the structure (this is actually quite a bit more exact, but falls much more in the realm of geology/geophysics than petroleum engineering). Remember, our first exploration hole is drilled strictly to determine if oil is present as well as provide some other key reservoir parameters (more on that later).

We have selected the location for our first well, secured all pertinent permitting, and either contracted a drilling rig or ramped up our own rig and crew. In the interest of limiting exposure to risk, we will likely choose to drill a vertical well with a conventional rotary rig (this link contains more than you would probably ever want to know about conventional rotary drilling rigs). The first thing we must do is design our well based on what limited information we have. We know with a fair degree of certainty (say, +/- 10m) where the top of our target formation is and how thick it is based on our seismic data. However, we are lacking a very key piece of information that can have devastating consequences: Pressure in the reservoir. To get some idea of what to expect, we will consult very experienced geologists who have much experience in estimating pressure in formations based on analogous fields that have already been drilled and prior drilling experience. Wow, how many times can I use experience in one sentence? Probably not enough to emphasize the importance of these geologists. Also, less dangerous but equally likely to ruin the well we are drilling, are zones with low pressure, which will result in lost circulation (i.e. the fluid used for drilling goes into the formation never to return to surface - no returns, no drilling). With our compiled data and estimates, we set out to "spud" our well. "Spudding" a well is the term given to the act of the start of drilling. As an aside, I had no idea where this term came from until I read Upton Sinclair's Oil!. It is actually an onomatopoeia describing the sound that the early blade bits made as they first broke ground (spud, spud, spud). The first thing we must do is set a surface conductor. This is a large piece of casing (say, 42+") that is set shallowly (typically <100') into the ground. This "string" (name given to lengths of pipe, be it a casing string, or a drilling string, or a tubing string) serves as a structural component to the well. It keeps the shallow, unconsolidated (non-rock) sediment from caving into the well bore when drilling begins, but it is typically not cemented in place. Having set our conductor, we will begin drilling through the shallow subsurface and aquifers. After we have drilled through freshwater aquifers to a depth largely determined by the regulations set by the EPA and DEQ for the area being drilled, we run surface casing. This could go anywhere from, say 500-2000', depending on aquifers present, and it serves to protect said aquifers. This casing also serves as a place to mount the blowout preventer (BOP) and directs return fluid up the wellbore. Once this casing is run in the hole, cement is pumped down it and displaced up the backside (annular space between the drilled hole and the outside diameter of the surface casing). Ideally this cement would be pumped all the way back to surface. The volume of cement needed is calculated such that we only leave the bottom of the surface casing filled after it is displaced behind the pipe. The cement cures, and a pressure test is performed to make sure an adequate seal was achieved prior to continuing drilling. Now we mount the BOP stack. This equipment is used in the case of a well control event (i.e. blowout). Normally, downhole pressure is balanced by the weight of the drilling fluid (mud) in the drill string. The density of the mud is managed by a mud engineer, who is continually monitoring the density and viscosity of the mud and adding components as necessary to weight up or weight down fluid. If a formation is encountered that has a higher pressure than the column of drilling fluid in the drill pipe, a blowout can occur. In the case of a blowout, the formation pressure forces the mud up and out of the hole. Without the BOP, the mud and reservoir fluids would blow out at the surface and continue to do so until the pressure is equalized. Think of the old pictures of "gushers" in Texas. These are avoided at all cost today as they result in massive environmental damage, injuries and fatalities to rig workers, and often the loss of the entire rig as blowouts usually ending up igniting and burning everything down. To combat all this we have the BOP stack. The first line of defense is the hydril, which is basically an inflatable rubber donut that seals around the outside of the drillstring hopefully holding back the pressure. If that doesn't work, above the hydril is a set of pipe rams which are hydraulically actuated rams that seal steal plates around the drillstring. If these fail, your last option is the shear rams. These are like the pipe rams but rather than fitting around the drillstring, they actually cut it off and form a blind seal across the BOP stack. This equipment is tested at the least every few days as it could mean life or death for the workers on the rig.

So, as we continue drilling, a geologist (mud-logger) is continually monitoring the cuttings returning with the drilling mud. Sorry, I should clarify that the drilling mud is pumped down the drillstring, where it cools and lubricates the bit, and returns up the annulus between the drillsting and surface casing carrying with it the finely pulverized rock. Drilling operations go on 24/7 usually with trips into and out of the hole with the drillstring to change bits. If any zones of lost circulation (drilling mud does not return) are encountered, the mud engineer will mix additives into the mud (walnut shells, wheat hulls, other fibers) to try to plug the offending zone. If the zone cannot be plugged, an intermediate casing string will be run and cemented to cover the "thief" zone. This can also be done for overpressured zones. The danger here is that each subsequent casing string must fit inside the previous one, so the internal diameter is shrinking with each additional string of casing that is run; as such, a smaller diameter bit must be run with each additional casing string. Typically this is not a problem, however in the Gulf of Mexico, many overpressured zones are encountered, resulting in numerous casing strings. There have been incidences of wells never reaching the planned depth because so many casing strings were run, that the internal diameter became too small to continue drilling. This is a very, very costly mistake. Okay, sorry to keep side-tracking, but this topic could be a thread unto itself. So we are drilling, and the geologist is watching the cutting, looking for a particular formation (probably a sandstone) that is our target reservoir. When the geologist thinks we are nearing the top of our target formation, he will tell the driller to slow down the rate of penetration (ROP). He will begin to watch the drilling mud returns for "shows" of oil as we enter the target zone. If oil is present, we are (potentially) rich, if not, we perhaps just wasted $20-200 million. Let's assume we hit oil. We continue drilling until we are through the entire thickness of the target formation, and drill a "rat hole," which leaves enough length for the tools in the subsequent logging run to pass by our entire reservoir formation. Our next step will be to log our exploration well, perform well testing, and, maybe, drill a couple of appraisal wells. All these topics and (maybe) more shall be covered in the next installment.

A few notes:
There are far better resources available on the web than my clunky explanation of drilling. For a much better explanation try this primer.
Diagram of things that can go wrong, such as lost circulation, blowouts, etc.
Wellbore Diagram (note: our well is not cased to the bottom, so that we can run open hole logs)
Also, here is a diagram showing the sequential casing runs.


Wow, sorry this one is such a rambling mess. As always, I will be happy to answer any questions that anyone might have, and I will, in the future, try to have my updates more regularly and expeditiously posted.