SPE-198406-MS

Workflow of Oil Spill Estimation

Eghbal Motaei, Dharshini Yeap May Chen, and Raj Deo Tewari, PETRONAS Carigali Sdn Bhd; Elnur Mammadov,
PETRONAS Azerbaijan; Jazael Ballina Ruvalcaba, PETRONAS Carigali Sdn Bhd

Copyright 2019, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Annual Caspian Technical Conference held in Baku, Azerbaijan, 16 – 18 October 2019.

This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents
of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect
any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written
consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may
not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

Abstract
Blowouts have catastrophic consequences and can potentially occur during any exploration or development
projects. The situations that could lead to a blowout are underestimated pore pressure, rapid change in
pressure, abnormal pressure or operation complexities such as complete fluid loss and thief zones. Blowouts
are the most destructive and dangerous disaster in the oil and gas operations. Apart from causing fatalities
and injuries, a blowout also causes ecological and environmental damages. While there is remedial work
that can be done to manage the side effect of a blowout, an analysis has to be done to evaluate the extent
of the damage. The oil spill analysis has to include all parameters and take into consideration all scenarios
which would cater for the remedial work needed to address the aftermath of the blowout.
In this paper, the entire spill process beginning from the reservoir up to the surface is reviewed. The
review covers the phenomenon involved in a spill process, the related HSE concerns, remedial work and
spill modelling. The initial conditions, boundary conditions, and media transfer functions such as porous
media properties, sand face and wellbore geometry are fully captured in the study and is reflected in the
workflow. These parameters would affect the plume shape, size and geometry, blowout duration and oil
spill volume estimation. The workflow presented in this paper is an effective technique for efficient decision
making and remedial work in case of an oil spill resulting from an uncontrolled blowout.

Introduction to Blowout and Spill

A kick is the flow of formation fluids, be it oil or gas, into the wellbore. A blowout is the uncontrolled
release of formation fluid from a reservoir into the sea for offshore wells or into the air for onshore wells,
gained through the kick. Situations that can cause a kick are:

• Pore pressure higher than expected

• Mud pressure lower than intended

• Mud pressure reduced temporarily for operational reasons

2 SPE-198406-MS

Once a kick is detected, the well will be shut in by closing the blowout preventer (BOP) and well control
operation is started to remove formation fluid from the mud column and re-create overpressure against the
formation. However, if these efforts fail, a blowout happens.

Blowout Phenomenon
During drilling operations, there is a possibility of encountering various issues including oil and gas
blowouts. Among the problems are:

• Downhole conditions that cause sticking which hinder rotation and/or reciprocation

• Encountering weaker than usual formation that is unable to support the wellbore drilling fluid
densities needed to create an overbalance formation pressures
• Drilling fluid contaminations caused by excessive solids concentrations, chemical substances that
result in adverse reactions in the fluid properties or toxic components of the formation fluids
• Penetrating rock formations where the fluid pressure is greater than the hydrostatic pressure exerted
by the drilling fluid which causes an undesirable influx of formation fluid into the wellbore (Adams
and Economides, 2003)
The important components of an offshore spill initiating from the seabed sources after a blowout event
starts with a buoyant jet where there is also potential for gas separation and gas hydrates formation. At the
terminal level, the jet either devolves into individual droplets and forms a subsurface plume which later go
on to create an oil slick at the surface (Sim et al., 2015). Figure 1 illustrates the blowout phenomenon that
happens from the seabed up to the surface.

Figure 1—Blowout Phenomenon

SPE-198406-MS 3

HSE Concerns
The main concern of a blowout and an oil spill is to the health, safety and environment. The main concern
are fatalities and injuries. The Piper Alpha incident in 1988 killed 167 people while the Deepwater Horizon
explosion in 2010 caused the death of 11 workers and injured 17 others.
Blowouts also cause long lasting HSE risks that may last up to decades. Oil spills may introduce
polycyclic aromatic hydrocarbons (PAH) into the environment which are carcinogenic in high
concentrations and will interfere with biological processes. They also bring about ecological and
environmental system toxicities. Living things are harmed by spilled oil due to its chemical constituents by
both internal exposures through inhalation or ingestion as well as external exposure through eye and skin
irritants. Besides that, oil can suffocate smaller species and coat the feathers of birds inhibiting them from
maintaining their body temperatures. Not only are marine life and birds harmed by spills, but plants such
as mangroves and coral reefs are at risk as well.
Heavy metals such as lead (Pb), nickel (Ni), vanadium (V), zinc (Zn) and cadmium (Cd) are often detected
in oil spills which are detrimental to the human health with the side effects being mainly cancer (Mustafa et
al., 2015). Once these heavy metals are released to the environment after a blowout, they may be taken into
the body by ingestion as well as inhalation. As the accumulation in body tissues surpasses the detoxification,
a gradual build-up of these metals take place. An indirect way of the heavy metals being harmful to humans
is where the heavy metal is introduced to the food chain through fishes (Chinedu and Chukwuemeka, 2018).

Remedial Work
Oil spill response strategies and remedial work are intended to minimize the impact of the spill to the
environment as much as possible. Each option must be assessed for its operational limitations, effectiveness,
side effects, the application under different scenarios and also the safety and health of the responders
(Nedwed et al., 2012). There are a few remedial works for oil spills such as in situ burning of oil,
bioremediation, mechanical collectors and spraying of chemical dispersants and oil sorbents that will be
discussed further in this section.
In-situ burning (ISB) is a spill countermeasure involving the burning of the oil at the location of the spill
under controlled conditions. It has been acknowledged for many years as an effective way of eliminating
large quantities of oil over a short period of time. (Guénette, 1997). It significantly reduces the amount of
oil on the surface of water and diminishes the effect of the spill to the environment. Bioremediation is where
contaminants are removed using microbes or plants under natural circumstances. It can also be improved
by artificially adding microorganisms which would result in a higher degradation rate compared to natural
degradation (Al Mujaini et al, 2018). Mechanical recovery is the industry's primary response to oil spills.
The spills are contained using booms and skimmers remove the oil from the surface of the water. However,
these mechanical collectors become less efficient when the spills are larger in size (Nedwed, 2013). The
use of dispersants will break the surface slick into micron-sized droplets. This dilute mixture of oil-in-water
provides better access to the oil degrading bacteria compared to a thick surface accumulation (Nedwed et
al., 2012). Sorbents are materials that have the ability to sorb, uniformly distribute and trap oil within its
structure. One of the main characteristics of sorbent is its hydrophobicity which defines the effectiveness
of oil sorption from the surface of the water (Anuzyte and Vaisis, 2018).

Blowout Control History

Blowouts have been an issue in the oil and gas industry from the beginning of time, Blowouts still occur
despite the improvement in safety measures such as blow out preventers (BOPs) and other equipment and
drilling procedures. Figure 2 shows that since 1965, blowouts have been happening at a fairly stable rate
which has not changed despite the advancement in blowout prevention procedures and equipment (Skalle
and Podio, 1998).
4 SPE-198406-MS

Figure 2—Number of Wells and Frequency of Blowouts/100 wells

There are many ways an offshore blowout is controlled. Among them are bridging (39%), the use of
BOP (9%), cement (11%), depletion (9%), installing equipment (5%), using mud (19%) and drilling relief
well (5%) (Skalle, et al., 1999) (Figure 3). It is interesting to note that the chances to control a blowout
using the BOP is somewhat similar to drilling a new well. Bridging, which is the most common blowout
control, is where the fluid flow through the wellbore is obstructed because of well collapse or buildup of
material such as sand.

Figure 3—Blowout Control Methods

Factors Affecting Blowout and Spill Volume

Blowouts will differ significantly in character depending on the factors below.

• Flow medium: The potential harm that a blowout can bring upon the people, environment and
assets depends on the fluid that the outflow comprises of.
• Flow rate: The flow rate is affected by fluid types and properties, flow regime, completion and
reservoir properties. The fluid types and properties refer to the oil density, viscosity and gas content.
These would determine how easily the fluid flows through the reservoir and up the well. Flow
regimes can be divided into three namely the transient flow, pseudosteady state flow and steady
state flow. Completion is mainly the well design including the inner diameter, the measured depth
(MD) along the well path and the true vertical depth (TVD) of the well path. Lastly, reservoir
SPE-198406-MS 5

properties encompass the pore and fracture pressures, temperature, porosity, permeability and sand
production.
• Release point: Whether the fluid is released into the air or the water has a big impact on the flow
rate as well as the spread of pollution
• Duration: Together with the flow rate, the duration will also affect the volume of the spill and the
possible harm it brings. The duration refers to the amount of time taken to contain the blowout
• Flow path: The flow path is the channel through which the fluid flows from the reservoir to the
release point be it an open hole, the annulus between the casing and the string or the continuous
space between the casing and the formation
• Restrictions: Items limiting the potential flow along a path such as well equipment, partially closed
valves or fragments from the formation
• Reservoir: The reservoir size that the blowout is originating from is the ultimate limitation on the
volume of the spill. Flow conditions and pressures will largely impact the flowrates
• Reservoir exposure: This refers to the well length that extends into the reservoir, the diameter of
the well and the properties of the completion equipment (Nilsen, 2014).
• All these factors that would control the blowout have to be taken into consideration and it would
also determine the spill volume.

Oil Spill vs Gas Spill

Oil and gas spills have different properties. Oil spills start of as a jet at the base and experience phase change
at shallower depths. Gas spills, on the other hand, rises sharply due to higher slip velocity. Unlike oil, gas
and water do not mix. Figures 4 and 5 show the difference between gas and oil spills with and without
crossflows respectively.

Figure 4—Gas spill (left) and oil spill (right) with cross flow

Figure 5—Gas spill and oil spill with no cross flow

6 SPE-198406-MS

Surface Phenomenon of Spill

"Weathering" is the series of changes that oil spills on water undergo in chemical and physical properties.
The weathering process, shown in Figure 6, begins immediately after a spill happens although the rates may
differ. Among the processes involved in weathering are emulsifications, evaporation, spreading, dispersion,
dissolution and biodegradation.

Figure 6—Weathering Process

Evaporation is the process that effects the remaining amount of oil on surface most especially the lighter
oils. The rate of oil evaporation depends largely on the composition of the oil. The more volatile the
components, the faster the rate of evaporation. Emulsification is where one liquid is dispersed into another
in the form of small droplets. During an oil spill, the sea energy forces the entry of small water droplets into
the oil after which the droplets gravitate towards the bottom of the oil layer. This emulsification process
significantly increases the actual volume of the oil spill and also the viscosity of the oil which in return
makes the cleaning up efforts more complicated as emulsified oil is difficult to disperse or recover with
skimmers. Evaporation too slows down.
Dispersion happens when the fine oil droplets are created and transferred into the water column by
turbulence or wave action. Small oil droplets are fairly stable in water while larger droplets rise to the surface
faster. Similar to evaporation, dispersion is dependent on the oil composition where heavier oils are harder to
disperse and lighter oils disperse easily. Another factor effecting dispersion is the wave action. Dissolution
causes some of the soluble components of the oil to be lost to the water under the slick. It usually happens to
the lower molecular weight aromatics which are toxic to marine life. Dissolution takes place immediately
after a spill but the rate decreases rapidly as soluble components are depleted quickly.
Biodegradation is where species such as fungi, bacteria and yeast metabolize petroleum hydrocarbons
as food energy source. The rate of biodegradation depends on several factors such as temperature, carbon
number, presence of oxygen, availability of nutrients like phosphorus and nitrogen and accessibility of the
oil to the microorganism or bacteria. As biodegradation is a very slow process, it is not considered as an
important component of weathering.
Following an oil spill, the oil will spread into a slick over the surface of the water. Lighter products form
thin slicks and heavier products form thicker slicks. The spreading process is sped up by winds and currents.
In the absence of currents and wind, gravitational forces and interfacial tension between oil and water assist
SPE-198406-MS 7

spreading. Slicks are usually thinner at the outer edges compared to the inside but the effect lessens after
some time (Fingas, 2017).

Droplet Size Distribution

Droplet sizes are proportional to the buoyancy velocity. As such, smaller droplets will rise slower compared
to larger droplets which will rise faster. As a result, larger droplets reach the surface faster and will not reach
places further places hence the spill area is minimum. Comparatively, smaller sized droplets can be carried
further away by currents making the spill cover a larger area.
The formula to calculate droplet size distribution is (Dunn et al., 2018)

(1)

The global size of droplets is between 20 microns to 10mm. At the Deepwater Horizon, the droplet sizes
at about 50 microns have a good agreement with the measured subsurface plumes. Due to the decreasing
pressure, droplets expand as they rise. All in all, droplet size is a function of temperature, pressure and
current velocity. The particle size and distribution data obtained can be used to support operational decisions
during an oil spill response operation.

Hydrate Formation
Hydrates are solid ice like crystals that form when water and gas are combined under high pressures and
low temperatures. They form a closed cage structure that traps gas molecules within its interior. These
closed cage structure may further combine with others resulting in larger cavities. Turbulences such as high
velocities, pressure pulsations and agitation can serve as catalyst and aid the formation of hydrates. Presence
of H2S and CO2 also promote the hydrate formation as acid gases are more soluble in water compared to
hydrocarbons. In the instance of a blowout, hydrates may have a significant effect on the behavior of the jet.
In the Deepwater Horizon incident, attempts to contain the blowout using a dome structure was foiled
due to the formation of hydrates. The dome that was deployed trapped a mixture of oil, gas and water which
are the components required for the hydrates to form. The failure of the dome was due to gas hydrates
clogging up the containment structure.

Spill Models and Tools

There are different types of models available for spill modeling which are transient modeling, steady state
modeling and probabilistic modeling. Among these three, the most reliable is the transient modeling which is
done using Olga and Computational Flow Dynamics (CFD). However, transient modeling is time consuming
and costly. As a cost and time effective method, simpler models, namely the steady state modeling and
probabilistic modeling are used.

Transient Modelling
The transient modeling has several functions including evaluating the spilling process and the subsurface
phenomenon such as hydrate formation and plume propagation. This model also estimates the cease time
using correlations as well as the oil spreading and diffusion. Although it provides the most detailed analysis,
it is also the most expensive and takes a longer time to complete.

Steady State Modelling

Steady state modeling is done using PROSPER and WellFlo. Using the well model, the well spill potential
is estimated and the cease condition can be evaluated.
8 SPE-198406-MS

Probabilistic Modelling
The probabilistic model is done subsequent to the steady state model. Once the steady state model has been
created, a Monte Carlo simulation is run on top of it. Doing this will give a probability distribution of the
spill volume estimations. This addressed all the possible spills and the corresponding remedial work.
As transient modelling is the most expensive and takes the longest time to be completed, it is rarely
chosen for spill modeling. The most convenient method would be to use the steady state modelling together
with the probabilistic modelling for oil spill estimations. In this paper, the workflow establish is based on
this method.

Plume Characterization
A deepwater blowout can lead to the release of uncontrolled volumes of hycrocarbons at the seabed. The
fluid that is released will move upwards in a shape known as a bubble plume. Initially, the fluids move due
to the jetting action of the high velocity flow. However, the surrounding sea water quickly dampens the
jet. At this point, gravity segregation takes place due to the density difference between the seawater and
hydrocarbon, resulting in an upward movement.
As the plume moves upwards, the confining pressure decreases and the volume increases. Although for
a large section between the seabed and below the sea level the expansion is linear, however, the volume
increment accelerates near the sea level till it reached the surface. The plume behavior can be further broken
down into three main regions shown in Figure 7.

Figure 7—Blowout Plume

• Zone of Flow Establishment (ZOFE): Velocity in core region is assumed to be the same as the
blowout exit velocity at the seabed
• Zone of Established Flow (ZOEF): Plume gradually widens with height due to expanding core
plume volume and entrained water volume and are usually in conical shape
• Zone of Surface Flow (ZOSF): Plume dynamics change as boundary condition is reached as sea
level. It changes from an upward movement to a radial flow (Adams & Economides, 2003)
SPE-198406-MS 9

Workflow
The workflow of oil spill estimation has 5 main components which can be simplified as follows:
i. Building blowout scenarios
ii. Oil and gas spill rate calculations
iii. Blowout duration
iv. Oil and gas spill volume distribution
v. Spread area estimation
Each of these components in Figure 8 would be discussed in detail in the next section.

Figure 8—Oil Spill Estimation Workflow

Building Blowout Scenarios

There are many different scenarios which have the prpobability of happening. Figure 9 is the representation
of 12 blowout scenarios. Each junction represents the probability split between possible events. There are
4 factors taken into consideration which are the size of the hole that exposes the reservoir, the length of
the aforementioned hole section, the possibility of restrictions which will limit the flow of the well such as
presence of drill string and the status of well control equipment or systems such as the BOP closure. Each
of these scenarios will yield a different result.

Figure 9—Development of Blowout Scenarios with Probabilities

The first branch is the probability of a blowout happening. The second branch is the probability of a
blowout happening at various degrees of penetration. The third branch is divided equally between swabbing
in and detecting a kick when the bit is near the bottom or swabbing in reservoir fluid on the way out which
is not detected prior to removing the string from the well. At the fourth branch, the probability is divided
10 SPE-198406-MS

between the BOP failing to close and the well cross section being 95% closed. Every combination yields
different results.

Oil and Gas Spill Rate Calculations

Each of the 12 scenarios illustrated above have their corresponding blowout rates. Acknowledging the
challenges the nodal modeling was based in pressure drops from the reservoir to the ocean, the well model
can be built using data such as water depth, reservoir pressure, temperature, fluid composition, GOR,
permeability, viscosity, porosity, drill string diameters, pore pressure, geometries of the well, BOP data, riser
data, tubing head pressure, and temperature conditions which would give and estimated range of possible
flows, based of different scenarios of how the fluid will be flowing through the well and the blowout rates,
the most significan factor impacting the well model result is the bottom hole pressure.
The workflow in Figure 10 represents the steps to evaluate the rate calculations. The reservoir and well
modelling is an excellent adjunct of flow rate information. Running all these scenarios and permuting
variables from surface, well and reservoir will give multiple rates that will be organized and evaluated by
probabilistic distribution in order to get the desirable rate calculations.

Figure 10—Dynamic workflow representing the steps to get the possible oil and gas rate calculations.

The intersection between the inflow performance relationship (IPR) and the vertical lift performance
(VLP) curves is the rate (Figure 11). The model results from the varios scenarios for base case, low case and
high case and was clustered into probability distributions with the best estimate for flow rate, the calculated
rates are summarized in Table 1.
SPE-198406-MS 11

Figure 11—Oil and Gas Spill Rate Calculation using Well Mode l

Table 1—Calculated blowout rates for each scenario

Scenario Probability Rate (m3/d)

1 2.90% 1,912
2 6.70% 1,705
3 12.80% 5,011
4 29.90% 4,435
5 6.40% 5,414
6 15.00% 4,781
7 6.40% 13,248
8 15.00% 11,520
9 0.80% 2,707
10 1.80% 2,615
11 0.80% 3,283
12 1.80% 3,000

The strength of the fluid flow which is the rate dictates the total volume of spill as well as the scale of
potential damage to the environment. Only the initial rate is assessed and it is assumed that it would be
constant throughout the time taken to kill the well. Pressure drop and other factors that would affect the
rate over time is not taken into account.

Blowout Duration
The blowout duration depends on the responses to the blowout by human introduced or reservoir and well
performance. There are a few mechanisms to breaking a blowout.

• Active rig barriers : Mechanically shutting in the well or killing it with mud and cement

• Bridging : No human intervention and change of flow conditions causes the blowout to cease due
to either collapse of the well or unconsolidated material clogging the well
• Drilling of relief well : A new well is drilled into the bottom part of the well experiencing the
blowout and is used as a work channel for stopping the blowout using mud and cement
• Natural cessation : The flow of oil ceases either because of drop of pressure in the reservoir, increase
in pressure in the well or water or gas coning that results in change of flow medium
The assessment of a blowout duration may be expressed as follows
12 SPE-198406-MS

(2)
Using this equation and determining the uncertainty allows the distribution for the blowout duration to be
calculated. The duration range should also cover all P90 to P10 cases. Similar to the flow rate, the blowout
duration also determines the spill volume and the extent of the damage caused.

Oil Spill Volume Distribution

Using the spill rates and the blowout duration calculated, the oil volume distribution can be generated. The
formula "=NORM.INV(probability, mean, standard deviation)" is used to normally distribute the oil spill
rates (Figure 12), gas spill rates (Figure 13) and blowout durations (Figure 14) calculated above. The oil
spill volume (m3) (Figure 15) is then calculated by simply multiplying the spill rate (m3/day) by the duration
(day). The similar can be done to calculate the gas spill volume by using the oil oil rates, GOR and the
blowout duration (Figure 16).

Figure 12—Oil Spill Rate Frequency

Figure 13—Gas Spill Rate Frequency

SPE-198406-MS 13

Figure 14—Spill Duration Frequency

Figure 15—Cumulative Oil Frequency

Figure 16—Cumulative Gas Frequency

14 SPE-198406-MS

The probabilistic method of Monte Carlo simulation is applied on the steady state model to determine
the spill rate and blowout duration distributions which results in the distribution of volume discharged.

Spread Area Estimation

Once these have been established, now, it is possible to estimate the oil spill spread area. The spread area
is highly dependent on the current vortex, wave strengths and heights and also the wave sped.
This spread area can be calculated using the equation

(3)

where Vspill is the spilled oil volume and Hmax is the maximum thickness of the spilled oil which could be
estimated to be 1/6 to 1/3 of the wave height.
Once the spread area of the oil spill is estimated, it will now be possible to cater for all necessary steps
that should be taken for remedial works and efficient decision working should a blowout happen.

Rate Dependency on Wellbore Integrity

Skin Build up and wellbore collapse are the main issues of un-restricted flow and un-limited draw down
which leads to:

• Wellbore fill up with sand

• Reducing effective area to flow

• Wellbore collapse

The rate discount that should be applied for each wellbore integrity is shown in Figure 17. Rate
discount needs to be defined in probabilistic approach terms to cater the wellbore integrity as per 1/3
in unconsolidated rock, 100% for consolidated rock with intermediate case of ½. In other words, for
unconsolidated rocks the ranges could be discounting rate by 1/3, ½, 1/1 as a range of discount. While for
consolidated sand, rate discount could be limited to ½ for low case, ¾ for intermediate case and 1 for the
high case. As for carbonates or deep reservoirs, the distribution curve is skewed left with the discounting
rates ranging from 50% to 100%.

Figure 17—Rate discount probability for wellbore integrity

Spill Control Points

Spill control points refer to the tubing head pressure (THP) of the well during a blowout. For onshore wells
or wells in shallow water, the THP during a blowout is zero. For wells in deep water fields, the THP will
be a function of water depth. As such, tubing head pressure can be calculated using the following equation:
SPE-198406-MS 15

(4)

Figure 18 shows the spill control point and tubing head pressure calculations. This THP is among factors
that determine the blowout rate discussed earlier.

Figure 18—Spill Control Point

Well Control Barriers

A barrier is a well control system that prevents fluids from the formation from flowing unintentionally to
the surface. Barriers can be divided into two which are the primary barrier and the secondary barrier where
the primary barrier is the barrier that is closest to the reservoir, the source of flow. The barriers may vary
depending on the operation stages be it drilling, completion, production, intervention or workover.
During the drilling and completion operation the primary and secondary barriers are the circulating or
static drilling fluid and the blow out preventer (BOP) respectively.
For a production well, the primary barriers available are the packer that seals off the annulus, the tubing
flow in the subsurface safety valve (SSSV) and the SSSV and the secondary barriers are the tubing above
the SSSV, the X-mas tree main flow side, the casing or the wellhead and the annulus side of the X-mas
tree (Holland, 2017).
As for intervention and workover, the barriers change according to the operations being carried out
which encompass the barriers of drilling and production operations. In short, production, intervention and
workover operations have the maximum number of barriers while drilling and operations have the minimum
number of barriers.
In the event the primary and secondary barriers fail, it may lead to a blowout. Similar to how each
operation has its own barrier, it also has different blowout rates. By applying the method discussed in the
earlier sections, the probability of the blowout rates ranges can be calculated.
16 SPE-198406-MS

By plotting the spill rates according to the operation as done in Figure 19, it is noticed that operations
with lower spills such as production, intervention and workover have the maximum number of barriers.
This is because they have a faster cure.

Figure 19—Rate distributions for different operation types

Flame Height Analysis

Oil spill rate and volume estimation is only applicable for blowouts in offshore wells. For onshore wells,
a flame height analysis is done. Knowing the flame height is important to quantify the magnitude of the
discharge and evaluate the damage it would cause which would assist in preparing rescue actions.
The three main factors that determine the flame height are flow rate, wellbore diameter and the type of
fluid, whether oil or gas. The formula for calculating flame height is as follows (Evans, et al., 1994):
(5)

(6)

The normalized flame height may be calculated by dividing the flame height by the pipe diameter. The
normalized flame height can then be plotted against the dimensionless flow number, U* where
SPE-198406-MS 17

(7)

From Figure 20, it is observed that with increasing U*, there are 3 different flow regimes:

Figure 20—Normalized flame height as a function of U*

• A buoyancy-dominated laminar and pool flame regime that eventually merges into a turbulent,
subsonic flow
• A transition from momentum-dominated turbulent subsonic, flow to

• A choked, turbulent supersonic flow (Bradley, et al., 2016)

Conclusion
A complete workflow that fully addresses the steps in oil spill estimation is established. The outcome of
this workflow is used to evaluate the risk associated with the oil spill. This study is an effective analysis and
technical evaluation that takes into consideration all scenarios and parameters through proper modelling to
cater the necessary steps to be taken and remedial work to be carried out after a blowout. It will also assist
in efficient decision making. The workflow begins with building different blowout scenarios, followed by
oil and gas spill rate calculation, blowout duration calculation, the oil and gas spill volume distribution
and finally the oil spill spread area estimation. The spill spread area estimation is crucial in selecting the
remedial work to be carried out in the event of a blowout be it in-situ burning, bioremediation, mechanical
recovery, dispersants or sorbents. The proper selection of remedial work would reduce the HSE concerns of
the blowout such as fatalities, injuries, ecological and environmental system toxicities and release of heavy
metals to the environment. This workflow should be applied in all E&P activities to address the concerns
of a blowout. In conclusion, it is important that we understand the environmental implications of oil spills
to raise awareness and prevent future mishaps.
18 SPE-198406-MS

Nomenclature
Dmax Maximum droplet diameter (m)
σ Surface tension (kg/s2)
ρg Density of gas (kg/m3)
ρf Density of liquid (kg/m3)
jgVolumetric flux of gas (m3/s.m2)
Reg Gas Reynolds number
µg Viscosity of gas (kg/m.s)
µf Viscosity of liquid (kg/m.s)
Z Flame height (m)
C Fuel dependent constant
d Nozzle diameter (m)
Frm Modified Froude number
rho Mass flow of oil (kg/s)
rhg Mass flow of gas (kg/s)
u Velocity (m/s)
g Gravitational constant (m/s2)
u fuel flow mean velocity at the exit plane of pipe for subsonic flow. For ratios of
atmospheric pressure to pipe pressure equal to, or less than, the critical pressure ratio,
sonic velocity after isentropic expansion (m/s)
SL maximum laminar burning velocity of the fuel-air mixture under conditions of ambient
atmosphere (m/s)
ReL Reynolds number based on SL
Pi Initial stagnation pressure (Pa)
Pa Atmospheric pressure (Pa)

References
Adams, N., & Economides, M. J. 2003. Characterization of Blowout Behavior in Deepwater Environments. Presented at
2003 IADC/SPE Drilling Conference, Amsterdam, The Netherlands, 19-21 February. SPE/IADC 79879.
Sim, L., Graham, J., Rose K. et al. 2015. Developing a Comprehensive Deepwater Blowout and Spill Model. EPAct
Technical Report Series. National Energy Technology Laboratory, Albany, Oregon.
Mustafa, A.D., Juahir, H., Yunus, K. et al. 2015. Oil Spill Related Heavy Metal: A Review. Malaysian Journal of Analytical
Sciences 19 (6): 1348–1360.
Chinedu, E. and Chukwuemeka, C.K. 2018. Oil Spillage and Heavy Metals Toxicity Risk in the Niger Delta, Nigeria.
Journal of Health & Pollution 8 (19): 1–8.
Nedwed, T., Coolbaugh, T. and Demarco, G. 2012. The Value of Dispersants for Offshore Oil Spill Response. Presented
at 2012 Offshore Technology Conference, Houston, Texas, USA, 30 April-3 May. OTC 23359.
Guénette, C. 1997. In-Situ Burning: An Alternative Approach to Oil Spill Clean-Up in Artic Waters. Proc., International
Offshore and Polar Engineering Conference, Honolulu, USA, 25-30 May, 587–593.
Al Mujaini, M., Joshi, S.J., Sivakumar, N. et al. 2018. Potential Application of Crude Oil Degrading Bacteria in Oil Spill
and Waste Management. Presented at 2018 SPE International Conference on Health, Safety, Security, Environment,
and Social Responsibility, Abu Dhabi, UAE, 16-18 April. SPE-190564-MS.
Nedwed, T. 2013. Recent Advances in Oil Spill Response Technologies. Presented at 2013 International Petroleum
Technology Conference, Beijing, China, 26-28 March. IPTC 16845.
Nedwed, T., Tidwell, A., Buist, I. et al. 2012. Advances in Treating Agents for Oil Spill Response. Presented at 2012 SPE/
APPEA Internation Conference on Health, Safety and Environment in Oil and Gas Exploration and Production, Perth,
Australia, 11-13 September. SPE 158087.
Anuzyte, E. and Vaisis, V. 2018. Natural oil sorbents modification methods for hydrophobicity improvement. Energy
Procedia 147 (2018): 295–300
Skalle, P. and Podio, A.L. 1998. Trends extracted from 1,200 Gulf Coast blowouts during 1960–1996. Presented at 1998
IADC/SPE Drilling Conference, Dallas, Texas, 3-6 March. SPE 39354.
SPE-198406-MS 19

Skalle, P., Jinjun, H. and Podio, A.L. 1999. Killing Methods and Consequences of 1120 Gulf Coast Blowouts During
1960-1996. Presented at 1999 SPE Latin America and Caribbean Petroleum Engineering Conference, Caracas,
Venezuela, 21-23 April. SPE 53974.
Nilsen, T. 2014. Guidelines for estimating blowout rates and duration for environmental risk analysis purposes. Final
report, Norwegian Oil and Gas, Stavanger, Norway.
Fingas, M. 2017. Oil Spill Science and Technology, 2nd Edition. Texas, United States: Gulf Publishing Company.
Dunn, M.D., Fitzgerald, S. and Garner, J.B. 2018. Predicting Hydrocarbon Burn Efficiency of Ignited Blowout for Oil
Spill Source Control. Presented at 2018 IADC/SPE Drilling Conference and Exhibition, Fort Worth, Texas, 6-8 March.
IADC/SPE-189610-MS.
Holland, P. 2017. Loss of Well Control Occurrence and Size Estimators, Phase I and I. Final Report, ExproSoft, Trondheim,
Norway
Evans, D., Madrzykowski, D. and Haynes G.A. 1994. Flame heights and heat release rates of 1991 Kuwait oil field fires.
Presented at Fourth International Symposium on Fire Safety Science, Ottawa, Canada. https://dx.doi.org/10.3801/
IAFSS.FSS.4-1279
Bradley, D., Gaskell, P.H., Gu, X. et al. 2016. Jet flame heights, lift-off distances, and mean flame surface density for
extensive ranges of fuels and flow rates. Combustion and Flame 164 (2016): 400–409. http://dx.doi.org/10.1016/
j.combustflame.2015.09.009