IPTC 11958

Diagnosis of Excessive Water Production in Horizontal Wells Using WOR

Plots
Majid A. Al Hasani, Petroleum Development Oman, SPE, Saif R. Al Khayari, SPE, and Rashid S. Al Maamari,
Majid A. Al Wadhahi, Sultan Qaboos University, SPE

Copyright 2008, International Petroleum Technology Conference cut in general impacts both the inflow and outflow curves
This paper was prepared for presentation at the International Petroleum Technology negatively. Higher water production increases the cost of
Conference held in Kuala Lumpur, Malaysia, 3–5 December 2008.
fluids lifting to surface, water treatment, and disposal. Water
This paper was selected for presentation by an IPTC Programme Committee following review production is also related to scale problems at various
of information contained in an abstract submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the International Petroleum Technology Conference production system components.
and are subject to correction by the author(s). The material, as presented, does not necessarily
reflect any position of the International Petroleum Technology Conference, its officers, or
members. Papers presented at IPTC are subject to publication review by Sponsor Society Excessive water production can be resulted from either a
Committees of IPTC. Electronic reproduction, distribution, or storage of any part of this paper
for commercial purposes without the written consent of the International Petroleum Technology problem of the well (mechanical failure) or other reasons
Conference 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 related to the reservoir like Water channelling from water
acknowledgment of where and by whom the paper was presented. Write Librarian, IPTC, P.O. table to the well through natural fractures or faults, water
Box 833836, Richardson, TX 75083-3836, U.S.A., fax +1-972-952-9435.
breakthrough in high permeability zones, or water coning.

In general, water production problems related to wells

Abstract
integrity easier to solve and it gets more complicated to
Many oil fields in Oman are developed with horizontal wells
control water production if it is related to the reservoir
to maximize productivity and develop wider drainage areas for
characteristics. Various water control techniques were
more cost effective recovery. Premature water breakthrough
developed to shut-off or reduce excessive water production.
either from water injectors or from water aquifer reduces the
However, the rate of success of water shut-off jobs is still
wells profitability because of both reduction in net oil rate and
considered to be low. In some publications, that was reasoned
additional cost for water handling. To determine the best
as the mechanisms that contribute to excessive water
solution to shut-off, source and nature of the water entries
production are not well understood (1). Water coning,
must be well identified.
multilayer channeling and near wellbore problems are the
main three contributors to excessive water production (2).
The flow dynamics and fluid entry mechanisms in horizontal
Obviously, the understanding of excessive water production
wells are complex and identification of water inflow zones is
mechanism and identifying the water entry in the well are the
challenging even when using the best production logging
two major factors make the shut-off job successful.
technology available today. Traditionally, when the water cut
increase becomes abnormally high, production logging tool
The common practice for many operators is using production
(PLT) is run to identify water inflow zones. However
logging tools (PLT) to define water entries and then select the
interpretation of PLT logs in horizontal wells is not an easy
shut-off technique and design the job. It is very important to
task for log analysts because of the complex flow dynamics
notice that the high technology PLT available in the market
and logging tool’s limitations in measuring downhole fluid
still have some limitations in horizontal wells due to complex
velocities and fluid holdups coverage across the borehole.
fluid entry mechanisms and flow dynamics of multi-phase
flow in the wellbore. To aid understanding excessive water
This paper discusses the limitations of PLT and focuses on
mechanism, several methods and techniques have been
diagnosing excessive water production problems using water-
developed. Majority of the techniques are specialized plots
oil-ratio (WOR) plots, which is commonly used for screening
such as linear water cut vs. time (3);, linear WOR plots (4),
and selecting water shut-off candidates. A 3-D simulation
semi-log WOR (5), X-plot (6), Wilhite's WOR (7), Novotny's
model was used to investigate the effect of water arial coning
method (8), log-log plots of the WOR (2), Egbe and Dulu
and channeling through fractures on WOR and WOR
method (9), Yortsos et al. method (10) .
derivative trends in vertical and horizontal wells. Simulation
results are in good agreement with field data, which are also
Use and limitation of Production Logging Tool (PLT)
presented in this paper.
PLT is one of the most common practices to identify the
source of excessive water production. The most advanced
Introduction
PLTs have very complex tools that measure instantaneous
Excessive water production is one of the major factor
fluid hold up, fluid density, flow velocity and temperature in
contribute in reduction of wells productivity. Increasing water
highly deviated and horizontal wells (11 - 13).
2 IPTC 11958
There are some factors influence the conclusiveness of PLT
results such as:
1. Well completion and integrity problems: In slightly
inclined wells, fluid segregations are dependent on
well deviation angle. For pipe deviations between 80
and 90 degrees from the vertical, the hold-up can
vary up to more than 60% and the slip velocity
between oil and water can reverse its sign (14).
2. Well accessibility and PLT conveyance (coil tubing
vs. wireline): The presence of the PLT tool with
coiled tubing in the well bore can influence the flow
rate and pressure measurements. In low-conductivity
horizontal wells, the PLT may not indicate the low-
permeability zones along the well (15).
3. Fluid segregation and flow dynamics in horizontal
wells: In the case of having gas in addition to oil and
water, (Figure 1), fluid segregation and turbulent flow
can make interpretation of the PLT data very
difficult. For horizontal wells, when the deviation is
below 90 degree, the gas usually flows as slugs.
When the deviation is above 90 degree, the flow is
stratified. For vertical and deviated wells, the flow
can be bubbly, plug or slug (16). The gas bubbles or
plugs are confined to the high side of the wellbore
(annular flow) for relatively small gas entries. The
same is true for gas wells, with a small liquid entry
(17)
. The PLT tools are centred in the wellbore and can
easily miss such flow since the tips of probes (where
measurements are taken) cannot make physical
contacts with such film flow. Also,
4. Horizontal well with high-conductivity features: The
interest in the wellbore flow-rate profile disappears if
it is known that the horizontal well displays high-
conductivity features. In such case, the only visible
use of the PLT is to determine the low-productivity
zones along the well, which in some cases is affected
by the tool presence in the wellbore (18).
5. Data interpretation related issues: The complicity of
the interpretation of the PLT data will possibly lead
to wrong conclusions.
Still there are many other factors that can affect the accuracy
of the PLT, including borehole environment, severe corrosion,
scale, mechanical obstructions, formation rock types and
presence of shale and washouts in case of open hole (19).
Study Objectives
The water-oil ratio plots (WOR and WOR') are popular in
diagnosis of excessive water production mechanism. These
plots were proposed by Chan (2) in 1995 based on a single
vertical well simulation model. In this study we have
investigated the applicability of these plots for horizontal
wells. A 3-D simulation model was developed to study
different excessive water production scenarios and establish
WOR and WOR’ diagnostic plots. These plots were applied to
diagnose water conformance problems in well from an Omani
field. The model properties are attached as appendix (I)
Methodology
The numerical model is a two-phase (oil and water) single-
well model and has a producer located at the centre of the
model.
The water-aquifer is represented by one large cell at the
bottom of the oil grid cells. To simulate coning in vertical
wells, two scenarios are modelled. One model has a single
well that is completed across all reservoir layers and has a
strong water support from an active bottom aquifer. The
second scenario is similar to the first one except that there is
no bottom aquifer support. This model represents the ideal
case for channeling/fractures behaviours. In both models,
immobile initial water saturation was used.
In the numerical models, a channel is represented by a streak
of thin reservoir layer (layer two (Z =2)) that has a very high
permeable (80000 mD in X direction). The permeabilities in Y
and Z directions are 800 mD and 400 mD, respectively. The
channel layer is 300m × 300m × 1m in the X-Y plane. The
aquifer flux through this channel was set constant at 30
SM3/day, entering from one side towards the vertical producer
well.
To simulate coning in horizontal wells, a two dimensional
model with smaller cell size was used. Weak bottom aquifer
support is used in the model. Similar well control and internal
diameter to vertical well model are used.
The fracture in the horizontal well is modelled from edge at
layer three (Z =3), X = 30 and between Y = 1 and Y= 60, with
permeability in X and Y directions equals to 80000 mD and in
the Z direction equals to 40000 mD. The fracture zone is 5m ×
300m × 1m in the X-Y plane. The water flux is only in the
fracture zone in Y direction towards the centre of the wellbore
and has a constant rate of 5 SM3/day. The well is completed in
layer three (Z = 3).
Results and discussion
Vertical well coning with strong bottom aquifer support
The WOR (shown in upper side of Figure 2) increases with
time as a result of continuous high pressure support from the
bottom aquifer. All the water produced by the well is replaced
immediately by the recharge from the aquifer. As a result, the
WOR trend does not reach a plateau and the WOR derivative
(WOR’) decreases linearly. This behavior is expected as a
result of the high bottom aquifer support.
Vertical well coning with no bottom aquifer support
The WOR, shown in upper side of Figure 2, increases at
the beginning as the case with high pressure bottom aquifer
support scenario. After some time; here at t § 60 days; the
WOR trend reaches a plateau. This time depends on the drive
energy provided by oil expansion and the drawdown of the
well. For high oil expansion and low well drawdown, this
time increases. The WOR derivative decreases until it curves
down when the system starts to loss its drive energy. This
behavior is expected as a result of no bottom aquifer support.
IPTC 11958
WOR and WOR' results obtained earlier, are verified
against published work (2) for coning in vertical well. As can
be seen in bottom of Figure 2, there is a good match between
our results with no aquifer support and published work (2).
Although, in previous work (2), the model properties were not
sated, it was possible to get the same trends using our simple
model. It is correct to say that the aquifer support that was
used in previous work is very weak. The difference in the
magnitude of the WOR and WOR’ between our model results
and previous model is due to the difference in the properties
between the two models.
Vertical well channeling/fracture
Upper part of Figure 3 is for the period of water
breakthrough and after. Both WOR and WOR' increase
gradually. This is caused by the first layer breakthrough. At
about 381 days (about 100 days in upper part of Figure 3), a
second trend is observed. This is caused by the second layer
breakthrough. As more layers breakthrough, WOR' becomes
steeper. At some time later, the WOR' starts to decline as sign
of coning. That is expected as most layers are depleted and
they act as an edge aquifer.
As shown in Figure 3, the channel behaviour in our model
is similar to previous published work (2), bottom part of Figure
3. It appears that the previous model has movable water at
initial conditions (constant WOR § 0.1). When the first
channel breakthrough occurs, the WOR increases gradually
and also the derivative. Another trend is observed at t = 100
days when the second channel breakthrough. This time the
WOR is less steep. The same behaviour is observed in our
model. Initially, the model has no movable water. The water
breakthrough started later at about t = 286 days and as the
second layer breaks through, another trend is observed. In our
model, only one permeable layer is simulated, all other layers
have the same rock properties. Whereas in previous work (2),
layers of different permeabilities were modelled. Our model is
closer to fracture behaviour than multi channeling as the case
with previous model. Nevertheless, our model showed that one
permeable layer can cause the adjacent layers to behave as if
they have higher permeability than the other layers.
To verify the findings with examples from the field, three
vertical wells were studied. It was found from the field static
model that these wells crossed faults or fractures that
contributed to the excessive water production or completed
across layers of high permeability contrast. All these wells are
mechanically lifted with beam pump. The beam pump
parameters were checked to make sure that the pump is
operating in all times and its speed does not vary in a way that
affects the production data collected at surface. Also,
production data were cleaned from anomalies. These wells did
not have work-over history like change of completion, re-
perforation, wire wrap screen back-flush, or any job that can
alter the production behaviour. The WOR and WOR' of these
wells compared to our model are shown in Figure 4.
Horizontal well coning
When water has just reached the well’s perforations, the
WOR' is decreased as more water is produced. Water breaks
3
into the adjacent perforations and WOR' shows small peak.
However, the general trend continues decreasing in the WOR'.
At some other time, the cone stretches and the WOR' shows
smooth transition period until a peak is reached at which
WOR' decreases slowly where the next perforation watered
out. Finally, when water has reached all the perforations, the
cone will stretch horizontally. The cone is now acting as a
water channel until WOR' reached a maximum at about t =
550 days (313 days in upper left side of Figure 5). However,
since the reservoir pressure decreases, the WOR' decreases
afterward. Overall trends show a general decrease in the
WOR'.
The same approach as in vertical wells verification was
followed to check our results. The horizontal wells in the field
of our interest were carefully chosen based on available
history data. All horizontal wells presented here are having no
major work-over activities. All bad data due to wrong
production test or deferred oil activities are cleaned. The field
model was checked to know if there are fractures crossing
these wells. Any other data like time-lapse logs are also used
to derive the right conclusion. In addition, the pump
parameters were checked to make sure that the well is on 24
hour timer and the pump speed does not vary.
Three horizontal wells are verified for their trends against
our model findings for coning in horizontal wells. These wells
are chosen carefully where there is no fracture or fault
crossing them. They are completed in the same formation unit
and did not experience any possible channeling.
Figure 5 shows the WOR and WOR' for the three wells
compared to results from our model for horizontal coning
scenario. All three wells WOR and WOR' show good match
compared to our model results.
Horizontal well channeling/fracture
From the model results, we can see that the water
breakthrough in the horizontal well occurs faster than in the
vertical well. As more water get produced, the WOR and
WOR' continue to rise (upper part of Figure 6). At about 233
days, the WOR' decreases due to depleted zones acting like
water cone from the edge.
Figure 6 shows a good match obtained between the model
results of WOR and WOR’ and the actual field data.
Conclusions
Using simple simulation model of defined static and dynamic
properties, it was possible to get similar trends of WOR and
WOR' to those of published work (2) on single vertical well.
Three different scenarios were tested. Two scenarios were for
coning and one for channeling/fracture. The conning model
that has no support from a bottom water aquifer matches very
well the published WOR and WOR'(2). The third scenario is
about channeling/fracture in single vertical well which also
matches well the previous work for single vertical well.
Three vertical wells from an Omani oil field were selected
after screening to verify their trends against our findings for
4 IPTC 11958

channeling/fracture in vertical wells. All three wells showed (7) G. P, Wilhite: Waterflooding, Text Book Series, SPE.,
good agreement with our model. Hence, the model was further Richardson, TX (1986) 3, Chapter 5.
developed to study coning and channeling/fracture in single (8) R. J., Novotny: “Matrix flow evaluation technique for water
horizontal well for the first time. The same approach was used control application”, SPE paper 030094 presented at the
European Formation Damage Conference held in the Hague, the
to verify the trends of WOR and WOR' for the different Netherlands, May, 15-16, 1995.
scenarios. Four horizontal wells were chosen after proper (9) ThankGod Egbe and Dulu Appah; Petroleum and Gas
screening for verification. The WOR and WOR’ trends for all Engineering Department, University of Port Harcourt, "Water
four wells matched our findings for the single horizontal well Coning Diagnosis using Spectral Analysis", Society of
simulation model. Petroleum Engineers Inc, 2005, SPE 98816
(10) Y. C.,Yortsos, C.,Youngman, Y., Zhengming and P.C, Shah:
The good match between the model trends and the actual “Analysis and Interpretation of Water Oil Ratio in Water
trends from well history production suggests that these plots Floods”, SPE Journal Vol. 4, No. 4, Dec 1999. pp 413 -425
(11) H. Dandarawy, A. AlMutairi, G. Abdouche, and O. Khedr,
can be used to diagnose the excessive water production
Zadco, and A. Elkadi, Schlumberger, "A New Production
problems in vertical and horizontal wells. Logging Tool Allows a Superior Mapping of the Fluid
The WOR and WOR' trends observed are not affected if the Velocities and Holdups Inside the Wellbore", Society of
well is produced by electric submersible pump (ESP) or beam Petroleum Engineers, Inc., 2005, SPE 93556
pump (BP). This is valid when the BP parameters (running (12) H. Farran, Schlumberger, J. Harris and S. Al Jabri, Petroleum
time and speed) are almost constant during the life of the well Development Oman, and R. R. Jackson, S. Al Khayari, and T.
and the ESP has relatively constant intake pressure through the Thomas, SPE, Schlumberger, "An Integrated Approach for
well’s life. Evaluating and Characterizing Horizontal Well Inflow and
Productivity in Heterogeneous Carbonate Reservoirs",
International Petroleum Technology Conference, 2005, IPTC
Nomenclature
10492
BHP: Bottom Hole Pressure (13) A. AL-Amer, B. A. AL-Dossary, Y. A. AL-Furaidan, and M. K.
Bo: Oil formation volume factor Hashem, SPE, Saudi Aramco, "Tractoring – A New Era in
BP: Beam Pump Horizontal Logging for Ghawar Field, Saudi Arabia", Society of
Bw: Water formation volume factor Petroleum Engineers Inc, 2005, SPE 93260
Cr: Rock compressibility (14) Krimo Laieb/Sonatrach-PED, and Djebbar Tiab/U. of
Cw: Water compressibility Oklahoma, “Design and Performance of Miscible Flood
ESP: Electrical Submersible pump Displacement”, Society of Petroleum Engineers, 2001, SPE
Kh: Horizontal Permeability 70021
Kro: Oil relative permeability (15) R. M. McKinley: "Production Logging", Exxon Production
Research Co, 1982, SPE (10035)
Krw: Water relative permeability (16) Justin Rounce, Chris Lenn, Schlumberger Technical Services,
Kv: Vertical Permeability Gerard Catala, Schlumberger Riboud Product Center:
Pc: Capillary pressure "Pinpointing Fluid Entries in Producing wells" presented at the
PLT: Production Logging Tools 1999 SPE Middle East Oil Show held in Bahrain, 20–23
PVT: Pressure Volume Temperature February 1999, SPE 53249
WOR: Water Oil Ratio (17) J,R, Fanchi and R.L. Christiansen, Marathon Oil Co,
WOR': Derivative of Water Oil Ratio “Applicability of Fractals to the Description of Viscous
Fingering”, Society of Engineers, 1989, SPE 19782
(18) B. E. Theron and T. Unwin, Schlumberger Cambridge Research:
References
"Stratified Flow Model and Interpretation in Horizontal wells",
(1) James Pappas, SPE, Fina Oil & Chemical Co., Prentice Creel,
SPE 36560, presented at the 1996 SPE Annual Technical
SPE, and Ron Crook, SPE, Halliburton Energy Services,
Conference and Exhibition held in Derver, Colorado, U SA, 8-9
“Problem Identification and Solution Method for Water Flow
October 1996
Problems”, Society of Petroleum Engineers Inc., 1996, SPE
(19) E. Ozkan, Colorado School of Mines, C. Sarica, SPE, The
35249
Pennsylvania State University, M. Haci, Drilling Measurements,
(2) K. S. Chan: "Water Control Diagnostic Plots", presented at the
Inc. :"Interpretation of Horizontal-Well Production Logs:
SPE Annual Technical Conferences & Exhibition, Dallas, USA,
Influence of Logging Tool", presented at the 1998 SPE
22-25 October, 1995, SPE 30775
International Conference on Horizontal Well Technology held
(3) R-N., Hwan: "Numerical Simulation Study of Water Shutoff
in Calgary, Alberta, Canada, 1-4 November 1998, SPE 50395
Treatment Using Polymers" paper SPE 25854, presented at the
1993 SPE Rocky Mountain Regional/Low-Permeability
Reservoirs Symposium, Denver, CO, April 12-14.
(4) R. V., Higgins and R. V., Leighton,: "Matching Calculated With
Actual Waterflood Performance With Estimation of Some
Reservoir Properties", paper SPE 4412, presented at the 1973
SPE Rocky Mountain Regional Meeting, Casper, WY, May 15-
16.
(5) N., Mungan: "A Theoretical and Experimental Coning Study",
SPEJ (June 1975) 247-254.
(6) I., Ershaghi and D, Abdassah.: "A Prediction Technique for
Immiscible Process Using Field Performance Data", JPT (April
1984) 664-670.
IPTC 11958 5

Appendix (I) General Model Properties Appendix (II)

Tables Figures

Well type Grid Completion Control Well

interval method diameter
Vertical X,Y Z1, Z2 BHP (2.7 0.1143 m
30, 30 1, 20 barsa) (4 1/2")
Table 1: Well properties for strong bottom aquifer
support in vertical well-coning

Grid Area Length Porosity Permeability

X, Y, Z
20000 m2 2000 m 27 % 200 mD
60, 60, 41
Table 2: Aquifer properties for strong bottom aquifer
support in vertical well-coning

Initial Pressure 102.9 bars (at datum of 730m)

Horizontal Permeability (Kh) 800 mD
Vertical Permeability (Kv) 400 mD
Porosity 27 %
Hydrocarbone Saturation (Sh) 64 %
Oil density 929 kg/m3 (16)
Figure 1: Principle gas-liquid flow regimes
Water density 1006 kg/m3
Relative water volume (Bw) 1.02 rm3/sm3
Water compressibility (Cw) 4.35e-05 1/bars 100

Water viscosity 0.54 cP

The rock compressibility (Cr) 5.8e-05 1/bars 10

Table 3: Rock and fluid properties WOR (Aquifer Support)

WOR' (Aquifer Support)

WOR (No Aquifer Support)

1
WOR' (No Aquifer Support)
WOR'
WOR

Pressure (barsa) Bo (rm3/sm3) Viscosity (cP) 0.1

6.40 1.012 188

17.24 1.010 213 0.01

34.47 1.008 254

51.70 1.006 294
0.001
69.00 1.004 335 1 10
Time (Days)
100 1000

86.19 1.002 375

103.90 1.000 417
137.90 0.996 496 Coning:
Table 4: Model PVT data for dead oil Strong bottom aquifer support
No pressure support

Sw Krw Kro Pc (bars)

0.080 0.000 1.000 0.303
0.081 0.000 0.996 0.289
0.083 0.000 0.989 0.274
0.086 0.000 0.978 0.260
0.090 0.000 0.963 0.245
0.100 0.000 0.928 0.231
0.140 0.000 0.793 0.217
0.180 0.000 0.670 0.202
0.220 0.000 0.559 0.188
0.260 0.000 0.459 0.173
(Swc) 0.300 0.001 0.371 0.159
0.340 0.003 0.293 0.144
0.380 0.005 0.226 0.130
0.420 0.010 0.169 0.116
0.460 0.016 0.121 0.101
0.500 0.026 0.083 0.087
0.540 0.039 0.052 0.072 Coning: Chan’s paper (2)
0.580 0.058 0.030 0.058
0.620 0.084 0.014 0.044 Figure 2: Vertical well: WOR and WOR' coning trends
0.700 0.117 0.000 0.000 compared to published results
Table 5: Model relative permeability and capillary
data for dead oil
6 IPTC 11958

WOR' 0.1
WOR

0.01

0.001

0.0001
1 10 100 1000
Time (Days)

WOR WOR'

WOR and WOR' trend for channeling/fracture scenario in vertical well

Channeling: Chan trend of WOR and WOR' (2)

Figure 3: Vertical well: WOR and WOR' channeling/fracture trends compared to published results
IPTC 11958 7

1 1000

100

10
0.1

0.1
WOR'

WOR'
WOR

WOR
0.01
0.01

0.001

0.001
0.0001

0.00001

0.0001 0.000001
1 10 100 1000 1 10 100 1000 10000
Time (Days) Time (Days)

WOR WOR' WOR WOR'

WOR and WOR' trends for channeling/fracture scenario in WOR and WOR' for X-2
vertical well

1000 100

100
10

10
1

1
0.1
WOR'

WOR'
WOR

WOR

0.1

0.01
0.01

0.001
0.001

0.0001
0.0001

0.00001 0.00001
1 10 100 1000 10000 1 10 100 1000 10000
Time (Days) Time (Days)

WOR WOR' WOR WOR'

WOR and WOR' for X-1 WOR and WOR' for X-3
Figure 4: WOR and WOR' for three vertical wells compared to model’s trends
8 IPTC 11958

100 100

10 10

1 1
WOR'

WOR'
WOR

WOR
0.1 0.1

0.01 0.01

0.001 0.001
1 10 100 1000 1 10 100 1000 10000
Time (Days) Time (Days)

WOR WOR' WOR WOR'

WOR and WOR' trends for horizontal well coning scenario, the WOR and WOR' for X-5
time step is shifted -237 days

100 100

10 10

1 1
WOR'

WOR'
WOR

WOR

0.1 0.1

0.01 0.01

0.001 0.001

0.0001 0.0001
1 10 100 1000 10000 1 10 100 1000 10000
Time (Days) Time (Days)

WOR WOR' WOR WOR'

WOR and WOR' for X-4 WOR and WOR' for X-6
Figure 5: WOR and WOR' for three horizontal wells compared to model’s trends
IPTC 11958 9

10

Fracture
0.1
WOR'
WOR

0.01

0.001

Coning
0.0001
1 10 100 1000
Time (Days)

WOR WOR'

WOR and WOR' trends for horizontal well channeling/fracture scenario

1000.00

100.00

10.00

1.00

0.10

0.01

0.00
1 10 100 1000 10000

WOR WOR'

WOR and WOR' for X-3H2

Figure 6: WOR and WOR' for X-3H2 compared to model’s trend