See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/249883303

Potential Impacts of Integrated Oil and Gas Plant on

Ambient Air Quality

Article in Energy & Environment · July 2009

DOI: 10.1260/095830509788066394

CITATIONS READS

7 168

2 authors:

Jacob Sonibare Precious Ede

Obafemi Awolowo University Rivers State University
79 PUBLICATIONS 1,117 CITATIONS 59 PUBLICATIONS 290 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Precious Ede on 11 July 2014.

The user has requested enhancement of the downloaded file.

 331

POTENTIAL IMPACTS OF INTEGRATED OIL AND GAS

PLANT ON AMBIENT AIR QUALITY
Jacob Ademola Sonibare1 and Precious Nwobidi Ede2
1Environment Department, Shell Petroleum Development Company, Port Harcourt, Nigeria.
2Institute of Geosciences and Space Technology,

River State University of Science and Technology, Port Harcourt, Nigeria.

ABSTRACT
The ISCST-3 model has been used to study the ground level concentrations of
carbon monoxide (CO), oxides of nitrogen (NOX), particulate matter less than 10.0
microns in diameter (PM10) and sulphur dioxide (SO2). The purpose was to predict
air quality effects from an integrated oil and gas project (IOGP) proposed for the
Niger Delta, Nigeria. The emissions of 12 elevated point sources for criteria air
pollutants were considered with eight different scenarios. Both natural gas-fired
and diesel (AGO) fired equipment were considered for the identified emission
sources. Five years of hourly meteorological observations were used in the
prediction of the ground level concentration of the pollutants.
The model outputs show a general maximum ground level concentration between
0.5 – 3.5 km, with easterly and north-easterly receptors having the greatest
concentration. The pollutants emitted in the greatest concentration in their order of
magnitude were CO, NOx and PM10. Maximum ground level concentration for 1, 8
and 24 – hr periods for CO were below regulatory standards (less than 10 %) for
scenarios 1 – 4 when the fuel is natural gas. The 8 – hr ground level concentrations
were about 1.3 folds of the World Bank Standard in scenarios 5 – 8 when the fuel
sources are switched to AGO. However in these scenarios, the Nigerian environment
and petroleum ministries’ emission limits were not breached for all the averaging
periods. Modelled NOX concentrations were significant in all the scenarios except 3
and 4. Though NOX concentrations were within the standards for all the averaging
periods in scenarios 1 – 4, except the 1 – hr limit that was breached in scenarios 1
and 2, these concentrations were about 1.1 – 2.8 folds in all the averaging periods of
all the standards in scenarios 5 – 8. The predicted PM10 and SO2 concentrations are
all below the guideline limits. If the use of AGO is completely avoided in the entire
life of the proposed IOGP, the air quality of its immediate vicinity can be protected.
Furthermore, the declaration of a radius 1.75 km natural forest buffer around the
plant and continuous monitoring of ground level concentrations of CO, NOX, PM10,
and SO2 may further assist in guaranteeing the safety of people in terms of air
quality. The study shows the need for policy formulation that will integrate
comprehensive cumulative impacts assessment of air quality in the environmental
laws regulating the gas development projects in Nigeria.

1PermanentAddress: Environmental Engineering Research Laboratory, Chemical Engineering Department,

Obafemi Awolowo University, Ile-Ife. Nigeria. [email protected]; corresponding author.
332 Energy & Environment · Vol. 20, No. 3, 2009

INTRODUCTION
Nigeria is presently a major emitter of criteria air pollutants from routine flares in the
upstream petroleum operations. In a previous study [1], gaseous emissions from this
sector were predicted to be 1.72E+05, 5.48E+05, 1.611E+05, 2.252E+05, and
1.16E+03 kg for CO2, CO, NO, NO2, and SO2, respectively from an annual flaring rate
of 25838.35 m3. If the on-going efforts at eliminating routine flares from the upstream
petroleum operations in Nigeria are sustained, the current levels of air pollution due to
flaring activities in the Niger Delta may be significantly reduced. For the anticipated
gains of improved air quality to be accomplished, potential air pollution sources from
the gas development projects proposed for the area must be identified. This will assist
in incorporating adequate control measures at the Front End Engineering Design
(FEED) stages. In this study, emission inventory of a gas processing plant of an
Integrated Oil and Gas Project (IOGP) in the Niger Delta was undertaken. Dispersion
modelling was applied to the estimated criteria air pollutants for the purpose of ground
level concentration determination around adjoining settlements.
The Industrial Source Complex Short-Term (ISCST-3) Version 5.2 air dispersion
modelling tool was used to study the ground level concentrations of CO, NOX, PM10,
and SO2 from the IOGP using some operation scenarios. The ISCST-3 is an air
dispersion model from the United States Environmental Protection Agency (US EPA)
designed to support regulatory activities [2]. It uses the basic Gaussian equation to
estimate the concentration at a particular location, which is given by Eqn. (1):

Q  Y2   (Z − he )2 
C= exp − 0.5 2  exp − 0.5 

(1)
2psysz  sy   sz 2 

Where Q is the emission rate (g/s), σy and σz are the dispersion coefficients in
crosswind and vertical directions (m), Y and Z are the transverse and vertical distance
from the source (m), u is the wind speed (m/s), and he is the effective stack height (m).
Each hour is considered as a steady-state period and it requires input database
including: detailed characteristics of source emission and dimension (pollutant
emission rate, flue gas exit velocity and temperature, stack height and top inner
diameter) and hourly boundary layer meteorological data (wind speed and direction,
ambient temperature, atmospheric stability class, mixing height). The ISCST has been
used in air quality studies in the past [3,4,5]. Its capabilities include: evaluation of
output concentration from a source, deposition and depletion studies, and source
apportionment analyses. This model was applied to predict NOX, SO2, CO, and PM10
concentrations on a 1-hourly, 24-hourly and annual average basis to facilitate
comparison with ambient air quality standards. The prescribed standards are the
Environmental Guidelines and Standards in the Petroleum Industry (ENGASPIN) of
Nigeria’s Department of Petroleum Resources, the Federal Ministry of Environment,
Nigeria [6], the World Bank [7].
Global concern over the environmental status of the Niger Delta has never been
higher in the public mind because of its strategic location in the crude oil supply chain.
Implementation of control measures that will guarantee availability of safe air quality
Potential impacts of integrated oil and gas plant on ambient air quality 333

in the area can only be achieved if possible sources of air pollutants are identified in
every major project proposed for the area with potential ground level concentrations.
An additional benefit of the control of these air pollutants is the prevention of possible
secondary air pollutants that can be formed within the atmosphere. These are some of
the issues set to be addressed by the emission inventory and dispersion modelling
carried out in this study.

Methodology
Emission Inventory
This study considered emissions from equipment associated with the proposed plant
which require combustion activities for operation. The equipment identified were
within the Central Processing Facility (CPF) of the IOGP and include fired heaters,
gas turbine driven compressors and gas turbines required for power generation. The
CPF will have flares that will operate intermittently (Table 1). For some of these
equipments, emission rates were obtained from the manufacturers and for others,
estimates were carried out using emission factors approach (Eqn. 2) with emission
factors obtained from AP-42 [8].

E = EF x AR x
(100 − Co ) (2)
100
Where: E is the air pollutant emission rate from the equipment (kg/s); EF is the
emission factor (kg/unit); AR is the activity rate (unit/s); and Co is the control
efficiency (%).
Those with estimates include anticipated emissions from the compressors and
power generators when diesel (AGO) is used as alternate fuel. Table 2 summarizes the
anticipated emissions from identified sources and these serve as key inputs into the
modelling activity.

Table 1. Major Sources of Emissions and Parameters used as

Input to the Air Dispersion Model

S/No Emission Source Fuel Consumption (kg/hr) Stack Parameters Location

Gas Fuel AGO Ht. (m) Dia. (m) Temp. (K) Exit Vel. Xp(m) Yp(m)
(m/s)
1. Fired Heater 1 3865 – 20 1.5 493 12.5 0 507
2. Fired Heater 2 3865 – 20 1.5 493 12.5 0 492
3. Atmospheric Flare 1 2798 – 110 1.25 1273 20 492 435
4. Atmospheric Flare 2 1393200 – 110 1.25 1273 20
5. AG Compressor 1 2499 2774 11.4 11.4 780 14 101 163
6. AG Compressor 2 2499 2774 11.4 11.4 780 14 159 225
7. AG Compressor 3 2499 2774 11.4 11.4 780 14 38 163
8. AG Compressor 4 (future) 2499 2774 11.4 11.4 780 14 159 163
9. AG Compressor 5 (future) 2499 2774 11.4 11.4 780 14 101 225
10. Power Generator Turbine 1 2499 2774 11.4 11.4 758 22 101 225
11. Power Generator Turbine 2 2499 2774 11.4 11.4 758 22 101 163
12. Power Generator Turbine 3 2499 2774 11.4 11.4 758 22 38 163
334 Energy & Environment · Vol. 20, No. 3, 2009

Table 2. Computed Air Pollutants from the Identified Sources

Source From Natural Gas From AGO

NOX CO SO2 PM10 NOX CO SO2 PM10
Fired Heater (g/s) 2.1 1.77 – 0.16 34 1299 1.7 2.1
Atmospheric Flare (g/s) 624 3400 – – – – – –
Gas Turbine-driven AG Compressor (g/s) 5.6 2.7 – – 24 932 1 1.5
Power Generator Turbine (g/s) 7.3 1.8 – 0.3 24 932 1 1.5

Dispersion Modelling
For a comprehensive investigation of the dispersion modelling in this study, eight
different scenarios of facility operation and fuel types evaluated in emission sources were:
Scenario 1: Described as the “worst situation”, it was assumed that all the
associated point sources of emissions are working in the CPF and at 100 % fuel
consumption rate. However in this scenario, natural gas is assumed to be the fuel of
consumption.
Scenario 2: The plant is designed to operate with 2 heaters, 3 compressors, and 2
power generators at the initial stage with the atmospheric flare intermittently in use for
emergency purposes. To estimate the ground level concentrations of air pollutants at
this stage, this scenario was considered with natural gas as the fuel.
Scenario 3: Since the atmospheric flare is designed for emergency situations during
the life of the plant, this scenario assumed all other identified point emission sources
operating but with the gas flare not in use. As in scenarios 1 and 2, the fuel considered
for combustion is natural gas.
Scenario 4: In this scenario, the operating conditions were assumed similar to that
of scenario 2 but with the gas flares shut down. This scenario is important because it
looks like the likely operating condition in the plant since the gas flare is designed for
emergency purposes. As in scenario 2, the facilities are running on natural gas.
Scenario 5: As in scenario 1, another “worst situation” was assumed with all the
associated point sources of emissions in the CPF working at 100 % fuel consumption
rate. However in this scenario, it is assumed that the heaters, the compressors, and the
power generators are fuelled with AGO. The atmospheric flare also operates at full
capacity.
Scenario 6: During this operating condition of the plant, one of the three power
generators will be on stand-by while the other two compressors will be brought
upstream at full capacity at the later stage of the plant. This scenario was considered
with the heaters, the compressors, and the power generators fuelled with AGO.
Scenario 7: This scenario assumed all other identified point emission sources
operating but with the gas flare not in use. This is similar to scenario 3 in operating
conditions but for the fuel considered for combustion which is AGO.
Scenario 8: In this scenario, the operating conditions of the plant were assumed to
be similar to scenario 2 but with the gas flares shut down. This scenario is considered
important because of several problems associated with pipelines delivery of fuel in
Nigeria [9]. The facilities are working on AGO. Table 3 summarizes all these
scenarios.
Potential impacts of integrated oil and gas plant on ambient air quality 335

Table 3. Emissions Cases of Each Modelling Scenario

Scenario Emission Sources Number Included Contaminant Considered

1. Fired Heaters 2 – NG Fired CO, NOX, PM10
Atmospheric Flare 1 – NG Fired
Gas Turbine-driven AG Compressor 5 – NG Fired
Power Generator Turbine Drivers 3 – NG Fired
2. Fired Heaters 2 – NG Fired CO, NOX, PM10
Atmospheric Flare 1 – NG Fired
Gas Turbine-driven AG Compressor 3 – NG Fired
Power Generator Turbine Drivers 2 – NG Fired
3. Fired Heaters 2 – NG Fired CO, NOX, PM10
Gas Turbine-driven AG Compressor 5 – NG Fired
Power Generator Turbine Drivers 3 – NG Fired
4. Fired Heaters 2 – NG Fired CO, NOX, PM10
Gas Turbine-driven AG Compressor 3 – NG Fired
Power Generator Turbine Drivers 2 – NG Fired
5. Fired Heaters 2 – AGO Fired SO2, CO, NOX, PM10
Atmospheric Flare 1 – NG Fired
Gas Turbine-driven AG Compressor 5 – AGO Fired
Power Generator Turbine Drivers 3 – AGO Fired
6. Fired Heaters 2 – AGO Fired SO2, CO, NOX, PM10
Atmospheric Flare 1 – NG Fired
Gas Turbine-driven AG Compressor 3 – AGO Fired
Power Generator Turbine Drivers 2 – AGO Fired
7. Fired Heaters 2 – AGO Fired SO2, CO, NOX, PM10
Gas Turbine-driven AG Compressor 5 – AGO Fired
Power Generator Turbine Drivers 3 – AGO Fired
8. Fired Heaters 2 – AGO Fired SO2, CO, NOX, PM10
Gas Turbine-driven AG Compressor 3 – AGO Fired
Power Generator Turbine Drivers 2 – AGO Fired

Meteorological Data
An important input into ISCST-3 model is the meteorological information and in this
study, a five-year period of representative hourly meteorological data was considered
for the study area. Along the coast of Nigeria, meteorological analysis of the
prevailing wind patterns revealed that the wind direction persists from the southwest
for most of the year [10]. Similarly, a five year period considered for Port Harcourt,
the nearest synoptic meteorological station to the study location, agrees with these
(Figure 1). The radial arms of the windrose represent the directions from which the
wind is coming.
Lack of upper air observations in the Port Harcourt weather station made it
impossible to have data whose input would have been useful in this study. However,
there are two locations nearby at Douala, Cameroon and Cotonou, Benin with good
data recovery rates. While the location at Douala is on the west-facing coast of Africa
336 Energy & Environment · Vol. 20, No. 3, 2009

with winds showing a prevailing westerly direction, the Cotonou location is on a

south-facing coast with the winds having prevalence for a southwesterly direction as
the study area. Figure 2 shows the windrose based on five-year meteorological
observations at Cotonou. This is consistent with the winds observed at the study area.
Also for other meteorological parameters that affect dispersion, such as temperature
and cloud cover, Cotonou meteorological observations were used in this study.

1 m /s

N0
9
2.5 m /s
4.5 m /s
6.7 m /s
NW NE 9.3 m /s
> 9.3 m /s

C alm = 24.3 %

W
0
8
1

E0

SW SE
S
0
7
2

Figure 1. Windrose for the Study Area

Figure 2. Windrose for Cotonou, Benin

Potential impacts of integrated oil and gas plant on ambient air quality 337

The study area has two predominant surface wind sources, which include winds
from the southerly directions coming across from the Gulf of Guinea and overland
winds coming principally from over freshwater swamps. Tree heights range generally
from 10 to 40 meters. Roughness length in this study are taken to be 0.16 and 0.0001,
respectively, for overland fetch and over water fetch; while the average tree heights as
obtained from the vegetation study of the area is 35 m. Using land use pattern
classification by Auer [11], the study area is classified as rural. All these parameters
served as input for the modelling exercise.

RESULTS AND DISCUSSION

Since investigation of the contribution of criteria air pollutants to the ground level
concentrations in surrounding settlements is the focus of this study, the location of the
nearest settlement to the plant guided in the choice of area of interest for the modelling
exercise (0 – 60 km). Figure 3 is the surface plot of the maximum predicted 24-hr
concentration of CO for scenario 1 which is the “worst case” scenario. Whenever it is
favoured by operating conditions, its predicted CO concentrations were in the range
0.27 – 273.1 _g/m3 with the maximum concentrations east and north-eastward of the
plant. In scenario 2, the predicted 24-hr concentrations of CO ranged between 0.03 and
272.9 _g/m3 with the direction of maximum concentrations following the trend in
scenario 1 (surface plot available on request). For all the three averaging periods, the
maximum concentrations were obtained around the same distance of between 1 and
1.25 km from the IOGP site. Generally the predicted ground level concentrations of
CO are far below the FMENV, EGASPIN, and the World Bank standards (Table 4 and
Table 5) while the south, south-east, and north-western flank of the plant may be
receptors to CO emissions from the plant.

Figure 3. Maximum Predicted 24-hr CO Concentrations – Scenario 1

338 Energy & Environment · Vol. 20, No. 3, 2009

Table 4. Scenario 1 Output Compared to the Standards

S/No Pollutant Averaging Max. Conc. Receptor Location for (m) % FMENV % EGASPIN % World Bank
Period (µg/m3) X Y
1. CO 1 – Hr 2934.4 -1060.7 -1060.7 – 9.78 9.78
8 – Hr 894.3 -1237.44 -1237.44 3.92 – 8.94
24 – Hr 273.1 -1060.7 -1060.7 2.40 2.73 –
2. NOX 1 – Hr 549.6 -1060.7 -1060.7 – 1.4 fold –
24 – Hr 54.3 -1060.7 -1060.7 48.05–72.4 36.2 36.2
Annual 10.8 -3535.5 -3535.5 – – –
3. PM10 1 – Hr 20.6 -1237.44 -1237.44 – - –
24 – Hr 2.6 6500 0.00 – – 3.25
Annual 0.5 -3182 -3182 – – –

Table 5. Scenario 2 Output Compared to the Standards

S/No Pollutant Averaging Max. Conc. Receptor Location for (m) % FMENV % EGASPIN % World Bank
Period (µg/m3) X Y
1. CO 1 – Hr 2932.6 -1060.7 -1060.7 – 9.78 9.78
8 – Hr 893.5 -1237.44 -1237.44 3.92 – 8.94
24 – Hr 272.9 -1060.7 -1060.7 2.39 2.73 –
2. NOX 1 – Hr 545.6 -1060.7 -1060.7 – 1.4 fold –
24 – Hr 52.4 -1060.7 -1060.7 46.37–69.67 34.93 34.93
Annual 10.4 -3535.5 -3535.5 – – –
3. PM10 1 – Hr 19.9 -1237.44 -1237.44 – – –
24 – Hr 2.2 6500 0.0 – – 2.75
Annual 0.4 -3182 -3182 – – –

Due to the abundance of naturally occurring nitrogen in the atmosphere, operating

conditions play significant role in NOX formation in combustion activities. Predicted
annual maximum concentrations from the CPF indicates a range of 0.02 – 10.8 µg/m3
(Figure 4) from scenario 1 but when the facility is operated as scenario 2, this becomes
0.02 – 10.4 µg/m3 (surface plot available on request). There are no set limits for annual
NOX concentrations but when the predicted maximum 1 – hr and 24 – hr
concentrations are compared with the set limits (Table 4 and Table 5), the outcome
raises some level of concerns. At about 1 km from the plant, the 1 – hr maximum
ground level concentration of 549.6 µg/m3 is predicted while the 24 – hr maximum
ground level concentration of 54.3 µg/m3 is obtained around the same receptor
location for scenario 1. In scenario 2, these concentrations reduced to 545.6 µg/m3 and
52.4 µg/m3 for 1 – hr and 24 – hr averaging periods, respectively. For both scenario 1
and scenario 2, the 1 – hr concentrations were about 1.4 folds of the EGASPIN set
limit while the 24 – hr concentrations are about 70 % of the FMENV limit and 36 %
of both the EGASPIN and World Bank limits.
If the facility is strictly operated on natural gas during the life of the IOGP then,
particulates may not be a significant emission of the process. As shown by both
Potential impacts of integrated oil and gas plant on ambient air quality 339

scenario 1 (Figure 5) and scenario 2 (surface plot available on request), the predicted
ground level PM10 concentrations were respectively of the range 0.05 – 0.38 µg/m3
and 0.08 – 0.34 µg/m3 which were far below the World Bank limit (Tables 4 and 5).
As observed in the earlier two parameters, these maximum concentrations were
obtained at the north-easterly direction of the plant.
As expected, ground level concentrations resulting from emissions from scenario 3
and scenario 4 were lower than those from scenarios 1 and 2 (Tables 6 and 7). This can
be attributed to the decrease in the number of point emission sources from scenario 2

Figure 4. Maximum Predicted Annual NOX Concentrations – Scenario 1

Figure 5: Maximum Predicted Annual PM10 Concentrations – Scenario 1

340 Energy & Environment · Vol. 20, No. 3, 2009

down to scenario 4. The predicted maximum daily CO ground level concentrations

from scenarios 3 and 4 were of the range 0.27 – 11.1 µg/m3; 8 – hr concentrations was
17.7 µg/m3 for both and the 1 – hr concentration was about 26.4 µg/m3 from scenario
3 but from scenario 4, it reduced to 22. 3 µg/m3.

Table 6. Scenario 3 Compared to the Standards

S/No Pollutant Averaging Max. Conc. Receptor Location for (m) % FMENV % EGASPIN % World Bank
Period (µg/m3) X Y
1. CO 1 – Hr 26.4 -1750 0 – 0.09 0.09
8 – Hr 17.7 -750 0 0.08 – 0.18
24 – Hr 11.1 -750 0 0.10 0.11 –
2. NOX 1 – Hr 55 0 -2000 – 13.75 –
24 – Hr 13.1 -750 0 11.59–17.47 3.28 3.28
Annual 2.9 -1500 0 – – –
3. PM10 1 – Hr 6.5 0 -2000 – – –
24 – Hr 1.3 4242.6 4242.6 – – 1.63
Annual 0.2 -1500 0.0 – – –

Table 7. Scenario 4 Output Compared to the Standards

S/No Pollutant Averaging Max. Conc. Receptor Location for (m) % FMENV % EGASPIN % World Bank
Period (µg/m3) X Y
1. CO 1 – Hr 22.3 0 -250 – 0.07 0.07
8 – Hr 17.7 -750 0 0.08 – 0.18
24 – Hr 11.1 -750 0 0.10 0.11 –
2. NOX 1 – Hr 37.0 0 -2000 – 9.25 –
24 – Hr 13.1 -750 0 11.59-17.47 3.28 3.28
Annual 2.8 -1500 0 – – –
3. PM10 1 – Hr 4.2 0 -2000 – – –
24 – Hr 1 -750 0 – – 1.25
Annual 0.2 -1500 0 – – –

While the 8 – hr concentrations were predicted to be around 0.75 km from the

proposed plant, the 1 – hr concentrations were at 1.75 km in the x-direction for
scenario 3 but for scenario 4, it was at around 0.25 km in the y-direction. For both
scenarios 3 and 4, the predicted maximum daily concentrations were at about 0.75 km
in the x-direction of the plant. Unlike the observed trend in both scenario 1 and
scenario 2 where the 1 – hr set limit of EGASPIN was predicted to be exceeded by the
ground level NOX concentrations, whenever operating conditions favour scenario 3
and scenario 4, this limit will not be breached. The predicted concentrations from the
two scenarios were about 13.75 and 9.25 % respectively from scenario 3 and scenario
4. Similarly, the maximum 24 – hr predicted ground level concentrations of NOX from
both scenarios are about 11.59 – 17.47 % of the FEMENV limit and 3.28 % of both
the ENGASPIN and the World Bank limits. The predicted PM10 concentrations from
Potential impacts of integrated oil and gas plant on ambient air quality 341

both scenarios were not too different from what was obtained in scenarios 1 and 2.
This can be attributed to the fuel assumed for combustion, natural gas, which is known
for its very low capacity to generate particulates during combustion.
The predicted ground level concentrations of considered criteria air pollutants in
the dispersion modelling are summarized in Tables 8 – 11. Though scenarios 1 and 5
are similar in terms of identified number of combustion sources, the fuel available for
combustion are different and this might be responsible for the significant differences
in their contributed ground level concentrations for all the pollutants. In scenario 5, the
predicted ground level concentrations of CO increased to 71.24 % of the FMENV
limit from 2.4 % predicted for scenario 1. Also the 1 – hr concentrations increased
from 9.78 % to 54.19 % of both the EGASPIN and World Bank limits. In scenarios 6
– 8, the ground level concentrations wereabout 1.3 folds of the World Bank 8 – hr
limit. The predicted ground level concentrations of NOX from all scenarios 5 – 8 were
about 1.4 – 2.8 folds of all the three set 24 – hr limits of FMENV, EGASPIN, and the
World Bank. Though the predicted PM10 concentrations from scenarios 5 – 8 were still
within the set limits under consideration, they were all significantly elevated from
scenarios 1 – 4 where the point sources are fired with natural gas.

Table 8. Scenario 5 Output Compared to the Standards

S/No Pollutant Averaging Max. Conc. Receptor Location for (m) % FMENV % EGASPIN % World Bank
Period (µg/m3) X Y
1. CO 1 – Hr 16257 0 -250 – 54.19 54.19
8 – Hr 12990.8 -750 0 56.98 – 1.3 fold
24 – Hr 8121.5 -750 0 71.24 81.22 –
2. NOX 1 – Hr 672.5 -1500 0 – 1.7 fold –
24 – Hr 212.9 -750 0 1.9-2.8 fold 1.4 fold 1.4 fold
Annual 46 -1500 0 – – –
3. PM10 1 – Hr 28.2 -1750 0 – – –
24 – Hr 13.1 -750 0 – – 16.38
Annual 2.8 -1500 0 – – –
4. SO2 1 – Hr 21.2 0 -250 8.15 – –
24 – Hr 10.6 -750 0 40.77 – –
Annual 2.1 -1500 0 – – 1.68
342 Energy & Environment · Vol. 20, No. 3, 2009

Table 9. Scenario 6 Output Compared to the Standards

S/No Pollutant Averaging Max. Conc. Receptor Location for (m) % FMENV % EGASPIN % World Bank
Period (µg/m3) X Y
1. CO 1 – Hr 16232.1 0 -250 – 54.11 54.11
8 – Hr 12990.8 -750 0 56.98 – 1.3 fold
24 – Hr 8120.4 -750 0 71.23 81.20 –
2. NOX 1 – Hr 672.5 -1500 0 – 1.7 fold –
24 – Hr 212.9 -750 0 2.84 fold 1.4 fold 1.4 fold
Annual 45.5 -1500 0 1.9-2.8 fold – –
3. PM10 1 – Hr 28.2 -1750 0 – – –
24 – Hr 13.1 -750 0 – – 16.38
Annual 2.7 -1500 0 – – –
4. SO2 1 – Hr 21.2 0 -250 8.15 – –
24 – Hr 10.6 -750 0 40.77 – –
Annual 2.1 -1500 0 – – 1.68

Table 10. Scenario 7 Output Compared to the Standards

S/No Pollutant Averaging Max. Conc. Receptor Location for (m) % FMENV % EGASPIN % World Bank
Period (µg/m3) X Y
1. CO 1 – Hr 16257 0 -250 – 54.19 54.19
8 – Hr 12990.8 -750 0 56.98 – 1.3 fold
24 – Hr 8121.5 -750 0 71.24 81.22 –
2. NOX 1 – Hr 426.2 0 -250 – 1.1 fold –
24 – Hr 212.9 -750 0 1.9-2.8 fold 1.4 fold 1.4 fold
Annual 42.8 -1500 0 – – –
3. PM10 1 – Hr 26.3 0 -250 – – –
24 – Hr 13.1 -750 0 – – 16.38
Annual 2.6 -1500 0 – – –

Table 11. Scenario 8 Output Compared to the Standards

S/No Pollutant Averaging Max. Conc. Receptor Location for (m) % FMENV % EGASPIN % World Bank
Period (µg/m3) X Y
1. CO 1 – Hr 16,232.1 0 -250 – 54.11 54.11
8 – Hr 12,990.8 -750 0 56.98 – 1.3 fold
24 – Hr 8,120.4 -750 0 71.23 81.20 –
2. NOX 1 – Hr 425.5 0 -250 – 1.1 fold –
24 – Hr 212.8 -750 0 1.9-2.8 fold 1.4 fold 1.4 fold
Annual 42.3 -1500 0 – – –
3. PM10 1 – Hr 26.2 0 -250 – – –
24 – Hr 13.1 -750 0 – – 16.38
Annual 2.6 -1500 0 – – –
Potential impacts of integrated oil and gas plant on ambient air quality 343

The similarities in scenarios 2 and 6, scenarios 3 and 7, and scenarios 4 and 8 were
like those of scenarios 1 and 5. Though they were the same in terms of identified
number of point sources, there are differences in the available fuel involved in
combustion activities in the proposed IOGP. Apart from significant level of emission
of criteria air pollutants from scenarios 5 – 8, there is an additional pollutant expected
in the form of SO2 which is absent in scenarios 1 – 4 due to the “sweetness” of natural
gas of Nigeria origin [1]. The “worst case” scenario with AGO-fired facilities as
represented by scenario 5 (Table 8) indicates that the potential ground level
concentrations of SO2 when the facilities are fired will be about 8.15 %, and 40.77 %
of the 1 – hr and 24 – hr limits of the FMENV, respectively. Around the same plant,
the predicted concentrations will be about 1.68 % of the World Bank annual limit.
Presently, regulators in Nigeria have constraints in applying regulatory frameworks
to gas plant of the IOGP status as obtainable in other countries thus emissions from
projects with multiple emission sources and complex operational scenarios has
hitherto been treated fragmentally. The ISC model as used in this study has
demonstrated the possibility of modelling dispersion on a regional basis. This
approach will avail regulators a proactive tool for cumulating the combined or
incremental effects of past, present or future activities from the myriads of petroleum
and industrial activities that daily go on in the Nigeria’s Niger Delta region. It is
expected to influence policy formulation that will integrate comprehensive cumulative
impacts assessment of air quality in the environmental laws regulating the gas
development projects in the country.

CONCLUSIONS
The ISCST-3 model has been used to evaluate the ground level concentrations from an
IOGP proposed for the Niger Delta, Nigeria, for the purpose of air quality prediction.
The emissions of 12 elevated point sources for criteria air pollutants were considered
with eight different scenarios. Both natural gas-fired and AGO-fired were considered
for the identified emission sources. Five years of hourly meteorological observations
have been used in the prediction of ground level criteria air pollutants. The ground
level concentrations were predicted over averaging periods for which the standards
were available; these range from one hour to one year. The 1 – hr, 8 – hr, 24 – hr, and
annual averaged model-predicted concentrations have been compared with the
ambient air quality limits as set by the Federal Ministry of Environment (FMENV) in
Nigeria, the Department of Petroleum’s Environmental Guidelines (EGASPIN) and
the world Bank at some distances around the proposed plant.
The model outputs show a general maximum ground level concentration between
0.5 – 3.5 km, with easterly and north-easterly receptors having the greatest
concentration. The pollutants emitted in the greatest concentration in their order of
magnitude were CO, NOx and PM10. Maximum ground level concentration for 1, 8
and 24 – hr periods for CO were below regulatory standards (less than 10 %) for
scenarios 1 – 4 when the fuel is natural gas. The 8 – hr ground level concentrations
were about 1.3 folds of the World Bank Standard in scenarios 5 – 8 when the fuel
sources are switched to AGO. However in these scenarios, the FMENV and EGASPIN
 344 Energy & Environment · Vol. 20, No. 3, 2009

limits were not breached for all the averaging periods. Modelled NOX concentrations
were significant in all the scenarios except 3 and 4. Though its concentrations are
within the standards for all the averaging periods in scenarios 1 – 4 except the 1 – hr
limit that will be breached in scenarios 1 and 2, these concentrations are about 1.1 –
2.8 folds in all the averaging periods for all the standards in scenarios 5 – 8. The
predicted PM10 and SO2 concentrations were all below the guideline limits. It may be
concluded that for air quality in the settlements around the proposed IOGP not to be
significantly impacted during the life of the plant, the use of AGO should be
completely avoided while a radius 1.75 km around the plant should be declared as a
protected area with continuous monitoring of ground level concentrations of CO,
NOX, PM10, and SO2.

REFERENCES
1. Sonibare, J.A., and Akeredolu, F.A., 2004. A Theoretical Prediction of Non-methane
Gaseous Emissions from Natural Gas Combustion” Energy Policy, Netherlands, Volume
32 (14), pp. 1653 – 1665
2. EPA, 1995. User’s Guide For the Industrial Source Complex (Isc3) Dispersion Models;
Volume I – User Instructions. U.S. Environmental Protection Agency Office of Air
Quality Planning and Standards Emissions, Monitoring, and Analysis Division Research
Triangle Park, North Carolina 27711. EPA-454/B-95-003a.
3. Kumar, A., Bellam, N.K., Sud, A., 1999. Performance of an industrial source complex
model: predicting long-term concentration in an urban area. Atmospheric Progress 18 (2),
93–100.
4. Lorber, M., Eschenroeder, A., Robinson, R., 2000. Testing the USEPA’s ISCST–Version
3 Model on dioxins: a comparison of predicted and observed air and soil concentration.
Atmospheric Environment 34 (23), 3995–4010.
5. Sivacoumar, R., Bhanarkar, A.D., Goyal, S.K., Gadkari, S.K., Aggarwal, A.L., 2001. Air
pollution modeling for an industrial complex and model performance evaluation.
Environmental Pollution 111, 471 – 477.
6. FEPA, 1991. Guidelines to Standards for Environmental Pollution Control in Nigeria.
Federal Environmental Protection Agency (FEPA), Lagos.
7. World Bank, 1998. Oil and Gas Development – Pollution Prevention and Abatement
Handbook WORLD BANK GROUP,
8. EPA, 1995a. EPA. Compilation of Air Pollutant Emission Factors Volume I: Stationary
Point and Area Sources, 5th ed. United States Environmental Protection Agency, Office
of Air Quality Planning and Standards. Research Triangle Park NC 27711, USA, 1995.
9. Sonibare, J.A., Obanijesu, E.O., Adebiyi, F.M. and Akeredolu, F.A., 2006. Contribution
of Volatile Organic Compounds (VOCs) to Nigeria’s Airshed by Petroleum Refineries.
Accepted for Publication. Journal of Petroleum Science and Technology, UK.
10. Ojo, O., 1977. Climates of West Africa. Heinemann, Ibadan.
11. Auer, August H. Jr., 1978. Correlation of Land Use and Cover with Meteorological
Anomalies, Journal of Applied Meteorology, pp. 636-643.

View publication stats