Journal of King Saud University – Engineering Sciences xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Journal of King Saud University – Engineering Sciences

journal homepage: www.sciencedirect.com

Original article

Gas flaring attendant impacts of criteria and particulate pollutants: A

case of Niger Delta region of Nigeria
Solomon O. Giwa a,⇑, Collins N. Nwaokocha a, Sidikat I. Kuye a,b, Kayode O. Adama a
a
Department of Agricultural and Mechanical Engineering, College of Engineering and Environmental Studies, Olabisi Onabanjo University, Ibogun Campus, Ogun State, Nigeria
b
Department of Mechanical Engineering, College of Engineering, Federal University of Agriculture, Abeokuta, Ogun State, Nigeria

a r t i c l e i n f o a b s t r a c t

Article history: Gas flaring operations as experienced in the Niger Delta region (NDR) of Nigeria are characterized with
Received 8 February 2017 the release of gases, particulates, noise and heat that have adversely affected both human and environ-
Accepted 30 April 2017 ment. The purpose of this paper was to conduct an inventory of pollutants (NOx, SOx, CO, NMVOC, PM2.5,
Available online xxxx
and organic carbon (OC)) released over half a century in order to assess the environmental health risk of
the region. Data on volumes of gas flared in the country and emission factors of pollutants sourced from
Keywords: literature were used in carrying out emission inventory from 1965 to 2015. The air quality index (AQI) of
Gas flaring
NOx, SOx, CO and PM2.5 was estimated while the relative uncertainties associated with the emissions
Emissions
Air quality index
were obtained using AnalyticaÒ (4.6). The result showed that of the 1.78
1012 m3 of gas produced for
Niger Delta region a period of 51 years, 822.02
109 m3 utilized and 917.17
109 m3 were flared. Estimated total amount
Air pollution of 1.13
106, 5.10
106, 1.46
106, 1.05
104, 6.56
105 and 8.90
104 tons of NOx, CO, NMVOC, SOx,
Health risk PM2.5, and OC, respectively, were emitted due to flaring. Relative uncertainties linked to the pollutants
ranged from 99.53% to 335.02%. The environmental health risk assessment as indicated by the AQI
showed that the emissions of NOx and PM2.5 were hazardous, CO was unhealthy and SOx was good.
Statistical analysis on PM2.5 estimated in this study and life expectancy in NDR showed that both data
are significant and statistically independent with a correlation coefficient of 0.6717. The implementation
of sustainable development goals and other international initiatives aimed at reducing gas flaring in the
NDR will ensure the well-being and livelihood of the inhabitants as well as the environment.
Ó 2017 The Authors. Production and hosting by Elsevier B.V. on behalf of King Saud University. This is an
open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction sulphur dioxide, benzapryene, nitrogen dioxides, xylene etc. have

been identified with the flaring of gas (ICF, 2006). These pollutants
Gas flaring is the rapid oxidation of natural gas with the release especially PM and precursor gases emitted have been reported to
of gaseous, particulate and heat into the atmosphere. The compo- have adverse effects on both the environment and human health
sition and quantity of natural gas flared determines the amount (Giwa et al., 2014; Yaduma et al., 2013).
and type of pollutants emitted in addition to meteorological condi- Nigeria is endowed with natural gas proven reserves of 182 tril-
tions, flare design and geometry, and combustion variables (Fawole lion cubic metre, ranking her as the seventh in the world (Shaaban
et al., 2016; Torres et al., 2012). Globally, gas flaring is recognized and Petintin, 2014; Oyedepo, 2014). The country has been flaring
as a colossal waste of natural resource of which around 150 billion gas for over half a century and is currently ranked fifth globally
cubic metres are flared annually (Farina, 2011). Flared gas (FG) is a from the second position held over three decades (USEIA, 2016).
major contributor to global warming and climate change. This infa- A total of 102.3 million cubic metre (mcm) of gas was flared out
mous act is a foremost source of greenhouse gases, precursor gases, of 125.5 mcm in the Niger Delta region (NDR) between 1970 and
volatile organic chemicals (VOCs), polycyclic aromatic hydrocar- 1986 (Shaaban and Petintin, 2014). Gas reserves in the NDR of
bon (PAH), particulate matter (PM) and black carbon (Giwa et al., Nigeria cover wetlands of around 70,000 km2 with 20,000 km2 of
2014; McEwen and Johnson, 2012) that pollutes the air, soil and land. Over 20 million people inhabit the NDR. Due to incessant
water. Over 250 toxins such as hydrogen sulfide, toluene, benzene, gas flaring operations with more than 123 gas flaring sites
(Tawari and Abowei, 2012), the region has been plagued with
⇑ Corresponding author.
socio-economic issues, civil unrest, health problems, environmen-
E-mail addresses: sologiwa2002@yahoo.com, solomon.giwa@oouagoiwoye.edu.
tal and ecological degradation.
ng (S.O. Giwa).

http://dx.doi.org/10.1016/j.jksues.2017.04.003
1018-3639/Ó 2017 The Authors. Production and hosting by Elsevier B.V. on behalf of King Saud University.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Please cite this article in press as: Giwa, S.O., et al. Gas flaring attendant impacts of criteria and particulate pollutants: A case of Niger Delta region of Nige-
ria. Journal of King Saud University – Engineering Sciences (2017), http://dx.doi.org/10.1016/j.jksues.2017.04.003
2 S.O. Giwa et al. / Journal of King Saud University – Engineering Sciences xxx (2017) xxx–xxx

Anthropogenic sources of air pollution in the NDR includes bio- in this study were garnered from publications of Nigerian National
mass burning, oil exploration, industrial operations, road trans- Petroleum Corporation (NNPC) on oil and gas activities in the coun-
portation, gas flaring, pipeline explosion etc (Tawari and Abowei, try (NNPC, 1997; 2008–2015). Data of FG gas for the period of
2012). Few studies have been carried out to determine the level 51 years (1965 to 2015) were collected and utilized in this work
of air pollution in the region via aircraft campaign (Marais et al., to estimate criteria gases (NOx, CO, non-methane volatile organic
2014) and field-based work (Ana et al., 2012; Tawari and carbon (NMVOC) and SOx) and particulates (PM2.5) and organic car-
Abowei, 2012; Ede et al., 2011). Studies on in-situ ground-based bon (OC) released through gas flaring for years under considera-
measurement of emissions at gas flaring sites are scarce (Nwaichi tion. Using Tier 1 default approach, as recommended by
and Uzazobona, 2011; Obanijesu et al., 2009; Sonibare et al., Intergovernmental Panel on Climate Change (IPCC) guidelines
2010). The ambient levels of CO2, SO2, NH3, NO2, CO and PM at flow (IPCC, 2006), emission factor as provided in literature (Aasestad,
stations in various communities within the NDR were measured 2013; Trozzi, 2013; Pederstad et al., 2015) for each of the pollu-
and were found to be higher than recommended values especially tants aforementioned and the volume of gas flared were used in
at the flares but reduced with distance from the flares. In addition estimating the amounts of the pollutants emitted into the NDR
studies on the health impact of gas flaring were also scarce (Ana environment. Data estimation and analyses (correlation and analy-
et al., 2009; Gobo et al., 2009; Adienbo and Nwafor, 2010; sis of variance) were carried out using Microsoft Excel 2013.
Ovuakporaye et al., 2012). The effect of total suspended particles The formulas used in estimating the quantities of pollutants
(TSP), PM10 and PM7 at four stations located in Igwuruta/Umue- released into the atmosphere were given by Eqs. (1)-(6) according
chem Communities in Rivers State, Nigeria was studied and found to IPCC (2006):
to influence disease such as asthma, cough, respiratory difficulty,
eye and skin irritation (Gobo et al., 2009). Also, health implication ECO ¼ GF
EFCO
qfg
106 ð1Þ
of long time exposure to emissions from gas flaring on some
human haematological parameters investigated revealed striking
deterioration in haematological parameters (Adienbo and ENOx ¼ GF
EFNOx
qfg
106 ð2Þ
Nwafor, 2010). Exposure of residents around gas flare sites to emis-
sions from FG was also reported to cause reduced peak expiratory
flow rate, respiratory morbidities, skin disorders and other related ENMVOC ¼ GF
EFNMVOC
qfg
106 ð3Þ
health risks (Ovuakporaye et al., 2012).
However, due to the number of gas flares and the inaccessible
EPM2:5 ¼ GF
EFPM2:5
qfg
106 ð4Þ
terrain of the NDR, several methods have been employed in esti-
mating the quantity of emissions from FG (Fawole et al., 2016).
Of these methods (laboratory and field study, emission factors ESOx ¼ GF
EFSOx
qfg
106 ð5Þ
and scaling calculations, and simulations), emission factors and
activity data is widely used (Zhou et al., 2014). Documentation
on the emission inventory in the country as well as that of gas flar- EOC ¼ GF
EFOC
qfg
106 ð6Þ
ing activities in the region is scarce in public domain except for
recent publications authored by us (Giwa et al., 2014, 2016). Owing where, GF = volume of gas flared (m3); ECO = CO emitted from FG
to the heavy air pollution of the NDR, the urgent need for emission (tons); ENOx = NOx emitted from FG (tons); ENMVOC = NMVOC emit-
inventory of the region has been stressed (Ana, 2011; Fagbeja et al., ted from FG (tons); EPM2:5 = PM2.5 emitted from FG (tons);
2013). However, to the best of our knowledge, the works of Fagbeja ESOx = SOx emitted from FG (tons); EOC = OC emitted from FG (tons);
et al. (2013) and Nwanya (2011) on the emission inventory for SO2, EFCO = CO emission factor (g/kg); EFNOx = NOx emission factor (g/kg);
VOCs, CO, NOx, PM10, CO2 and CH4 from domestic cooking and EFNMVOC = NMVOC emission factor (g/kg); EFPM2:5 = PM2.5 emission
lighting and, energy and emission inventory of energy wasted factor (g/kg); EFSOx = SOx emission factor (g/kg); EFOC = OC emission
and CO2 from gas flaring, respectively, are available publicly. factor (g/kg); qfg = Density of FG (0.882 kg/m3) for Nigeria (Nwanya,
At local and regional levels, the concentrations of these pollu- 2011).
tants from FG can be significant, thereby influencing micro-
environment considerably (Obanijesu et al., 2009; Sonibare et al.,
2010). Subject to the above human health impacts and other doc- 2.2. Uncertainty estimation and propagation
umented facts associated with adverse effects of gas flaring on ani-
mals, vegetation and ecosystem (Tawari and Abowei, 2012; In the estimation of the value of uncertainty linked to the emis-
Ologunorisa, 2001), coupled with the widespread flare sites in sion of NOx, CO, NMVOC, SOx, PM2.5 and OC via flaring operations in
the NDR, there is the urgent need to undertake the emission inven- the NDR, the volume gas flared (GF) was employed as the input
tory of criteria pollutants and particulates that are directly linked variable for modeling the output variables (NOx, CO, NMVOC,
to these problems. This study was carried out in relation to the glo- SOx, PM2.5 and OC). EasyFitÒ 5.6 (evaluation version) was employed
bal mandate of sustainable development goals. to fit the input data (GF) into the appropriate probability distribu-
tion function while AnalyticaÒ (4.5) software was employed in
modeling the uncertainty of the emission estimate. Subsequently,
2. Materials and methods the probability distribution model of the input variable (GF) was
developed. The formulas given in Eqs. (1)-(6) were utilized in Ana-
2.1. Study area and data processing lyticaÒ (4.5) for modeling the uncertainty related to each pollutant
considered in this study. The uncertainty was propagated from the
The study area, NDR, remains the hub of the country’s oil and input variable so as to model the output variables with the use of
gas reserves and it is the lifeblood of the economy (Fig. 1). Over Tier 2 technique as stipulated in IPCC Guidelines (IPCC, 2006).
95% of export income earnings and around 85% of national rev- Finally, the uncertainty estimates associated with the pollutants
enues comes from crude oil and natural gas proceeds (Oyedepo, were evaluated in numerical terms. The detail of the procedure uti-
2014). NDR is prominent for its exceptional biodiversity which lized in this work has been given in a previous paper authored by
has been threatened due to oil and gas operations. The data used us (Giwa et al., 2016).

Please cite this article in press as: Giwa, S.O., et al. Gas flaring attendant impacts of criteria and particulate pollutants: A case of Niger Delta region of Nige-
ria. Journal of King Saud University – Engineering Sciences (2017), http://dx.doi.org/10.1016/j.jksues.2017.04.003
 S.O. Giwa et al. / Journal of King Saud University – Engineering Sciences xxx (2017) xxx–xxx 3

Fig. 1. Gas flaring sites in Niger Delta region.

2.3. Estimation of air quality index plots of all the variables in Fig. 2 showed two distinct cycles – a
small and a large one – with a trough between the plateaus (peak)
Results of the estimated amounts of pollutants released into the of the cycles. The first and the small cycle are displayed by the
environment were compared with ambient air quality standards plots from 1965 to 1983. The plots of variables represent sharp
(World Health Organization (WHO) and other relevant organiza- increase from 1968 to 1974 after relatively constant values from
tion) for conformity with values recommended by these standards. 1965 to 1968, and with relative reduction from 1979 to 1983.
Also, health risk assessment of CO, SOx, NOx and PM2.5 was evalu- The large and second cycle as represented by the plots from
ated using Eq. (7) as provided in the guidelines for reporting air 1983 to 2015 shows considerable increase in values of the vari-
quality index (AQI) (Rim-Rukeh, 2015). The descriptor of AQI as ables from 1983 to 2006 spanning 24 years. Thereafter, significant
stated in the guidelines is given in Table 1. reduction is observed till 2015. The decrease in the values of the
  variables was due to increased efficient utilization of gas in the
IHi  ILo country, especially in the energy sector where electricity is gener-
Ip ¼ ðCp  BPLo Þ þ ILo ð7Þ
BPHi  BPLo ated using gas turbines across the country. It should be noted that
the peak of all variables occurred in 2006 when the highest quan-
where: Ip = index for pollutant P; Cp = rounded concentration of pol-
tity (28.58 bcm) of gas was flared.
lutant P; BPHi = breakpoint that is greater than or equal to Cp ;
As can be seen in Fig. 2, the amounts of CO emitted is the high-
BPLo = breakpoint that is less than or equal to Cp ; IHi = AQI value cor-
est followed by NMVOC, NOx and lastly SOx, for the entire duration
responding to BPHi ; ILo = AQI value corresponding to BPLo .
of 51 years. This can be linked to the flare combustion efficiency of
60%–80% documented for gas flaring operations in the country
3. Results and discussion (Abdulkareem, 2005). Numerically, total quantity of CO, NMVOC,
NOx and SOx released into the environment was 5.10
106 tons,
3.1. Volume of gas flared and amounts of criteria pollutants emitted 1.46
106 tons, 1.13
106 tons and 1.05
104 tons, respectively.
Annual average of each gas amounts to 2.22
104 tons for NOx,
From the data garnered for the volume of GF for the 51-year 9.99
104 tons for CO, 2.86
104 tons for NMVOC and 206.2 tons
period, a sum of 917.17 bcm was flared. On average basis, this cor- for SOX. CO concentrations of 5.0–61.0 ppm and 14.63 mg/m3
responds to annual flaring of 17.98 bcm of gas and 49.27 bcm daily. (7.99 ppm) were recorded for communities close to flaring sites
The volume of GF and quantities of NOx, CO, NMVOC and SOx dis- (Ana, 2011) and 20 m from flaring point (Ogwu et al., 2015),
charged into the NDR environment for the period in view is illus- respectively, as against yearly average of 9.99
104 tons estimated
trated in Fig. 2. Obviously, the same pattern is observed for all in this work. An estimate of 25.9
107 tons/yr of VOC was
the other variables (GF, NOx, CO, NMVOC and SOx). This can be recorded to be discharged into the atmosphere as a result of gas
related to the volume of GF as it is the main variable determining flaring in Rivers State, Nigeria (Nwachukwu and Ugwuanyi,
the quantity of pollutants emitted as expressed in Eqs. (1)-(7). The

Table 1
Air Quality Index Descriptor (Rim-Rukeh, 2015).

Breakpoints
PM2.5 (mg/m3) CO (ppm) SO2 (ppm) NO2 (ppm) AQI Category
0.0–15.4 0.0–4.4 0.000–0.034 – 0–50 Good
15.5–40.4 4.5–9.4 0.035–0.144 – 51–100 Moderate
40.5–65.4 9.5–12.4 0.145–0.224 – 101–150 Unhealthy
65.5–150.4 12.5–15.4 0.225–0.304 – 151–200 Unhealthy
150.5–250.4 15.5–30.4 0.305–0.604 0.65–1.24 201–300 Very unhealthy
250.5–350.4 30.5–40.4 0.605–0.804 1.25–1.64 301–400 Hazardous
350.5–500.4 40.5–50.4 0.805–1.004 1.65–2.01 401–500 Hazardous

Please cite this article in press as: Giwa, S.O., et al. Gas flaring attendant impacts of criteria and particulate pollutants: A case of Niger Delta region of Nige-
ria. Journal of King Saud University – Engineering Sciences (2017), http://dx.doi.org/10.1016/j.jksues.2017.04.003
4 S.O. Giwa et al. / Journal of King Saud University – Engineering Sciences xxx (2017) xxx–xxx

Fig. 2. Volume of gas flared (NNPC, 1997; 2008–2015) and quantity of criteria pollutants emitted.

2010). This value is considerably higher than what was evaluated for OC. Fig. 3 shows the particles released through the burning of
in this work. gas in the NDR for 51 years. Similar pattern as earlier stated in
In 2000, United Nations reported total SO2 emission of sub-section 3.1 for the criteria pollutants was observed in Fig. 3
1.9
105 tons due to air pollution in the country (UN, 2015), which illustrating the amounts of PM2.5 and OC discharged in the region.
is far more than 278 tons of SOx obtained in this study resulting Under the period in view, 6.56
105 tons of PM2.5 and 8.90
104 -
from FG. Also, gas flaring in Rivers State of Nigeria was reported tons of OC were estimated to be released into the environment due
to contribute 1.96
104 tons of SO2 per annum and this amount to flaring operations. It has been stated by Ede and Edokpa (2015)
of emission was observed to be remarkably higher than the value that over 6000 tons of PM2.5 was emitted via gas flaring in the
of 206 tons estimated in this work. Data from the UN put total region on annual basis. Also, 2590 tons/yr of PM2.5 were reported
NOx emission in the country at 1.01
106 tons for the year 2000 to be released in Rivers State of Nigeria due to flaring activities
(UN, 2015). Though this NOx quantity is more than the estimated in the state alone (Nwachukwu and Ugwuanyi, 2010). These values
NOx of 3.0
104 tons from FG obtained in this work, nevertheless, are far less than the annual quantity (1.29
104 tons) of PM2.5 esti-
they seems to be in the same range of that of UN. Due to gas flaring mated in this study as the studies appear to underestimate the par-
in Rivers State, NOx estimate of 1.04
105 tons/yr was achieved, ticulate. As no previous work has reported or estimated the
which is moderately higher than the annual average of 2.22
104 - emission of OC from FG, no comparison could be made of the quan-
tons obtained in this study. The amounts of VOC, SO2, and NOx esti- tity of the particles emitted into the region.
mated for Rivers State, Nigeria from gas flaring source were found
to be relatively higher than those evaluated by this work. This may 3.3. Uncertainty estimation
be due to the method employed in the estimation of the gases.
3.3.1. Estimation of uncertainty for the emission model
3.2. Amounts of particles emitted The uncertainty estimation was propagated using Tier 2 since
the input variable was not normally distributed and have a coeffi-
Emission of PM2.5 and OC as particles from gas flaring is being cient of variation >0.3. Goodness of fit tests conducted on the input
reported in this study seemingly for the first time in literature variable using EasyFitÒ resulted in Weibull distribution for the

Fig. 3. Volume of gas flared (NNPC, 1997; 2008–2015) and quantity of particles (OC and PM2.5) emitted.

Please cite this article in press as: Giwa, S.O., et al. Gas flaring attendant impacts of criteria and particulate pollutants: A case of Niger Delta region of Nige-
ria. Journal of King Saud University – Engineering Sciences (2017), http://dx.doi.org/10.1016/j.jksues.2017.04.003
 S.O. Giwa et al. / Journal of King Saud University – Engineering Sciences xxx (2017) xxx–xxx 5

variable. Fig. 4 provides the influence diagram developed in mod- with estimation of emission. Since a single input variable was
elling the estimated uncertainty connected to the pollutants using employed in this work, the GF remains the only sensitive source
AnalyticaÒ. On running the model, the range of the mean of NOx, of uncertainty associated with the estimation of emissions from
CO, NMVOC, SOx, PM2.5 and OC at 95% confidence level based on gas flaring activities. This signifies the need to seriously assess
Weibull distribution was obtained as given in Table 2. Table 2 gives the method and instrumentation used in data collection relating
the simulated and estimated mean, relative uncertainties of the to the volume of GF.
mean, lower and upper confidence levels of the mean for the pol-
lutants emitted into the region from gas flaring. It is pertinent to
3.4. Environmental health risk assessment of emissions
know that the simulation was run at 250, 500, 750, 1000, 2000
and 3000 iterations. Since both the standard deviation and mean
The evaluation of the AQI for NOx, CO, SOx and PM2.5 in order to
of the pollutants were the same at 2000 and 3000 iterations, values
assess the environmental health risk linked with the emissions dis-
corresponding to 2000 iterations were reported for this study.
charged into the NDR atmosphere due to flaring of gas was carried
out. AQI of >500.40, 196.96, 48.59 and >500 for PM2.5, CO, SOx and
3.3.2. Estimation uncertainty for the model output NOx, respectively, was obtained. These values showed that human
It should be noted that similar cumulative distribution as pro- exposure to PM2.5 and NOx is hazardous, to CO is unhealthy while
vided in Fig. 5 (for NOx) was noticed for CO, NMVOC, SOx, PM2.5 to SOx in relation to the estimated amounts of these pollutants as
and OC (Figs. 6–9). As the model yielded Weibull distribution, reported in this work is good. However, indoor AQI of NMVOC con-
the ends of the distribution for each pollutant (as illustrated in centration of 10 ppm (3190 mg/m3) and above has been adjudged
Figs. 5 and 6) reflect the simulated means of the pollutants. It to be poor, meaning, it is hazardous to human exposure (IAQI,
can be noticed in Table 2 that the estimated means of the pollu- 2017). PM2.5 contained in polluted ambient air has been strongly
tants (NOx, CO, NMVOC, SOx, PM2.5 and OC) were slightly more linked with human health problems and life expectancy
than the simulated means. The form and statistical characteristics (Dziubanek et al., 2017). The estimated PM2.5 in this work was cor-
of the input variable in terms of the collection and mode of collec- related with data on life expectancy in the NDR as garnered from
tion of the data used could be responsible to the incongruity in the literature (Effiong and Etowa, 2012). Correlation coefficient of
means. 0.6717, showing a positive and moderately strong relationship
Uncertainties stemming from GF were propagated in evaluating between PM2.5 and life expectancy in the region was found.
the uncertainty accompanying the emissions of NOx, CO, NMVOC, The total amounts of NOx, CO, NMVOC, SOx, PM2.5 and OC esti-
SOx, PM2.5 and OC. The simulated means of 2.37
104 tons, mated in this work translate to 60.84 tons, 273.78 tons, 78.2 tons,
1.07
105 tons, 3.05
104 tons, 220 tons, 1.37
104 tons and 0.56 tons, 35.24 tons and 4.78 tons, respectively, on daily average.
1.86
103 tons were obtained for NOx, CO, NMVOC, SOx, PM2.5 Year 2015 data on GF and estimated quantities of pollutants were
and OC, respectively, from the output model (see Table 2). In this used for clarity of discussion. For this year, pollutant concentra-
work, relative uncertainty range of 99.53% to 335.02% was estab- tions of 1234.8 mg/m3 (NOx), 5556.6 mg/m3 (CO), 1587.6 mg/m3
lished for the pollutants using the model. The range represents (NMVOC), 11.47 mg/m3 SOx), 715.3 mg/m3 (PM2.5) and 17.46 mg/
both the lower and upper limits of relative uncertainties which m3 (OC) were obtained. The estimated average daily concentra-
were estimated at 95% confidence level as given in Table 2. It tions of pollutants at flare point and 3 km from flare and WHO
should be noted that the magnitude and wide range of relative standard for compliance are presented in Table 3. The estimated
uncertainties related to the emission of pollutants as realized in amounts/concentrations of the pollutants are assumed to be the
this present study emanated for the input variable (GF). However, amounts/concentrations at the flaring point. According to
the quality of the data on GF and other oil and gas related informa- Strosher (1996), average concentration of pollutants at 3 km from
tion has been reported in literature to be relatively correct for the flare is 20% of the maximum concentration. It is based on this
reporting emissions (Hassan and Kouhy, 2013). assertion that the concentrations of the pollutants at 3 km from
As sensitivity analysis is a technique for identifying the most flare are obtained as given in Table 3. At both the flare point and
appropriate source for possible reduction of uncertainty related 3 km away, it can be seen that the concentrations of PM2.5, NOx

Fig. 4. Influence diagram developed for this study.

Please cite this article in press as: Giwa, S.O., et al. Gas flaring attendant impacts of criteria and particulate pollutants: A case of Niger Delta region of Nige-
ria. Journal of King Saud University – Engineering Sciences (2017), http://dx.doi.org/10.1016/j.jksues.2017.04.003
6 S.O. Giwa et al. / Journal of King Saud University – Engineering Sciences xxx (2017) xxx–xxx

Table 2
Uncertainties of Emissions (1965–2015) in the NDR of Nigeria (number of trails = 2000).

Emission Min. (2.5th CL) Simulated Mean Max. (97.5th CL) Relative uncertainty Estimated mean
NOx 110.7 23.7 K 103.1 K 99.53% to 335.02% 22.1 K
CO 498.1 106.6 K 463.7 K 99.53% to 334.99% 99.93 K
NMVOC 142.3 30.47 K 132.5 K 99.53% to 334.85% 28.55 K
SOx 1.028 220.0 956.9 99.53% to 334.95% 206.20
PM2.5 64.13 13.73 K 59.7 K 99.53% to 334.81% 12.86 K
OC 8.698 1862 8097 99.53% to 334.85% 1745
Total 825 176.6 K 768 K 99.53% to 334.88% 165.50 K

Note: CL = Confidence level: Negative random error = (2.5th percentile-mean)/mean; positive random error = (97.5th percentile-mean)/mean.

Fig. 5. Cumulative distribution of NOx.

Fig. 6. Cumulative distribution of NMVOC.

and CO were remarkably higher than the values recommended by NMVOC is known to play a key role in ozone and photochemical
WHO, whereas the concentration of SOx (31.41 mg/m3) was consid- smog formation. Also, aromatic compounds of benzene, toluene,
erably lower than 125 mg/m3 specified by WHO (Table 3). The high xylene and 1, 3-butadiene are major constituents of NMVOC and
values of PM2.5, NOx and CO can be attributed to incomplete com- are carcinogenic, and may lead to leukemia and cancer in human
bustion of FG while the low concentration of SOx is due to the low through prolonged exposure (Tawari and Abowei, 2012). Moreso,
sulphur content of GF. As can be noticed in Table 3, no recommen- OC is an important component of particulate matter in polluted
dation was made by WHO for both NMVOC and OC. air as a result of incomplete combustion of fuels and can reduce
atmospheric visibility (Cao et al., 2006). It has adverse health

Please cite this article in press as: Giwa, S.O., et al. Gas flaring attendant impacts of criteria and particulate pollutants: A case of Niger Delta region of Nige-
ria. Journal of King Saud University – Engineering Sciences (2017), http://dx.doi.org/10.1016/j.jksues.2017.04.003
 S.O. Giwa et al. / Journal of King Saud University – Engineering Sciences xxx (2017) xxx–xxx 7

Fig. 7. Cumulative distribution of SOX.

Fig. 8. Cumulative distribution of PM2.5.

Fig. 9. Cumulative distribution of OC.

Please cite this article in press as: Giwa, S.O., et al. Gas flaring attendant impacts of criteria and particulate pollutants: A case of Niger Delta region of Nige-
ria. Journal of King Saud University – Engineering Sciences (2017), http://dx.doi.org/10.1016/j.jksues.2017.04.003
8 S.O. Giwa et al. / Journal of King Saud University – Engineering Sciences xxx (2017) xxx–xxx

Table 3 responsible to impact climate on continental and global scales

Average daily concentrations of pollutants in 2015. (Giwa et al., 2014, 2016). In this context, the implementation of
Pollutant Concentration (mg/m3) WHO standard (mg/m3) the United Nations’ sustainable development goals (SDGs), World
At flare point 3 km from flare point Bank campaign (Global Gas Flaring Reduction) in reducing gas flar-
ing and World Bank, United Nations, governments and oil compa-
PM2.5 1959.73 391.95 10a (annual); 25a (daily)
NOx 3383.01 676.60 200b (annual); 50b (daily)
nies initiative of ‘‘Zero Routine Flaring by 2030” by the
SOx 31.41 6.28 125b (annual); 50b (daily) Government of Nigeria, being a signatory to all will spur the drive
NMVOC 4349.59 869.92 to increase utilization of natural gas in the country, which will con-
CO 15223.56 3044.71 60b (daily) sequently reduce the volume of gas flared (Fawole et al., 2016).
OC 265.81 53.16 NA
In connection to this present study, the primary mitigation
a
WHO (2014). method to the negative environmental impacts of gas flaring in
b
Ede et al. (2011); NA = not available. the country is reducing the amounts of pollutants at the flaring
sites through increased utilization of natural gas. Beside the con-
tinuous use of natural gas for power generation, its potential as
effects as it can penetrate deep into the lungs. It is a short-lived
transport fuel (compressed and liquefied natural gas), substitute
submicron carbonaceous particle with atmospheric residence time
for other fossil fuels (diesel, kerosene, gasoline etc.) in engines,
of 3–7 days and be transported and spread to hundreds to thou-
and as feedstock for industrial use is huge and should be explored.
sands of kilometers (Cao et al., 2006). Considering the aforemen-
This would subsequently add both commodity and monetary value
tioned health effects of NMVOC and OC and their quantities as
to natural gas. Also, more stringent penalty should be imposed on
given in Table 3, they seem to be hazardous to human health on
the oil companies involved in the flaring operations in order to
exposure. It is therefore evident that exposure to degraded air from
compel them to find possible alternatives to this wasteful act.
the point of flaring to 3 km away and possibly beyond can impact
Finally, the Petroleum Information Bill should be passed into law
human health negatively. This is consistent with the report of Ede
without further delay to address salient issues in the oil and gas
et al. (2011) that at around 10 km from the flaring site (even under
sector such as gas flaring, as this is expected to ameliorate the envi-
unstable atmospheric conditions) exposure to concentration of
ronmental footprints and the attendance effects of flared gas in the
pollutants would be injurious to health.
nearest future.

3.5. Gas flaring and sustainable development goals 4. Conclusion

Long protracted gas flaring operations in the NDR have left the Gas flaring activities in the NDR is a barbaric act renowned to
soil, air and water polluted through gases, particles and heat have impoverished, destroyed and degraded the region. This work
released from FG with ineradicable impact on ecosystem, vegeta- undertook an emission inventory of NOx, CO, NMVOC, SOx, PM2.5
tion, environment and public wellbeing. Heat from gas flaring and OC released into the environment for 51 years of gas flaring
has been reported to affect farmlands to a distance of 2.06 km from using FG data and emission factors of the selected pollutants. In
flare point (Anomohanran, 2012) with particulates transported and addition, the AQI and the attendant environmental health effect
disseminated to 2.61
106 km from flare with mean wind speed of of gas flaring in the region were explored. The magnitude of esti-
3 m/s (Ede et al., 2011). Pollutants from gas flaring activities have mated CO was found to be highest of all the pollutants and this
been documented to be the cause of acidic rain, asthma, cough, can be related to the low efficiency of flare in the country. Compar-
breathing difficulty, eye/skin irritation, roof and structural defor- ison of the estimated quantities of pollutants with previous studies
mation, change in haematological parameters of human, lung mal- showed good conformity of the results obtained in this work.
functioning, increase in gamma radiation level, economic loss and ANOVA tests also revealed significance of the results at 95% confi-
environmental degradation (Anomohanran, 2012; Nwachukwu dence limits. In addition, strong correlation was established
and Ugwuanyi, 2010; Ana et al., 2009; Gobo et al., 2009; Adienbo between life expectancy in the region and PM2.5 estimated in this
and Nwafor, 2010; Ovuakporaye et al., 2012; Tawari and Abowei, study. On the environmental health risk assessment, only SOx of
2012; Onosohwo et al., 2014). all the pollutants were adjudged to be safe on human exposure.
Data on life expectancy in the NDR and that of the country SDGs among other environmental initiatives has been deemed to
(World Bank, 2016) both from 1979 to 2008 were correlated and, have most influence in improving the socio-economic, environ-
a strong and positive interaction between the two data existed mental and public health of the region vis-à-vis reduction in FG
with correlation coefficient of 0.8434. Also, the ANOVA showed through increased natural gas utilization in the country.
that both data are statistically not the same and are significant
with P-value 0.05. It is apparent from the two sets of data on life
expectancy that those of the country were moderately higher than
those of the NDR. According to World Bank, life expectancy for the References
country in 2015 was 52.8 years while that of the NDR have been
Aasestad, K., 2013. Emissions of black carbon and organic carbon in Norway 1990-
reported to be around 40 years (Ukala, 2011), which is consistent 2011. Statistics Norway. <https://www.ssb.no/en/natur-og-miljo/artikler-og-
with life expectancy data for the region. And this is strongly cor- publikasjoner/emissions-of-black-carbon-and-organic-carbon-in-norway-
roborated using the evidence that an increase of 10 mg/m3 of 1990-2011?fane =om>.
Abdulkareem, A.S., 2005. Evaluation of ground level concentration of pollutant due
PM2.5 reduces life expectancy between 6 months to 2 years to gas flaring by computer simulation: a case study of Niger-Delta area of
(Caiazzo et al., 2013; Pope et al., 2009). Nigeria. Leornado Electronic J. Practices Technol. 6, 29–42.
Though increase in natural gas utilization in the country has Adienbo, O.M., Nwafor, A., 2010. Effect of prolong exposure to gas flaring on some
haematological parameters of humans in the Niger Delta Region of Nigeria. J.
drastically reduced the volume of gas flared, especially in the last
Appl. Sci. Environ. Manage. 14 (1), 13–15.
nine years as reported in this study and literature (USEIA, 2016), Ana, G.R., 2011. Air pollution in the Niger Delta area: scope, challenges and
much still need to be done in this regard. As short-lived pollutants remedies, the impact of air pollution on health, economy, environment and
were examined in this work, they are known for their significant agricultural sources, Khallaf, M. (Ed.), ISBN: 978-953-307-528-0, InTech.
Available from: <http://www.intechopen.com/books/the-impact-of-
influence on micro-environment in this case the NDR and the airpollution-on-health-economy-environment-and-agricultural-sources/air-
country as against the long-lived pollutants (CO2, CH4 and N2O) pollution-in-the-niger-delta-area-scopechallenges-and-remedies>.

Please cite this article in press as: Giwa, S.O., et al. Gas flaring attendant impacts of criteria and particulate pollutants: A case of Niger Delta region of Nige-
ria. Journal of King Saud University – Engineering Sciences (2017), http://dx.doi.org/10.1016/j.jksues.2017.04.003
 S.O. Giwa et al. / Journal of King Saud University – Engineering Sciences xxx (2017) xxx–xxx 9

Ana, G.R.E.E., Sridhar, M.K.C., Bamgboye, E.A., 2009. Environmental risk factors and Nwanya, S.C., 2011. Climate change and energy implications of gas flaring for
health outcomes in selected communities of the Niger delta area, Nigeria. Nigeria. Int. J. Low-Carbon Technol. 6, 193–199.
Perspec. Public Health 129 (4), 183–191. Obanijesu, E., Adebiyi, F., Sonibare, J., Okelana, O., 2009. Air-borne SO2 pollution
Ana, G.R.E.E., Sridhar, M.K.C., Emerole, G.O., 2012. Polycyclic aromatic hydrocarbon monitoring in the upstream petroleum operation areas of Niger-Delta, Nigeria.
burden in ambient air in selected Niger Delta communities in Nigeria. J. Air Energy Sour. Part A 31 (3), 223–231.
Waste Manag. Assoc. 62 (1), 18–25. Ogwu, F.A., Peters, A.A., Aliyu, H.B., Abubakar, N., 2015. An investigative approach
Anomohanran, O., 2012. Thermal effect of gas flaring at Ebedei Area of Delta State, on the effect of air pollution on climate change and human health in the Niger
Nigeria. Pacific J. Sci. Technol. 13 (2), 555–560. Delta region of Nigeria. Int. J. Sci. Res. Innov. Technol. 2 (5), 37–49.
Caiazzo, F., Ashok, A., Waitz, I.A., et al., 2013. Air pollution and early deaths in the Ologunorisa, T.E., 2001. A review of the effects of gas flaring on the Niger Delta
United States. Part I: Quantifying the impact of major sectors in 2005. Atmos. environment. Int. Sustain. Dev. World Ecol. 8. 249-5.
Environ. 79, 198–208. Onosohwo, B.U., Mangset, E.W., Khandaker, M.U., 2014. In situ measurements of
Cao, G., Zhang, X., Zheng, F., 2006. Inventory of black carbon and organic carbon atmospheric gamma radiation in the gas flaring environment of the Ologbo
emissions from China. Atmos. Environ. 40, 6516–6527. town, Nigeria. Malay. J. Sci. 33 (1), 99–106.
Dziubanek, G., Spychała, A., Marchwińska-Wyrwał, E., et al., 2017. Long-term Ovuakporaye, S.I., Aloamaka, C.P., Ojieh, A.E., Ejebe, D.E., Mordi, J.C., 2012. Effect of
exposure to urban air pollution and the relationship with life expectancy in gas flaring on lung function among residents in gas flaring community in Delta
cohort of 3.5 million people in Silesia. Sci. Total Environ. 580, 1–8. State, Nigeria. Res. J. Environ. Earth Sci. 4 (5), 525–528.
Ede, P.N., Edokpa, D.O., 2015. Regional air quality of the Nigeria’s Niger delta. Open J. Oyedepo, S.O., 2014. Towards achieving energy for sustainable development in
Air Pollut. 4, 7–15. Nigeria. Renew. Sustain. Energy Rev. 34, 255–272.
Ede, P.N., Edokpa, D.O., Ayodeji, O., 2011. Aspects of air quality status of Bonny Pederstad, A., Smith, J.D., Jackson R., et al., 2015. Assessment of flare strategies,
Island, Nigeria attributed to an LNG plant. Energy Environ. 22 (7), 891–909. techniques for reduction of flaring and associated emissions, emission factors
Effiong, S.A., Etowa, U.E., 2012. Oil spillage cost, gas flaring cost and life expectancy and methods for determination of emissions from flaring. Carbon Limits, 2013.
rate of the Niger Delta people of Nigeria. Advan. Manag. Appl. Econ. 2 (2), 211– Norwegian Environment Agency CL-2013-29. <http://www.miljodirektoratet.
228. no/no/Publikasjoner/2015/Januar1/Assessment-of-flare-strategies-techniques-
Fagbeja, M.A., Hill, J., Chatterton, T., et al., 2013. Residential-source emission for-reduction-of-flaring-and-associated-emissions-emission-factors-and-
inventory for the Niger Delta – a methodological approach. J. Sustain. Dev. 6, methods-for-determination-of-emissions-to-air-from-flaring/>.
98–120. Pope, C.A., Ezzati, M., Dockery, D.W., 2009. Fine-particulate air pollution and life
Farina 2011. Global Gas Flaring Reduction Public-Public Partnership Global Gas expectancy in the United States. N. Eng. J. Med. 360 (4), 376–386.
Flaring Reduction Public-Private Partnership. Washington, D.C. Available at: Rim-Rukeh, A., 2015. An assessment of indoor air quality in selected households in
<http//web.worldbank.org/WBSITE/EXTERNAL/NEWS/0>, squatter settlements Warri, Nigeria. Adv. Life Sci. 5 (1), 1–11.
contentMDK:4607,00.html (visited last on 27/11/16). Shaaban, M., Petintin, J.O., 2014. Renewable energy potentials in Nigeria: meeting
Fawole, O.G., Cai, X.M., MacKenzie, A.R., 2016. Gas flaring and resultant air rural energy needs. Renew. Sustain. Energy Rev. 29, 72–84.
pollution: a review focusing on black carbon. Environ. Pollut. 216, 182–197. Sonibare, J., Adebiyi, F., Obanijesu, E., Okelana, O., 2010. Air quality index pattern
Giwa, S.O., Adama, O.O., Akinyemi, O.O., 2014. Baseline black carbon emissions for around petroleum production facilities. Manag. Environ. Qual. Int. J. 21 (3), 379–
gas flaring in the Niger Delta region of Nigeria. J. Nat. Gas Sci. Eng. 20, 373–379. 392.
Giwa, S.O., Nwaokocha, C.N., Layeni, A.T., 2016. Assessment of millennium Strosher, M., 1996. Investigations of flare gas emissions in Alberta. Final report to
development goal 7 in the Niger Delta region of Nigeria via emissions environment Canada, conservation and protection, the Alberta energy and
inventory of flared gas. Nigerian J. Technol. 2016. http://dx.doi.org/10.4314/ utilities board and the Canadian association of petroleum product. Canada:
Njt.V35i1.1. environment technologies.
Gobo, A.E., Richard, G., Ubang, I.U., 2009. Health impact of gas flares on Igwuruta/ Tawari, C.C., Abowei, J.F.N., 2012. Air pollution in the Niger Delta Area of Nigeria.
Umuechem communities in Rivers State. J. Appl. Sci. Environ. Manage. 13 (3), Intl. J. Fish. Aquat. Sci. 1, 94–117.
27–33. Torres, V.M., Herndon, S., Allen, D.T., 2012. Industrial flare performance at low flow
Hassan, A., Kouhy, R., 2013. Gas flaring in Nigeria: analysis of changes in its conditions. Steam-and air-assisted flares. Ind. Eng. Chem. Res. 51 (39), 12569–
consequent carbon emission and reporting. Acc. Forum 37, 124–134. 12576.
IAQI (Indoor air quality index). <http://www.iaqindex.com/faqlist/10-iaqfaqvoc. Trozzi, C. EMEP/EEA emission inventory guidebook 2013. <http://www.eea.europa.
html>. eu/publications/emep-eea-guidebook-2013>.
ICF (2006). Nigeria: Guidebook for carbon credit development for flare reduction Ukala, E., 2011. Gas flaring in Nigeria‘s Niger Delta: failed promises and reviving
projects. International. June 2006. community voices. J. Energy Clim. Environ. 97, 97–126.
IPCC (Intergovernmental Panel on Climate), 2006. IPCC Guidelines for National UN (United Nations), 2015. Air and climate pollution, NOX and SO2 emissions.
Greenhouse Gas Inventories, Volume 1: General Guidance and Reporting. IPCC, <http://uns tats.un.org/unsd/ environment/qindicators.htm>.
Geneva, Switzerland. USEIA (United States Energy Information Administration). Country analysis brief:
Marais, E.A., Jacob, D.J., Wecht, K., et al., 2014. Anthropogenic emissions in Nigeria Nigeria. May, 2016. <https://www.eia.gov/beta/international/analysis_
and implications for atmospheric ozone pollution: a view from space. Atmos. includes/countries_long/Nigeria/nigeria.pdf>.
Environ. 99, 32–40. World Health Organization, 2014. WHO guidelines for indoor air quality: household
McEwen, J.D., Johnson, M.R., 2012. Black carbon particulate matter emission factors fuel combustion.
for buoyancy-driven associated gas flares. J. Air Waste Manag. Assoc. 62 (3), World Bank, 2016. World development indicators. Nigeria. <http://data.worldbank.
307–321. org/country/nigeria>.
Nigerian National Petroleum Corporation (NNPC), 1997. Annual Statistical Bulletin. Yaduma, N., Kortelainen, M., Wossink, A., 2013. Estimating mortality and economic
Corporate Planning and Development Division (CPDD). 2008–2015. costs of particulate air pollution in developing countries: the case of Nigeria.
Nwachukwu, A.N., Ugwuanyi, J.U., 2010. Air pollution and its possible health effects Environ. Resour. Econ. 54, 361–387.
on rural dwellers in Rivers State, Nigeria. Afr. J. Phys. 3, 217–240. Zhou, Y., Cheng, S., Chen, D., Land, J., Zhao, B., Wei, W., 2014. A new statistical
Nwaichi, E.O., Uzazobona, M.A., 2011. Estimation of the CO2 level due to gas flaring approach for establishing high-resolution emission inventory of primary
in the Niger Delta. Res. J. Environ. Sci. 5 (6), 565–572. gaseous air pollutants. Atmos. Environ. 94. 392-40.

Please cite this article in press as: Giwa, S.O., et al. Gas flaring attendant impacts of criteria and particulate pollutants: A case of Niger Delta region of Nige-
ria. Journal of King Saud University – Engineering Sciences (2017), http://dx.doi.org/10.1016/j.jksues.2017.04.003