Environ Monit Assess (2017) 189:148
DOI 10.1007/s10661-017-5833-6
An assessment of subsurface contamination of an urban
coastal aquifer due to oil spill
Indumathi M. Nambi & Bokam Rajasekhar &
Vijay Loganathan & R. RaviKrishna
Received: 18 May 2016 / Accepted: 9 February 2017
# Springer International Publishing Switzerland 2017
Abstract Incidences of leakages of chemicals from
underground oil storage tanks or oil-carrying pipelines
have posed huge threat to the coastal aquifers around the
world. One such leak was recently identified and notified by the people of Tondiarpet, Chennai, India. The
assessment of the contamination level was done by
obtaining electrical resistivity maps of the subsurface,
drilling of 20 new borewells for soil and water analysis,
and testing the water quality of 30 existing borewells.
Samples were collected from the borewells, and observations were made that included parameters such as
odor, moisture, contamination characteristics, lithology,
groundwater level, thickness of the free product that are
used to demarcate the extent of soil, and water contamination. Furthermore, a multigas detector was used to
detect hydrocarbon presence as soil vapor. Moreover, to
capture the transport of dissolved hydrocarbons, 10
samples were collected in the periphery of the study
area and were analyzed for the presence of petroleum
hydrocarbon and polyaromatic hydrocarbon. Analysis
Electronic supplementary material The online version of this
article (doi:10.1007/s10661-017-5833-6) contains supplementary
material, which is available to authorized users.
I. M. Nambi (*) : B. Rajasekhar : V. Loganathan
Department of Civil Engineering, Indian Institute of Technology
Madras, Environmental and Water Resources Engineering,
Chennai, Tamil Nadu 600036, India
e-mail: indunambi@iitm.ac.in
R. RaviKrishna
Department of Chemical Engineering, Indian Institute of
Technology Madras, Chennai 600036, India
of the data indicated the presence of free-phase hydrocarbon in soil and groundwater close to the junction of
Thiruvottiyur high (TH) road (TH) and Varadaja
Perumal Koil (VPK) street. Although the contaminant
plume is confined to a limited area, it has spread more to
the southern and eastern side of the pipeline possibly
due to continuous abstraction of groundwater by residential apartments. After cutting a trench along the VPK
street and plotting of the plume delineation map, observations indicated that the source of the hydrocarbon leak
is present in VPK street close to TH road. A multipronged strategy was suggested targeting the remediation of oil in various phases.
Keywords Hydrocarbon . Oil spill . Environmental
assessment . LNAPL . North Chennai . Tamil Nadu
Introduction
Apart from oil spill incidents in refineries and onshore
drilling sites, highly urbanized areas are also being
impacted due to transport of petroleum products through
leaking pipelines, leaks from storage terminals, leaks in
underground petrol tanks in retail outlets, and accidents
of oil tankers. When an oil spill occurs and if the volume
of release is high and continuous, it infiltrates the soil
and reaches the capillary fringe zone and eventually to
the water table. The oil floating on the groundwater
table can migrate laterally in the direction of groundwater flow (Mercer and Cohen 1990) and remains as a
continuous source of contamination for several years
148
Page 2 of 17
(Hunt et al. 1988). The hydrocarbon fuels unlike other
pollutants exist as a separate phase above water table
which is technically known as light non-aqueous phase
liquids (L-NAPLs). L-NAPLs are those that are lighter or
less dense than water and tend to float on water table while
undergoing different processes such as volatilization into
soil pores, dissolution into water, biodegradation, and
sorption onto the soil surfaces. In saturated zone of subsurface, NAPL constituents can be found in three phases
such as aqueous phase, pure NAPL, and sorbed phase in
the soil, whereas in unsaturated zone (vadose zone),
NAPL compounds exist in four phases including gas
phase, air in soil pores that contains volatile compounds.
NAPL undergoes continuous mass transfer processes
among these phases (Newell et al. 1995). In summer, there
is excessive water withdrawal in wells which causes water
table to drop, and during monsoon, the recharge of freshwater causes the water table to rise. Due to water table
fluctuation, the oil layer also is pulled up and down which
causes the smearing of the oil, leading to the formation of
smear zones. These smear zones are identified by the oil
trapped into the soil pore spaces due to the capillary forces
(Newell et al. 1995). Characterization of the site in terms
of physical, chemical, and biological attributes is an immediate and essential component of the contaminated site
assessment (Masakorala et al. 2014).
Khan and Abbasi (1999) conducted a study on industrial accidents that had occurred between 1926 and 1997
and reported that land environments are more prone to
accidents than marine environments. Nearly 200-m2 area
of subsoil was contaminated due to leaking of underground storage tank in a petrochemical industry at Vadodara district, India (Shah et al. 2003). In a report to Goa
Pollution Control Board, National Institute of Oceanography reported the presence of petroleum compounds in
water wells in the area of Bhimut ward at Bogmalo, India
(D’Silva et al. 2008). There could be many such instances
of subsurface contamination which were neither noticed
nor reported, and its effect cannot be realized for several
months due to very slow movement of groundwater flow.
Out of 242 storage tank accidents occurred in industrial
zones, Chang and Lin (2006) found that most of the
accidents occurred at petroleum refineries (48%) followed by terminals and pumping stations (26%).
Chennai, the capital city of Tamil Nadu, is one of the
top four metropolitan cities of India with the population of
about 4.7 million (Census data 2011). Exhaustive studies
on the water quality of the shallow Chennai aquifer and
other parts of the state have been previously reported
Environ Monit Assess (2017) 189:148
(Somasundaram et al. 1993; Vasanthavigar et al. 2012;
Brindha et al. 2014; Jebastina and Prince Arulraj 2016;
Parameswari et al. 2016). These studies indicated gross
level of pollution of the Chennai aquifer with regard to
heavy metals, major cations, anions such as Mg2+, Ca2+,
F−, NO3−, PO43−, Cl−, etc., and microbial contamination.
Also, those investigations were confined to single-phase
contamination study which is either groundwater or soil
phase. Moreover, the long-term existence of benzene,
toluene, ethyl benzene, and xylene (BTEX) compounds
and polyaromatic hydrocarbons (PAHs) in northern part of
Chennai district has been observed (Clement 1993;
Brindha and Elango 2014). Despite the contamination,
the Chennai aquifer has been serving as the source of water
to many residents, at least for non-potable purposes especially at times of water scarcity. Groundwater has a natural
hydraulic gradient toward the east due to the seashore, but
it can also be influenced by the localized hydraulic gradients due to pumping. The water table fluctuates up and
down due to rainwater infiltration and pumping.
Studies related to oil spills in India have also been
reported in literature. Sukumaran et al. (2014) reported
the oil spill in coastal waters due to the collision of two
ships in the Mumbai harbor region, India. Sediment and
water samples had been analyzed for petroleum
hydrocarbons, and their resultant affects on marine
ecological species were discussed. Kankara et al.
(2016) in their study identified the sensitive zones and
most impacted marine organisms due to oil spills along
the Chennai coastal stretch. Sukhdhane et al. (2013)
summarized the different oil spill incidents in Indian
coastal waters and their effects on marine habitats.
Similarly, Rekadwad and Khobragade (2015) discussed
the impact of oil spills on marine ecosystem and biodiversity in Goan beaches, India, which could lead to the
decrease in economy of the state from the tourism point
of view. However, these studies focused on oil spills in
seawaters and the resultant effects on coastal water
quality and their ecosystem. Petroleum contamination
in groundwater due to oil spills has not been addressed
in these studies.
Whenever a particular site is contaminated, there is
always a necessity of conducting detailed site investigation to find out the different factors like extent of contamination, type of phase effected (soil, air, and water),
potential exposed receptors, etc. Such site investigation
could be very much useful to quantify the health risk
posed by the contaminants and for future remediation
projects in setting up the site cleanup goals that will not
Environ Monit Assess (2017) 189:148
produce health risks. The contaminants can be of both
carcinogenic and non-carcinogenic in nature. For instance, BTEX compounds which are part of gasoline
fuel pose health effects of lung cancer, leukemia, etc.
(Budavari et al. 2001). Similarly, polycyclic aromatic
hydrocarbons could produce health disorders such as
respiratory tract tumor, leukemia, lung cancer, bladder
cancer, etc. (Bonassi et al. 1989; Armstrong et al. 1994;
Nadon et al. 1995; Boffetta et al. 1997; US EPA 2016).
Recently, a potential hydrocarbon leak was reported
at Tondiarpet, Chennai, in July 2013. Tondiarpet area is
the northern most part of Chennai district. The underground pipelines of major oil companies reach their
respective terminals at Tondiarpet area. The objectives
of the present study are (1) to identify the source of the
recent oil spill and to delineate its extent and (2) to
characterize the contamination associated with various
phases of the subsurface, i.e., soil, groundwater, and air.
Materials and methods
Study area
The study area for environmental site assessment is located in Chennai district of Tamil Nadu state and falls under
Fort Tondiarpet Taluk (Fig. 1a). The site is a congested
area located in Tondiarpet and lies close to the junction of
Thiruvottiyur high road (TH road) and Varadaraja
Perumal Koil Street (VPK street). Domestic houses and
commercial properties lay on either side of the pipeline. It
is located at a distance of 2850 ft west of the coastal line. It
is bounded by Thiruvottiyur high road on the west and
Kannakar street running parallel in the east (Fig. 1b).
The borewells in the area extend up to the rock bed to
a depth of 60 ft with slotted pipes placed from 20 ft. The
depth of the slotted pipe could have varied from one
well to the other. The unique feature of this site is that
the borewells are placed at very close intervals, approximately 10 ft from each other since the property is highly
fragmented and each owner has his/her own borewell.
Geology and hydrogeology of the site
The geological formations of the Chennai district consist of ancient archaean crystalline rocks to the recent
alluvium with varying degrees and depths of weathering
in crystalline rocks. In Tondiarpet, groundwater is
drawn mostly from the alluvium which covers 80-ft
Page 3 of 17 148
depth from the surface, and the depth of borewells in
alluvium ranged between 60 and 80 ft, and the yield
ranges from 63 to 95 gal (US)/min (CGWB 2008). The
permeability of the aquifer ranges between 4.26 and
6.56 ft/day at a depth of ca. 20 ft. The lithology of the
site is described in Fig. 2. The hydrogeology of the area
when investigated showed alternating layers of sand of
different grain sizes each 10 ft deep up to a depth of 30 ft
beyond which the sand layer alternates with clay layers
up to 60 ft. The thickness of the sand layers may vary
from place to place. Beyond 60 to 80 ft, it was hard rock.
Overall, the top soil was filled material with graded
gravels, cobbles of building material mixed with brown
to dark brown sandy silt. Boreholes were drilled to a
depth of 30 ft. The local geology encountered was
unconsolidated alluvium comprising predominantly alternating layers of brown to light brown fine and medium silty sand. It was also found that there are occurrences of discontinuous clay patches in some regions of
top sand layer as shown in Fig. 2. The particle size
distribution indicated the soil to be medium to fine sand,
and the gradation does not vary significantly over the
study area. At depths beyond 22 ft, the color of the sand
changed from light brown to gray to whitish gray fine
sand. At depth of 27 to 28 ft, fine sandy silt was
encountered. Water table was encountered at depth of
27 to 29 ft below ground level. Some of the geological
aspects were also explained in the section BElectrical
resistivity tomography.^
Well gauging and inventory
An inventory of wells was conducted to map the plume,
i.e., to map wells that had hydrocarbon contamination and
wells that were free of hydrocarbon contamination. An oilwater interface meter (Solinst) was used to identify the
presence and depth of free-phase hydrocarbons in groundwater. The interface meter provides a continuous beep
when free-phase hydrocarbons are present and an intermittent beep when water is detected. The sensitivity of the
interface probe is 0.1968 in. About 30 borewells around
the contaminated site were identified to map the free-phase
hydrocarbon plume. The existing borehole locations where
well gauging was carried out is shown in Fig. 3.
Drilling of exploratory boreholes and soil sampling
In addition to well inventory, 20 boreholes in open areas
were dug to determine the presence of soil and
148
Page 4 of 17
Environ Monit Assess (2017) 189:148
N
Fig. 1 a Location of the site. b Site and its surrounding landmarks
groundwater contamination. The aspects considered
while locating the 20 soil boring locations were topographical characteristics, regional and local groundwater flow direction, and the presence of stationary structures like metal posts/buildings, etc. The drilling of
boreholes could not be carried out in grid pattern due
to congested nature of the area, absence of open space,
and presence of roads with live electric cables, sewer
pipes, telephone cables, drinking water pipelines, etc.
These were field limitations. Firstly, trial pits were made
to ensure that there were no live electric cables and/or
pipes. After ensuring safety, the pits were deepened until
the groundwater table (∼30 ft) was reached. Representative soil samples were collected at every 1-ft depth.
The samples were collected, and lithological observations such as odor, moisture, contamination characteristics, lithology, extent of visible contamination, etc. were
recorded. These 20 exploratory soil borings were labeled as EBW 2 to EBW 21. A field Global Positioning
System (GPS) kit was used to map the borehole locations. The coordinates of soil and groundwater sampling
locations were marked using the WGS84 UTM datum.
The site map showing the exploratory borewell locations are shown in the Fig. 4.
Environ Monit Assess (2017) 189:148
Fig. 1 (continued)
Page 5 of 17 148
148
Page 6 of 17
Environ Monit Assess (2017) 189:148
Fig. 2 Cross-sectional view of contaminated site geology
Gas sampling
When each boring reached the bottom of each 1-ft
interval, representative soil sample was collected
using hand auger, and as the sample was retrieved
in a plastic bag, a sampling tube with a silt filter
was inserted that was attached to a gas monitoring
device (AreaRAE) into the plastic bag. The end of
the tube was held at this depth for 1 to 2 min, and
the maximum concentration of the parameters tested for was recorded. To minimize loss of volatiles
during the transfer of samples to laboratory, we
tested the samples for volatile organic compounds
(VOCs) in situ.
Laboratory analysis
The preliminary analytical studies were intended to
qualitatively assess the free oil and groundwater
samples collected from the site. The free oil collected at the borewells were diluted with dichloromethane and analyzed in gas chromatograph with
mass spectrograph (GC-MS). The groundwater
samples were extracted with dichloromethane as
the solvent and analyzed in the GCMS. In addition,
commercially available petrol and diesel were analyzed for comparison.
Electrical resistivity tomography survey
In the electrical resistivity tomography (ERT) method, a
DC current is applied into the ground (subsurface) using
a pair of current electrodes and resulting potential difference is measured by two other electrodes. Based on
the voltage and current applied, resistivity can be calculated. Each material has its own resistivity values depending on the composition and structure. For example,
resistivity of a wet soil is nearly 10 Ω m, for sand stone is
108 Ω m, and for a copper which is good conductor,
resistivity value is 10−8 Ω m (Herman 2001). Similarly,
contaminated soil has different resistivity compared to
the base soil which can be useful to find out the petroleum contamination in soil layers (Cassiani et al. 2014).
When current is applied to subsurface w.r.t. depth, it is
possible to map soil geological profile along with contamination details. In the present study, field data were
collected at two transects (ERT 1 and ERT 2) both
running in the east–west direction. The data were gathered to obtain a continuous coverage of the subsurface
along the line of investigation. A Wenner electrode
configuration was employed in the present study.
Environ Monit Assess (2017) 189:148
Page 7 of 17 148
Fig. 3 Location of existing borewells examined for hydrocarbon contamination
Results and discussion
Interpretation of field results
Well inventory was carried out in about 30 wells surrounding the contaminated area. The contaminated
borewell locations were identified and shown in Fig. 5.
Free-phase hydrocarbon contamination was observed in
all the nine wells of the narrow lane located in TH road
and in few houses on VPK street. The wells in the
narrow lane lie very close to each other within distances
of 7 ft. The product thickness was observed to be in the
range of 2 to 24 in. The maximum product thickness
was observed in well number BW 222 with a thickness
of 24 in. The nature of well construction could be the
reason for varying thickness of product in this close
cluster of borewells. Normally, in a small area, the
product thickness is expected to be uniform. For monitoring product thickness in contaminated areas, the
borewell must be constructed with slotted casings from
above water table to the bottom of the well (for the entire
water column). Floating product on groundwater can
then be accurately recorded in these wells. In case of
plain casings, intercepting water-level product thickness
will be very much reduced than the actual. In VPK
street, few houses exhibited high level of free-phase
product contamination. The maximum product thickness recorded was 2.75 ft. A dal mill located in the
VPK street showed 3 in. of free product, and four other
houses in the lane have recorded 2 to 6 in. of product.
148
Page 8 of 17
Environ Monit Assess (2017) 189:148
Fig. 4 Locations of exploratory borewells
Field observations during soil boring
Soil profiles that are encountered during drilling of a
borewell have been explained in BGeology and hydrogeology of the site^ section earlier. Shallow unconfined groundwater table conditions prevail in the
area. The water level ranges from 18 ft below ground
level during post monsoon and 28 ft during premonsoon period. Soil contamination spread is predominantly due to initial downward oil movement
by gravity immediately after spill and horizontal
plume movement due to gradient. Subsequently, due
to monsoon fluctuation in water level, the oil could
have also moved up and down vertically. The majority of the oil pools above the water table and the
maximum volatile organic compound concentration
are expected close to the water table or in regions
of coarse sand with high permeability. Oil saturation
in the soil was estimated onsite using the VOC
monitor. It was also cross-checked with analysis of
volatile suspended solid in the soil samples brought
back to the laboratory.
The VOC readings with their concentration ranges
observed in different exploratory boreholes are summarized in Table 1 where as depthwise concentration
profiles for VOCs are shown in Fig. 6. In most of the
wells, hydrocarbon contamination in soil was observed from depth of 19 to 28 ft (water table depth).
In the case of EBW no. 18, soil contamination was
observed at a shallow depth of 6 ft and extended up to
the water table (∼29 ft) indicating that this location
(VPK street-TH road intersection) could be closer to
the source of the spill. In some boreholes, moisture
with decomposed odor was observed between 15 and
20-ft depth probably indicating sewerage pollution.
Wherever soil contamination was observed, heavy
odor of hydrocarbon due to release of VOCs was
observed. The MultiRAE gas detector equipment
Environ Monit Assess (2017) 189:148
Page 9 of 17 148
recorded high VOC readings in the range of 0 to
212 ppm. Contours for gas-phase concentration at
two different depths (near to water table and away
from water table) predicting the distribution of gas
phase in subsurface are shown in Figs. S2 and S3.
The soil borelogs are shown in Fig. 7a.
Figure 7b shows the oil contamination profile in the
vadose zone and the capillary fringe. It is observed that
in well EBW no. 2, no VOC was measured up to 15 ft,
and then, the values increased up to the water table. All
borewells away from the pipeline showed a similar
volatile suspended solid (VSS) profile. Close to the
pipeline where the leak could have originated, the
VOC/VSS traces are likely to be present in the top layers
of the vadose zone since the spilled free-phase hydrocarbons penetrate soil and reach groundwater vertically.
Subsequently, plume moves along the direction where
there is greater groundwater abstraction if there were no
recovery of oil. This was observed in the borewells
drilled below the pipeline (EBW nos. 17 and 18). Unlike
the other borewells, the oil was observed in the soil
matrix just below the pipeline and extended all the
way to the water table (Fig. 7b). Maximum VSS was
observed close to the groundwater table in the region
just below the pipeline indicating that the source could
be somewhere below the pipeline. The VSS values are
quite high in the range of 5–20 g/kg which may lead to
large release of toxic/flammable compounds in the atmosphere during excavation operations and slow release
in unpaved areas under normal conditions.
It can also be noted that the VSS data is high at the
zone where the coarse-grained sand is present confirming
the capillary entrapment phenomena. The coarse sand
will hold the oil to a larger extent because the fine sand
below with the small pore size will exert a very high
capillary resistance to the entry of oil into their pores.
Gas chromatography
The gas chromatograms indicated that the oil collected
from the borewell matches with commercial diesel and
commercial petrol. Petrol has only the lighter fractions of
the crude with compounds having less than nine carbon
atoms. These compounds have low boiling point and elute
out at shorter retention times. Diesel with a large range of
compounds with higher number of carbon atoms has a
Table 1 Contamination characteristics of exploratory wells
EBW no.
Depth to product
Depth to water (ft)
Status of soil contamination
VOC range (ppm)
Depth (ft)
2
ND
27.6
HC contaminated
98.9–212
15–30
3
ND
27.5
HC contaminated
10.5–11
29–30
4
ND
27.4
No HC contamination
0
0
5
ND
27.5
HC contaminated
3.1–61.7
18–28
6
28.0
28.2
HC contaminated
11.1–169
20–28
7
ND
27.5
No HC contamination
0
0
8
ND
28.5
No HC contamination
0
0
9
ND
29.0
No HC contamination
0
0
10
ND
27.2
HC contaminated
22.6–190
19–26
11
ND
27.6
No HC contamination
0
0
12
ND
26.7
HC contaminated
11.9–141
19–26
13
ND
28.0
No HC contamination
0
0
14
ND
27.0
HC contaminated
20.4–173
21–26
15
ND
27.0
No HC contamination
0
0
16
28.2
28.3
HC contaminated
11–194
18.0–29
17
27.0
27.1
HC contaminated
6.5–129
11–29
18
27.6
27.7
HC contaminated
29–195
6–29
19
ND
30.5
No HC contamination
0
0
20
ND
ND
HC contaminated
10.5–132
12–25
21
ND
ND
HC contaminated
2–207
18–27
HC hydrocarbon, ND not determined
148
Page 10 of 17
Environ Monit Assess (2017) 189:148
Fig. 5 Identification of hydrocarbon-contaminated borewells
distinct signature (Fig. S1). The groundwater extracts
from borewell also indicated compounds from both the
lighter and heavier fraction dissolved in the groundwater.
Electrical resistivity tomography
The results of the electrical resistivity tomography indicated a four-layer model of the subsurface (Fig. 8). The
tomograms show low resistivity contours up to the 3.2-ft
depth from the surface in the top part of the profile
indicating the presence of saturated silty sand with cobbles and pebbles. High resistivity contour suggests the
presence of medium to fine unsaturated silty sand up to
the depth of ca. 13 ft. Saturated oil sands generally have
low resistive value than the alluvium, and some of the
value is overlapped between the two. The tomogram
Environ Monit Assess (2017) 189:148
Fig. 6 Depthwise gas-phase
concentration of volatile organic
compounds in each borewell
Page 11 of 17 148
Bore
No/Depth
()
2
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
shows slightly low resistive value of 100–400 Ω suggesting the presence of fine to silty alluvium with oil
spill on the sands at a depth of around 30 ft. Most of the
oil seepage is visible at chainage 52 ft and around 105 to
121 ft. The low resistive region with contour 10–100 Ω
suggests the presence of the groundwater aquifer. The
contrast in the resistivity of the sand at third and fourth
layer shows the presence of oil only up to third layer,
i.e., above groundwater table.
Delineation of the plume
The site harbors a heavy cluster of residential and commercial units which poses a huge challenge in the demarcation of contaminated plume and its subsurface
movement. One cannot completely rely on the collected
data to quantify the extent of contamination and/or
remediation since the depth of screening of the existing
borewells is an unknown. Plume delineation was done
based on the collected free-phase hydrocarbon product
thickness in existing borewells and newly drilled exploratory borewells. The plume outputs were created
using 3D analysis GIS software tools. The plume output
for the free-phase product floating on groundwater as of
3
4
5
6
7
8
9
10 11 12 13 14 15 16 17 18 19 20 21
Legend
Not measured
No Contaminaon
1-50 ppm (VOC Value)
50 -100 ppm (Voc value)
101-150 ppm (VOC value)
151-240 ppm(VOC Value)
October 2013 is shown in Fig. 9a. The areal extent of the
plume is about 76,826 ft2. There are various analytical
and semi-analytical methods available for estimating the
volume of oil trapped in the subsurface (Farr et al. 1990;
Lenhard and Parker 1990; Kemblowski and Chiang
1990; Al-Suwaiyan et al. 2002). Based on de Pastrovich
et al. (1979) method, the volume of the spill is estimated
to be around 9528 ft3 (or) 71,327 gal (US), wherein in
this method, the oil in the formation is estimated to be
one fourth of the oil present in the monitoring wells.
When a trench was excavated below the pipeline on
VPK street-TH road junction, there was oil-soaked soil
at shallow depths. Assuming that an equivalent amount
of oil could be present there as in the adjacent well,
about 30 in. of oil right beneath the trench was assumed
and a simulation was performed (Fig. 9b). The simulation indicated the area and volume of contamination
plume to be around 81,549 ft2 and 11,292 ft3, respectively. This scenario can be considered as an upper
bound of the contamination. Moreover, the available
data indicates that the movement of the oil plume is
predominantly eastward during non-pumping periods
following the groundwater gradient. During pumping
times specifically in summer months, it is governed by
148
Page 12 of 17
Environ Monit Assess (2017) 189:148
Fig. 7 a Soil borelog indicating
borewell contamination. b
Volatile solids extracted from the
contaminated soils
borewell pumping rates. Further, extraction of the oil
and oil-mixed water has influenced the oil plume movement since the time the recovery operations started. This
indicates that the extraction of water from the aquifer
plays a huge role in the mobility of groundwater, and
hence, remediation efforts should be planned
accordingly.
Extent of groundwater contamination
The groundwater samples were collected from regions
within the oil contamination and beyond. The samples
were analyzed for 60 VOCs including petroleum hydrocarbons. The dissolved phase plume map for benzene
and toluene is shown (Fig. 10). Benzene is a wellknown carcinogen, and toluene is a potential neurotoxin. The USEPA set maximum contaminant level for
benzene and toluene are 5 μg/L and 1 mg/L, respectively. The dissolved plume map indicates a maximum
concentration of the benzene and toluene in groundwater as high as 23 and 108 mg/L, respectively. Moreover,
the maximum concentration for both benzene and toluene is found close to VPK street-TH road intersection. It
can also be noted that the benzene and toluene-dissolved
Environ Monit Assess (2017) 189:148
Page 13 of 17 148
Fig. 8 Electrical resistivity tomography (ERT) of Tondiarpet oil spill site Transect-1
plume extends beyond the oil plume. This is because it
is in the dissolved phase and transported along with
water.
The dissolved benzene and toluene plumes seem to
be moving in the northeastern direction which could be
driven by local hydraulic gradients due to pumping. The
future predictions of these plume migrations are uncertain due to the extensive pumping in the region with
multiple borewells. But the contamination of the
groundwater will be a continuous phenomenon as long
as there is oil in the aquifer. Literature suggests that
immediate action must be taken since the water shows
high levels of benzene, toluene, and other VOCs that are
extremely detrimental to human health (Pavithra and
Nambi 2012).
In comparison with previous literatures on oil spills
(Sukhdhane et al. 2013; Sukumaran et al. 2014;
Rekadwad and Khobragade 2015; Kankara et al.
2016), the merits and contributions of the present study
include the following: This is the first time the
petroleum-contaminated site had been investigated in
India; the presence of the site in a densely populated
and old part of the metropolitan city with 100% developed area and narrow lanes makes it a challenging and
unique site; since leaking pipelines carried different
petroleum fuels at different times of its operation, nature
of contamination is also unique with mixture of fuel
compounds that includes gasoline, kerosene, diesel,
and lube oil. Site investigation was challenging considering the fact that the local residents were complacent
and accepting the fact that oil is present in their groundwater since they have been living in similar conditions
for several years; site investigation was started soon
after the contamination was reported by residents; detailed study to find out the extent of contamination in all
the three phases of subsurface (soil, water, and gas
phase); and development of contours for predicting free
phase and dissolved phase concentrations, for better
understanding of extent of contamination, using GIS
tools.
Conclusions
A comprehensive evaluation of the site was performed
wherein data were collected from existing boreholes
belonging to residents, drilling of 20 new exploratory
boreholes in available open spaces, and by conducting
resistivity survey. The geology of the contaminated area
consists of fill material up to 3 ft followed by different
grades of alluvial sand up to 30 ft. Groundwater table
beyond 30 ft was contaminated with many hydrocarbon
compounds that include benzene and toluene at levels as
high as 23 and 108 ppm, respectively. Free-phase hydrocarbon is present in 15 existing borewells and three
newly drilled exploratory boreholes. The product thickness in borewells ranges from 1 to 33 in. The free-phase
hydrocarbon is present close to the junction of TH road
and VPK street. The free product in the borewells in the
residences indicated diesel range organics.
A review of plume delineation map indicates that
the source of hydrocarbon leak is somewhere in VPK
street close to TH road. This was confirmed by drilling
borewells below the pipeline in that stretch of the road.
148
Page 14 of 17
Environ Monit Assess (2017) 189:148
(a)
(b)
Fig. 9 a Map showing the thickness of oil plume floating above the groundwater table as of October 2013. b Map showing the thickness of
oil plume considering the presence of oil in the trench
Environ Monit Assess (2017) 189:148
(a)
(b)
Fig. 10 Dissolved phase plume map. a Benzene. b Toluene
Page 15 of 17 148
148
Page 16 of 17
The areal extent of the oil spread in the aquifer is
around 76,825–81,549 ft2, and volume of the freephase hydrocarbon spill is approximately 9528 to
11,292 ft3, respectively. These estimates are based
on the interpolation of the oil depth data measured
on the field. Soil contamination was observed in
12 out of 20 newly drilled exploratory wells. The
depth of soil contamination starts from ca. 16 ft
below ground level in all exploratory borewells
except the zones below the pipeline where it was
much shallow and close to the bottom of the
pipeline. The contaminated soil thickness ranges
between 10 and 12 ft spread over a limited area.
Contamination of soil in the range 5 to 20 g/kg
was observed above and below the water table
indicating the smearing of the oil pool due to
water table fluctuation.
Based on the assessment, we have found oil
contamination of the aquifer in the following three
forms: (1) a large volume of oil floating as a pool
on top of the groundwater table as a light nonaqueous phase liquid, (2) a significant volume of
oil entrapped to the soil as blobs in the vicinity of
the water table both above and below in the vadose zone and saturated layer, and (3) dissolved
phase hydrocarbon in the groundwater in the vicinity of the oil blobs and oil pool and moving
downstream.
Based on our results and observations, it is recommended that the following steps could be undertaken for
complete cleanup of the area:
1. Installing recovery wells in the trench close to the
source and in the locations which show maximum
standing oil depths
2. Removal of free product through a combination of
proven techniques such as dual-phase pumping,
surfactant-enhanced removal, or other proven
technologies.
3. Removal of trapped oil from the vadose zone
through proven techniques such as soil vapor
extraction.
4. In situ treatment of contaminated groundwater and
removal of blobs in saturated zone by chemical or
biological methods.
5. Installation of monitoring wells at the higher contamination zones and periphery of the dissolved
plume to monitor the effect of oil removal and
groundwater remediation.
Environ Monit Assess (2017) 189:148
It has to be noted that in Chennai, potable water has
been a precious commodity. Since the demand for water
exceeds the available water resources (both surface and
groundwater), the water needs of the city are managed
by borrowing water from adjacent states and/or by
adopting cost-intensive treatment technologies like desalination. Besides the quantity, the quality of groundwater has been marred by processes like saltwater intrusion resulting due to unregulated overextraction of
groundwater from coastal aquifers. In this situation, it
is important that we protect our existing water resources
from scenarios such as oil spills since complete remediation of the contaminated site is not only costly and
challenging but also a time-consuming one that could
run up to decades for a complete cleanup. This particular
investigated region is extremely depending on groundwater. Early groundwater was the main source for usage
for the entire region which has now become unusable.
Although the investigation used established technologies, it is the first time the petroleum pipeline spill issue
has been taken up for investigation in India. The study
has generated a lot of learnings, and awareness has
happened among the common public, regulators, and
judicial authorities. This study also paved way for remedial action by the polluters as enforced by the National Green Tribunal (NGT), India.
Acknowledgements The authors would like to acknowledge
Tamilnadu Pollution control Board who initiated the study, Bharat
Petroleum Corporation Limited for funding and EGSS Pvt. Ltd
who assisted in the field investigation services.
References
Al-Suwaiyan, M. S., Bashir, K., Aiban, S. A., & Ishaq, A. M.
(2002). Analytical model to quantify crude oil spill volume in
sandy layered aquifers. Journal of Environmental
Engineering, 128(4), 320–326.
Armstrong, B., Tremblay, C., Baris, D., & Theriault, G. (1994).
Lung cancer mortality and polynuclear aromatic hydrocarbons: a case-cohort study of aluminium production workers
in Arvida, Quebec, Canada. American Journal of
Epidemiology, 139(3), 250–262.
Boffetta, P., Jourenkova, N., & Gustavsson, P. (1997). Cancer risk
from occupational and environmental exposure to polycyclic
aromatic hydrocarbons. Cancer Causes & Control, 8, 444–
472.
Bonassi, S., Merlo, F., Pearce, N., & Puntoni, R. (1989). Bladder
cancer and occupational exposure to polycyclic aromatic
hydrocarbons. International Journal of Cancer, 44, 648–651.
Environ Monit Assess (2017) 189:148
Brindha, K., & Elango, L. (2014). PAHs contamination in groundwater from a part of metropolitan city, India: a study based on
sampling over a 10-year period. Environmental Earth
Sciences., 71, 5113–5120.
Brindha, K., Vaman, K. V. N., Srinivasan, K., Babu, M. S., &
Elango, L. (2014). Identification of surface watergroundwater interaction by hydrogeochemical indicators
and assessing its suitability for drinking and irrigational
purposes in Chennai, Southern India. Applied Water
Science., 4(2), 159–174.
Budavari, S., O’Neil, M. J., Smith, A., & Heckelman, P. A. (2001).
The Merck index: an encyclopedia of chemicals, drugs and
biologicals. Rahway, NJ: Merck & Co..
Cassiani, G., Binley, A., Kemna, A., Wehrer, M., Orozco, A.F.,
Deiana, R., Boaga, J., Rossi, M., Dietrich, P., Werban, U.,
Zschornack, L., Godio, A., Gandomi, A.J., & Deidda, G.P.
(2014). Noninvasive characterization of the Trecate (Italy)
crude-oil contaminated site: links between contamination and
geophysical signals. Environmental Science and Pollution
Research, 1–18. doi:10.1007/s11356-014-2494-7.
Census data (2011). Directorate of census operations—Tamilnadu.
http://www.census.tn.nic.in. Accessed 11 July 2014.
CGWB (Central Ground Water Board) (2008). District groundwater brochure, Chennai district, Tamil Nadu, India.
Chang, J. I., & Lin, C. C. (2006). A study of storage tank accidents. Journal of Loss Prevention in the Process Industries.,
19, 51–59.
Clement, T. P. (1993). Discussion of groundwater pollution of the
Madras urban aquifer, India. Groundwater, 31, 1029.
D’Silva, C., Paropkari, A. L., Naik, B. G., Fernandes, B., Dalvi,
H., & D’Souza, C. (2008). Sampling of wells and spring
contaminated with petroleum products At Bhimut Ward–
Bogmolo. Goa. National Institute of Oceanography, Goa.
Farr, A. M., Houghtalen, R. J., & McWhorter, D. B. (1990).
Volume estimation of light nonaqueous phase liquids in
porous media. Ground Water, 28(1), 48–56.
Herman, R. (2001). An introduction to electrical resistivity in
geophysics. American Journal of Physics, 69(9), 943–952.
Hunt, J. R., Sitar, N., & Udell, K. S. (1988). Non aqueous phase
liquid transport and clean up:1. Analysis of mechanisms.
Water Resources Research, 24(8), 1247–1258.
Jebastina, N., & Prince Arulraj, G. (2016). Contamination analysis
of groundwater in Coimbatore district, India: a statistical
approach. Environmental Earth Sciences, 75, 1447.
Kankara, R. S., Arockiaraj, S., & Prabhu, K. (2016).
Environmental sensitivity mapping and risk assessment for
oil spill along the Chennai coast in India. Marine Pollution
Bulletin, 106(1–2), 95–103.
Kemblowski, M. W., & Chiang, C. Y. (1990). Hydrocarbon thickness fluctuations in monitoring wells. Ground Water, 28(2),
244–252.
Khan, F. I., & Abbasi, S. A. (1999). Major accidents in process
industries and an analysis of causes and consequences.
Journal of Loss Prevention in the Process Industries., 12,
361–378.
Lenhard, R. J., & Parker, J. C. (1990). Estimation of free hydrocarbon volume from fluid levels in monitoring wells. Ground
Water, 28(1), 57–67.
Page 17 of 17 148
Masakorala, K., Yao, J., Chandankere, R., Liu, H., Liu, W., Cai,
M., & Choi, M. M. F. (2014). A combined approach of
physicochemical and biological methods for the characterization of petroleum hydrocarbon-contaminated soil.
Environmental Science and Pollution Research., 21, 454–
463.
Mercer, J. W., & Cohen, R. M. (1990). A review of immiscible
fluids in the subsurface: properties, models, characterization,
and remediation. Journal of Contaminant Hydrology., 6,
107–163.
Nadon, L., Siemiatycki, J., Dewar, R., Krewski, D., & Gerin, M.
(1995). Cancer risk due to occupational exposure to polycyclic aromatic hydrocarbons. American Journal of Industrial
Medicine, 28(3), 303–324.
Newell, C. J., Acree, S. D., Ross, R. R. & Huling, S. G. (1995).
Light nonaqueous phase liquids, U.S. Environmental
Protection Agency Ground Water Issue Paper, R.S. Kerr
Environmental Research Laboratory, Ada OK. EPA/540/S95/500.
Parameswari, K., Mudgal, B. V., & Padmini, T. K. (2016).
Awareness on usage of contaminated groundwater around
Perungudi dumpsite, Tamil Nadu, India. Environment,
Development and Sustainability, 18(3), 657–668.
de Pastrovich, T. L., Baradat, Y., Barthel, R., Chiarelli, A., &
Fussel, D. R. (1979). Protection of groundwater from oil
p o l l u t i o n . C O N C AW E R e p . N o . 3 / 7 9 , H a g u e ,
The Netherlands.
Pavithra, M., & Nambi, I. M. (2012). NAPL contamination in soil
and aquifer—an endeavor towards creating awareness.
International Journal of Emerging Trends in Engineering
and Development., 2(2), 445–454.
Rekadwad, B. N., & Khobragade, C. N. (2015). A case study on
effects of oil spills and tar-ball pollution on beaches of Goa
(India). Marine Pollution Bulletin, 100(1), 567–570.
Shah, J. S., Shroff, A. V., Jignesh, V. P., Tiwari, K. C., &
Ramakrishnan, D. (2003). Stabilisation of fuel oil contaminated soil—a case study. Geotechnical and Geological
Engineering., 21(4), 415–427.
Somasundaram, M. V., Ravindran, G., & Tellam, J. H. (1993).
Ground-water pollution of the Madras urban aquifer, India.
Ground Water, 31(1), 4–11.
Sukhdhane, K. S., Priya, E. R., Raut, M. S., & Jayakumar, T.
(2013). Status of oil pollution in Indian coastal waters.
Fishing Chimes, 33(5), 53–54.
Sukumaran, S., Mulik, J., Rokade, M. A., & Kamble, A. (2014).
Impact of ‘Chitra’ oil spill on tidal pool macrobenthic communities of a tropical rocky shore (Mumbai, India). Estuaries
and Coasts, 37(6), 1415–1431.
US EPA (2016). Poly cyclic organic matter. Available at:
https://www3.epa.gov/ttn/atw/hlthef/polycycl.html.
Accessed 22 June 2016.
Vasanthavigar, M., Srinivasamoorhty, K., & Prasanna, M. V.
(2012). Evaluation of groundwater suitability for domestic,
irrigational, and industrial purposes: a case study from
Thirumanimuttar river basin, Tamilnadu, India.
Environmental Monitoring and Assessment, 184(1), 405–
420.