Ecological Engineering 51 (2013) 237–243

Contents lists available at SciVerse ScienceDirect

Ecological Engineering
journal homepage: www.elsevier.com/locate/ecoleng

Enhancing bioremediation of oil-contaminated soils by controlling nutrient

dispersion using dual characteristics of soil pore structure
Yasushi Mori a,∗ , Atsushi Suetsugu a , Yuko Matsumoto b , Atsushi Fujihara b , Kosuke Suyama b
a
Graduate School of Environmental Science, Okayama University, Okayama, Japan
b
Graduate School of Life and Environmental Science, Shimane University, Matsue, Japan

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

Article history: Soil structure is heterogeneous with cracks or macropores allowing bypass flow, which may lead to applied
Received 26 April 2012 chemicals avoiding interaction with soil particles or the contaminated area. We investigated the biore-
Received in revised form 26 August 2012 mediation efficiency of oil-contaminated soils by applying suction at the bottom of soil columns during
Accepted 1 December 2012
bioremediation. Unsaturated flow conditions were investigated so as to avoid bypass flow and achieve
Available online 7 January 2013
sufficient dispersion of chemicals in the soil column. The boundary conditions at the bottom of the soil
columns were 0 kPa (saturated flow) and −3 kPa (unsaturated flow), and were applied to a volcanic ash soil
Keywords:
with and without macropores. Unsaturated flow was achieved with −3 kPa and an injection rate of 1/10
Bioremediation
Oil contaminated soils
of the saturated hydraulic conductivity. The resultant biological activities of the effluent increased dra-
Unsaturated zone matically in the unsaturated flow with macropores condition. Unsaturated conditions prevented bypass
Bypass flow flow and allowed dispersion of the injected nutrients. Unsaturated flow achieved 60–80% of saturation,
Solute transport which enhanced biological activity in the soil column. Remediation results were better for unsaturated
conditions because of greater biological activity. Also, unsaturated flow with macropores achieved even
remediation efficiency from upper through lower positions in the column. Finally, taking the applied
solution volume into consideration, unsaturated flow with −3 kPa achieved a 10 times higher efficiency
when compared with conventional saturated flow application. These results suggest that effective use of
nutrients or remediation chemicals is possible by avoiding bypass flow and enhancing biological activity
using relatively simple and inexpensive techniques.
© 2012 Elsevier B.V. All rights reserved.

1. Introduction
Soils are an important environmental resource, along with air
and water. They affect not only the global hydrologic cycle, but
also regional water environments. Environmentally sound water
and solute circulation in watersheds is desirable for conservation,
protection and recovery of the natural environment in regional
societies. In recent years, various regulations, such as the Soil Con-
tamination Countermeasures Act (Ministry of the Environment,
Japan, 2002) and CERCLA (U.S. EPA, 1980) were established to
protect the soil environment and prevent soil contamination by
environmentally toxic chemicals.
If soils are contaminated by oil or volatile organic compounds,
effective remediation is sometimes challenging. When the con-
centration of contaminants is relatively high, ex situ treatment
∗ Corresponding author at: Graduate School of Environmental Science, Okayama
University, 3-1-1 Tsushimanaka, Okayama 700-8530, Japan. Tel.: +81 86 251 8875;
fax: +81 86 251 8875.
E-mail address: yasushim@cc.okayama-u.ac.jp (Y. Mori).
after soil excavation or soil vapor extraction can be applied (U.S.
EPA, 1988). At lower concentrations, many physical/chemical civil
engineering techniques are not cost-effective, because most of the
contaminants remain in small micropores. In these cases, biore-
mediation is a promising site treatment tool (Balba et al., 1998),
and its application to the removal of pollutants is typically less
expensive than other civil engineering methods (Russell, 1992).
Bioremediation enhances the degradation process by injecting air
or nutrients, or even cultured microbes that are specific for cer-
tain degradation processes (Vidali, 2001). It has less impact on the
environment and enables degradation of hazardous compounds to
innocuous by-products (Vidali, 2001). In situ bioremediation tech-
niques have major advantages in the remediation of contaminated
soil and groundwater because they do not require site excavation
(U.S. EPA, 1990).
The success of bioremediation of contaminated soils depends
on how much chemical solution is conducted into the finer pores
where the contaminants are usually located. However, the soil
pore structure is heterogeneous, and macropores or cracks may
rapidly conduct water flow below the remediation zone (Beven
and Germann, 1982), in contrast to Darcy’s law (Darcy, 1856)

0925-8574/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ecoleng.2012.12.009
238 Y. Mori et al. / Ecological Engineering 51 (2013) 237–243

and Richards equation (Richards, 1931). Soil macropore networks

establish a dual-domain type of transport in which water and
solutes are preferentially channeled through soil macropores,
while slowly diffusing into and out of the bulk soil matrix (Gerke
and van Genuchten, 1993; Köhne and Mohanty, 2005). There are
reports that elucidate the role of macropores in solute transport
(Natsch et al., 1996; Pivetz and Steenhuis, 1995; Ronkanen and
Klove, 2009). However, generally it is a challenge to control solute
transport in soils so that applied chemicals are delivered into
desired positions.
The motivation for this study was the need for solute transport
control in soils with macropores, where bypass flow is likely to be
dominant, with a consequent high risk of bioremediation failure.
Engineers at remediation sites sometimes inject nutrients into the
ground at high rates. However, most of the injected materials will
not undergo bioremediation because of bypass flow. In this context,
it is fundamental to study and utilize the natural structure of soils
in order to achieve environmentally sound remediation.
Convection and dispersion are fundamental processes under-
pinning solute transport in porous media, which governs the
distribution of solutes in soils (Ishiguro, 1992; Toride et al., 2003). If
dispersion is sufficiently developed in a soil column, solutes will be
well distributed and breakthrough curves will show a normal dis-
tribution. On the contrary, if solute transport in soils is governed by
structure-induced bypass flow, dispersion will not be well devel-
oped in the soil column and bypass flow will deliver the solutes
through the soil without interaction with most of the soil body Fig. 1. Experimental setup for the biostimulation experiments.
(Kutilek and Nielsen, 1994).
In a previous report (Mori and Higashi, 2009), we introduced
solutes to the soil matrix and successfully enhanced dispersion,
even in soils with macropores, while avoiding bypass flow. Bimodal 10.0 cm) to give bulk densities of 0.90 cm3 cm−3 that simulated nat-
distribution in the breakthrough curve (bypass flow dominant) for ural conditions. The columns were placed in pressure cells with
a macroporous soil column was greatly changed to a normal dis- membrane filters at the bottom. The experimental setup is shown
tribution (matrix flow dominant), when a suction of −3 kPa and an in Fig. 1.
infiltration rate of 1/10 of the saturated hydraulic conductivity (Ks) Two columns contained only micropores with no macropores.
were applied. Application of this technique to contaminated soils In the other two soil columns, seven vertical macropores (simulat-
could deliver chemicals to desired locations within the soil profile. ing a natural field soil) were artificially created by gently pushing
Moreover, if control could be achieved by using a simple technique, a stainless steel rod of 1 mm diameter into the soils following the
it would be beneficial for cost effective remediation. technique of Mori and Higashi (2009). The macropores were suffi-
In this study, biostimulation was applied to oil-contaminated ciently small that they were not visible from the surface. Physical
soils to investigate the effect of structural differences on bioreme- properties of the soil columns with and without the macropores
diation. Cutting oil was used as the contaminant as it is frequently are shown in Table 1. Saturated hydraulic conductivities were mea-
used for lubrication in industrial processes (Shashidhara and sured using the constant head method (Klute and Dirksen, 1986).
Jayaram, 2010) and contamination by it is often found in the vadose After saturation with distilled water, samples were drained by
zone (unsaturated zone) (Hillel, 1998). Macropore structure is more applying gravitational pressure to the bottom of the tubes using an
often the rule than the exception in the unsaturated vadose zone outlet tube. Nutrients were then applied to the top of the columns
and thus the effect of soil structure needs to be carefully exam- for 30 days. The applied pressure and infiltration intensity were
ined. The two objectives of this study were to examine differences 0 kPa and 1.4 × 10−4 cm s−1 , respectively, for the saturated infiltra-
in biological activity by changing the soil structure and controlling tion conditions, and −3 kPa and 1.4 × 10−5 cm s−1 , respectively, for
the infiltration process, and to examine how to obtain cost-effective the unsaturated infiltration conditions. The four columns were con-
conditions for biostimulation in structured soils where macropores sidered to represent unsaturated or saturated flow with or without
are predominant. macropores.

2. Materials and methods Table 1

Physical properties of the examined soils.
2.1. Soils
Soil Kuroboku soil (Andisol)

Bioremediation experiments were conducted using four soil Pore structure w/o macropores w/ macropores
Particle density (Mg m−3 ) 2.59 2.59
columns containing a Kuroboku soil (a volcanic ash soil) from Porosity (m3 m−3 ) 0.654 0.654
Shimane, Japan. Soils were sampled from depths of 30–50 cm. Con- Macro-porosity (m3 m−3 ) 0 0.003
taminated soils were prepared by mixing the soil with cutting oil Ks (cm s−1 ) 1.40 × 10−5 1.43 × 10−4
at a concentration of 5000 mg kg−1 to enable effective observation Bulk density (Mg m−3 ) 0.90 0.90
of the bioremediation processes. Four repacked soil columns were Macroporosity: the volume of macropores (d > 75 ␮m) in a soil sample divided by
prepared in small stainless steel columns (diameter 5.0 cm, height the bulk volume of the sample (after Soil Sci. Soc. Am., 1996).
 Y. Mori et al. / Ecological Engineering 51 (2013) 237–243
Table 2
Constituents of injected nutrients.
Anions Concentrations (mg/L) Cations Concentrations (mg/L)
NO3 −
99.43 NH4 +
134.28
PO4 3− 344.25 K+ 117.56
SO4 2− 0.61 Mg2+ 0.81
Cl− 4.21 Ca2+ 0.77
Na+ 3.50
2.2. Biostimulation experiments
A micro quantifying pump (AP-2250, Advantec, Tokyo, Japan)
was used for nutrient injection. Nutrient solutions with an N:P:K
ratio of 1:2:1 were prepared for the biostimulation experiments
(Table 2). To prevent damage to the soil surface from the impact
of water drops, nylon mesh was placed over the samples. To main-
tain a constant pressure head at the bottom boundary, an outlet
tube was placed 5 cm below the bottom (−0.5 kPa) for the satu-
rated flow condition and 35 cm (−3.5 kPa) below the bottom for
the unsaturated flow condition. Hydraulic resistance at the bottom
of the pressure cell was 0.5 kPa, and the resultant suctions at the
bottom of the soil samples were 0 kPa for saturated flow and −3 kPa
for unsaturated flow, respectively.
Solute transport was carried out under steady state condi-
tions. Because of the water saturation dependence of hydraulic
conductivity in soils, the resultant achieved outflow rates were
1.4 × 10−4 cm s−1 for saturated flow and 1.4 × 10−5 cm s−1 for the
−3 kPa unsaturated flow conditions. These rates corresponded well
with field values of saturated and unsaturated hydraulic con-
ductivity values at −3 kPa (Mori and Higashi, 2009). The room
temperature was maintained at 25 ◦ C.
At the conclusion of the experiments, the experimental systems
were taken apart and soil samples were collected by depth from
the upper, middle, and lower parts of the column. Water content,
FDA hydrolysis activity, and residual oil weight were measured. The
hexane extraction method (Central Environmental Council, 2006)
was used for estimating the residual oil weight. All soil volumes
were used for residual oil weight measurement to avoid uncer-
tainty arising from the small sampling volumes.
2.3. X-ray radiography of infiltration process
X-ray radiography was undertaken according to Mori et al.
(1999a,b). X-ray images were taken during both the saturated and
unsaturated infiltration experiments for the macropore columns,
which were conducted separately from the bioremediation exper-
iments with the soils packed in low density acrylic columns rather
than the high density stainless columns to obtain clear images.
Methylene iodide (CH2 I2 ) was used as a contrast agent to obtain
clear images. It was applied drop-wise onto the soil surface during
the experiments.
2.4. FDA hydrolysis activity method
Effluent was collected each day and biological activity was
measured using the FDA hydrolysis activity method. Fluorescein
diacetate (FDA) hydrolysis is a non-specific reaction that repre-
sents esterase activity, involving proteases, esterases and lipases.
These enzymes are usually found in the biodegradation processes
of organic materials. Microbial degradation activity was deter-
mined by measuring FDA hydrolysis, following the methodology
described by Schnurer and Rosswall (1982). Fluorescein diacetate
is hydrolyzed into fluorescein by carboxyl esterase. Fluorescein
239
emits fluorescence, which was measured by spectrophotometry at
490 nm.
Fujihara et al. (2010) developed the FDA hydrolysis method
for measuring biological activity in effluent solutions. The effluent
solution needed to be incubated at 25 ◦ C to obtain clear differ-
ences in FDA hydrolysis activity, since the bacterial population in
the effluent was small and differences were hard to distinguish.
Fujihara et al. (2010) showed that fluorescence linearly increased
with incubation over 48 h. Thus we incubated the solution for 24 h
at 25 ◦ C so that we could obtain sufficient fluorescent intensity at
490 nm.
The FDA hydrolysis activity method was also applied to the soil
sample after the biostimulation experiment as described in Section
2.2.
2.5. EC measurement
Characteristics of effluent solute transport were observed. If the
dynamics of solute transport are different, the resultant effluent EC
fluctuations will be different. A Wenner array of four-electrode EC
sensors was installed at the middle of the drainage tubes to monitor
effluent EC fluctuations (see Fig. 1 for the detail).
2.6. Polymerase chain reaction-denaturing gradient gel
electrophoresis (PCR-DGGE)
To understand microbial community shifts in the biostimula-
tion experiments, PCR-DGGE analysis of soil DNA samples from
the oil-contaminated and the uncontaminated soils was conducted.
Twenty grams of each soil sample was incubated with continuous
shaking in 100-mL of the nutrient solution. Samples were dupli-
cated for each treatment. The concentration of the cutting oil in
each soil sample was 2.5% (w/w). The soil DNA samples were pre-
pared by using an ISOIL for Beads Beating kit (NIPPON GENE Co.,
Ltd). The primer set of the PCR was F338GC and 907r (Huong et al.,
2008) for amplification of the V3-V5 region of 16S ribosomal RNA
genes. Primer F338GC included a GC clamp at the 5 end. Other con-
ditions for the PCR followed those described by Huong et al. (2008).
The amplification products were electrophoresed in 6% polyacryl-
amide gel at 65 V and 333 K for 21 h. The denaturing gradient was
from 30% to 75% (7 mol L−1 urea and 40% formamide mixture was
standardized as the 100% denaturing solution). TAE buffer (40 mM
Tris–Cl, 20 mM acetic acid, 1 mM Na2 EDTA) was used for the elec-
trophoresis. The lanes in DGGE were duplicated for each treatment.
3. Results and discussion
3.1. X-ray radiography during the infiltration process
Fig. 2 shows the flow paths highlighted by the contrast agent.
The two radiographs were obtained from the same soil, and show
the differences in flow domains during saturated flow and −3 kPa
unsaturated flow. Black areas correspond to the imbibed water
pathways. The vertical, straight flow paths in the left figure (sat-
urated flow) denote the bypass flow induced by the artificial
macropores, while they disappear when unsaturated flow occurred
because matrix (micropore) flow was dominant in these condi-
tions. The same process was successfully achieved here as reported
by Mori and Higashi (2009), with bypass flow being dominant
in saturated flow conditions, and matrix flow being dominant in
unsaturated flow conditions. Nutrients would be expected to be
delivered to finer pores in the convection and dispersion process.
Thus contaminants present in the finer pores would be expected to
be effectively degraded by microbes.
240 Y. Mori et al. / Ecological Engineering 51 (2013) 237–243

Fig. 2. X-ray radiography under different suction controls. White straight lines at the left figure show macropore-induced bypass flow. Photos were taken at soil surface to
5 cm depth.

3.2. Biostimulation experiments with the FDA hydrolysis activity

method

The results of biostimulation experiments with the FDA hydrol-

ysis activity method for effluent are shown in Fig. 3. Fig. 3
(upper) shows effluent volume measurements during the bios-
timulation experiments. Flow rates were successfully controlled
for one month during the bioremediation experiment at 240
and 24 ml day−1 corresponding to flow rates of 1.4 × 10−4 and
1.4 × 10−5 cm s−1 , respectively. Nutrient injection was controlled
at about 1/10 of the saturated hydraulic conductivity, which
maintained the unsaturated boundary condition. These rates corre-
sponded well with hydraulic conductivity values for undisturbed
macroporous field soils at 0 (saturated flow) and −3 kPa suction
(unsaturated flow), respectively (Mori and Higashi, 2009).
Most of the columns, except for the case of saturated flow
without macropores, maintained a stable outflow rate during the
experiments. The column corresponding to saturated flow without
macropores showed a decrease in effluent volume (Fig. 3) and sur-
face ponding was observed after several days. It never recovered to
the original drainage rate.
Fig. 3 (lower) shows biological activity in the effluent water
during the biostimulation experiments. Generated fluorescein did
not show significant differences until 16 days had elapsed, after
which fluorescein activity markedly increased day by day under
the condition of unsaturated flow with macropores. Fig. 2 sug-
gests that, under unsaturated flow conditions, nutrients were
successfully delivered into the finer pores by dispersion processes
accompanying matrix flow, while avoiding bypass flow. Even solute
distribution in the soil columns was achieved regardless of the pore
structures.
There were clear characteristics of an increasing trend of
biological activity, in which fluctuations were observed. There
was a generally increasing trend for the unsaturated flow with
macropores condition, but biological activity sometimes returned
to base line values during the overall increasing trend. Possi-
Fig. 3. Effluent volume (upper) and biological activity (lower) in the effluent during
ble mechanisms for this are unclear. However, with increased the bioremediation experiments after Fujihara et al. (2010).
microbial biomass, they may develop colonies and biofilms along
the soil pores, which may form and disappear from time to The observed biological activity in the FDA hydrolysis assay was
time. Alternatively, microbes involved in this bio-remediation may confirmed by the nucleotide-based PCR-DGGE analysis. The results
change their family groups, causing fluctuations in microbiological of PCR-DGGE analysis of the oil-contaminated and uncontaminated
activity. soils showed some specific bands induced by the oil-contamination
 Y. Mori et al. / Ecological Engineering 51 (2013) 237–243 241

Fig. 5. EC fluctuations in the effluent solutions.

3.4. Water content distribution and biological activity in the soils

Fig. 6 shows water content distribution and FDA hydrolysis

Fig. 4. DGGE banding pattern of amplified 16S rRNA gene fragments from biostim-
activity in the soil columns at the conclusion of the experiments.
ulated soils. Lanes were duplicated for each sample.
Biological activity is considered to be enhanced at water contents of
0.6–0.8 of saturation (Nishio, 1989). For the saturated column with-
out macropores, water content exceeded saturation (Fig. 6), with
ponding observed at the soil surface during the experiment, and
treatment (Fig. 4). Bacterial 16S rRNA gene fragments repre-
Fig. 3 (upper) also showed a trend of decreasing effluent volume per
sented by band (D) and band (E) were selectively amplified in
day, suggesting clogging. On the other hand, the unsaturated flow
the oil-contaminated soil (Fig. 4). This indicated that some oil-
columns showed applicable water contents for bioremediation. For
tolerant bacterial species had increased their population in the
unsaturated flow with macropores, water contents at almost all
oil-contaminated soil. The band (D) had a sequence similarity to
sampled depth positions were appropriate for biological activity.
Rhodococcus erythropolis that has distinctive abilities, including
However, the water content slightly increased toward the surface
hydrocarbon oxidation (Lang and Philp, 1998) and bioflocculant
for unsaturated flow without macropores.
production (Kurane et al., 1995). Hydrocarbon oxidation positively
works for bioremediation while bioflocculant production nega-
tively works for infiltration. In contrast, band (F) and band (G) were
eliminated by the oil contamination (Fig. 4), implying growth inhi-
bition by the oil-contamination treatment. Band (A) and band (B)
were not affected by the oil contamination treatment (Fig. 4). Band
(C) was specifically amplified in the oil-contaminated soil sam-
ples, but the intensity of the bands for duplicated samples was
quite different (Fig. 4). The fluctuation in gene amplification implies
unstable growth conditions for the oil-degrading species in the
soils.

3.3. EC measurement

Fig. 5 shows EC fluctuations in the effluent solutions. The satu-

rated flow conditions with and without macropores show different
curves. EC values remained different from the unsaturated flow
conditions even when the horizontal axis was expressed as pore
volume (effluent volume/water content in the soil column). Prob-
ably, bypass flow occurred under saturated flow conditions with
macropores, while clogging occurred under the saturated flow con-
dition without macropores as shown in Fig. 3 (upper). On the other
hand, for unsaturated flow conditions the effluent EC curves were
very similar to each other for the totally different pore structures.
As shown in Fig. 2 and reported by Mori and Higashi (2009), bypass
Fig. 6. Volumetric water content and biological activity in the soil columns after the
flow was effectively prevented and matrix flow successfully con- bioremediation experiments. Vertical lines denote 60, 80 and 100% of soil saturation,
ducted the solutes to the finer pores. while L, M and U denote lower, middle and upper positions in the columns.
242 Y. Mori et al. / Ecological Engineering 51 (2013) 237–243

Table 3
Amount of nutrients leached out from the column.

NO3 − (mg) PO4 3− (mg)

Unsat. w/o macropore 362 0

Sat. w/o macropore 409 213
Unsat. w/ macropore 290 0
Sat. w/ macropore 2676 526

There was a trend in water content distribution in the

columns, that is, water content increased toward the surface in
the without-macropore column while it decreased in the with-
macropore column. Although the macropore volume (r = 0.05 cm,
length = 10 cm, number = 7, V = 0.55 cm3 ) has minor effects on water
content measurement, the macropore structure affected water
content distribution in the soil profile.
Biological activity in the soil column (Fig. 6 vertical axis) was
relatively high for unsaturated flow compared with saturated flow
conditions. Saturated flow has a risk of bypass flow and resul-
tant nutrient loss, which are shown in Table 3. Unsaturated flow
with macropore conditions showed highest biological activity with
the matrix flow without bypass flow shown in Fig. 2. Unsaturated
Fig. 7. Biological activity in the soil columns after the bioremediation experiments.
flow without macropore conditions probably also had enhanced
L, M and U denote lower, middle and upper positions in the columns.
dispersion, but showed slightly less activity than the unsatu-
rated flow with macropore conditions. Fig. 6 shows an increasing
water content toward the soil surface for unsaturated flow without efficiency for transporting solutes to finer pores, as well as mini-
macropores, which indicates the formation of clogging. mizing clogging.
Bioremediation, which enhances biological activity, also causes Remediation efficiencies were similar under unsaturated con-
increasing microbial numbers with a consequent increased risk of ditions with and without macropores. Macropores sometimes
clogging (Thullner, 2010). Seki et al. (1998) reported rapid bio- seriously affect the bioremediation process because of bypass flow.
clogging formation at the soil surface. If there were macropores However, as long as bypass flow was prevented, similar effective-
through the column, they will avoid clogging, which enables infil- ness was obtained even where there were large differences in
tration over longer times. Also, macropores have advantages for air structure. Also, in this experiment we found that even remedia-
conduction (Pivetz and Steenhuis, 1995). At the beginning of this tion efficiency through the column was achieved with unsaturated
research, we assumed macropores were undesirable because they flow with macropores, and uneven remediation with unsaturated
potentially caused bypass flow and resultant nutrient loss. Thus flow without macropores. We investigated the effect of infiltration
we developed infiltration techniques to avoid bypass flow. When process control on bioremediation over a 1-month period. Tak-
bypass flow was regulated, however, Figs. 3 and 6 suggest that ing into consideration that the soil column without macropores
macroporous structure maintained the designed infiltration rate, showed signs of clogging, while columns containing macropores
kept an appropriate water content without clogging, and probably showed increasing biological activity, it is likely that the observed
worked as pathways for air transport. As long as bypass flow is regu- differences would become more pronounced with longer-term
lated, macropores enhance bioremediation by preventing clogging experiments.
and helping air intrusion. It is easy to imagine that the higher the suction, the less
the chances of bypass flow occurring. Thus it was interesting
and potentially quite beneficial that the simple technique of
3.5. Efficiency of bioremediation

Fig. 7 shows the remediation results for different flow and struc-
ture conditions. As expected from Fig. 6, the unsaturated condition
shows higher remediation results than the saturated flow condi-
tion. Moreover, unsaturated flow with macropores shows even
remediation results throughout the whole column. On the other
hand, soil columns without macropores show uneven remediation
results for each position in the column. This is probably because the
macropore structure avoided clogging, in addition to bypass flow
prevention.
Finally, bioremediation efficiency was also calculated as the
remediated oil concentration divided by the applied nutrient vol-
ume (Fig. 8). As expected, unsaturated flow conditions showed
higher remediation efficiencies compared with saturated flow con-
ditions. Remediation efficiency was lower under saturated flow
conditions, because large amounts of nutrients leached out from
the column without contributing to biological activation (Table 3).
Also, when saturated flow continued, either bypass flow or risk
of clogging increased. Unsaturated flow therefore provides greater Fig. 8. Remediation efficiency based on applied nutrients.
 Y. Mori et al. / Ecological Engineering 51 (2013) 237–243
applying a small suction (−3 kPa) enhanced dispersion regardless
of the macropore structure.
4. Conclusions
To investigate effective bioremediation processes, biostimula-
tion experiments were conducted for a soil with different pore
structures (with and without macropores) and different nutrient
injection conditions (saturated flow and unsaturated flow). Biolog-
ical activities along with water contents were measured to examine
appropriate conditions for enhancing biodegradation. The follow-
ing conclusions were obtained for the examined soils:
1. Flow rates were successfully controlled for 1 month during
the bioremediation experiment. X-ray radiography showed that
bypass flow and matrix flow were successfully controlled using
a relatively simple technique.
2. Biological activities in the effluent and soil were greatest under
unsaturated flow conditions with macropores because unsat-
urated conditions prevented bypass flow in the soil, and the
macropores prevented clogging.
3. Water contents after the experiments showed that 0.6–0.8 of the
saturated water content were almost achieved for the unsat-
urated flow condition with macropores. These water contents
enhanced biological activity.
4. Resultant remediation efficiencies per unit volume of imbibed
nutrients for unsaturated flow were higher than for saturated
flow conditions and they were similar for unsaturated conditions
with and without macropores. However, the soil column without
macropores had a higher water content near the soil surface,
suggesting the formation of clogging.
5. Macropores were useful in enhancing biological activity while
avoiding clogging, as long as the infiltration process was con-
trolled.
X-ray radiography observations reflected the solute trans-
port processes well. Relatively simple techniques were used for
more than a month to beneficially minimize bypass flow and
allow matrix flow. Controlling the solute transport processes
successfully enhanced biostimulation. We intend to investigate
contaminant remediation in situ and for longer periods in future
studies.
Acknowledgements
The authors are grateful to bachelors students who supported
the flow experiment for this paper in 2005. We are also grateful
to Ayayo Ohira for her support with the biological experiment.
This work was partially supported by the Japan Society for the
Promotion of Science, NEXT program (GS021), 2011–2014, and a
Grant-in-Aid for Scientific Research (C), 18510074, 2006–2008.
References
Balba, M.T., Al-Awadhi, N., Al-Daher, R., 1998. Bioremediation of oil-contaminated
soil: microbiological methods for feasibility assessment and field evaluation. J.
Microbiol. Methods 32 (2), 155–164.
Beven, K.J., Germann, P.F., 1982. Macropores and water flow in soils. Water Resour.
Res. 18, 1311–1325.
243
Central Environmental Council, 2006. A Guideline for Oil Contamination – Real
Estate Correspondence for Stink and Oil Film Issue Originated from Mineral Oil
Contamination. U.S. Central Environmental Council.
Darcy, H., 1856. Les Fontaines Publiques de la Ville de Dijon. Dalmont, Paris.
Fujihara, A., Mori, Y., Matsumoto, Y., 2010. Enhancing biological activity for biore-
mediation by controlling nutrient infiltration area. Jpn. Soc. Civil Eng. 66 (4),
187–193.
Gerke, H.H., van Genuchten, M.T., 1993. A dual-porosity model for simulating the
preferential movement of water and solutes in structured porous-media. Water
Resour. Res. 29 (2), 305–319.
Ministry of the Environment, Government of Japan, 2002. Soil Contamination Coun-
termeasures Act (Act No. 53 of 2002). Soil Environment Management Division,
Environmental Management Bureau.
Hillel, D., 1998. Environmental Soil Physics. Academic Press, San Diego, pp. 677–690.
Huong, N.L., Itoh, K., Suyama, K., 2008. 2,4-Dichlorophenoxyacetic acid (2,4-D)- and
2,4,5-trichlorophenoxyacetic acid (2,4,5-T)-degrading bacterial community in
soil–water suspension during the enrichment process. Microbes Environ. 23
(2), 142–148.
Ishiguro, M., 1992. Ion transport in soil with ion exchange reaction: effect of distri-
bution ratio. Soil Sci. Soc. Am. J. 56, 1738–1743.
Klute, A., Dirksen, C., 1986. Hydraulic conductivities and diffusivity: laboratory
methods. Methods of soil analysis. Part 1. In: Klute, A. (Ed.), Physical and Min-
eralogical Methods. , 2nd ed. ASA-SSSA, Madison, WI, pp. 687–734.
Köhne, J.M., Mohanty, B.P., 2005. Water flow processes in a soil column with a cylin-
drical macropore: experiment and hierarchical modeling. Water Resour. Res. 41,
W03010, http://dx.doi.org/10.1029/2004WR003303.
Kurane, R., Hatamochi, K., Kakuno, T., Kiyohara, M., Tajima, T., Hirano, M., Taniguchi,
Y., 1995. Chemical structure of lipid bioflocculant produced by Rhodococcus
erythropolis. Biosci. Biotechnol. Biochem. 59, 1652–1656.
Kutilek, M., Nielsen, D., 1994. Transport of solutes in soils. In: Soil Hydrology.
CATENA VERLAG, Germany, pp. 274–324.
Lang, S., Philp, J.C., 1998. Surface-active lipids in rhodococci. Antonie Van Leeuwen-
hoek 74, 59–70.
Mori, Y., Higashi, N., 2009. Controlling solute transport processes in soils by using
dual-porosity characteristics of natural soils. Colloid Surf. A: Physicochem. Eng.
Aspects 347 (1–3), 121–127.
Mori, Y., Iwama, K., Maruayma, T., Mitsuno, T., 1999a. Discriminating the influence
of soil texture and management-induced changes in macropore flow using soft
X-rays. Soil Sci. 164, 467–482.
Mori, Y., Maruyama, T., Mitsuno, T., 1999b. Soft X-ray radiography of drainage pat-
terns of structured soils. Soil Sci. Soc. Am. J. 63, 733–740.
Natsch, A., Keel, C., Troxler, J., Zala, M., Von Albertini, N., Defago, G., 1996. Importance
of preferential flow and soil management in vertical transport of a biocon-
trol strain of pseudomonas fluorescens in structured field soil. Appl. Environ.
Microbiol. 62 (1), 33–40.
Nishio, M., 1989. Fundamental of Soil Microbes. Rural Culture Association, Tokyo,
pp. 166–185.
Pivetz, B.E., Steenhuis, T.S., 1995. Soil matrix and macropore biodegradation of 2,4-D.
J. Environ. Qual. 24, 564–570.
Richards, L.A., 1931. Capillary conduction of liquids through porous mediums.
Physics 1 (5), 318–333.
Ronkanen, A.K., Klove, B., 2009. Long-term phosphorus and nitrogen removal pro-
cesses and preferential flow paths in Northern constructed peatlands. Ecol. Eng.
35 (5), 843–855.
Russell, D.L., 1992. Remediation Manual for Petroleum-Contaminated Sites. Tech-
nomic Publishing Co., Lancaster, PA, USA.
Schnurer, J., Rosswall, T., 1982. Fluorescein diacetate hydrolysis as a measure of total
microbial activity in soil and litter. Appl. Environ. Microbiol. 43 (6), 1256–1261.
Seki, K., Miyazaki, T., Nakano, M., 1998. Effects of microorganisms on hydraulic
conductivity decrease in infiltration. Eur. J. Soil Sci. 49 (2), 231–236.
Shashidhara, Y.M., Jayaram, S.R., 2010. Vegetable oils as a potential cutting fluid—an
evolution. Tribol. Int. 43 (5–6), 1073–1081.
Soil Science Society of America, 1996. Glossary of Soil Science Terms. Soil Science
Society of America, Madison, WI.
Thullner, M., 2010. Comparison of bioclogging effects in saturated porous media
within one- and two-dimensional flow systems. Ecol. Eng. 36 (2), 176–196.
Toride, N., Inoue, M., Leij, F.J., 2003. Hydrodynamic dispersion in an unsaturated
dune sand. Soil Sci. Soc. Am. J. 67, 703–712.
U.S. Environmental Protection Agency, 1980. The Comprehensive Environmental
Response, Compensation and Liability Act of 1980. EPA, U.S.A.
U.S. Environmental Protection Agency, 1988. Technology Screening Guide for Treat-
ment of CERCLA Soils and Sludges. EPA/540/288/004.
U.S. Environmental Protection Agency, 1990. Handbook on In Situ Treatment of
Hazardous Waste Contaminated Soils. EPA/540/2-90/002.
Vidali, M., 2001. Bioremediation. An overview. Pure Appl. Chem. 73 (7), 1163–1172.