Environ Monit Assess (2015) 187: 437 DOI 10.1007/s10661-015-4633-0
Risk assessment of gas oil and kerosene contamination on some properties of silty clay soil M. Fallah & M. Shabanpor & M. Zakerinia & S. Ebrahimi
Received: 4 January 2015 / Accepted: 21 May 2015 / Published online: 18 June 2015 # Springer International Publishing Switzerland 2015
Abstract Soil and ground water resource pollution by petroleum compounds and chemical solvents has multiple negative environmental impacts. The aim of this research was to investigate the impacts of kerosene and gas oil pollutants on some physical and chemical properties, breakthrough curve (BTC), and water retention curve (SWRC) of silty clay soil during a 3-month period. Therefore, some water-saturated soils were artificially contaminated in the pulse condition inside some glassy cylinders by applying half and one pore volume of these pollutants, and then parametric investigations of the SWRC were performed using RETC software for Van Genukhten and Brooks-Corey equations in the various suctions and the soil properties were determined before and after pollution during 3 months. The results showed that gas oil and kerosene had a slight effect on soil pH and caused the cumulative enhancement in the soil respiration, increase in the bulk density and organic M. Fallah (*) Faculty of Agriculture, Guilan University, Rasht, Iran e-mail: [email protected] M. Shabanpor Department of Soil Science, Guilan University, Rasht, Iran e-mail: [email protected] M. Zakerinia Department of Water Engineering, Gorgan University, Gorgan, Iran e-mail: [email protected] S. Ebrahimi Department of Soil Science, Gorgan University, Gorgan, Iran e-mail: [email protected]
matter, and reduction in the soil porosity and electrical and saturated hydraulic conductivity. Furthermore, gas oil retention was significantly more than kerosene (almost 40 %) in the soil. The survey of SWRC indicated that the contaminated soil samples had a little higher amount of moisture retention (just under 15 % in most cases) compared to the unpolluted ones during this 3-month period. The parametric analysis of SWRC demonstrated an increase in the saturated water content, Θs, from nearly 49 % in the control sample to just under 53 % in the polluted ones. Contaminants not only decreased the residual water content, Θr, but also reduced the SWRC gradient, n, and amount of α parameter. The evaluation of both equations revealed more accurate prediction of SWRC’s parameters by Van Genukhten compared to those of Brooks and Corey. Keyword Break through curve . Gas oil . Kerosene . Soil properties . Water retention curve
Introduction Soil contamination is caused by the presence of chemicals or other alteration in the natural soil environment. In fact, soil and water pollution refers to an increased level of harmful human-made chemicals in the natural soil and groundwater environment (Norris et al. 1994). It is typically caused by industrial activity, agricultural chemicals, or improper disposal of waste
437 Page 2 of 13
material. Petroleum products and by-products have been widely used chemicals in recent years. Leakage from underground storage tanks and accidental spills of hydrocarbons, such as light nonaqueous phase liquids (LNAPLs), pose a serious risk of subsurface contamination (Sharma and Mohamed 2003a). As petroleum hydrocarbons release in the environment, these pollutants can be absorbed by the soil particles and move into the soil body and affect some physical, chemical, and biological properties of soil (Marins Pala et al. 2006; Mohan et al. 2007). A LNAPL migrates downwards under the influence of gravity and viscous and capillary forces, contaminating the subsurface region around the spill as well as deteriorating the quality of groundwater (Sharma and Mohamed 2003b). The study of the breakthrough curve (BTC) is one of the important ways to investigate the emission of pollutants through the soil porous media. In fact, it is a plot of relative concentration versus time, where relative concentration is defined as C/Co with C as the concentration at a point in the groundwater flow domain, and Co as the source concentration (Haj Abasi 2007). Water retention curve (SWRC) is an applicable method to study on the behavior of soil-pollutant system. SWRC shows the relationship between the water content, θ, and the soil water potential, ψ. This curve is characteristic for different types of soil and is also called the soil moisture characteristic. It is used to predict the soil water storage, water supply to the plants (field capacity), and soil aggregate stability. Because of the hysteretic effect of water filling and draining the pores, different wetting and drying curves may be distinguished (Gupta and Larson 1979). In recent years, several researchers have been worked on the hydrocarbon contaminant emission in the porous media of soil, the BTC of the petroleum hydrocarbons, and their fate in the soil composition, but more consideration and research should be taken into account on the moisture retention capacity in the soil-pollutant composition (Ebrahimi et al. 2010). In a study which has been carried out by Burckhard et al. (2004), a comparison of the contaminated SWRC with the uncontaminated SWRC showed that petroleum pollution not only prevented the movement of soil water but also caused some changes on the hydraulic properties of soil. Moreover, the contaminated soil samples
Environ Monit Assess (2015) 187: 437
had a higher water-holding capacity compared to the uncontaminated ones at higher suctions (Burckhard et al. 2004). In recent years, interest in the unsaturated zone has significantly increased because of growing evidence and public concern that the quality of the subsurface environment is being adversely affected by the petroleum contaminants. These days, applications of computer models have been common in many research and management to predict the movement of water and chemicals into and through the unsaturated zone of soils. The RETC computer code is one of the software to analyze the SWRC and hydraulic conductivity functions of unsaturated soils (Van Genuchten et al. 1991). These hydraulic properties are key parameters in any quantitative description of the water flow into and through the unsaturated zone of soils (Van Genuchten and Nielsen 1991). Iran is a major oil-producing country in the world with an annual production capability of large amounts of crude oil and has been exposed to the contamination by these hydrocarbon components through the extraction, transportation, and refining process (Ebrahimi 2009). In this regard, the need for more comprehensive studies to consider and solve this problem is inevitable. Hence, the aim of this research was to determine the impact of kerosene and gas oil as available hydrocarbon pollutants on some physical and chemical properties of a silty clay soil, BTC, and SWRC. Then, the results from SWRC were studied by RETC software quantitatively and parametrically.
Methods and material Soil In this study, the silty clay soil was taken from an area in the University of Gorgan located in Golestan Province, Iran. After the homogenization of the soils, some physical and chemical properties of soils were measured before artificial applying of the pollutants and within the equilibrium time (3 week after polluting and the first 3 months (Ebrahimi 2009) such as the bulk density (BD) using clod method (Brasher et al. 1966), the actual density of soil using pyknometer, pH of the soil saturation extract using the pH meter, the soil
Environ Monit Assess (2015) 187: 437
saturation extract electrical conductivity using the gauges conductivity apparatus (Page et al. 1987), the hydraulic conductivity using RETC code (Van Genuchten et al. 1991), the organic matter using Walkey and Block method (Walkey and Black 1934), the soil respiration using Stotzky methods (Stotzky 1965), and porosity using calculation from the bulk density and actual density of soil). Petroleum hydrocarbon contaminant Kerosene and gas oil were chosen as soil pollutants because of abundant consumption of these contaminants and subsequently their contaminations in the soil porous media especially in Iran as one of the main oilproducing countries. Artificial contamination method of soil and determination of BTC The contaminants were applied separately on the silty clay soils in the pilot scale and pulse condition. Hence, the equipment including glassy cylinders were designed in the height of 120 cm and the diameter of 16 cm with the filter and special outlet at the bottom of the cylinders and the bulk density and porosity of soils were determined (Fallah et al. 2013). The glassy cylinders were filled with the soil homogeneously in the given depth based on the silty clay soil bulk density, and then, soil samples were saturated by water from the top to the bottom of soil column. In this stage of procedure, the amounts of half and one pore volume porosity of silty clay soil, kerosene, and gas oil were applied separately on the saturated soil and let the pollutants pass through the soil. Eventually, output volume of gas oil, kerosene, and water was measured per time and breakthrough curves were drawn (Fallah et al. 2012a, 2013). Then, the polluted samples of soil were placed outdoors and these samples were allowed to reach equilibrium with the natural environment (Ebrahimi 2009). Petroleum hydrocarbon analysis The total petroleum hydrocarbon (TPH) concentration was measured by dichloromethane according to US
Page 3 of 13 437
Environmental Protection Agency within a certain time (EPA 2000).
SWRC analysis After polluting the soil, soil samples were collected from the contaminated soils within the equilibrium time (3 weeks after contamination), first, second, and third months, and unpolluted soil which was considered as a control sample. Then, the pressure plate apparatus (Tuller and Dani 2003) was used to determine the soil water retention curves in 0-, 0.3-, 0.5-, 1-, 5-, and 15-bar suction within the pre-contamination time (control sample); the equilibrium time, first, second, and third months after the soil contamination; and the SWRCs were drawn by Microsoft Excel. Then, data obtained from SWRC was analyzed by RETC software. The RETC computer code is the computer software to analyze the SWRC and hydraulic conductivity functions of unsaturated soils (Van Genuchten et al. 1991). These hydraulic properties are key parameters in any quantitative description of water flow into and through the unsaturated zone of soils. The program uses the parametric models of Brooks-Corey and Van Genuchten to represent the SWRC and the theoretical pore-size distribution models of Mualem and Burdine to predict the unsaturated hydraulic conductivity function from the observed soil water retention data.
Table 1 Some soil properties Sample Texture
Silty clay
Clay (%)
50
Silt (%)
43.3
Sand (%)
6.7
Bulk density (g/cm3) 3
1.26
Actual density (g/cm )
2.35
Porosity (%)
47
Organic carbon (%)
1.22
Organic matter (%)
2.10
pH
7.73
EC (ds/m)
1.12
Respiration (mg CO2/day)
2.1
437 Page 4 of 13
Environ Monit Assess (2015) 187: 437
Fig. 1 Output volume ratio of kerosene and gas oil to water for half pore volume of silty clay soil, pollutant
Brooks and Corey (1964) used this three-parametric equation to describe the SWRC: θw ¼ θr þ ðθs −θr Þ
a b Ψ
ð1Þ
where a is bubbling pressure or potential at air entire value, Θr is residual moisture, Θs is saturation moisture, and Θw is volumetric moisture of soil. This equation can be used in some situations that Ψ is larger than or equal to a. If the potential is less than a, Θw is, then, equal to Θs (Ebrahimi Fig. 2 Output volume ratio of kerosene and gas oil to water for one pore volume of silty clay soil, pollutant
2004, 2009).Van Genukhten (1980) introduced the following four-parametric equation: θs −θr c 1 þ aΨ b This equation can be written as
θw ¼ θr þ
θw ¼ θr þ h
θs −θr b ic 1 þ Ψa
ð2Þ
ð3Þ
where b and c are dimensionless parameters. The slop factor coefficient b (=n) changes the curve shape around
Environ Monit Assess (2015) 187: 437
Page 5 of 13 437
Table 2 Mean comparison of bulk density (g cm−3) and porosity (%) for both pollutants in different time Time
Silty clay soil Bulk density
Porosity
1/262d
0/47a
Control Equilibrium
a
1/298
0/454d
First month
1/287b
0/459c
c
0/464b
c
1/270
0/466b
0/009
0/003
Second month
1/274
Third month e
LSD (5 %) a‐d e
Show the difference between values of parameters
Least significant difference
the turning point. The a (=a) and c (=m) change the curve shape around the bubbling pressure point. It is assumed that a (=a) is the function of matric potential in r a point with the water content of θs þθ 2 . Suction at this point can be shown by Ψ50. If, in the above equation, θw ¼
ðθs þθr Þ 2
a¼
and Ψ=Ψ50, then Ψ 50
2
ð1=cÞ
−1
ð4Þ
ð1=bÞ
In the last, Van Genuchten and Brooks-Corey equation parameters for each sample were calculated by RETC code (Ebrahimi 2004, 2009). In this study, assuming m=1−1/n, the α, θr, θs, and n parameters were estimated for these equations by the RETC program (Van Genuchten et al. 1991).
Statistical design Statistical analysis was performed with three replications based on a split plot design in time for the silty clay soil with two contaminants (gas oil and kerosene) and two levels of pollution (half and one pore volume of soil). SPSS software was used to study the correlation, analysis of variance, and comparison of treatment mean.
Results and discussions Some initial properties of soil were determined before the pollution. Table 1 represents the properties of silty clay soil. According to Table 1, the soil studied in this research was silty clay one (with 50 % clay) and it was considered as a heavy-textured soil which was in the neutral acidity range. The porosity of silty clay soil was also 47 % and had a nearly high content of organic matter compared to the light-textured soil. The electrical conductivity of silty clay soil showed that this soil had a slight salinity and was not in the range of salty soils. Pollutants emission through the soil column The same volume of kerosene and gas oil was used for polluting the soil because of the same porosity of the soil columns. The results indicated that after a short delay, the amount of gas oil and kerosene was observed in the drainage water output from the silty clay soil while the
Table 3 Mean comparison of different levels of contaminants on the pH values in different time Pollution
Time Control
Equilibrium
First month
Second month
Third month
0.5 VPGe
7/74a
7/62c
7/75a
7/33b
7/71a
f
a
a
a
a
0.5 VPK
7/74
7/92
7/70
7/69
7/37b
1 VPGg
7/74a
7/83b
7/40b
7/73a
7/45b
a
b
b
a
h
1VPK
7/74
7/83
7/40
7/73
7/45b
LSD (5 %)
–
0/024
0/098
0/156
0/16
a‐d
Show the difference between values of parameters
e
Half pore volume of gas oil
f
Half pore volume of kerosene
g
One pore volume of gas oil
h
One pore volume of kerosene
437 Page 6 of 13 Table 4 Mean comparison of various pollution levels on the electrical conductivity in different time (ds/m)
Environ Monit Assess (2015) 187: 437
Pollution
0.5VPGa b
0.5VPK
a‐d
Show the difference between values of parameters
Time Control
Equilibrium
First month
Second month
Third month
1/12
0/598a
0/464b
0/596a
0/704a
1/12
ab
a
a
0/677a
b
a‐d
Show the difference between values of parameters
b
0/618
1/12
0/456
0/417
0/453
0/526b
1 VPKd
1/12
0/591a
0/594a
0/643a
0/680a
LSD (5 %)
–
0/105
0/62
0/52
0/78
Pollution
b
0/594
1 VPG
pollution emission enhancement rate was followed by reduction in the rate of water drainage output. Figure 1 showed that in the application of half volume of pollutants, the gas oil retention was more than the kerosene and kerosene flowed out of the silty clay soil by sharper gradient and less retardation and higher velocity (Fallah et al. 2012a, 2013). Figure 2 showed the same procedure as Fig. 1; on the other hand, at the same time, kerosene to water output ratio was substantially more than gas oil to water ratio (Fallah et al. 2012a, 2013). The only difference between the two graphs was relevant to the application of different amounts of pollutants (half and one pore volume) which brought about more pollutant emission from the soil in the usage of one pore volume porosity of silty clay soil, pollutants (Fig. 2) at the same time (approximately double). Emission rate and permeability of petroleum contaminants through the porous media of soil are dependent on the contaminant type, particle size, bulk density, porosity, and compaction of soil, and the result from the emission rate and permeability for these pollutants was in accordance with the results reported by other scientists on the emission rate of petroleum hydrocarbons through the porous medium of soil (Wang et al. 2000; Wang and Shao 2009). Table 5 Mean comparison of various pollution levels on the soil organic matter (percentage) in different time
0/50
c
Survey of changes on some physical and chemical properties and microbial respiration of soil After applying half and one pore volume of gas oil and kerosene pollutants, microbial respiration and some physical and chemical properties of soil samples were examined within the mentioned period of time and the results were analyzed by split-plot design in time with SPSS software. Then, analysis of variance for silty clay soil showed where the main effect (time and pollution) was significant and when the interaction effect was not substantial (time in pollution), mean comparison of the main effects was performed (Table 2). Where the interaction effects were significant, it was not necessary to do the mean comparison of main effects individually, but it was better to compare the different levels of contamination at any time, and in the following, the SWRCs were drawn by Microsoft Excel and analyzed by RETC program. The effect of gas oil and kerosene on the soil bulk density and porosity Considering that the only main effect of time was significant for the bulk density and porosity and there was no effect of pollutant type, the main comparison of mentioned parameter values (mean
Time Control
Equilibrium
First month
Second month
Third month
0.5VPG
2/10a
6/26b
6/05b
5/77b
5/09b
0.5VPK
a
2/10
d
4/37
c
4/25
d
3/98
3/69d
1 VPG
2/10a
8/47a
7/61a
6/79a
6/85a
1 VPK
a
2/10
c
4/98
c
4/75
c
4/58
4/17c
LSD (5 %)
–
0/213
0/543
0/172
0/286
Environ Monit Assess (2015) 187: 437 Table 6 Mean comparison of pollution levels on the microbial respiration values (mg CO2/day)
a‐d
Show the difference between values of parameters
Pollution
Page 7 of 13 437
Time Control
Equilibrium
First month
Second month
Third month
0.5VPG
0/154a
0/119c
0/165c
0/152c
0/149b
0.5VPK
a
b
b
0/190
b
0/167a
c
c
c
0/154
a
0/134
1 VPG
0/154
0/121
0/17
0/157
0/152b
1 VPK
0/154a
0/149a
0/213a
0/19a
0/181a
LSD (5 %)
–
0/013
0/0078
0/0091
0/0144
tended to remain in the neutral range. The result from laboratory tests revealed that there was a shift toward neutral in pH of different types of soils after being contaminated by the petroleum hydrocarbon contaminants (Everett 1978). Another research showed that the pH status of soil in the contaminated (heavy impact and moderate impact) zones varied from acidic to near neutral (Ahmadi et al. 1996).
values of these parameters for half and one pore volume of kerosene and gas oil in each time individually) was performed in different time (Table 2). According to Table 2, the most and least amounts of alteration in the bulk density and porosity were related to the equilibrium time and 3 months after the equilibrium, respectively; furthermore, a significant difference was not observed between the second and third months for these factors in the silty clay soil. A research which was conducted by Kayode et al. (2009) showed that high concentration of hydrocarbon pollutants affected and altered the physical and chemical properties of soil, increased the soil bulk density a little, and reduced the porosity of soil and root penetration.
The effect of gas oil and kerosene on the electrical conductivity of the soil Survey of mentioned contaminants showed that since the interaction effect of the electrical conductivity was significant, the comparison of pollution levels was done in different time. As it can be observed from Table 4, the contaminants caused the reduction in the electrical conductivity and over time, the electrical conductivity has slightly increased and the most increase was related to the third month after pollution; moreover, no significant difference was found between the use of different levels of contaminants at the same time within this period of time in the silty clay soil. The results from Table 4 were in accordance with the experiments conducted by
The effect of gas oil and kerosene on the soil pH The survey of contaminants impacts indicated that the interaction effect was significant for the soil pH, and the comparison of contamination levels was performed in different time (Table 3). Table 3 showed that during different period of time, application of gas oil and kerosene had no significant impact on the pH of silty clay soil. In fact, these contaminants actually made a slight change on the silty clay soil pH and the soil pH Table 7 Mean comparison of the pollution levels on the contaminant content in different time (mg/ 1000 g soil)
Pollution
0.5VPG
a‐d
Show the difference between values of parameters
0/180
Time Control
Equilibrium
First month
Second month
Third month
0
135b
91/66b
70b
63/3cb
d
c
0.5VPK
0
88
1 VPG
0
178a
1 VPK
0
c
115
LSD (5 %)
–
13/62
b
75
158a b
65
58/3c
128a
96/6a
93/33
b
75
68/3b
9/84
12/57
7/62
437 Page 8 of 13
Environ Monit Assess (2015) 187: 437
Fig. 3 SWRC for one pore volume of kerosene
Benka-Coker and Ekunday (1995) in Nigeria which demonstrated the electrical conductivity reduction in the petroleum-polluted soil. The effect of gas oil and kerosene on the organic matter of soil Table 5 represents that the organic matter has enhanced after the soil contamination in comparison with the control sample. The most increase was relevant to the application of one pore volume of gas oil compared to other values of the contaminants in the initial time after contamination. This increase was due to more gas oil retention and longer carbon chain of gas oil, and this increase showed a downward trend over time. The most increase in the Fig. 4 SWRC for half-pore volume of kerosene
soil organic matter was related to one pore volume of gas oil, half pore volume of gas oil, one pore volume of kerosene, and half pore volume of kerosene, respectively, in the early time after the contamination. The results of various researches on the effects of petroleum distribution showed an increase in the organic carbon and organic matter in the petroleum-polluted soils (Okonokhua et al. 2007; Wang et al. 2013).
The effect of gas oil and kerosene on the microbial respiration of soil Table 6 revealed that the soil microbial respiration showed reduction in the early time after the soil contamination.
Environ Monit Assess (2015) 187: 437
Page 9 of 13 437
Fig. 5 SWRC for one pore volume gas oil
After 1 month, soil microorganisms were accustomed to the new condition and the microbial respiration has increased. This increment in the microbial activity could be because of the microorganism population increase and that was probably due to the presence of oil-eating bacteria in the soil which referred to the usage of the hydrocarbon as a carbon resource (Seklemova et al. 2001; Fallah et al. 2012b). More volume of kerosene resulted in more soil respiration. As observed in Table 6, there was a declining trend of the soil microbial respiration from the first month to the third month. However, the cumulative increase in the microbial respiration in the contaminated soil indicated the microbial Fig. 6 SWRC for half-pore volume of gas oil
respiration increase in these soil samples (Fallah et al. 2012b; Labud et al. 2007). Variation of gas oil and kerosene values in the silty clay soil Table 7 represents the contaminant variation over time. As you see Table 7, the contaminant content in the gas oil-polluted samples was more than the kerosenepolluted ones. Over time, the amount of pollution has reduced and in the third month, it has reached the lowest level during this period. As observed, in the first 2 months, changes on the pollution level had a significant difference from each other. However;
437 Page 10 of 13
Environ Monit Assess (2015) 187: 437
Fig. 7 Mean comparison of different periods for the keroseneand gas oil-polluted soil
this difference between the levels of contaminants has been reduced in the next 2 months. The effect of petroleum pollutants on the SWRC Kerosene effect on the SWRC In the silty clay soil, the survey of the SWRC for half and one pore volume of kerosene showed that by increasing the suction, water started to flow out of the soil with slow gradient in the control soil. As it can be easily seen from Figs. 3 and 4, the most reduction in the soil water retention was relevant to the suction between 0 and 0.5 and at the higher suction, the soil water loss was less than lower suction in both contaminated soil and control sample. In addition, in the contaminated samples, the intensity of moisture loss was less than control sample, and the polluted soils had slightly higher volumetric moisture content in the different suctions. In fact, in the heavy texture soil, the moisture content of soil became more in the unpolluted soil in all applied pressure or in the natural condition without any suction. For one pore volume of kerosene (Fig. 3), the soil water Table 8 Van Genuchten equation parameters for the silty clay soil
Sample
Θr (L3 L−3)
retention was a bit more than half volume in different suction within this period of time. In Figs. 3 and 4, graphs showed decreasing gradient over time and this moisture reduction continued until the third month but never reached the initial value of water retention compared to the control sample during this period.
Gas oil effect on the SWRC Figures 5 and 6 illustrate the effect of gas oil emission on the SWRC for one pore volume of gas oil (Fig. 5) and half pore volume of gas oil (Fig. 6). As it can be observed from these figures, the SWRC followed exactly the same trend as Figs. 3 and 4. By increasing the suction in the control sample, the silty clay soil has lost its moisture by slow gradient in the various suctions due to approximately 50 % clay whereas the contaminated samples demonstrated less moisture loss than the uncontaminated soil and a little greater volume of the water retention in different suctions within various time.
Θs (L3 L−3)
α (L−)
Number
m
Ks (cm/day)
Control
0.157
0.49
0.056
1.24
0.19
22.26
1 VPG
0.102
0.526
0.0275
1.16
0.13
10.99
0.5VPG
0.108
0.512
0.0307
1.17
0.145
13.77
1 VPK
0.144
0.51
0.053
1.19
0.165
16.7
0.5VPK
0.117
0.507
0.036
1.18
0.15
16
Environ Monit Assess (2015) 187: 437 Table 9 Brooks-Corey equation parameters for the silty clay soil
Page 11 of 13 437
Sample
Θr (L3 L3)
Θs (L3 L3)
α (L−)
Number
m
Ks (cm/day)
Silty clay
0.056
0.423
0.0292
0.127
1
2.16
In addition, this moisture retention represented more increase particularly in the lower suction as an illustration, in the early time after contamination and higher level of pollution (one pore volume of gas oil compared to half volume of gas oil) in the lower suction. In other words, over time, the soil moisture content gradually declined; nevertheless, it has never attained the initial values of the water retention and held a bit more moisture compared to control sample similar to Figs. 3 and 4. Mean comparison of the water retention curve of different time for the contaminated silty clay soil compared to the control one The results of the mean comparison of the SWRC for half and one pore volume porosity of soil, kerosene, and gas oil (Fig. 7) demonstrated that the addition of pollutants into soil samples caused increase in the moisture retention capability in different suctions. According to Fig. 7, the maximum increase was related to one pore volume of gas oil and then, half pore volume of gas oil, one pore volume of kerosene, and half pore volume of kerosene, respectively, at all suctions during different periods of time. Parametric analysis of the soil-pollutant water retention curve After drawing the SWRC for the contaminated samples and control sample (uncontaminated soil) by Excel Microsoft, Brooks-Corey and Van Genuchten parameters were determined by RETC computer program. It was also given the nonlinear least-squares parameter optimization method to estimate the unknown coefficients in the hydraulic models. The program also predicted the hydraulic conductivity from the observed soil water retention data assuming that one observed conductivity value (not necessarily at saturation) was available. Van Genuchten model of Mualem and BrooksCorey equation parameters were determined using volumetric moisture content by RETC computer program (Van Genuchten et al. 1991). Results from RETC software for the parameters of Van Genuchten and BrooksCorey equation are presented in Tables 8 and 9 for the
silty clay soil contaminated by kerosene and gas oil hydrocarbon pollutants. The parametric study of soil-pollutant water retention curves indicated that the saturated hydraulic conductivity of the polluted soil samples has decreased due to higher viscosity of contaminants compared to water and lower capacity of soil retention for these fluids (Nouri et al. 2013). The saturation moisture (Θs) has increased by pollutant addition in the silty clay soil which was in accordance with the expectation. However, there is no significant difference between the amount of saturation moisture in the half and one pore volume of the applied kerosene on the silty clay soils by RETC. As it is observed from Table 8, the change on the residual moisture (Θr) showed decline by increasing the amount of contamination in the soil samples and the most and the least reduction in Θr was pertaining to the application of one pore volume of gas oil and one pore volume of kerosene, respectively. The enhancement in amount of pollutant treatment caused decrease in the soil-pollutant water retention curve slope, n parameter. The changes in α parameter are strictly dependant on the soil type. Increase in amount of pollutants caused enhancement in air pressure and decreased amounts of α parameter in the silty clay soil especially in the application of gas oil compared to kerosene pollutant. As it can be observable from Table 9, Brooks-Corey equation did not make any difference for the same textures with different conditions in terms of contamination with these pollutants and one type of data was only given for all soil conditions contaminated with gas oil and kerosene. Accordingly, Brooks-Corey equation cannot give so accurate and detailed data and correlation to evaluate various conditions.
Conclusion The result from this research represented that all apparent soil physical and chemical properties changed by emission of kerosene and gas oil on the soil surface. The soil analysis indicated a slight impact of gas oil and kerosene on the soil pH, and it had a tendency to shift toward neutral (Vonloocke et al. 1975; Ahmadi et al.
437 Page 12 of 13
1996) and caused the cumulative enhancement in the soil respiration (Fallah et al. 2012b; Labud et al. 2007), increase in the BD (Kayode et al. 2009), organic matter (Okonokhua et al. 2007; Wang et al. 2013), and the reduction in the soil porosity (Kayode et al. 2009) and the electrical (Benka-Coker and Ekunday 1995) and saturated hydraulic conductivity; moreover, the BTC results revealed that the gas oil retention was significantly more than that of kerosene and kerosene flowed out of the soil by sharper gradient and less retardation especially in the application of higher amount of kerosene (Wang et al. 2000; Wang and Shao 2009). The reason is that the reactivity of gas oil with the silty clay soil is substantially more than that of kerosene which is probably because of the structural form and its carbonic chain number, and on the other hand, it caused more intense surface absorption (Fallah et al. 2012a, 2013). The survey of SWRC showed higher amount of moisture retention in the contaminated soil samples compared to the unpolluted sample (Nouri 2012; Burckhard et al. 2004), and the maximum increase was related to one pore volume of gas oil and then, half pore volume of gas oil, one pore volume of kerosene, and half pore volume of kerosene, respectively, within this period. In fact, some parts of contaminants were strongly absorbed to the clay mineral as a residual pollutant and increased suction in the capillary pipes of clay and reduced the water loss from the soil, and some authors attributed this to the nonpolar characteristic of oil (Volk 1980). The parametric analysis of soil-pollutant water retention curve by RETC software for Van Genukhten and Brooks-Corey equations demonstrated increase in the saturated water content, Θs, in silty clay soil. The saturated hydraulic conductivity of the polluted soil samples has decreased (Nouri et al. 2013). The residual water content, Θr, and the SWRC gradient, n, has diminished by increasing amount of contaminants. Increase in amount of pollutants caused enhancement in the air pressure and decreased amount of α parameter especially in the application of gas oil compared to kerosene. Evaluation of the model validity used to predict water retention curve equation using van Genuchten model revealed better results than Books-Corey model. Brooks-Corey equation represented close and similar results for both types of contaminants and did not show any significant difference for the usage of half and one pore volume of both pollutants for all soil samples (the unpolluted and polluted soil with the same texture). So, more accurate predictions were obtained from Van
Environ Monit Assess (2015) 187: 437
Genuchten model compared to Brooks-Corey one, because different pollutants and values of pollutants have been considered by Van Genuchten equation parameters while there was no such an accuracy in the estimation of Brooks-Corey equation. The results of this research can be used to evaluate the emission impact of pollution at any point during the contamination (particularly when abrupt contamination occurs) and the reaction of the soil. Due to the importance of available water range and few experiments which have been conducted in this area for different types of contaminants and soils, it is suggested that further studies should be carried out in the same domain in terms of various conditions. Finally, the results from this experiment can be generalizable to a global scale for approximately the same condition to assist in solving crisis and problem made by these contaminants. Acknowledgments The authors thank the Guilan University research fund for providing the financial support of this work.
References Ahmadi, A., Abbey, S., & Anthony, N. (1996). Chronic effects of oil spill on soil properties and microflora of a rainforest ecosystem in Nigeria. Water, Air, and Soil Pollution, 86(1– 4), 1–11. Benka-Coker, M. O., & Ekunday, J. A. (1995). Effects of an oil spill on soil physico-chemical properties of a spill site in the Niger Delta Area of Nigeria. Environmental Monitoring and Assessment, 36(2), 93–104. Brasher, B. R., Franzmeier, D. P., Valassis, V., & Davidson, S. E. (1966). Use of Saran resin to coat natural soil clods for Bulkdensity water retention measurements. Soil Science. 101– 108. Brooks, R. H., & Corey, A. T. (1964). Hydraulic properties of porous media. Hydrology Papers. Colorado State University (3), 27. Burckhard, S. R., Pirkl, D., Schaefer, V. R., Kulakow, P., & Leven, B. (2004). Civil and Environmental Engineering Department. Box 2219, South Dakota State University, Brookings, SD 57007. Ebrahimi, S. (2004). Cyclic swelling behavior of superabsorbent polymers in soil porous media. 2004. Master Thesis, Faculty of Agricultural Engineering. Tarbiat Modares University, 107 pages. Ebrahimi, S. (2009). Spatial-temporal variability of hydrocarbon pollutants and chemical solvents behavior in soil porous media. Phd Thesis. Tarbiyat Modares University. 150p. Ebrahimi, S., Shayegan, J., Malakouti, M. J., Akbari, A., & Atashjameh, A. (2010). Hydrocarbon pollution emission in soil around sarkoun refinery. Journal of Water and Soil Conservation, 4, 101–124.
Environ Monit Assess (2015) 187: 437 Everett, K. (1978). Some effects of oil on the physical and chemical characteristics of wet tundra soils. ARCTIC, 31(3), 260–276. Fallah, M., Ebrahimi, S., & Shabanpour, M. (2012a). Comparison of gasoline and kerosene transport in the subsurface, a laboratory soil column study. International Soil Science Congress. BLand Degradation and Challenges in Sustainable Soil Management^. 4p. Fallah, M., Ebrahimi, S., & Shabanpour, M. (2012b). Effect of kerosene on two type soil respiration. International Soil Science Congress. BLand Degradation and Challenges in Sustainable Soil Management^. 5p. Fallah, M., Ebrahimi, S., & Shabanpour, M. (2013). Hydrocarbon pollution emission in the pilot and pulse condition in saturated porous media of soil. Journal of Water and Soil Conservation, 20(3), 227–240. Gupta, S. C., & Larson, W. E. (1979). Estimating soil water retention characteristics from particle size distribution, organic matter content, and bulk density. Water Resources Research, 15, 1633–1635. Haj Abasi, M. (2007). Soil physics. Iran: University Press. 523p. Kayode, J., Oyedeji, A. A., & Olowoyo, O. (2009). Evaluation of the effects of pollution with spent lubricating oil on the physical and chemical properties of soil. Pacific Journal of Science and Technology, 10(1), 387–391. Labud, V., Garcia, C., & Hernandez, T. (2007). Effect of hydrocarbon pollution on the microbial properties of a sandy and a clay soil. Chemosphere, 66(10), 1863–1871. Marins Pala, D., Carvalho, D., & Lippel, G. (2006). A suitable model to describe bioremediation of petroleum contaminated soil. International Biodeterioration, 58, 254–260. Mohan, S., Ramakrishna, M., Shailja, S., & Sarma, P. (2007). Influence of soil water ratio on the performance of slurry phase bioreactor treating herbicide contaminated soil. Bioresource Technology, 98, 2584–2589. Norris, R. D., Hinchee, R. E., Brown, R. A., McCarty, P. L., Semprini, L., Thomas, J. M., & War, C. H. (1994). Handbook of bioremediation. Boca Raton: CRC Press. Nouri, M. (2012). Parametric estimation of soil hydraulic functions in the presence of hydrocarbon pollutants. Master’s thesis, Faculty of Agriculture, Tarbiat Modarres University. 131p. Nouri, M., Homaee, M., & Bybordi, M. (2013). Parametric study of soil hydraulic properties in the presence of kerosene pollutant. Journal of Soil and Water Conservation. 12 Pp. Okonokhua, B. O., Ikhajiagbe, B., Anoliefo, G. O., & Emede, T. O. (2007). The effects of spent engine oil on soil properties and growth of maize (Zea mays L.). Journal of Applied Sciences and Environmental Management, 11(3), 147–152. Page, M. C., Sparks, D., Noll, M. R., & Hendricks, G. J. (1987). Kinetics and mechanisms of potassium release from sandy
Page 13 of 13 437 middle Atlantic Coastal plain soils. Soil Science Society of America Journal, 51, 1460–1465. Seklemova, E., Pavlova, A., & Kovacheva, K. (2001). Biostimulation based bioremediation of diesel fuel: field demonstration. Biodegradation, 12, 311–316. Sharma, R. S., & Mohamed, M. H. A. (2003a). Patterns and mechanisms of migration of light non-aqueous phase liquid in an unsaturated sand. Geotechnique, (in press). Sharma, R. S., & Mohamed, M. H. A. (2003b). An experimental investigation of LNAPL migration in an unsaturated/ saturated sand. Engineering Geology, 70, 305–313. Stotzky, G. (1965). Microbial respiration. In C. A. Black (Ed.), Methods of soil analysis, part 2 (pp. 1550–1572). Madison: American Society of Agronomy Inc. Tuller, M., & Dani, O. (2003). Retention of water in soil and the soil water characteristic curve soil. Soil Water Characteristic Curve. NO 1–20. U.S. Environmental Protection Agency (EPA). (2000). Profile of the oil and gas extraction industry. Washington: EPA (EPA Docket, A-93-02, V-B-7). Van Genuchten, M. T., & Nielsen, D. R. (1991). On describing and predicting the properties of unsaturated soils. Annales Geophysicae, 3(5), 615–628. Van Genukhten, M. T. (1980). A closed-form equation for predicting the hydraulic conductivity of unsaturated soils. Soil Science Society of America Journal, 44(5), 892–898. Van Genuchten, M. Th., Leij, F. J., & Yates, S. R. (1991). The RETC code for quantifying the hydraulic functions of unsaturated soils. Agricultural research service, EPA/600. December. pp 2–91. Volk, V. V. (1980). Oily waste and plant growth. Resource systems and management association (pp. 123–138). New Jersey: Northfield. Vonloocke, R., Deborger, R., Voets, J. P., & Verstraete, W. (1975). Soil and groundwater contamination y oil spills; problems and remedies. International Journal of Environmental Studies, 8, 99–111. Walkey, A., & Black, I. A. (1934). An examination of Degtjareff method for determination of soil organic matter and a proposed modification of the chromic acid in soil analysis. I. Experimental Journal Soil Science, 79, 459–465. Wang, Y., & Shao, M. (2009). Infiltration characteristics of nonaqueous phase liquids in undisturbed loessal soil cores. Journal of Environmental Sciences, 21, 1424–1431. Wang, H. T., Luo, J., Li, Y. S., Niu, D. J., Li, L. Q., & Li, G. F. (2000). Numerical simulation of petroleum pollutant transport in soil. Acta Scientiae Circumstantiae, 20, 755–760. Wang, Y., Feng, J., Lin, Q., Lyu, X., Wang, X., & Wang, G. (2013). Effects of crude oil contamination on soil physical and chemical properties in Momoge wetland of China. Chinese Geographical Science, 23(6), 708–715.
Risk assessment of gas oil and kerosene contamination on some properties of silty clay soil M. Fallah & M. Shabanpor & M. Zakerinia & S. Ebrahimi
Received: 4 January 2015 / Accepted: 21 May 2015 / Published online: 18 June 2015 # Springer International Publishing Switzerland 2015
Abstract Soil and ground water resource pollution by petroleum compounds and chemical solvents has multiple negative environmental impacts. The aim of this research was to investigate the impacts of kerosene and gas oil pollutants on some physical and chemical properties, breakthrough curve (BTC), and water retention curve (SWRC) of silty clay soil during a 3-month period. Therefore, some water-saturated soils were artificially contaminated in the pulse condition inside some glassy cylinders by applying half and one pore volume of these pollutants, and then parametric investigations of the SWRC were performed using RETC software for Van Genukhten and Brooks-Corey equations in the various suctions and the soil properties were determined before and after pollution during 3 months. The results showed that gas oil and kerosene had a slight effect on soil pH and caused the cumulative enhancement in the soil respiration, increase in the bulk density and organic M. Fallah (*) Faculty of Agriculture, Guilan University, Rasht, Iran e-mail: [email protected] M. Shabanpor Department of Soil Science, Guilan University, Rasht, Iran e-mail: [email protected] M. Zakerinia Department of Water Engineering, Gorgan University, Gorgan, Iran e-mail: [email protected] S. Ebrahimi Department of Soil Science, Gorgan University, Gorgan, Iran e-mail: [email protected]
matter, and reduction in the soil porosity and electrical and saturated hydraulic conductivity. Furthermore, gas oil retention was significantly more than kerosene (almost 40 %) in the soil. The survey of SWRC indicated that the contaminated soil samples had a little higher amount of moisture retention (just under 15 % in most cases) compared to the unpolluted ones during this 3-month period. The parametric analysis of SWRC demonstrated an increase in the saturated water content, Θs, from nearly 49 % in the control sample to just under 53 % in the polluted ones. Contaminants not only decreased the residual water content, Θr, but also reduced the SWRC gradient, n, and amount of α parameter. The evaluation of both equations revealed more accurate prediction of SWRC’s parameters by Van Genukhten compared to those of Brooks and Corey. Keyword Break through curve . Gas oil . Kerosene . Soil properties . Water retention curve
Introduction Soil contamination is caused by the presence of chemicals or other alteration in the natural soil environment. In fact, soil and water pollution refers to an increased level of harmful human-made chemicals in the natural soil and groundwater environment (Norris et al. 1994). It is typically caused by industrial activity, agricultural chemicals, or improper disposal of waste
437 Page 2 of 13
material. Petroleum products and by-products have been widely used chemicals in recent years. Leakage from underground storage tanks and accidental spills of hydrocarbons, such as light nonaqueous phase liquids (LNAPLs), pose a serious risk of subsurface contamination (Sharma and Mohamed 2003a). As petroleum hydrocarbons release in the environment, these pollutants can be absorbed by the soil particles and move into the soil body and affect some physical, chemical, and biological properties of soil (Marins Pala et al. 2006; Mohan et al. 2007). A LNAPL migrates downwards under the influence of gravity and viscous and capillary forces, contaminating the subsurface region around the spill as well as deteriorating the quality of groundwater (Sharma and Mohamed 2003b). The study of the breakthrough curve (BTC) is one of the important ways to investigate the emission of pollutants through the soil porous media. In fact, it is a plot of relative concentration versus time, where relative concentration is defined as C/Co with C as the concentration at a point in the groundwater flow domain, and Co as the source concentration (Haj Abasi 2007). Water retention curve (SWRC) is an applicable method to study on the behavior of soil-pollutant system. SWRC shows the relationship between the water content, θ, and the soil water potential, ψ. This curve is characteristic for different types of soil and is also called the soil moisture characteristic. It is used to predict the soil water storage, water supply to the plants (field capacity), and soil aggregate stability. Because of the hysteretic effect of water filling and draining the pores, different wetting and drying curves may be distinguished (Gupta and Larson 1979). In recent years, several researchers have been worked on the hydrocarbon contaminant emission in the porous media of soil, the BTC of the petroleum hydrocarbons, and their fate in the soil composition, but more consideration and research should be taken into account on the moisture retention capacity in the soil-pollutant composition (Ebrahimi et al. 2010). In a study which has been carried out by Burckhard et al. (2004), a comparison of the contaminated SWRC with the uncontaminated SWRC showed that petroleum pollution not only prevented the movement of soil water but also caused some changes on the hydraulic properties of soil. Moreover, the contaminated soil samples
Environ Monit Assess (2015) 187: 437
had a higher water-holding capacity compared to the uncontaminated ones at higher suctions (Burckhard et al. 2004). In recent years, interest in the unsaturated zone has significantly increased because of growing evidence and public concern that the quality of the subsurface environment is being adversely affected by the petroleum contaminants. These days, applications of computer models have been common in many research and management to predict the movement of water and chemicals into and through the unsaturated zone of soils. The RETC computer code is one of the software to analyze the SWRC and hydraulic conductivity functions of unsaturated soils (Van Genuchten et al. 1991). These hydraulic properties are key parameters in any quantitative description of the water flow into and through the unsaturated zone of soils (Van Genuchten and Nielsen 1991). Iran is a major oil-producing country in the world with an annual production capability of large amounts of crude oil and has been exposed to the contamination by these hydrocarbon components through the extraction, transportation, and refining process (Ebrahimi 2009). In this regard, the need for more comprehensive studies to consider and solve this problem is inevitable. Hence, the aim of this research was to determine the impact of kerosene and gas oil as available hydrocarbon pollutants on some physical and chemical properties of a silty clay soil, BTC, and SWRC. Then, the results from SWRC were studied by RETC software quantitatively and parametrically.
Methods and material Soil In this study, the silty clay soil was taken from an area in the University of Gorgan located in Golestan Province, Iran. After the homogenization of the soils, some physical and chemical properties of soils were measured before artificial applying of the pollutants and within the equilibrium time (3 week after polluting and the first 3 months (Ebrahimi 2009) such as the bulk density (BD) using clod method (Brasher et al. 1966), the actual density of soil using pyknometer, pH of the soil saturation extract using the pH meter, the soil
Environ Monit Assess (2015) 187: 437
saturation extract electrical conductivity using the gauges conductivity apparatus (Page et al. 1987), the hydraulic conductivity using RETC code (Van Genuchten et al. 1991), the organic matter using Walkey and Block method (Walkey and Black 1934), the soil respiration using Stotzky methods (Stotzky 1965), and porosity using calculation from the bulk density and actual density of soil). Petroleum hydrocarbon contaminant Kerosene and gas oil were chosen as soil pollutants because of abundant consumption of these contaminants and subsequently their contaminations in the soil porous media especially in Iran as one of the main oilproducing countries. Artificial contamination method of soil and determination of BTC The contaminants were applied separately on the silty clay soils in the pilot scale and pulse condition. Hence, the equipment including glassy cylinders were designed in the height of 120 cm and the diameter of 16 cm with the filter and special outlet at the bottom of the cylinders and the bulk density and porosity of soils were determined (Fallah et al. 2013). The glassy cylinders were filled with the soil homogeneously in the given depth based on the silty clay soil bulk density, and then, soil samples were saturated by water from the top to the bottom of soil column. In this stage of procedure, the amounts of half and one pore volume porosity of silty clay soil, kerosene, and gas oil were applied separately on the saturated soil and let the pollutants pass through the soil. Eventually, output volume of gas oil, kerosene, and water was measured per time and breakthrough curves were drawn (Fallah et al. 2012a, 2013). Then, the polluted samples of soil were placed outdoors and these samples were allowed to reach equilibrium with the natural environment (Ebrahimi 2009). Petroleum hydrocarbon analysis The total petroleum hydrocarbon (TPH) concentration was measured by dichloromethane according to US
Page 3 of 13 437
Environmental Protection Agency within a certain time (EPA 2000).
SWRC analysis After polluting the soil, soil samples were collected from the contaminated soils within the equilibrium time (3 weeks after contamination), first, second, and third months, and unpolluted soil which was considered as a control sample. Then, the pressure plate apparatus (Tuller and Dani 2003) was used to determine the soil water retention curves in 0-, 0.3-, 0.5-, 1-, 5-, and 15-bar suction within the pre-contamination time (control sample); the equilibrium time, first, second, and third months after the soil contamination; and the SWRCs were drawn by Microsoft Excel. Then, data obtained from SWRC was analyzed by RETC software. The RETC computer code is the computer software to analyze the SWRC and hydraulic conductivity functions of unsaturated soils (Van Genuchten et al. 1991). These hydraulic properties are key parameters in any quantitative description of water flow into and through the unsaturated zone of soils. The program uses the parametric models of Brooks-Corey and Van Genuchten to represent the SWRC and the theoretical pore-size distribution models of Mualem and Burdine to predict the unsaturated hydraulic conductivity function from the observed soil water retention data.
Table 1 Some soil properties Sample Texture
Silty clay
Clay (%)
50
Silt (%)
43.3
Sand (%)
6.7
Bulk density (g/cm3) 3
1.26
Actual density (g/cm )
2.35
Porosity (%)
47
Organic carbon (%)
1.22
Organic matter (%)
2.10
pH
7.73
EC (ds/m)
1.12
Respiration (mg CO2/day)
2.1
437 Page 4 of 13
Environ Monit Assess (2015) 187: 437
Fig. 1 Output volume ratio of kerosene and gas oil to water for half pore volume of silty clay soil, pollutant
Brooks and Corey (1964) used this three-parametric equation to describe the SWRC: θw ¼ θr þ ðθs −θr Þ
a b Ψ
ð1Þ
where a is bubbling pressure or potential at air entire value, Θr is residual moisture, Θs is saturation moisture, and Θw is volumetric moisture of soil. This equation can be used in some situations that Ψ is larger than or equal to a. If the potential is less than a, Θw is, then, equal to Θs (Ebrahimi Fig. 2 Output volume ratio of kerosene and gas oil to water for one pore volume of silty clay soil, pollutant
2004, 2009).Van Genukhten (1980) introduced the following four-parametric equation: θs −θr c 1 þ aΨ b This equation can be written as
θw ¼ θr þ
θw ¼ θr þ h
θs −θr b ic 1 þ Ψa
ð2Þ
ð3Þ
where b and c are dimensionless parameters. The slop factor coefficient b (=n) changes the curve shape around
Environ Monit Assess (2015) 187: 437
Page 5 of 13 437
Table 2 Mean comparison of bulk density (g cm−3) and porosity (%) for both pollutants in different time Time
Silty clay soil Bulk density
Porosity
1/262d
0/47a
Control Equilibrium
a
1/298
0/454d
First month
1/287b
0/459c
c
0/464b
c
1/270
0/466b
0/009
0/003
Second month
1/274
Third month e
LSD (5 %) a‐d e
Show the difference between values of parameters
Least significant difference
the turning point. The a (=a) and c (=m) change the curve shape around the bubbling pressure point. It is assumed that a (=a) is the function of matric potential in r a point with the water content of θs þθ 2 . Suction at this point can be shown by Ψ50. If, in the above equation, θw ¼
ðθs þθr Þ 2
a¼
and Ψ=Ψ50, then Ψ 50
2
ð1=cÞ
−1
ð4Þ
ð1=bÞ
In the last, Van Genuchten and Brooks-Corey equation parameters for each sample were calculated by RETC code (Ebrahimi 2004, 2009). In this study, assuming m=1−1/n, the α, θr, θs, and n parameters were estimated for these equations by the RETC program (Van Genuchten et al. 1991).
Statistical design Statistical analysis was performed with three replications based on a split plot design in time for the silty clay soil with two contaminants (gas oil and kerosene) and two levels of pollution (half and one pore volume of soil). SPSS software was used to study the correlation, analysis of variance, and comparison of treatment mean.
Results and discussions Some initial properties of soil were determined before the pollution. Table 1 represents the properties of silty clay soil. According to Table 1, the soil studied in this research was silty clay one (with 50 % clay) and it was considered as a heavy-textured soil which was in the neutral acidity range. The porosity of silty clay soil was also 47 % and had a nearly high content of organic matter compared to the light-textured soil. The electrical conductivity of silty clay soil showed that this soil had a slight salinity and was not in the range of salty soils. Pollutants emission through the soil column The same volume of kerosene and gas oil was used for polluting the soil because of the same porosity of the soil columns. The results indicated that after a short delay, the amount of gas oil and kerosene was observed in the drainage water output from the silty clay soil while the
Table 3 Mean comparison of different levels of contaminants on the pH values in different time Pollution
Time Control
Equilibrium
First month
Second month
Third month
0.5 VPGe
7/74a
7/62c
7/75a
7/33b
7/71a
f
a
a
a
a
0.5 VPK
7/74
7/92
7/70
7/69
7/37b
1 VPGg
7/74a
7/83b
7/40b
7/73a
7/45b
a
b
b
a
h
1VPK
7/74
7/83
7/40
7/73
7/45b
LSD (5 %)
–
0/024
0/098
0/156
0/16
a‐d
Show the difference between values of parameters
e
Half pore volume of gas oil
f
Half pore volume of kerosene
g
One pore volume of gas oil
h
One pore volume of kerosene
437 Page 6 of 13 Table 4 Mean comparison of various pollution levels on the electrical conductivity in different time (ds/m)
Environ Monit Assess (2015) 187: 437
Pollution
0.5VPGa b
0.5VPK
a‐d
Show the difference between values of parameters
Time Control
Equilibrium
First month
Second month
Third month
1/12
0/598a
0/464b
0/596a
0/704a
1/12
ab
a
a
0/677a
b
a‐d
Show the difference between values of parameters
b
0/618
1/12
0/456
0/417
0/453
0/526b
1 VPKd
1/12
0/591a
0/594a
0/643a
0/680a
LSD (5 %)
–
0/105
0/62
0/52
0/78
Pollution
b
0/594
1 VPG
pollution emission enhancement rate was followed by reduction in the rate of water drainage output. Figure 1 showed that in the application of half volume of pollutants, the gas oil retention was more than the kerosene and kerosene flowed out of the silty clay soil by sharper gradient and less retardation and higher velocity (Fallah et al. 2012a, 2013). Figure 2 showed the same procedure as Fig. 1; on the other hand, at the same time, kerosene to water output ratio was substantially more than gas oil to water ratio (Fallah et al. 2012a, 2013). The only difference between the two graphs was relevant to the application of different amounts of pollutants (half and one pore volume) which brought about more pollutant emission from the soil in the usage of one pore volume porosity of silty clay soil, pollutants (Fig. 2) at the same time (approximately double). Emission rate and permeability of petroleum contaminants through the porous media of soil are dependent on the contaminant type, particle size, bulk density, porosity, and compaction of soil, and the result from the emission rate and permeability for these pollutants was in accordance with the results reported by other scientists on the emission rate of petroleum hydrocarbons through the porous medium of soil (Wang et al. 2000; Wang and Shao 2009). Table 5 Mean comparison of various pollution levels on the soil organic matter (percentage) in different time
0/50
c
Survey of changes on some physical and chemical properties and microbial respiration of soil After applying half and one pore volume of gas oil and kerosene pollutants, microbial respiration and some physical and chemical properties of soil samples were examined within the mentioned period of time and the results were analyzed by split-plot design in time with SPSS software. Then, analysis of variance for silty clay soil showed where the main effect (time and pollution) was significant and when the interaction effect was not substantial (time in pollution), mean comparison of the main effects was performed (Table 2). Where the interaction effects were significant, it was not necessary to do the mean comparison of main effects individually, but it was better to compare the different levels of contamination at any time, and in the following, the SWRCs were drawn by Microsoft Excel and analyzed by RETC program. The effect of gas oil and kerosene on the soil bulk density and porosity Considering that the only main effect of time was significant for the bulk density and porosity and there was no effect of pollutant type, the main comparison of mentioned parameter values (mean
Time Control
Equilibrium
First month
Second month
Third month
0.5VPG
2/10a
6/26b
6/05b
5/77b
5/09b
0.5VPK
a
2/10
d
4/37
c
4/25
d
3/98
3/69d
1 VPG
2/10a
8/47a
7/61a
6/79a
6/85a
1 VPK
a
2/10
c
4/98
c
4/75
c
4/58
4/17c
LSD (5 %)
–
0/213
0/543
0/172
0/286
Environ Monit Assess (2015) 187: 437 Table 6 Mean comparison of pollution levels on the microbial respiration values (mg CO2/day)
a‐d
Show the difference between values of parameters
Pollution
Page 7 of 13 437
Time Control
Equilibrium
First month
Second month
Third month
0.5VPG
0/154a
0/119c
0/165c
0/152c
0/149b
0.5VPK
a
b
b
0/190
b
0/167a
c
c
c
0/154
a
0/134
1 VPG
0/154
0/121
0/17
0/157
0/152b
1 VPK
0/154a
0/149a
0/213a
0/19a
0/181a
LSD (5 %)
–
0/013
0/0078
0/0091
0/0144
tended to remain in the neutral range. The result from laboratory tests revealed that there was a shift toward neutral in pH of different types of soils after being contaminated by the petroleum hydrocarbon contaminants (Everett 1978). Another research showed that the pH status of soil in the contaminated (heavy impact and moderate impact) zones varied from acidic to near neutral (Ahmadi et al. 1996).
values of these parameters for half and one pore volume of kerosene and gas oil in each time individually) was performed in different time (Table 2). According to Table 2, the most and least amounts of alteration in the bulk density and porosity were related to the equilibrium time and 3 months after the equilibrium, respectively; furthermore, a significant difference was not observed between the second and third months for these factors in the silty clay soil. A research which was conducted by Kayode et al. (2009) showed that high concentration of hydrocarbon pollutants affected and altered the physical and chemical properties of soil, increased the soil bulk density a little, and reduced the porosity of soil and root penetration.
The effect of gas oil and kerosene on the electrical conductivity of the soil Survey of mentioned contaminants showed that since the interaction effect of the electrical conductivity was significant, the comparison of pollution levels was done in different time. As it can be observed from Table 4, the contaminants caused the reduction in the electrical conductivity and over time, the electrical conductivity has slightly increased and the most increase was related to the third month after pollution; moreover, no significant difference was found between the use of different levels of contaminants at the same time within this period of time in the silty clay soil. The results from Table 4 were in accordance with the experiments conducted by
The effect of gas oil and kerosene on the soil pH The survey of contaminants impacts indicated that the interaction effect was significant for the soil pH, and the comparison of contamination levels was performed in different time (Table 3). Table 3 showed that during different period of time, application of gas oil and kerosene had no significant impact on the pH of silty clay soil. In fact, these contaminants actually made a slight change on the silty clay soil pH and the soil pH Table 7 Mean comparison of the pollution levels on the contaminant content in different time (mg/ 1000 g soil)
Pollution
0.5VPG
a‐d
Show the difference between values of parameters
0/180
Time Control
Equilibrium
First month
Second month
Third month
0
135b
91/66b
70b
63/3cb
d
c
0.5VPK
0
88
1 VPG
0
178a
1 VPK
0
c
115
LSD (5 %)
–
13/62
b
75
158a b
65
58/3c
128a
96/6a
93/33
b
75
68/3b
9/84
12/57
7/62
437 Page 8 of 13
Environ Monit Assess (2015) 187: 437
Fig. 3 SWRC for one pore volume of kerosene
Benka-Coker and Ekunday (1995) in Nigeria which demonstrated the electrical conductivity reduction in the petroleum-polluted soil. The effect of gas oil and kerosene on the organic matter of soil Table 5 represents that the organic matter has enhanced after the soil contamination in comparison with the control sample. The most increase was relevant to the application of one pore volume of gas oil compared to other values of the contaminants in the initial time after contamination. This increase was due to more gas oil retention and longer carbon chain of gas oil, and this increase showed a downward trend over time. The most increase in the Fig. 4 SWRC for half-pore volume of kerosene
soil organic matter was related to one pore volume of gas oil, half pore volume of gas oil, one pore volume of kerosene, and half pore volume of kerosene, respectively, in the early time after the contamination. The results of various researches on the effects of petroleum distribution showed an increase in the organic carbon and organic matter in the petroleum-polluted soils (Okonokhua et al. 2007; Wang et al. 2013).
The effect of gas oil and kerosene on the microbial respiration of soil Table 6 revealed that the soil microbial respiration showed reduction in the early time after the soil contamination.
Environ Monit Assess (2015) 187: 437
Page 9 of 13 437
Fig. 5 SWRC for one pore volume gas oil
After 1 month, soil microorganisms were accustomed to the new condition and the microbial respiration has increased. This increment in the microbial activity could be because of the microorganism population increase and that was probably due to the presence of oil-eating bacteria in the soil which referred to the usage of the hydrocarbon as a carbon resource (Seklemova et al. 2001; Fallah et al. 2012b). More volume of kerosene resulted in more soil respiration. As observed in Table 6, there was a declining trend of the soil microbial respiration from the first month to the third month. However, the cumulative increase in the microbial respiration in the contaminated soil indicated the microbial Fig. 6 SWRC for half-pore volume of gas oil
respiration increase in these soil samples (Fallah et al. 2012b; Labud et al. 2007). Variation of gas oil and kerosene values in the silty clay soil Table 7 represents the contaminant variation over time. As you see Table 7, the contaminant content in the gas oil-polluted samples was more than the kerosenepolluted ones. Over time, the amount of pollution has reduced and in the third month, it has reached the lowest level during this period. As observed, in the first 2 months, changes on the pollution level had a significant difference from each other. However;
437 Page 10 of 13
Environ Monit Assess (2015) 187: 437
Fig. 7 Mean comparison of different periods for the keroseneand gas oil-polluted soil
this difference between the levels of contaminants has been reduced in the next 2 months. The effect of petroleum pollutants on the SWRC Kerosene effect on the SWRC In the silty clay soil, the survey of the SWRC for half and one pore volume of kerosene showed that by increasing the suction, water started to flow out of the soil with slow gradient in the control soil. As it can be easily seen from Figs. 3 and 4, the most reduction in the soil water retention was relevant to the suction between 0 and 0.5 and at the higher suction, the soil water loss was less than lower suction in both contaminated soil and control sample. In addition, in the contaminated samples, the intensity of moisture loss was less than control sample, and the polluted soils had slightly higher volumetric moisture content in the different suctions. In fact, in the heavy texture soil, the moisture content of soil became more in the unpolluted soil in all applied pressure or in the natural condition without any suction. For one pore volume of kerosene (Fig. 3), the soil water Table 8 Van Genuchten equation parameters for the silty clay soil
Sample
Θr (L3 L−3)
retention was a bit more than half volume in different suction within this period of time. In Figs. 3 and 4, graphs showed decreasing gradient over time and this moisture reduction continued until the third month but never reached the initial value of water retention compared to the control sample during this period.
Gas oil effect on the SWRC Figures 5 and 6 illustrate the effect of gas oil emission on the SWRC for one pore volume of gas oil (Fig. 5) and half pore volume of gas oil (Fig. 6). As it can be observed from these figures, the SWRC followed exactly the same trend as Figs. 3 and 4. By increasing the suction in the control sample, the silty clay soil has lost its moisture by slow gradient in the various suctions due to approximately 50 % clay whereas the contaminated samples demonstrated less moisture loss than the uncontaminated soil and a little greater volume of the water retention in different suctions within various time.
Θs (L3 L−3)
α (L−)
Number
m
Ks (cm/day)
Control
0.157
0.49
0.056
1.24
0.19
22.26
1 VPG
0.102
0.526
0.0275
1.16
0.13
10.99
0.5VPG
0.108
0.512
0.0307
1.17
0.145
13.77
1 VPK
0.144
0.51
0.053
1.19
0.165
16.7
0.5VPK
0.117
0.507
0.036
1.18
0.15
16
Environ Monit Assess (2015) 187: 437 Table 9 Brooks-Corey equation parameters for the silty clay soil
Page 11 of 13 437
Sample
Θr (L3 L3)
Θs (L3 L3)
α (L−)
Number
m
Ks (cm/day)
Silty clay
0.056
0.423
0.0292
0.127
1
2.16
In addition, this moisture retention represented more increase particularly in the lower suction as an illustration, in the early time after contamination and higher level of pollution (one pore volume of gas oil compared to half volume of gas oil) in the lower suction. In other words, over time, the soil moisture content gradually declined; nevertheless, it has never attained the initial values of the water retention and held a bit more moisture compared to control sample similar to Figs. 3 and 4. Mean comparison of the water retention curve of different time for the contaminated silty clay soil compared to the control one The results of the mean comparison of the SWRC for half and one pore volume porosity of soil, kerosene, and gas oil (Fig. 7) demonstrated that the addition of pollutants into soil samples caused increase in the moisture retention capability in different suctions. According to Fig. 7, the maximum increase was related to one pore volume of gas oil and then, half pore volume of gas oil, one pore volume of kerosene, and half pore volume of kerosene, respectively, at all suctions during different periods of time. Parametric analysis of the soil-pollutant water retention curve After drawing the SWRC for the contaminated samples and control sample (uncontaminated soil) by Excel Microsoft, Brooks-Corey and Van Genuchten parameters were determined by RETC computer program. It was also given the nonlinear least-squares parameter optimization method to estimate the unknown coefficients in the hydraulic models. The program also predicted the hydraulic conductivity from the observed soil water retention data assuming that one observed conductivity value (not necessarily at saturation) was available. Van Genuchten model of Mualem and BrooksCorey equation parameters were determined using volumetric moisture content by RETC computer program (Van Genuchten et al. 1991). Results from RETC software for the parameters of Van Genuchten and BrooksCorey equation are presented in Tables 8 and 9 for the
silty clay soil contaminated by kerosene and gas oil hydrocarbon pollutants. The parametric study of soil-pollutant water retention curves indicated that the saturated hydraulic conductivity of the polluted soil samples has decreased due to higher viscosity of contaminants compared to water and lower capacity of soil retention for these fluids (Nouri et al. 2013). The saturation moisture (Θs) has increased by pollutant addition in the silty clay soil which was in accordance with the expectation. However, there is no significant difference between the amount of saturation moisture in the half and one pore volume of the applied kerosene on the silty clay soils by RETC. As it is observed from Table 8, the change on the residual moisture (Θr) showed decline by increasing the amount of contamination in the soil samples and the most and the least reduction in Θr was pertaining to the application of one pore volume of gas oil and one pore volume of kerosene, respectively. The enhancement in amount of pollutant treatment caused decrease in the soil-pollutant water retention curve slope, n parameter. The changes in α parameter are strictly dependant on the soil type. Increase in amount of pollutants caused enhancement in air pressure and decreased amounts of α parameter in the silty clay soil especially in the application of gas oil compared to kerosene pollutant. As it can be observable from Table 9, Brooks-Corey equation did not make any difference for the same textures with different conditions in terms of contamination with these pollutants and one type of data was only given for all soil conditions contaminated with gas oil and kerosene. Accordingly, Brooks-Corey equation cannot give so accurate and detailed data and correlation to evaluate various conditions.
Conclusion The result from this research represented that all apparent soil physical and chemical properties changed by emission of kerosene and gas oil on the soil surface. The soil analysis indicated a slight impact of gas oil and kerosene on the soil pH, and it had a tendency to shift toward neutral (Vonloocke et al. 1975; Ahmadi et al.
437 Page 12 of 13
1996) and caused the cumulative enhancement in the soil respiration (Fallah et al. 2012b; Labud et al. 2007), increase in the BD (Kayode et al. 2009), organic matter (Okonokhua et al. 2007; Wang et al. 2013), and the reduction in the soil porosity (Kayode et al. 2009) and the electrical (Benka-Coker and Ekunday 1995) and saturated hydraulic conductivity; moreover, the BTC results revealed that the gas oil retention was significantly more than that of kerosene and kerosene flowed out of the soil by sharper gradient and less retardation especially in the application of higher amount of kerosene (Wang et al. 2000; Wang and Shao 2009). The reason is that the reactivity of gas oil with the silty clay soil is substantially more than that of kerosene which is probably because of the structural form and its carbonic chain number, and on the other hand, it caused more intense surface absorption (Fallah et al. 2012a, 2013). The survey of SWRC showed higher amount of moisture retention in the contaminated soil samples compared to the unpolluted sample (Nouri 2012; Burckhard et al. 2004), and the maximum increase was related to one pore volume of gas oil and then, half pore volume of gas oil, one pore volume of kerosene, and half pore volume of kerosene, respectively, within this period. In fact, some parts of contaminants were strongly absorbed to the clay mineral as a residual pollutant and increased suction in the capillary pipes of clay and reduced the water loss from the soil, and some authors attributed this to the nonpolar characteristic of oil (Volk 1980). The parametric analysis of soil-pollutant water retention curve by RETC software for Van Genukhten and Brooks-Corey equations demonstrated increase in the saturated water content, Θs, in silty clay soil. The saturated hydraulic conductivity of the polluted soil samples has decreased (Nouri et al. 2013). The residual water content, Θr, and the SWRC gradient, n, has diminished by increasing amount of contaminants. Increase in amount of pollutants caused enhancement in the air pressure and decreased amount of α parameter especially in the application of gas oil compared to kerosene. Evaluation of the model validity used to predict water retention curve equation using van Genuchten model revealed better results than Books-Corey model. Brooks-Corey equation represented close and similar results for both types of contaminants and did not show any significant difference for the usage of half and one pore volume of both pollutants for all soil samples (the unpolluted and polluted soil with the same texture). So, more accurate predictions were obtained from Van
Environ Monit Assess (2015) 187: 437
Genuchten model compared to Brooks-Corey one, because different pollutants and values of pollutants have been considered by Van Genuchten equation parameters while there was no such an accuracy in the estimation of Brooks-Corey equation. The results of this research can be used to evaluate the emission impact of pollution at any point during the contamination (particularly when abrupt contamination occurs) and the reaction of the soil. Due to the importance of available water range and few experiments which have been conducted in this area for different types of contaminants and soils, it is suggested that further studies should be carried out in the same domain in terms of various conditions. Finally, the results from this experiment can be generalizable to a global scale for approximately the same condition to assist in solving crisis and problem made by these contaminants. Acknowledgments The authors thank the Guilan University research fund for providing the financial support of this work.
References Ahmadi, A., Abbey, S., & Anthony, N. (1996). Chronic effects of oil spill on soil properties and microflora of a rainforest ecosystem in Nigeria. Water, Air, and Soil Pollution, 86(1– 4), 1–11. Benka-Coker, M. O., & Ekunday, J. A. (1995). Effects of an oil spill on soil physico-chemical properties of a spill site in the Niger Delta Area of Nigeria. Environmental Monitoring and Assessment, 36(2), 93–104. Brasher, B. R., Franzmeier, D. P., Valassis, V., & Davidson, S. E. (1966). Use of Saran resin to coat natural soil clods for Bulkdensity water retention measurements. Soil Science. 101– 108. Brooks, R. H., & Corey, A. T. (1964). Hydraulic properties of porous media. Hydrology Papers. Colorado State University (3), 27. Burckhard, S. R., Pirkl, D., Schaefer, V. R., Kulakow, P., & Leven, B. (2004). Civil and Environmental Engineering Department. Box 2219, South Dakota State University, Brookings, SD 57007. Ebrahimi, S. (2004). Cyclic swelling behavior of superabsorbent polymers in soil porous media. 2004. Master Thesis, Faculty of Agricultural Engineering. Tarbiat Modares University, 107 pages. Ebrahimi, S. (2009). Spatial-temporal variability of hydrocarbon pollutants and chemical solvents behavior in soil porous media. Phd Thesis. Tarbiyat Modares University. 150p. Ebrahimi, S., Shayegan, J., Malakouti, M. J., Akbari, A., & Atashjameh, A. (2010). Hydrocarbon pollution emission in soil around sarkoun refinery. Journal of Water and Soil Conservation, 4, 101–124.
Environ Monit Assess (2015) 187: 437 Everett, K. (1978). Some effects of oil on the physical and chemical characteristics of wet tundra soils. ARCTIC, 31(3), 260–276. Fallah, M., Ebrahimi, S., & Shabanpour, M. (2012a). Comparison of gasoline and kerosene transport in the subsurface, a laboratory soil column study. International Soil Science Congress. BLand Degradation and Challenges in Sustainable Soil Management^. 4p. Fallah, M., Ebrahimi, S., & Shabanpour, M. (2012b). Effect of kerosene on two type soil respiration. International Soil Science Congress. BLand Degradation and Challenges in Sustainable Soil Management^. 5p. Fallah, M., Ebrahimi, S., & Shabanpour, M. (2013). Hydrocarbon pollution emission in the pilot and pulse condition in saturated porous media of soil. Journal of Water and Soil Conservation, 20(3), 227–240. Gupta, S. C., & Larson, W. E. (1979). Estimating soil water retention characteristics from particle size distribution, organic matter content, and bulk density. Water Resources Research, 15, 1633–1635. Haj Abasi, M. (2007). Soil physics. Iran: University Press. 523p. Kayode, J., Oyedeji, A. A., & Olowoyo, O. (2009). Evaluation of the effects of pollution with spent lubricating oil on the physical and chemical properties of soil. Pacific Journal of Science and Technology, 10(1), 387–391. Labud, V., Garcia, C., & Hernandez, T. (2007). Effect of hydrocarbon pollution on the microbial properties of a sandy and a clay soil. Chemosphere, 66(10), 1863–1871. Marins Pala, D., Carvalho, D., & Lippel, G. (2006). A suitable model to describe bioremediation of petroleum contaminated soil. International Biodeterioration, 58, 254–260. Mohan, S., Ramakrishna, M., Shailja, S., & Sarma, P. (2007). Influence of soil water ratio on the performance of slurry phase bioreactor treating herbicide contaminated soil. Bioresource Technology, 98, 2584–2589. Norris, R. D., Hinchee, R. E., Brown, R. A., McCarty, P. L., Semprini, L., Thomas, J. M., & War, C. H. (1994). Handbook of bioremediation. Boca Raton: CRC Press. Nouri, M. (2012). Parametric estimation of soil hydraulic functions in the presence of hydrocarbon pollutants. Master’s thesis, Faculty of Agriculture, Tarbiat Modarres University. 131p. Nouri, M., Homaee, M., & Bybordi, M. (2013). Parametric study of soil hydraulic properties in the presence of kerosene pollutant. Journal of Soil and Water Conservation. 12 Pp. Okonokhua, B. O., Ikhajiagbe, B., Anoliefo, G. O., & Emede, T. O. (2007). The effects of spent engine oil on soil properties and growth of maize (Zea mays L.). Journal of Applied Sciences and Environmental Management, 11(3), 147–152. Page, M. C., Sparks, D., Noll, M. R., & Hendricks, G. J. (1987). Kinetics and mechanisms of potassium release from sandy
Page 13 of 13 437 middle Atlantic Coastal plain soils. Soil Science Society of America Journal, 51, 1460–1465. Seklemova, E., Pavlova, A., & Kovacheva, K. (2001). Biostimulation based bioremediation of diesel fuel: field demonstration. Biodegradation, 12, 311–316. Sharma, R. S., & Mohamed, M. H. A. (2003a). Patterns and mechanisms of migration of light non-aqueous phase liquid in an unsaturated sand. Geotechnique, (in press). Sharma, R. S., & Mohamed, M. H. A. (2003b). An experimental investigation of LNAPL migration in an unsaturated/ saturated sand. Engineering Geology, 70, 305–313. Stotzky, G. (1965). Microbial respiration. In C. A. Black (Ed.), Methods of soil analysis, part 2 (pp. 1550–1572). Madison: American Society of Agronomy Inc. Tuller, M., & Dani, O. (2003). Retention of water in soil and the soil water characteristic curve soil. Soil Water Characteristic Curve. NO 1–20. U.S. Environmental Protection Agency (EPA). (2000). Profile of the oil and gas extraction industry. Washington: EPA (EPA Docket, A-93-02, V-B-7). Van Genuchten, M. T., & Nielsen, D. R. (1991). On describing and predicting the properties of unsaturated soils. Annales Geophysicae, 3(5), 615–628. Van Genukhten, M. T. (1980). A closed-form equation for predicting the hydraulic conductivity of unsaturated soils. Soil Science Society of America Journal, 44(5), 892–898. Van Genuchten, M. Th., Leij, F. J., & Yates, S. R. (1991). The RETC code for quantifying the hydraulic functions of unsaturated soils. Agricultural research service, EPA/600. December. pp 2–91. Volk, V. V. (1980). Oily waste and plant growth. Resource systems and management association (pp. 123–138). New Jersey: Northfield. Vonloocke, R., Deborger, R., Voets, J. P., & Verstraete, W. (1975). Soil and groundwater contamination y oil spills; problems and remedies. International Journal of Environmental Studies, 8, 99–111. Walkey, A., & Black, I. A. (1934). An examination of Degtjareff method for determination of soil organic matter and a proposed modification of the chromic acid in soil analysis. I. Experimental Journal Soil Science, 79, 459–465. Wang, Y., & Shao, M. (2009). Infiltration characteristics of nonaqueous phase liquids in undisturbed loessal soil cores. Journal of Environmental Sciences, 21, 1424–1431. Wang, H. T., Luo, J., Li, Y. S., Niu, D. J., Li, L. Q., & Li, G. F. (2000). Numerical simulation of petroleum pollutant transport in soil. Acta Scientiae Circumstantiae, 20, 755–760. Wang, Y., Feng, J., Lin, Q., Lyu, X., Wang, X., & Wang, G. (2013). Effects of crude oil contamination on soil physical and chemical properties in Momoge wetland of China. Chinese Geographical Science, 23(6), 708–715.
