1 Introduction
Soil water hydrological processes control infiltration and surface runoff, and also the transport of pollutants, such as pesticides from agricultural land, either in overland flow or in percolating water that may enter the ground water [2,9,16]. Concerns about these processes and their consequences have motivated numerous experiments and theoretical studies relating soil properties to soil water movement and chemical transport.
Rock fragments are often present in soil as a result of soil forming processes and human activity. Rock fragments in the soil may influence water movement and solute transport by affecting the soil structure and the tortuosity of water flow paths, which may make water movement and solute transport more complex than in a stone-less soil [16,30]. Moreover, with increasing rock fragment content, the effects may become more significant. Since rock fragments in the soil can affect water movement and solute transport, they can in turn influence land use and site productivity [7,19,21]. Since the 1950s, scientists have studied the effect of rock fragments on several soil physical processes. Corey and Kemper [7] studied the reduction in evaporation rate attributable to rock fragments covering the soil surface and noted that, as with other mulch materials, they had the greatest inhibiting effect upon evaporation during the first few days after a rainfall event [7]. The effects of rock fragment cover in both increasing and decreasing infiltration were discussed in detail by Brakensiek and Rawls [4]. Valentin [28] studied the influence of rock fragments on surface sealing and concluded that relationships between sealing and rock fragments depended upon the pedoclimatic conditions [28]. Most studies, however, focused on the effect of the cover percentage of rock fragments on surface hydrological processes [18,19] whereas fewer have considered the effects of rock fragments in the soils on water movement and solute transport processes.
A particular characteristic of the soils of the Loess Plateau is that they are formed from loess deposits that have a high calcium carbonate content [26]. The source of these carbonates is both detrital, deposited following eolian transportation by monsoon winds, and authigenic [14]. Calcium carbonate concretions, resulting from pedogenesis, are often the main representatives of the authigenic carbonates and soils containing them are widespread [12]. The concretions form as a result of dissolution of calcium minerals in the presence of carbonic acid formed during the wet and humid weather of the summer and autumn from plant residues. The dissolved minerals are leached by percolating rainwater to lower depths where they are concentrated, coalesce, and dry out to form the calcium carbonate concretions. Processes such as uplifting and cultivation have subsequently redistributed these concretions, or rock fragments, within the upper soil profile where they can influence water movement and solute transport.
In this study, we determined water movement and solute transport processes through columns of disturbed soil samples containing rock fragments. The objective was to assess the effect of various rock fragment contents on soil water infiltration rates over time, the saturated hydraulic conductivity (Ks) and the solute transport process. The experimental data were also compared with existing models for Ks and solute transport in order to test their effectiveness in predicting these properties for a soil from the Loess Plateau containing rock fragments.
2 Material and methods
Samples of the A-horizon (10–30 cm depth) of a silty clay loam semi-luvisol (Chinese genetic soil classification system) [15] were collected at Xibo Village, Yangling District, Shaanxi Province, on the Loess Plateau, China. The soil samples were air-dried and the rock fragments (> 2 mm) were then separated from the finer soil (< 2 mm) by sieving. The rock fragments were further sieved to obtain a fraction with an equivalent diameter of 10–30 mm. Selected soil properties are shown in Table 1; the particle size analysis was determined by sieving and the pipette method; organic matter was determined by potassium dichromate titration; cation exchange capacity was determined by barium chloride exchange; exchangeable sodium percentage was determined by ammonium acetate exchange [15]. The rock fragments were mainly composed of CaCO3 (> 90%) and had a saturated water content of 0.072 m3 m−3.
Selected properties of the silty clay loam soil (International soil science society textural class).
Tableau 1 Quelques propriétés du limon argilo-silteux (classes de texture de la Société Internationale des Sciences du Sol).
| Saturated water content (m3 m−3) | Sand content (kg kg−1) | Silt content (kg kg−1) | Clay content (kg kg−1) | Organic matter (kg kg−1) | Cation exchange capacity cmol kg−1 | Exchangeable sodium percentage (%) |
| 0.235 | 0.263 | 0.482 | 0.364 | 0.015 | 14.6 | 0.3 |
Air-dried fine soil samples (< 2 mm) were mixed well with the rock fragments (10–30 mm) to give seven different gravimetric contents (0, 10, 20, 30, 40, 50, and 60%). The soil–rock fragment mixtures were then uniformly packed into acrylic plastic columns (height 50 cm, inner diameter 20 cm), the perforated bases of which were covered by a coarse filter paper (Fig. 1), to obtain a bulk density of 1.4 g cm−3. The initial moisture contents of the fine soil and of the rock fragments were 0.012 m3 m−3 and 0.0048 m3 m−3, respectively. The surfaces of the soil mixtures were covered with a circle of filter paper to reduce disturbance from inflowing solutions.
Experimental setup for water flow and solute transport study.
Fig. 1. Dispositif expérimental d’étude de l’écoulement de l’eau et du transport de soluté.
Flow experiments were carried out by rapidly establishing and then maintaining a constant 4.5 cm head of deionized water on the surface of the soil using a Mariotte bottle (Fig. 1). The cumulative infiltration rate was initially determined from the volume of water supplied from the Mariotte bottle measured at timed intervals. After the wetting front had reached the base of the soil column the effluent was collected every 30 min and the volume measured until the flow rate became constant. The soil columns were then left to saturate for at least 24 h.
The distilled water in the Mariotte bottle was replaced with 0.15 mol L−1 CaCl2 solution, acting as a tracer, and the flow was resumed while maintaining a constant water head of 4.5 cm. The effluent was collected continuously in 150 ml volumetric flasks over timed intervals. The effluent samples were then titrated with silver chloride solution to determine the Cl− concentrations until they attained a stable level, close to 0.15 mol L−1.
The experiments were carried out in three replications in a laboratory where the average temperature was 19 ± 3 °C. Relative humidity was not controlled but remained at 40 ± 10%.
The saturated hydraulic conductivity of the soil–rock fragment mixtures (Ks, L T−1) was calculated from the infiltration data using Darcy's law [20]:
| (1) |
We wanted to compare our measured Ks values with those predicted by models devised for stony soils although literature on the hydraulic properties of such soils is limited. Peck and Watson [18] used a heat flow analogy for a medium in which there were spherical inclusions that were noninteractive to derive the following equation for the ratio of the field, or bulk, soil saturated hydraulic conductivity, Kb, to the fine-soil fraction saturated conductivity, Ks:
| (2) |
Brakensiek and Rawls [8] pointed out that Eq. (2) was over-simplified and inaccurate. Bouwer and Rice (1984) used data from laboratory studies on sand-gravel mixtures to establish an expression that used void ratios [3]:
| (3) |
Therefore, in order to further discuss the effect of rock fragment contents on Ks, we calculated Ks using Eqs. (2) and (3) and then compared those theoretical solutions with our experimentally determined Ks values.
The experimental data of the solute transport were analyzed using two transport models, described below, that were fitted to the data using an iteration technique based on achieving the lowest sum of squares residual value (SSQ).
Model 1: The simplified convection–dispersion equation (CDE) [13], given by:
| (4) |
Model 2: The two-region model (T-R) is given by [11]:
| (5) |
| (6) |
| (7) |
By taking the effluent concentration to be the flux-averaged concentration of the soil column, the initial and boundary conditions for this study can be assumed to be:
| (8) |
| (9) |
| (10) |
Using the derived values of the dispersion coefficient (D or Di) from each of the fitted solute transport models, we also calculated the dispersivity, λ, (L) and the Peclet number, Pe, defined by the following equations:
| (11) |
| (12) |
3 Results and discussion
3.1 Infiltration
Fig. 2 presents the mean cumulative infiltration as a function of time for the rock fragment contents studied. Maximum values of the coefficients of variance for the cumulative infiltration occurring for a particular rock content were 2.4, 9.0. 6.5, 9.0, 8.8, 7.0 and 8.7% for the 0, 10, 20, 30, 40 50, and 60% rock fragment contents, respectively. Smooth cumulative infiltration curves were obtained for all rock fragment contents, which follow a pattern similar to those typical for homogeneous soils without stones and indicate that the fine soil matrix was continuous in all cases. Cumulative infiltration increased at a significantly lesser rate when rock fragments were present than for the stoneless soil (P < 0.05) (Fig. 2). This indicates that the presence of rock fragments in the soil impeded infiltration. Additionally, during the early stage of the infiltration process, the effect of rock fragment content on cumulative infiltration was less noticeable but became so with time as the process of infiltration progressed. The effect of 10 and 20% rock fragment contents on cumulative infiltration was considerably smaller than that of higher rock fragment contents. Statistically, there was no significant difference between cumulative infiltration in the rock mixtures for Rw values of 10 and 20%, or for Rw values of 50 and 60%. We observed the smallest rates of infiltration when Rw was 40% (Fig. 2).
Effect of rock fragment contents on cumulative infiltration over time. Symbols represent mean data points; solid lines are fitted curves using the Philip's infiltration equation; error bars for data points at about 12,000 s represent ± 1 standard deviation.
Fig. 2. Effet de la teneur en fragments de roche sur l’infiltration cumulée en fonction du temps. Les symboles représentent la moyenne des données expérimentales ; les lignes continues représentent le modèle d’infiltration de Philip calé sur ces données ; les barres d’erreur pour les mesures à 12 000 s. représentent ± un écart-type.
In order to analyze the cumulative infiltration data, we fitted the experimental data with Philip's infiltration equation (Eq. [14]), the parameters of which are presented in Table 2:
| (13) |
Parameters for the Philip infiltration equation determined for soils with various rock fragment contents.
Tableau 2 Paramètres pour l’équation d’infiltration de Philip déterminés pour des sols à différentes teneurs en fragments de roche.
| Parameter | Rock fragment content | ||||||
| 0% | 10% | 20% | 30% | 40% | 50% | 60% | |
| S (cm/s1/2) × 10−2 | 4.13 ± 0.02 | 4.76 ± 0.01 | 4.56 ± 0.02 | 3.13 ± 0.01 | 2.68 ± 0.01 | 2.81 ± 0.01 | 3.16 ± 0.04 |
| A (cm/s) × 10−2 | 0.05 ± 0.00 | 0.04 ± 0.00 | 0.04 ± 0.00 | 0.04 ± 0.00 | 0.02 ± 0.00 | 0.03 ± 0.00 | 0.03 ± 0.00 |
| R2 | 0.998 | 0.999 | 0.998 | 0.999 | 0.997 | 0.997 | 0.997 |
Philip's infiltration equation considers both the downward movement of water, due to gravity, and advection due to matric forces. The coefficient of determination, R2, values were all above 0.99 indicating that the equation effectively described the relation between cumulative infiltration and time during the course of our experiment (Table 2). Both sorptivity and steady state infiltration were affected in a similar manner by increasing rock fragment contents (Table 2).
We regarded the rock fragments as impermeable. Thus, a main effect of the presence of the rock fragments is to reduce the area for water transmission and to increase the tortuosity of the water flow paths [6,17,27]. The values of sorptivity for rock fragment contents of 10 and 20% determined by Philip's equation are higher than that for the soil alone and may be an indication of this increased tortuosity. During the initial stage of the infiltration process, water movement into the dry soil is faster due to the presence of the matrix potential that exists at the wetting front between wet and dry soils. Since the presence of rock fragments might be expected to represent a discontinuity in the pore system, matrix potential might also be expected to decrease with increasing rock content. However, it appears from our data that, at a critical rock fragment content that appears to be about 40%, the rock fragments begin to contribute to a more continuous pore system. We attribute this to a greater degree of interconnection between the macropores that form around the fragments.
When the soil is wet, the increase in the volume occupied by the macropores associated with the increase in the rock fragment content may result in complex trends in water flow [22]. Continuous macropore systems can promote water flow whereas isolated macropores do not and, furthermore, are more likely to trap air, which interferes with water movement. Additionally, as the rock fragment content increases, there is a decrease in the effective porosity, i.e., the porosity where actual movement of water is possible and where isolated pores, for example, are excluded. Thus, we observe a decrease in infiltration due to decreasing effective porosity and the presence of isolated macropores as the rock fragment content increases. However, when the rock fragment content is about 40%, the macropore system appears to become more continuous to the extent that it facilitates water movement and, thus, infiltration subsequently increases with further increases in rock fragment content.
3.2 Saturated hydraulic conductivity (Ks)
Fig. 3 shows the effect of rock fragments on Ks and the comparisons of the experimental data with the calculated values determined by the Peck–Watson and Bouwer–Rice equations. The experimental values of Ks initially decreased with increasing rock fragment contents to a minimum value and then increased (Fig. 3). When Rw = 40%, Ks had the smallest value. The trends in Ks values follow those observed for the cumulative infiltration and are explained in a similar manner. When the rock fragment contents are small, the presence of nonporous fragments only reduces the available area for the flow and increases the tortuosity of the water flow. The increased tortuosity also results from the necessity for water to move laterally in order to wet the soil under the rock fragments. Although macropores are more likely to exist at the soil-to-rock interface, these are not continuous when the rock fragments are not in contact with each other, as would be the case in the mixtures with low rock fragment contents. Thus, increasing the rock fragment content initially increases the impedance of water flow and consequently Ks values become smaller. However, at a critical rock fragment content, which we observed to be about 40%, there are sufficient rock fragments in the soil to create more continuous macropores. Furthermore, fragment-to-fragment contact is more likely to create larger voids than those existing at the fragment-to-soil interfaces. Therefore, for rock fragment contents greater than this critical content, even though tortuosity increases, the creation of such larger and more continuous macropores appears to be beneficial for water flow and results in the observed increase in saturated hydraulic conductivity with further increases in rock fragment contents [8,21].
Comparison of Ks determined experimentally and from equations proposed by Peck and Watson [18] and Bouwer and Rice [5]. Error bars represent ± 1 standard deviation.
Fig. 3. Comparaison de la perméabilité Ks déterminé expérimentalement et à partir des équations de Peck et Watson [18] et de Bower et Rice [5] ; les barres d’erreur représentent ± un écart-type.
From Fig. 3 it can also be seen that the Ks values calculated by the Peck–Watson equation were consistently smaller than those predicted by the Bouwer–Rice equation, which was also found to be the case by Sauer and Logsdon [24]. Furthermore, neither equation correctly predicted the trend in the relation between Ks and rock fragment content although there were some coincidental agreements between the experimental data and the calculated values, notably for the two lower rock fragment contents. Zhou and Shao [29] similarly reported that the Peck–Watson and Bouwer–Rice equations were only valid for estimating Ks for soil containing small amounts of elliptical rock fragments and that, with increasing content of these elliptical rock fragments, both equations over-estimated the value of Ks. Bagarello et al. [1] pointed out that determining the void ratio measurements used in the Bouwer–Rice equation are not necessary to predict Kb/Ks since the es/eb ratio equals and can thus be simply determined by the direct measurement of [2]. Thus our results and the findings of these other studies suggest that the use of a simple experimental approach to determine the saturated hydraulic conductivity of a soil containing rock fragments can be advantageous in the absence of accurate models.
3.3 Solute transport
Fig. 4a–g presents the breakthrough curves of the 0.15 mol L−1 CaCl2 solution used as a flow tracer for the soil samples with different rock fragment contents. The data in these figures showed that, even when the rock fragment content was as great as 60% (Fig. 4a–g), smooth breakthrough curves were obtained that followed a similar trend to those typical for a homogeneous soil without stones. Russo, experimenting on a soil containing 50–70% rock fragments by weight, similarly observed that the solute transport process in the stony soil was very similar to that in the homogeneous stoneless soil [23].
(a–g) Comparison of the ratio of concentrations of Cl− in the effluent to that in the leaching solution when determined experimentally and when derived by the classic diffusion equation (CDE) or the two-region model (T-R).
Fig. 4. (a–g) Comparaison du rapport de la concentration en Cl− dans l’effluent à celle de la solution entrante, valeur expérimentale et celles obtenues avec le modèle CDE classique ou avec le modèle T-R à deux régions.
Our experimental data were fitted using both the CDE and the T-R. Fig. 4a–g indicates that the T-R model provided a slightly better fit to the data than the CDE, especially in its ability to describe the steep initial increases in relative concentration and the tailing phenomenon. From Fig. 4a–g, it can be seen that Cl− was detected earliest in the effluent from soil–rock fragment mixtures when Rw was 60%, which was probably due to the greater preponderance of continuous macropores that facilitated rapid solute flow. The presence of Cl− was detected latest in the effluent from soil–rock fragment mixtures when Rw was 40%, which was the critical content noted above as the level at which Ks was the lowest value while the presence of macropores was probably least beneficial because they did not yet form a continuous macropore system.
When Rw was 0%, breakthrough occurred in the shortest time in the soil columns while the longest time for breakthrough was observed when Rw was 40% (Fig. 4a–g). Comparing the various breakthrough curves in Fig. 4a–g, it can be seen that, when there were no rock fragments in the soil, Cl−1 was present in the effluent in a relatively short time and also took a short time to break through. This can be attributed to the homogeneity of the pore system and the high Ks value. In contrast, when Rw was 60%, although Cl−1 was present in the effluent in a relatively short time, it took a relatively long time for breakthrough to occur. The former can be attributed to the rapid transport of solute through the macropore system while the latter results from the increased tortuosity, which led to the longer time required in order to saturate the zones with finer pores created below the rock fragments and the immobile regions. The longest time between the initial detection of Cl− and breakthrough was observed when Rw was 50%.
Table 3 compares the two models describing solute transport. The SSQ values indicate that both models predicted values for relative concentrations of Cl− that fitted the experimental data well. However, in general, the T-R model fits the experimental data slightly better than the CDE. This can be attributed to the greater number of parameters used in the T-R model than were used in the CDE that more accurately describe the solute transport processes.
Values of the Peclet number, Pe, dispersivity, λ, the mobile-immobile partition coefficient, β, and the mass transfer coefficient, ω, when derived by either the convection dispersion equation (CDE) or the two region model (T-R).
Tableau 3 Valeurs du nombre de Péclet, Pe, de la dispersivité, λ, du coefficient de partage mobile-immobile β, et du coefficient de transfert de masse ω, obtenus à partir de l’équation de convection–dispersion CDE ou du modèle à deux régions T-R.
| Rw (%) | SSQ | Pe | λ | β | ω | |||||
| CDE | T-R | CDE | T-R | CDE | T-R | CDE | T-R | CDE | T-R | |
| 0 | 3.53E-03 | 2.19E-03 | 160.64 | 210.49 | 0.42 | 0.25 | - | 0.72 | - | 0.065 |
| 10 | 1.04E-02 | 6.48E-03 | 61.73 | 71.70 | 1.02 | 0.59 | - | 0.67 | - | 0.099 |
| 20 | 1.15E-02 | 2.44E-03 | 33.02 | 39.99 | 1.85 | 0.94 | - | 0.61 | - | 0.13 |
| 30 | 1.59E-02 | 3.32E-03 | 33.05 | 43.25 | 1.62 | 0.84 | - | 0.66 | - | 0.15 |
| 40 | 1.72E-02 | 3.13E-02 | 150.61 | 172.70 | 0.48 | 0.27 | - | 0.88 | - | 0.01 |
| 50 | 1.11E-02 | 5.32E-03 | 19.79 | 20.22 | 2.77 | 1.70 | - | 0.80 | - | 0.15 |
| 60 | 3.47E-02 | 4.40E-03 | 29.89 | 27.60 | 1.77 | 0.63 | - | 0.86 | - | 0.46 |
Table 3 lists the Peclet number, Pe, and the dispersivity, λ, determined from the values of the dispersivity coefficients obtained by fitting either the CDE or the T-R model, and the mobile–immobile partition coefficient, β, and the mass transfer coefficient, ω, derived by the T-R model alone.
From Table 3, it can be seen that the values of the dispersivity, λ, when calculated by the CDE were about twice as great as those obtained with the T-R model and followed the same trend (Table 3). This can be attributed to the fact that the CDE combines the effect of solute exchange between the mobile and immobile regions into one term, the equivalent dispersion coefficient, while the processes of hydrodynamic dispersion and diffusional exchange are treated separately by the T-R model [29]. Thus, the value for dispersivity in the T-R model accounts for only part of the dispersion phenomenon, the other part being accounted for by the exchange coefficient, and is therefore always smaller than the dispersivity value obtained by the CDE. The values of the dispersivity are an indication of the distance travelled by the solute. Thus, the value was small when the path taken by the solute in the column was less tortuous as was the case for the stoneless soil, the most homogeneous medium studied. It was also relatively small when the rock fragment content was 40% with values comparable to those of the stoneless soil (Table 3). It appears that the solute was totally excluded from regions within that soil–rock fragment matrix and was confined to direct, almost vertical, pathways. Since such pathways would have only existed in a fraction of the volume occupied by the medium, this explanation could also account for the low Ks value observed.
The Pe values of the soil column varied with the various rock fragment contents (Table 3). When the Rw reached 40%, the value of the Pe was the largest of all the mixtures and comparable to that of the stoneless soil. The values of the Peclet number determined using values for the dispersion coefficient derived from the CDE and the T-R model generally followed the same trend for both models, although those predicted by the former are typically lower than those obtained by the latter, and reflect the inverse relationship between the Peclet number and the dispersivity (Table 3). The Peclet number is a measure of the movement of the solute by mass flow moving vertically in the columns compared to dispersion that involves advective movement and higher numbers indicate that the latter process dominates. Thus, in all the columns, diffusive flow was the dominant component of solute transport but was especially so for those that contained stone-free soil and the soil with an Rw of 40%. This indicates that there was minimal impedance to sideways movement in the homogeneous stoneless soil, as expected, and also that in the Rw = 40% mixture, which suggests that possibly the macropores present in this mixture may have been dominated by horizontally oriented larger pores. It appears that even low rock fragment contents impeded diffusive flow considerably and that increases in the rock fragment content above Rw = 40% also impeded diffusive flow (Table 3).
The mobile–immobile partition coefficient, β, of the T-R model represents the fraction of solute present in the mobile region under equilibrium conditions [25]. Table 3 showed that this fraction in mobile water initially declined with increasing rock content but values did not vary greatly for contents between 10 and 30%. The fraction became significantly larger when Rw was 40% but again did not vary greatly when Rw was between 40 and 60%. This suggests that, once the rock fragment content volume reaches the critical level of about 40%, which corresponded to the creation of a significant macropore system, the volume of the column occupied by the immobile regions increased significantly and was not subsequently greatly affected by the further development of that system. These immobile regions might usually be expected to result in increased advection of solute, which was indicated by the highest Peclet number derived for Rw = 40%.
The mass transfer coefficient, ω, of the T-R model is conceptually related to the dispersivity, λ. Thus, in general, the trends in ω are a reflection of those seen in λ although the changes in magnitude are not directly proportional (Table 3).
4 Conclusions
This study shows that the presence of rock fragments in soil affects the infiltration process but that the overall continuity of the medium was sufficient to result in smooth cumulative infiltration curves, described well by a power function, even when 60% consisted of rock fragments by weight. Infiltration rates over time and the Ks values first decreased with increasing rock fragment content to an observed minimum, when the content was 40%, and then increased. Both the Peck–Watson equation and Bouwer–Rice equation provided estimates of Ks that were close to the experimental data when the rock fragment content was less than 20% but failed to predict the trends observed experimentally for higher contents. The increases in Ks and infiltration above rock fragment contents of about 40% were ascribed to the development of a more continuous macropore system.
The solute transport experiment confirmed the overall continuity of the medium even for the most heterogeneous soil–rock mixture as indicated by the smooth breakthrough curves. For larger rock fragment contents, breakthrough curves were described slightly better by the T-R model than by the CDE. However, the simpler CDE proved to be sufficiently accurate for simulating solute movement in the soil containing rock fragments. The Peclet number was largest when Rw was 40% corresponding to the lowest observed Ks, indicating that advection occurred at the greatest degree in this mixture of those studied. The mobile–immobile partition coefficient, β, values were similar when the rock fragment contents were either below or were above or equal to 40% but differed significantly around this critical content.
This study provides evidence that can enhance the understanding of the role of rock fragments in water movement and solute transport in soils on the Loess Plateau. Knowledge of the effect of the infiltration and solute transport processes in soils in relation with the content of rock fragments can facilitate management strategies that can reduce runoff and associated soil erosion, and/or nutrient losses on the Loess Plateau. Further in depth and systematic research is required as the distribution and range of the rock fragments in soils is more complex in nature and results need to be transferred to the field.
Acknowledgements
The work in this paper is supported by National Natural Science Foundation of China (50479063, 40025106, 90102012, 90502006), Shao Ming-An's Changjiang Scholar Innovation Team Projects of Education Ministry of China and Innovation Team Group Projects of Northwest A&F University, the International Cooperative Partner Plan of Chinese Academy of Sciences, and One-hundred-Talent Plan of Chinese Academy of Sciences.The authors thank the four reviewers for the conductive comments on the manuscript, especially Professor G. de Marsily (at University Paris VI, Member, French Academy of Sciences) for the excellent editing and correcting of the revision. We also extended our sincere thanks to Assistant Editor, Marina Jimenez for the kind suggestions and guidance.