- Email: [email protected]
Use of rice husk ash as a stabilizer to reduce soil loss and runoff rates on sub-base materials of forest roads from rainfall simulation tests
Catena 150 (2017) 116–123
Contents lists available at ScienceDirect
Catena journal homepage: www.elsevier.com/locate/catena
Use of rice husk ash as a stabilizer to reduce soil loss and runoff rates on sub-base materials of forest roads from rainfall simulation tests Mehran Nasiri a,⁎, Majid Lotfalian a, Amir Modarres b, Wei Wu c a b c
Department of Forest Engineering, Sari University of Agricultural Sciences and Natural Resources, Sari, Iran Department of Civil Engineering, Babol University of Technology, Babol, Iran Institute of Geotechnical Engineering, University of Natural Resources and Life Sciences, Vienna, Austria
a r t i c l e
i n f o
Article history: Received 10 December 2015 Received in revised form 16 October 2016 Accepted 7 November 2016 Available online xxxx Keywords: Unpaved roads Stabilization Sediment concentration Runoff coefficient CBR MDD
a b s t r a c t The impact of rice husk ash (RHA) as a stabilizer to reduce soil loss and surface runoff rates on sub-base materials of forest roads has been studied using a portable rainfall simulator at an intensity of 52 mmh−1. Thirty rainfall simulations were carried out on different combinations of materials: on the natural sub-base soils (5), on the materials stabilized with pure lime (5), and on the materials stabilized with different percent of RHA and lime (20). Results indicated that on natural sub-base soils, the runoff coefficient was 53.6% and mean time to runoff was 87 s. On the materials stabilized with pure lime, the runoff coefficient and mean time to runoff were measured 58.7% and 63 s, respectively. The lowest runoff coefficient was measured 36.6% on the combination of soil + 6% lime + 9% RHA. However, the highest mean time to runoff was recorded 198 s on the combination of soil + 4% lime + 9% RHA. The maximum and minimum soil loss rates were found on the natural sub-base soils (212.2 g m−2) and on the combination of soil + 6% lime + 9% RHA (162.4 g m−2), mainly due to changes in maximum dry density (MDD), plasticity index (PI), optimum moisture content (OMC) and CBR of materials. On the basis of the results, we concluded that the rice husk ash not only increases the materials quality in soil stabilization methods with lime, but also reduces the runoff and soil loss rates on unpaved forest roads. © 2016 Published by Elsevier B.V.
1. Introduction Road-related factors, such as the road construction and timber harvest on unstable soils, dense road networks, poor drainage and inappropriate pavement materials contribute to a high probability of observing changes in peak flows and sedimentation (Selkirk and Riley, 1996; Ramos-Scharron and MacDonald, 2005; Safari et al., 2016). Unpaved forest roads can create an impermeable layer for the initiation of surface flow. Then, large volumes of overland flow may travel downslope toward the stream network (Reid and Dunne, 1984; Croke et al., 1999; Jordán et al., 2009). The rates of soil loss and surface runoff on unpaved forest roads depend on terrain features and topography, geometric design, age of road, hydraulic parameters, drainage condition and traffic volume (Coker et al., 1993; Cao et al., 2009). The most important pavement layer of forest roads (access roads) is sub-base layers, since the construction of base layer on forest roads is not justified according to its standards due to economic situation of forestry projects. Unconsolidated material with poor structure is susceptible to erosion during precipitation events. A stable structure with larger voids in the pavement ⁎ Corresponding author at: Sari University of Agricultural Sciences and Natural Resources, PO Box: 737, Sari, Iran. E-mail address: [email protected] (M. Nasiri).
http://dx.doi.org/10.1016/j.catena.2016.11.010 0341-8162/© 2016 Published by Elsevier B.V.
layer promotes infiltration of water during rainstorms and results in reduced runoff and soil loss (Ziegler et al., 2000). Materials of the pavement layer have an important role on the rates of soil loss and runoff. The production of sediment from the road surface depends on materials of road surface, traffic density, road dimensions and road gradient. Luce and Black (1999) stated that “contributing segment area and road materials are the key controlling factors for road surface-related sediment productions”. Sheridan and Noske (2007) found that soil-surface roads contributed 25 times more sediment than gravel paved roads and Demir et al. (2012) stated that total sediment production of unpaved forest road was 1.96 times higher than that of paved forest road. The materials used for the sub-base layer of forest roads generally contain N 12% silt and clay. The kinetic energy of raindrops break up soil aggregates and fine particles detached from soil. The detachment of soil particles can increase the soil erosion after rainstorms (Rimal and Lal, 2009; Jordán et al., 2009). To have the required strength to resist the kinetic energy of raindrops, tensile stresses and strains spectrum the materials used for the sub-base layer should have suitable specification. One way to improve the mechanical properties of these soils is the use of soil stabilizers. Pozzolanic materials such as rice husk ash (RHA) with silica and high specific surface can be used as an inexpensive stabilizer (RHA is not suitable for cattle feeding and it is also non-biodegradable) in soil stabilization method with lime (Chobbasti et al., 2010;
M. Nasiri et al. / Catena 150 (2017) 116–123
Weiting et al., 2012). Low cost and availability of RHA have led many researchers to investigate RHA as an alternative for soil stabilization (Basha et al., 2005; Nair et al., 2008; Harichane et al., 2011; Hossain and Mol, 2011; Weiting et al., 2012; Jamil et al., 2013). They stated that addition of lime and RHA can improve the mechanical properties of soil including optimum moisture content (OMC), maximum dry density (MDD), California Bearing Ratio (CBR), unconfined compressive strength (UCS) and plasticity index (PI). Several studies have reported the effect of pozzolanic materials on mechanical properties' improvement and durability of sub-base soils (Onyango et al., 2007; Alhassan, 2008; Chobbasti et al., 2010; Trivedi et al., 2013). Over the past two decades, many studies have been carried out to determine runoff and soil loss rates from unpaved roads (Reid and Dunne, 1984; Selkirk and Riley, 1996; Ziegler et al., 2001; Arnaez et al., 2004; Jordán and Martínez-Zavala, 2008; Cao et al., 2009; Safari et al., 2016). However, very few studies have been conducted on the impact of pozzolanic materials on runoff and soil loss rates. The aim of this research is to quantify runoff and soil erosion from unpaved forest roads according to the different combinations of sub-base materials. In this
117
work, rainfall simulation tests were used to (a) study the hydrological and erosive response from different combinations of sub-base materials (soil, lime and rice husk ash) and (b) correlate these data to the soil mechanical properties of materials. 2. Materials and method 2.1. Study area This study was carried out in a road of Caspian forest (Azarrood basin), in northern Iran, approximately on the coordinates 36° 19′ 72″ N and 52° 83′ 22″ E. In this region, there is a moderate mountainous climate with cold winters and humid summers and the mean annual rainfall is about 800 mm. Alborz earth dam with a height of 72 m was constructed below the confluence of several rivers including the Skelimrood, Azarrood and Karsangrood (Fig. 1,a). Total length of forest roads (surrounding the Alborz dam) and density of these roads are 17 km and 10.5 m per hectare, respectively. There is low-volume of traffic on these roads by forestry machinery. However, this network is used
Fig. 1. Alborz earth dam and construction of forest roads in the Azarrood forest of Iran (a); produced RHA and rice husk (b); laboratory tests (unconfined compression apparatus(Afrazma)) to identify specification of different combinations (c); rainfall simulations to measure the runoff and soil loss (d) and use of mobile pump for field experiments (e).
118
M. Nasiri et al. / Catena 150 (2017) 116–123
by public transport on holidays due to presence of tourism areas such as Gazoo waterfall and natural forests. This traffic can increase the sediment delivery from unpaved roads. Our observation showed that large volumes of runoff moved downslope toward the earth dam after a rain storm in September 2015. Therefore, due to the presence of earth dam and soil erosion issues, one of the most important solutions to reduce sediment delivery from forest roads as well as strengthening the road materials is the use of stabilizers. 2.2. Identification of soil, lime and RHA About 350 kg of soil were collected from the study area. Soils were put in 50 kg bags and transferred to the soil mechanics laboratory. In the laboratory, soils were remixed well together. In order to identify the soil, laboratory tests such as sieve analysis and Aterberg tests were done. According to the AASHTO classification and unified system, the studied soil is classified as A-6 and SC (Silty sand with gravel), respectively. The materials used for the access roads are usually placed in classes of A-4, A-5, A-6 and A-7 according to AASHTO classification. These soils generally contain silt and clay and are classified in poor to average classes in terms of road construction. In order to stabilize the sub-base materials, at first rice husk and hydrated lime were prepared from farm fields of northern Iran and the Alborz industrial factory, respectively. The type of used lime is slaked lime and some characteristics of that are shown in Table 1. To make rice husk ash, rice husk must be burned at high temperatures (about 500 °C). After burning, the obtained RHA were placed in the open air to complete the ash production process (Fig. 1,b). Some chemical and physical characteristics of rice husk ash are shown in Tables 2 and 3. 2.3. Mix design and laboratory studies By examining different studies it can be understood that one of the most important factors of mix design is soil properties. The soil type has an important effect on pozzolanic activity. Fine-grained soils with clayey mineral are suitable in order to stabilize soils with rice husk ash. The clay minerals have the property of absorbing certain anions and cations and retaining them in an exchangeable state. The exchangeable ions are held around the outside of the silica-alumina clay mineral structural unit. Therefore, materials mixed well together (Table 4) according to the soil properties, performance of lime and RHA in improving materials quality (Alhassan, 2008; Chobbasti et al., 2010; Muntohar, 2002). Soil laboratory tests were done to evaluate the effect of stabilizers after mixing materials. Then, different combinations of materials were studied with 5 replications using tests of CBR or California Bearing Ratio (%), maximum dry density or MDD (Mg m−3) and optimum moisture content or OMC (%), unconfined compressive strength or UCS (KN m−2) and plasticity index or PI (%). The California Bearing Ratio test, is an empirical test, which is used as an important criterion in pavement design. With this test, the bearing value of road sub-base and subgrade, can be estimated. According to the purpose of the study, this test can examine the strength of stabilized samples compared to the natural soil samples (control). This test was conducted (ASTM D1883-07) using the load-measuring device which is connected to the compression machine with loading rate of 1 mm/min. The proctor compaction tests (MDD and OMC) were performed with the standard of ASTM D069807E01 in CBR mould; by conducting this test, it can be found that how much free space of aggregates is filled by stabilizers and how
Table 1 Characteristics of the used lime.
Table 2 Chemical characteristics of the rice husk ash (*10−6). P (%)
Mg2+ (%)
Ca2+ (%)
K (%)
Na (%)
SiO2 (%)
24.7
66.1
0.016
0.124
0.114
83.7
affect the optimum water content. One of the most useful methods of evaluating the strength of stabilized soil is UCS test (Fig. 1, c). The required amount of additive to be used in stabilization of soil could be determined by this test. In this study, the test was performed for samples with 28 days curing time to consider the time factor (Alhassan, 2008). This test was performed with standard of ASTM D 2166. The compression device (a hydraulic-actuated loading piston with the capability of infinite rates of strain and stress loads) was used to measure the UCS rates. The Aterberg limit (Plastic Limit (PL), liquid limit (LL) and plasticity index (PI)) test was performed by standard of ASTM D4318-05. The liquid limit and Plastic Limit were measured using a Casagraunde cup and rolling method, respectively. The Plastic Limit is often used together with the liquid limit to determine the plasticity index. Plastic limit of fine-aggregates in sandy materials have a major influence on the strength of used materials. Materials with a PI more than allowed values should not be used on base and sub-base layers, because materials with higher PI have lower shear strength. 2.4. Rainfall simulation Rainfall simulations were carried out to measure runoff and soil loss from the different combinations of materials. In order to simulate rainfall, a squared plot of 60 × 60 cm2 that is delineated by a steel frame was fabricated (Fig. 1, d). The frame was carefully tapped into the soil to prevent leakage and the simulator was covered with a plastic cover to eliminate the wind effect. A minimum slope of 3% (the minimum longitudinal slope of forest roads) was considered for soil surface in each plot and longitudinal gradient was determined using a Suunto clinometer. A portable single nozzle rainfall simulator (Schlick r18650), which is connected through a rubber pipe to an automatic pump (Fig. 1, e) was used to simulate rainfall. According to weather stations of the Alborz dam and Gharakheil, the mean rainfall intensity for the experiments was selected 52 mm h−1. For each mixture the duration and replication of the rainfall simulations were considered 30 min and 5 replication, respectively. Time to runoff, runoff coefficient, runoff rate, sediment concentration and soil loss rate were determined for each plot. Usually, a steady-state runoff rate was reached after 15–20 min. So, Runoff coefficients were calculated as the runoff rate divided by the rainfall rate for each sample. Runoff samples were collected every 5 min in order to determine the sediment concentration in runoff and to calculate the soil loss rates for each plot. Each sample was filtered using Whatman filter papers and then the filter oven dried at 105 °C for 24 to 48 h and weighed. Then, sediment mass per volume of runoff was determined for each sample. 2.5. Data analysis The effects of different combinations of materials on runoff and soil loss rates were evaluated by statistical analysis of the data. All computations were done using SPSS version 16. This included assumptions of normality test using the Shapiro–Wilk, an ANOVA test to compare means among treatments and the Pearson test to identify correlations
Table 3 Physical characteristics of the rice (Tarom) husk ash.
Non hydrated CaO (%)
Non hydrated MgO (%)
pH
CO2 (%)
Ca(OH)2 (%)
Density (kg m−3)
Matter size/μm
Special surface (m2 g−1)
1.62
0.8
11.2
0
92.8
2200–2300
4.8–8.9
60–130
M. Nasiri et al. / Catena 150 (2017) 116–123
and Mol, 2011. He stated that addition of rice husk ash reduced the plasticity of soil and improved the desired engineering characteristics of soil such as shear strength, unconfined compressive strength and load bearing capacity. The addition of 9% RHA was able to reduce soil MDD to 1.23 Mg m3. Reduction in MDD can be related to the replacement of soil by RHA in mixtures which have relatively lower specific gravity compared to natural soil (Osinubi and Katte, 1997).
Table 4 Investigated mixtures and mentioned indices. Mix design
Index
Natural soil Soil + 6% lime Soil + 4% lime + 7% RHAa Soil + 6% lime + 9% RHA Soil + 6% lime + 7% RHA Soil + 6% lime + 9% RHA
S S S S S S
a
+ + + + +
119
6L 4L + 7R 4L + 9R 6L + 7R 6L + 9R
Rice husk ash.
3.2. Hydrological and erosive response between runoff, sediment and soil mechanical properties of materials. Also, the graphical displays were made using the Excel and SigmaPlot softwares.
3. Results and discussion 3.1. Characteristics of materials Mechanical properties of natural soil and different combinations of materials are shown in Table 5. After the ANOVA, some significant differences between the different combinations of materials were detected. The increase in OMC, UCS and CBR was a result of additive materials. Lime and RHA can reduce the quantity of free silt and clay and increase the coarser materials with larger surface areas. The RHA content has an important effect on OMC values. According to Table 5, it can be found that the soil OMC for combinations of S + 4L + 9R and S + 6L + 9R had increased 19 and 21.9%, respectively compared to the natural soil samples. The major chemical compounds present in Rice Husk ash contribute active role in chemical reactions with constituents of soil is Silica Si-ions. Calcium is another additive to form binding compounds along with soil. Lime is added at different doses to supplement Ca to form binding compounds along with Silica. When lime is mixed with water, it forms calcium hydroxide, called slaked lime. This chemical reaction increased the OMC value and requires more water to supply more Ca ions for the cation exchange reaction. This result is in agreement with Ramezanianpour et al., 2009; Chobbasti et al., 2010 and Purbi et al., 2011. The UCS value did not have significant changes with increasing lime content (4% lime or 6% lime). The increase in UCS is attributed to the formation of cementations compounds between the CaOH present in the soil and RHA as well as the pozzolans present in the RHA (Jamil et al., 2013). According to the laboratory results the maximum and minimum values of plasticity index were recorded for the natural soil and combination of S + 6L + 9R respectively. Table 5 shows the lime can be much more effective on reduction of the plasticity index if mixed with RHA. The reason for reduction in plasticity index could be the chemical reaction between limes and soil that require more water (Jamil et al., 2013; Ramezanianpour et al., 2009). This result is in agreement with Hossain
Hydrological response of natural soil and other combinations are shown in Table 6. The ANOVA analysis showed significant differences in time to runoff data between the different combinations of materials (p b 0.0001). The mean time to runoff for the natural soils (without additive materials) and combination of S + 6L were recorded 87 s (57–104 s) and 63 s (35–95 s) respectively. In other studies, time to runoff was recorded 25–89 (Jordán and Martínez-Zavala, 2008) and 23–60 (Jordán et al., 2009). According to result the mean time to runoff was longer for the combinations of S + 4L + 9R (198 s) and S + 6L + 9R (173 s). The ANOVA also showed significant differences for runoff coefficients between the materials stabilized with RHA and other combinations. Materials stabilized with RHA can increase the infiltration rates due to large particles with larger voids. Several researchers studied the runoff coefficients on unpaved forest roads and gravel surface roads. The rate of runoff coefficient was measured 31% on gravel surface roads under a rainfall intensity of 100 mm·h−1 (Selkirk and Riley, 1996). Ziegler et al. (2001) in a rainfall simulation study (100 mm·h−1 rainfall intensity) reported that the rate of runoff coefficients on unpaved forest roads was 84%. Also, in similar studies the rate of runoff coefficients were measured 50.7% under a rainfall intensity of 75 mm·h− 1 (Arnaez et al., 2004), 46.6% under a rainfall intensity of 72 mm h−1 (Jordán and Martínez-Zavala, 2008) and 58.6% under a rainfall intensity of 90 mm h−1 (Jordán et al., 2009) on unpaved forest roads. Table 6 shows sub-base soils stabilized with RHA have a response similar (35.9% on the combination of S + 4L + 9R) to gravel surface roads (Selkirk and Riley, 1996) and soils without additive materials have a response similar (53.66% on the natural sub-base materials) to unpaved roads (Arnaez et al., 2004; Jordán et al., 2009). Fig. 2 shows the highest rate of runoff was recorded for natural soils (combinations without additive materials) at the beginning of experiments (5–15 min). Runoff rate increased slowly and in materials stabilized with RHA, a steady-state runoff was reached after 15–20 min. Addition of pure lime to the natural soil can decrease the swelling and soil contraction as well as increase the optimum water content and soil resistance including compressive and tensile strengths (Alhassan, 2008). Absorption rate of materials stabilized with pure lime was more than the natural soils at the beginning of experiments because of more optimum water content. After that the runoff rate increased quickly during the experiments (8–20 min) due to decrease in plasticity
Table 5 Soil mechanical properties of different combinations of materials. Parameter PI⁎ (%) OMC (%) MDD (Mg m−3) UCS-28d* (KN m−3) CBR (%)
M⁎ SD⁎ M SD M SD M SD M SD
Soil
S + 6L
S + 4L + 7R
S + 4L + 9R
S + 6L + 7R
S + 6L + 9R
ANOVA, p
10.82d 0.65 16.3a 1.65 1.54c 0.005 213a 3.4 9.24a 0.97
4b 0.16 24.4ab 1.74 1.42bc 0.06 225ab 4.65 36.3b 2.6
4.24b 0.37 35.02c 3.7 1.37abc 0.08 232b 3.96 46.6b 2.1
2.36a⁎ 0.1 36.2c 0.84 1.23a 0.02 234b 3.36 43.8b 3.6
5.66c 0.49 32.3bc 3.7 1.37abc 0.03 231b 3.8 42.2b 4.1
1.13a 0.3 39.4c 2.54 1.24ab 0.06 235b 3.8 47b 4.2
0.000
⁎Within-a-row means followed by the same letter are not significantly different; M⁎: mean; SD⁎: standard deviation; PI⁎: plasticity index; 28d⁎: 28 days curing time.
0.000 0.000 0.000 0.000
120
M. Nasiri et al. / Catena 150 (2017) 116–123
Table 6 Average time to runoff, runoff coefficients, and total soil loss according to the combinations of materials. Parameter Time to runoff (s)
M SD M SD M SD
Runoff coefficient (%) Total soil loss (g m
−2
)
Soil
S + 6L
S + 4L + 7R
S + 4L + 9R
S + 6L + 7R
S + 6L + 9R
ANOVA, p
86.7a 8.6 53.66b 1.88 212.2b 2.27
63.4a⁎ 12.3 58.7c 1.43 171.4ab 7.6
166bc 14.3 38. 5a 1.25 176.4ab 6.7
198c 12.4 35.9a 0.71 173.1ab 5.6
144.1b 11.4 37.1a 0.34 175.4ab 2.33
172.7bc 10.4 36.6a 0.68 162.4a 5.9
0.0000 0.0000 0.0000
⁎Within-a-row means followed by the same letter are not significantly different; M⁎: mean; SD⁎: standard deviation.
index and porosity of materials. After addition of RHA and lime, fine particles of soil stick to each other which result in large particles with larger voids (Chobbasti et al., 2010). Gradual increase in runoff rates on materials stabilized with RHA can be related to these voids. In this study higher quantity of RHA resulted in lower runoff rates because it takes more time for the water to infiltrate into the soil voids. The minimum and maximum rates of soil loss were recorded on the combination of S + 6L + 9R (162.4 g·m2) and natural soil (212.2 g·m− 2) respectively (Table 6). The materials stabilized with RHA and lime had the same behavior in sediment generation. So that, the ANOVA analysis showed significant differences in total soil loss between the natural soils and materials stabilized with RHA and lime. The RHA and lime can increase the strength properties of materials. Reduction in soil loss can be related to the soil strength properties which have higher CBR rate and lower MDD and plasticity indices compared to natural soil. Fig. 3 shows sediment concentration in runoff from different combinations of sub-base materials. Sediment concentration increased during the first 5–10 min of rainfall on the combinations of S + 4L + 7R, S + 4L + 9R, S + 6L + 7R and S + 6L + 9R. However, the rates of sediment concentration decreased after 10 min from the beginning of the rainfall simulation. Fine particles among the soils were washed out at
the beginning of the test. Therefore, the rate of sediment concentration increased at the beginning of the rainfall simulation and then decreased due to the loss of fine surface soil particles (Reid and Dunne, 1984; Arnaez et al., 2004). The rate of sediment concentration was greater for the natural soils (12.52 g L−l) than materials stabilized with RHA and lime. Several researchers studied the sediment concentration in runoff during the rainfall simulation. The sediment concentration was measured 1.8 g L−l for gravel surface road under a rainfall intensity of 100 mm·h−1 (Selkirk and Riley, 1996). This rates were reported on unpaved roads 4.4 g L−l under a rainfall intensity of 72 mm h−1 (Jordán and Martínez-Zavala, 2008), 12.2 g L-l under a rainfall intensity of 33 mm h−1 (Lotfalian et al., 2013). In this study the erosive response of subbase materials stabilized with RHA and lime was similar to gravel surface roads (Selkirk and Riley, 1996) and materials without additive have a response similar to unpaved roads (Lotfalian et al., 2013; Jordán et al., 2009). Fig. 4 shows the relationship between soil loss rates and runoff coefficients for different combinations of materials. The lowest values of steady-state runoff and soil loss rates were measured on the materials stabilized with RHA and lime. These materials (63.3%) have same behavior under rainfall simulation tests. The highest runoff coefficients and soil losses were recorded on the materials stabilized with pure 30
45
25
35
Runoff rate (mm h -1)
Runoff rate (mm h -1)
40
30 25 20 15 10 5
Soil S+6L S+4L+7R
0 0
5
10
15
20
25
30
20 15 10 5 Soil S+4L+9R
0
35
0
5
10
30
30
25
25
20 15 10 5 0
Soil S+6L+7R 0
5
10
15
15
20
25
30
35
Time (min)
Runoff rate (mm h -1)
Runoff rate (mm h -1)
Time (min)
20
Time (min)
25
30
20 15 10 5 Soil S+6L+9R
0
35
0
5
10
15
20
25
Time (min)
Fig. 2. Behavior of surface runoff from rainfall simulations on different combinations of sub-base materials.
30
35
20
Sediment Concentration (gr L-1)
Sediment Concentration (gr L-1)
M. Nasiri et al. / Catena 150 (2017) 116–123
Soil S+6L S+4L+7R
18 16 14 12 10 8 6 4 2 0 0
5
10
15
20
25
30
121
20 Soil S+4L+9R
18 16 14 12 10 8 6 4 2 0
0
35
5
20 Soil S+6L+7R
18 16 14 12 10 8 6 4 2 0 0
5
10
15
20
10
15
20
25
30
35
Time (min)
Sediment Concentration (gr L-1)
-1
Sediment Concentration (gr L )
Time (min)
25
30
35
Time (min)
20 Soil Soil+6L+9R
18 16 14 12 10 8 6 4 2 0 0
5
10
15
20
25
30
35
Time (min)
Fig. 3. Sediment concentration in runoff from different combination of materials.
lime and natural soils, respectively. In this study it appears that the mixture of S + 6L + 9R was the best combination to reduce runoff and soil loss rates. Road engineers make use of optimum moisture content when they apply water to soil to compact. Excessively compacted soil results in problems such as reduced rainfall infiltration. When soil voids are filled with pure lime and RHA, compaction naturally occurs in materials due to pozzolanic reaction between materials. The compaction of the materials stabilized with lime and RHA can partly explain the lower soil loss rates. Our observation during the study showed that compaction can facilitate runoff generation on materials stabilized with pure lime due to less voids space than when using RHA.
65
Runoff coefficients (%)
60 55 50 45
Soil S+6L S+4L+7R S+4L+9R S+6L+7R S+6L+9R
40 35 30 140
150
160
170
180
190
200
210
220
Soil loss rate (gr m -2) Fig. 4. Relationship between runoff coefficients and soil loss rates for different combinations of materials.
3.3. Correlation between runoff, soil loss and materials Table 7 shows the R-Pearson coefficients between runoff rate, soil loss rate and soil mechanical parameters of different combinations of materials. The MDD and OMC indices show how much free space of aggregates is filled by stabilizers and how that affects the optimum water content. According to the results, soil compaction characteristics (the OMC and MDD) have an important effect on runoff rates due to changes in soil voids, swelling and PI. Some significant and negative correlations were found between runoff rates, the PI and MDD indices of materials stabilized with RHA and lime, but not significant for natural soil. In addition, some positive significant correlations were found between runoff rate and OMC index of materials stabilized with RHA. We found some significant and negative correlations between soil loss rates and the CBR as well as some significant and positive correlations between soil loss rates and the MDD of materials stabilized with RHA and lime (Table 7). Also, no significant correlation was detected between soil loss rates and the OMC of materials. The CBR value is used as an index of compacted soil strength and bearing capacity. Stabilized soil with lime and RHA, increased the CBR values and decreased the MDD values. Lime and RHA can reduce fine particles and increase the cohesion among aggregates (Harichane et al., 2011). Therefore, in this study addition of lime and RHA to the soils increased the strength of materials and consequently the rate of soil loss decreased. Table 7 shows some significant and positive correlations between PI and the MDD as well as some significant and negative correlations between OMC and the PI and MDD. We found some significant and negative correlations between CBR and the PI and MMD of materials stabilized with RHA and lime. The CBR value of soil may depends on many factors such as MDD, OMC, liquid limit, Plastic Limit, PI, type of soil and permeability of soil. The CBR value decreases with the increase in the plasticity index and optimum moisture content of soil but increases with the increase in the maximum dry density. Talukda (2014) stated that there
122
M. Nasiri et al. / Catena 150 (2017) 116–123
Table 7 R-Pearson correlation between runoff rates, soil loss rate and soil mechanical parameters of materials. Variables Runoff rate Soil S + 6L S + 4 L + 7R S + 4 L + 9R S + 6 L + 7R S + 6 L + 9R Soil loss rate Soil S + 6L S + 4 L + 7R S + 4 L + 9R S + 6 L + 7R S + 6 L + 9R Soil properties MDD OMC UCS-28 day CBR
Soil loss rate R
PI R
MDD R
OMC R
UCS-28 day R
CBR R
0.86* 0.48 0.78* 0.69 0.7 0.65
−0.68 −0.79* −0.52 −0.65 −0.75* −0.53
−0.63 −0.73* −0.8* −0.88* −0.76* −0.87*
0.54 0.35 0.85* 0.86* 0.73* 0.63
0.53 0.65 0.37 0.42 −0.01 0.55
0.69 0.71 0.29 0.32 0.52 0.66
0.35 0.19 −0.28 0.29 0.68 0.23
0.39 0.77* 0.53 0.59 0.87* 0.82*
0.37 −0.29 −0.41 −0.56 0.35 −0.02
0.03 −0.83* −0.56 −0.7 −0.61 −0.34
−0.64 −0.87* −0.49 −0.76* −0.91* −0.54
0.82* −0.79* −0.61 −0.88*
−0.89* 0.58 0.83*
0.44 −0.47
0.7
⁎ Marked R-Pearson coefficients are significant (b0.05).
is some significant correlation between the CBR of fine grained soil (ML and MI) with PI, MDD and OMC values.
4. Conclusion 1. Additive materials have an important effect on the runoff and soil loss rates. The highest soil loss rate was recorded for the control (natural soil). Given the rice husk is an agricultural waste material and it is available in many countries at little or no cost ($1 U.S. per kg in Iran), the produced ash from rice husk can be used to improve road surface quality and decrease the runoff and soil loss rates. 2. The combination of S + 6L + 9R was selected as the best combination of materials to reduce runoff and soil loss rates. Materials stabilized with RHA and lime decreased the runoff and soil loss rates due to large particles with larger voids, fewer fine aggregates and more strength among soil particles. 3. The highest runoff coefficient was recorded on the materials stabilized with pure lime. These results show that the pure lime is not suitable for stabilization of unpaved forest roads due to high runoff coefficients. 4. Some significant and positive or negative correlations were found between the CBR, dry density, optimum moisture content, plasticity index and the rates of runoff and soil loss. These correlations show the relationship between soil mechanical properties and hydrological and erosive response of materials which are associated with changes in infiltration rates, moisture content, swelling, porosity, shear strength, cohesion and internal friction angle of soils. On the basis of the results, we concluded that the addition of rice husk ash not only increases the materials quality in soil stabilization, but also reduces the runoff and soil loss rates.
Acknowledgements This research was supported by the University of Sari Agricultural Sciences and Natural Resources. We would like to thank Mr. Aghajantabar and Mr. Bozorgpour for their suggestions about pozzolanic materials, field experiments and laboratory tests. We also would like to thank volunteers (Mr. Valizadeh, Mr. Aghajani and Mr. Ahmadi) who devoted their time and knowledge during the field experiments.
References Alhassan, M., 2008. Permeability of lateritic soil treated with lime and rice husk ash. Assumption Univ. J. Thailand 12 (2), 115–120. Arnaez, J., Larrea, V., Ortigosa, L., 2004. Surface runoff and soil erosion on unpaved forest roads from rainfall simulation tests in northeastern Spain. Catena 57, 1–14. Basha, E.A., Hashim, R., Mahmud, H.B., Muntohar, A.S., 2005. Stabilization of residual soil with rice husk ash and cement. Constr. Build. Mater. 19 (6), 448–453. Cao, L., Zhang, K., Zhang, W., 2009. Detachment of road surface soil by flowing water. Catena 76, 155–162. Chobbasti, A.J., Ghodrati, H., Vahdatirad, M.J., Firouzian, S., Barari, A., Torabi, M., Bagherian, A., 2010. Influence of using rice husk ash in soil stabilization methodwith lime. Front. Earth Sci. China 4 (4), 471–480. Coker, R.J., Fahey, B.D., Payne, J.J., 1993. Fine sediment production from truck traffic, Queen Charlotte Forest, Marlborough Sounds. New Zealand. J. Hydrol. 31, 56–64. Croke, J.C., Hairsine, P., Fogarty, P., 1999. Runoff generation and redistribution on disturbed forest hill slopes, south eastern Australia. J. Hydrol. 216, 55–77. Demir, M., Makineci, E., Kartaloglu, M., 2012. Temporal sediment production of paved and unpaved forest road. Fresenius Environ. Bull. 21, 1180–1185. Harichane, K., Ghrici, M., Kenai, S., 2011. Effect of curing time on shear strength of cohesive soils stabilized with combination of lime and natural pozzolana. Int. J. Civil Eng. 9, 90–96. Hossain, K.M.A., Mol, L., 2011. Some engineering properties of stabilized clayey soils incorporating natural pozzolans and industrial wastes. Constr. Build. Mater. 25, 3495–3501. Jamil, M., Kaish, A.B.M.A., Raman, S.N., Zain, M.F.M., 2013. Pozzolanic contribution of rice husk ash in cementitious system. Constr. Build. Mater. 47, 588–593. Jordán, A., Martínez-Zavala, L., 2008. Soil loss and runoff rates on unpaved forest roads in southern Spain after simulated rainfall. For. Ecol. Manag. 255, 913–919. Jordán, A., Martínez-Zavala, L., Bellinfante, N., 2009. Impact of different parts of unpaved forest roads on runoff and sediment yield in a Mediterranean area. Sci. Total Environ. 407, 937–944. Lotfalian, M., Parsakhoo, A., Kavian, A., Hosseini, S.A., 2013. Runoff and sediment concentration of different parts of a road in Hyrcanian forests. Forest Sci. Pract. 15 (2), 144–151. Luce, C.H., Black, T.A., 1999. Sediment production from forest roads in western Oregon. Water Resour. Res. 35, 2561–2570. Muntohar, A.S., 2002. Utilization of uncontrolled burnt rice husk ash in soil improvement. Dimensi Teknik Sipil J. 4, 100–105. Nair, D.G., Fraaij, A., Klaassen, A.K., Kentgens, P.M., 2008. A structural investigation relating to the pozzolanic activity of rice husk ashes. J. Cem. Concr. Res. 38 (6), 861–869. Onyango, M., Macha, I., Busch, C., 2007. Use of Naturally Occurring Pozzolans for Road Construction in Tanzania. 9th International Conference on Low-volume Roads. June 24–27, Austin, Texas . Osinubi, K.J., Katte, V.Y., 1997. Effect of elapsed time after mixing on grain size and plasticity characteristics, soil lime mixes. Nigeria Soc. Eng. Tech. Trans. 32 (4), 65–76. Purbi, S., Mukesh, N., Dixit, M., 2011. Evaluation of strength characteristics of clayey soil by adding soil stabilizing additives. Int. J. Earth Sci. Eng. 4, 1060–1064. Ramezanianpour, A.A., Khani, M.M., Ahmadibeni, G., 2009. The effect of rice husk ash on mechanical properties and durability of sustainable concretes. Int. J. Civil Eng. 7 (2), 83–91. Ramos-Scharron, C., MacDonald, L.H., 2005. Measurement and prediction of sediment production from unpaved roads, St. John, U.S. Virgin Islands. Earth Surf. Process. Landf. 30, 1283–1304. Reid, L.M., Dunne, T., 1984. Sediment production from forest road surfaces. Water Resour. Res. 20, 1753–1761.
M. Nasiri et al. / Catena 150 (2017) 116–123 Rimal, B.K., Lal, R., 2009. Soil and carbon losses from five different land management areas under simulated rainfall. Soil Tillage Res. 106, 62–70. Safari, A., Kavian, A., Kavian, A., Parsakhoo, A., Saleh, I., Jordán, A., 2016. Impact of different parts of skid trails on runoff and soil erosion in the Hyrcanian forest (northern Iran). Geoderma 263, 161–167. Selkirk, J.M., Riley, S.J., 1996. Erodibility of road batters under simulated rainfall. Hydrol. Sci. J. 41, 363–376. Sheridan, G.J., Noske, P.J., 2007. A quantitative study of sediment delivery and stream pollution from different forest road types. Hydrol. Process. 21 (3), 387–398. Talukda, K., 2014. A study of correlation between California Bearing Ratio (CBR) value with other properties of soil. Int. J. Emerg. Technol. Adv. Eng. 2, 559–562.
123
Trivedi, J.S., Sandeep, N., Chakradhar, I., 2013. Optimum utilization of fly ash for stabilization of sub-grade soil using genetic algorithm. Procedia Eng. 51, 250–258. Weiting, X.u., Tommy, Y.L., Shazim, A.M., 2012. Microstructure and reactivity of rich husk ash. Constr. Build. Mater. 29, 541–547. Ziegler, A.D., Sutherland, R.A., Giambelluca, T.W., 2000. Runoff generation and sediment production on unpaved roads, footpaths and agricultural land surfaces in northern Thailand. Earth Surf. Process. Landf. 25, 519–534. Ziegler, A.D., Sutherland, R.A., Giambelluca, T.W., 2001. Interterm surface preparation and sediment detachment by vehicle traffic on unpaved mountain roads. Earth Surf. Process. Landf. 26, 235–250.
Contents lists available at ScienceDirect
Catena journal homepage: www.elsevier.com/locate/catena
Use of rice husk ash as a stabilizer to reduce soil loss and runoff rates on sub-base materials of forest roads from rainfall simulation tests Mehran Nasiri a,⁎, Majid Lotfalian a, Amir Modarres b, Wei Wu c a b c
Department of Forest Engineering, Sari University of Agricultural Sciences and Natural Resources, Sari, Iran Department of Civil Engineering, Babol University of Technology, Babol, Iran Institute of Geotechnical Engineering, University of Natural Resources and Life Sciences, Vienna, Austria
a r t i c l e
i n f o
Article history: Received 10 December 2015 Received in revised form 16 October 2016 Accepted 7 November 2016 Available online xxxx Keywords: Unpaved roads Stabilization Sediment concentration Runoff coefficient CBR MDD
a b s t r a c t The impact of rice husk ash (RHA) as a stabilizer to reduce soil loss and surface runoff rates on sub-base materials of forest roads has been studied using a portable rainfall simulator at an intensity of 52 mmh−1. Thirty rainfall simulations were carried out on different combinations of materials: on the natural sub-base soils (5), on the materials stabilized with pure lime (5), and on the materials stabilized with different percent of RHA and lime (20). Results indicated that on natural sub-base soils, the runoff coefficient was 53.6% and mean time to runoff was 87 s. On the materials stabilized with pure lime, the runoff coefficient and mean time to runoff were measured 58.7% and 63 s, respectively. The lowest runoff coefficient was measured 36.6% on the combination of soil + 6% lime + 9% RHA. However, the highest mean time to runoff was recorded 198 s on the combination of soil + 4% lime + 9% RHA. The maximum and minimum soil loss rates were found on the natural sub-base soils (212.2 g m−2) and on the combination of soil + 6% lime + 9% RHA (162.4 g m−2), mainly due to changes in maximum dry density (MDD), plasticity index (PI), optimum moisture content (OMC) and CBR of materials. On the basis of the results, we concluded that the rice husk ash not only increases the materials quality in soil stabilization methods with lime, but also reduces the runoff and soil loss rates on unpaved forest roads. © 2016 Published by Elsevier B.V.
1. Introduction Road-related factors, such as the road construction and timber harvest on unstable soils, dense road networks, poor drainage and inappropriate pavement materials contribute to a high probability of observing changes in peak flows and sedimentation (Selkirk and Riley, 1996; Ramos-Scharron and MacDonald, 2005; Safari et al., 2016). Unpaved forest roads can create an impermeable layer for the initiation of surface flow. Then, large volumes of overland flow may travel downslope toward the stream network (Reid and Dunne, 1984; Croke et al., 1999; Jordán et al., 2009). The rates of soil loss and surface runoff on unpaved forest roads depend on terrain features and topography, geometric design, age of road, hydraulic parameters, drainage condition and traffic volume (Coker et al., 1993; Cao et al., 2009). The most important pavement layer of forest roads (access roads) is sub-base layers, since the construction of base layer on forest roads is not justified according to its standards due to economic situation of forestry projects. Unconsolidated material with poor structure is susceptible to erosion during precipitation events. A stable structure with larger voids in the pavement ⁎ Corresponding author at: Sari University of Agricultural Sciences and Natural Resources, PO Box: 737, Sari, Iran. E-mail address: [email protected] (M. Nasiri).
http://dx.doi.org/10.1016/j.catena.2016.11.010 0341-8162/© 2016 Published by Elsevier B.V.
layer promotes infiltration of water during rainstorms and results in reduced runoff and soil loss (Ziegler et al., 2000). Materials of the pavement layer have an important role on the rates of soil loss and runoff. The production of sediment from the road surface depends on materials of road surface, traffic density, road dimensions and road gradient. Luce and Black (1999) stated that “contributing segment area and road materials are the key controlling factors for road surface-related sediment productions”. Sheridan and Noske (2007) found that soil-surface roads contributed 25 times more sediment than gravel paved roads and Demir et al. (2012) stated that total sediment production of unpaved forest road was 1.96 times higher than that of paved forest road. The materials used for the sub-base layer of forest roads generally contain N 12% silt and clay. The kinetic energy of raindrops break up soil aggregates and fine particles detached from soil. The detachment of soil particles can increase the soil erosion after rainstorms (Rimal and Lal, 2009; Jordán et al., 2009). To have the required strength to resist the kinetic energy of raindrops, tensile stresses and strains spectrum the materials used for the sub-base layer should have suitable specification. One way to improve the mechanical properties of these soils is the use of soil stabilizers. Pozzolanic materials such as rice husk ash (RHA) with silica and high specific surface can be used as an inexpensive stabilizer (RHA is not suitable for cattle feeding and it is also non-biodegradable) in soil stabilization method with lime (Chobbasti et al., 2010;
M. Nasiri et al. / Catena 150 (2017) 116–123
Weiting et al., 2012). Low cost and availability of RHA have led many researchers to investigate RHA as an alternative for soil stabilization (Basha et al., 2005; Nair et al., 2008; Harichane et al., 2011; Hossain and Mol, 2011; Weiting et al., 2012; Jamil et al., 2013). They stated that addition of lime and RHA can improve the mechanical properties of soil including optimum moisture content (OMC), maximum dry density (MDD), California Bearing Ratio (CBR), unconfined compressive strength (UCS) and plasticity index (PI). Several studies have reported the effect of pozzolanic materials on mechanical properties' improvement and durability of sub-base soils (Onyango et al., 2007; Alhassan, 2008; Chobbasti et al., 2010; Trivedi et al., 2013). Over the past two decades, many studies have been carried out to determine runoff and soil loss rates from unpaved roads (Reid and Dunne, 1984; Selkirk and Riley, 1996; Ziegler et al., 2001; Arnaez et al., 2004; Jordán and Martínez-Zavala, 2008; Cao et al., 2009; Safari et al., 2016). However, very few studies have been conducted on the impact of pozzolanic materials on runoff and soil loss rates. The aim of this research is to quantify runoff and soil erosion from unpaved forest roads according to the different combinations of sub-base materials. In this
117
work, rainfall simulation tests were used to (a) study the hydrological and erosive response from different combinations of sub-base materials (soil, lime and rice husk ash) and (b) correlate these data to the soil mechanical properties of materials. 2. Materials and method 2.1. Study area This study was carried out in a road of Caspian forest (Azarrood basin), in northern Iran, approximately on the coordinates 36° 19′ 72″ N and 52° 83′ 22″ E. In this region, there is a moderate mountainous climate with cold winters and humid summers and the mean annual rainfall is about 800 mm. Alborz earth dam with a height of 72 m was constructed below the confluence of several rivers including the Skelimrood, Azarrood and Karsangrood (Fig. 1,a). Total length of forest roads (surrounding the Alborz dam) and density of these roads are 17 km and 10.5 m per hectare, respectively. There is low-volume of traffic on these roads by forestry machinery. However, this network is used
Fig. 1. Alborz earth dam and construction of forest roads in the Azarrood forest of Iran (a); produced RHA and rice husk (b); laboratory tests (unconfined compression apparatus(Afrazma)) to identify specification of different combinations (c); rainfall simulations to measure the runoff and soil loss (d) and use of mobile pump for field experiments (e).
118
M. Nasiri et al. / Catena 150 (2017) 116–123
by public transport on holidays due to presence of tourism areas such as Gazoo waterfall and natural forests. This traffic can increase the sediment delivery from unpaved roads. Our observation showed that large volumes of runoff moved downslope toward the earth dam after a rain storm in September 2015. Therefore, due to the presence of earth dam and soil erosion issues, one of the most important solutions to reduce sediment delivery from forest roads as well as strengthening the road materials is the use of stabilizers. 2.2. Identification of soil, lime and RHA About 350 kg of soil were collected from the study area. Soils were put in 50 kg bags and transferred to the soil mechanics laboratory. In the laboratory, soils were remixed well together. In order to identify the soil, laboratory tests such as sieve analysis and Aterberg tests were done. According to the AASHTO classification and unified system, the studied soil is classified as A-6 and SC (Silty sand with gravel), respectively. The materials used for the access roads are usually placed in classes of A-4, A-5, A-6 and A-7 according to AASHTO classification. These soils generally contain silt and clay and are classified in poor to average classes in terms of road construction. In order to stabilize the sub-base materials, at first rice husk and hydrated lime were prepared from farm fields of northern Iran and the Alborz industrial factory, respectively. The type of used lime is slaked lime and some characteristics of that are shown in Table 1. To make rice husk ash, rice husk must be burned at high temperatures (about 500 °C). After burning, the obtained RHA were placed in the open air to complete the ash production process (Fig. 1,b). Some chemical and physical characteristics of rice husk ash are shown in Tables 2 and 3. 2.3. Mix design and laboratory studies By examining different studies it can be understood that one of the most important factors of mix design is soil properties. The soil type has an important effect on pozzolanic activity. Fine-grained soils with clayey mineral are suitable in order to stabilize soils with rice husk ash. The clay minerals have the property of absorbing certain anions and cations and retaining them in an exchangeable state. The exchangeable ions are held around the outside of the silica-alumina clay mineral structural unit. Therefore, materials mixed well together (Table 4) according to the soil properties, performance of lime and RHA in improving materials quality (Alhassan, 2008; Chobbasti et al., 2010; Muntohar, 2002). Soil laboratory tests were done to evaluate the effect of stabilizers after mixing materials. Then, different combinations of materials were studied with 5 replications using tests of CBR or California Bearing Ratio (%), maximum dry density or MDD (Mg m−3) and optimum moisture content or OMC (%), unconfined compressive strength or UCS (KN m−2) and plasticity index or PI (%). The California Bearing Ratio test, is an empirical test, which is used as an important criterion in pavement design. With this test, the bearing value of road sub-base and subgrade, can be estimated. According to the purpose of the study, this test can examine the strength of stabilized samples compared to the natural soil samples (control). This test was conducted (ASTM D1883-07) using the load-measuring device which is connected to the compression machine with loading rate of 1 mm/min. The proctor compaction tests (MDD and OMC) were performed with the standard of ASTM D069807E01 in CBR mould; by conducting this test, it can be found that how much free space of aggregates is filled by stabilizers and how
Table 1 Characteristics of the used lime.
Table 2 Chemical characteristics of the rice husk ash (*10−6). P (%)
Mg2+ (%)
Ca2+ (%)
K (%)
Na (%)
SiO2 (%)
24.7
66.1
0.016
0.124
0.114
83.7
affect the optimum water content. One of the most useful methods of evaluating the strength of stabilized soil is UCS test (Fig. 1, c). The required amount of additive to be used in stabilization of soil could be determined by this test. In this study, the test was performed for samples with 28 days curing time to consider the time factor (Alhassan, 2008). This test was performed with standard of ASTM D 2166. The compression device (a hydraulic-actuated loading piston with the capability of infinite rates of strain and stress loads) was used to measure the UCS rates. The Aterberg limit (Plastic Limit (PL), liquid limit (LL) and plasticity index (PI)) test was performed by standard of ASTM D4318-05. The liquid limit and Plastic Limit were measured using a Casagraunde cup and rolling method, respectively. The Plastic Limit is often used together with the liquid limit to determine the plasticity index. Plastic limit of fine-aggregates in sandy materials have a major influence on the strength of used materials. Materials with a PI more than allowed values should not be used on base and sub-base layers, because materials with higher PI have lower shear strength. 2.4. Rainfall simulation Rainfall simulations were carried out to measure runoff and soil loss from the different combinations of materials. In order to simulate rainfall, a squared plot of 60 × 60 cm2 that is delineated by a steel frame was fabricated (Fig. 1, d). The frame was carefully tapped into the soil to prevent leakage and the simulator was covered with a plastic cover to eliminate the wind effect. A minimum slope of 3% (the minimum longitudinal slope of forest roads) was considered for soil surface in each plot and longitudinal gradient was determined using a Suunto clinometer. A portable single nozzle rainfall simulator (Schlick r18650), which is connected through a rubber pipe to an automatic pump (Fig. 1, e) was used to simulate rainfall. According to weather stations of the Alborz dam and Gharakheil, the mean rainfall intensity for the experiments was selected 52 mm h−1. For each mixture the duration and replication of the rainfall simulations were considered 30 min and 5 replication, respectively. Time to runoff, runoff coefficient, runoff rate, sediment concentration and soil loss rate were determined for each plot. Usually, a steady-state runoff rate was reached after 15–20 min. So, Runoff coefficients were calculated as the runoff rate divided by the rainfall rate for each sample. Runoff samples were collected every 5 min in order to determine the sediment concentration in runoff and to calculate the soil loss rates for each plot. Each sample was filtered using Whatman filter papers and then the filter oven dried at 105 °C for 24 to 48 h and weighed. Then, sediment mass per volume of runoff was determined for each sample. 2.5. Data analysis The effects of different combinations of materials on runoff and soil loss rates were evaluated by statistical analysis of the data. All computations were done using SPSS version 16. This included assumptions of normality test using the Shapiro–Wilk, an ANOVA test to compare means among treatments and the Pearson test to identify correlations
Table 3 Physical characteristics of the rice (Tarom) husk ash.
Non hydrated CaO (%)
Non hydrated MgO (%)
pH
CO2 (%)
Ca(OH)2 (%)
Density (kg m−3)
Matter size/μm
Special surface (m2 g−1)
1.62
0.8
11.2
0
92.8
2200–2300
4.8–8.9
60–130
M. Nasiri et al. / Catena 150 (2017) 116–123
and Mol, 2011. He stated that addition of rice husk ash reduced the plasticity of soil and improved the desired engineering characteristics of soil such as shear strength, unconfined compressive strength and load bearing capacity. The addition of 9% RHA was able to reduce soil MDD to 1.23 Mg m3. Reduction in MDD can be related to the replacement of soil by RHA in mixtures which have relatively lower specific gravity compared to natural soil (Osinubi and Katte, 1997).
Table 4 Investigated mixtures and mentioned indices. Mix design
Index
Natural soil Soil + 6% lime Soil + 4% lime + 7% RHAa Soil + 6% lime + 9% RHA Soil + 6% lime + 7% RHA Soil + 6% lime + 9% RHA
S S S S S S
a
+ + + + +
119
6L 4L + 7R 4L + 9R 6L + 7R 6L + 9R
Rice husk ash.
3.2. Hydrological and erosive response between runoff, sediment and soil mechanical properties of materials. Also, the graphical displays were made using the Excel and SigmaPlot softwares.
3. Results and discussion 3.1. Characteristics of materials Mechanical properties of natural soil and different combinations of materials are shown in Table 5. After the ANOVA, some significant differences between the different combinations of materials were detected. The increase in OMC, UCS and CBR was a result of additive materials. Lime and RHA can reduce the quantity of free silt and clay and increase the coarser materials with larger surface areas. The RHA content has an important effect on OMC values. According to Table 5, it can be found that the soil OMC for combinations of S + 4L + 9R and S + 6L + 9R had increased 19 and 21.9%, respectively compared to the natural soil samples. The major chemical compounds present in Rice Husk ash contribute active role in chemical reactions with constituents of soil is Silica Si-ions. Calcium is another additive to form binding compounds along with soil. Lime is added at different doses to supplement Ca to form binding compounds along with Silica. When lime is mixed with water, it forms calcium hydroxide, called slaked lime. This chemical reaction increased the OMC value and requires more water to supply more Ca ions for the cation exchange reaction. This result is in agreement with Ramezanianpour et al., 2009; Chobbasti et al., 2010 and Purbi et al., 2011. The UCS value did not have significant changes with increasing lime content (4% lime or 6% lime). The increase in UCS is attributed to the formation of cementations compounds between the CaOH present in the soil and RHA as well as the pozzolans present in the RHA (Jamil et al., 2013). According to the laboratory results the maximum and minimum values of plasticity index were recorded for the natural soil and combination of S + 6L + 9R respectively. Table 5 shows the lime can be much more effective on reduction of the plasticity index if mixed with RHA. The reason for reduction in plasticity index could be the chemical reaction between limes and soil that require more water (Jamil et al., 2013; Ramezanianpour et al., 2009). This result is in agreement with Hossain
Hydrological response of natural soil and other combinations are shown in Table 6. The ANOVA analysis showed significant differences in time to runoff data between the different combinations of materials (p b 0.0001). The mean time to runoff for the natural soils (without additive materials) and combination of S + 6L were recorded 87 s (57–104 s) and 63 s (35–95 s) respectively. In other studies, time to runoff was recorded 25–89 (Jordán and Martínez-Zavala, 2008) and 23–60 (Jordán et al., 2009). According to result the mean time to runoff was longer for the combinations of S + 4L + 9R (198 s) and S + 6L + 9R (173 s). The ANOVA also showed significant differences for runoff coefficients between the materials stabilized with RHA and other combinations. Materials stabilized with RHA can increase the infiltration rates due to large particles with larger voids. Several researchers studied the runoff coefficients on unpaved forest roads and gravel surface roads. The rate of runoff coefficient was measured 31% on gravel surface roads under a rainfall intensity of 100 mm·h−1 (Selkirk and Riley, 1996). Ziegler et al. (2001) in a rainfall simulation study (100 mm·h−1 rainfall intensity) reported that the rate of runoff coefficients on unpaved forest roads was 84%. Also, in similar studies the rate of runoff coefficients were measured 50.7% under a rainfall intensity of 75 mm·h− 1 (Arnaez et al., 2004), 46.6% under a rainfall intensity of 72 mm h−1 (Jordán and Martínez-Zavala, 2008) and 58.6% under a rainfall intensity of 90 mm h−1 (Jordán et al., 2009) on unpaved forest roads. Table 6 shows sub-base soils stabilized with RHA have a response similar (35.9% on the combination of S + 4L + 9R) to gravel surface roads (Selkirk and Riley, 1996) and soils without additive materials have a response similar (53.66% on the natural sub-base materials) to unpaved roads (Arnaez et al., 2004; Jordán et al., 2009). Fig. 2 shows the highest rate of runoff was recorded for natural soils (combinations without additive materials) at the beginning of experiments (5–15 min). Runoff rate increased slowly and in materials stabilized with RHA, a steady-state runoff was reached after 15–20 min. Addition of pure lime to the natural soil can decrease the swelling and soil contraction as well as increase the optimum water content and soil resistance including compressive and tensile strengths (Alhassan, 2008). Absorption rate of materials stabilized with pure lime was more than the natural soils at the beginning of experiments because of more optimum water content. After that the runoff rate increased quickly during the experiments (8–20 min) due to decrease in plasticity
Table 5 Soil mechanical properties of different combinations of materials. Parameter PI⁎ (%) OMC (%) MDD (Mg m−3) UCS-28d* (KN m−3) CBR (%)
M⁎ SD⁎ M SD M SD M SD M SD
Soil
S + 6L
S + 4L + 7R
S + 4L + 9R
S + 6L + 7R
S + 6L + 9R
ANOVA, p
10.82d 0.65 16.3a 1.65 1.54c 0.005 213a 3.4 9.24a 0.97
4b 0.16 24.4ab 1.74 1.42bc 0.06 225ab 4.65 36.3b 2.6
4.24b 0.37 35.02c 3.7 1.37abc 0.08 232b 3.96 46.6b 2.1
2.36a⁎ 0.1 36.2c 0.84 1.23a 0.02 234b 3.36 43.8b 3.6
5.66c 0.49 32.3bc 3.7 1.37abc 0.03 231b 3.8 42.2b 4.1
1.13a 0.3 39.4c 2.54 1.24ab 0.06 235b 3.8 47b 4.2
0.000
⁎Within-a-row means followed by the same letter are not significantly different; M⁎: mean; SD⁎: standard deviation; PI⁎: plasticity index; 28d⁎: 28 days curing time.
0.000 0.000 0.000 0.000
120
M. Nasiri et al. / Catena 150 (2017) 116–123
Table 6 Average time to runoff, runoff coefficients, and total soil loss according to the combinations of materials. Parameter Time to runoff (s)
M SD M SD M SD
Runoff coefficient (%) Total soil loss (g m
−2
)
Soil
S + 6L
S + 4L + 7R
S + 4L + 9R
S + 6L + 7R
S + 6L + 9R
ANOVA, p
86.7a 8.6 53.66b 1.88 212.2b 2.27
63.4a⁎ 12.3 58.7c 1.43 171.4ab 7.6
166bc 14.3 38. 5a 1.25 176.4ab 6.7
198c 12.4 35.9a 0.71 173.1ab 5.6
144.1b 11.4 37.1a 0.34 175.4ab 2.33
172.7bc 10.4 36.6a 0.68 162.4a 5.9
0.0000 0.0000 0.0000
⁎Within-a-row means followed by the same letter are not significantly different; M⁎: mean; SD⁎: standard deviation.
index and porosity of materials. After addition of RHA and lime, fine particles of soil stick to each other which result in large particles with larger voids (Chobbasti et al., 2010). Gradual increase in runoff rates on materials stabilized with RHA can be related to these voids. In this study higher quantity of RHA resulted in lower runoff rates because it takes more time for the water to infiltrate into the soil voids. The minimum and maximum rates of soil loss were recorded on the combination of S + 6L + 9R (162.4 g·m2) and natural soil (212.2 g·m− 2) respectively (Table 6). The materials stabilized with RHA and lime had the same behavior in sediment generation. So that, the ANOVA analysis showed significant differences in total soil loss between the natural soils and materials stabilized with RHA and lime. The RHA and lime can increase the strength properties of materials. Reduction in soil loss can be related to the soil strength properties which have higher CBR rate and lower MDD and plasticity indices compared to natural soil. Fig. 3 shows sediment concentration in runoff from different combinations of sub-base materials. Sediment concentration increased during the first 5–10 min of rainfall on the combinations of S + 4L + 7R, S + 4L + 9R, S + 6L + 7R and S + 6L + 9R. However, the rates of sediment concentration decreased after 10 min from the beginning of the rainfall simulation. Fine particles among the soils were washed out at
the beginning of the test. Therefore, the rate of sediment concentration increased at the beginning of the rainfall simulation and then decreased due to the loss of fine surface soil particles (Reid and Dunne, 1984; Arnaez et al., 2004). The rate of sediment concentration was greater for the natural soils (12.52 g L−l) than materials stabilized with RHA and lime. Several researchers studied the sediment concentration in runoff during the rainfall simulation. The sediment concentration was measured 1.8 g L−l for gravel surface road under a rainfall intensity of 100 mm·h−1 (Selkirk and Riley, 1996). This rates were reported on unpaved roads 4.4 g L−l under a rainfall intensity of 72 mm h−1 (Jordán and Martínez-Zavala, 2008), 12.2 g L-l under a rainfall intensity of 33 mm h−1 (Lotfalian et al., 2013). In this study the erosive response of subbase materials stabilized with RHA and lime was similar to gravel surface roads (Selkirk and Riley, 1996) and materials without additive have a response similar to unpaved roads (Lotfalian et al., 2013; Jordán et al., 2009). Fig. 4 shows the relationship between soil loss rates and runoff coefficients for different combinations of materials. The lowest values of steady-state runoff and soil loss rates were measured on the materials stabilized with RHA and lime. These materials (63.3%) have same behavior under rainfall simulation tests. The highest runoff coefficients and soil losses were recorded on the materials stabilized with pure 30
45
25
35
Runoff rate (mm h -1)
Runoff rate (mm h -1)
40
30 25 20 15 10 5
Soil S+6L S+4L+7R
0 0
5
10
15
20
25
30
20 15 10 5 Soil S+4L+9R
0
35
0
5
10
30
30
25
25
20 15 10 5 0
Soil S+6L+7R 0
5
10
15
15
20
25
30
35
Time (min)
Runoff rate (mm h -1)
Runoff rate (mm h -1)
Time (min)
20
Time (min)
25
30
20 15 10 5 Soil S+6L+9R
0
35
0
5
10
15
20
25
Time (min)
Fig. 2. Behavior of surface runoff from rainfall simulations on different combinations of sub-base materials.
30
35
20
Sediment Concentration (gr L-1)
Sediment Concentration (gr L-1)
M. Nasiri et al. / Catena 150 (2017) 116–123
Soil S+6L S+4L+7R
18 16 14 12 10 8 6 4 2 0 0
5
10
15
20
25
30
121
20 Soil S+4L+9R
18 16 14 12 10 8 6 4 2 0
0
35
5
20 Soil S+6L+7R
18 16 14 12 10 8 6 4 2 0 0
5
10
15
20
10
15
20
25
30
35
Time (min)
Sediment Concentration (gr L-1)
-1
Sediment Concentration (gr L )
Time (min)
25
30
35
Time (min)
20 Soil Soil+6L+9R
18 16 14 12 10 8 6 4 2 0 0
5
10
15
20
25
30
35
Time (min)
Fig. 3. Sediment concentration in runoff from different combination of materials.
lime and natural soils, respectively. In this study it appears that the mixture of S + 6L + 9R was the best combination to reduce runoff and soil loss rates. Road engineers make use of optimum moisture content when they apply water to soil to compact. Excessively compacted soil results in problems such as reduced rainfall infiltration. When soil voids are filled with pure lime and RHA, compaction naturally occurs in materials due to pozzolanic reaction between materials. The compaction of the materials stabilized with lime and RHA can partly explain the lower soil loss rates. Our observation during the study showed that compaction can facilitate runoff generation on materials stabilized with pure lime due to less voids space than when using RHA.
65
Runoff coefficients (%)
60 55 50 45
Soil S+6L S+4L+7R S+4L+9R S+6L+7R S+6L+9R
40 35 30 140
150
160
170
180
190
200
210
220
Soil loss rate (gr m -2) Fig. 4. Relationship between runoff coefficients and soil loss rates for different combinations of materials.
3.3. Correlation between runoff, soil loss and materials Table 7 shows the R-Pearson coefficients between runoff rate, soil loss rate and soil mechanical parameters of different combinations of materials. The MDD and OMC indices show how much free space of aggregates is filled by stabilizers and how that affects the optimum water content. According to the results, soil compaction characteristics (the OMC and MDD) have an important effect on runoff rates due to changes in soil voids, swelling and PI. Some significant and negative correlations were found between runoff rates, the PI and MDD indices of materials stabilized with RHA and lime, but not significant for natural soil. In addition, some positive significant correlations were found between runoff rate and OMC index of materials stabilized with RHA. We found some significant and negative correlations between soil loss rates and the CBR as well as some significant and positive correlations between soil loss rates and the MDD of materials stabilized with RHA and lime (Table 7). Also, no significant correlation was detected between soil loss rates and the OMC of materials. The CBR value is used as an index of compacted soil strength and bearing capacity. Stabilized soil with lime and RHA, increased the CBR values and decreased the MDD values. Lime and RHA can reduce fine particles and increase the cohesion among aggregates (Harichane et al., 2011). Therefore, in this study addition of lime and RHA to the soils increased the strength of materials and consequently the rate of soil loss decreased. Table 7 shows some significant and positive correlations between PI and the MDD as well as some significant and negative correlations between OMC and the PI and MDD. We found some significant and negative correlations between CBR and the PI and MMD of materials stabilized with RHA and lime. The CBR value of soil may depends on many factors such as MDD, OMC, liquid limit, Plastic Limit, PI, type of soil and permeability of soil. The CBR value decreases with the increase in the plasticity index and optimum moisture content of soil but increases with the increase in the maximum dry density. Talukda (2014) stated that there
122
M. Nasiri et al. / Catena 150 (2017) 116–123
Table 7 R-Pearson correlation between runoff rates, soil loss rate and soil mechanical parameters of materials. Variables Runoff rate Soil S + 6L S + 4 L + 7R S + 4 L + 9R S + 6 L + 7R S + 6 L + 9R Soil loss rate Soil S + 6L S + 4 L + 7R S + 4 L + 9R S + 6 L + 7R S + 6 L + 9R Soil properties MDD OMC UCS-28 day CBR
Soil loss rate R
PI R
MDD R
OMC R
UCS-28 day R
CBR R
0.86* 0.48 0.78* 0.69 0.7 0.65
−0.68 −0.79* −0.52 −0.65 −0.75* −0.53
−0.63 −0.73* −0.8* −0.88* −0.76* −0.87*
0.54 0.35 0.85* 0.86* 0.73* 0.63
0.53 0.65 0.37 0.42 −0.01 0.55
0.69 0.71 0.29 0.32 0.52 0.66
0.35 0.19 −0.28 0.29 0.68 0.23
0.39 0.77* 0.53 0.59 0.87* 0.82*
0.37 −0.29 −0.41 −0.56 0.35 −0.02
0.03 −0.83* −0.56 −0.7 −0.61 −0.34
−0.64 −0.87* −0.49 −0.76* −0.91* −0.54
0.82* −0.79* −0.61 −0.88*
−0.89* 0.58 0.83*
0.44 −0.47
0.7
⁎ Marked R-Pearson coefficients are significant (b0.05).
is some significant correlation between the CBR of fine grained soil (ML and MI) with PI, MDD and OMC values.
4. Conclusion 1. Additive materials have an important effect on the runoff and soil loss rates. The highest soil loss rate was recorded for the control (natural soil). Given the rice husk is an agricultural waste material and it is available in many countries at little or no cost ($1 U.S. per kg in Iran), the produced ash from rice husk can be used to improve road surface quality and decrease the runoff and soil loss rates. 2. The combination of S + 6L + 9R was selected as the best combination of materials to reduce runoff and soil loss rates. Materials stabilized with RHA and lime decreased the runoff and soil loss rates due to large particles with larger voids, fewer fine aggregates and more strength among soil particles. 3. The highest runoff coefficient was recorded on the materials stabilized with pure lime. These results show that the pure lime is not suitable for stabilization of unpaved forest roads due to high runoff coefficients. 4. Some significant and positive or negative correlations were found between the CBR, dry density, optimum moisture content, plasticity index and the rates of runoff and soil loss. These correlations show the relationship between soil mechanical properties and hydrological and erosive response of materials which are associated with changes in infiltration rates, moisture content, swelling, porosity, shear strength, cohesion and internal friction angle of soils. On the basis of the results, we concluded that the addition of rice husk ash not only increases the materials quality in soil stabilization, but also reduces the runoff and soil loss rates.
Acknowledgements This research was supported by the University of Sari Agricultural Sciences and Natural Resources. We would like to thank Mr. Aghajantabar and Mr. Bozorgpour for their suggestions about pozzolanic materials, field experiments and laboratory tests. We also would like to thank volunteers (Mr. Valizadeh, Mr. Aghajani and Mr. Ahmadi) who devoted their time and knowledge during the field experiments.
References Alhassan, M., 2008. Permeability of lateritic soil treated with lime and rice husk ash. Assumption Univ. J. Thailand 12 (2), 115–120. Arnaez, J., Larrea, V., Ortigosa, L., 2004. Surface runoff and soil erosion on unpaved forest roads from rainfall simulation tests in northeastern Spain. Catena 57, 1–14. Basha, E.A., Hashim, R., Mahmud, H.B., Muntohar, A.S., 2005. Stabilization of residual soil with rice husk ash and cement. Constr. Build. Mater. 19 (6), 448–453. Cao, L., Zhang, K., Zhang, W., 2009. Detachment of road surface soil by flowing water. Catena 76, 155–162. Chobbasti, A.J., Ghodrati, H., Vahdatirad, M.J., Firouzian, S., Barari, A., Torabi, M., Bagherian, A., 2010. Influence of using rice husk ash in soil stabilization methodwith lime. Front. Earth Sci. China 4 (4), 471–480. Coker, R.J., Fahey, B.D., Payne, J.J., 1993. Fine sediment production from truck traffic, Queen Charlotte Forest, Marlborough Sounds. New Zealand. J. Hydrol. 31, 56–64. Croke, J.C., Hairsine, P., Fogarty, P., 1999. Runoff generation and redistribution on disturbed forest hill slopes, south eastern Australia. J. Hydrol. 216, 55–77. Demir, M., Makineci, E., Kartaloglu, M., 2012. Temporal sediment production of paved and unpaved forest road. Fresenius Environ. Bull. 21, 1180–1185. Harichane, K., Ghrici, M., Kenai, S., 2011. Effect of curing time on shear strength of cohesive soils stabilized with combination of lime and natural pozzolana. Int. J. Civil Eng. 9, 90–96. Hossain, K.M.A., Mol, L., 2011. Some engineering properties of stabilized clayey soils incorporating natural pozzolans and industrial wastes. Constr. Build. Mater. 25, 3495–3501. Jamil, M., Kaish, A.B.M.A., Raman, S.N., Zain, M.F.M., 2013. Pozzolanic contribution of rice husk ash in cementitious system. Constr. Build. Mater. 47, 588–593. Jordán, A., Martínez-Zavala, L., 2008. Soil loss and runoff rates on unpaved forest roads in southern Spain after simulated rainfall. For. Ecol. Manag. 255, 913–919. Jordán, A., Martínez-Zavala, L., Bellinfante, N., 2009. Impact of different parts of unpaved forest roads on runoff and sediment yield in a Mediterranean area. Sci. Total Environ. 407, 937–944. Lotfalian, M., Parsakhoo, A., Kavian, A., Hosseini, S.A., 2013. Runoff and sediment concentration of different parts of a road in Hyrcanian forests. Forest Sci. Pract. 15 (2), 144–151. Luce, C.H., Black, T.A., 1999. Sediment production from forest roads in western Oregon. Water Resour. Res. 35, 2561–2570. Muntohar, A.S., 2002. Utilization of uncontrolled burnt rice husk ash in soil improvement. Dimensi Teknik Sipil J. 4, 100–105. Nair, D.G., Fraaij, A., Klaassen, A.K., Kentgens, P.M., 2008. A structural investigation relating to the pozzolanic activity of rice husk ashes. J. Cem. Concr. Res. 38 (6), 861–869. Onyango, M., Macha, I., Busch, C., 2007. Use of Naturally Occurring Pozzolans for Road Construction in Tanzania. 9th International Conference on Low-volume Roads. June 24–27, Austin, Texas . Osinubi, K.J., Katte, V.Y., 1997. Effect of elapsed time after mixing on grain size and plasticity characteristics, soil lime mixes. Nigeria Soc. Eng. Tech. Trans. 32 (4), 65–76. Purbi, S., Mukesh, N., Dixit, M., 2011. Evaluation of strength characteristics of clayey soil by adding soil stabilizing additives. Int. J. Earth Sci. Eng. 4, 1060–1064. Ramezanianpour, A.A., Khani, M.M., Ahmadibeni, G., 2009. The effect of rice husk ash on mechanical properties and durability of sustainable concretes. Int. J. Civil Eng. 7 (2), 83–91. Ramos-Scharron, C., MacDonald, L.H., 2005. Measurement and prediction of sediment production from unpaved roads, St. John, U.S. Virgin Islands. Earth Surf. Process. Landf. 30, 1283–1304. Reid, L.M., Dunne, T., 1984. Sediment production from forest road surfaces. Water Resour. Res. 20, 1753–1761.
M. Nasiri et al. / Catena 150 (2017) 116–123 Rimal, B.K., Lal, R., 2009. Soil and carbon losses from five different land management areas under simulated rainfall. Soil Tillage Res. 106, 62–70. Safari, A., Kavian, A., Kavian, A., Parsakhoo, A., Saleh, I., Jordán, A., 2016. Impact of different parts of skid trails on runoff and soil erosion in the Hyrcanian forest (northern Iran). Geoderma 263, 161–167. Selkirk, J.M., Riley, S.J., 1996. Erodibility of road batters under simulated rainfall. Hydrol. Sci. J. 41, 363–376. Sheridan, G.J., Noske, P.J., 2007. A quantitative study of sediment delivery and stream pollution from different forest road types. Hydrol. Process. 21 (3), 387–398. Talukda, K., 2014. A study of correlation between California Bearing Ratio (CBR) value with other properties of soil. Int. J. Emerg. Technol. Adv. Eng. 2, 559–562.
123
Trivedi, J.S., Sandeep, N., Chakradhar, I., 2013. Optimum utilization of fly ash for stabilization of sub-grade soil using genetic algorithm. Procedia Eng. 51, 250–258. Weiting, X.u., Tommy, Y.L., Shazim, A.M., 2012. Microstructure and reactivity of rich husk ash. Constr. Build. Mater. 29, 541–547. Ziegler, A.D., Sutherland, R.A., Giambelluca, T.W., 2000. Runoff generation and sediment production on unpaved roads, footpaths and agricultural land surfaces in northern Thailand. Earth Surf. Process. Landf. 25, 519–534. Ziegler, A.D., Sutherland, R.A., Giambelluca, T.W., 2001. Interterm surface preparation and sediment detachment by vehicle traffic on unpaved mountain roads. Earth Surf. Process. Landf. 26, 235–250.