| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Institute of Soil, Water and Environmental Sciences, the Volcani Centre, Agricultural Research Organization, P.O. Box 6, Bet Dagan, 50250, Israel
* Corresponding author (marcos{at}volcani.agri.gov.il)
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: AMC, antecedent moisture content • FC, field capacity • IR, infiltration rate • WR, wetting rate
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Raindrop kinetic energy effects on seal formation, IR, and soil loss have been studied extensively (e.g., Mohammed and Kohl, 1987; Agassi et al., 1994). Agassi et al. (1985) showed that when soil was exposed to drops with kinetic energy <0.01 J mm–1 m–2, a seal was not formed. However, when the kinetic energy was 23 J mm–1 m–2, a seal with a very low hydraulic conductivity was formed. Al-Durrah and Bradford (1982a)( 1982b) studying several soils and falling heights and three sizes of drops, concluded that splash erosion was correlated with raindrop kinetic energy and inversely proportional to soil shear strength. Shainberg et al. (2003) suggested that the raindrop impact energy plays a dominant role in seal formation only in sandy loams to silt clays (soils with <300 g kg–1 clay). In clay soils with stable structure fast wetting was essential for seal formation and raindrop impact played a secondary role. The physicochemical mechanism in seal formation is controlled mainly by cations concentration and composition in the soil and applied water (Agassi et al., 1981; Kazman et al., 1983). The role of this mechanism in seal formation has been studied in detail and can be predicted satisfactorily from basic mineralogical and chemical properties of the soil (Shainberg and Letey, 1984; Stern et al., 1991).
Aggregate disintegration caused by fast wetting (slaking) is well documented (Le Bissonnais and Arrouays, 1997; Abu-Sharar et al., 1987; Loch, 1994; Lado et al., 2004). Slaking occurs when the aggregate is exposed to stresses produced by differential swelling, entrapped air explosion, rapid release of heat during wetting, and the mechanical action of moving water (Emerson, 1977; Kay and Angers, 1999). Slaking is determined by the WR of the soil; the faster the WR the higher the slaking forces (Le Bissonnais and Arrouays, 1997). Aggregate slaking is essential to seal formation in soils with high clay content and high aggregate stability (Loch, 1994; Shainberg et al., 2003).
The effect of soil WR on aggregate stability, seal properties, and soil erosion depends on the AMC of the soil. Fast wetting of a moist soil causes less slaking compared with fast wetting of a dry soil. Also, the method by which soil AMC was maintained is important. Fast wetting of a dry soil to a given AMC may disintegrate the aggregates before exposing the soil to a rainstorm. Consequently, studies on the effect of WR and AMC on seal formation lead to contradicting results (e.g., Wischmeier and Mannering, 1969; Luk, 1985; Truman and Bradford, 1990; Truman et al., 1990; Le Bissonnais and Singer, 1992; Reichert and Norton 1994). For example, Luk (1985) reported that increasing AMC from near wilting point to saturation produced a five-fold increase in soil loss in two silt loam soils. Reichert and Norton (1994) reported that wetting the soil for 2 h at –0.5 kPa tensions before the rain, generally increased or had little or no effect on runoff and soil loss in most soils except for a stable Oxisol for which wetting decreased soil loss. Truman and Bradford (1990) found that for four out of five soils studied, prewetting the soil for 48 h reduced significantly splash and wash erosion compared with air-dried soil. Le Bissonnais and Singer (1992) analyzed runoff and soil loss generated in two soils with 298 and 417 g kg–1 clay under two initial conditions: air-dry and 24 h prewetted by capillary action, and found that increasing AMC reduced runoff and soil loss.
Recently, Levy et al. (1997) prewetted two soils, a silt loam and a sandy clay, with 5 mm of deionized water and WR of 1 and 30 mm h–1. Thereafter, the soils were exposed to rainfall of approximately 32 mm h–1. A decrease in WR from 30 to 1 mm h–1 decreased the total runoff and soil loss in both soils (Levy et al., 1997). Mamedov et al. (2001)(2002) showed that when prewetting the soil until saturation, the effect of WR (2, 8, and 64 mm h–1) on runoff and interrill erosion depends on clay content of the soil. The effect of WR was pronounced in clay soils, negligible in sandy loam and silt loam, and moderate in soils with intermediate clay content. Similarly, the effect of AMC on seal formation and soil loss could be influenced by WR, and the contradicting results found by many authors could be due to the use of soils with different properties (e.g., clay content) and WRs before the application of rain.
Clay can act as a cementing material that holds particles together in the aggregate (Emerson, 1977). Therefore, increasing the clay content in the soil increases aggregate stability (Kemper and Koch, 1966; Boix-Fayos et al., 2001). Ben-Hur et al. (1985), studying the effect of clay content on the susceptibility of soils to seal formation and soil loss, found that in soils with low clay content (<100 g kg–1), IR was high and runoff was low because the amount of clay in the soil was not enough to clog the pores and form a seal. In soils with clay content >300 to 400 g kg–1 the rate of seal formation was low, because the clay acted as a cementing material that increased aggregate stability and diminished seal formation (Ben-Hur et al., 1985). Consequently, high IR and low runoff were observed in clay soils. Soils with intermediate clay content (200 g kg–1) were the most susceptible to seal formation because the amount of clay was too low to stabilize the aggregates, but was enough to clog the pores at the soil surface (Ben-Hur et al., 1985).
An increase in clay content in the soil could also increase slaking forces during soil wetting. Under fast wetting, an increase in clay content in the aggregate also increase the extent of differential swelling and the volume of entrapped air that, in turn, increase aggregate slaking. Therefore, an increase in clay content in the soil might have two opposing effects on seal formation: (i) an increase in aggregate stability and diminishing seal formation and (ii) an increase in aggregate slaking, on wetting, and an increase in soil susceptibility to sealing. Thus, the objectives of this study were: (i) to determine the effects of clay content on aggregate strength and on aggregate slaking; (ii) to study the effects of WR, AMC, and clay content on seal formation, IR, and interrill erosion.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
Aggregate Stability Study
The aggregate stability of the six soils was determined by two methods: (i) Fast-wetting in water as proposed by Le Bissonnais (1996) and (ii) Slow wetting (1 mm h–1) by mist.
(i) Fast Wetting.
Samples of aggregates 3 to 5 mm in size were put in the oven at 40°C for 24 h, and 5 g of the oven-dry aggregates were immersed in a beaker containing 50 cm3 of deionized water (fast wetting). Following saturation for 10 min the water was sucked off with a pipette. The soil material was transferred to a 50-µm sieve that had previously been immersed in ethanol and gently moved up and down in ethanol five times to separate <50-µm fragments from >50-µm ones. The >50-µm fraction was oven dried and then gently dry sieved by hand on a column of sieves of mesh sizes 2, 1, 0.5, 0.25, 0.1, and 0.05 mm. The weight of each fraction was measured; and that of the <50-µm fraction was calculated as the difference between the initial weight and the sum of the weights of the other six fractions.
(ii) Slow Wetting.
Samples of aggregates 3 to 5 mm in size were wetted with 1 mm h–1 mist to saturation. A detailed description of the wetting methods is presented in the rain simulation section. Following wetting, the aggregates were transferred to a 50-µm sieve immersed in ethanol, and the same sieving procedure as described in the fast wetting test was used.
The aggregate stability for each soil sample in the two methods was expressed by calculating the mean weight diameter (MWD) of the seven classes:
 | [1] |
i. The higher the MWD was the higher the aggregate stability (Le Bissonnais, 1996).
Rainfall Simulator Study
This study was conducted with three cultivated soils: A loam with 230 g kg–1 clay from Nevatim, a clay soil with 410 g kg–1 clay from Hafetz Haim, and clay soil with 620 g kg–1 clay from Eilon (Table 1). The reported values of moisture contents of the three soils at FC were 200, 343, and 480 g kg–1 by weight, respectively (Koyumdjisky and Soriano, 1988).
The IR, runoff, and soil loss from the three soils were determined by means of a rotary disc rainfall simulator (Morin et al., 1967). The typical mechanical parameters of the simulated rainfall were: rainfall intensity, 42 mm h–1; raindrop mean diameter, 1.9 mm; median drop velocity, 6.2 m s–1; and kinetic energy, 18.1 J mm–1 m–2.
The air-dry soils were crushed and sieved to obtain samples with aggregate size <4 mm. The samples were packed in perforated trays measuring 0.30 by 0.50 m and 0.02 m deep. The packed soil trays were placed on an 8-cm thick layer of coarse sand, in a box positioned under the rainfall simulator at a slope of 9%. The bulk densities in the trays of the soils with 230, 410, and 620 g kg–1 clay were 1.2, 1.15, and 1.1 Mg m–3, respectively.
The soil samples were prewetted from the top with deionized water applied as mist. With this procedure, the mechanical breakdown of the aggregates by the raindrop impact was prevented. Three levels of mist depths were applied to obtain three average levels of AMC: air-dry, 0.25, and 0.5 of the FC. To obtain these AMC levels, the depths of mist applied to the soil with 230 g kg–1 clay were 0, 1.2, and 2.4 mm, respectively; to the soil with 410 g kg–1 clay, 0, 1.95, and 3.9 mm, respectively; and to the soil with 620 g kg–1 clay, the depths of mist applied were 0, 2.7, and 5.4 mm, respectively. It is expected that the upper layer of the soil in the tray will be wetter than the lower layer in each AMC level, and that the AMC levels of 0.25 and 0.5 of the FC are only average values for the entire soil depth. Nevertheless, the water content of the upper 2 mm of the soil surface at AMC of 0.5 FC was higher than at AMC of 0.25 FC.
For each AMC, wetting of the soils with mist was done with two WRs, 1 and 5 mm h–1. The WR of 1 mm h–1 was obtained by exposing the soil trays to 6 s of 30 mm h–1 mist in every 3-min cycle. The WR of 5 mm h–1 consisted of applying 30 mm h–1 mist for 30 s in 3 min cycle. The prewetting duration varied between 0.24 h in the soil with 230 g kg–1 clay with AMC of 0.25 FC (1.2 mm mist) and WR of 5 mm h–1, and 6.6 h in the soil with 620 g kg–1 clay, with AMC of 0.5 FC (5.4 mm mist) and WR of 1 mm h–1.
After prewetting and 15 min break for moisture redistribution, the trays were exposed to a simulated rainstorm of 80 mm of deionized water. When rainwater percolating through the soil was detected at the outlet of the trays, leachate was collected and measured at regular times during the rainstorm to determine the IR. Likewise, the runoff from the entire rainstorm was collected and measured. The weight of soil that was eroded from each soil tray was measured by drying the runoff samples and weighing the dry soil. The measured soil loss could be assumed to represent interrill erosion because the trays in the rainfall simulator were short (0.5 m) (Meyer and Harmon, 1984).
Infiltration data obtained from the rainfall simulator were analyzed using Eq. [1] proposed by Morin and Benyamini (1977):
 | [2] |
is soil coefficient related to surface aggregate stability (mm–1); t is time from the beginning of the storm (h), and p is rain intensity (mm h–1).
A nonlinear regression program used measured It, If, and p values to calculate Ii and that gave the best coefficient of determination (R2 > 0.9) between paired calculated and measured It values.
Runoff volume (Roff) for any given depth of rain (N) from each single event was calculated using Eq. [3]:
 | [3] |
Statistical Analysis
All the studies were conducted in three replicates. Runoff and soil loss data were subjected to a multifactor analysis of variance (SAS Institute, 1995). In cases where interactions were noted among treatments, differences among runoff or soil loss of individual treatments were determined using a single confidence interval value at level P 
0.05 (SAS Institute, 1995).
|
RESULTS AND DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
|
As a result of the increase in aggregate stability with the increase in clay content, we should expect the final IR of the soil with 410 g kg–1 clay to be intermediate between the soils with 230 and 620 g kg–1 clay. Also, the response of the soil with 410 g kg–1 clay to WR should be intermediate between the two soils. However, the results in Fig. 2 indicate that the soil with 410 g kg–1 clay was equally susceptible to sealing and equally insensitive to WR as the soil with 230 g kg–1 clay. These discrepancies are explained by considering the effect of clay content on aggregate strength, aggregate slaking, and the response of soils to raindrop impact energy.
Aggregate stability could be affected by soil properties like organic matter, clay, and CaCO3 content or exchangeable sodium percentage (Kay and Angers, 1999). Linear regressions between MWD, as determined by slow wetting of the aggregates in water, and these soil properties were used to identify their effect of these properties on aggregate stability in these soils. The parameters of the regressions are shown in Table 2. Only the relation between MWD after slow wetting and clay content of the soil was found to be significant (Fig. 3A , Line a), which indicates that in these soils, this soil property is the main one responsible for aggregate stability. When the soil was wetted slowly, no slaking of the aggregates took place and MWD increased linearly with clay content (Fig. 3A, Line a), likely due to the cementing effect of the clay within the aggregates (Kemper and Koch, 1966; Kay and Angers, 1999).
|
|
The effect of soil properties (e.g., clay content) and soil conditions (e.g., WR and AMC) on IR and seal properties (Fig. 1 and 2) can be explained by the interaction between clay content, aggregate strength and aggregate slaking, and rain impact energy. In the soils with 230 and 410 g kg–1 clay, aggregate stability is relatively low and raindrop impact is enough to break down the aggregates and to form a seal (Betzalel et al., 1995). Thus, the differences in the final IR values between the air-dry soils exposed to 42 mm h–1 rain and soils prewetted with WR of 1 mm h–1 to AMC of 0.5 FC were small (Fig. 2). Conversely, in the soil with 620 g kg–1 clay, aggregate strength was high and aggregate breakdown by raindrop impact was not enough to disintegrate completely the aggregates and to form a seal (Fig. 1). Thus, slaking of the aggregates by fast WR before raindrop impact enhanced the formation of a seal with low final IR (Fig. 2).
The interaction between AMC (dry soil, 0.25 and 0.5 FC) and WR (1 and 5 mm h–1) in their effect on seal formation for the three soils with different clay contents is presented in Fig. 4 . The effect of AMC and WR on the final IR values of the soils with 230 and 410 g kg–1 clay were small and not significant (Fig. 4). In these soils, aggregate stability was so low that raindrop impact was enough to form a seal and the final IRs of both soils were not affected by WR and AMC (Fig. 4). Conversely, in the soil with 620 g kg–1 clay a significant interaction between AMC and WR on seal formation and final IR values was observed (Fig. 4). In this soil, increasing the WR from 1 to 5 mm h–1 decreased the final IR values for the high AMC level, and had no significant effect in the low AMC level (Fig. 4). In the soil with 620 g kg–1 clay and 0.5 FC AMC level, 10 mm of water were applied. This depth of water applied at WR of 5 mm h–1 was enough to disintegrate the aggregates and the final IR was similar to that of dry soil exposed to simulated rain (Fig. 4). When the soil with 620 g kg–1 clay was wetted to AMC level of 0.25 FC (application of only 5-mm water), there was not enough water to disintegrate the aggregates in either WR treatment; therefore the final IR of the clay soil was not affected by WR (Fig. 4). However, this wetting was enough to decrease slaking by high intensity rain, causing the final IR to be significantly higher than that of dry soil exposed to high intensity rain. Similarly, the effect of AMC depended on WR. At WR of 1 mm h–1, increasing the AMC from 0.25 to 0.5 FC increased significantly the final IR (Fig. 4). Slowly increasing the moisture content of the soil before the rain prevented the aggregates from slaking by the rapid wetting (42 mm h–1) during the rainstorm. Consequently, at AMC of 0.5 FC, the soil surface became less susceptible to seal formation and IR reduction than at AMC of 0.25 FC. In contrast, for the fast WR (5 mm h–1), increasing the AMC from 0.25 to 0.5 FC decreased significantly the final IR (Fig. 4). In this case, probably, fast wetting (5 mm h–1) of the soil up to 0.5 FC before the simulated rain caused more aggregate slaking than wetting up to 0.25 FC. Consequently, the total aggregate disintegration during the prewetting and the rain storm in the treatment of AMC of 0.5 FC and WR of 5 mm h–1 was larger than in the treatment of AMC of 0.25 FC and WR of 5 mm h–1.
|
|
|
For each soil, soil loss was the highest in the air-dry treatment and decreased on prewetting before the simulated rainfall (Fig. 5). This decrease was more pronounced as the WR was slower, in spite of the fact that runoff in the soils with 230 and 410 g kg–1 clay was not affected significantly by WR (Table 3). The notable effect of WR on soil loss and the smaller effect of WR on runoff in these soils suggest that WR had no significant effect on the hydraulic properties of the seal in them, but had a significant effect on the detachment rate and entrainment of soil particles in runoff. In soils with low to moderate aggregate stability, the high impact energy of raindrops was enough to disintegrate the aggregates and to form a seal whose hydraulic properties were unaffected by WR and AMC. However, during the slow WR and high AMC, increase in cohesion between soil particles took place (Sirjacobs et al., 2001) resulting in higher shear strength of surface seal and bigger detached particles so that sediment entrainment and soil loss decreased. Mamedov et al. (2002) also reported that a decrease in WR decreased the amount of soil loss, and this effect was more evident as clay content in the soil increased.
The response of soil loss to AMC depended on clay content and WR (Fig. 5). In the soil with 230 g kg–1 clay, a trend (not significant) in soil loss was found between the 0.25 and 0.5 FC treatments in both WR levels (Fig. 5). In this soil, the seal strength controlled soil detachment, and seal strength was slightly affected by AMC (Sirjacobs et al., 2001). With increase in AMC, seal strength increased and soil loss decreased. In the soil with 410 g kg–1 clay, the role of seal strength and soil detachment in sediment delivery was small and no effect of AMC, on both runoff and soil loss was observed. In the soil with 620 g kg–1 clay, AMC had an effect on both, soil loss and runoff and this effect was explained above. Prewetting the clay soil with 5 mm h–1 resulted in some aggregate slaking and a slight increase in both, runoff and erosion (Table 2 and Fig. 5). When this soil was prewetted with 1 mm h–1 mist, a decrease of soil loss and runoff with increasing AMC was observed. This phenomenon was explained by assuming that WR of 5 mm h–1 was fast enough to produce some slaking of the aggregates and increasing AMC increased the amount of slaked aggregates and detached clay particles, which contribute to seal development.
|
CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The effect of the wetting treatments on soil loss during the 80 mm of simulated rain was more pronounced than the effect of these treatments on runoff. Whereas the IR and runoff production depended mostly on seal formation, soil erosion is controlled by detachment of soil material from soil mass by raindrop impact and the transport of the resulting sediments by runoff or splash. Detachment of sediments from the soil surface depends also on seal strength, and with an increase in seal strength there is less detachment. Thus, soil loss, being dependent on two mechanisms that are controlled by seal formation (runoff production and detachment) was more sensitive than runoff production to differences in WR and AMC.
Most of the studies conducted on seal formation, IR, and interrill erosion were based on results obtained in air-dry soils that were exposed to fast WRs. However, in field conditions the soils can be wetted at different rates, according to the different rain intensities in different seasons, to different antecedent moisture content before they are exposed to high intensity rainstorms. These different wetting conditions can change the response of the soil to seal formation after tillage. For example, in Mediterranean climate, it can be expected that soil erosion will be higher in fall, when events of heavy rain fall on dry soil, than in spring, when events of rain start with low intensity, wetting the soil before the high intensity rain.
The implications of this study are quite relevant for soil erosion modeling. Most of the existing models have developed experimental regressions, which consider a constant detachability and decline of IR of the soil, related to intrinsic properties like sand or clay content. However, as indicated in this study, soil detachability, final infiltration of the seal and soil erodibility are dependent not only on these intrinsic properties, but also on soil conditions before rainfall, like WR and AMC. Therefore, to improve the accuracy of the erosion models, it is suggested that these parameters should be taken into account.
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Received for publication January 15, 2004.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. K. Udeigwe, J. J. Wang, and H. Zhang Predicting Runoff of Suspended Solids and Particulate Phosphorus for Selected Louisiana Soils Using Simple Soil Tests J. Environ. Qual., July 17, 2007; 36(5): 1310 - 1317. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Blanco-Canqui, R. Lal, and M. J. Shipitalo Aggregate Disintegration and Wettability for Long-Term Management Systems in the Northern Appalachians Soil Sci. Soc. Am. J., April 5, 2007; 71(3): 759 - 765. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. I. Mamedov, C. Huang, and G. J. Levy Antecedent Moisture Content and Aging Duration Effects on Seal Formation and Erosion in Smectitic Soils Soil Sci. Soc. Am. J., March 29, 2006; 70(3): 832 - 843. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. A. Fox and R. Metla Soil Property Analysis using Principal Components Analysis, Soil Line, and Regression Models Soil Sci. Soc. Am. J., September 29, 2005; 69(6): 1782 - 1788. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Lado, M. Ben-Hur, and S. Assouline Effects of Effluent Irrigation on Seal Formation, Infiltration, and Soil Loss during Rainfall Soil Sci. Soc. Am. J., August 4, 2005; 69(5): 1432 - 1439. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
= 0.05).
