Published in Soil Sci. Soc. Am. J. 68:1162-1168 (2004).
© 2004 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
DIVISION S-1—SOIL PHYSICS
A Conceptual Model to Predict the Deflation Threshold Shear Velocity as Affected by Near-Surface Soil Water
II. Calibration and Verification
Wim M. Cornelis*,
Donald Gabriels and
Roger Hartmann
Ghent University, Dep. Soil Management and Soil Care, Coupure links 653, B-9000 Gent, Belgium
* Corresponding author (wim.cornelis{at}UGent.be).
 |
ABSTRACT
|
|---|
A conceptual model to predict the threshold shear velocity, which should be overcome to initiate deflation of moist sediment, was recently developed by Cornelis et al. The model relates the threshold shear velocity to the ratio between water content and the water content at a matric potential of –1.5 MPa, and contains one proportionality coefficient that accounts for the effect of near-surface wetness. The present study was conducted (i) to determine that proportionality coefficient and hence calibrate the model, and (ii) to verify the calibrated model. The model calibration was achieved through curve fitting the expression against a data set from wind-tunnel experiments that were conducted on different sized sand particles and soil aggregates. Each sediment sample tray was prewetted, and subjected to different shear velocities and hence to different evaporation rates that dried the sediment. Once particle entrainment became sustained as recorded with a saltiphone, samples were taken to a depth of 1 mm to determine water content gravimetrically. To verify the calibrated model, threshold shear velocities simulated with our expression were compared with values obtained with Chepil's model of 1956. It was observed that the soil had to dry out to 75% of the water content at a matric potential of –1.5 MPa for deflation to occur. A single proportionality coefficient value could be used for the different sized sand and soil particles. Very good agreement was observed between our model and the model of Chepil.
|
INTRODUCTION
|
|---|
ASSESSMENT OF THE threshold shear velocity, which should be overcome to initiate deflation, that is, the process of particle detachment by wind shear at the surface, is essential for wind erosion prediction. Near-surface soil water contributes greatly to the binding forces keeping particles together, and hence it significantly affects the deflation threshold shear velocity of wetted sediment.
Several authors have studied the influence of near-surface wetness on a deflation threshold, such as the threshold shear velocity or the threshold wind velocity measured at a given height. Chepil (1956), Bisal and Hsieh (1966), Gregory and Darwish (1990), and Saleh and Fryrear (1995) observed a critical water content close to the water content at a matric potential of –1.5 MPa, above which no wind erosion occurs. Such a critical water content was also reported by Tanaka et al. (1954), Azizov (1977), Horikawa et al. (1982), Chen et al. (1996), and Shao et al. (1996) for the particular sediment they were using in their experiments. However, they did not relate this critical water content to its corresponding matric potential. Except for Bisal and Hsieh (1966), and Gregory and Darwish (1990), these workers all presented exponential, polynomial, or power-law growth functions to predict the deflation threshold. Such functions seem to be in contrast with the widely cited models of Belly (1964), Hotta et al. (1984), McKenna-Neuman and Nickling (1989), and Fécan et al. (1999). In the latter models, u*tw increases asymptotically to reach a maximum value even if water content further increases.
Recently, Cornelis et al. (2004) developed a new conceptual model to predict the deflation threshold shear velocity as affected by near-surface soil water. They related the deflation threshold shear velocity to the ratio between water content and water content at a matric potential of –1.5 MPa using following expressions:
 | [1] |
where u*tw is the deflation threshold shear velocity as affected by near-surface soil water, 
s is the particle density (Mg m–3), f is the fluid density (Mg m–3), g is the gravitational acceleration (m s–2), d is the particle diameter (m), and
 | [2] |
where A1, A2, and A3 are model coefficients, 
is the surface tension of the liquid (N m–1), 
md is the matric potential at oven dryness (approximately –103 MPa), w is the gravimetric water content (kg kg–1), and w1.5 the gravimetric water content at –1.5 MPa (kg kg–1). The two coefficients A1 and A2 are associated with respectively aerodynamic forces and interparticle forces between dry particles. Using wind-tunnel data from dry sediment experiments, Cornelis and Gabriels (2004) found A1 = 0.013 and A2 = 1.7 x 10–4 N m–1. The coefficient A3 is associated with the effects of soil water through capillary and adsorptive forces. It depends on particle shape, particle roughness, particle diameter, interparticle distance, and liquid surface tension of a given bulk soil, and needs to be determined by curve fitting. The water content at –1.5 MPa is a constant for a given sediment and can be assessed experimentally, from databases or by using pedotransfer functions (Cornelis et al., 2001).
The objective of the present study was (i) to determine the proportionality coefficient A3 in the conceptual model presented by Cornelis et al. (2004) and hence calibrate the model, and (ii) to verify the calibrated model. The model calibration was achieved through curve fitting the expression against a data set from wind-tunnel experiments that were conducted on different sized sand particles and soil aggregates. To verify the calibrated model, threshold shear velocities simulated with our expression were compared with values obtained with the model of Chepil (1956).
|
MATERIALS AND METHODS
|
|---|
Model Calibration
The sediment used was dune sand and sandy loam soil aggregates. The sand was collected from the Belgian coastal dunes (Bredene), whereas the aggregates were taken from a sandy loam soil on an agricultural field near Melle (Belgium). The physicochemical properties of both sediments as well as their fully dispersed particle-size distribution are given in Table 1. Particle-size distribution was determined with the pipette method (Gee and Bauder, 1986). Organic-matter measurements were based on the Walkley and Black (1934) method, carbonate content was determined using an excess of H2SO4 and subsequent titration with NaOH, and the electrical conductivity at 25°C was measured on a saturated extract with Pt electrodes.
View this table:
[in this window]
[in a new window]
|
Table 1. Fully dispersed particle-size distribution and physicochemical properties of the two sediments used in the experiments.
|
|
The sediments were oven-dried and sieved to obtain three particle-size ranges per sediment: 100 to 200, 200 to 500, and 50 to 500 for the dune sand, and 100 to 200, 200 to 300, and 300 to 500 µm for the soil aggregates. The bulk densities were 1.66, 1.66, 1.68, 1.44, 1.44, and 1.34 Mg m–3, respectively. The water content at –1.5 MPa was determined in three replicates using a pressure chamber (Soilmoisture Equipment, Santa Barbara, CA).
The experiments to calibrate the model were conducted in the wind tunnel of the International Centre for Eremology, Ghent University, Belgium. It is a closed-circuit blowing-type wind tunnel with a 12-m long, 1.2-m wide, and 3.2-m high working section (Gabriels et al., 1997). The boundary layer, that is, that layer close to the surface of the sediment in which the wind velocity is retarded due to the shear exerted by the surface, was set at a height of 0.60 m by using a combination of spires and roughness elements (Cornelis, 2002).
A sample tray 0.95 m long, 0.40 m wide, and 0.02 m deep was located at a distance of 6.0 m downwind from the entrance of the wind-tunnel test section. The tray was filled with oven-dried sediment and the sample was smoothed and leveled to the test-section's false floor by drawing a straight edge across its surface. The tunnel floor windward of the roughness elements was covered with commercial emery paper with a roughness length similar to that of the sediment. The sample tray was then wetted by spraying a fine mist, a wetting method also applied by Tanaka et al. (1954), Chepil (1956), Logie (1982), Horikawa et al. (1982), and McKenna-Neuman and Nickling (1989). Spraying allowed the soil aggregates to remain loose, whereas wetting from below (Bisal and Hsieh, 1966; Saleh and Fryrear, 1995), or thoroughly mixing the soil with water (Chen et al., 1996) resulted in disruption of individual aggregates as was observed in preliminary tests. Following wetting, the samples were covered with a perspex sheet and were allowed to equilibrate for 24 h. The sample trays were then exposed to wind with different shear velocities u* ranging from a value close to the sediment's threshold shear velocity u*t to a value of 0.6 m s–1. During each test run, the wind velocity was held constant and the sample was allowed to evaporate.
A saltiphone was continuously monitoring any particle deflating from the sample tray. The saltiphone was located along the centerline of the test section at a height of 0.035 m and at a distance of 6.85 m from the entrance of the test section, that is, at a distance of 10 cm windward of the leeward edge of the sample tray. It is an acoustic sediment sensor that measures the number of saltating particles that bounce against a microphone (Spaan and van den Abeele, 1991). The saltiphone was connected to a Campbell-Scientific datalogger (Campbell Scientific, Logan UT), which was connected to a personal computer. By using PC208W software, the impacts on the saltiphone sensor could be followed on the computer's screen at a frequency of 1 Hz. From the instant of continuous particle entrainment, that is, from the moment that three impacts were recorded within 1 min, the sample tray was covered with the perspex sheet, and the wind tunnel was switched off. A few minutes later, when wind velocity was almost zero, three small samples were taken by scraping of the upper 1 mm of the sediment using a sharp knife. Since we could visually observe differences in water content over the tray due to a difference in color, two samples were taken at a dry section and one sample at a wetter section. The samples were then weighed, oven-dried at 105°C during at least 24 h, and weighed again to determine water content gravimetrically.
Continuous deflation only occurred once the first dry sections appeared. In Fig. 1, an example of the evolution of entrained particles is given as a function of time. It appears that at the initial stages of the test run impacts already are recorded, but rather sporadically. These impacts could be associated with isolated surface particles that remain perched in a precarious position during sample preparation as was suggested by Greeley et al. (1977). In these initial stages, no dry sections were observed. In the case of the example shown in Fig. 1, a dry section appeared and deflation became steady only 42 min after the first impact was observed. It should be noted that the saltiphone was not able to detect the soil aggregates with a diameter between 100 and 200 µm at the low to medium wind velocities. The impact energy of these particles and hence the frequency of the pulse was too small to be detected by the saltiphone. When the shear velocity was higher than 0.5 m s–1, impact energy was sufficiently high for these particles to be recorded. In the case of soil aggregates ranging from 100 to 200 µm, dislodgement of particles was therefore determined visually by using a noncounting Ne-He laser beam (Logie, 1982).
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 1. Particle impact on the saltiphone (in number of particles per second i) vs. time from the first impact. Notice that the value of i was always equal to 10.1 which is the smallest possible value that can be measured with the saltiphone, and which is associated with the algorithm that is used (see Spaan and van den Abeele, 1991). This does not necessarily mean that the number of impacting particles was 10.1.
|
|
It is also important to note here that the relative humidity was almost constant during the experiments. It was measured at a 1-Hz frequency with a humidity probe with capacitive sensor (Testo, Lenzkirch, Germany), and it ranged from 31 to 34%. To have an idea about the evaporation rate during the test runs, tests were performed with four 7.5-cm wide cylindrical evaporation pans that were located in an empty tray, which was placed at the same location as the sediment tray. The potential evaporation rate from the pan Epan ranged from 0.7 to 1.4 mm h–1 depending on the temperature and the shear velocity (Fig. 2). If a pan coefficient of 0.5 were taken (Allen et al., 1998), this would correspond to a reference evapotranspiration Eo of 0.4 to 0.7 mm h–1. For comparison, Skidmore et al. (1969) calculated for Manhattan, KS, shortly past midday, Eo values of about 0.8 and 1.3 mm h–1 for what they called a ‘non-windy’ and a ‘windy’ day, respectively. The air temperature in our study was measured with a NiCr-Ni temperature sensor incorporated into the anemometers and ranged from 24 to 34°C.
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 2. Pan evaporation Epan vs. temperature T at different free-stream wind velocities u , under the experimental conditions in the wind tunnel.
|
|
The wind velocity was measured with 16-mm vane probes (Testo, Lenzkirch, Germany) at a 1-Hz frequency. They were mounted at heights z of 0.025, 0.096, 0.170, 0.256, and 0.377 m at a downwind distance of 5.90 m along the test-sections centerline. The time-average wind velocity was then calculated at each height over the whole test run and shear velocity was determined using the well-known Prandt–von Kármán logarithmic law and the method of Ling and Untersteiner (1974).
The coefficient A3 (see Eq. [2]) associated with the effect of near-surface soil water on deflation was determined by a linear transformation of Eq. [2] in combination with Eq. [1], in which A3 corresponds to the slope of the regression equation:
 | [3] |
where
 | [4] |
and
 | [5] |
Note that A3X is the additional term in Eq. [2] that takes into account the effect of near-surface wetness, and is associated with the interparticle force due to wet bonding. The term Y is associated with the forces acting on dry particles resting on a surface bed and being exposed to a fluid stream.
Model Verification
Comparison of predictions applying the model of Cornelis et al. (2004) with data from other studies reported in literature was difficult since these studies were conducted on sediment with particle-size distributions and water retention properties different than in our study. Our model was therefore verified against simulations with the empirical model of Chepil (1956), which indexes water content to matric potential using the water content at –1.5 MPa:
 | [6] |
where u*t is the threshold shear velocity of dry particles (m s–1). The model of Cornelis et al. (2004), which also uses the water content at –1.5 MPa, can therefore be compared easily with Chepil's model for soils of different particle size, particle or aggregate density, and water retention properties, performing simulations at different water contents. Two hundred simulations were done for particles with sizes of 25, 50, 100, 250, and 500 µm, particle densities of 2.65 Mg m–3 (sand grains) and 1.47 Mg m–3 (sandy loam aggregates), and ten w/w1.5 ratios ranging from 0 to 1 in steps of 0.1. The u*t values in Eq. [6] were computed using the conceptual model of Cornelis and Gabriels (2004) for dry sediment, which is equal to Eq. [1] with w = 0 in Eq. [2]. It is worth mentioning that Chepil used sediments ranging from dune sand to silty clay.
The model of Chepil was further chosen because of its good performance in predicting the threshold shear velocity as a function of the water content of the uppermost surface layer (Cornelis and Gabriels, 2003). One of the reasons for this good performance compared with other studies, was that in those studies, the water content was not determined at the very instant or location at which deflation occurred, although it was directly related to the deflation threshold. Since the sediment surface dries out rather quickly under normal conditions in wind tunnels, the deflation thresholds observed in many studies are therefore often overestimated (Cornelis and Gabriels, 2003). However, Chepil (1956) determined the deflation threshold indirectly by measuring mass transport rates above sediment with different water contents and under different shear velocities. As a result, the transport rate became almost zero at relative high average near-surface water contents of the sample tray, even if there would have been some irregular deflation from dry spots and hence the determined threshold shear velocity is not expected to be underestimated. The transport rate became significantly larger than zero once sustained deflation takes place.
|
RESULTS AND DISCUSSION
|
|---|
Model Calibration
The deflation appeared to be sustained, that is, when at least three impacts were recorded on the saltiphone in 1 min, once the first relatively dry sections were observed. The occurrence of relatively dry and wetter regions was most pronounced at high shear velocities. In Fig. 3, the observed u*tw values are plotted as a function of w, making the distinction between dry-section and wet-section samples. Since we were merely interested in the wet-bonding force that must be overcome for deflation to occur, the water contents of the relatively dry sections will be considered in the further discussion, except otherwise mentioned. Also plotted is Eq. [1] with A from Eq. [2], and Cornelis and Gabriels' (2004) data for oven-dry particles. In Fig. 4, the term associated with the forces acting on dry particles resting on a surface bed and exposed to a fluid stream (Y in Eq. [3]) is plotted against the term associated with the force due to wet bonding between two particles (X in Eq. [3]). The slope of the regression line corresponds to A3 and was found to be equal to 3 x 1014 N–1 m–1 (significant below the 0.0001 level). Figure 5 illustrates the effect of A3 on u*tw as a function of w for the case of a mean particle diameter of 250 µm and w1.5 = 0.019 kg kg–1. As the value of A3 rises, u*tw at w = 0 slightly rises as well. When A3 exceeds 3 x 1014 N–1 m–1, the increase in u*tw starts to become higher than 2%.
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 3. Observed threshold shear velocity u*tw data vs. gravimetric water content w data for dune sand with particle-size range of (a) 100 to 200, (b) 50 to 500, (c) 200 to 500 µm, and (d) sand loam soil aggregates with particle-size range of 100 to 200 µm, (e) 200 to 300, and (f) 300 to 500 µm. Also plotted is Eq. [1] with A from Eq. [2] and A3 = 3 x 1014 N–1 m–1. The oven-dry data are from Cornelis and Gabriels (2004).
|
|
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 4. The term for the forces acting on dry particles resting on a surface bed and exposed to a fluid stream (Y in Eq. [1]) vs. the term for the force due to wet bonding between two particles (X in Eq. [1]).
|
|
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 5. Simulations of u*tw using Eq. [1] and [2] for varying values of the coefficient A3 vs. gravimetric water content w. The simulations are for sand with d = 250 µm and w1.5 = 0.019 kg kg–1.
|
|
Although the data show a lot of scatter and the overall R2 was 0.46 only, Fig. 3 illustrates that the model followed the data rather well. At low water contents, the increase in u*tw with w was gradual. But once water content reached a given value, a steep increase in u*tw became apparent when considering the data that are associated with sustained particle dislodgment. The threshold shear velocity soon reached a very high value and a critical water content, above which no wind erosion occurs, could be observed. This is in line with observations of Tanaka et al. (1954), Chepil (1956), Bisal and Hsieh (1966), Azizov (1977), Horikawa et al. (1982), Gregory and Darwish (1990), Saleh and Fryrear (1995), Chen et al. (1996), and Shao et al. (1996). Therefore, there was no need to quantify the upper boundary conditions of our model.
When considering the water content at the wetter sections, u*tw seemed to increase asymptotically to reach a maximum value even if water content further increased, as is the case in the models of Belly (1964), Hotta et al. (1984), McKenna-Neuman and Nickling (1989), and Fécan et al. (1999). This was also the case when taking the average water content of all water content samples per test run and indicated that wind erosion could occur, even if the average near-surface water content was rather high. However, as we observed that deflation was limited to the drier spots, the total mass transport rate could be expected to be rather low. These observations could also explain why for example, Sarre (1988) reported sand transport on beaches even if water content was 0.14 kg kg–1. The existence of a high spatial variability in surface water content, even on bare dune sand, which at first sight show the greatest homogeneity, was observed by Ritsema and Dekker (1994). To illustrate the occurrence of wet and dry spots and their possible effect on the water content measurements, the w values of the wet spots are plotted in Fig. 3 as well.
Model Verification
In Fig. 6, the threshold shear velocities u*tw predicted with the model of Chepil (1956)(Eq. [6]) are plotted against the threshold shear velocities u*tw predicted with the model of Cornelis et al. (2004)(Eq. [1] and [2]). Figure 6 clearly illustrates that both models agree very well for threshold shear velocities not exceeding 0.5 m s–1, which was about the maximum shear velocity generated in Chepil's study. This value corresponds to a w/w1.5 ratio of 0.75 and is close to the critical water content above which no wind erosion will take place.
Since there was such a good agreement between our data and Chepil's model, at least within the calibration data range of Chepil's experiments, our observations could further explain the discrepancy between Chepil's results and subsequent work from other researchers. In an attempt to explain these differences, Namikas and Sherman (1995) argued that it is possible that some cementation of the surface layer took place in Chepil's experiments, although Chepil reported that the soil samples were thoroughly mixed before each run. They further mention the difference in water content at –1.5 MPa of Chepil's dune sand compared with McKenna-Neuman and Nickling's (1989) data for well-sorted sand fractions. The value measured by Chepil was 0.013 kg kg–1, which is about the same as we measured, even for well-sorted sand. This is significantly higher than the 0.0005 to 0.002 kg kg–1 of McKenna-Neuman and Nickling (1989). The main reason for the discrepancy is, as we believe, the fact that most researchers probably measured occasional dislodgement of particles rather than sustained deflation. It is worthwhile to stress once more that the sustained deflation as was defined in this study implies barely three impacts per minute on the saltiphone, and hence it does not correspond to fully initiated sediment transport. Deflation as is defined in this study is apparently a more appropriate definition in the context of wind-erosion prediction than considering deflation as the moment that some particles are occasionally blown away. Our observations are also supported by the findings of Logie (1982) who observed that the wind had to dry the surface to a very low water content, which was 0.012 kg kg–1 in the case of the dune sand used in her study, before its erosive action could start.
The above considerations indicate that the theoretically derived model of Cornelis et al. (2004) is adequate to predict the increase of the threshold shear velocity with water content. Though this model was only calibrated for dune sand and sandy loam data, it has the potential to be valid for all type of soils, as it followed the Chepil (1956) model very well.
Finally, Fig. 7 illustrates, for terrestrial conditions and for particles with a density of 2.65 Mg m–3, the relative importance of the different moments that act upon wetted particles of different size as they are exposed to the wind. The different moments are divided by b1KG to avoid any assumption about the angle of repose and the shape of the particles, and are expressed in terms of their moment factors MF (N m) as defined in Table 2. The aerodynamic moment factor represent the moments associated with the horizontal drag force, the lift force, and the aerodynamic moment force, the gravimetric moment factor is associated with the particle weight including the weight of adsorption water around the particles, the dry-bonding moment factor accounts for the van der Waals forces, and the wet-bonding moment factor is responsible for the capillary and adhesive forces that keep particles together. As an example, the aerodynamic moment factor was calculated for a low and a high shear velocity u* (m s–1), being 0.25 and 1.8 m s–1, respectively, whereas the wet-bonding moment factor was computed for w/w1.5 ratios of 0.25, 0.50, and 1.00. Since the weight of adsorption water around the particles seemed to be negligible compared with the weight a dry particle, the three curves for the gravitational MF related to the three different water contents could not be distinguished visually. As can be seen from Fig. 7, near-surface wetness does not affect at all particle entrainment for w/w1.5 = 0.25 (or smaller). At such low water content (and under the given circumstances of f = 1.2 x 10–3 Mg m–3 and s = 2.65 Mg m–3), the wet-bonding moment is always lower than the other moments. The dominant retarding moment is due to dry-bonding forces (solid line) for d less than approximately 80 µm, whereas for d greater than approximately 80 µm particle weight (dotted line) becomes more important. With respect to the second wetness condition considered here, that is, for w/w1.5 = 0.5, an effect of near-surface wetness can now be observed. If u* = 0.25 m s–1, the wet-bonding moment always exceeds the aerodynamic moment. However, once particle size drops below approximately 20 µm, the dry-bonding moment becomes the dominant retarding moment, while the gravitational moment prevails for particles larger than approximately 500 µm. But for u* = 1.8 m s–1, the aerodynamic moments always dominate, and deflation occurs for all particle sizes. The latter is still the case when considering the third wetness condition, being water content at a matric potential of –1.5 MPa or w/w1.5 = 1.0. However, the wet-bonding moment is now the dominant retarding moment at all particle sizes and is almost as large as the aerodynamic moment corresponding to a shear velocity of 1.8 m s–1. This means that under such wetness conditions, deflation can only occur once u* 
1.8 m s–1, which corresponds to a wind velocity of 110 km h–1 at a 10-m height, if a roughness length z0 = 10–2 m is considered. From a practical perspective, no wind erosion is to be expected if the surface wetness is larger than –1.5 MPa.
View this table:
[in this window]
[in a new window]
|
Table 2. Definition of the moment factors associated with the different moments acting on wetted particles exposed to the wind.
|
|
|
CONCLUSIONS
|
|---|
The deflation threshold model developed by Cornelis et al. (2004) to predict entrainment of moist sediment by wind, relates the threshold shear velocity to the ratio between water content and the water content at a matric potential of –1.5 MPa, and contained one proportionality coefficient A3 that accounts for the effect of near-surface wetness. To determine the coefficient A3 and hence calibrating the model, wind-tunnel experiments were conducted with six fractions of dune sand and sandy loam. The deflation threshold was defined as the instant at which at least three particle impacts were recorded on a saltiphone within 1 min. It was observed that entrainment of particles became sustained only once the surface layer dried to a water content w that was below –1.5 MPa, even at relatively high shear velocities and that a critical water content above which dislodgment of particles will cease exists. Its value was found to be about 75% of the water content w1.5 at a matric potential of –1.5 MPa. By curve fitting the model against wind tunnel data, the coefficient A3 was found to be equal to 3 x 1014 N–1 m–1 for both sediment types.
To verify the calibrated model, threshold shear velocities simulated with our expression were compared with values obtained with the model of Chepil (1956). There was a very good agreement between both models for w/w1.5 ratios below 0.75. Notwithstanding this, there seemed to be a discrepancy between both our data and the data of Chepil (1956) on the one hand, and the observations of other workers such as Hotta et al. (1984), Chen et al. (1996) and Azizov (1977)—they found lower threshold velocity values at a given water content—on the other hand. This was probably due to the fact that in our study each test run continued until deflation was sustained. Particles that were occasionally dislodged were not considered as true deflation. Furthermore, we observed some spatial variation in water content at the end of each experiment. It was visually observed that deflation was sustained only once the drier sections appeared. It is probable that these phenomena occurred as well in the other above-mentioned studies. If these variations in water content are not taken into consideration, or if large or deep water content samples are taken, the average water content corresponding to the wind or shear velocity at the instant of particle motion will be higher.
Finally, it can be concluded that the wind should dry the surface layer to water contents slightly lower than wetness conditions at –1.5 MPa before wind erosion will occur. Once this value is reached, a dramatic decrease in threshold shear velocity was observed. As was pointed out in Cornelis and Gabriels (2003), a tiny surface layer will then be removed and a surface layer with a higher water content appears. Unless temporal changes in wetness of the uppermost 1 mm of the soil layer and its removal cannot be accurately modeled, wind-erosion models will fail to produce realistic predictions.
|
NOTES
|
|---|
Mention of the company name is for the convenience of the reader and does not constitute any endorsement in whatever sense from the authors.
Received for publication April 8, 2003.
|
REFERENCES
|
|---|
- Allen, R.G., L.S. Pereira, D. Raes, and M. Smith. 1998. Crop evapotranspiration, Guidelines for computing crop water requirements. FAO Irrigation and drainage paper 56. FAO, Rome.
- Azizov, A. 1977. Influence of soil moisture in the resistance of soil to wind erosion. Soviet Soil Sci. 1:105–108.
- Belly, P.Y. 1964. Sand movement by wind. Tech. Memo. 1. U.S. Army Corps of Engineers, Coastal Engineering Research Center, Washington, DC.
- Bisal, F., and J. Hsieh. 1966. Influence of moisture on erodibility of soil by wind. Soil Sci. 102:143–146.
- Chen, W., Z. Dong, Z. Li, and Z. Yang. 1996. Wind tunnel test of the influence of moisture on the erodibility of loessial sandy loam soils by wind. J. Arid Environ. 34:391–402.
- Chepil, W.S. 1956. Influence of moisture on erodibility of soil by wind. Soil Sci. Soc. Am. Proc. 20:288–292.
- Cornelis, W.M. 2002. Erosion processes of dry and wet sediment as induced by wind and wind-driven rain: A wind-tunnel study. Ph.D. dis. Ghent University, Ghent, Belgium.
- Cornelis, W.M., and D. Gabriels. 2003. The effect of surface moisture on the entrainment of dune sand by wind: An evaluation of selected models. Sedimentology 50:771–790.
- Cornelis, W.M., and D. Gabriels. 2004. A simple model to predict the deflation threshold shear velocity of loose particles. Sedimentology 51:39–51.
- Cornelis, W.M., D. Gabriels, and R. Hartmann. 2004. A conceptual model to predict the deflation threshold shear velocity as affected by near-surface soil water: 1. Theory. Soil Sci. Soc. Am. J. 68:1154–1161 (this issue).[Abstract/Free Full Text]
- Cornelis, W.M., J. Ronsyn, M. Van Meirvenne, and R. Hartmann. 2001. Evaluation of pedotransfer functions for predicting the soil moisture retention curve. Soil Sci. Soc. Am. J. 65:638–648.[Abstract/Free Full Text]
- Fécan, F., B. Marticorena, and G. Bergametti. 1999. Parametrization of the increase of the aeolian erosion threshold wind friction velocity due to soil moisture for arid and semi-arid areas. Ann. Geophys. 17:149–157.
- Gabriels, D., W.M. Cornelis, I. Pollet, T. Van Coillie, and M. Ouessar. 1997. The I.C.E. wind tunnel for wind and water erosion studies. Soil Technol. 10:1–8.
- Gee, G.W., and J.W. Bauder. 1986. Particle-size analysis. p. 383–411. In A. Klute (ed.) Methods of soil analysis. Part 1. 2nd ed. Agron. Monogr. No. 9. ASA and SSSA, Madison, WI.
- Greeley, R., B.R. White, J.B. Pollack, J.D. Iversen, and R.N. Leach. 1977. Dust storms on Mars: Considerations and simulations. Tech. memo. 78423, Ames Research Center, NASA, Moffet Field, CA.
- Gregory, J.M., and M.M. Darwish. 1990. Threshold friction velocity prediction considering water content. Pap. ASAE 902562. ASAE, St. Joseph, MI.
- Horikawa, K., S. Hotta, and S. Kubota. 1982. Experimental study of blown sand on a wetted sand surface. Coastal Eng. Japan 25:177–195.
- Hotta, S., S. Kubota, S. Katori, and K. Horikawa. 1984. Sand transport by wind on a wet sand surface. p. 1263–1281. In Proceedings of the 19th Coastal Engineering Conference. ASCE, New York.
- Ling, C.H., and N. Untersteiner. 1974. On the calculation of the roughness parameter of sea ice. J. Geophys. Res. 79:4112–4114.
- Logie, M. 1982. Influence of roughness elements and soil moisture on the resistance of sand to wind erosion. Catena. Suppl. 1:161–174.
- McKenna-Neuman, C., and W.G. Nickling. 1989. A theoretical and wind tunnel investigation of the effect of capillary water on the entrainment of sediment by wind. Can. J. Soil Sci. 69:79–96.
- Namikas, S.L., and D.J. Sherman. 1995. A review of the effects of surface moisture content on aeolian sand transport. p. 269–293. In V.P. Tchakerian (ed.) Desert aeolian processes. Chapman & Hall, London.
- Ritsema, C.J., and L.W. Dekker. 1994. Soil moisture and dry bulk density patterns in bare sands. J. Hydrol. (Amsterdam) 154:107–131.
- Saleh, A., and D.W. Fryrear. 1995. Threshold wind velocities of wet soils as affected by wind blown sand. Soil Sci. 160:304–309.
- Sarre, R.D. 1988. Evaluation of aeolian sand transport equations using intertidal zone measurements, Saunton Sands, England. Sedimentology 35:671–679.
- Shao, Y., M.R. Raupach, and J.F. Leys. 1996. A model for predicting aeolian sand drift and dust entrainment on scales from paddock to region. Aust. J. Soil Res. 34:309–342.
- Skidmore, E.L., H.S. Jacobs, and W.L. Powers. 1969. Potential evaporation as influenced by wind. Agron. J. 61:543–546.[Abstract/Free Full Text]
- Spaan, W.P., and G.D. van den Abeele. 1991. Wind borne particle measurements with acoustic sensors. Soil Technol. 4:51–63.
- Tanaka, T., H. Sano, and S. Kakinuma. 1954. Study on the control of wind erosion. (In Japanese.) Agric. Meteorol. 10:24–30.
- Walkley, A., and I.A. Black. 1934. An examination of the Degtjareff method for determining soil organic matter and a proposed modification of the chromic acid titration method. Soil Sci. 37:29–38.
This article has been cited by other articles:
|
|
|
|
 
J. E. Bullard
Arid geomorphology
Progress in Physical Geography,
August 1, 2006;
30(4):
542 - 552.
[PDF]
|
|
|
|
|
|
|
D. A. Abrahamson, D. E. Radcliffe, J. L. Steiner, M. L. Cabrera, J. D. Hanson, K. W. Rojas, H. H. Schomberg, D. S. Fisher, L. Schwartz, and G. Hoogenboom
Calibration of the Root Zone Water Quality Model for Simulating Tile Drainage and Leached Nitrate in the Georgia Piedmont
Agron. J.,
November 17, 2005;
97(6):
1584 - 1602.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
W. M. Cornelis, D. Gabriels, and R. Hartmann
A Conceptual Model to Predict the Deflation Threshold Shear Velocity as Affected by Near-Surface Soil Water: I. Theory
Soil Sci. Soc. Am. J.,
July 1, 2004;
68(4):
1154 - 1161.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
