Effects of surfactant treatments on the wettability of ... - CSIRO Publishing
CSIRO PUBLISHING www.publish.csiro.au/journals/ajsr
Australian Journal of Soil Research, 2005, 43, 383–395
Effects of surfactant treatments on the wettability of a water repellent grass-covered dune sand Louis W. DekkerA,C , Klaas OostindieA , Stanley J. KostkaB , and Coen J. RitsemaA A Alterra,
Soil Science Center, Green World Research, PO Box 47, 6700 AA Wageningen, The Netherlands. B Aquatrols Corporation of America, Paulsboro, NJ 08066, USA. C Corresponding author. Email address: [email protected]
Abstract. The objective of this study was to evaluate the effectiveness of the surfactant formulation Primer604 for amelioration and management of soil water repellency in grass-covered dune sand. The soil is severely to extremely water repellent to a depth of >0.50 m during dry periods. Primer604 was applied 12 times between 22 April and 23 November 1999. During that period, soil samples were taken in the untreated and treated plot— 8 times in transects and 2 times in soil blocks. A total of 4950 samples were collected for assessment of the actual water repellency and for the spatial and temporal variability of the water content of the soil. Resistance to wetting was determined by measuring the wetting rate of field-moist samples. Measurements of water repellency revealed that applications of Primer604 resulted in less persistent water repellency in the surface layer to a depth of 0.05 m. No effects were observed deeper in the soil profile, likely due to adsorption of the surfactant material in the surface layer. In the surface layer (0–0.025 and 0.025–0.05 m), the critical soil water content below which the soil is actually water repellent in the field was lowered distinctly by the application of Primer604, potentially due to coating of water repellent particle surfaces by the surfactant. This suggests that the surface layer (0–0.05 m) in the Primer604-treated soil can dry to lower water contents than in the untreated soil before water repellency is induced. The thatch layer (0–0.025 m) of the treated soil was often found to have slightly higher water contents than of the untreated soil. The surfactant did not equalise the uneven moisture distribution in the soil below the surface layer (0–0.05 m). Primer604 applications increased the wetting rate of field-moist samples from the thatch layer. This may result in a more effective wetting of the root-zone during rain events or irrigation, and a reduction in runoff. Additional keywords: actual water repellency; critical soil water content; irregular wetting; preferential flow; water drop penetration time (WDPT) test; wetting rate.
Introduction The phenomenon of soil water repellency has been recognised in various parts of the world (DeBano 2000; Doerr et al. 2000; Jaramillo et al. 2000) including the Netherlands (Hooghoudt 1950; Dekker and Jungerius 1990; Dekker and Ritsema 1996, 2000) and has resulted in serious land-use problems in agriculture (Blackwell 2000) and an ongoing management problem on sand-based turfgrass systems (Cisar et al. 2000). It has been recognised for many years that soil water repellency is often a function of the type of organic matter incorporated in the soil, and that certain types of organic matter cause water repellency by several means (e.g. Doerr et al. 2000; Dekker et al. 2001; Hallett et al. 2001). Soil water repellency may dramatically affect fieldscale water and solute movement, and has often been underestimated (Bauters et al. 2000). Water repellency and © CSIRO 2005
its spatial variability have been shown to cause non-uniform wetting and preferential flow in many soils (Dekker and Ritsema 1994, 1996, 2000; Ritsema and Dekker 1996; Ritsema et al. 1998b). Based on experimental observations, Ritsema et al. (1993) proposed a conceptual model for water flow in water repellent sandy soils. According to this model, the initially uniform water infiltration is disrupted within the first few centimeters of the water repellent soil, causing water to move laterally towards microdepressions and regions with lower water repellency, where fingers are formed. Water is transported along these preferential flow pathways until decreasing water repellency in the soil at increasing depth causes divergence of the flow lines (Ritsema et al. 1998a; Nguyen et al. 1999). This process, occurring at the onset of an infiltration or leaching event, may drastically affect subsequent water movement (Ritsema and Dekker 1995). 10.1071/SR04090
0004-9573/05/030383
384
Australian Journal of Soil Research
Soil surfactants have been developed as a possible means for overcoming the problems caused by water repellent soils (Moore 1981; Rieke 1981; Kostka et al. 1997; Dekker et al. 2000; Kostka 2000). Wetting agents that have a strong affinity for the surface of hydrophobic soil particles and adsorb strongly at the soil surface will enhance infiltration rates at the soil surface interface. On the other hand, good water dispersion throughout the profile would require uniformity of penetration of the wetting agent in the profile. It is clear that a true test of the effectiveness of a soil wetting agent must include the assessment of the uniformity of distribution of the water in the soil, as well as the increase in infiltration rate and water content. The objective of our study was to evaluate the effectiveness of Primer604 for amelioration and management of soil water repellency in a pasture on a native dune sand. The present paper describes the influence of the surfactant in reducing the severity of soil water repellency and increasing the wetting rate, and its effect on the spatial variability in soil water content, water flow, and wetting patterns. Materials and methods Field-soil and treatment The wetting agent was applied to a plot (25 by 5 m) of a dune sand having a grass cover near Ouddorp in the south-western part of the Netherlands. An untreated adjacent plot was used for comparison. The soil consisted of fine sand with 3 m, and was classified as mesic Typic Psammaquent (Dekker 1998). Organic matter contents of 18 and 10% (wt/wt) were found at depths of 0–0.025 and 0.025–0.05 m, respectively. At depths of 0.05–0.075 m organic matter content was 2%, and from 0.075 m downwards approximately 0.5%. The soil was known to be severely to extremely water repellent to a depth of >0.50 m during dry periods (Dekker and Ritsema 1994; Dekker et al. 2001). Water repellency of this sandy soil is due to a coating of the sand grains with hydrophobic material and the presence of hydrophobic particulate organic matter (Bisdom et al. 1993). Primer604 (Aquatrols Corporation of America, Paulsboro, NJ, USA) treatments at a rate of 1.85 mL/m2 (volume solution 70 mL/m2 ) were applied with a Mesto Pico backpack-type sprayer 12 times during the period 22 April–23 November 1999. Soil sampling Between 22 April and 12 October 1999, the spatial and temporal variability of the volumetric soil water content was evaluated 8 times in vertical transects by intensive sampling of the treated and the untreated plots. To allow more detailed determination of the wetting patterns in the dune sand, soil blocks (0.25 by 0.75 by 0.19 m) were sampled in both plots on 25 October and 23 November 1999 (Dekker et al. 2000). The soil of transects and blocks was sampled at 6 depths (0–0.025, 0.025–0.05, 0.07–0.095, 0.095–0.12, 0.14–0.165, and 0.165–0.19 m), using steel cylinders with a diameter of 50 mm. In each transect and at each depth, 35 adjacent samples were taken over a distance of approximately 1.8 m. A total of 3150 samples were collected in the transects. In each of the 4 soil blocks, 75 samples were taken at each depth along a horizontal plane in a regular grid of 15 by 5 samples. A total of 1800 samples were collected from all blocks. The cylinders were pressed vertically into the soil, emptied into plastic bags, and re-used. Plastic bags were tightly sealed to minimise evaporation. The field-moist bagged soil was weighed. ‘Actual’ soil water repellency
L. W. Dekker et al.
was measured (Dekker and Ritsema 1994), and after drying in a fan-oven during 1 week at 25◦ C, ‘potential’ water repellency was measured. Samples were further dried at 105◦ C and weighed again to calculate the water content and dry bulk density of each sample. A total of 4950 samples were collected and measured in this way. Water Drop Penetration Time (WDPT) test The persistence or stability of water repellency of the soil samples was examined using the WDPT test (e.g. King 1981). Using a standard medicine dropper, 3 drops of distilled water were placed on the smoothed surface of a soil sample, and the time that elapsed until the drops infiltrated was determined. Soil water repellency of all 4950 samples were measured under controlled conditions at a constant temperature of 20◦ C and a relative air humidity of 50%. In general, a soil is considered to be water repellent if WDPT exceeds 5 s (Dekker 1998). We applied an index allowing a quantitative classification of the persistence of soil water repellency as described by Dekker and Jungerius (1990). Thus, 7 classes of repellency were distinguished, based on the time for water drops to infiltrate the soil: class 0, wettable, non-water repellent (infiltration within 5 s); class 1, slightly water repellent (5–60 s); class 2, strongly water repellent (60–600 s); class 3, severely water repellent (600 s–1 h); and extremely water repellent (>1 h), further subdivided into class 4 (1–3 h), class 5 (3–6 h), and class 6 (>6 h). Water repellency was measured on the field-moist samples (‘actual’ water repellency), and again after drying at 25◦ C. The severity of water repellency measured on dried soil samples, the so-called ‘potential’ water repellency, is considered to be the most appropriate parameter for comparing soils with respect to their sensivity to water repellency (Dekker and Ritsema 1994), because differences in water content are wiped out. We measured the ‘actual’ water repellency on the field-moist samples immediately after recording their wet weight. By measuring the water content of the samples, we could assess ‘critical soil water contents’ for the different depths of the intensively sampled transects and soil blocks. The soil is wettable above, and water repellent below, these values (Dekker and Ritsema 1994). Wetting rate measurements Resistance to wetting was determined several times by measuring the wetting rate of field-moist samples collected at depths of 0–0.025 m in the treated and untreated plots prior to surfactant applications. The samples were collected in steel cylinders (50 cm3 ) with a height of 25 mm and a diameter of about 50 mm. To measure wetting rate, these samples, within their steel cylinders, were subjected to a constant pressure head of −2.5 cm water applied at the bottom of the samples (Dekker et al. 1998). The experimental set-up was designed in such way that water content changes in 1.0 vol% increments were recorded automatically. All measurements were performed in a controlled environment laboratory with a constant temperature of 20◦ C and a relative humidity of 50%.
Results Actual water repellency The entire soil profile was wet at the beginning of the study on 22 April 1999. All 210 samples taken at depths between 0 and 0.19 m were wettable or non-water repellent, exhibiting WDPT values 6 h on 17 May 1999 (Fig. 2). A decrease in severity of actual water repellency occurred in the soil layers sampled below 0.12 m on almost all sampling dates (Fig. 2). The variability in actual water repellency was high in the soil over short distances at all depths in the untreated and treated plot, with WDPT values often varying between 6 h on all sampling days (Figs 1, 2). Relatively small differences in actual water repellency between the untreated and treated plot were determined for samples
collected at depths of 0.07 to 0.19 m (Fig. 2). The thatch layer (0–0.025 m depth) was found to be wettable for both plots on 22 April, 12 August, 12 and 25 October, and 23 November 1999. More wettable samples were recorded at depths of 0–0.025 m and 0.025–0.05 m for the Primer604-treated plot compared with the untreated plot on 1 June and 8 July 1999 (Fig. 1). Distinctly lower WDPT values were recorded for the samples taken in the thatch layer and in the surface layer at depths of 0.025–0.05 m in the treated plot compared with the untreated plot on 2 and 21 September 1999 (Fig. 1). The spatial and the temporal variability of the persistence of actual water repellency in transects of the untreated and treated plot were remarkable (Fig. 2). Within a horizontal distance of only 1.8 m, extreme water repellency was detected in dry soil areas, and wettable soil in preferential
Actual water repellency Untreated
Treated
0–0.025 m 17 May 1 Jun 8 Jul 12 Aug 2 Sep 21 Sep 12 Oct 25 Oct 23 Nov
0.025–0.05 m 17 May 1 Jun 8 Jul 12 Aug 2 Sep 21 Sep 12 Oct 25 Oct 23 Nov 0%
20%
40%
60%
80%
100%
385
0%
20%
40%
60%
80%
100%
WDPT 6h
600–3600 s
Fig. 1. Relative frequency of the persistence of actual water repellency of field-moist samples taken at depths of 0–0.025 and 0.025–0.05 m in the untreated and treated dune sand on 9 sampling days (n = 35 at both depths for the 17 May to 12 October 1999 transects, and n = 75 for the 25 October and 23 November 1999 soil blocks).
386
Australian Journal of Soil Research
L. W. Dekker et al.
Untreated
Treated 17 May
1 June
8 July
12 August
2 September
21 September
12 October
0%
20%
40%
60%
80%
100%
0%
20%
40%
60%
80%
100%
WDPT 6h
Fig. 2. Relative frequency of the persistence of actual water repellency of field-moist samples (n = 35) taken at 4 depths (top to bottom: 0.07–0.095, 0.095–0.12, 0.14–0.165, 0.165–0.19 m) in the untreated and treated dune sand on 7 sampling days.
Surfactant treatment and wettability
Australian Journal of Soil Research
flow paths, the fingers. Examples are the 2 September and 12 October 1999 transects of the untreated plot and the 1 June, 8 July, and 2 September 1999 transects of the treated plot (Fig. 2). The temporal variability in actual water repellency of the soil in both plots is evident when comparing the diagrams of the transects sampled on different days. Spatial variability in severity of actual water repellency is demonstrated by the contour plots of the 25 October 1999 soil blocks, with a size of only 0.75 by 0.25 by 0.19 m, or 0.036 m3 (Fig. 3). It is worthy of note that the actual water repellency at depths of 0.0825–0.1775 m in the treated plot was more severe and distributed over larger areas than in the untreated plot. We assume that the applications of Primer604 resulted
in a movement of the surfactant from the thatch layer into the preferential flow paths and thereby induced an enhancement of wetting of these flow paths. As a consequence, dry pockets in the treated soil were provided with less rainwater and were therefore more persistent than in the untreated soil. Although the severity of actual water repellency decreased during the autumn rains, extreme repellency with WDPT values >1 h still occurred in dry soil pockets of the soil block of 23 November 1999 (not shown here). In conclusion, the application of Primer604 resulted in a decrease in water repellency in the surface layer to a depth of 0.05 m, due to adsorption of the surfactant in this zone. Deeper in the profile no positive effects could be observed.
Untreated
Treated
Distance (m)
0
Distance (m)
0
0.75
0.75
Depth (m) 0.0125
0.0375
0.0825
0.1075
0.1525
0.1775
WDPT
6h
60–600 s
387
600–3600 s
1–3 h
Fig. 3. Contours of the persistence of actual water repellency in horizontal planes (0.25 by 0.75 m) at 6 depths in the untreated and treated 25 October 1999 soil blocks.
Australian Journal of Soil Research
L. W. Dekker et al.
Untreated
Treated
0–0.025 m
60 50
)
40 30 20 10 0
0.025–0.05 m 50 40 30 20 10 0
0.07–0.095 m 40 35 30 25 20
Soil water content (vol%)
388
15 10 5 0
0.095–0.12 m 30 25 20 15 10 5 0
0.14–0.165 m
18 16 14 12 10 8 6 4 2 0
0.165–0.19 m 16 14 12 10 8 6 4 2 0 31-Mar
20-May
9-Jul
28-Aug
17-Oct
6-Dec
31-Mar
20-May
9-Jul
28-Aug
17-Oct
6-Dec
Date
◦
Fig. 4. Minimum (–), mean ( ), and maximum (–) soil water contents of samples taken at 6 depths in the 8 transects (n = 35) and 2 soil blocks (n = 75) of the untreated and treated dune sand between 17 May and 23 November 1999.
Surfactant treatment and wettability
Australian Journal of Soil Research
Soil water content A high degree of spatial and temporal variability of the soil water content was often found in all layers sampled (Fig. 4). The highest water contents were found in the surface layers, which possessed higher organic matter contents and lower dry bulk densities than the deeper layers. Relatively high soil water contents were established at depths between 0 and 0.05 m in April (not shown), October, and November; however, spatial variability remained high. For instance, the soil water content at depths of 0–0.025 m varied between 31.4 and 69.3 vol%, and at depths of 0.025–0.05 m between 26.1 and 57.5 vol% at the start of the experiment, on 22 April 1999. At the end of the experiment, on 23 November 1999, the water content of the thatch layer in the untreated plot varied between 29.9 and 47.5 vol% and in the treated plot between 33.7 and 54.3 vol% (Fig. 4). The mean soil water content in the thatch layer was often slightly higher in the treated plot than the untreated plot (Fig. 5). For example, in the treated plot, the mean soil water content was 43.8 vol% at depths of 0–0.025 m on 25 October 1999, compared with 40.8 vol% in the untreated plot. The plot treated with Primer604 applications exhibited slightly
higher mean soil water contents at all depths and in most transects compared with the untreated plot (Fig. 5). The diagrams of these transects also show clearly the wide variability which often existed between the soil water contents in the surface layer and in the soil at depths of 0.095–0.19 m. Relatively dry transects, with small variations in water content, were sampled on 17 May and 2 September 1999 (Fig. 4). Large differences in soil water content were found at depths of 0.07–0.19 m in the untreated and treated plots between 17 May and 23 November 1999 (Fig. 4). In this zone, wet fingers and dry soil areas were evident in the soil profile in both the untreated and treated plots. Surfactant treatment did not prevent irregular wetting of the water repellent subsoil. The variation in water content in the treated transects and soil blocks was also high, as demonstrated by the diagrams on the right-hand side of Fig. 4. Irregular wetting patterns were also encoutered at depths of 0.0825–0.1775 m in the soil blocks of the untreated and treated plots on 25 October 1999 (Fig. 6). Between 1 October and 25 October, precipitation amounted to 122 mm, yet soil
0.025–0.05 m
0–0.025 m 40
50 40
30
30 20 20 10
10
0
0.095–0.12 m
0.07–0.095 m 9
18 16 14 12 10 8 6 4 2 0
8 7 6 5 4 3 2 1 0
0.165–0.19 m
0.14–0.165 m 7
8
6
7
5
6 5
4
4
3
3
2
2
1
1 23 -N ov
25 -O ct
ov -N 23
12 -O ct
ct -O 25
-S ep
ct -O 12
21
ep -S 21
ug
p Se 2-
-A
ug -A 12
12
l Ju 8-
8Ju l
n Ju 1-
17 -M ay
ay
1Ju n
0
0
2Se p
Soil water content (vol%)
0
-M 17
389
Date Untreated
Treated
Fig. 5. Mean soil water contents at 6 depths in the untreated and treated dune sand between 17 May and 23 November 1999.
390
Australian Journal of Soil Research
L. W. Dekker et al.
was locally dry at depths of 0.0825–0.1775 m. Water content in these regions was generally 23 vol%, were determined as wettable (Fig. 7, upper left-hand diagram). All samples with a soil water content 25
Fig. 6. Contours of the volumetric soil water content in horizontal planes (0.25 by 0.75 m) at 6 depths in the untreated and treated dune sand on 25 October 1999.
Surfactant treatment and wettability
Australian Journal of Soil Research
critical soil water content of the thatch layer of the untreated plot is variable and ranges between 18 and 23 vol%, most likely depending on the wetting history of the soil, weather sequence, etc. The critical soil water content of the soil in the untreated plot at depths of 0.025–0.05 m was found to be between 14 and 19 vol% (Fig. 7, lower left-hand diagram). Although there were large differences in severity of actual water repellency at specific soil moisture contents, there was a distinct increase in severity with decreasing soil water contents, as shown in the diagrams of Fig. 7. Treatment with Primer604 caused a significant decrease in the critical soil water content of the thatch layer, as can be seen in the upper right-hand diagram of Fig. 7. The transition zone varied in this case between 12 and 16.5 vol%, compared with variation between 18 and 23 vol% in the untreated plot. For instance, all soil samples with a water content of 17 vol% were determined as wettable in the Primer604-treated plot, whereas all samples with this water content in the untreated plot exhibited slight to extreme repellency. Also at depths of 0.025–0.05 m there was a slight shift in the critical soil water content. Soil samples with water contents of 8–14 vol% were water repellent in the untreated plot, whereas in the Primer604-treated plot a number of samples with these water contents were still wettable.
Untreated
In conclusion, the critical soil water content in the thatch layer was lowered distinctly by the applications of Primer604. This means that the surfactant-treated soil dried to lower water contents than the untreated soil before water repellency was induced. Resistance to wetting of field-moist samples An instantly high wetting rate of the surface layer is important for the effective infiltration of rain and irrigation water as well as for prevention of erosion and runoff. Measurements with the wetting rate device showed that the water uptake of the thatch layer was generally more rapid in the beginning for samples from the plot treated with Primer604 than for samples from the untreated plot. Differences in instant wetting rate were observed between thatch layer samples from the untreated and treated plot on 5 of the 6 sampling dates (Fig. 8). The uptake of water (in mm) gives an indication of the amount of rainwater that can be absorbed readily. It is evident from the diagrams that water infiltrated more effectively into the thatch layer of the treated plot than the control. Initial soil water content of the samples played an important role for the wetting rate during the first hour (Fig. 8). Especially, samples from the untreated plot with soil water contents below the critical soil water content exhibited less affinity for water absorption.
Treated
0–0.025 m
6 5 4 3 2
WDPT class
1 0
0.025–0.05 m 6 5 4 3 2 1 0 0
20
40
391
60
0
20
40
60
Soil water content (vol%)
Fig. 7. Relationship between soil water content and persistence of actual water repellency at depths of 0–0.025 and 0.025–0.05 m in the untreated and treated dune sand. The transition zone, with critical soil water contents, is indicated with a grey tone.
392
Australian Journal of Soil Research
L. W. Dekker et al.
Soil water content 80
20 16 June
17 May 60
15
40
10
20
5
0
0
80
20 27 July
(vol%)
8 July 60
15
40
10
20
5
0
0
80
20 2 September
21 September
60
15
40
10
20
5
0
0 0
10
20
30
40
50
60
0
10
20
30
40
50
60
Time (min) Untreated
Treated
Fig. 8. Increase in water content (vol%) of field-moist samples taken on 6 sampling days in the untreated and treated thatch layer (0–0.025 m) during 1 h of wetting. The initial water content and water uptake are also indicated in mm.
Note that the wetting rate of the thatch layer did not increase after the first Primer 604 application, as shown by the diagram of 17 May 1999, but a distinct increase was observed after additional applications, as shown by the diagrams of 16 June–21 September 1999 (Fig. 8). In conclusion, the application of Primer604 increased the wetting rate of the thatch layer, which results in a more effective wetting of the root-zone during rain events and/or irrigation events, thereby decreasing runoff. Persistence of potential water repellency The persistence of potential water repellency of samples taken at depths of 0–0.025 and 0.025–0.05 m both in the transects and in the soil blocks of the untreated and treated plots was measured with the WDPT test after drying at 25◦ C. All field-moist samples from the 22 April 1999 transect
were wettable, but the WDPT varied between 60–600 s (class 2) and 3–6 h (class 5) after drying at 25◦ C (data not shown). Differences in potential water repellency (samples dried at 25◦ C prior to WDPT) occurred between samples taken at the same depths but also between samples taken on different sampling dates (Fig. 9). Notable differences in persistence of water repellency were found for both the 0–0.025 and 0.025–0.05 m depths in the untreated plot between the 2 September and 12 October 1999 transects. All samples from 2 September 1999 exhibited extreme water repellency with WDPTs between 1–3 and >6 h, whereas the 12 October 1999 samples showed strong to severe water repellency with WDPTs varying between 60 s and 1 h (Fig. 9). High spatial and temporal variability in potential water repellency were also determined for the samples taken at depths of 0.025–0.05 m in the
Surfactant treatment and wettability
Australian Journal of Soil Research
393
Potential water repellency (25°C) Untreated
Treated
0–0.025 m 17 May 1 Jun 8 Jul 12 Aug 2 Sep 21 Sep 12 Oct 25 Oct 23 Nov
0.025–0.05 m 17 May 1 Jun 8 Jul 12 Aug 2 Sep 21 Sep 12 Oct 25 Oct 23 Nov 0% WDPT
20%
40%
60%
80%
100%
0%
6h
20%
40%
60%
80%
100%
600–3600 s
Fig. 9. Relative frequency of the persistence of potential water repellency of samples dried at 25◦ C, and taken at depths of 0–0.025 and 0.025–0.05 m in the untreated and treated dune sand on 9 sampling days (for n see Fig. 1).
treated plot between 17 May and 23 November 1999, as demonstrated in Fig. 9. Because all samples were dry, the differences in water repellency must be due to differences in initial water content of the samples and a process of initiating water repellency in the field. The persistence of potential water repellency of the samples dried at 25◦ C is clearly negatively related to the initial soil water content and positively to the persistence of the actual water repellency of the samples. For example, the relatively dry and severe to extreme actual water repellency of the 17 May and 2 and 21 September 1999 transects (Fig. 1) resulted in locally extreme potential water repellency (Fig. 9).
Distinctly lower WDPT values after drying at 25◦ C were detected for the samples taken at both depths in the Primer604-treated plot between 1 June and 23 November 1999 compared with the untreated plot (Fig. 9). A majority of the samples taken from the thatch layer of the treated plot between 8 July and 25 October 1999 exhibited only slight water repellency after drying at 25◦ C, whereas most samples from the untreated plot exhibited severe to extreme water repellency (Fig. 9). In most cases the persistence of potential water repellency of the samples was distinctly higher than the actual water repellency after the samples had been dried at 25◦ C.
394
Australian Journal of Soil Research
However, it is worthy of note that the potential water repellency of samples on some days was less severe than the actual water repellency of field-moist samples on other days, thus underestimating the maximal persistence of water repellency that can occur in the field (see Fig. 1 and Fig. 9). This indicates that processes which are taking place in the field during dry weather cannot be artificially generated during drying in a laboratory oven over a time span of several days. Regardless of how water repellency was measured in soil samples, surfactant treatment generally shifted water repellency classes (actual or potential) to more wettable classes. Surfactant-induced shifts in water repellency classes were most evident in potential water repellency results. In conclusion, the spatial and temporal variability in persistence of water repellency after drying at 25◦ C was high at both depths for the treated and untreated plot. But more important, distinctly lower WDPT values were detected for samples at depths of 0–0.025 and 0.025–0.05 m from the Primer604-treated plot than the untreated plot. Discussion and conclusions Water repellency of soils may dramatically affect water and solute movement, due to non-uniform wetting and forming of preferential flow paths, so-called fingers, as also often occur in the dune sand studied. Soil wetting agents have been developed as a possible means for overcoming the problems caused by water repellency (e.g. Kostka et al. 1997; Cisar et al. 2000). It is evident that a test of the effectiveness of a soil wetting agent must include the assessment of the uniformity of distribution of the water in the soil. Soil moisture measurements in the untreated plot and in the treated plot with applications of Primer604 revealed that after rain events the thatch layer in the treated plot was wetter than the untreated control. However, the applications did not improve the uneven distribution in the soil below the surface layer. The differences between lowest and highest soil water contents were even larger in the treated plot than the untreated plot. A more homogeneous wetting of the treated soil may be realised by using a surfactant that penetrates deeper into the soil profile, or by combining treatments with sprinkler irrigations to prevent drying of the soil below the critical soil water content, and thereby preventing the soil from becoming water repellent. This is in accordance to the statement of Moore (1981): ‘Once an area receives an effective wetting agent program and has a treated root-zone, surface applied rain and irrigation penetrates rapidly, wets and drains through thatch, and uniformly wets the entire profile’. The variability in actual water repellency was high over short distances at all depths in the untreated and treated dune sand, with WDPT values often varying between 6 h on all sampling days (Fig. 1, Fig. 2). Relatively small
L. W. Dekker et al.
differences in actual water repellency between the untreated and treated plot were determined for samples collected at depths of 0.07–0.19 m (Fig. 2). Applications of Primer 604 resulted in less persistent water repellency in the thatch layer and surface layer to a depth of 0.05 m (Fig. 1). Water content has a large effect on the actual water repellency of a soil. The critical soil water content introduced by Dekker and Ritsema (1994) appears not to be a sharp static threshold above which a soil is wettable and below which a soil is water repellent, but rather a transitional range value. This range of critical soil water contents for a certain depth has been introduced by Dekker et al. (2001) as the ‘transition zone’. Soil samples can be either wettable or water repellent within the transition zone, depending on the wetting history, sequence of weather conditions, etc. In the untreated plot of the dune sand studied the transition zone was assessed at depths of 0–0.025 and 0.025–0.05 m as being between soil water contents of 18–23 and 14–20 vol%, respectively. Applications of Primer604 lowered these transition zones to 12–16.5 and 8–20 vol%, respectively (Fig. 7). This implies that the surface layer in the treated soil can dry to lower water contents than in the untreated soil before water repellency is induced. Primer604 applications also increased distinctly the wetting rate of the field-moist samples from the thatch layer (Fig. 8). This may result in a more effective supply of water (irrigation), thereby inducing a better grass growth and a reduction in runoff. The spatial and temporal variability in persistence of potential water repellency after drying at 25◦ C was high at depths of 0–0.025 and 0.025–0.05 m in the treated and untreated plot (Fig. 9). More important, distinctly lower WDPT values were detected for the samples from the Primer604-treated plot compared with the untreated plot. Acknowledgments This study has been partly financed by Aquatrols Corporation of America, Paulsboro, NJ, USA 08066, and the Commission of the European Community under work program FAIR (ref. 4027). The project does not necessarily reflect the Commission’s views and in no way anticipates its future policy in this area. References Bauters TWJ, Steenhuis TS, DiCarlo DA, Nieber JL, Dekker LW, Ritsema CJ, Parlange J-Y, Haverkamp R (2000) Physics of water repellent soils. Journal of Hydrology 231–232, 233–243. doi: 10.1016/S0022-1694(00)00197-9 Bisdom EBA, Dekker LW, Schoute JFTh (1993) Water repellency of sieve fractions from sandy soils and relationships with organic material and soil structure. Geoderma 56, 105–118. doi: 10.1016/0016-7061(93)90103-R
Surfactant treatment and wettability
Australian Journal of Soil Research
Blackwell PS (2000) Management of water repellency in Australia, and risks associated with preferential flow, pesticide concentration and leaching. Journal of Hydrology 231–232, 384–395. doi: 10.1016/S0022-1694(00)00210-9 Cisar JL, Williams KE, Vivas HE, Haydu JJ (2000) The occurrence and alleviation by surfactants of soil-water repellency on sandbased turfgrass systems. Journal of Hydrology 231–232, 352–358. doi: 10.1016/S0022-1694(00)00207-9 DeBano LF (2000) Water repellency in soils: A historical overview. Journal of Hydrology 231–232, 4–32. doi: 10.1016/S00221694(00)00180-3 Dekker LW (1998) Moisture variability resulting from water repellency in Dutch soils. Doctoral thesis, Wageningen Agricultural University, The Netherlands. Dekker LW, Doerr SH, Oostindie K, Ziogas AK, Ritsema C (2001) Water repellency and critical soil water content in a dune sand. Soil Science Society of America Journal 65, 1667–1674. Dekker LW, Jungerius PD (1990) Water repellency in the dunes with special reference to the Netherlands. Catena Supplement 18, 173–183. Dekker LW, Oostindie K, Ritsema CJ (2000) Effects of surfactant treatments on the wettability of the surface layer and the wetting patterns in a water repellent dune sand with grass cover. Alterra Report 079, Alterra, Green World Research, Wageningen-UR, 75 pp. Dekker LW, Ritsema CJ (1994) How water moves in a water repellent sandy soil. 1. Potential and actual water repellency. Water Resources Research 30, 2507–2517. doi: 10.1029/94WR00749 Dekker LW, Ritsema CJ (1996) Variation in water content and wetting patterns in Dutch water repellent peaty clay and clayey peat soils. CATENA 28, 89–105. doi: 10.1016/S0341-8162(96)00047-1 Dekker LW, Ritsema CJ (2000) Wetting patterns and moisture variability in water repellent Dutch soils. Journal of Hydrology 231–232, 148–164. doi: 10.1016/S0022-1694(00)00191-8 Dekker LW, Ritsema CJ, Oostindie K, Boersma OH (1998) Effect of drying temperature on the severity of soil water repellency. Soil Science 163, 780–796. doi: 10.1097/00010694-199810000-00002 Doerr SH, Shakesby RA, Walsh RPD (2000) Soil water repellency: its causes, characteristics and hydro-geomorphological significance. Earth Science Reviews 51, 33–65. doi: 10.1016/S00128252(00)00011-8 Hallett PD, Ritz K, Wheatly RE (2001) Microbial derived water repellency in golf course soil. International Turfgrass Society Research Journal 9, 518–524. Hooghoudt SB (1950) Irreversibly desiccated peat, clayey peat, and peaty clay soils: the determination of the degree of reversibility. In ‘4th International Congress of Soil Science’. Amsterdam, 24 July–1 August 1950. Transactions Vol. II, pp. 31–34. (Hoitsema Brothers: Groningen, The Netherlands)
395
Jaramillo DF, Dekker LW, Ritsema CJ, Hendrickx JMH (2000) Occurrence of soil water repellency in arid and humid climates. Journal of Hydrology 231–232, 105–111. doi: 10.1016/S00221694(00)00187-6 King PM (1981) Comparison of methods for measuring severity of water repellence of sandy soils and assessment of some factors that affect its measurement. Australian Journal of Soil Research 19, 275–285. Kostka SJ (2000) Amelioration of water repellency in highly managed soils and the enhancement of turfgrass performance through the systematic application of surfactants. Journal of Hydrology 231–232, 359–370. doi: 10.1016/S0022-1694(00)00208-0 Kostka SJ, Cisar JL, Short JR, Mane S (1997) Evaluation of soil surfactants for the management of soil water repellency in turfgrass. International Turfgrass Society Research Journal 8, 485–494. Moore R (1981) Wetting agents: A tool for influencing water behavior in soil. Golf Course Management 49, 26–28. Nguyen HV, Nieber JL, Odura P, Ritsema CJ, Dekker LW, Steenhuis TS (1999) Modeling solute transport in a water repellent soil. Journal of Hydrology 215, 188–201. doi: 10.1016/S00221694(98)00270-4 Rieke PE (1981) Wetting agents: Applications vary for different soils. Golf Course Management 49, 27–30. Ritsema CJ, Dekker LW (1995) Distribution flow: A general process in the top layer of water repellent soils. Water Resources Research 31, 1187–1200. doi: 10.1029/94WR02979 Ritsema CJ, Dekker LW (1996) Water repellency and its role in forming preferred flow paths in soils. Australian Journal of Soil Research 34, 475–487. Ritsema CJ, Dekker LW, Hendrickx JMH, Hamminga W (1993) Preferential flow mechanism in a water repellent sandy soil. Water Resources Research 29, 2183–2193. doi: 10.1029/ 93WR00394 Ritsema CJ, Dekker LW, Nieber JL, Steenhuis TS (1998a) Modeling and field evidence of finger formation and finger recurrence in a water repellent sandy soil. Water Resources Research 34, 555–567. doi: 10.1029/97WR02407 Ritsema CJ, Nieber JL, Dekker LW, Steenhuis TS (1998b) Stable or unstable wetting fronts in water repellent soils -effect of antecedent soil moisture content. Soil and Tillage Research 47, 111–123. doi: 10.1016/S0167-1987(98)00082-8
Manuscript received 25 June 2004, accepted 7 January 2005
http://www.publish.csiro.au/journals/ajsr
Australian Journal of Soil Research, 2005, 43, 383–395
Effects of surfactant treatments on the wettability of a water repellent grass-covered dune sand Louis W. DekkerA,C , Klaas OostindieA , Stanley J. KostkaB , and Coen J. RitsemaA A Alterra,
Soil Science Center, Green World Research, PO Box 47, 6700 AA Wageningen, The Netherlands. B Aquatrols Corporation of America, Paulsboro, NJ 08066, USA. C Corresponding author. Email address: [email protected]
Abstract. The objective of this study was to evaluate the effectiveness of the surfactant formulation Primer604 for amelioration and management of soil water repellency in grass-covered dune sand. The soil is severely to extremely water repellent to a depth of >0.50 m during dry periods. Primer604 was applied 12 times between 22 April and 23 November 1999. During that period, soil samples were taken in the untreated and treated plot— 8 times in transects and 2 times in soil blocks. A total of 4950 samples were collected for assessment of the actual water repellency and for the spatial and temporal variability of the water content of the soil. Resistance to wetting was determined by measuring the wetting rate of field-moist samples. Measurements of water repellency revealed that applications of Primer604 resulted in less persistent water repellency in the surface layer to a depth of 0.05 m. No effects were observed deeper in the soil profile, likely due to adsorption of the surfactant material in the surface layer. In the surface layer (0–0.025 and 0.025–0.05 m), the critical soil water content below which the soil is actually water repellent in the field was lowered distinctly by the application of Primer604, potentially due to coating of water repellent particle surfaces by the surfactant. This suggests that the surface layer (0–0.05 m) in the Primer604-treated soil can dry to lower water contents than in the untreated soil before water repellency is induced. The thatch layer (0–0.025 m) of the treated soil was often found to have slightly higher water contents than of the untreated soil. The surfactant did not equalise the uneven moisture distribution in the soil below the surface layer (0–0.05 m). Primer604 applications increased the wetting rate of field-moist samples from the thatch layer. This may result in a more effective wetting of the root-zone during rain events or irrigation, and a reduction in runoff. Additional keywords: actual water repellency; critical soil water content; irregular wetting; preferential flow; water drop penetration time (WDPT) test; wetting rate.
Introduction The phenomenon of soil water repellency has been recognised in various parts of the world (DeBano 2000; Doerr et al. 2000; Jaramillo et al. 2000) including the Netherlands (Hooghoudt 1950; Dekker and Jungerius 1990; Dekker and Ritsema 1996, 2000) and has resulted in serious land-use problems in agriculture (Blackwell 2000) and an ongoing management problem on sand-based turfgrass systems (Cisar et al. 2000). It has been recognised for many years that soil water repellency is often a function of the type of organic matter incorporated in the soil, and that certain types of organic matter cause water repellency by several means (e.g. Doerr et al. 2000; Dekker et al. 2001; Hallett et al. 2001). Soil water repellency may dramatically affect fieldscale water and solute movement, and has often been underestimated (Bauters et al. 2000). Water repellency and © CSIRO 2005
its spatial variability have been shown to cause non-uniform wetting and preferential flow in many soils (Dekker and Ritsema 1994, 1996, 2000; Ritsema and Dekker 1996; Ritsema et al. 1998b). Based on experimental observations, Ritsema et al. (1993) proposed a conceptual model for water flow in water repellent sandy soils. According to this model, the initially uniform water infiltration is disrupted within the first few centimeters of the water repellent soil, causing water to move laterally towards microdepressions and regions with lower water repellency, where fingers are formed. Water is transported along these preferential flow pathways until decreasing water repellency in the soil at increasing depth causes divergence of the flow lines (Ritsema et al. 1998a; Nguyen et al. 1999). This process, occurring at the onset of an infiltration or leaching event, may drastically affect subsequent water movement (Ritsema and Dekker 1995). 10.1071/SR04090
0004-9573/05/030383
384
Australian Journal of Soil Research
Soil surfactants have been developed as a possible means for overcoming the problems caused by water repellent soils (Moore 1981; Rieke 1981; Kostka et al. 1997; Dekker et al. 2000; Kostka 2000). Wetting agents that have a strong affinity for the surface of hydrophobic soil particles and adsorb strongly at the soil surface will enhance infiltration rates at the soil surface interface. On the other hand, good water dispersion throughout the profile would require uniformity of penetration of the wetting agent in the profile. It is clear that a true test of the effectiveness of a soil wetting agent must include the assessment of the uniformity of distribution of the water in the soil, as well as the increase in infiltration rate and water content. The objective of our study was to evaluate the effectiveness of Primer604 for amelioration and management of soil water repellency in a pasture on a native dune sand. The present paper describes the influence of the surfactant in reducing the severity of soil water repellency and increasing the wetting rate, and its effect on the spatial variability in soil water content, water flow, and wetting patterns. Materials and methods Field-soil and treatment The wetting agent was applied to a plot (25 by 5 m) of a dune sand having a grass cover near Ouddorp in the south-western part of the Netherlands. An untreated adjacent plot was used for comparison. The soil consisted of fine sand with 3 m, and was classified as mesic Typic Psammaquent (Dekker 1998). Organic matter contents of 18 and 10% (wt/wt) were found at depths of 0–0.025 and 0.025–0.05 m, respectively. At depths of 0.05–0.075 m organic matter content was 2%, and from 0.075 m downwards approximately 0.5%. The soil was known to be severely to extremely water repellent to a depth of >0.50 m during dry periods (Dekker and Ritsema 1994; Dekker et al. 2001). Water repellency of this sandy soil is due to a coating of the sand grains with hydrophobic material and the presence of hydrophobic particulate organic matter (Bisdom et al. 1993). Primer604 (Aquatrols Corporation of America, Paulsboro, NJ, USA) treatments at a rate of 1.85 mL/m2 (volume solution 70 mL/m2 ) were applied with a Mesto Pico backpack-type sprayer 12 times during the period 22 April–23 November 1999. Soil sampling Between 22 April and 12 October 1999, the spatial and temporal variability of the volumetric soil water content was evaluated 8 times in vertical transects by intensive sampling of the treated and the untreated plots. To allow more detailed determination of the wetting patterns in the dune sand, soil blocks (0.25 by 0.75 by 0.19 m) were sampled in both plots on 25 October and 23 November 1999 (Dekker et al. 2000). The soil of transects and blocks was sampled at 6 depths (0–0.025, 0.025–0.05, 0.07–0.095, 0.095–0.12, 0.14–0.165, and 0.165–0.19 m), using steel cylinders with a diameter of 50 mm. In each transect and at each depth, 35 adjacent samples were taken over a distance of approximately 1.8 m. A total of 3150 samples were collected in the transects. In each of the 4 soil blocks, 75 samples were taken at each depth along a horizontal plane in a regular grid of 15 by 5 samples. A total of 1800 samples were collected from all blocks. The cylinders were pressed vertically into the soil, emptied into plastic bags, and re-used. Plastic bags were tightly sealed to minimise evaporation. The field-moist bagged soil was weighed. ‘Actual’ soil water repellency
L. W. Dekker et al.
was measured (Dekker and Ritsema 1994), and after drying in a fan-oven during 1 week at 25◦ C, ‘potential’ water repellency was measured. Samples were further dried at 105◦ C and weighed again to calculate the water content and dry bulk density of each sample. A total of 4950 samples were collected and measured in this way. Water Drop Penetration Time (WDPT) test The persistence or stability of water repellency of the soil samples was examined using the WDPT test (e.g. King 1981). Using a standard medicine dropper, 3 drops of distilled water were placed on the smoothed surface of a soil sample, and the time that elapsed until the drops infiltrated was determined. Soil water repellency of all 4950 samples were measured under controlled conditions at a constant temperature of 20◦ C and a relative air humidity of 50%. In general, a soil is considered to be water repellent if WDPT exceeds 5 s (Dekker 1998). We applied an index allowing a quantitative classification of the persistence of soil water repellency as described by Dekker and Jungerius (1990). Thus, 7 classes of repellency were distinguished, based on the time for water drops to infiltrate the soil: class 0, wettable, non-water repellent (infiltration within 5 s); class 1, slightly water repellent (5–60 s); class 2, strongly water repellent (60–600 s); class 3, severely water repellent (600 s–1 h); and extremely water repellent (>1 h), further subdivided into class 4 (1–3 h), class 5 (3–6 h), and class 6 (>6 h). Water repellency was measured on the field-moist samples (‘actual’ water repellency), and again after drying at 25◦ C. The severity of water repellency measured on dried soil samples, the so-called ‘potential’ water repellency, is considered to be the most appropriate parameter for comparing soils with respect to their sensivity to water repellency (Dekker and Ritsema 1994), because differences in water content are wiped out. We measured the ‘actual’ water repellency on the field-moist samples immediately after recording their wet weight. By measuring the water content of the samples, we could assess ‘critical soil water contents’ for the different depths of the intensively sampled transects and soil blocks. The soil is wettable above, and water repellent below, these values (Dekker and Ritsema 1994). Wetting rate measurements Resistance to wetting was determined several times by measuring the wetting rate of field-moist samples collected at depths of 0–0.025 m in the treated and untreated plots prior to surfactant applications. The samples were collected in steel cylinders (50 cm3 ) with a height of 25 mm and a diameter of about 50 mm. To measure wetting rate, these samples, within their steel cylinders, were subjected to a constant pressure head of −2.5 cm water applied at the bottom of the samples (Dekker et al. 1998). The experimental set-up was designed in such way that water content changes in 1.0 vol% increments were recorded automatically. All measurements were performed in a controlled environment laboratory with a constant temperature of 20◦ C and a relative humidity of 50%.
Results Actual water repellency The entire soil profile was wet at the beginning of the study on 22 April 1999. All 210 samples taken at depths between 0 and 0.19 m were wettable or non-water repellent, exhibiting WDPT values 6 h on 17 May 1999 (Fig. 2). A decrease in severity of actual water repellency occurred in the soil layers sampled below 0.12 m on almost all sampling dates (Fig. 2). The variability in actual water repellency was high in the soil over short distances at all depths in the untreated and treated plot, with WDPT values often varying between 6 h on all sampling days (Figs 1, 2). Relatively small differences in actual water repellency between the untreated and treated plot were determined for samples
collected at depths of 0.07 to 0.19 m (Fig. 2). The thatch layer (0–0.025 m depth) was found to be wettable for both plots on 22 April, 12 August, 12 and 25 October, and 23 November 1999. More wettable samples were recorded at depths of 0–0.025 m and 0.025–0.05 m for the Primer604-treated plot compared with the untreated plot on 1 June and 8 July 1999 (Fig. 1). Distinctly lower WDPT values were recorded for the samples taken in the thatch layer and in the surface layer at depths of 0.025–0.05 m in the treated plot compared with the untreated plot on 2 and 21 September 1999 (Fig. 1). The spatial and the temporal variability of the persistence of actual water repellency in transects of the untreated and treated plot were remarkable (Fig. 2). Within a horizontal distance of only 1.8 m, extreme water repellency was detected in dry soil areas, and wettable soil in preferential
Actual water repellency Untreated
Treated
0–0.025 m 17 May 1 Jun 8 Jul 12 Aug 2 Sep 21 Sep 12 Oct 25 Oct 23 Nov
0.025–0.05 m 17 May 1 Jun 8 Jul 12 Aug 2 Sep 21 Sep 12 Oct 25 Oct 23 Nov 0%
20%
40%
60%
80%
100%
385
0%
20%
40%
60%
80%
100%
WDPT 6h
600–3600 s
Fig. 1. Relative frequency of the persistence of actual water repellency of field-moist samples taken at depths of 0–0.025 and 0.025–0.05 m in the untreated and treated dune sand on 9 sampling days (n = 35 at both depths for the 17 May to 12 October 1999 transects, and n = 75 for the 25 October and 23 November 1999 soil blocks).
386
Australian Journal of Soil Research
L. W. Dekker et al.
Untreated
Treated 17 May
1 June
8 July
12 August
2 September
21 September
12 October
0%
20%
40%
60%
80%
100%
0%
20%
40%
60%
80%
100%
WDPT 6h
Fig. 2. Relative frequency of the persistence of actual water repellency of field-moist samples (n = 35) taken at 4 depths (top to bottom: 0.07–0.095, 0.095–0.12, 0.14–0.165, 0.165–0.19 m) in the untreated and treated dune sand on 7 sampling days.
Surfactant treatment and wettability
Australian Journal of Soil Research
flow paths, the fingers. Examples are the 2 September and 12 October 1999 transects of the untreated plot and the 1 June, 8 July, and 2 September 1999 transects of the treated plot (Fig. 2). The temporal variability in actual water repellency of the soil in both plots is evident when comparing the diagrams of the transects sampled on different days. Spatial variability in severity of actual water repellency is demonstrated by the contour plots of the 25 October 1999 soil blocks, with a size of only 0.75 by 0.25 by 0.19 m, or 0.036 m3 (Fig. 3). It is worthy of note that the actual water repellency at depths of 0.0825–0.1775 m in the treated plot was more severe and distributed over larger areas than in the untreated plot. We assume that the applications of Primer604 resulted
in a movement of the surfactant from the thatch layer into the preferential flow paths and thereby induced an enhancement of wetting of these flow paths. As a consequence, dry pockets in the treated soil were provided with less rainwater and were therefore more persistent than in the untreated soil. Although the severity of actual water repellency decreased during the autumn rains, extreme repellency with WDPT values >1 h still occurred in dry soil pockets of the soil block of 23 November 1999 (not shown here). In conclusion, the application of Primer604 resulted in a decrease in water repellency in the surface layer to a depth of 0.05 m, due to adsorption of the surfactant in this zone. Deeper in the profile no positive effects could be observed.
Untreated
Treated
Distance (m)
0
Distance (m)
0
0.75
0.75
Depth (m) 0.0125
0.0375
0.0825
0.1075
0.1525
0.1775
WDPT
6h
60–600 s
387
600–3600 s
1–3 h
Fig. 3. Contours of the persistence of actual water repellency in horizontal planes (0.25 by 0.75 m) at 6 depths in the untreated and treated 25 October 1999 soil blocks.
Australian Journal of Soil Research
L. W. Dekker et al.
Untreated
Treated
0–0.025 m
60 50
)
40 30 20 10 0
0.025–0.05 m 50 40 30 20 10 0
0.07–0.095 m 40 35 30 25 20
Soil water content (vol%)
388
15 10 5 0
0.095–0.12 m 30 25 20 15 10 5 0
0.14–0.165 m
18 16 14 12 10 8 6 4 2 0
0.165–0.19 m 16 14 12 10 8 6 4 2 0 31-Mar
20-May
9-Jul
28-Aug
17-Oct
6-Dec
31-Mar
20-May
9-Jul
28-Aug
17-Oct
6-Dec
Date
◦
Fig. 4. Minimum (–), mean ( ), and maximum (–) soil water contents of samples taken at 6 depths in the 8 transects (n = 35) and 2 soil blocks (n = 75) of the untreated and treated dune sand between 17 May and 23 November 1999.
Surfactant treatment and wettability
Australian Journal of Soil Research
Soil water content A high degree of spatial and temporal variability of the soil water content was often found in all layers sampled (Fig. 4). The highest water contents were found in the surface layers, which possessed higher organic matter contents and lower dry bulk densities than the deeper layers. Relatively high soil water contents were established at depths between 0 and 0.05 m in April (not shown), October, and November; however, spatial variability remained high. For instance, the soil water content at depths of 0–0.025 m varied between 31.4 and 69.3 vol%, and at depths of 0.025–0.05 m between 26.1 and 57.5 vol% at the start of the experiment, on 22 April 1999. At the end of the experiment, on 23 November 1999, the water content of the thatch layer in the untreated plot varied between 29.9 and 47.5 vol% and in the treated plot between 33.7 and 54.3 vol% (Fig. 4). The mean soil water content in the thatch layer was often slightly higher in the treated plot than the untreated plot (Fig. 5). For example, in the treated plot, the mean soil water content was 43.8 vol% at depths of 0–0.025 m on 25 October 1999, compared with 40.8 vol% in the untreated plot. The plot treated with Primer604 applications exhibited slightly
higher mean soil water contents at all depths and in most transects compared with the untreated plot (Fig. 5). The diagrams of these transects also show clearly the wide variability which often existed between the soil water contents in the surface layer and in the soil at depths of 0.095–0.19 m. Relatively dry transects, with small variations in water content, were sampled on 17 May and 2 September 1999 (Fig. 4). Large differences in soil water content were found at depths of 0.07–0.19 m in the untreated and treated plots between 17 May and 23 November 1999 (Fig. 4). In this zone, wet fingers and dry soil areas were evident in the soil profile in both the untreated and treated plots. Surfactant treatment did not prevent irregular wetting of the water repellent subsoil. The variation in water content in the treated transects and soil blocks was also high, as demonstrated by the diagrams on the right-hand side of Fig. 4. Irregular wetting patterns were also encoutered at depths of 0.0825–0.1775 m in the soil blocks of the untreated and treated plots on 25 October 1999 (Fig. 6). Between 1 October and 25 October, precipitation amounted to 122 mm, yet soil
0.025–0.05 m
0–0.025 m 40
50 40
30
30 20 20 10
10
0
0.095–0.12 m
0.07–0.095 m 9
18 16 14 12 10 8 6 4 2 0
8 7 6 5 4 3 2 1 0
0.165–0.19 m
0.14–0.165 m 7
8
6
7
5
6 5
4
4
3
3
2
2
1
1 23 -N ov
25 -O ct
ov -N 23
12 -O ct
ct -O 25
-S ep
ct -O 12
21
ep -S 21
ug
p Se 2-
-A
ug -A 12
12
l Ju 8-
8Ju l
n Ju 1-
17 -M ay
ay
1Ju n
0
0
2Se p
Soil water content (vol%)
0
-M 17
389
Date Untreated
Treated
Fig. 5. Mean soil water contents at 6 depths in the untreated and treated dune sand between 17 May and 23 November 1999.
390
Australian Journal of Soil Research
L. W. Dekker et al.
was locally dry at depths of 0.0825–0.1775 m. Water content in these regions was generally 23 vol%, were determined as wettable (Fig. 7, upper left-hand diagram). All samples with a soil water content 25
Fig. 6. Contours of the volumetric soil water content in horizontal planes (0.25 by 0.75 m) at 6 depths in the untreated and treated dune sand on 25 October 1999.
Surfactant treatment and wettability
Australian Journal of Soil Research
critical soil water content of the thatch layer of the untreated plot is variable and ranges between 18 and 23 vol%, most likely depending on the wetting history of the soil, weather sequence, etc. The critical soil water content of the soil in the untreated plot at depths of 0.025–0.05 m was found to be between 14 and 19 vol% (Fig. 7, lower left-hand diagram). Although there were large differences in severity of actual water repellency at specific soil moisture contents, there was a distinct increase in severity with decreasing soil water contents, as shown in the diagrams of Fig. 7. Treatment with Primer604 caused a significant decrease in the critical soil water content of the thatch layer, as can be seen in the upper right-hand diagram of Fig. 7. The transition zone varied in this case between 12 and 16.5 vol%, compared with variation between 18 and 23 vol% in the untreated plot. For instance, all soil samples with a water content of 17 vol% were determined as wettable in the Primer604-treated plot, whereas all samples with this water content in the untreated plot exhibited slight to extreme repellency. Also at depths of 0.025–0.05 m there was a slight shift in the critical soil water content. Soil samples with water contents of 8–14 vol% were water repellent in the untreated plot, whereas in the Primer604-treated plot a number of samples with these water contents were still wettable.
Untreated
In conclusion, the critical soil water content in the thatch layer was lowered distinctly by the applications of Primer604. This means that the surfactant-treated soil dried to lower water contents than the untreated soil before water repellency was induced. Resistance to wetting of field-moist samples An instantly high wetting rate of the surface layer is important for the effective infiltration of rain and irrigation water as well as for prevention of erosion and runoff. Measurements with the wetting rate device showed that the water uptake of the thatch layer was generally more rapid in the beginning for samples from the plot treated with Primer604 than for samples from the untreated plot. Differences in instant wetting rate were observed between thatch layer samples from the untreated and treated plot on 5 of the 6 sampling dates (Fig. 8). The uptake of water (in mm) gives an indication of the amount of rainwater that can be absorbed readily. It is evident from the diagrams that water infiltrated more effectively into the thatch layer of the treated plot than the control. Initial soil water content of the samples played an important role for the wetting rate during the first hour (Fig. 8). Especially, samples from the untreated plot with soil water contents below the critical soil water content exhibited less affinity for water absorption.
Treated
0–0.025 m
6 5 4 3 2
WDPT class
1 0
0.025–0.05 m 6 5 4 3 2 1 0 0
20
40
391
60
0
20
40
60
Soil water content (vol%)
Fig. 7. Relationship between soil water content and persistence of actual water repellency at depths of 0–0.025 and 0.025–0.05 m in the untreated and treated dune sand. The transition zone, with critical soil water contents, is indicated with a grey tone.
392
Australian Journal of Soil Research
L. W. Dekker et al.
Soil water content 80
20 16 June
17 May 60
15
40
10
20
5
0
0
80
20 27 July
(vol%)
8 July 60
15
40
10
20
5
0
0
80
20 2 September
21 September
60
15
40
10
20
5
0
0 0
10
20
30
40
50
60
0
10
20
30
40
50
60
Time (min) Untreated
Treated
Fig. 8. Increase in water content (vol%) of field-moist samples taken on 6 sampling days in the untreated and treated thatch layer (0–0.025 m) during 1 h of wetting. The initial water content and water uptake are also indicated in mm.
Note that the wetting rate of the thatch layer did not increase after the first Primer 604 application, as shown by the diagram of 17 May 1999, but a distinct increase was observed after additional applications, as shown by the diagrams of 16 June–21 September 1999 (Fig. 8). In conclusion, the application of Primer604 increased the wetting rate of the thatch layer, which results in a more effective wetting of the root-zone during rain events and/or irrigation events, thereby decreasing runoff. Persistence of potential water repellency The persistence of potential water repellency of samples taken at depths of 0–0.025 and 0.025–0.05 m both in the transects and in the soil blocks of the untreated and treated plots was measured with the WDPT test after drying at 25◦ C. All field-moist samples from the 22 April 1999 transect
were wettable, but the WDPT varied between 60–600 s (class 2) and 3–6 h (class 5) after drying at 25◦ C (data not shown). Differences in potential water repellency (samples dried at 25◦ C prior to WDPT) occurred between samples taken at the same depths but also between samples taken on different sampling dates (Fig. 9). Notable differences in persistence of water repellency were found for both the 0–0.025 and 0.025–0.05 m depths in the untreated plot between the 2 September and 12 October 1999 transects. All samples from 2 September 1999 exhibited extreme water repellency with WDPTs between 1–3 and >6 h, whereas the 12 October 1999 samples showed strong to severe water repellency with WDPTs varying between 60 s and 1 h (Fig. 9). High spatial and temporal variability in potential water repellency were also determined for the samples taken at depths of 0.025–0.05 m in the
Surfactant treatment and wettability
Australian Journal of Soil Research
393
Potential water repellency (25°C) Untreated
Treated
0–0.025 m 17 May 1 Jun 8 Jul 12 Aug 2 Sep 21 Sep 12 Oct 25 Oct 23 Nov
0.025–0.05 m 17 May 1 Jun 8 Jul 12 Aug 2 Sep 21 Sep 12 Oct 25 Oct 23 Nov 0% WDPT
20%
40%
60%
80%
100%
0%
6h
20%
40%
60%
80%
100%
600–3600 s
Fig. 9. Relative frequency of the persistence of potential water repellency of samples dried at 25◦ C, and taken at depths of 0–0.025 and 0.025–0.05 m in the untreated and treated dune sand on 9 sampling days (for n see Fig. 1).
treated plot between 17 May and 23 November 1999, as demonstrated in Fig. 9. Because all samples were dry, the differences in water repellency must be due to differences in initial water content of the samples and a process of initiating water repellency in the field. The persistence of potential water repellency of the samples dried at 25◦ C is clearly negatively related to the initial soil water content and positively to the persistence of the actual water repellency of the samples. For example, the relatively dry and severe to extreme actual water repellency of the 17 May and 2 and 21 September 1999 transects (Fig. 1) resulted in locally extreme potential water repellency (Fig. 9).
Distinctly lower WDPT values after drying at 25◦ C were detected for the samples taken at both depths in the Primer604-treated plot between 1 June and 23 November 1999 compared with the untreated plot (Fig. 9). A majority of the samples taken from the thatch layer of the treated plot between 8 July and 25 October 1999 exhibited only slight water repellency after drying at 25◦ C, whereas most samples from the untreated plot exhibited severe to extreme water repellency (Fig. 9). In most cases the persistence of potential water repellency of the samples was distinctly higher than the actual water repellency after the samples had been dried at 25◦ C.
394
Australian Journal of Soil Research
However, it is worthy of note that the potential water repellency of samples on some days was less severe than the actual water repellency of field-moist samples on other days, thus underestimating the maximal persistence of water repellency that can occur in the field (see Fig. 1 and Fig. 9). This indicates that processes which are taking place in the field during dry weather cannot be artificially generated during drying in a laboratory oven over a time span of several days. Regardless of how water repellency was measured in soil samples, surfactant treatment generally shifted water repellency classes (actual or potential) to more wettable classes. Surfactant-induced shifts in water repellency classes were most evident in potential water repellency results. In conclusion, the spatial and temporal variability in persistence of water repellency after drying at 25◦ C was high at both depths for the treated and untreated plot. But more important, distinctly lower WDPT values were detected for samples at depths of 0–0.025 and 0.025–0.05 m from the Primer604-treated plot than the untreated plot. Discussion and conclusions Water repellency of soils may dramatically affect water and solute movement, due to non-uniform wetting and forming of preferential flow paths, so-called fingers, as also often occur in the dune sand studied. Soil wetting agents have been developed as a possible means for overcoming the problems caused by water repellency (e.g. Kostka et al. 1997; Cisar et al. 2000). It is evident that a test of the effectiveness of a soil wetting agent must include the assessment of the uniformity of distribution of the water in the soil. Soil moisture measurements in the untreated plot and in the treated plot with applications of Primer604 revealed that after rain events the thatch layer in the treated plot was wetter than the untreated control. However, the applications did not improve the uneven distribution in the soil below the surface layer. The differences between lowest and highest soil water contents were even larger in the treated plot than the untreated plot. A more homogeneous wetting of the treated soil may be realised by using a surfactant that penetrates deeper into the soil profile, or by combining treatments with sprinkler irrigations to prevent drying of the soil below the critical soil water content, and thereby preventing the soil from becoming water repellent. This is in accordance to the statement of Moore (1981): ‘Once an area receives an effective wetting agent program and has a treated root-zone, surface applied rain and irrigation penetrates rapidly, wets and drains through thatch, and uniformly wets the entire profile’. The variability in actual water repellency was high over short distances at all depths in the untreated and treated dune sand, with WDPT values often varying between 6 h on all sampling days (Fig. 1, Fig. 2). Relatively small
L. W. Dekker et al.
differences in actual water repellency between the untreated and treated plot were determined for samples collected at depths of 0.07–0.19 m (Fig. 2). Applications of Primer 604 resulted in less persistent water repellency in the thatch layer and surface layer to a depth of 0.05 m (Fig. 1). Water content has a large effect on the actual water repellency of a soil. The critical soil water content introduced by Dekker and Ritsema (1994) appears not to be a sharp static threshold above which a soil is wettable and below which a soil is water repellent, but rather a transitional range value. This range of critical soil water contents for a certain depth has been introduced by Dekker et al. (2001) as the ‘transition zone’. Soil samples can be either wettable or water repellent within the transition zone, depending on the wetting history, sequence of weather conditions, etc. In the untreated plot of the dune sand studied the transition zone was assessed at depths of 0–0.025 and 0.025–0.05 m as being between soil water contents of 18–23 and 14–20 vol%, respectively. Applications of Primer604 lowered these transition zones to 12–16.5 and 8–20 vol%, respectively (Fig. 7). This implies that the surface layer in the treated soil can dry to lower water contents than in the untreated soil before water repellency is induced. Primer604 applications also increased distinctly the wetting rate of the field-moist samples from the thatch layer (Fig. 8). This may result in a more effective supply of water (irrigation), thereby inducing a better grass growth and a reduction in runoff. The spatial and temporal variability in persistence of potential water repellency after drying at 25◦ C was high at depths of 0–0.025 and 0.025–0.05 m in the treated and untreated plot (Fig. 9). More important, distinctly lower WDPT values were detected for the samples from the Primer604-treated plot compared with the untreated plot. Acknowledgments This study has been partly financed by Aquatrols Corporation of America, Paulsboro, NJ, USA 08066, and the Commission of the European Community under work program FAIR (ref. 4027). The project does not necessarily reflect the Commission’s views and in no way anticipates its future policy in this area. References Bauters TWJ, Steenhuis TS, DiCarlo DA, Nieber JL, Dekker LW, Ritsema CJ, Parlange J-Y, Haverkamp R (2000) Physics of water repellent soils. Journal of Hydrology 231–232, 233–243. doi: 10.1016/S0022-1694(00)00197-9 Bisdom EBA, Dekker LW, Schoute JFTh (1993) Water repellency of sieve fractions from sandy soils and relationships with organic material and soil structure. Geoderma 56, 105–118. doi: 10.1016/0016-7061(93)90103-R
Surfactant treatment and wettability
Australian Journal of Soil Research
Blackwell PS (2000) Management of water repellency in Australia, and risks associated with preferential flow, pesticide concentration and leaching. Journal of Hydrology 231–232, 384–395. doi: 10.1016/S0022-1694(00)00210-9 Cisar JL, Williams KE, Vivas HE, Haydu JJ (2000) The occurrence and alleviation by surfactants of soil-water repellency on sandbased turfgrass systems. Journal of Hydrology 231–232, 352–358. doi: 10.1016/S0022-1694(00)00207-9 DeBano LF (2000) Water repellency in soils: A historical overview. Journal of Hydrology 231–232, 4–32. doi: 10.1016/S00221694(00)00180-3 Dekker LW (1998) Moisture variability resulting from water repellency in Dutch soils. Doctoral thesis, Wageningen Agricultural University, The Netherlands. Dekker LW, Doerr SH, Oostindie K, Ziogas AK, Ritsema C (2001) Water repellency and critical soil water content in a dune sand. Soil Science Society of America Journal 65, 1667–1674. Dekker LW, Jungerius PD (1990) Water repellency in the dunes with special reference to the Netherlands. Catena Supplement 18, 173–183. Dekker LW, Oostindie K, Ritsema CJ (2000) Effects of surfactant treatments on the wettability of the surface layer and the wetting patterns in a water repellent dune sand with grass cover. Alterra Report 079, Alterra, Green World Research, Wageningen-UR, 75 pp. Dekker LW, Ritsema CJ (1994) How water moves in a water repellent sandy soil. 1. Potential and actual water repellency. Water Resources Research 30, 2507–2517. doi: 10.1029/94WR00749 Dekker LW, Ritsema CJ (1996) Variation in water content and wetting patterns in Dutch water repellent peaty clay and clayey peat soils. CATENA 28, 89–105. doi: 10.1016/S0341-8162(96)00047-1 Dekker LW, Ritsema CJ (2000) Wetting patterns and moisture variability in water repellent Dutch soils. Journal of Hydrology 231–232, 148–164. doi: 10.1016/S0022-1694(00)00191-8 Dekker LW, Ritsema CJ, Oostindie K, Boersma OH (1998) Effect of drying temperature on the severity of soil water repellency. Soil Science 163, 780–796. doi: 10.1097/00010694-199810000-00002 Doerr SH, Shakesby RA, Walsh RPD (2000) Soil water repellency: its causes, characteristics and hydro-geomorphological significance. Earth Science Reviews 51, 33–65. doi: 10.1016/S00128252(00)00011-8 Hallett PD, Ritz K, Wheatly RE (2001) Microbial derived water repellency in golf course soil. International Turfgrass Society Research Journal 9, 518–524. Hooghoudt SB (1950) Irreversibly desiccated peat, clayey peat, and peaty clay soils: the determination of the degree of reversibility. In ‘4th International Congress of Soil Science’. Amsterdam, 24 July–1 August 1950. Transactions Vol. II, pp. 31–34. (Hoitsema Brothers: Groningen, The Netherlands)
395
Jaramillo DF, Dekker LW, Ritsema CJ, Hendrickx JMH (2000) Occurrence of soil water repellency in arid and humid climates. Journal of Hydrology 231–232, 105–111. doi: 10.1016/S00221694(00)00187-6 King PM (1981) Comparison of methods for measuring severity of water repellence of sandy soils and assessment of some factors that affect its measurement. Australian Journal of Soil Research 19, 275–285. Kostka SJ (2000) Amelioration of water repellency in highly managed soils and the enhancement of turfgrass performance through the systematic application of surfactants. Journal of Hydrology 231–232, 359–370. doi: 10.1016/S0022-1694(00)00208-0 Kostka SJ, Cisar JL, Short JR, Mane S (1997) Evaluation of soil surfactants for the management of soil water repellency in turfgrass. International Turfgrass Society Research Journal 8, 485–494. Moore R (1981) Wetting agents: A tool for influencing water behavior in soil. Golf Course Management 49, 26–28. Nguyen HV, Nieber JL, Odura P, Ritsema CJ, Dekker LW, Steenhuis TS (1999) Modeling solute transport in a water repellent soil. Journal of Hydrology 215, 188–201. doi: 10.1016/S00221694(98)00270-4 Rieke PE (1981) Wetting agents: Applications vary for different soils. Golf Course Management 49, 27–30. Ritsema CJ, Dekker LW (1995) Distribution flow: A general process in the top layer of water repellent soils. Water Resources Research 31, 1187–1200. doi: 10.1029/94WR02979 Ritsema CJ, Dekker LW (1996) Water repellency and its role in forming preferred flow paths in soils. Australian Journal of Soil Research 34, 475–487. Ritsema CJ, Dekker LW, Hendrickx JMH, Hamminga W (1993) Preferential flow mechanism in a water repellent sandy soil. Water Resources Research 29, 2183–2193. doi: 10.1029/ 93WR00394 Ritsema CJ, Dekker LW, Nieber JL, Steenhuis TS (1998a) Modeling and field evidence of finger formation and finger recurrence in a water repellent sandy soil. Water Resources Research 34, 555–567. doi: 10.1029/97WR02407 Ritsema CJ, Nieber JL, Dekker LW, Steenhuis TS (1998b) Stable or unstable wetting fronts in water repellent soils -effect of antecedent soil moisture content. Soil and Tillage Research 47, 111–123. doi: 10.1016/S0167-1987(98)00082-8
Manuscript received 25 June 2004, accepted 7 January 2005
http://www.publish.csiro.au/journals/ajsr
