Published in Soil Sci. Soc. Am. J. 68:1982-1991 (2004).
© 2004 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
Division S-6—Soil & Water Management & Conservation
Spectral Reflectance Properties of Crusted Soils under Solar Illumination
G. Eshela,*,
G. J. Levyb and
M. J. Singera
a Dep. of Land, Air, and Water Resources, Univ. of California, Davis, CA 95616
b Institute of Soil, Water, and Environmental Sciences, Agricultural Research Organization (ARO), The Volcani Center, P.O. Box 6, Bet Dagan 50250, Israel
* Corresponding author (geshel{at}ucdavis.edu)
 |
ABSTRACT
|
|---|
Reflectance of soil crusts has been studied under artificial light but not under solar illumination. Our objectives were to (i) compare reflectance from crusted and noncrusted soil surfaces under solar illumination, and (ii) explore the relationship between crust permeability and spectral signature. Two California soils were studied, Capay (fine, smectitic, thermic Typic Haploxererts) and Reiff (coarse-loamy, mixed, superactive, nonacid, thermic Mollic Xerofluvents), with samples for the latter taken from plots under organic (ReiffOM) and conventional (ReiffCM) management. A laboratory rainfall simulator was used to form crusts which were sampled at different stages of development and permeability. A portable spectroradiometer was used to collect spectral data in direct sunlight for wet and dry crusted samples. Baseline spectra (albedo) of the dry samples were, in most cases, higher than spectra from the corresponding moist samples. Crusted samples exhibited higher baseline spectra compared with the noncrusted samples. The absorption feature at 
1400 nm (related to lattice OH of montmorillonite) suggested accumulation of clay in the crust of the ReiffCM, and clay depletion from the crusts of the Capay and ReiffOM. An inverse linear relationship existed between reflectance at many wavelengths and crust permeability at different stages of crust development. The reflectance at 1700 nm provided highly significant correlations with infiltration rate (IR) for all three soils, and reflectance at 2130 nm also had a highly significant correlation for the Capay. Our data suggest that spectral measurements of soil surfaces under solar illumination can differentiate between crusted and noncrusted surfaces and assist in evaluating the degree of crust development.
Abbreviations: IFOV, instrument field of view • IR, infiltration rate
|
INTRODUCTION
|
|---|
REFLECTANCE OF SOILS across the entire spectral region of solar illumination (400–2400 nm) carries much information about the soil (Baumgardner et al., 1985; Ben-Dor et al., 1998). For instance, information regarding the nature and chemical composition of soils can be derived from measurements of their spectral reflectance (Stoner et al., 1980). Reflectance data have been successfully used for prediction of numerous soil properties, such as soil moisture and organic matter content, in laboratory studies (Dalal and Henry, 1986; Ben-Dor and Banin, 1995).
It was suggested nearly two decades ago that soil reflectance measurements could assist in measuring quantitative changes in soil surface conditions (Baumgardner et al., 1985). Crust formation on the surface of bare soils exposed to rain or overhead sprinkler irrigation is one condition that may have significant adverse effects on soil behavior, such as inducing runoff and soil erosion. Formation of such surface crusts is a common phenomenon in many soils, particularly in arid and semiarid regions (Singer, 1991; Singer and Warrington, 1992). Surface crusts are thin (<2 mm) layers characterized by greater density, higher shear strength, finer pores, and lower saturated hydraulic conductivity than the bulk soil below (Bradford et al., 1987). Crust formation is caused by two mechanisms that act simultaneously and enhance each other: (i) a physical breakdown of soil aggregates caused by fast wetting of dry aggregates and the mechanical impact of water drops; and (ii) a physicochemical dispersion of clay particles which migrate and clog the conducting pores immediately beneath the surface (McIntyre 1958; Agassi et al., 1981). The formation of a crust is commonly characterized by decreased soil IR; the more developed the crust, the lower the IR (Shainberg and Levy, 1995, and references cited therein).
Studies of crust micromorphology have revealed that soil crusts may vary in their structure, depending on soil properties and conditions prevailing in the soil during crust formation. Under conditions that enhance clay dispersion (i.e., soil with high sodicity, or rain water without electrolytes = distilled water) the crust consists of two distinct layers (McIntyre, 1958; Gal et al., 1984; Onofiok and Singer, 1984): (i) an upper layer (0.1–0.25 mm) of high porosity composed predominantly of individual sand- and silt-size grains, and (ii) a deeper layer of considerably lower porosity composed of accumulated fine material that was washed out from the upper layer, termed washed-in zone. Under conditions where clay dispersion was prevented, only a thin skin layer at the soil surface comprised mostly of compacted fine particles was noted with no evidence of accumulation of fine particles in the washed-in zone (Chen et al., 1980; Gal et al., 1984). Levy et al. (1988) noted that the structure of crusted surfaces is heterogeneous, comprising of small mounds protruding from relatively smooth plains. Surface and cross-section SEM micrographs showed that structure of the mounds was relatively unaffected by raindrop impact. In the plains, a clearly compacted layer with primary particles partly stripped of clay was visible. The differences in structure between mounds and plains explained the significantly higher permeability measured in the mounds compared with that for the plains (Levy et al., 1988).
Spectral changes due to soil crusting have received little attention. Recently, Goldshleger et al. (2001) have noted for three Israeli soils that reflectance values in the shortwave infrared (1100–2500 nm) from crusted surfaces were significantly higher than those from noncrusted soils. A further in-depth study of the same soils (Ben-Dor et al., 2003), has shown that (i) the magnitude of the difference in reflectance values between crusted and noncrusted conditions among the soils tested depended on soil texture, (ii) differences in crust micromorphology could be detected from their reflectance spectra, and (iii) a significant relationship existed between crust permeability and its reflectance spectra.
Most measurements of the spectral responses of soils have been performed in the laboratory under artificial illumination. A notable exception is the study of Stoner et al. (1980) that demonstrated that for wavelengths in the range of 520 to 1750 nm, reflectance under solar illumination for two moist soils was directly proportional (about 1.5 times higher) to the reflectance obtained under laboratory conditions. The objectives of our study were to (i) compare reflectance measured under solar illumination from crusted surfaces to that from noncrusted ones, and (ii) evaluate whether changes in the spectral response of a crust at different stages of its development correlate with the corresponding IRs of the crust.
|
MATERIALS AND METHODS
|
|---|
Soils
Two California soils, the Reiff and Capay series, were sampled on the UC Davis campus. Two Reiff series samples were collected, one that had been under purely organic management for 12 yr (ReiffOM) and the other from a nearby plot that had been under conventional management for the same period (ReiffCM). The Capay series soil was sampled from a field that has been under conventional management for many years. The A horizon of each soil from the soil surface to 200 mm below the surface was sampled from each site. The samples were dry at the time of collection, and were sieved in the field to recover <4-mm aggregates (a size fraction that generally represents the prepared seed bed in these soils). After collection, the soils were stored in closed containers. A few characteristics of each soil are shown in Table 1.
View this table:
[in this window]
[in a new window]
|
Table 1. Particle size distribution, cation exchange capacity, organic matter content, CaCO3 content, and exchangeable sodium percentage of the soils studied.
|
|
Rainfall Simulation
Soil was loosely packed into 185-mm-diam. by 50-mm-deep columns by filling each column to the rim and gently tapping its side to settle the soil. Each column had a solid bottom with a central fitting where water could drain. Soil in each column was first saturated using zero-energy distilled-water rain applied as a mist by four nozzles (Spraying Systems Company, Wheaton, IL), placed at a 0.9-m height. Water began to flow through each column after approximately 16 min. To ensure that the soil was fully saturated before energy rain was applied, application of zero-energy rain continued for approximately 30 min after water began to flow out of the columns.
After completion of the saturation phase, the columns were transferred to the energy-carrying rain simulator which was immediately adjacent to the zero-energy one, and placed at a slope of 5% to promote runoff and reduce ponding. The columns were exposed to 3.2-mm-diam. distilled-water drops that fell from a height of 1 m through 18-gage hypodermic needles. Water was delivered to hypodermic needles mounted in a chamber by a variable speed electric pump. Raindrops were applied at a 70 mm h–1 rate as measured before and after each experiment. Different rainfall durations were applied to subject the soil surface to different levels of energy, and thus form crusts of different permeability. Two sets of 10 columns were packed for each soil. Nine columns from each set were placed randomly in the rainfall simulator. The columns were removed from the rainfall simulator, one at a time at predetermined time intervals to obtain crusts from nine different rainfall durations. The additional column was not subjected to rain with energy and has been designated the zero-rainfall-energy (noncrusted) sample. The IR was periodically determined during the entire rain event by measuring the volume of water that drained from the central fitting during a known period of time. For each column, immediately after the column was removed from the rainfall simulator it was taken outdoors for collection of the spectral data.
Spectral Reflectance Measurements
An ASD Field Spec FR (Analytical Spectral Devices, Boulder, CO, model 6100) portable spectroradiometer with a spectral range of 400 to 2400 nm, fitted with an 18-degree foroptic was used for the spectral data collection. The spectroradiometer's fiber optic was positioned 45 cm above and at the nadir to each sample to provide a 14.5-cm-diam. instrument field of view (IFOV).
Digital images of each sample were collected using a Nikon (Melville, NY) Cool-Pix digital camera mounted with the fiber optic (Fig. 1) . Subsequent measurements on the dry samples were made with the same instrument and configuration.
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 1. Schematic presentation of the reflectance measurement apparatus and arrangement for collecting spectral data.
|
|
All measurements were made in direct sunlight between the hours 1000 and 1500. These measurements were made close to the summer solstice in Davis, CA, when the sun is at its maximum height in the sky. Every day was cloud free and conditions were uniform for all of the digital data collection.
Spectral data for all samples were collected in radiance mode. A reference scan of a 25- by 25-cm diffuse reflectance panel (Spectralon, Labsphere, Inc., North Sutton, NH) and six sample scans were collected for each soil and energy sample across the range of 450 to 2400 nm in 1-nm steps. Initial postprocessing reflectance percentages were calculated using the ratio of mean sample radiance to reference panel radiance.
Two sets of spectral data were collected for every column: (i) immediately after completion of the rainfall simulation, and (ii) after 48 h of oven drying at 40°C. In cases where 48 h of drying were proven insufficient to fully dry the surfaces, the columns were allowed to dry for an additional period of 24 h to reach 72 h of drying in total. A digital image representing the IFOV of the spectral scan was collected at the same time that spectral response was measured for each surface at each moisture content. A small soil sample (about 5 g) of the crust was taken near the perimeter of every column after each spectral measurement for determination of soil moisture content. It is important to note that because of technical problems, not all samples were completely dry after 72 h in the oven. To reduce the interference of water content with the reflectance measurements, only the 10 driest samples for each soil studied were selected for the dry reflectance measurements.
|
RESULTS AND DISCUSSION
|
|---|
Surface images of dry noncrusted (zero energy) and crusted (1044 J m–2 cumulative rain energy) samples from the three soils illustrate the differences in the untreated and treated surfaces (Fig. 2)
. Surfaces of the noncrusted samples were rough compared with the crusted samples. In the Capay soil, the rough surface was composed mainly of macroaggregates; in the ReiffCM and ReiffOM soils, part of the surface was occupied by aggregates (macro- and micro-) and part by sand-size material (Fig. 2). When crusted, the surface of the Capay soil was smoother than the noncrusted one, and showed some cracks that developed during the drying phase. In the Reiff soils, the crusted surfaces showed a limited degree of cracking, and they seemed smoother than the crust in the Capay soil (Fig. 2). Levy et al. (1994) demonstrated that the degree of crust smoothness was inversely related to soil clay content. The lower clay content in the Reiff soil compared with the Capay soil may explain the observed smoother surfaces in the crusted Reiff soils (Table 1).
View larger version (88K):
[in this window]
[in a new window]
|
Fig. 2. Images of dry (1) noncrusted and (2) crusted surfaces of (a) Capay, (b) ReiffCM, and (c) ReiffOM soils.
|
|
The crusted surfaces had a lighter color than the noncrusted surfaces (Fig. 2). The change in color suggests that the composition of the particles at the soil surface changed during the crust formation process. In all three soils, light-color spots of naked silt and sand grains (Gal et al., 1984; Onofiok and Singer, 1984), separated by darker areas were observed. The darker areas were considered to represent regions of compacted fine material (Chen et al., 1980; Levy et al., 1988) with a limited depletion of clay.
Soil Spectral Reflectance
Field measurements of spectral curves do not contain data in the 1850- to 1950-nm region because in this region solar illumination is almost completely absorbed (Stoner et al., 1980). Therefore, only data in the 400- to 1850-nm and 1950- to 2400-nm ranges, that provided useful information, were used in our analyses (Fig. 3)
. Soil spectral response is comprised of a spectral baseline (albedo) and absorption features differing in position, shape, and intensity. Changes in the spectral baseline are associated with physical characteristics of the soil surface (aggregate size and soil moisture). Changes in the absorption features depend on the chemical composition of the soil (e.g., OH in clay lattice, iron oxide, carbonates, organic matter). In the range of the wavelength used in our study (400–1750 and 1950–2400 nm), some absorption features assigned to OH in adsorbed water can be found in the soil reflectance curves, including a strong absorption band at 1450 nm and occasionally weaker absorption bands at 970 and 1200 nm (Baumgardner et al., 1985). In addition, soils containing smectitic minerals may exhibit absorption bands at around 1300 to 1400 and 2200 to 2500 nm. These absorption bands are assigned to OH groups in the clay mineral lattice because they are the only groups among the various elements of clay minerals that provide a spectra across the entire spectral region of solar illumination (Ben-Dor, 2002).
Moist Samples
Reflectance measurements of the moist samples were made immediately after the completion of the rainfall simulation, thus moisture content in these samples was near saturation (42.4, 36.1, and 37.2% for Capay, ReiffCM, and ReiffOM, respectively). In all three soils, the moist crusted and noncrusted samples clearly showed a lower spectral baseline than their corresponding dry equivalents (Fig. 3). This observation was in agreement with previously published data (e.g., Hoffer and Johannsen, 1969; Planet, 1970; Lobell and Asner, 2002). It has been suggested that the primary reason for the lower reflectance of moist soil is the decrease in the relative refractive index of the medium surrounding the soil particles as a result of water replacing air (Twomey et al., 1986; Ishida et al., 1991).
No difference in color could be noted between the crusted and noncrusted moist samples, yet reflectance from the crusted samples was, generally, higher than that from the noncrusted ones (Fig. 3). The higher reflectance from the crusted samples could be ascribed to (i) its smoother surface (Fig. 2), which decreased the shadowing effect (leading to reduced reflection) of the rougher and more aggregated surfaces of the noncrusted samples, and (ii) possible accumulation of fine material in the crusted layer (Chen et al., 1980; Levy et al., 1988). Although differences in clay content between the Capay soil and the Reiff soils, and differences in organic matter content between the two Reiff soils were measured (Table 1), reflectance spectra for the crusted samples and those for the noncrusted samples were similar in the three soils (Fig. 3).
A spectral absorption feature (decrease in spectral reflectance) at around 1450 nm was noted in both the crusted and noncrusted samples of the three soils (Fig. 3). This feature was assigned to OH in adsorbed water. The large size of the absorption feature in all the samples was ascribed to the high water content in the samples. An additional small absorption feature at 2200 nm was assigned to OH groups in the clay lattice and indicates the presence of montmorillonite in the clay fraction of the soils (Ben-Dor, 2002).
Dry Samples
Reflectance measurements of the dry crusted and noncrusted samples (Fig. 3) were taken after the samples were allowed to dry 48 or 72 h. Moisture contents of the dry samples were 6.2, 2.9, and 3.5% for Capay, ReiffCM, and ReiffOM, respectively. Reflectance measurements for the crusted ReiffOM in the 1950- to 2400-nm region showed a high noise-to-signal ratio and are not presented (Fig. 3). The spectral baseline of the dry samples was, in most cases, higher than that of the corresponding moist samples, and the crusted samples were characterized by higher baseline spectra compared with the noncrusted samples, almost across the entire spectrum (Fig. 3). The higher baseline is explained by the smoother surfaces and lighter color of the crusted samples compared with the noncrusted samples (Fig. 2). The lighter color in the crusted samples suggested that some changes in the composition of the soil surface took place during crust formation. The possible effects of these changes on the spectral response of the soils are discussed later.
Unlike the wet samples, differences in reflectance were noted among the three dry soils. Comparison of the baseline spectra of the ReiffCM and ReiffOM indicate that the spectra of the crusted and noncrusted ReiffOM were lower than the ReiffCM (Fig. 3b, 3c). No differences were observed by eye between the crusted samples of the two soils (Fig. 2). Therefore, the lower spectra in the ReiffOM samples compared with the ReiffCM samples was ascribed to the higher organic matter in the former (Table 1). Soil organic matter plays an important, yet at times complicated, role in the spectral response of soils (Ben-Dor, 2002). However, in general, it has been observed that as organic matter content increases soil reflectance decreases, especially when organic matter content is >2% (e.g., Hoffer and Johannsen, 1969; Stoner and Baumgardner, 1981; Baumgardner et al., 1985).
The baseline spectra of the crusted and noncrusted Capay samples were also lower than those from the ReiffCM samples (Fig. 3a, 3b). This observation is explained by the rougher and darker crusted surface in the Capay soil (Fig. 2), which is attributed to its higher clay content (Table 1). Goldshleger et al. (2001) also observed that the baseline spectra of soils decreased with an increase in soil clay content.
In addition to studying changes in the baseline spectra (albedo) in the crusted samples, it is also of interest to explore the absorption features of the crusted and noncrusted surfaces. Similar to the wet samples, the dry crusted and noncrusted samples from the three soils exhibited an absorption feature at 2200 nm (Fig. 3) belonging to lattice OH in montmorillonite. The intensity of this absorption feature in the dry samples was greater than in the wet ones. In addition, the dry samples of the Capay and ReiffCM also exhibited absorption features around 1410 nm (Fig. 3a, 3b) that is related to the lattice OH of montmorillonite (Ben-Dor, 2002).
To improve the ability to detect changes in the absorption features between the crusted and noncrusted samples, we followed Goldshleger et al. (2001) and divided the reflectance at each wavelength of the noncrusted spectra by the reflectance of the crusted spectra at the same wavelength (Fig. 4)
. Potential peaks in the ratio curve at regions of absorption features related to lattice OH enable us to examine the possibility of accumulation or depletion of clay from the crusted layer. A downward peak indicates that the height of the absorption peak in the noncrusted layer was greater than that in the crusted layer; suggesting clay depletion from the crusted layer compared with the noncrusted one. An upward peak indicates the opposite trend regarding the height of the absorption peak, thus signifying the enrichment of the crusted layer with clay compared with the noncrusted layer. Our samples exhibited a peak in the ratio curve around 1400 nm but not at 2200 nm. The absence of a peak in the ratio curve at 2200 nm may stem from the fact that the absorption peaks at this wavelength were of lower intensity than those at 1400 nm (Fig. 3), a phenomenon also presented in the review of Ben-Dor (2002) for the data on smectitic soils of Kruse et al. (1991) and observed in the data of Ben-Dor et al. (2003). It is postulated that this lower intensity hinders the detection of changes in clay content between the noncrusted and crusted layer based on spectral reflectance data.
View larger version (29K):
[in this window]
[in a new window]
|
Fig. 4. Ratio of noncrusted to crusted reflectance percentage of dry samples from (a) Capay, (b) ReiffCM, and (c) ReiffOM soils.
|
|
The observed peak height and direction at 1400 nm varied among the soils (Fig. 4). The ratio for the ReiffCM sample was a high upward peak with a value of >1 (Fig. 4b), which indicated the possible accumulation of a substantial amount of clay in the crusted layer. This accumulation of clay was not observed in the image of the crusted surface of this soil (Fig. 2b). Unlike the ReiffCM sample, the Capay and ReiffOM samples displayed small downward peaks in their respective ratio curves (Fig. 4a, 4c). These downward peaks (implying some depletion of clay material from the crusted layer) were in agreement with the images of the crusted surfaces of these soils (Fig. 2a, 2c), which showed spots of lighter color belonging to silt and sand grains from which clay was stripped. Goldshleger et al. (2001) also noted downward peaks in the ratio curve around 1400 nm for three Israeli soils. Albeit our inability to explain the upward peak in the ratio curve of the ReiffCM sample by the surface image, the existing agreement between (i) the surface images and the downward peaks in the ratio curves for the Capay and ReiffOM soils, and (ii) the peak direction in the ratio curve in our study and that of Goldshleger et al. (2001), promotes further exploration of the possible use of reflectance data for identifying morphological changes in the surface layer of crusting soils.
Spectral Reflectance and Infiltration Rate of Crusts
Infiltration rates of the three soils at 10 levels of cumulative energy are presented in Table 2. Similar to other observations (Agassi et al., 1981; Betzalel et al., 1995), the increase in rain cumulative energy was accompanied by a decrease in IR, indicating the development of a crust. Assessment of the ability to use spectral response to determine the degree of crust development was obtained by performing a linear regression analysis between the IR of crusts exposed to different levels of cumulative rain energy and their respective spectra. Coefficients of determination (R2) and the slopes of the regression lines for the three soils are presented in Fig. 5
. An inverse relation (negative slope) was found between reflectance and crust permeability, that is, reflectance decreased with an increase in IR. In all three soils, R2 values > 0.81 (corresponding to correlation coefficients > 0.9) were found for many different wavelengths.
View this table:
[in this window]
[in a new window]
|
Table 2. Infiltration rate (IR) at progressively greater cumulative rain energy for the Capay, ReiffCM, and ReiffOM soils.
|
|
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 5. Coefficient of determination (R2) and the slope (a) resulting from linear regression analysis between reflectance values at every spectral wavelength and the infiltration rate for the dry samples of (a) Capay, (b) ReiffCM, and (c) ReiffOM soils.
|
|
From the data presented in Fig. 5, we identified for each soil the wavelength at which both the coefficient of determination was >0.81, and the absolute value of the slope was high. The latter criterion was used to enhance the sensitivity of the reflectance as a predictor for changes in crust IR. It was found that the reflectance maxima (Fig. 3) between the absorption features gave the best correlations. The absorption maxima at 1700 nm (which is between the absorption features assigned to the OH of the absorbed water) provided highly significant correlations for all three soils. The absorption maxima at 2130 nm also provided a highly significant correlation with the clay-rich Capay. This is near the 2200-nm adsorption feature assigned to the OH of the clay lattice. We then plotted the reflectance at this wavelength as a function of IR (Fig. 6)
and found that logarithmic regression provided even higher coefficients of determination (R2). The slope of the fitted line in the Capay and ReiffOM soils was similar and more than double that in the ReiffCM (Fig. 6). The greater slope in the Capay and ReiffOM soils indicated greater changes in the reflectance with changes in the IR; that is, reflectance was more sensitive to the different stages in crust development and hence changes in the structure of the soil surface. The changes in the reflectance spectra between the dry crusted and noncrusted samples from the three soils (Fig. 3) point out that the main source of difference in the Capay and ReiffOM were in the baseline (albedo), while in the ReiffCM the notable change was in the absorption feature at 1400 nm. Changes in albedo are associated with changes in the physical characteristics of the soil surface and are best correlated with the IR. It is suggested, therefore, that in the Capay and ReiffOM soils, the processes of crust formation yielded greater changes in physical characteristics of the soil surface compared with the ReiffCM soil. It is further postulated that the observed greater changes in the physical characteristics of the Capay and ReiffOM soil surfaces during crust formation in comparison with the ReiffCM soil is associated with the presence of more aggregate stabilizing agents in the former two soils, high clay content in the Capay and high organic mater content in the ReiffOM (Table 1).
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 6. Reflectance at 1700 and 2130 nm as a function of infiltration rate for the dry samples of (a) Capay, and at 1700 nm for (b) ReiffCM and (c) ReiffOM soils.
|
|
|
SUMMARY AND CONCLUSIONS
|
|---|
Reflectance measurements from crusted and noncrusted soil samples from different soils were made under direct sunlight across the entire spectral region of solar illumination (400–2400 nm). Reflectance from crusted surfaces was generally higher than that from noncrusted surfaces. The source of the observed difference was mainly in the baseline of the spectra (albedo), which is associated with physical changes that occur in the surface during the process of crust development. Analysis of the absorption features in the reflectance spectra indicated that in two of our soils there was a depletion of fine material from the crusted layer when compared with the noncrusted samples. In the third soil, the analysis implied that enrichment of the crust with clay occurred.
Our data suggest that detection of the development of crusts at soil surfaces via reflectance measurements is not restricted to laboratory conditions only (as has been shown in previous studies), but can also be obtained when measured under natural direct sun light. We found that the reflectance around 1700 nm provided a statistically significant correlation with IR for all three soils studied and around 2130 nm also provided a statistically significant correlation with IR for the high clay Capay soil. Although we used just three soils, our finding are supported by interpretation of other soils studied before (e.g., Lobell and Asner, 2002; Ben Dor et al., 2003). The results of our study provide solid evidence that crust development can be examined using reflectance spectroscopy. Further studies, mainly under natural conditions and at a field scale are needed to be able to use this tool for monitoring soil conservation programs or to help in mapping regions susceptible to soil degradation.
|
ACKNOWLEDGMENTS
|
|---|
The work reported here was supported in part by the Binational Agricultural Research and Development Fund (BARD) grant IS-3111-99. We appreciate the help of Rick Perk from the University of Nebraska for making the reflectance measurements and initial data management. We also thank Bryan Leavitt from the University of Nebraska for assistance with data handling.
Received for publication January 7, 2004.
|
REFERENCES
|
|---|
- Agassi, M., I. Shainberg, and J. Morin. 1981. Effect of electrolyte concentration and soil sodicity on infiltration rate and crust formation. Soil Sci. Soc. Am. J. 45:848–851.[Abstract/Free Full Text]
- Baumgardner, M.F., L.F. Silva, L.L. Biehl, and R.E. Stoner. 1985. Reflectance properties of soils. Adv. Agron. 38:1–44.
- Ben-Dor, E. 2002. Quantitative remote sensing of soil properties. Adv. Agron. 75:173–243.
- Ben-Dor, E., and A. Banin. 1995. Near infrared analysis as a rapid method to simultaneously evaluate several soil properties. Soil Sci. Soc. Am. J. 59:364–372.[Abstract/Free Full Text]
- Ben-Dor, E., N. Goldshleger, Y. Benyamini, M. Agassi, and D.G. Blumberg. 2003. The spectral reflectance properties of soil structural crusts in the 1.2–2.5 µm spectral region. Soil Sci. Soc. Am. J. 67:289–299.[Abstract/Free Full Text]
- Ben-Dor, E., J.A. Irons, and A. Epema. 1998. Soil spectroscopy. p. 111–189. In A. Renez (ed.) Manual of remote sensing. 3rd ed. Wiley and Sons, New York.
- Betzalel, I., J. Morin, Y. Benyamini, M. Agassi, and I. Shainberg. 1995. Water drop energy and seal properties. Soil Sci. 159:13–22.
- Bradford, J.M., J.E. Ferris, and P.A. Remley. 1987. Interrill soil erosion processes: I. Effect of surface sealing on infiltration, runoff, and soil splash detachment. Soil Sci. Soc. Am. J. 51:1566–1571.[Abstract/Free Full Text]
- Chen, Y., J. Tarchitzky, and A. Banin. 1980. Scanning electron microscope observation on soil crusts and their formation. Soil Sci. 130:49–55.
- Dalal, R.C., and R.J. Henry. 1986. Simultaneous determination of moisture or organic carbon and total nitrogen by near infrared reflectance spectroscopy. Soil Sci. Soc. Am. J. 50:120–123.[Abstract/Free Full Text]
- Eshel, G., G.J. Levy, U. Mingelgrin, and M.J. Singer. 2004. Critical evaluation of the use of laser diffraction for particle-size distribution analysis. Soil Sci. Soc. Am. J. 68:736–743.[Abstract/Free Full Text]
- Gal, M., L. Arcan, I. Shainberg, and R. Keren. 1984. Effect of exchangeable sodium and phosphogypsum on crust structure—Scanning electron microscope observations. Soil Sci. Soc. Am. J. 48:872–878.[Abstract/Free Full Text]
- Goldshleger, N., E. Ben-Dor, Y. Benyamini, M. Agassi, and D. Blumberg. 2001. Characterization of soil's structural crust by spectral reflectance in the SWIR region (1.2–2.5 µm). Terra Nova 13:12–17.
- Hoffer, R.M., and C.J. Johannsen. 1969. Ecological potentials in spectral signature analysis. p. 1–29. In P.L. Johnson (ed.) Remote sensing in ecology. Univ. of Georgia Press, Athens.
- Ishida, T., H. Ando, and M. Fukuhara. 1991. Estimation of complex refractive index of soil particles and its dependence on soil chemical properties. Rem. Sens. Environ. 38:173–182.
- Kruse, F.A., M. Thiry, and P.L. Hauff. 1991. Spectral identification (1.2–2.5 µm) and characterization of Paris basin kaolinite/smectite clays using a field spectrometer. p. 181–184. In Proc. of the 5th Int. Colloquium, Physical Measurements and Signatures in Remote Sensing, Courchevel, France. Vol. I. 14–18 Jan. 1991.
- Levy, G.J., P. Berliner, H.M. du Plessis, and H.V.H. van der Watt. 1988. Microtopographical characteristics of artificially formed crusts. Soil Sci. Soc. Am. J. 52:784–791.[Abstract/Free Full Text]
- Levy, G.J., J. Levin, and I. Shainberg. 1994. Seal formation and interrill soil erosion. Soil Sci. Soc. Am. J. 58:203–209.[Abstract/Free Full Text]
- Lobell, D.B., and G.P. Asner. 2002. Moisture effects on soil reflectance. Soil Sci. Soc. Am. J. 66:722–727.[Abstract/Free Full Text]
- McIntyre, D.S. 1958. Permeability measurements of soil crusts formed by raindrop impact. Soil Sci. 85:185–189.
- Onofiok, O., and M.J. Singer. 1984. Scanning electron microscope studies of soil surface crusts formed by simulated rainfall. Soil Sci. Soc. Am. J. 48:1137–1143.[Abstract/Free Full Text]
- Planet, W.G. 1970. Some comments on reflectance measurements of wet soils. Rem Sens. Environ. 1:127–129.
- Shainberg, I., and G.J. Levy. 1995. Infiltration and seal formation processes. p. 1–22. In M. Agassi (ed.) Soil erosion and rehabilitation. Marcel Dekker, New York.
- Singer, M.J. 1991. Physical properties of arid region soils. p. 81–109. In J. Skujins (ed.) Semiarid lands and deserts, soil resource and reclamation. Marcel Dekker, New York.
- Singer, M.J., and D. Warrington. 1992. Crusting in the western United States. p. 179–204. In M.E. Sumner and B.A. Stewart (ed.) Soil crusting: Chemical and physical processes. Lewis Publ., Boca Raton, FL.
- Stoner, E.R., and M.F. Baumgardner. 1981. Characteristic variations in reflectance of surface soils. Soil Sci. Soc. Am. J. 45:1161–1165.[Abstract/Free Full Text]
- Stoner, E.R., M.F. Baumgardner, R.A. Weismiller, L.I. Biehl, and B.F. Robinson. 1980. Extension of laboratory measured soil spectra to field conditions. Soil Sci. Soc. Am. J. 44:572–574.[Abstract/Free Full Text]
- Twomey, S.A., C.F. Bohren, and J.L. Mergenthaler. 1986. Reflectances and albedo differences between wet and dry surfaces. Appl. Opt. 25:431–437.
This article has been cited by other articles:
|
|
|
|
 
D. G. Sullivan, J. N. Shaw, and D. Rickman
IKONOS Imagery to Estimate Surface Soil Property Variability in Two Alabama Physiographies
Soil Sci. Soc. Am. J.,
September 29, 2005;
69(6):
1789 - 1798.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
