SSSAJ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Scalenghe, R.
Right arrow Articles by Ugolini, F. C.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Scalenghe, R.
Right arrow Articles by Ugolini, F. C.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Scalenghe, R.
Right arrow Articles by Ugolini, F. C.
Related Collections
Right arrow Soil Compaction
Right arrow Pedology
Right arrow Soil Physics
Published in Soil Sci. Soc. Am. J. 68:204-214 (2004).
© 2004 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA

DIVISION S-5—PEDOLOGY

Segregated Ice and Liquefaction Effects on Compaction of Fragipans

Riccardo Scalenghe*,a, Giacomo Certinib, Giuseppe Cortic, Ermanno Zaninid and Fiorenzo C. Ugolinib

a Dipartimento di Agronomia, Coltivazioni Erbacee e Pedologia, Università di Palermo, Viale delle Scienze, 90128 Palermo, Italy
b Dipartimento di Scienza del Suolo e Nutrizione della Pianta, Università di Firenze, Piazzale delle Cascine 28, 50144 Firenze, Italy
c Dipartimento di Scienze Ambientali e delle Produzioni Vegetali, Università delle Marche, Via Brecce Bianche, Monte Dago, 60100 Ancona, Italy
d Dipartimento di Valorizzazione e Protezione delle Risorse Agroforestali, Università di Torino, via Leonardo da Vinci 44, 10095 Grugliasco (TO), Italy

* Corresponding author (scalengh{at}unipa.it).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The brittleness of fragipans is generally ascribed to the close-packing fabric arrangement acquired at the initial step of pedogenesis thanks to physical processes. However, there is an on-going debate over the agent causing soil densification. In this work, we tested the plausibility that ice segregation or liquefaction could have been the cause of the compaction of four fragipans. Two of them are located in nonseismic areas that have experienced periglacial conditions; one is from a strongly seismic area not affected by periglacial conditions, while the fourth site underwent moderate seismic activity and slight periglacial conditions. After disaggregation in the laboratory, soil specimens were submitted to freeze–thaw cycles and vibrations at different amplitude and duration, either dry or water-saturated. Analyses of aggregate stability, bulk density, porosity, and pore-size distribution were made on natural and treated specimens. Results indicated that the compactness arose mainly from the close-packing arrangement of particles. The freeze–thaw cycles were able to reproduce only some of the features in the water-saturated specimens, independent of whether they came from periglacial or seismic areas, while those from seismic areas successfully acquired the original arrangement after vibrations-induced liquefaction. This different behavior could be partly explained by the fact that consolidation after liquefaction occurs only if a material with proper particle-size distribution and mineralogical assemblage is saturated by solutions able to disperse phyllosilicates and promote their face-to-face arrangement. Our findings support the hypothesis that liquefaction of soil material due to earthquakes could indeed provide a dense parent material in which the fragipan may develop through pedogenesis.

Abbreviations: COLE, coefficient of linear extensibility


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
THE CHARACTERISTICS OF FRAGIPANS consist chiefly of a high bulk density, a low content of organic matter, and a texture usually ranging from fine sand to silt. When dry, fragipan peds are able to resist compression, but yield suddenly under a considerable load; when wet, peds generally collapse under their own weight. Fragipans are distributed worldwide and have been intensively studied (e.g., Nikiforoff et al., 1948; Smalley and Davin, 1982; James et al., 1995). Despite this, the origin of the hardness of fragipans is still debated and numerous theories have been proposed (Bryant, 1989). This hardness is usually ascribed to the close-packing arrangement of particles, although some authors have found evidences of inorganic cements, for example, of clay minerals (Knox, 1957; Lozet and Herbillon, 1971; Wang et al., 1974) and silica or aluminosilicate polymers (Karathanasis, 1987; Norfleet and Karathanasis, 1996; Duncan and Franzmeier, 1999). In any case, initial mechanical compaction is thought to be necessary for fragipan formation and four compaction mechanisms have been invoked: (i) action of permafrost (FitzPatrick, 1956; Van Vliet and Langhor, 1981; Payton, 1992), (ii) load exerted by glacial ice masses (Bryant, 1989), (iii) collapse of water-saturated sediments under their own weight (Barden et al., 1973; Assallay et al., 1998), and (iv) alternation of wetting-drying cycles (Jha and Cline, 1963; Bruckert and Bekkary, 1992; Attou and Bruand, 1998; Boulet et al., 1998), favored by an initial hydroconsolidation (Bryant, 1989). In areas intensively affected in the past by glacial and periglacial conditions, the first two mechanisms are considered the most plausible explanations for the high density acquired by the fragipans.

In Italy, glacial and periglacial conditions may have been responsible for the formation of fragipans in the northern alpine area. In fact, several soils developed on glacial and fluvioglacial deposits in northern Italy contain fragipans (Arduino et al., 1982; Previtali, 1985; Casati et al., 1987; Ente Regionale di Sviluppo Agricolo della Lombardia [ERSAL], 1988; Ajmone Marsan et al., 1994). In Tuscany (central Italy), fragipans have been described in areas where periglacial conditions were moderate or absent. They occur in soils derived from Oligocene sandstone (Certini et al., 1997), Pliocene conglomerates and sandy marine deposits in the Chianti region (Lulli et al., 1987), Pleistocene colluvia and alluvia near Siena (Costantini et al., 1996), Middle Pleistocene fluviolacustrine terraces in the Mugello valley (Dimase and Desideri, 1983) and fluvial terraces in the Ampio valley near Grosseto (Sanesi, 2000). In addition, we found not described fragipans on the highest terraces of Pleistocene fluvial and fluviolacustrine deposits in the Arno valley (from Florence to Arezzo) and on Trias limestone breccias reworked in Neogene times in the Merse valley (close to Siena). The Tuscan fragipans have developed in parent materials that (i) moved along slopes, as in the cases of colluvia and alluvia, or (ii) remained in place after deposition, as for fluvial and fluviolacustrine deposits. To explain consolidation of the fragipans in the latter case, many theories have been invoked, but a new one can also be proposed. Most of these fragipans are in seismic areas and it is proposed that earthquakes could have been responsible for their formation through liquefaction and successive compaction of the parent material. The liquefaction of water-saturated sediments due to earthquakes is a well-known phenomenon (Committee on Earthquake Engineering, 1985) and has been experimentally demonstrated by Faccioli and Resendiz (1976). Liquefaction is consequent on the annulment of contact forces between particles by water (Kramer and Seed, 1988). Once liquefied, sediments on flat surfaces remain in situ, while those on slopes >3° can move downward up to distances of several kilometers (Seed, 1976; Youd, 1992). Dewatering increases the density of the sediments (Ishihara, 1985). Consequently, seismic events might be a cause of the initial mechanical compaction of parent materials that then developed into a fragipan.

The aim of this work was to investigate the extent to which ice segregation in soil or liquefaction of earthy material might have been responsible for imparting the close-packing arrangement of particles in four fragipans from areas that experienced different climatic conditions and seismic activity.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Study Areas
This study was conducted on four soils from northern and central Italy where the fragipan showed different thickness and development of bleached seams. The soils were (Soil Survey Staff, 1999): (i) Fragic Paleudult from Venegono (Varese, Lombardy), (ii) Aquic Fragiudalf from Maggiora (Vercelli, Piedmont), (iii), Fragic Paleudult from Scarperia (Florence, Tuscany), and (iv) Fragic Dystrudept from Vallombrosa (Florence, Tuscany). The first three fragipans showed a similar development of bleached seams with a progressive decrease of thickness, while the fourth one did not show any seam. The soils were previously described and classified by other authors: ERSAL (1988) for Venegono, Ajmone Marsan and Torrent (1989) for Maggiora, Dimase and Desideri (1983) for Scarperia, Corti et al. (2001) for Vallombrosa. However, for this work a new description of a representative profile (Table 1) for each area was made according to the guidelines of the Soil Survey Division Staff (1993). The soils of Venegono and Maggiora are from northern Italy and are developed on fluvioglacial terraces formed 400 to 200 kyr BP and 500 to 300 kyr BP, respectively. In both cases, the parent material consists of deposits made up from a number of different lithologies (Table 1). The sites of Scarperia and Vallombrosa are in central Italy. The soil of Scarperia is developed on a fluviolacustrine terrace formed 500 to 400 kyr BP and consists of sediments of sandstone, siltstone, and flintstone. The soil of Vallombrosa (950 m above sea level) developed on soliflucted soil material derived from sandstone and siltstone that moved along sloping surfaces, 36 to 25 kyr ago according to radiocarbon data. No information about the weathering status of the sediments at the time of deposition is available, but the age of the deposition of the parent materials of Vallombrosa appears to be much younger than the sediment-parent materials of the other soils. This accounts for a shorter pedogenesis of the Vallombrosa soil. However, it is assumed that, since consolidation, pedogenic processes have produced only minor changes in the physical, chemical, and mineralogical properties of the four fragipans.


View this table:
[in this window]
[in a new window]
  		Table 1. Morphological description of the sites investigated.

 
The areas of Venegono and Maggiora experienced intense periglacial conditions (Tomadin and Varani, 1998) and slight seismic activity (Albini et al., 2000). The area of Scarperia was scarcely affected by periglacial conditions while earthquakes were frequent and strong (Albini et al., 2000). Vallombrosa underwent slight periglacial conditions (Losacco, 1949) and moderate seismic activity (Albini et al., 2000). The fragipans from Venegono, Maggiora, and Scarperia formed on parent materials that remained in situ after deposition. The fragipan of Vallombrosa often occurs as tongues (not recognizable through the present geomorphology) aligned along the lines of maximum slope (Bolla, 2001). The intermittent spatial occurrence of this horizon led us to suppose that it originated by the consolidation of mudflows.

The four areas were covered by dicot forest species until the medieval times. At present, Venegono and Scarperia have conserved their original vegetation (Table 1), Maggiora is cultivated with cereals and Vallombrosa has been replanted with conifers. According to the phytosociological classification of Braun-Blanquet (1964), all sites originally belonged to the class Querco fagetea, order Quercetalia pubescentis for Venegono, Maggiora and Scarperia, and order Fagetalia for Vallombrosa.

Evaluation of Aggregate Stability as a Rationale for Sampling
The slaking of aggregates in water is the most common field-test to identify the fragipan (Lindbo and Rhoton, 1996). Laboratory analyses such as determination of bulk density, saturated hydraulic conductivity or soil strength may help to define the boundary between the fragipan and the adjacent horizons (Habecker et al., 1990), but not be used as principal criteria. For example, Lindbo et al. (1994) reported fragipans that exhibit bulk densities not significantly different from those of the overlying or underlying horizons of the same pedons.

Many methods based on wet sieving have been proposed to measure water-stable aggregates. In this work, to evaluate aggregate stability, we used the model proposed by Zanini et al. (1998a), which has been successfully checked by Zanini et al. (1998b) to identify the best-expressed fragic character within a fragipan as recognized in the field. This model allows quantitative evaluation of aggregate breakdown and uses a system where dry aggregates are gradually wetted, flooded, and subjected to disruptive actions of both flowing water and eroding suspended material. The net amount of disintegration is limited by the content of coarse primary particles and is determined by the progressive breakdown caused both by dissolution of bonding substances and mechanical abrasion of particles. The mathematical model that results from a quantitative evaluation of these assumptions is an equation describing the asymptotic approach to the maximum decay of the aggregates in time t. For the purpose of estimating appropriate values of the parameters from experimental data, it is assumed that the breakdown of the aggregates could be described by the following equation

[1]
where y is the aggregate breakdown or loss (grams), t is the time of abrasion (minutes), a is the incipient failure of aggregates during saturation with water (grams), b is the maximum abrasion loss of aggregates (grams), and c (minutes) is a time controlling factor equal to 1/3 of the time interval at which approximately 95% of the sill (a + b) is reached (Isaaks and Srivastava, 1989). From this consideration, the more useful 3c parameter is derived, which indicates the time needed to destroy 90% of the aggregates (Zanini et al., 1998a). Practically, about 10 g of air-dried aggregates of 1 to 2 mm in diameter were wetted at atmospheric pressure and repeatedly wet-sieved in deionized water by Yoder's multiple-sieve apparatus at 40 cycles min–1 for 5, 10, 15, 20, 40, and 60 min. Then, the amount of aggregates >0.2 mm in diameter that resisted the abrasion were air-dried and weighed; this value was corrected for coarse sand after dispersion with sodium hexametaphosphate. The estimated values of parameters a, b, and c of the Eq. [1] were obtained using the iterative nonlinear regression procedure (SPSS package).

In the soils examined, the model was applied to intact peds collected at 15-cm intervals from the upper boundary of the fragipan to the bottom of the profile. From the fragipan layer of each soil that showed the maximum aggregate breakdown (data not shown), we collected undisturbed samples to perform analyses and experiments.

General Soil Characterization
A portion of the soil samples was air-dried and sieved at 2 mm. On the fine earth, textural analyses were performed after treatments with 3 M H2O2 to remove organic matter; the sand was separated using a 53-µm sieve and the clay was separated from the silt by sedimentation (Stoke's law) in a suspension adjusted to pH 8.8 to 8.9 with 0.01 M NaOH and maintained in a warm bath (Gee and Bauder, 1986). The drop-cone penetrometer method was used to determine the liquid limit (Sherwood and Ryley, 1970). The pH in water was determined potentiometrically using a solid/liquid ratio of 1:2.5. Organic C was measured by combustion using a Carlo Erba NA 1500 N/C/S Analyzer, Series 2 (Carlo Erba, Milan, Italy). The mineralogical investigation was conducted by a Philips PW 1710 diffractometer using a Fe-filtered Co-K{alpha}1 radiation (Philips, Eindhoven, the Netherlands). Treatments of the powders (saturation with Mg and K, solvation with glycerol, and heating at 550°C) and semi-quantitative estimation of the mineralogical composition of the whole fine earth was obtained according to Ugolini et al. (2001). The coefficient of linear extensibility (COLE) was determined according to USDA-NRCS (1996). From the other portion of the soil samples, undisturbed peds were separated to assess their aggregate stability, porosity, and pore-size distribution. Aggregate stability was determined according to the previously described method of Zanini et al. (1998a). Bulk density was determined by the clod method of Blake and Hartge (1986), based on the weight obtained at 105°C; for each sample, 15 replicates were used. The porosity and the pore-size distribution in the range 0.006 to 100 µm of diameter were measured on dried (40°C) and degassed peds of about 2 cm3 in size by a Hg intrusion Carlo Erba 2000 WS porosimeter equipped with a Carlo Erba 120 macropore unit (Carlo Erba, Milan, Italy); the tension of Hg was 0.0048 N cm–1 (480 dyne cm–1), the soil-Hg contact angle was 140°.

Simulation of Ice Segregation and Earthquake Effects
For the following experiments, the fine earth was gently disaggregated using a rotating gum-cylinder apparatus. Specimens of 100 g (four replicates per soil) were placed in finely pierced PVC insulated beakers and water-saturated by immersing the bottom of the beakers in a container, at atmospheric pressure. Water absorbed by capillary rise up to field capacity was determined gravimetrically; it accounted to 56 mL for Venegono and Maggiora, 48 mL for Scarperia, and 29 mL for Vallombrosa. On these specimens two experimental treatments were conducted as summarized in Fig. 1 . The effects of ice segregation were simulated by five sequential 24-h cycles; during each cycle, temperature was maintained for 12 h at –10°C then for 12 h at 25°C. The cold temperature was applied from bottom and top of the containers, while the warm temperature was irradiated only from the top. Some not significant quantities of water were lost during melting of the frozen material, both for evaporation and drainage trough the holes at the base of the beakers. Bulk density, aggregate stability, porosity, and pore-size distribution were measured on the treated specimens. Earthquakes were simulated by vibrating the specimens with a vibrant head. Vibrations at 50 Hz were applied for 1 min; one specimen was treated at 0.4 mm of amplitude and a duplicate specimen was treated at 1 mm of amplitude. These treatments simulated peak ground accelerations of 0.35 and 0.98 m s–2, comparable with VIII and IX degrees of the 12-degrees Mercalli-Cancani-Sieberg (MCS) scale of seismic intensity (Davison, 1921). Basing on the recollection of the earthquakes effects in Italy since 2400 yr BP (Albini et al., 2000), the two studied sites in seismic areas, Vallombrosa and Scarperia, experienced earthquakes with intensities up to VI and IX degree, respectively.



View larger version (19K):
[in this window]
[in a new window]
  		Fig. 1. Experimental design of the treatments to simulate effects of ice-segregation in soil and liquefaction of soil induced by earthquakes on fragipan specimens.

 
The bulk density was measured on both treated specimens, while aggregate stability, porosity, and pore-size distribution were obtained only for the specimens vibrated at 0.4 mm of amplitude. Bulk density was also determined on air-dried specimens vibrated at 50 Hz for 1 min at 0.4 or 1 mm of amplitude, and for 5 min at 0.4 mm of amplitude. The freeze–thaw treatment was conducted in an attempt to recreate the effects of ice segregation on soils. However, the treatment was extreme, with a temperature difference of 35°C in 24 h, a rare event in nature. Concerning earthquakes, most of them occur with a release of energy corresponding to frequencies ranging from 1 to 12 Hz (Peacock and Seed, 1968). Depending on water content, texture, mineralogy, and stability of aggregates, it is likely that there is a threshold intensity that results in liquefaction of the soil. We did not want to identify this threshold for the samples under analysis, but merely ensure that liquefaction had in fact occurred. Consequently, the strong motion of 50 Hz for 1 min was applied so as to ensure that all the water-saturated specimens were liquefied. The specimens treated in wet conditions were then oven-dried (50°C) before analysis. As for the freeze–thaw treatment, each sample was run in quadruplicate.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
General Characterization
The layer showing the maximum aggregate breakdown occurred at a depth from the surface of 175 to 190 cm at Venegono and Scarperia, 190 to 205 cm at Maggiora, 100 to 115 cm at Vallombrosa.

All the collected samples had a texture with sand and silt individually exceeding clay fractions (Table 2). This implies that these materials are susceptible to liquefaction since the intermolecular forces of cohesion decrease as the dimensions of particles increase (Castro, 1969). The specimens from seismic areas (Scarperia and Vallombrosa) reached the liquid limit (Table 2) at a percentage of water content lower than that of the specimens that experienced periglacial conditions (Venegono and Maggiora). The samples from seismic areas also had lower pH values (Table 2). As is typical for fragipans, organic C was scarce in all samples, especially at Scarperia where it accounted for only 0.2 g kg–1 (Table 2). Mineralogical analysis indicated that primary minerals such as quartz, plagioclases, micas, and chlorite prevailed in all samples (Table 2). Kaolinite was always represented, while hydroxy-Al interlayered vermiculites (HIV) were present only at Venegono and Vallombrosa. Of the expandable clay minerals, vermiculite was present at Venegono, vermiculite and smectite at Maggiora, while small amounts of an assemblage of disordered 2:1 expandable clay minerals (mostly smectitic) occurred at Scarperia. The presence of expandable clay minerals seems to be incongruent with the acquisition of a dense state, but several papers have reported fragipans that contain high amounts of vermiculite (Bruckert and Bekkary, 1992; Payton, 1993; Ajmone Marsan et al., 1994) and smectites (Yassoglou and Whiteside, 1960). However, even though the samples of the four soils were considered to be very similar to the original sediments in which the fragipans formed, it cannot be excluded that some mineralogical changes could have occurred in those sediments consolidated more than 200000 yr ago (Venegono, Maggiora and Scarperia). However, at least for the fragipan of Scarperia, the formation of smectite since consolidation seems unlikely because of the acidic pH (Table 2), although there is insufficient information to speculate about the role of pedogenesis in the formation of secondary minerals after the compaction of the parent material.


View this table:
[in this window]
[in a new window]
  		Table 2. Particle-size distribution, liquid limit, pH, organic C, semi-quantitative estimate of the mineralogical composition, and coefficient of linear extensibility (COLE) for the fine earth of fragipan specimens. Numbers in brackets are the standard errors.

 
The COLE values appeared to be related to both the age and mineralogy of the deposits. Thus, the Vallombrosa sample showing the lowest COLE value (0.026 cm cm–1), came from the soil developed on the youngest parent material (25–36 kyr BP) and did not contain expandable clay minerals. The other samples that displayed higher COLE values came from older parent materials and contained expandable minerals.

Aggregate Stability
The loss of aggregates during the water-saturation and abrasion procedure (parameter a + b) after freezing–thawing was similar to that of the natural specimens for Venegono and Maggiora, higher for Scarperia and lower for Vallombrosa (Table 3). Vibrations decreased the original value of the parameter a + b for all the specimens. For the specimens submitted to freeze–thaw, the time necessary to destroy 90% of the aggregates (parameter 3c) was similar or slightly higher than that required for the natural specimens, while liquefied specimens required a considerably longer lapse of time (Table 3).


View this table:
[in this window]
[in a new window]
  		Table 3. Effect of freeze–thaw treatment and liquefaction on disaggregation kinetic parameters of fragipan specimens.

 
A value of the parameter a + b similar to or lower than that of the natural specimens indicates that the treatments have imparted the same or a more stable aggregation to the specimens. In these cases, it can be inferred that the aggregate stability in the natural specimens can hardly be due to relatively strong, irreversible, binding agents such as Fe and/or Al oxyhydroxides or silica. As both treatments involved water-saturation, the acquisition of a more resistant aggregation could be partly attributed to the formation of reversible clay mineral bridges, as reported by Attou and Bruand (1998). Thus, it might be expected that liquefaction is able to promote face-to-face rearrangement of the phyllosilicates in the water-saturated specimens more effectively than freeze–thaw treatment. It is possibly for this reason that vibration treatment did produce the highest aggregate stability, as shown by a greater diminution of the a + b parameter as measured in the natural specimen (Table 3). The specimen from Scarperia showed ambiguous behavior, suggesting the presence of cementing agents during the freeze–thaw treatment, but indicating the loss of such cements after liquefaction.

Bulk Density
Fragipans often display high bulk densities, ranging from 1.5 or 1.6 to 2.15 Mg m–3 (Lindbo and Veneman, 1989; Ajmone Marsan et al., 1994), although lower values have been reported (Lindbo et al., 1994). Our samples had bulk densities from about 1.60 Mg m–3 at Scarperia to 1.70 Mg m–3 at Maggiora (Fig. 2) . The freeze–thaw treatment succeeded in reproducing the original bulk density in the specimens from Venegono and Vallombrosa, while the treated specimens from Maggiora and Scarperia became less dense than the natural ones (Fig. 2). Vibrations of dry specimens produced bulk densities considerably lower than those of the original samples, while increase of amplitude or application time did not produce significant changes (Fig. 2). Vibrations of water-saturated specimens always led to a denser state compared with the equivalent dry specimens, hence demonstrating the fundamental role of water in lubricating the particles and enabling their close arrangement after liquefaction. However, only the liquefied specimens from Scarperia and Vallombrosa reached bulk densities similar to the original ones, while those from Venegono and Maggiora became less dense. Evidently, there are other intrinsic properties of the soil materials that are fundamental in reproducing their original fabric.



View larger version (51K):
[in this window]
[in a new window]
  		Fig. 2. Effect of freeze–thaw treatment and liquefaction on bulk density of fragipan specimens. Bars indicate standard errors.

 
Vibration of water-saturated specimens at 1-mm amplitude rather than at 0.4 mm resulted in no further increase of bulk density (Fig. 2), indicating that the arrangement of the particles after liquefaction did not depend on the amplitude of vibration.

Porosity
Pore volume determined in the natural and treated samples was in the range 0.006- to 100-µm diameter (Table 4). We took into consideration two classes of pores—the "storage" pores, 0.5 to 50 µm in diameter, and the "residual" pores, <0.5 µm in diameter (Greenland, 1977)—because Olson (1985) found that fragipans often have lower total porosity but higher amounts of residual pores than the adjacent horizons. This latter author suggested using pore-size distribution as a test for fragipan identification. As field porosity includes visible pores with diameters ranging from 0.5 to 1 mm or more (Table 1), which are not measured by Hg intrusion, the porosity reported in Table 4 is not total porosity. Consequently, correlation with bulk density data cannot be made.


View this table:
[in this window]
[in a new window]
  		Table 4. Effect of freeze–thaw treatment and liquefaction on porosity of fragipan specimens. Numbers in brackets are the standard errors.

 
For both natural and treated specimens, summation of the storage and residual classes of pores (0.006–50 µm) represented 99 to 100% of the porosity in the 0.006- to 100-µm range, indicating that the number of pores in the 50- to 100-µm size range was negligible.

Freeze–thaw treatment of the Maggiora soil resulted in no significant change in porosity both in terms of the 0.006- to 100-µm pores and residual and storage classes. The same treatment decreased the 0.006- to 100-µm porosity in the other samples because of the diminution of both residual and storage pores for the Venegono soil, the sole residual pores for Scarperia, and the storage ones for Vallombrosa (Table 4). After liquefaction, the Scarperia and Vallombrosa specimens reacquired porosity similar to the natural one. For Venegono and Maggiora liquefaction provoked a decrease in the 0.006- to 100-µm porosity due to the loss of storage porosity, partly counterbalanced by an increase in the residual porosity (Table 4). Summarizing, the freeze–thaw treatment recreated the same porosity of the natural soil only at Maggiora, while liquefaction did the same at Scarperia and Vallombrosa. These findings again suggest that the properties of the materials from which fragipans generated played a key role in the compaction.

Pore-Size Distribution
The distribution of pores within the range 0.006 to 100 µm in diameter, shown in Fig. 3 as the percentage of cumulative curves of frequencies, indicates that neither freeze–thaw nor vibrations treatments produced substantial changes in the pore-size distribution of the specimens from Scarperia and Vallombrosa. However, major changes after both treatments were observed in the Venegono and Maggiora specimens (Fig. 3). In particular, compared with the natural sample, freeze–thaw treatment of the Venegono soil showed an increase of pores with a diameter between 0.04 and 0.3 µm, while for the Maggiora soil there was an increase of the pores in the same range and between 7 and 20 µm. Compared with the original samples, liquefaction of the Venegono soil induced a remarkable increase of the pores <0.01 µm and of those ranging from 0.6 and 3 µm; for the Maggiora specimens a large increment of pores with a diameter <0.01 µm and of those between 0.5 and 0.8 µm was observed. These results indicate that ice segregation and liquefaction may or may not reproduce the natural pore-size distribution depending on the soil composition. When the pore-size distribution changes, there is at least a percentage increase of the pores between 0.04 and 0.3 µm after freeze–thaw treatment, while liquefaction causes a more substantial increase of the <0.01-µm pores.



View larger version (29K):
[in this window]
[in a new window]
  		Fig. 3. Effect of freeze–thaw treatment and liquefaction on pore-size distribution of fragipan specimens. Bars indicate standard errors.

 

   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
The experimental freeze–thaw and vibration treatments used in this work represent an over-simplification of natural events. In fact, differences of 35°C in 24 h and earthquakes with frequencies of 50 Hz are very rare events. In addition, both treatments used neglected the presence of horizons overlying the fragipan, which could have played a role in the compaction of the fragipan itself. However, at least in the case of vibration treatment, the energy applied corresponded only to the intent to cause liquefaction of the water-saturated specimens.

Analyses of aggregate stability indicated that the consolidation of the four fragipans was mainly due to a close-packing arrangement of particles while the inorganic cements probably played a secondary role.

The freeze–thaw treatment used reproduced only some of the original features of each soil sample, independently of whether the material experienced periglacial conditions or came from seismic areas (Table 5). A possible explanation for these results is that the complexity of the natural phenomena of ice segregation was not sufficiently mimicked by the adopted procedure. However, even natural cycles of freezing and thawing per se are not able to fully explain fragipan genesis as suggested by Van Vliet and Langhor (1981). On the contrary, a close-packing arrangement similar to the original one was successfully recreated following liquefaction after vibration treatment of water-saturated specimens from the seismic areas of Scarperia and Vallombrosa. After liquefaction, these two soils reacquired bulk density, total porosity, and pore-size distribution very similar to those of the natural samples (Table 5). In contrast, the same treatment did not succeed in recreating any of these features on the fragipans that experienced periglacial conditions, namely Venegono and Maggiora. A possible reason for this discrepant behavior is that a given consolidation effect after liquefaction can occur only when a material with appropriate particle-size distribution and mineralogical assemblage is saturated by solutions able to disperse the phyllosilicates and so to promote their face-to-face arrangement.


View this table:
[in this window]
[in a new window]
  		Table 5. Schematic outline of the efficiency of freeze–thaw treatment and liquefaction in reproducing the natural features of fragipan specimens. The plus sign (+) indicates that the natural feature was reproduced by the treatment, while minus (–) indicates that the natural feature was not reproduced.

 
Whatever the cause, the densification of the deposits can be considered the first stage of fragipan formation, although the brittleness associated with this horizon may have another genesis altogether. The results reported here support the hypothesis that liquefaction produced by earthquakes may have been responsible for the consolidation of the parent materials that then became fragipans. However, it is reasonable to consider that overall fragipans are the product of the combined influence of different climatic, geomorphologic, and pedologic processes.


   ACKNOWLEDGMENTS
 
We are indebted to B. Biasiol and M. Vieri for laboratory assistance, L. Calamai and E. Menegatti for enlightening advices, J. Bouma, A.C. Edwards, G. Sanesi, and M.J. Wilson for helpful criticisms on the manuscript and D.L. Lindbo for driving the paper. Funding was provided by the Italian Ministry of University and Research (MIUR) 2000 "Indicatori chimici, mineralogici e biologici di qualità del suolo nei confronti dell'inquinamento da metalli pesanti".

Received for publication February 15, 2002.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Scalenghe, R.
Right arrow Articles by Ugolini, F. C.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Scalenghe, R.
Right arrow Articles by Ugolini, F. C.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Scalenghe, R.
Right arrow Articles by Ugolini, F. C.
Related Collections
Right arrow Soil Compaction
Right arrow Pedology
Right arrow Soil Physics

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome