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 Bravo-Garza, M. R.
Right arrow Articles by Bryan, R. B.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Bravo-Garza, M. R.
Right arrow Articles by Bryan, R. B.
Agricola
Right arrow Articles by Bravo-Garza, M. R.
Right arrow Articles by Bryan, R. B.
Related Collections
Right arrow Soil Fertility and Productivity
Right arrow Ecosystem Management
Right arrow Soil Organic Matter
Published in Soil Sci. Soc. Am. J. 69:473-481 (2005).
© 2005 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA

Division S-6—Soil & Water Management & Conservation

Soil Properties along Cultivation and Fallow Time Sequences on Vertisols in Northeastern Mexico

Maria R. Bravo-Garza* and Rorke B. Bryan

Faculty of Forestry, Univ. of Toronto, 33 Willcocks St., Toronto, ON, M5S 3B3, Canada

* Corresponding author (maria.bravo.garza{at}utoronto.ca)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Rain-fed agriculture in alternation with natural fallow is widespread in semiarid northeastern Mexico, but little information on changes in soil properties, soil degradation, and natural rehabilitation is available. The effects of rain-fed agriculture and fallow on soil quality indicators on vertisols in a semiarid area near Linares, Nuevo León, were studied. One cultivated time sequence representing 3 to 30 yr of use and one fallow time sequence of 2 to 22 yr were selected. Fifty percent of soil organic carbon (SOC) and 56% of total nitrogen (TN) were lost during the first 4 yr of cultivation, but loss reached approximate equilibrium after 6 yr. The SOC showed 34% recovery and TN showed 62% recovery on sites abandoned for 22 yr. Despite this recovery after 22 yr of fallow, SOC and TN levels reached only 50% of those observed under native vegetation. Water-stable macroaggregation declined by 14% under cultivation, but increased swiftly during fallow, showing no significant correlation with SOC. The levels of SOC and TN depletion under conventional rain-fed agriculture observed are very difficult to mitigate by natural fallows in economically-viable time periods. However, the precise impact of these changes on aggregation properties of these vertisols and the long-term sustainability of present cultivation–fallow practices is not clear. Further research to determine the precise influence of individual soil organic constituents on physical properties of these vertisols is in progress.

Abbreviations: PET, potential evapotranspiration • SOC, soil organic carbon • SOM, soil organic matter • TN, total nitrogen • WSA, water-stable aggregates


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
SOIL DEGRADATION due to land mismanagement is a major global concern and threatens economic and rural development, especially in third-world countries (El-Swaify, 1994). Arid and semiarid regions are particularly susceptible to soil degradation and often show low resilience (Seybold et al., 1998). Scant susceptible vegetation, once removed, recovers with great difficulty (Bryan, 1994) and disturbance due to population pressure, inappropriate land use, and recurrent drought all make dryland regions highly susceptible to soil degradation.

Drylands cover 80 to 90 million ha (40%) of Mexico, and are of considerable economic and ecological importance. Half the representative ecosystems are arid and semiarid, with high plant and animal diversity, and are extremely vulnerable to disturbance, with slow subsequent recovery (Arguello-Sosa et al., 1997). These characteristics are particularly significant in northeastern Mexico, where extensive subsistence rain-fed agriculture is one of the principal economic activities. Rain-fed cultivation is primarily conducted in rural communities or ejidos, where uncertain rainfall and ambitious crop production programs have encouraged landowners (ejidatarios) to convert extensive areas of natural shrubland vegetation or matorral. During the past 30 yr, accelerated deforestation and land use changes have caused clearance rates to reach 3000 ha yr–1 in some areas (Treviño et al., 1997). Ejidos typically practice land rotation with cultivation based on natural soil fertility continuing until the yield–cost ratio is no longer positive (typically 40–60 yr of conventionally-tilled cropping). The soil is then allowed to recover its fertility under natural fallow vegetation. Natural fallows are generally dominated by grasses during the first years of abandonment and Acacia spp. shrub communities or huizachales after a few years of fallow (González et al., 1997). Natural fallows typically vary from 2 to 7 yr, based on owners' needs or government policy rather that on nutrient status. Old areas are frequently added to new ones and cultivated after only a few years of fallow. New management strategies for sustainable agriculture that reduce the impact of shrubland deforestation are urgently needed in the region, but these strategies should be based on accurate understanding of soil property changes during cultivation and fallow.

The negative effects of conventional tillage on soil physical and chemical properties, and resulting soil degradation, have been widely recognized, but poorly documented in dryland regions. Observations on Australian vertisols have shown reductions of >40% in soil C fractions (Whitbread et al., 1998), >50% decrease in the amount of water-stable macroaggregates > 250 µm and reduced hydraulic conductivity, on cultivation (Blair and Crocker, 2000). Conversely, although natural fallow cycles have been used to restore soil fertility for millennia, their effects on soil property recovery are still poorly understood (Buresh and Cooper, 1999). Mbagwu and Auerswald (1999) reported that rapid percolation stability (>250 mL 10 min–1) was positively correlated with an increase on SOC on plots covered by a secondary forest > 10 yr old in Nigerian vertisols, while in semiarid areas of northeastern Mexico native N-fixing legumes might restore N and C during the fallow period in the matorral (Kass and Somarriba, 1999).

Several soil physical, chemical, and biological indicators have been proposed to assess changes in soil quality, including water-holding capacity, bulk density, total organic C and N, and microbial biomass C and N (Doran and Parkin, 1994). Soil organic matter is an early indicator of soil degradation (Campbell et al., 1999) and a valuable indicator of overall soil quality, especially in cultivated drylands where it is the most important source of plant nutrients. Organic matter is very sensitive to changes in soil management (Römkens et al., 1999) and strongly influences properties affecting resilience, such as soil structure, nutrient status, and microbial populations (Tiessen et al., 1994; Masciandaro and Ceccanti, 1999).

This project was undertaken to examine deterioration and subsequent recovery of soil quality under cultivation and natural fallow on vertisols in a representative semiarid area of northeastern Mexico. As long-term monitored plots were unavailable, the study followed the chronosequence approach (Jenny, 1941) frequently used in pedogenic studies. A chronosequence is a sequence of soils developed on similar parent material and relief, under similar climatic conditions and vegetation type, so that differences in soil character can be attributed to the lapse of time (Stevens and Walker, 1970). Although the chronosequence approach was followed, the term time sequence was employed, as it is difficult to demonstrate strict adherence to the definition of chronosequence and to avoid confusion with soil formation studies. In addition, it was adopted because many rain-fed cultivated areas of different age exist in juxtaposition with patches of fallow and native vegetation on vertisols that developed under similar climatic conditions on extensive low-relief areas. The objective was to evaluate changes in the main soil physical and chemical properties on continuous rain-fed cultivated and natural fallowed sites by integrating two time sequences on vertisols.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Experimental sites were selected near Linares, Nuevo León, on the Gulf Coastal Plains of northeastern Mexico (Fig. 1). This is a semiwarm, subhumid [Köppen classification (A)C(Wo)] climate (García, 1988), with 790 mm average annual precipitation and an annual average of 85 (±15) rain days (Návar et al., 2001). Rainfall distribution is concentrated in two rainy seasons, April to May and September, the latter being most important for agriculture. Mean annual temperature is 22.5°C, and annual potential evapotranspiration (PET) is 2200 mm (González-Rodríguez et al., 2000), giving a precipitation to PET index of 0.35, characteristic of semiarid regions. Soils were developed on Mesozoic slatey marlstones eroded from the Sierra Madre Oriental and redeposited (Ruiz-Martínez and Wegner, 1997). Pellic-vertisols (Food and Agriculture Organization of the United Nations, 1998) or fine to very-fine Udic Haplusterts (Soil Survey Staff, 1999) dominate flat gently sloping areas and are extensively used for crop production. Experimental sites were selected on vertisols in similar landscape positions (<1% slope) with similar soil characteristics and cultural practices, but showing no evidence of soil erosion and profile truncation.



View larger version (17K):
[in this window]
[in a new window]
  		Fig. 1. Area of study and location of selected sites representing rain-fed cultivation and fallow time sequences in northeastern Mexico.

 
Cultivation and Fallow Time Sequences
Cultivation and fallow time sequences were integrated using 18 study sites. In addition, three undisturbed plots with native vegetation cover, dominated by leguminous woody species such as Acacia rigidula Benth., Havardia pallens (Benth.) Britton & Rose, Eysenhardtia polystachya (Ortega) Sarg., Zanthoxylum fagara L., and Cordia boissieri A. DC. (Jurado and Reid, 1989) formed the control sites (M1–M3). Site selection was conducted on ejidos land since land management is quite homogeneous. Ejidatarios traditionally grow the same crops, share the same markets, and were provided with similar machinery during a government program in the 1970s. The homogeneous basis for the time sequence methodology was established by soil profile examination and sample analysis at 11 sites, using the Food and Agriculture Organization of the United Nations system of description and classification (Siebe et al., 1996) (Table 1).


View this table:
[in this window]
[in a new window]
  		Table 1. Description of a representative soil profile of a control site (M1), an 18-yr rain-fed cultivation plot (A5), and a 2-yr fallow site (F1), belonging to soil time sequences in a semiarid area of northeastern Mexico according to Food and Agriculture Organization of the United Nations soil classification system. Single samples were analyzed per horizon.

 
The cultivation time sequence included nine sites (A1–A9) spanning 3 to 30 yr of continuous rain-fed cultivation (Table 2). This was intended to cover the typical cultivation period in the area and the time interval when the most dramatic changes in soil properties usually occur (Cambardella and Elliott, 1992). The main characteristics of all the cultivated plots included (i) sites with no record of past disturbance, covered by native vegetation before rain-fed cultivation; (ii) sites managed with similar cultivation intensity and tillage practices; and (iii) sites where there have been no regular additions of pesticides or fertilizers. Because rain-fed cultivation is restricted by climatic conditions, two crops may be grown in a wet year, but usually only one crop is grown during the fall. Typical tillage involves soil inversion by moldboard plowing to the 20- to 25-cm depth performed in April to May for fall cropping, and/or in December for spring cropping. Seedbed preparation involves one or two disc harrows followed by planting using tine cultivators. Tillage is carried every year in preparation for any rainfall event even if, in some years, a crop is not grown. All the cultivated plots selected were used for maize (Zea mays L.) production, with exception of A1, where bean (Phaseolus vulgaris L.) was the only crop. However, farmers mentioned occasionally growing beans, sorghum [Sorghum halepense (L.) Pers.], and/or peanuts (Arachis hypogaea L.) in other plots (Table 2).


View this table:
[in this window]
[in a new window]
  		Table 2. Main crops and dominant vegetation occurring in 30-yr rain-fed cultivation and 22-yr fallow time sequences.

 
Fallow time sequence sites were selected from old rain-fed fields cultivated for >20 yr, then abandoned for several years. Nine fallow sites (F1–F9) were selected (Table 2). The F1 to F5 covered the most common fallow period in the area (2–9 yr), while the F6 to F9 fallow duration exceeded 10 yr. Information on cropping condition and fallow duration was obtained through interviews with landowners. This information was cross-checked by examination of the age and morphology (diameter, height, cover) of trees in secondary plot vegetation using Nájera's (1999) mathematical model for matorral species (Table 2).

Soil Analysis
Soil samples were randomly collected from the 0- to 15-cm depth on 50- by 50-m plots at each site. Samples were air dried and passed through a 2-mm sieve for pH and particle size analysis, and through a 500-µm sieve for SOC and TN analysis. Three to five samples per plot were used to determine soil reaction (pH) and particle size analysis. The former was determined in a 1:2 soil–water suspension, and the latter by the pipette method (Gee and Bauder, 1986) with sand fractions separated by sieving. The SOC was evaluated on three to five samples, with three replicates each by the Walkley-Black wet combustion method (Nelson and Sommers, 1982), and TN on three to 10 samples by the micro-Kjeldahl digestion procedure (Bremner and Mulvaney, 1982) and an autoanalyzer (Autoanalyzer AAII, Technicon Engineering, Atlanta, GA). Aggregate stability was tested on three to five samples per plot by wet sieving (Yoder, 1936). Twenty to 30 g of air-dried soil aggregates > 2 mm were placed on the top of a nest of four sieves (2.00, 1.00, 0.50, and 0.25 mm) and immersed in water for 1 min, then wet sieved for 20 min at a speed of 71 oscillations per minute. After sieving, the water-stable aggregates remaining on the top of the sieves were oven-dried and weights of the aggregates were corrected by subtracting the weight of sand. Four water-stable aggregate (WSA) classes: 2.00 to 1.00, 1.00 to 0.50, 0.50 to 0.25, and <0.25 mm were obtained, and three replicates per sample were conducted.

Because of the impossibility of getting undisturbed soil samples for water retention capacity determination, repacked samples were used following Richards (1965). Three randomly selected samples per plot were air dried and passed through a 2-mm sieve (not crushed) to fill 1-cm-deep rings. Samples were presaturated >8 h on ceramic plates before applying 0.1, 0.3, and 1.5 MPa of pressure. Three replicates per sample were conducted.

Infiltration capacity was determined in triplicate at field sites with a set of two steel thin-wall rings, with 20- and 30-cm diam., respectively, and 30 cm tall (Bouwer, 1986). Rings were inserted approximately 5 cm in the soil, and a falling head infiltration in the inner ring was recorded.

Statistical Analysis
Multiple regression analyses were performed with all data obtained, and were used to explain the trend in the soil chemical and physical properties along the time sequences. Regression models were fitted using SigmaPlot for Windows 6.00 (SPSS Science, 2000). Data analysis included calculation of variation between sites. However, despite attempts to obtain more than one site for each selected land use period in each time sequence (true replicates), this proved impossible for all sites. Therefore, for comparison purposes, the ANOVA was performed with data from sites having at least two replicates. The undisturbed sites (M1, M2, M3), the 4-yr plots (A2, A3), and the 30-yr plots (A7, A8, A9) were included in the analysis of the cultivation time sequence. While for the fallow time sequence, the 2-yr sites (F1, F2), 16-yr sites (F6, F7), and the 22-yr sites (F8, F9), as well as the undisturbed sites, were used in the analysis. SigmaStat for Windows 2.03 (SPSS Science, 1997) was used for ANOVA, and means were tested using Tukey's significant differences at {alpha} = 0.05.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Examination of the SOC content in the soil profiles shows clearly a similar distribution of C with depth and that changes in C content between control sites and cultivated and fallow sites have primarily occurred near the soil surface, presumably due to differences in soil management (Fig. 2).



View larger version (23K):
[in this window]
[in a new window]
  		Fig. 2. Distribution of soil organic C along soil profiles in 30-yr rain-fed cultivation and 22-yr fallow time sequences. M1 and M3 represent two control sites.

 
Soil alkalinity increased slightly but significantly during the period of cultivation (P = 0.041). Statistical significant differences (P = 0.038) were found between the average of control sites and the 30-yr plots (7.82) (Table 3). No difference was observed between the pH values of 4-yr plots (7.73) and either control and 30-yr plots P = 0.675 and P = 0.154, respectively. Conversely, results along the fallow time sequence were highly variable and statistically not significant (P = 0.261). Two-year fallow sites had mean pH values of 7.76, while 16-yr fallow sites had a mean pH of 7.69. After 22 yr, fallow pH averaged 7.73 (Table 3). The high buffer capacity of vertisols may have prevented high fluctuations in soil pH (Woerner, 1991) and the greater pH under cultivation may reflect a loss of SOM exchange sites. The fine clay-rich texture of the vertisols was confirmed by particle size analysis, and most soils were silty-clay (USDA system) (Table 3).


View this table:
[in this window]
[in a new window]
  		Table 3. Soil characteristics of control sites, along a 30-yr rain-fed cultivation time sequence, and along a 22-yr fallow time sequence in vertisols from northeastern Mexico. Values in parentheses represent standard deviation.

 
Change in Soil Organic Carbon and Total Nitrogen with Length of Cultivation and Fallow
Perhaps the most important and well-documented effect of cultivation on soil is reduction in SOM content (Nye and Greenland, 1960; Smith and Elliott, 1990). Generally, the decay rate varies with differences in management intensity, climate, and soil characteristics (Burke et al., 1989). In this study, where semiarid conditions and tillage practices increased SOM decomposition, SOC and TN declined swiftly during the first 6 yr of cultivation, followed by an apparent equilibrium (Fig. 3a). Control sites showed a mean of 40.8 g kg–1 SOC, while 20.05 g kg–1 was observed on cultivated 4-yr-old plots, indicating a significant 50% reduction (P < 0.001), while there was no statistically significant difference (P = 0.262) between SOC in 4- and 30-yr-old plots (15.75 g kg–1). The final reduction from initial SOC content after 30 yr of continuous cultivation was 60%. According to a best-fit hyperbolic decay function for SOC dynamics during cultivation (r2 = 0.95), the first year of cultivation caused the highest SOC loss (9.46 g kg–1), declining to 2.63 g kg–1 yr–1 from Years 2 to 3, and maintaining a quasiequilibrium of 0.70 g kg–1 yr–1 after Year 6 (Fig. 3a). Observed SOC decline rates agree with values of 20 to 60% reported from other studies (Tiessen and Stewart, 1983; Burke et al., 1989), but were faster than reported by Paustian et al. (1997) and Haynes and Tregurtha (1999).



View larger version (15K):
[in this window]
[in a new window]
  		Fig. 3. Soil organic carbon (SOC) in time sequences of (a) 30-yr rain-fed cultivation and (b) 22-yr fallow sites in vertisols from northeastern Mexico. Bars represent confidence intervals (P = 0.05).

 
In contrast with SOC depletion rates during cultivation, recovery was slow under fallow (Fig. 3b). After 2 yr of fallow, SOC averaged 15.97 g kg–1, increasing to 20.48 and 21.47 g kg–1 after 16 and 22 yr, respectively (increase of 28 and 34%, respectively). The SOC recovered at the end of the fallow period (21.47 g kg–1) was only 53% of the mean SOC (40.8 g kg–1) at the control sites, indicating very slow natural recovery of SOC under dryland conditions.

Total N showed similar behavior to SOC under cultivation and fallow, decreasing 56% (from 2.52 g kg–1 to 1.11 g kg–1) during the first 4 yr of cultivation (Fig. 4a). After 30 yr of continuous land use, mean TN dropped to 0.95 g kg–1 (62%), statistically no different (P = 1.00) from the 4-yr-old plots. According to a rational regression model (r2 = 0.83), the greatest reduction (1.37 g kg–1) in TN occurred during the first year of cultivation, declining to around 0.017 g kg–1 yr–1 by Year 4, 0.007 g kg–1 yr–1 by Year 6, and 0.0003 g kg–1 yr–1 by Year 30. On the other hand, an increase of 47% in the TN content was observed after 16 fallow years (Fig. 4b). A significant difference (P = 0.028) was observed between the 2-yr (0.781 g kg–1) and the 16-yr fallow plots (1.145 g kg–1), while no difference was observed between the 16- and 22-yr fallow plots (P = 0.478). Total N recovered after 22-yr fallow was 1.269 g kg–1, about half of the content before cultivation.



View larger version (14K):
[in this window]
[in a new window]
  		Fig. 4. Total soil nitrogen (TN) in time sequences of (a) 30-yr rain-fed cultivation and (b) 22-yr fallow sites in vertisols from northeastern Mexico. Bars represent confidence intervals (P = 0.05).

 
The hyperbolic decline patterns observed for SOC and TN after several years of cultivation (Fig. 3a and 4a) are consistent with initial depletion of the active and easily decomposable soil organic matter fractions, leaving a residue of more resistant fractions (Six et al., 2000). A labile or active fraction, composed of free organic matter not intimately associated with mineral soil constituents (Magdoff, 1996), has been found to be strongly affected by cultivation practices, becoming depleted after only a few years of cultivation (Tiessen and Stewart, 1983; Cambardella and Elliott, 1992). After such depletion, remaining humic substances are more resistant to decomposition and are normally protected from microbial attack in soil microaggegates < 250 µm (Jastrow and Miller, 1997; Six et al., 1998). These residual components were not analyzed in this study, but other reports suggest this fraction is mainly characterized by clay–organic complexes formed between humic substances and smectites (Dalal and Bridge, 1996; Jastrow and Miller, 1997; Six et al., 2000).

Formation of clay-organic complexes has been reported as the main mechanism for SOC storage in vertisols, and >50% of the SOM content has been found in the clay-size fraction (Dalal and Bridge, 1996). Leinweber et al. (1999) found that 70 to 91% of organic C in vertisols was stored in the clay fraction, 8 to 28% in the silt fraction, and 1 to 28% in the sand fraction. The authors concluded that rapid decomposition of plant residues in warm climates, combined with shrink–swell pedoturbation, influenced the distribution and composition of organic matter in vertisols. In this study, formation of strong SOM–clay complexes may have prevented complete SOC depletion, accounting for the relatively high level of SOC preservation after 30 yr of cultivation (1.6% of SOC {approx} 2.7% SOM).

Changes in Soil Aggregation and Hydraulic Soil Properties
Physical properties on vertisols are particularly affected by the moisture content and show great seasonal or annual variation (Coulombe et al., 1996). In this study, the shrink–swell phenomenon that is attributed to changes in interparticle and intraparticle porosity on wetting and drying (Coulombe et al., 1996) might have contributed to the high variability of soil aggregation, infiltration and water retention capacities, and the inconclusive statistical results.

Although no statistically significant differences in the amount of water-stable macroaggregates > 250 µm were detected between cultivated and control plots with length of cultivation (P = 0.193), a trend of reduction can be observed on the values of Table 3 and Fig. 5a. A reduction in WSA > 250 µm occurred along the cultivation time sequence, from an average of 80% at control sites (Table 3) to 66% at 30-yr-old plots (Table 3). Large aggregate sizes showed the greatest reductions, a decrease of 18 to 29% for the WSA > 2000 µm and between 2 to 9% for the 1000- to 2000-µm-diam. WSA (data not shown).



View larger version (17K):
[in this window]
[in a new window]
  		Fig. 5. Distribution of water-stable macroaggregates > 250 µm in time sequences of (a) 30-yr rain-fed cultivation and (b) 22-yr fallow sites in vertisol from northeastern Mexico. Bars represent confidence intervals (P = 0.05).

 
Breakdown of macroaggregates > 250 µm under tillage has been widely reported (e.g., Beare et al., 1994; Six et al., 1998), and in vertisols could contribute to high rates of soil erosion. Návar and Synnott (2000) reported a 41-fold increase erosion under intense rainfall on cultivated plots in the study area, compared with native vegetation plots. Thus, soil structure preservation is crucial on vertisols where physical characteristics such as high swelling, reduced infiltration, and enhanced slaking may promote erosion.

On the other hand, no significant differences were observed in the amount of water-stable macroaggregates during the fallow stage (P = 0.84). However, the values in Table 3 (with exception of F1) and Fig. 5b indicate clear improvement in aggregation. Large macroaggregates > 2000 µm increased 32% from 2- to 22-yr fallow sites, while water-stable microaggregates < 250 µm dropped from 28 to 13% during the same period (data not shown). In total, water-stable macroaggregates > 250 µm increased 14% after 22 yr of fallow.

Although the influence of SOM on soil aggregation has been demonstrated in many studies, the data for control sites show that the relationship is not always clear or direct (Fig. 3a and 5a). McGarry (1996) does not consider that organic matter per se is important for structural development in vertisols, but the more labile fractions may have a positive influence. The importance of materials such as fungal hyphae, bacterial cells, algae, roots, and microbial and plant-derived polysaccharides in binding soil macroaggregates that form part of the labile SOM fraction has been widely recognized (Beare et al., 1994; Angers and Chenu, 1997; Six et al., 1998). Although this relationship has been demonstrated for several soil types, no significant correlation was observed between the amount of macroaggegates and SOC in this study (P = 0.06). These results support Whitbread et al. (1998) and Blair and Crocker (2000), who reported no significant correlation between aggregate stability and any SOM fraction, including the active fraction, on an Australian black Vertisol. Possible explanations include (i) negligible or limited influence of SOC in water-stable aggregation, or the importance of only some SOM fractions; (ii) greater influence of clay minerals on soil aggregation in clay-rich soils compared with organic materials (Blair and Crocker, 2000); and (iii) greater importance of the shrink–swell process on structure formation on vertisols, so that wetting and drying cycles act as a mechanistic method of counteracting compaction in such soils (Pillai and McGarry, 1999).

High variability and not significant differences were reported in the final soil infiltration capacity data for the control sites and the two time sequences (Table 3), showing the difficulty of measuring this property in vertisols. Besides known disadvantages of the method, such as a small area of measurement (0.07 m2), lateral divergence of flow, soil disturbance by cylinder insertion, and absence of raindrop impact, possible explanations might include differences in surface conditions at the time of measurement, or in crack size and density. In this study, some sites were recently plowed for weed control (A5 and A6), potentially increasing soil porosity and measured infiltration rates, while others were plowed several months before the measurement. Návar and Synnott (2000) attributed similar variability in infiltration rates on cultivated plots to the presence of larger, more abundant cracks on tilled plots than on undisturbed plots. This observation agrees with the results obtained at two control sites (M1 and M2). Overall, infiltration rates on fallow sites were higher than cultivated plots (Table 3), which agree with the trend for aggregates stabilization.

Despite observed differences in SOC between land uses, no significant difference was observed in soil water retention capacity between 0.03 and 1.5 MPa, P = 0.803 and P = 0.285 for the cultivation and fallow time sequences, respectively. Undisturbed soils showed mean water retention of 0.129 kg kg–1 soil, while 30-yr cultivated plots showed 0.116 kg kg–1 (Table 3). Under fallow, 2 yr-old sites averaged 0.134 kg kg–1, and averages of 0.138 and 0.109 kg kg–1, respectively, were observed after 16 and 22 yr (Table 3). The surprisingly low water retention on the oldest fallow site may reflect obstruction of swelling capacity by SOM, inhibiting the absorption of water molecules in the clay interlayer space (four H2O layers). Davidson and Page (1956) reported a 52% increase in soil swelling pressure in Houston Black Clay (vertisol) after SOM removal. However, the general effect of SOM on the shrink–swell capacity of vertisols is still unclear and requires further study. In this study, disruption of soil structure combined with the capacity of smectite clays to absorb several times their own weight in water (Borchardt, 1989) produced water contents > 100% at the saturation point (>1.0 g H2O g–1 soil), much higher than normally reported for vertisols in natural conditions.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Continuous rain-fed cultivation with conventional tillage on vertisols in semiarid areas of northeastern Mexico promoted significant reduction on SOC and TN. A hyperbolic decline pattern is consistent with initial depletion of active, easily decomposed soil organic matter fractions, leaving a more resistant residue. The exact composition of soil organic matter in vertisols is still unclear and requires more study.

Soil organic matter depletion appears to significantly affect soil aggregation, as shown by WSA proportions on cultivated land, but its influence on aggregation during long fallow periods is not clear. In addition, recovery of SOC and TN through natural revegetation occurs at very slow rates, indicating a need for very long, uneconomic fallow periods. However, soil physical properties may recover faster than those of soil fertility, due to shrink–swell processes combined with root and biological activity.

In view of the strong influence of soil moisture content on determinations of physical properties in vertisols, several issues should be considered before using the standard methodology for nonswelling soils. Determinations involving soil volume, such as bulk density, infiltration capacity, and water retention capacity will be seriously affected by changes in soil moisture due to the shrink–swell behavior. Evaluation of water retention capacity through the repacked method (Richards, 1965) is also affected, since the pore system is disrupted and the increase of volume on saturation may exceed the initial sample core volume. Modified, improved analytical and sampling methodology is a clear priority for sustainable management of these fertile but physically problematic soils.


   ACKNOWLEDGMENTS
 
We thank the Faculty of Forest Sciences of the University of Nuevo León in Mexico for support in the fieldwork, with special thanks to the faculty members Jose Návar, Humberto González, Israel Cantú, and Eduardo Treviño. This research was supported by CONACYT, Mexico, and an operating grant to R.B. Bryan from the Natural Sciences Engineering and Research Council, Canada.

Received for publication December 22, 2003.


   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 Bravo-Garza, M. R.
Right arrow Articles by Bryan, R. B.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Bravo-Garza, M. R.
Right arrow Articles by Bryan, R. B.
Agricola
Right arrow Articles by Bravo-Garza, M. R.
Right arrow Articles by Bryan, R. B.
Related Collections
Right arrow Soil Fertility and Productivity
Right arrow Ecosystem Management
Right arrow Soil Organic Matter

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