HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (15) Citing Articles via Google Scholar Google Scholar Articles by Lilienfein, J. Articles by Zech, W. Search for Related Content PubMed Articles by Lilienfein, J. Articles by Zech, W. Agricola Articles by Lilienfein, J. Articles by Zech, W.

DIVISION S-6-SOIL & WATER MANAGEMENT & CONSERVATION

Annual Course of Matric Potential in Differently Used Savanna Oxisols in Brazil

^a Inst. of Soil Science and Soil Geography, Univ. of Bayreuth, D-95440 Bayreuth, Germany
^b CIAT-Laderas, Apartado 1410, Tegucigalpa, Honduras
^c Federal Univ. of Uberlândia, 38406-210 Uberlândia-MG, Brazil
^d EMBRAPA-Cerrado, Caixa Postal 08223, Planaltina-DF, Brazil

julia.lilienfein{at}uni-bayreuth.de

	ABSTRACT

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 
Sustainable land use in periodically dry Brazilian savannas requires a water-saving management. We hypothesized that the annual course of matric potentials (_M) in very-fine, isohyperthermic Anionic Acrustoxes of Pinus plantations (PI), degraded (DP) and productive pastures (PP), no-till (NT) and conventional tillage (CT) cropping, and natural savanna (Cerrado, CE) differed significantly. On three plots in each of these land-use systems water input and _M at the 0.15-, 0.30-, 0.80-, 1.2-, and 2.0-m depths was measured with tensiometers weekly between 27 Mar. 1997 and 28 Apr. 1998. Precipitation between 29 Apr. 1997 and 28 Apr. 1998 was 1562 mm, with only 210 mm in May to September, when _M at the 0.15- and 0.30-m depths decreased to less than -80 kPa in all systems; the lowest _M at 2-m depth was -57 kPa. During the monitored period, the PI soils had lower average _M at the 0.8- to 2-m depths (-60 kPa) than those in CE (-46), indicating higher rainfall interception losses and higher transpiration. In CT, average _M values at the 0.8- to 2-m depths (-29) were higher than in NT (-51) because of different crops and different soil management. Between June and November, _M at the 2-m depth in CE decreased to a lower value (-42) than in vegetation-free CT (-22) and NT (-27). In DP and PP soils, _Ms were similar to those in CE soils at all depths. The estimated average water storage in the upper 2 m during the monitored period was: 565 mm (CT) > 553 (PP) > 541 (DP) > 537 (CE) > 526 (NT) > 479 (PI). Our results show that mainly the vegetation type and tillage practices control the annual course of matric potential in differently used savanna Oxisols.

Abbreviations: Al_o, oxalate-extractable Al • CE, Cerrado • CT, conventional tillage • DP, degraded pasture • Fe_d, dithionite-citrate-extractable Fe • NT, no-till • PI, Pinus caribaea plantation • PP, productive pasture • _M, matric potential

	INTRODUCTION

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 
IN THE LAST 30 YR, intensive agriculture developed on the nutrient-poor soils of the Brazilian savanna region (the Cerrados). Large parts of natural Cerrado were cleared and afforested or transformed to pasture and cropping systems (Goedert, 1983). Agriculture in this region is water-limited. In soils under native vegetation, matric potentials at 0.5-m soil depth decrease below the permanent wilting point, that is, the matric potential at which plants wilt, only one month after the end of the rainy season (Franco et al., 1996). If dry periods occur during critical stages of crop growth, such as flowering, water stress may result in severe yield losses (Goedert, 1983). Land-use practices which result in changed water budgets may therefore affect crop yields and, in the long run, may also have impacts on the groundwater level and storage.

The soil water content is influenced by the water consumption of plants and by soil management (McGowan and Williams, 1980a, 1980b; Azooz et al., 1996). In Great Britain, McGowan and Williams (1980a, 1980b) found decreasing evapotranspiration along the line: woodland > pasture > barley. Because forests generally consume more water and have a greater interception capacity resulting in higher evaporation losses, lower soil water contents are often found in soils under forest than under cropping systems or pastures (Blume and Zimmermann, 1975; Hodnett et al., 1995). In a study by Lima (1983), conducted in the state of São Paulo, Brazil, the water content of the uppermost 2 m of soils under different Pinus species was significantly lower than under natural Cerrado vegetation. Pinus forests use up to 93% of the annual precipitation for transpiration (Schiller and Cohen, 1998). Lima et al. (1990) reported a higher annual evapotranspiration of 5-yr-old Pinus caribaea Morelet plantations (716 mm yr^-1) than of native Cerrado vegetation (576 mm yr^-1) in Brazil. Hodnett et al. (1995) compared evapotranspiration rates of Amazonian rain forest and pasture and found similar evapotranspiration rates of 4 mm d^-1 at high soil water contents. At low soil water contents, evapotranspiration rates in pasture were substantially reduced to a minimum of 1.2 mm d^-1, while those of the forest remained above 3.5 mm d^-1. In cropping systems transpiration rates depend on crop species and yield. The economically most important crops cultivated in the study region are corn (Zea mays L.) and soybean [Glycine max (L.) Merr.]. The transpiration coefficient (i.e., the ratio between dry biomass and transpiration water loss) of soybean is almost two times (>700 kg H₂O kg^-1 dry biomass) that of corn (300–400 kg kg^-1; Geisler, 1980).

In CT soils, pore continuity is reduced compared with NT soils, resulting in reduced evaporation (Lal et al., 1980; Logsdon et al., 1990; Azooz et al., 1996). Azooz and Arshad (1996) reported higher infiltration rates and a higher saturated hydraulic conductivity in NT than in CT soils. While the volume of macropores (>14 µm) in the study of Azooz and Arshad (1996) was not significantly different between NT and CT soils, Hill (1990) found a higher macropore volume (>15 µm) in CT than in NT soils and, in contrast to the conclusions of Azooz and Arshad (1996), inferred faster drainage in the former than in the latter. Furthermore, the pore space for plant-available water was greater in CT than in NT soils (Hill, 1990). The effects of tillage practices also depend on climatic conditions. For example, during humid winters in Great Britain, Goss et al. (1978) found no difference in water content between the tillage variants, whereas in dry winters more water was stored in NT systems than in the tilled ones.

The objective of this work was to compare the annual course of matric potentials in differently used Brazilian savanna Oxisols and to evaluate the systems with regard to water availability for plants and water storage in the rooting zone.

	Materials and methods

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 
Study Sites
The study area is located southeast of Uberlândia (State of Minas Gerais) 400 km south of Brasilia (Fig. 1) . Within an area of 100 km², three plots of each of the following six land-use and natural systems were selected: (i) Pinus plantations (PI), (ii) degraded (DP) and (iii) productive pastures (PP), (iv) no-till (NT) and (v) conventional tillage (CT) cropping, and (vi) natural savanna (Cerrado, CE). We considered these land-use systems the most important ones in the study region. To allow for statistical evaluation with variance analysis we aimed at selecting independent replicates of each system. As our objective was to conduct an on-farm experiment, we had to select the experimental plots in existing land-use systems. Thus, an entirely randomized plot selection was not possible. However, we only chose replicate plots of each land-use and natural system which were separated from each other by a distance of at least 300 m and, except for PI, by an area which was differently used between the replicate plots. The PI forest covered a large area without intermixed plots with different land-use. We assume that the prerequisites for variance analysis have been met by this experimental design.

View larger version (122K): [in this window] [in a new window]   Fig. 1 Locations of the study sites (CE, native Cerrado; PI, Pinus; NT, no-till cropping; CT, conventional tillage; DP, degraded pasture; and PP, productive pasture)

All study sites had slopes below 1°; they have been continuously used for the same purposes for 12 (DP, PP, NT, CT) or 20 (PI) yr and passed directly from natural vegetation to the current land-use system except for the NT soils. All study soils developed from fine limnic sediments of the lower Tertiary. The soils were homogeneously weathered to a depth of several meters.

Equipment and Measurements
On each of the 18 plots a 10 by 10 m area was fenced and equipped with five replicate tensiometers at each of 0.15-, 0.3-, 0.8-, 1.2-, and 2-m depths. All plots were equipped with five rain collectors consisting of a sampling bottle and a funnel with a diameter of 115 mm at 0.3-m height above the soil surface to measure soil water input. Each sampling bottle was protected against larger particles and small animals with a polyethylene net (0.5-mm mesh width). A table-tennis ball was used to reduce evaporation. Tensiometers and rain collectors were placed within and between the crop rows in the cropping systems and near and between the trees in CE and PI. During the rainy season (March–April 1997 and October 1997–April 1998) and during the dry season (May–October 1997), _Ms were read and precipitation measured every 7 and 14 d, respectively. The soils in CT were hoed manually instead of plowing on 9 October and 21 Nov. 1997 within the fenced plots as tensiometers could not be removed. Weed in NT was controlled by the application of 1.2 kg ha^-1 glyphosate [N-(phosphonomethyl)glycine; Roundup, Monsanto, St. Louis, MO] on 27 Nov. 1997.

Meteorological Data
Precipitation data, and daily maximum and minimum temperatures were provided by the Fazenda Pinusplan, a commercial farm producing mainly corn, soybean, and Pinus. The meteorological station of the farm is located near Plot CE2 and CT2 (Fig. 1) in the study area.

Soil Physical Characterization
Particle-size distribution was determined in the 0- to 0.15-, 0.15- to 0.3-, 0.3- to 0.8-, 0.8- to 1.2-, 1.2- to 2-m layers after removal of the oxides with Na-dithionite and of organic matter with H₂O₂ at 90°C, following dispersion with hexametaphosphate. Coarse and fine sand were sieved to 250 to 2000 and 50 to 250 µm, respectively. Silt and clay concentrations were determined with the pipette method (Gee and Bauder, 1986). Soil density was determined gravimetrically by taking five undisturbed 100-mL soil cores.

To determine soil water characteristic curves, five undisturbed soil samples were taken from the 0- to 0.15-, 0.15- to 0.3-, 0.3- to 0.8-, 0.8- to 1.2-, 1.2- to 2-m soil layers with 100-mL steel rings from one selected soil in each of CE and NT (CE1 and CT2 in Fig. 1). The five replicate cores per layer were taken from various depths to account for the heterogeneity within one layer. The gravimetrical soil water content was determined at _Ms of -0.316, -1.0, -3.16, -10.0, and -31.6 kPa after equilibrating on ceramic plates. Water content at -1584.8 kPa was determined in a pressure pot using disturbed samples (Klute, 1986).

Soil Chemical Characterization
Soil organic carbon (SOC) was determined with a CHNS-analyzer (Vario EL, Elementar Analysensysteme GmbH, Hanau, Germany). Aluminum in poorly ordered Fe oxides was extracted with the oxalate-buffer method (Al_o; Schwertmann, 1964), crystalline Fe oxides with the cold dithionite-citrate-buffer (DCB) method (Fe_d; Holmgren, 1967). Aluminum and Fe were measured with flame atomic absorption spectrophotometry (Varian AA 400, Varian, Mulgrave, Australia).

Calculations and Statistical Evaluation
To fit soil water characteristic curves to the measured values, the Soil Hydraulic Properties Fitting (SHYPFIT) program (Durner, 1994) was used. The soil water characteristic curves obtained for a specific layer of one soil in each of CE and NT were averaged to estimate water storage in the soil layers. Estimates were calculated for the dry season (May–October 1997) and the rainy season (November 1997–April 1998) separately:

(1)

where WS_sl is the water storage in the soil layer (mm), _sl the soil bulk density (Mg m^-3), l_sl the depth of the soil layer (m), and _sl average water content in the soil layer (g kg^-1).

Water storages of the soil layers were summed to get the water storage of the uppermost 2 m:

(2)

Main and interactive means of the soil water characteristic curves, Fe_d and Al_o concentrations, texture, and the calculated water contents were tested with Tukey's honestly significant difference (HSD) mean separation test (Hartung and Elpelt, 1989). Mean _Ms were tested with the Wilcoxon matched pairs test. Significance was set at P < 0.05. Statistical analyses were performed with STATISTICA for Windows 5.1 (StatSoft, 1995, Hamburg, Germany).

	Results and discussion

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 
Comparability of the Soils
Differences in _Ms between land-use systems can only be interpreted as caused by land-use practices if the soils were similar prior to the beginning of land use. To test this prerequisite, we compared soil characteristics that are not or only little influenced by land use. Such characteristics include the soil classification, the particle-size distribution, and the mineralogical composition. As indicators of the mineralogical composition, we determined the concentration of Fe_d, which is a measure for the concentration of Fe oxides formed after the release of Fe from primary minerals (Mehra and Jackson, 1960). Additionally, we used the concentration of Al_o as a measure for the concentration of the Al that isomorphously substitutes for Fe in Fe oxides (Parfitt and Childs, 1988).

All studied soils were very-fine isohyperthermic Anionic Acrustoxes (Soil Survey Staff, 1997) with high clay concentrations (615–885 g kg^-1). There was no significant change in clay content with increasing soil depth in any individual soil (data not shown). Clay, silt, coarse, and fine sand concentrations were not significantly different between the land-use systems within any individual soil layer (Table 1) . The Al_o and Fe_d concentrations of the same depth layer were not significantly different between any of the studied soils of all land-use systems. Additionally, the average standard deviations of the Al_o and Fe_d concentrations between the replicate plots of one system (Fe_d: 9.1 g kg^-1, Al_o: 0.27 g kg^-1) were similar to those between all soils (Fe_d: 9.2, Al_o: 0.33). Thus, we assume that the soils had similar properties prior to land use, and differences in _M and water storage can be interpreted as the result of land-use practices, which also are the reason for small but significant differences in SOC concentrations.

View larger version (19K): [in this window] [in a new window]   Fig. 2 Precipitation between 1 Apr. 1997 and 28 Apr. 1998 recorded at the Fazenda Pinusplan, a commercial farm

View larger version (90K): [in this window] [in a new window]   Fig. 3 Course of matric potentials (M) at (a) 0.15-, (b) 0.30-, (c) 0.80-, (d) 1.20-, and (e) 2.00-m depth in soils of native Cerrado (CE), Pinus (PI), no-till cropping (NT), conventional tillage cropping (CT), degraded pasture (DP), and productive pasture (PP) between 27 Mar. 1997 and 28 Apr. 1998. Shown values are averages of three replicates per treatment

Average _M of the whole monitored period at the 0.15-m depth were not significantly different between CT, NT, and CE (Fig. 3a). Average _M in CT were significantly higher than in CE at the 0.3-, 1.2-, and 2-m depths (Fig. 3b, 3d, and 3e) and in NT at 0.3- to 2-m depths (Fig. 3b–3e). Average _M in NT at the 0.8- and 1.2-m depths were significantly lower than in CE (Fig. 3c and 3d). However, differences in soil _M between the cropping systems and CE varied in the course of the year.

The reasons for differences in _M between CE, CT, and NT are related to different plant cover and different soil management. Between the 15 March and the end of June 1997, the NT soils were cropped with corn while the CT soils were without crops. The water consumption of the corn probably lowered _M. This may, in part, also be the reason for the lower _M in NT than in CT at the 0.3- to 2-m depths during the dry season (June–November 1997; Fig. 3b–3e). However, evaporation from NT soils may also have been higher than from CT soils because plowing destroys the pore continuity and thus reduces evaporation (Azooz and Arshad, 1996). After the harvest in CT and in NT, the water consumption of the partly evergreen vegetation in CE was higher and _M consequently decreased to a lower value than in the cropping systems.

Lower _M at the 0.8- to 2-m depths in NT than in CT during the period of high biomass production (February–April 1998; Fig. 3c, 3d, and 3e) can be explained by higher transpiration coefficients of soybean than of corn (Geisler, 1980) and similar average biomass production (soybean: 11.3 Mg dry mass ha^-1, corn: 12.8 Mg ha^-1) on our plots during the 1997-1998 vegetation period. Lower _M in NT than in CT soils could also be attributable to a lower infiltration rate and a lower saturated hydraulic water conductivity in NT than in CT, and therefore to slower soil water refill, particularly at greater depth. However, infiltration in all study soils was rapid, and we never observed any surficial water even during strong rainfall events. Evaporation losses from surficial water may thus be excluded. Furthermore, NT soils should display a higher water conductivity than CT soils because of the higher pore continuity, and we measured a higher soil water input into NT soils than into CT soils (Table 2). An additional reason for lower _M in NT than in CT between February and April 1998 may be the lower water storage in NT soils as a result of water consumption at the end of the preceding cropping period.

Differences in _M between the pasture systems and CE were smaller than between the other systems and CE. There were no significant differences in average _M of the monitored period at the 0.15- and 0.3-m depths between CE, PP, and DP (Fig. 3a and 3b). Between 0.8- and 1.2-m depths, small but significant differences between the three land-use systems were recorded (Fig. 3c and 3d). Between December 1997 and April 1998, the average _M in DP at the 0.8- and 2-m depths were significantly lower than in PP and CE (Fig. 3c and 3e); at the 1.2-m depth, the average _M in DP was significantly lower than in PP (Fig. 3d). Between May and November 1997, average _M at the 0.8- to 2-m depth in PP were significantly lower than in DP (Fig. 3c–3e) and also significantly lower than in CE (Fig. 3c and 3d). Lower _M in DP than in PP at greater depth during the rainy season may be related to the water uptake of the invaded natural Cerrado brushes and trees, which may have a greater rooting depth than the grass in PP. In contrast, differences in _M between the pasture systems during the dry season indicate a higher water consumption of the PP vegetation, which is denser and shows more intensive grass rooting.

Water Storage
The soil water content can be estimated from _M with the help of the soil water characteristic curve. The shape of the soil water characteristic curve is influenced by soil texture and bulk density. We found that there were only small differences in texture between the studied soils and between the soil depths within one soil (Table 1). At the 0- to 0.15-m soil depth, the average soil density in PI was significantly smaller than in NT, CT, and DP. There was no other significant difference in soil density between the various layers of the soils of the same land-use system and between the same layers among the different land-use systems. As a result of the great similarity in soil texture and bulk densities among the studied soils, we did not expect significant differences in the soil water characteristic curve. Therefore, we chose only one soil in each of the CE and CT treatments (at the plots CE1 and CT2 in Fig. 1) to determine the soil water characteristic curves for five soil layers of each of the selected two soils (Fig. 4) .

View larger version (24K): [in this window] [in a new window]   Fig. 4 Measured (data points) and fitted (lines) soil water characteristic curves in two selected soils in each of Cerrado (CE) and conventional tillage (CT) treatments (CE1 and CT2 in Fig. 1) at 0- to 0.15-, 0.15- to 0.30-, 0.30- to 0.80-, 0.80- to 1.20-, and 1.20- to 2.00-m depths

The water content at the monitored _M range was not significantly different between the selected CE and CT soils at the 0.8- to 1.2- and 1.2- to 2.0-m depths. For the layers above the 0.8-m depth, small but significant differences in water content at a given _M between the selected CT and CE soils and between different layers within the same soil occurred at _M above -3.16 kPa and below -1584.9 kPa but not between -10.0 and -3.16 kPa, which covered the most frequently measured range of _M in the field. We concluded that for this range of _M, the water content at a given _M was comparable between the two soils and differences in _M measured with the tensiometers can directly be interpreted as differences in soil water contents. The lack of significant differences between the water characteristic curves of the selected CE and CT soils led us to the assumption that, at the range of _M observed in the field, an appropriate estimate of the water content of the soils in all studied land-use systems is possible by using only one water characteristic curve (averaged from the two determined curves) for each individual soil layer. Because the tensiometers only work down to -80 kPa, we could not estimate plant-available water contents in soils at _Ms less than -80 kPa. However, the difference in water contents between -31.6 and -1584.9 kPa was only 20 to 46 g kg^-1 for all studied soil samples, in the same range as those reported for Oxisols of northeastern Brazil (20–40 g kg^-1; Bui et al., 1989). Therefore, the difference in water contents between -80 kPa and the permanent wilting point at -1584.9 kPa by which we underestimate the plant-available water contents in our study soils is small.

With the help of the averaged water characteristic curves and the soil densities of the five soil layers (0–0.15, 0.15–0.3, 0.3–0.8, 0.8–1.2, and 1.2–2 m) the average water storage of each system in the uppermost 2 m was estimated using Eq. [1] and [2]. During the rainy season, the average water storages were not significantly different among all studied land-use systems, except for PI where the average water storage in the uppermost 2 m was significantly smaller than in all other land-use systems (Table 3) . During the dry season differences were more pronounced; the water storage decreased along the line: CT > DP = PP > CE > NT > PI.

	Conclusions

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

	ACKNOWLEDGMENTS

 
We gratefully acknowledge the funding of the German Research Foundation (DFG Ze 154/36-1: CE, PI, NT, CT) and the MAS (Managing Acid Soils) program (DP, PP) convened by CIAT (Centro Internacional de la Agricultura Tropical, Cali, Colombia). We thank C. Benicke, P.H. da Costa, A.C. Frascoli, T. Glotzmann, A. Schill, and U. Schwantag for their support. We also thank H.L.A. Bessa and C.R. Cage (Fazenda Planalto Hirofume), R. Domício (Fazenda Brasmix), J. Fonseca (Fazenda Rancharia alegre), H. Fuzaro (Fazenda Passarinho), G. Guimarães (Fazenda Bomjardim), H. Guimarães (Fazenda Luzía), A. Mauro (Fazenda Pinusplan), W. Ribeira da Sá (Fazenda Beija Flor), and A.F. Santes (Fazenda Estancia Recanto das Flores).

Received for publication September 24, 1998.

	REFERENCES

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 

• Azooz R.H., Arshad M.A. Soil infiltration and hydraulic conductivity under long-term no-tillage and conventional tillage systems. Can. J. Soil Sci. 1996;76:143-152.
• Azooz R.H., Arshad M.A., Franzluebbers A.J. Pore-size distribution and hydraulic conductivity affected by tillage in northwestern Canada. Soil Sci. Soc. Am. J. 1996;60:1197-1201.[Abstract/Free Full Text]
• Bäumler R., Zech W. Atmospheric deposition and impact of forest thinning on the throughfall of mountain ecosystems in the Bavarian Alps. For. Ecol. Manage. 1997;95:243-251.
• Blume H.-P., Zimmermann U. Zum Einfluß landwirtschaftlicher Nutzung auf die Wasser- und Luftverhältnisse typischer Böden aus Löß Nordwürttembergs. Z. Pflanzenernähr. Bodenkd. 1975;138:541-550.
• Bui E.N., Mermut A.R., Santos M.C.D. Microscopic and ultramicroscopic porosity of an Oxisol as determined by image analysis and water retention. Soil Sci. Soc. Am. J. 1989;53:661-665.[Abstract/Free Full Text]
• Durner, W. 1994. Shypfit user's manual. Research Rep. 94.1. Univ. of Bayreuth, Institute of Hydrology, Bayreuth, Germany.
• Ellies A., Huber A. Wasserhaushalt und Bodenwasserverteilung bei einer Futtermaiskultur in Südchile. Z. Pflanzenernähr. Bodenkd. 1991;154:9-12.
• Franco, A.C., G.B. Nardoto, and M.P. Souza. 1996. Patterns of soil water potential and seedling survival in the Cerrados of Central Brazil. p. 277–280. In R.C. Pereira and L.C.B. Nasser (ed.) Proceedings of the 1st International Symposium on Tropical Savannas. Brasília-DF, Brazil. 24–29 Mar. 1996. EMBRAPA, Planaltina-DF, Brazil.
• Gee G.W., Bauder J.W. Particle-size analysis. In: Klute A.K., ed. Methods of soil analysis. Madison, WI: Part 1. Agron. Monogr. 9. ASA and SSSA, 1986:383-411.
• Geisler G. Pflanzenbau. Berlin, Hamburg, Germany: Verlag Paul Parey, 1980.
• Goedert W.J. Management of the Cerrado soils in Brazil: A review. J. Soil Sci. 1983;34:405-428.
• Goss M.J., Howse K.R., Harris W. Effects of cultivation on soil water retention and water use by cereals in clay soils. J. Soil Sci. 1978;29:475-488.
• Hartung J., Elpelt B. Multivariate Statistik. München, Germany: Oldenbourg-Verlag, 1989.
• Hill R.L. Long-term conventional and no-tillage effects on selected soil physical properties. Soil Sci. Soc. Am. J. 1990;54:161-166.[Abstract/Free Full Text]
• Hodnett M.G., L. Pimentel da Silva, H.R. da Rocha, and R.C. Senna Seasonal soil water storage changes beneath central Amazonian rainforest and pasture. J. Hydrol. 1995;170:233-254.
• Holmgren G.G.S. A rapid citrate-dithionite-extractable iron procedure. Soil Sci. Soc. Am. Proc. 1967;31:210-212.
• Klute A.K. Water retention: Laboratory methods. In: Klute A.K., ed. Methods of soil analysis. Madison, WI: Part 1. Agron. Monogr. 9. ASA and SSSA, 1986:635-662.
• Lal R., De Vleeschauwer D., Nganje R.M. Changes in properties of a newly cleared tropical Alfisol as affected by mulching. Soil Sci. Soc. Am. J. 1980;44:827-833.[Abstract/Free Full Text]
• Lima W.P. Soil moisture regime in tropical pine plantations and in "Cerrado" vegetation in the state of São Paulo, Brazil. Piracicaba 1983;23:5-10 Instituto de Pesquisa e Estudos Florestais da Escola Superior de Agricultura Luiz de Queiroz (IPEF), Piracicaba, São Paulo, Brazil..
• Lima W.P., Zakia M.J.B., Libardi P.L., Filho A.P.S. Comparative evapotranspiration of Eucalyptus, pine and natural "Cerrado" vegetation measured by the soil water balance method. Piracicaba International 1990;1:5-11 Instituto de Pesquisa e Estudos Florestais da Escola Superior de Agricultura Luiz de Queiroz (IPEF), Piracicaba, São Paulo, Brazil..
• Logsdon S.D., Allmaras R.R., Wu W., Swan J.B., Randall G.W. Macroporosity and its relation to saturated hydraulic conductivity under different tillage practices. Soil Sci. Soc. Am. J. 1990;54:1096-1101.[Abstract/Free Full Text]
• McGowan M., Williams J.B. The water balance of an agricultural catchment. II. Crop evaporation: Seasonal and soil factors. J. Soil Sci. 1980;31:231-244 a.
• McGowan M., Williams J.B. The water balance of an agricultural catchment. III. The water balance. J. Soil Sci. 1980;31:245-262 b.
• Mehra O.P., Jackson M.L. Iron oxide removal from soils and clays by dithionite-citrate systems buffered with sodium bicarbonate. Clays Clay Miner. 1960;7:317-327.
• Parfitt R.L., Childs C.W. Estimation of forms of Fe and Al: A review and analysis of contrasting soils by dissolution and Moessbauer methods. Austr. J. Soil Res. 1988;26:121-144.
• Parfitt R.L., Percival H.J., Dahlgren R.A., Hill L.F. Soil solution chemistry under pasture and radiata pine in New Zealand. Plant Soil 1997;191:279-290.
• Rosa R., Lima S.d.C., Assunção W.A. Abordagem preliminar das condições climáticas de Uberlândia (MG, in Portuguese). Soc. Nat. 1991;3:91-108.
• Sarmiento G. The ecology of neotropical savannas. Cambridge, MA: Harvard University Press, 1984.
• Schiller G., Cohen Y. Water balance of Pinus halepensis Mill. afforestation in an arid region. For. Ecol. Manage. 1998;105:121-128.
• Schwertmann U. Differenzierung der Eisenoxide des Bodens durch Extraktion mit Ammoniumoxalat-Lösung. Z. Pflanzenernähr. Düng. Bodenkd. 1964;105:194-202.
• Sharma M.L., Uehara G. Influence of soil structure on water relations in low humic latosols: I. Water retention. Soil Sci. Soc. Am. Proc. 1968;32:765-770.
• Soil Survey Staff. Keys to soil taxonomy, 7th ed Blacksburg, VA: Pocahontas Press, 1997 SMSS technical monograph no. 19..
• Tsuji G.Y., Watanabe R.T., Sakai W.S. Influence of soil microstructure on water characteristics of selected Hawaiian soils. Soil Sci. Soc. Am. Proc. 1975;39:28-33.
• Van Wambeke A. Soils of the tropics. New York: McGraw-Hill, 1992.

This article has been cited by other articles:

				N. Volland-Tuduri, A. Bruand, M. Brossard, L. C. Balbino, M. I. L. de Oliveira, and E. de Souza Martins Mass Proportion of Microaggregates and Bulk Density in a Brazilian Clayey Oxisol Soil Sci. Soc. Am. J., September 1, 2005; 69(5): 1559 - 1564. [Abstract] [Full Text] [PDF]

				W. Wilcke and J. Lilienfein Nutrient Leaching in Oxisols Under Native and Managed Vegetation in Brazil Soil Sci. Soc. Am. J., June 2, 2005; 69(4): 1152 - 1161. [Abstract] [Full Text] [PDF]

				W. Wilcke, M. Krauss, J. Lilienfein, and W. Amelung Polycyclic Aromatic Hydrocarbon Storage in a Typical Cerrado of the Brazilian Savanna J. Environ. Qual., May 1, 2004; 33(3): 946 - 955. [Abstract] [Full Text] [PDF]

				W. Wilcke and J. Lilienfein Soil Carbon-13 Natural Abundance under Native and Managed Vegetation in Brazil Soil Sci. Soc. Am. J., May 1, 2004; 68(3): 827 - 832. [Abstract] [Full Text] [PDF]

				J. Lilienfein, W. Wilcke, L. Vilela, M. A. Ayarza, S. d. C. Lima, and W. Zech Soil Fertility under Native Cerrado and Pasture in the Brazilian Savanna Soil Sci. Soc. Am. J., July 1, 2003; 67(4): 1195 - 1205. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (15) Citing Articles via Google Scholar Google Scholar Articles by Lilienfein, J. Articles by Zech, W. Search for Related Content PubMed Articles by Lilienfein, J. Articles by Zech, W. Agricola Articles by Lilienfein, J. Articles by Zech, W.