sign in sign up
<<
Home , Laterite, Latosol

Reprinted from the Science Society ofAmerica Journal Volume 51. no, 3. May-June 1987 677 South Segoe Rd.. Madison, WI 53711 USA

Morphology, Mineralogy, and Genesis of a Hydrosequence of in Brazil

J. MACEDO AND R. B. BRYANT

/ Morphology, Mineralogy, and Genesis of a Hydrosequence of Oxisols in Brazil'

2 J. MACEDO AND R. B. BRYANT with a vast peneplain produced by denudation be­ ABSTRACT tween the late and mid-Tertiary, termed Research iias conducted to establish the genetic relationship be- twenteItlA ericand mid-eia ter and natural drainage chara,2teristics in a 'superflcie Sul-Americana" (South-Ameri'can sur­ tween soil color patterns Tertiary, the continent wa, uplifted hydrosequence of Oxisols formed in Tertiary/ sediments face). In the late Region of the Central of Brazil. and polycyclic stream incision carved valleys into the of tie Cerrado () upland. Dur­ The classify as Acrustox and Plinthaquox in Soil Taxonomy surface, transforming it into a dissected and as Dark-Red Latosols, Red-Yellow Latosols, and Hydromorphic ing this time, the tablelands underwent '-artial strip­ Laterite in the Brazilian classification scheme. In addition to stan- ping and reweathering. Braun (1971) considered the dard soil characterization analyses, x-ray diffraction and tablelands of central Brazil to be remnants of the citrate-bicarbonate-dithionite extractions were used to determine South-American surface. soil mineralogy. Precipitation was recorded and weekly positions of Feuer (1956) reported the occurrence of distinct ero­ the water table at sampling sites were measured in observation wells sion surfaces in the area of the Federal District. The for a period of I yr. , , , and Fe are oldest erosion surface, wiich corresponds to the South the domirant in all sites. Relative amounts of these min- American surface described by King (1956) and Braun erals show similar desilicatioa trends with depth at all sites in the (1 571), occupies the highest part of the landscape. It hydrosequence. Well-drained soils with water tables >3 m had red- is dominated by gently sloping to nearly level table­ dish hues (2.5YR or 5YR) indicating an oxidizing environment and lands, referred to locally as "chapadas", at elhvations a coduminant and Fe mineralogy. The upper or- of about 1000 to 1200 in. fhis .rface terminates ab­ ganic rich solum at sites with seasonally high water tables (<2 m) ruptly at an escarpment; a second surface slopes toward had yellowish hues (10YR or 7.5YR) and a dominantly goethitic Fe the major streams. This second surface, presumably a mineralogy. In the deeper perennially saturated zones of all profiles consequence of a more recent uplift, is being dissected, the matrix color has reddish hues (2.5YR or IOR) and a dominaady and a third erosion surface, dominated by valley slopes, hematitic Fe mineralogy. Stratigraphic, geomorphic, and pedogenic is found at elevations below 800 m. The youngest geo­ that soils having reddish (2.5YR evidence supports the hypothesis may be characterized as a narrow or 10R) hues and codominant hematite and goethite Fe mineralogy morphic surface the streams. environment, then underwent changes in Fe flood plain along some of major formed In a prior of the landscape of mineralogy in response to changes in hydrology brought about by Figure 2 depicts the evolution change. The organic rich upper sola the study area. The surface deposits at the study sites landscape evolution and from of soils having a seasonally high water table show color adjustment are Tertiary/Quaternary sediments preweathered to yellowish (10YR or 7.5YR) hues and dominantly goethitic Fe the regional rocks, which are interbed­ mineralogy. ded mixtures of slate, phyllite, , and quartzite These sediments underwent a long Indev Words: Acrustox, LatosoL oxides, soil color, (Barbosa, 1969). Additional process of that destroyed most of the evi­ water table, hematite, goethite. dence of the original lithology and resulted in uniform throughout the hydro­ Macedo, J.,and R.B. Bryant. 1987. Morphology, mineralogy, and morphology and mineralogy genesis ofa hydrosequence of Oxisols in Brazil. Soil Sci. Soc. Am. sequence. J.51:690-698. The present climate istropical continental with rainy summers and dry winters. Annual precipit.' is around 1500 mm, concentrated in the periu. from September to April. The mean annual air tem: !rature 0 difference of is 20 C (Domingues et al., 1968) and the T HIS STUDY CHARACTERIZEs a hydrosequence and mean winter temper­ Oxisols in the Cerrado region of Brazil. The study between the mean summer area is located on the EMBRAPA-CPAC (Cerrado Ag- ' Contribution from the Dep. of Agronomy, Cornell Univ., Ith­ ricultural Research Center) research station in the cen- aca, NY 14853. Supported in part by the TropSoil Program. Re­ (Savanna) Region, in the Fed- ceived 3 July 1986. tral part of the Cerrado DSoil Scientist, EMBRAPA-CPAC, Planaltina, DF,73300, Brazil, eral District of Brazil (Fig.1). Accordingto King (1956) now Graduate Research Assistant, and Assistant Professor, Dep. of the origin of the modern Brazilian landscape begins Agronomy, Cornell Univ., Ithaca, NY 14853, respectively. 690 MACEDO & BRYANT: A HYDROSEQUENCE OF OXISOLS IN BRAZIL 691

780 620 460

ELt , BRAZIL

I • _ ___ _

FEL) E A _ 'TI!C

i ---, ---. - 16-0

CERRADO [Savannal REGION

0 250 500 750 1000km

i ig. 1. Location of the study area. atures is 2.5'C. The dominant form of natural vege­ tation is 'Cerrado", a savanna-type vegetation. Var--______iations include: "Cez-radlio", a xeromorphic woodlanid, W/ -' "Campo Cerrado", a sparse savanna, and "Campos", ARETACEOUS to MID-TERTIARY: pianation of the surface and a grassland. Seasonal forests are also found in som "- deution of Tertiary sediment. areas of the Cerrado Region (Madeira Neto et al., ­ 1982). - Sampling sites were sd.cted along a transect across a hydrosequence of well-drainied to poorly drained soils on a chapada (Fig. 3). The hydrosequence includes * some Acrustox, previously described by Couto et al. ," ",,,'",,,' (1985) as having yellowish hues (drYR and 7.5YR) and seasnnally high water tables but showing ro evi- LATE TERTIARY: regional uplift and dac, weathering of the urface dence of Fe reduction or grey mottles commonly as- 2 km sociated with wetness in soils. The authors discussedet a.chap.-da" the factors affecting oxidation-reduction processes in E Oxisols and proposed that the lack pofan energy source 8 . for microbial activity at depths below 40 cm is the. most important factor preventing the reduction of Fe. However, the genesis of these soils and the relatin- LUATERNARY: distection of the surface ship between soil color and drainage isnot completely- Tertiary understood. The purpose of this study was to inves- i Sediments tigate the pedogenetic relationship between soil color [isPrecambrian Rocks l0'Pedilsedlments patterns and drainage characteristics in a hydrosequ- Fig. 2. Schematic representation of landscape evolution in the Brazil ence of Oxisols in the Cerrado region of Brazil. Central Plateau. 692 SOIL SCI. SOC. AM. I., VOL 51, 1987

BRAZILIAN CENTRAL PLATEAU CPAC Landscape

" ) Sites Location = Tertiary Sedliments 0 CPAC Headquarters M Precambrian Rocks

Fig. 3. Locations of the sampling sites.

MATERIALS AND METHODS preted using guidelines set forth by Brown and Brindley Ten pedons were selected along a hydrosequence crossing (1980), and by Brown (1980). a tableland to represent a range in natural drainage class and RESULTS AND DISCUSSION associated morphology. Pits were dug to a 2-m depth and described according to the procedures outlined in the Soil Soil Characteristics and Classification Survey Manual (USDA-Soil Conservation Service, 1981). Characterization data for Sites A, E, sites A, F, and H are given F, H, and J (Fig. 3)were sampled for character- in Table 1. These sites are representative of the three ization. A percussion core sampling rig was used to collect soils that comprise the hydrosequen-e. soil Site A is rep­ samples at 50-cm increments from the surface to bed­ rock in all 10 sites; depth to ranged from 8 to 15 resentative of the best drained end-member of the se­ m. Depth to the water table was measured at weekly inter- quence; these soils have reddish hues (2.5YR). Site F vals in observation wells, from July 1984 to September 1985. is representative ofthe poorly drained Gravimetric end-member of moisture contents were measured at 2D-cm in- the sequence. It has common, fine, distinct mottles* crements above the water table at site "J" in July 1985. (7.5YR 5/8) in the Precipitation A2 horizon (10-18 cm). Common, data were collected at the meteorological sta- fine, distinct mottles (10YR 5/8) and few, tion at CPAC. fine, prom­ inent mottles (2.5YR 4/8) were described in the Bo Physical and chemical analyses were run on representative horizon (34-52 cm). This samples from each soil has plinthite below 80 horizon of sites A, E, F, H, and J follow- cm. Site H is representative of the soil that occurs in ing standard procedures set forth by the USDA-Soil Con- positeoH interediate te silthataoccurshin servation Service (1984). The pipette method was used for positions intermediate between sites A and F. These particle size analysis of the <2-mm fraction. Soil pH was soils have yellowish hues (5YR-IOYR) but do not have measured in distilled water, I M KCI, and 0.01 M CaCI2, low chroma mottles or other characteristics associated using 1:1 (by weight) soil to solution ratio. Extractable bases with wetness in the upper and cation meter. Common, fine dis­ exchange capacity (CEC) were determined using tinct (5YR 5/8) mottles are present in the Bo2 herizon I MNH,OAc at pH 7.0 and I MNHCI (Peech et al., 1947). (150-200 Extractable acidity cm). was measured using BaCI-triethanola- Characteristics of all three of these soils fit the def­ mine at pH 8.2 (Mehlich, 1948), and I M KCI extractable inition of the Oxic horizon, Al was determined but, at site A, the require­ as described by Hsu (1963). Organic C ment for <16 cmole kg-I of clay by ammonium ace­ and total N were determined by thermal conductivity on a tate pH 7 is met only in the lower portion of the Oxic Perkin-Elmner Elemental Analyzer-Model 240 C(Nelson andmeolyithlwrpoinofheOc Sommers, 1982). The CBD (0.3 M sodium citrate, bicar- horizon. These soils have relatively high amounts of bonate, and dithionite) extraction method, as described by organic matter. Although they do not meet the re­ Mehra and Jackson (1960), was used for extracting free Fe quirements for the Humox suborder, oxides. Iron there is a sub­ was determined by atomic absorption. stantial amount ofCEC related to the organic fraction Mineralogic analyses were made using x-ray diffracto- in these soils. metry (XRD). Values in Table 1 are not corrected for Following dispersion and ultrasonic treat- organic matter contribution to CEC. The soils repre­ ment, clays were fractionated following procedures de- sented at sites A and scribed H ae clayey, oxidic, isothermic by Jackson (1974). The clays were dialyzed to remove Acrustox, and the poorly drained soil (site free salts and subsequently freeze-dried F) is a 2+ for storage. K and clayey, oxidic, isothermic Plinthaquox (Soil Survey Mg saturated clays were prepared for XRD as oriented mounts on glass slides. K* clays were x-rayed before and Staff, 1975). after heat treatments at 300 and 500 *C. Whole soil (<2- In the Brazilian classification scheme the soils at mm samples were prepared using randomly oriented pow- sites A der samples. and H are classified at the highest categorical Prepared mounts were run on a Norelco x-ray level as Soils with Latosolic B Horizons, not hydro­ diffiractometer using CuK-alpha radiation at a scanning rate riorphic (Bennema, of 1966), and the poorly drained soil 2*-20 per minute. X-ray diffraction patterns were inter- is classified as a Hydromorphic Soil (EMBRAPA, MACEDO & BRYANT: A HYDROSEQUENCE OF OXISOLS IN BRAZIL 693

Table I. Soil characterization data of selected pedons. CEC 11<2.gm Exch. fraction) Water content acid Extrac. Sum Or- pH Exch. bases U1M NH.OAc) 0.55 Al of I M Hori- Co'r 0.033 1.5 ganic (1:1 BaCI,- (0 M exch. NHOAc 1 M zon Depth (moist) Sr.nd MP& MPa carbon H,O) Ca Mg K Na TEA KCI) base, pH 7 NH.CI cm -% cmol kg-') Acrustox-Dark-Red Latosol (Site A) A 0O0-0It, 5YR3/4 18.6 22.0 59.4 31 21 2.8 4.8 0.010 0.033 0.067 0.012 16 0.5 0.122 27.7 16.9 AB 016-032 5YR4/8 18.0 20.1 61.9 29 22 2.0 5.0 O.0:fO 0.024 0.037 0.015 12 0.1 0.086 22.6 15.6 BA 032-066 2.5YR4/8 19.8 16.9 63.3 30 22 1.3 5.2 0.010 0.009 0.008 0.010 10 0.0 0.037 18.9 13.0 Bol 066-122 2.5YRE18 18.5 16.6 64.9 32 23 1.0 3.4 0.010 0.005 0.004 C.007 9 0.0 0.026 16.2 12.1 Bo2 122-167 2.5YR5/8 18.0 20.9 61.1 32 23 0.8 5.5 0.010 0.004 0.004 0.013 7 0.0 0.031 16.4 10.8 Bo3 167-210 2.5YR7.8 18.0 23.2 58.t 30 24 0.7 5.8 0.010 0.102 0.077 0.009 6 0.0 0.098 17.8 11.4 Bo4 210-300 2.5YR4/8 16.5 17.4 66.1 24 0.8 5.9 0.010 0.004 0.011 0 110 6 0.0 0.035 15.9 13.1 Acruqtox-Red-YeJow Latosol (Site H) Ap 000-013 10YR4/3 11.6 18.1 70.3 31 24 2.4 4.7 0.010 0.001 0.065 0.006 16 0.4 0.082 22.0 17.3 AE 013-035 10YR514 10.6 21.4 68.0 30 25 1.8 5.0 0.01G 0.026 0.043 0.013 13 0.1 0.092 21.3 16.3 BAI 035-054 7.5YR516 14.7 20.1 64.9 30 25 1.2 5.0 0.010 0.018 0.011 0.018 10 0.0 0.055 15.4 13.8 BA2 054-091 7.5YR5/8 23.5 19.8 56.7 31 25 1.0 5.1 0.020 0.018 0.016 0.005 9 0.0 0.059 19.4 12.9 Bol 091-150 7.5YR5/8 10.6 24.5 64.9 32 26 0.9 5.0 0.020 0.025 0.010 0.004 8 0.0 0.053 15.4 12.9 Bo2 150-200 7.5YR5/8 8.1 21.9 70.0 36 28 0.7 5.5 0.010 0.010 0.020 0.002 7 0.0 0.042 14.3 1..3 Be3 200-300 5YR5/8 14.5 17.8 67.7 29 0.6 5.6 0.010 0.004 0.J23 0.005 7 0.0 0.042 14.P 14.2 Plinthagox-Hydromorphic Laterite (Site F) Al 000-010 1OYR2/1 20.6 27.9 51.5 55 34 6.8 5.1 0.010 0.040 0.084 0.045 29 0.9 0.179 54.3 A2 010-018 IOYR4/3 33.9 12.4 53.7 36 24 3.5 4.9 0.010 0.026 0.038 0.007 17 0.6 0.081 34.5 BA 018-034 1OYR6/4 22.5 21.6 55.9 40 25 1.3 5.6 0.010 0.013 0.016 0.005 9 0.0 0.044 20.6 Bo 034-052 1OYR6/3 17.6 16.4 66.0 41 27 1.1 5.8 0.010 0.010 0.013 0.008 6 0.0 0.041 14.4 Bg 052-080 1OYR7/2 13.4 14.4 72.2 35 28 06 6.0 0.010 0.006 0.007 0.006 4 0.0 0.029 12.5 Bvl 080-150 2.5YR4/8 17.5 18.8 63.7 34 27 0.4 6.0 0.010 0.004 0.002 0.005 4 3.0 0.021 14.9 Bv2 150-200 2.5YR4/8 15.1 16.5 G0.4 29 0.3 6.2 0.010 0.003 0.004 0.006 4 0.0 0.023 14.6 Bv3 200-250 15.4 15.1 69.5 20 0.3 6.3 0.010 0.905 0.005 0.014 4 0.0 0.034 13.7 Bv4 250-300 17.4 14.9 67.7 31 0.3 6.1 0.010 0.003 0.004 0.009 4 0.0 0.026 14.0

1978). At the great group level, the Latosols in this0 area include Dark-Red Latsols, characterized by red hues (2.5YR or reddei) and high amountE of Fe ox- 71 ides, and Red-Yellow Latosols, characterized by hues yellower than 2.5YR and lower amounts of Fe oxides (EMBRAPA, 1978). The poorly drained soils are clas­ sified as hydromorphic Laterites, characterized by the presence of low chroma colors, plinthite, and Fe cor,­ cretions in the soil surface (EMBRAPA, 1978). In ad­ dition, all soils were classified as dystrophic for having <50% base saturation (ammonium acetate method, pH 7), and they are in the clayey textural class. In this ._ paper we will refer to the Brazilian classification rather . than to the Soil Taxonomy because the terminology is descriptive of the color, which is the major differ- " ence in the morphology of these soils. Stone Line Not Soil Development Stratigraphy in Residuum pretent Figure 4 represents the stratigraphy of the uncon- at solidated deposits at the study site as determined by 12 White Kaolinite Clay deep borings. The uniform Tertiary/Quaternary sed- -Saprolit all iments are 6 to 9 m deep and are the 0000000000 Wqatwed Quartzite for soil development at all sites. These sediments have o00ooOCo awrae ) a(tea a very uniform particle size distribution with a dom- Quartzite Bedrock inantly clayey composition. Red colors are found 15 throughout the profiles of the well-drained sites and Fig. 4. Stratigraphy of the unconsoldated sediments as determined yellowish colors are found in the upper parts of the by deep roring. poorly drained sites and intermediary sites. A stone line, mostly formed by concretionary material, can be soil developed in overlying Tertiary/Quaternary sed­ found at lower depths in most sites, indicating an ero- imants. At these sites, the soils grade into white, ka­ sional surface. At some sites, soil development ex- oliliitic, clayey . This gradational zone of tends below the stoneline and into residuum. The soil weathering was described as having variegated color. developed in residuum has properties similar to the At soiae sites, the saprolite grades into partially weath­

p 694 SOIL SCI. SOC. AM. J.,VOL 51, 1987

CROSS SECTION P-F iDark-Red LMA Red-Yellow Latosol Herirph.e 120C-0

1190- - o .,c,*.,J. 0 05

E 0 0 0 0 -- 0- - F 0 1150­

0 100 200 300 010 YR 5lYRfl25 YRfVARIEGATED Distance Between 75 1COLORS Sites Ira MBedrock Fig. 5a. Cross section from sites A thiough F showing topography, soil distribution and properties of color, relative soil depths, and depth to the water table.

1200 CRoSS SECTION F-J "L Red-Yellow Latosol .E119o. G0 110> F ...... J I ------WT ;1170

: . 10

001160-02

1150 .! o ° . ° " ,o ,. . ..

Distance Between 0Colors Sites Iml Bedrock Fig. 5b. Cross section from sites F through J showing topography, soil distributions and properties of color, relative soil depths, and depth to the water table.

eked, wealy consolidated gneiss or slate. At other sites, rated to within I m from the clayey saprolite is underlain the soil surface throughout by coarse sand, which the year. The water table depth weathered from the underlying quartzite. for the Red-Yellow and Dark-Red Latosol varies in an annual cycle from Soil Morphology, Hydrology, and Landform 2 to 5 m and from 5 to 7 m respectively. Relations The data also show a 6- to 8-week delay in the rise of the water The relationship table in response to precipitation, and the among soils, topography, depth to level declines slowly water table bedrock, for a similar period of time at and average depth to water table are shown the beginning in Fig. 5-a and 5-b. The surface of the . level is shown at an The pattern of water table .esponse exaggerated scale; actual slope to precipitation is 0 to 2%. The bedrock was consistent at all sites on the transects. surface is displaced with respect to the soil Both from surface in the massive nature of the quartzite rock and the order to depict hues and relative depths to the be­ water havior of the water table in these soils, it may be in­ table at an exaggerated scale. ferred Precipitation and that the underlying bedrock is vitually im­ water table data (Fig. 6) show the permeable. During the rise of the water table rainy season, excess water is in response to the an.tual dis- being stored in the soil, causing tribution of precipitation at sites wter table levels to A, F, and H, which rise. In the central areas of the chapada, represent the three major soils in the hydrosequence. the decline of the water table occurs as water moves laterally, The 1500 mm ofprecipitation occur mainly and from Sep- is eventually discharged at the border of the chapada tember to April. The Hydromorphic Laterite is satu- as seepage. The hydrology at any site on the chapada MACEDO & BRYANT: A HYDROSEQUENCE OF OXISOLS IN BRAZIL 695

WATER TAILE DIPTH and PRECIPITATION (weekly)

I AUG I SEP I OCT I NOV I DEC I JAN I FEB I MAR I APR I MAY I JUN I JUL I 0......

HYDROMORIPHIC LATERITE 2- SITE F

3./ I 4­

------"RED-YELLOW 5- /" • SO SITE H 7

DARK-REDLATOSOL Sly[ A

P20.

I- I

j 1 AUG SEP OCT I NOV DEC I JAN I FEB I MARI APR I MAY I JN IJULI Fig. 6. Water tab!e depth and precipitation at sites A, H, and F.

7 ERTIARY SEDIMENTS -

EJDark-RedzL atioso I­

o Red-Yellow - Latosol - - -PRE-CAMBRIAN

l tHydrmorphic -.. Crystaline Rocks

,, , * 0 0 :o00e ~ : ~O• * . * • ,o :

Fig. 7. Generalized block diagram showing landform and soil distribution relationships.

is determined by its position in relation to the border DRL of the chapada. Figures 7 and 8 are generalized diagrams that depict ra T. ri SodIm*n t the patterns of soil distribution with respect to the.. / landscape at the study site and in the surrounding / / / / areas. All of the soils in this study are formed in the / / / / /Bedock / / / Tertiary/Quaternary sediments that cover the dis- 1 km sected remnants of the older geomorphic surface (ta­ Dark- Rd Red-Yellow blelands). The Hydromorphic Laterites are found in D-R2.5Y IlOYR HydromorphicOYa ow chroma areas where the water table is seeping at the shoulder Latosol Latosol of the tableland. Laterite These areas are slightly lower in the Fig. . Cross section of a table land showing landform and soil color landscape than other soils on the chapada, and the relationships. thickness of sediments over bedrock is less. The Dark- Red Latosol is found in the central )art of the table- pattern typifies the study area, the extent of each soil land where the water table fluctuation zone is deepest. is not consistent throughout these landscapes. Irreg­ The Red-Yellow Latosol is found in an intermediate ularities in topographies of the land surface or the un­ landscape position. Although this soil distribution derlying bedrock surface at the edge of the tableland 696 SOIL SCL. SOC. AM. I., VOL. 51, 1987

cause some variations in the hydrology and associated Latosol has a similar amount of free Fe oxides as that soil morphology in local areas. found throughout the The schematic profile of the Dark-Red Latosol. in Fig. 9 shows the relationship be- In Fig. 12, the relative proportions of tween soil color and hematite and water table fluctuation zones in goethite, as determined by XRD on whole soil these soils as may be (<2­ summarized from the data col- mm fraction), were used to partition the CBD extract­ lected in this study. Well-drained soils, located in the able Fe upper contents. There is a good correlation between part of the chapada (site A, Dark-Red Latosol), soil color and the kind of Fe present. have reddish hues (2.5YR The Dark-Red or redder) throughout the Latosol and the reddish zones of the Red-Yellow La­ solum. Poorly drained soils, located at the edge of the tosol have hematite and chapada (site F, Hydromorphic goethite mineralogy, whereas Laterite), have yellow- goethite was the only x-ray-detectable Fe component ish hues (lOYR) in the soil surface, gleyed colors im- in the yellowish zone of the Red-Yellow Latosol. Sim­ mediately below the surface horizon, and reddish hues ilar results have been reported (2.5YR in other studies of highly or redder) in the . Soils in intermediate weathered soils of Brazil (Weaver, drainage 1974; Bigham et positions (sites B, C, D, E, G, H, I, and J, al., 1978; Rodrigues and Klamt, Red-Yellow Latosols) 1978; Volkoff, 1978; have yellowish hues (I0YR or Kampf ard Schwertmann, 1983; Curi and Franz­ 75YR) in the soil surface and hue gradually reddens meier, 1984). with depth. Plinthite was described in the zone im- It should be noted that Fe concretions found mediately below the average in the depth of the water table, gravel-sized fractions were excluded from the CBD and XRD analyses of the soil fraction <2 mm. X-ray Soil Minerology diffraction of dusk red (10R)colored concretions Soil found mineralogy studies focused on the relative dif- in deeper zones of these soils and analyzed separately ferences between the Dark-Red Latosol and the Red- show hematite as the dominant . Although Yellow Latosol. the Kaolinite, quartz, gibbsite, Fe oxides, authors made no attempt to determine the mass bal­ and are the dominant minerals in these ance ofsecondary Fe oxides in these soils sola by including as determined by XRD. , corundum, and the Fe concretions, it did appear that the rutile were also detected. Fe concre­ The relative abundance of tions in the water table fluctuation zone might be the the dominant minerals in the whole soil (<2-mm frac- product of Fe redistribution in the sola. tion) from representative pedons of a Dark-Red La- Implications for Soil Genesis tosol and a Red-Yellow Latosol (sites estimated A and H), as by XRD, is given in Fig. 10. Greater Environmental conditions that favor amounts of gibbsite the crystalli­ and correspondingly smaller pro- zation of hematite from Fe released by portions of kaolinite were weathering of observed in the upper parts primary minerals in the soil environment include high of these sola. It may be interpreted that silica has been temperature, low moisture, lost from the upper and low organic matter sola as a result of weathering pro- content (Kampf and Schwertmann, 1983). Parent cesses that favor desilication. ma­ Aluminum in the upper terial characteristics that favor hematite formation sola might also have been gained in­ from biocycling. Pro- clude a relatively rapid rate of Fe release, and high cesses that are responsible for the redistribution of sil­ ica and Al have resulted in identical relative depth so L distributions of the dominant MINERALOGY silica- and Al-bearing a 4 2mmE rton minerals in these two soils. M aolinlto Quirtz lIbbl,|t* Mineralogical differences between the Dark-Red La- Dark.Red Latosol tosol and the Red-Yellow Latoso: Red.Yellow Latosol were found when 0 20 40 60% 0 20 40 60% the content and amount of free Fe oxides and hy­ droxides in these soils were compared. Free Fe, ex­ tracted by CBD (Mehra and Jackson, 1960), is shown in Fig. I1.The Dark-Red Latosol has twice as much free Fe as the yellowish upper zones ofthe Red-Yellow Latosol. However, the reddish zone of the Red-Yellow 1.5-2.0

Dark-Red Red-Y,,llo- Hydromorphic Latosol Latosal Laterite. 0. 0Y 3.0-3.5

2 7~~.5YR .. >..

.c 4.-. 5.3-5.5

= 6 .- ' -. -. -i,7/////

0 8. 7.P5nthl.-8.0

Fig. 9. Schematic diagram showing the general relationships be- Fig. 10. Comparative whole soil mineralogies (<2-mm fraction) of tween soil color and the water table fluctuation zone, a Dark-Red Latosol (site A) and a Red-Yellow Latosol (site H). MACEDO & BRYANT: A HYDROSEQUENCE OF OXISOLS IN BRAZIL 697

pH in the weathering environment (Schwertmann, table observed within 2 m of the surface in the Red- 1985). Yellow Latosol, as well as the high water table in the The study area is located in a geographical region Hydromorphic Laterite, is a relatively recent change of Brazil that is dominantly Dark-Red Latosols. By in the hydrology of these sites that occurred in re­ geographical association, and since hematitic miner- sponse to the encroachment of the plateau escarpment alogy in the lower solurn is common to all pedons in and perhaps due in part to a change from a drier ma­ the study, it is inferred that the environmental con- croclimate to a seasonally wet climate with conse­ ditions that prevailed during weathering and Fe re- quent rise of the water table. lease from primary minerals were re,'-atively w-r'm and In order to gain some indication of the possible in­ dry. The parent material characteristics in the study fluence of a wate:" table occurring at a depth of 2 to area also favored the forriation of hematit; from Fe 2.5 m in the Red-Yellow Latosol, gravimetric mois­ in solution. Gneiss, slate, and in the area are ture contents were measured at 20-cm increments moderately high base rocks and contain Fe-bearing above the water table at site "J" in July, 1985. The minerals. These paleoclimate and parent material con- water table was at 2.35 m, and the site had received ditions favor the formation of Dark-Red Latosols and no precipitation for a period of 5 weeks prior to the explains their occurrence in well-drained positions in sampling. The data are shown in Fig. 13. The 0.033­ the modem landscape. The Hydromorphic Laterite has and 1.5-MPa gravimetric moisture contents were de­ morphology that reflects the processes ofFe reduction, termined in the laboratory using the pressure mem­ transfer, and re-oxidation in the water table fluctua- brane apparatus. The results show that the 1.3-m zone tion zone in the form of plinthite and Fe concretions. above the water table is maintained at soil moisture Whereas the association of the morphology of the Hy- tensions <0.033 MPa due to capillary rise, and the I­ dromorphic Laterite with the presence of a high water m zone above that is maintained at soil moisture ten­ table is easily observed, the pedogenetic relationship sions < 1.5 MPa. In reference to the climatic data in between the morphology of the adjacent Red-Yellow Fig. 6, the Red-Yellow Latosol should attain maxi­ Latosol and the presence of a water table at a depth mum wetness conditions in February and March, when of 2 to 2.5 in is not as -eadily apparent. the water table is at its shallowest depth and weekly The lower Fe content and dominantly goethitic Fe precipitation frequently approaches 100 mm. Al­ mineralogy observed in the upper sola of the Red- though soil mo'sture data are not available for this Yellow Latosol (Fig. 1I,12), appears to be a pedogenic season, conditi- is might easily approach near satu­ response to the imposition of a water table on a soil ration for a period of several days or a few weeks which previously had characteristics of a Dark Red during this part of the season. Reducing conditions Latosol. There is a similarity of depth distributions of might occur in some parts of the soil profile during kaolinite, quartz, and gibbsite in the Dark-Red Lato- this part of the season. Field measurements of redox sol and the Red-Yellow Latosol (Fig. 10). This sup- potentials by Couto et al. (1985) did not detect re­ ports the hypothesis that both soils developed under ducing conditions in the Red-Yellow Latosol during relatively dry and well-drained weathering environ- even the wettest season of the year. However, they did ments for a period of time sufficient to establish the observed desilication trends and codominant hema- IRON MINERALOGY 1

Dark-Red Latosol Rd-Yellow Latosol0-. 0 2 4 6 8 1012 % 0 2 4 6 8 10 12% 00.5

1.5-2.0 2 2

3. 3. •3,0.3.5 4- 4

5" 55­

6, 6­

7 7 7.0-7.5 a. -C2mrm IractIon * IRON CONCRETION EXCLUDED 9 Iron Fig. 12. Comparative hematite and goethite distribution in the <2­ Fig. 11. Comparative iron contents (CBD extractable) of a Dark- mm fraction of a Dark-Red Lztosol (site A) and a Red-Yellow Red Latosol (site A) and a Red-Yellow Latosol (site H). Latosol (site H). Ic9\ 698 SOIL SCI.SOC. AM. J., VOL 51, 1987

0 20 Bennema, J. 1966. Report 40 60 to the Government of Brazil fication of Brazilian soils. FAO-EPTA on classi­ (Report 2127). Rome, It­ aly. 40 t Bigham, J.M., ....--....1.5 O.033Mpa MPa It. D.C. Golden, S.W. Buol, S.B. Weed, and Brn1978.Influence on color, 42:825-8301mineralogy . ibioa LH. Bowen surfaceof area, well-drained andgeomorfologia phosphate do andBrasil.R Oxisols: retention. Soil Sci. Soc.. . 42:825-830. Bras. SdBrown, de Geogr 32(3):3-39, Jul/Set., Rio G. 1980 Associated de Janeiro, Ri. ioldtu minerals. p. 361-410. In G.W. x-rayand G. Brown (ed.) Crystal Brindley identification. Min. Soc.structure of clay minerals and Society, London.• Monograph no. 5, Mineralogicaltheir so MOISTURE- ...,,,- u. , d eao ci Red-Yellow Latosml Brown, G., and G.W. Brindley. 1980. X-ray diffraction procedures for identification. G. p. 305-360. InG.W. Brindley Brown (ed.) Crystal structures of clay and ray identification. Min.Soc. minerals and their x­ Uciet, Monograph no. 5, Mineralogical So- 12 London. Couto,affecting W., C.Sanzanowicz, and SCirri, oxidation-reduction processesA. de 0. Barcelos. 1985. Factors sonalN., and in an with a waterD.P. table. Franzmeier. Soil Sci. Soc. 1984. Am. J. 49:1245-1248. sea­ Toposequence ofOxisols the Central Plateau from 160 of Brazil. Soil Sci. Soc. Am. J. IS Domingues, 48:341-346. A., J.E. Nimer, and T.A. Alonso. 1968. 6gicos. Instituto Brasileiro Dominios Ecol­ de Geografia: Subsidios a regionali­ zajo. 11-36 (10 EMBRAPA. 1978. Levantamentomap.) Distrito de reconhecimento dos solos do 200 Federal. Scrvico Nacional de Levantamento vaoo de Solos (Boletim e Conser­ Feuer, R. 1956. An exploratoryTknico, 53). Rio de Janeiro, RJ. investigation of soils and agricul­ tural potential of the soils of the future Federal [Diss.Cerrado Plateau of Brazil. District in the Abstr. B., 16(3)13071. Ph.D. diss. Cornell Univ. Ithaca, NY 240 ­ Hsu, P.H. 1963. Effect table of initial pH,phosphate, and silicate determination of aluminum on the Jackson, M.L 1974. with aluminon. Soil Sci. 96:230-238. Fig. 13. Grayimietric field moisture versifyed. 9th printing. Soil chemical analysis: Advanced course, conteant of a Red-Yellow Latosal. of Wisconsin,Published Madison. by author, Dep. of Soil Science, 2nd e.s9th pricning. 1 " Uni Publis ytorDn.fSiSi~cUi ,Uin conditionsmeasure reducing potential under when an organic energy unsaturated Soil southern to these source was added soils. Brazil and their application in classification soils. We propose that weak reducing Geoderma 29:27-39. of kaolinitic tions might condi- Kin , LC. 1956. Geomorfologia occur inthese soils in the close Rio de Janeiro, do Braiii k'ent,-i. R. Bras. Gcogr., of root surfaces vicinity Madeira Neto, J.S., 18(2):147-265,J. in a manner similar to that Macedo, andAbr/Jun. L.G. Azevedo. 1982. Brazilian by described problem soils: Distribution, Schwertinann and Taylor (1977). characteristics and utilization. Trop­ ical Agriculture Research Center Series no. 15. Yatabe, Tsukuba, tosolThe hydrologicTheconditionscoditonsydrlogi off thete Red-YellowRd-YelowJ and the presence of organic compounds La- ~ Mehlich,lbaraki, A.Japan.1948. Determination surfaces within near root properties of soils. of cation and anion exchange the zone above the water table may Soil Sci. 66:429-444. have induced the Mehra, O.P., and M.L. Jackson. hve induce tselectiveselectivereduction ethe reduction, dissolution, and 1960. Iron oxide removal from removal of ha-matite from p southo awith soils and clays by dithionite-citrate-bcarbonate the upper part of the Soil sodium bicarbonate. Clays system buffered profile and/or Nelson, D.W., and L.E. Sommers. Clay Minerals 7:317- . its transformation to goethite. Evidence 1982. Total carbon, organic327 car­ of bon, and organic matter. In the preferential reduction Peech, A.L. Page ,:tal. (ed.) Methods of quent migration of hematite and subse- analysis,M., L.T. Alexander, soil of Fe is supported by the Part 2. 2nd ed. AgronomyL.A. Dean, 9:539-579. and presence of ods of soil analysis J.F. Reed. 1947. Meth­ precipitated for soil-fertility investigations. USDA Cir. tions found Fe in the form of plinthite and concre- no. at the 2- to 3-rn depth, which is the water 757. table fluctuation Rodrigues,sequencia T.E., and E. Klamt. 1978. Mineralogia zone in the Red-Yellow Latosol. de oios no Distrito Federal. egenesis de uma ray X- Solo, Campinas, R. Brs. de Ciencia do diffraction patterns of these concretions 2(2):132-143. mounted) show that hematite (powder Schwertmann, U. is the dominant mineral 1985. The effect of pedogenic environments iron oxide minerals. In B.H. on (over 80%), and only quartz was Stwart (ed.) Advan. Soil Sci. 1:171­ ondary component. identified as a sec- &hwertmann, U., Localized hematite dissolution and R.M. Taylor. 1977. Iron oxides. p. 145-180. and/or InJ.B. Dixon and S.B. transformation to goethite, Weed (ed.) Minerals in soil environments. aided by biotur- American bation (i.e., activities Society of Agronomy and of termites, earthworms, ants, Soil Science Society ofAmer­ etc.), appears to have Soil Survey Staff. 1975. Soil Taxonomy: occurred for a period of time Abasic system of soil clas­ sufficient to remove sification for making and interpreting all ofthe hematite and change soil surve s.Agric. Handb. the 436, USDA-SCS, U.S. Government color of the from red Printing O ce, Washington, to yellow. USDA-Soil Conservation Service. 1981. Soil Survey Manual. Chap- ACKNOWLEDGET ter 4, working draft e (430-V). USDA 430 V. U.S. Government USDA-SoilPrinting Office.Conservation Washington, DC. STner aurthe Service. 1984. ield o anldg the Procedures for co!lecting Stoner during the field ianil suppDrt ofi Soil samples EMBRAPA/CPAC and workthe TropSoil/Acidand the financial support of and methods of analysis for soil survey. gram. pro- Investigation Report no. I (revised), USDA-SCS, Soii Survey ment Printin Office, Wahington, DC U.S. Govern- Volkoff, B. 1978. Os produtos Ierruginosos que determinam a cor dos REFERENCES Weaver,Latossolos da Bahia. R. Bras. de Ciencia Barbosa, 0. 1969. Projeto 2(0):55-59.R.M. 1974. Soils of do Solo, Campinas, Brasilia-Goias. Departamento Nacional the Central Plateau ofBrazil: de Pesquisas and mineralogical properties. Chemical Minerais. Rio de Janeiro, RI. Agron. Mimeo. 74-78, Cornell Univ., Ithaca, NY.