Printer Friendly

Mineralogy of volcanically derived alluvial soils at Moshi, Tanzania.


Soils in areas associated with volcanic activity often have a unique assembly of clay minerals comprising both nanocrystalline and layered silicate minerals (Shoji et al. 1993; Harsh et al. 2002; Armas-Espinel et al. 2003; Hepper et al. 2006; Drouza et al. 2007). The presence of nanocrystallinc minerals, even in small quantities, may have a strong effect on the soil chemical and physical properties because of the large reactive surface area of these minerals (Gama-Castro et al. 2000; Kaufhold et al. 2010) and their strong aggregating effect (Armas-Espinel et al. 2003). Soil properties, such as water retention and cation exchange capacity, have been found to increase greatly with the amount of nanocrystalline volcanic minerals (Shoji et al. 1993; GamaCastro et al. 2000; Armas-Espinel et al. 2003; Fontes et al. 2004; Hepper et al. 2006).

The highland regions of Tanzania form part of the East African rift system, which developed due to crustal tension, rift faulting and volcanic activity. The geology of the rift zone is mainly comprised of basaltic intrusive and volcanic rocks, with some rare sodic alkaline rocks and igneous carbonates. This geology greatly affects the groundwater quality in the rift zone, which is mostly alkaline, with low calcium and magnesium content and high sodium concentrations (Smedley 2004). Despite widespread, intensive irrigated commercial agriculture in the rift zone, studies on the mineralogical properties of these soils influenced by volcanic origins are scarce.

The aim of the present study was to investigate the mineralogy of soils, currently under intensively irrigated sugar cane, which arc likely to have been strongly influenced by the volcanic history of Mt Kilimanjaro and Mt Mcru, situated approximately 50 km to the north of the study site. The present study formed part of a wider investigation into the soil properties of the sugar estate and their effects on the water retention and transmission properties with the purpose of guiding improved irrigation management.

Materials and methods

Site and sampling details

The study was performed on a 16 000-ha sugar estate, located outside Moshi, Tanzania (3[degrees]20'06"S, 37[degrees]20'25"E). The estate is planted under 8000 ha of sugar cane and currently produces over 100 000 tons of sugar per annum. The area has a semi-arid climate with an average annual rainfall ranging between 500 mm in the south of the estate to 700 mm in the north (Meyer et al. 2009), and a maximum temperature between 17[degrees]C and 30[degrees]C (Brogna 2004). The potential evaporation A-pan reading averages 2314 mm per annum. The topography is gently undulating with a slight decline in altitude from the north (780 m above sea level (a.s.l.)) to the south (650 m a.s.l.; Meyer et al. 2009).

The sugar estate has been divided into five management areas, namely the north, east, south, west and Kahc (Fig. 1). The five management areas are based on location within the estate for case of management and not necessarily on any similar properties, such as yield, soil characteristics or water source. As such, it was expected that each area would have considerable variability between fields in terms of measured properties. However, a comparison between areas in the measured properties is useful to indicate general differences and any trends that may be present from north to south or east to west across the estate. Several measured soil properties have shown north-south and east-west trends (Meyer et al. 2009) and therefore it was decided for the present study to provide averages of measured properties for the respective areas for purposes of an overall comparison.

Approximately 66% of the soils are Cambisols, 17% Luvisols, 7% Plinthosols and the remainder Fluvisols (International Union of Soil Science Working Group World Reference Base (1USS Working Group WRB) 2014). The soils on the estate are mainly Oakleaf soil form with some Tukulu form, especially in the south area (Soil Classification Working Group 1991; Meyer et al. 2009); Typic and Aquic Haplustept respectively (Soil Survey Staff 2014). Alluvially derived clay soils are mostly found in the east and central parts of the estate and are mainly Valsrivicr and Scpane soil forms (Soil Classification Working Group 1991; Meyer et al. 2009); Typic and Aquic Haplustalf respectively (Soil Survey Staff 2014).

In all, 21 fields within the five broad areas (based on permanent sampling sites selected by the sugar estate management) and two buried ash layers were selected for sampling. Bulk soil samples were collected from the A and B horizons (in general 0-400 and 500-800 mm) using a spade and hand trowel, air dried and gently milled by mortar and pestle to pass a 2-mm sieve. The sample numbering in this paper retains the field numbering system in place at the estate, and the areas in which the fields are located are indicated in each case.

Mineralogical methods

The mineralogical composition of the clay fraction was characterised using X-ray diffraction (XRD) on oriented specimens. The clay fraction (<2 [micro]m) was obtained from the soil, after ultrasonic dispersion, by sedimentation applying Stokes' law. Prior to preparation of oriented specimens on glass slides, one-third of the separated clay fraction was saturated with [Mg.sup.2+], one-third was saturated with [K.sup.+] and the remaining one-third was stored for surface area determination. The Mg-saturatcd specimens were air dried, treated with ethylene glycol at 60[degrees]C overnight and then treated with glycerol at 85[degrees]C overnight; the K-saturated specimens were air dried and heated at 550[degrees]C for 4h (Buhmann et al. 1985). After the clay was separated, the remaining material was wet sieved through a 50-[mu]m sieve and the separated sand (2000-50 pm) and silt (50-2 [mu]m) fractions were collected and dried in an oven for 24 h. After crushing in a pestle and mortar, each sample fraction was analysed as a random powder specimen by XRD. All XRD analyses were performed using a Philips X-ray diffractometer with a graphite monochromator at 40 kV and 40 mA. Oriented and random powder specimens were scanned at a rate of 1[degree] min 1 with a step-scan of 0.02[degrees] between 3[degrees] and 45[degrees] 2[theta] and between 3[degrees] and 75[degrees] 2[theta], respectively. All data were captured using a Sictronics 122D automated microprocessor attached to the diffractometer.

After XRD analysis, the clay fraction of the A and B horizons from 12 of the sampled fields and the two ash layers was analysed by transmission electron microscopy (TEM; JEOL 1400).

In all samples, the silica, aluminium and iron were extracted using acid ammonium oxalate (McKeaguc and Day 1966). Aluminium and iron were also extracted using sodium citratcbicarbonate-dithionitc (CBD; Mchra and Jackson 1958) and sodium pyrophosphate (McKeaguc 1967). These extractions are intended to differentiate between the nanocrystalline, crystalline and organically bound fractions respectively.

The clay fraction from the A and B horizons of seven fields (one from each of the north, east and west areas and two from each of the south area and Kahe) were selected for the determination of specific surface area (SSA) using an ASAP 2010 (Accelerated Surface Area and Porosimetry System; Micromeritics Instrument Corp.) in accordance with the method of Carter et al. (1986). These fields were selected to be representative of their respective areas based on the results from the XRD analysis.

Results and discussion

X-Ray diffraction

The XRD analysis showed that all samples had similar mineralogical composition and that the X-ray patterns were weak, indicating the presence of nanocrystalline material. All samples contained kaolin, K-feldspar, illite and a small quantity of hematite or gocthite. The kaolin is probably present as halloysite, rather than kaolinite, because the 020 reflection is much stronger than the 001 reflection, which, in most samples, has a d-spacing >7.03[Angstrom], which would indicate a low-defect kaolinite. This, in conjunction with the broadness of the peaks, suggests that there is a high level of disorder within the kaolin mineral. The heights of the kaolin and illite peaks differ in the XRD patterns between clay samples, indicating variable quantities between fields. However, there seems to be no trend for specific areas. There is a small amount of calcite, talc and gibbsite in a few of the sampled fields, especially from the south area. The A and B horizons from field 51 (south area) have been chosen as representative examples of the oriented clay XRD patterns found for the soils analysed (Fig. 2).

It is well recognised that the mineralogy of volcanic ashenriched soils usually consists of nanocrystalline and poorly crystalline (high-defect) minerals (Hepper et al. 2006) and the dominance of such material in all the samples rather than more crystalline minerals indicates the strong influence of volcanic ash in the soils across the estate. Other studies in Tanzania have also found the presence of nanocrystallinc material in the soil, attributed to volcanic parent materials (Jager 1982; Funakawa et al. 2012).

The volcanic glass present in volcanic material is poorly crystalline in nature and weathers rapidly. This rapid weathering releases elements faster than crystalline minerals can form, and therefore the soil solution is often over-saturated with Al and Si. These may precipitate as poorly ordered solid phase minerals (Shoji et al. 1993; Harsh et al. 2002). Under high-rainfall or irrigation and well-drained conditions, Si is quickly leached, resulting in low soluble Si concentrations and the formation of allophane (Broquen et al. 2005). However, under conditions where weathering is slow, the amount of Si and Al in the soil solution is usually insufficient to form allophane (Hepper et al. 2006; Drouza et al. 2007). In some soils under a drier climate, and where there is high soluble Si, halloysite forms preferentially in place of allophane and imogolite (Shoji et al. 1993; Harsh et al. 2002; Ndayiragijc and Delvaux 2003). The presence of halloysite together with allophane in all the current samples is likely indicative of the alluvial nature of the soils, which may result in different layers of permeability within the soil profile; consequently, the amount of silica will vary between these layers due to their different drainage properties. Singleton et al. (1989) studied the drainage sequence of rhyolitic volcanic alluvium in New Zealand and found that when silicon in soil solution was less than 10 g [m.sup.-3], allophane was dominant, whereas silicon in soil solution greater than 10g [m.sup.-3] resulted in the dominance of halloysitc. Similarly, Vacca et al. (2003) suggested that the different hydraulic properties of parent materials determined pedogenesis and found that allophane formation is favoured in young, porous and penncablc ash deposits, whereas nonallophanic soils consisting of halloysite and hydroxyinterlayered vermiculite develop in older and less porous material, thus allowing the coexistence of allophanic and non-allophanic soils in the same environment.

Similar to the study of Sellitto et al. (2010), the illite in the clay fraction is likely to be inherited from the parent material. The presence of a small amount of gibbsite indicates that there has been weathering of some of the kaolin minerals. As weathering proceeds, the silica content diminishes as the quantity of glass is depleted. This results in the formation of more stable crystalline aluminosilicates and Al- or Fe-oxide minerals (Shoji et al. 1993; Harsh et al. 2002). Gibbsite often forms under conditions where there is strong leaching (Harsh et al. 2002; Watanabe et al. 2006) and desilication of kaolin (kaolinitc and halloysitc) or non-crystalline intermediates (Nicuwenhuyse et al. 2000; Ndayiragije and Delvaux 2003). Gibbsite may also form directly from volcanic minerals, especially in permeable parent materials under high rainfall (Nieuwcnhuysc et al. 2000). The combination of allophane and halloysite, together with a small quantity of gibbsite, may be a consequence of the alluvial nature of the estate soils. It is likely that the material deposited over the estate with time consists of layers that have undergone varying degrees of weathering, resulting in the presence of both first-alteration minerals, such as allophane and halloysite, and secondarystage alteration minerals, such as gibbsite, within the soil profile. Halloysitc is usually the first alteration stage from feldspar weathering, followed by gibbsite with the removal of silica from the halloysite (Bates 1960). Bates (1960) found large quantities of both halloysite and gibbsite in soils weathered from basalitic rocks in the Hawaiian Islands that was attributed to different local conditions in terms of rainfall, slope and rock texture, which had affected the degree of weathering. The quantities of both gibbsite and talc in the samples where they were found are very small and unlikely to play a significant role in any management practices on the estate.

All sand and silt samples have strong K-feldspar peaks of sanidine, a mineral commonly found in volcanic rocks (Sellitto et al. 2010). The XRD patterns indicate that the sanidine is mixed with other K.-feldspars, such as microclinc and orthoclase. The sand sample from the R8 ash layer (west area) also contained calcite.

The XRD patterns of four sand (Fig. 3) and four silt (Fig. 4) samples are shown as representatives of the mineralogical composition of all the sand and silt fractions respectively.

The presence of calcite is probably due to a high concentration of bicarbonate in the irrigation water (J. R. Lincoln, Group Agricultural Development Manager, Ciel Agro-industry, pers. comm.), which causes the calcium to precipitate in the form of calcium carbonate. Hardpans with precipitated calcium carbonate (calcrete) were also observed in some areas of the estate.

Transmission electron microscopy

TEM analysis was performed on the clay fractions from the A and B horizons of Fields I, I OK, 12C, I6B, 17B and G2 (south area), E6 (east area), KH4 and KH29 (Kahe arca), N65 and N84 (north area), R3S (west area) and the buried ash layers from field R8 (west area), as well as from the borehole in the north area. These samples were selected for TEM to ensure a sample from each area of the estate was represented, with the majority of the samples selected coming from the south because soils from this area were considered from XRD to contain the most nanocrystalline material. Images from five of the selected fields and the two ash layers are shown in Fig. 5 to indicate the minerals present, as well as the variation between the areas of the estate.

The TEM images confirm the XRD data and indicate that the dominant minerals in the soils from all areas of the estate are very small kaolinite (< <0.5 [mu]m) and halloysite. The size of the kaolinite particles, together with their high defect structure indicated by XRD, is typical of soils formed in the tropics and subtropics (Singh and Gilkes 1992; Hart et al. 2002, 2003; Kankct et al. 2005; Sei et al. 2006). The halloysite is present in both tubular and spheroidal form. The dominant morphology of halloysite that forms from alteration of feldspars and mica is tubular, whereas that formed from volcanic glass has a predominately spheroidal morphology (Adamo et al. 2001; Singer et al. 2004). A study in Nayarit, Mexico, by GamaCastro et al. (2000) found that the dominant crystalline material in soils formed from alluvial pumice parent material is halloysite in both spheroidal and tubular forms. The TEM images also clearly show the presence of volcanic glass in samples from the R8 ash layer and field 10K. Sellitto et al. (2010) suggested that fine ash cools rapidly, resulting in volcanic glass, whereas larger particles cool more slowly, resulting in sanidine and mica.

Extractable iron, aluminium and silicon

The ranges for the percentages of extractable Fe, A1 and Si found in the soils from the estate for the various selective dissolution methods are given in Table 1.

Ferrihydrite (acid oxalate-extractable Fe (FeQ) x 1.7; Parfitt et at. 1988) contents ranged between 0.48% and 1.84% in the A horizon and between 0.34% and 1.70% in the B horizon. With the exception of Field 12C (1.84%), the ash layers contain the highest amount of ferrihydrite at 1.97% and 1.62% for the BH Ash and R8 Ash respectively. These ferrihydrite amounts arc relatively low. Singer et ai (2004) also found low ferrihydrite (between 0.44% and 3.5%) in soils developed from basic pyroclastics in the Golan Heights. Gama-Castro et al. (2000) found ferrihydrite between 0.20% and 0.51% in pumiceous, alluvial soils in Nayarit, Mexico, and Sellitto et al. (2010) measured ferrihydrite contents in the range 0.5-1% in five pedons developed from volcanic ash on the Matese Massif in southern Italy. As weathering proceeds in volcanic ash soils, the first-stage alteration minerals, such as ferrihydrite, are transformed into their second-stage alteration minerals (i.c. more crystalline iron oxides) or may be the end product if they are preserved in buried deposits (Nieuwenhuyse et al. 2000). The higher amounts of ferrihydrite in the ash layers may be due to higher initial concentrations in these layers than the surrounding soil, as well as slower weathering rates due to being buried within the soil profile.

The greater amount of [Fe.sub.o] in the soils from the south and west areas and the lowest [Fe.sub.o], in the Kahe area (Fig. 6) indicate that the south and west areas have the highest ferrihydrite levels and thus have cither undergone slower weathering rates than the other areas of the estate or had higher initial concentrations of nanocrystalline material.

The lower-lying areas in the south and west arc likely to have received higher initial concentrations of nanocrystalline minerals through alluvial deposition compared with the other areas of the estate and thus contain higher proportions of nanocrystalline material, such as ferrihydrite and allophane. However, although the allophane content (acid oxalateextractable Si ([Si.sub.o]) x 7.14; Parfitt and Wilson 1985) is highest in the BH Ash, similar to ferrihydrite, the allophane content is highest in the east area (Fig. 7) rather than the south and west areas, as is the case for fcrrihydritc.

The standard deviation for the amount of allophane in all areas is large (Fig. 7), suggesting a heterogeneous distribution of nanocrystalline material across the estate. This may be due to the nature of the alluvial deposits and indistinct buried ash layers. The sugar estate is located over many old river channels that could have deposited more nanocrystal line material in some fields within a particular area. In addition, local drainage conditions will vary (depending on the nature of deposited alluvial material) and so the formation of allophane will also vary. The amount of allophane ranges between 0.52% and 6.84% in the A horizons and between 0.55% and 6.26% in the B horizons. Singer et al. (2004) considered the allophane

contents found in the Golan Heights (2.5-5.3%) to be low. Therefore, a low content of allophane can be attributed to weathering of volcanic ash minerals directly to more crystalline minerals, predominately in the form of halloysite. Halloysitc formation is favoured over allophane formation when high silica concentrations are maintained, typical in areas with low rainfall and restricted drainage (McDaniel et al. 2012). Thus, the semiarid climate and poor drainage in many areas of the estate (due to high water tables and high sodicity) is likely to result in the preferential formation of halloysitc. Although the ferrihydrite and allophane contents are low, their presence can play a significant role in soil properties, such as surface area and related soil characteristics (Gama-Castro et al. 2000).

The dominance of the crystalline mineral halloysitc is further supported by the (acid oxalate-extractable Al ([Al.sub.o]) 1/2 [Fe.sub.o]) percentage. According to Tsai et al. (2010), an ([Al.sub.o]+ 1/2 [Fe.sub.o]) percentage <1 indicates that Al and Fe are present in predominately crystalline forms, whereas a percentage >2 suggests that Fe and Al are in nanocrystalline forms. The 1USS Working Group WRB (2014) uses an ([Al.sub.o] + 1/2 [Fe.sub.o]) percentage of 2 as the diagnostic limit to define andic soil properties. For all samples taken front the estate, the ([Al.sub.o] + 1/2 [Fe.sub.o]) percentage is <2 (with many values <1) and, as with the ferrihydrite results, the highest average values are in the west and south of the estate and the lowest average values in the Kahe area. The BH Ash has an ([Al.sub.o] + 1/2 [Fe.sub.o]) percentage of 1.62, close to the threshold value of 2 for defining andic soils.

Specific surface area

The SSA of the measured samples ranged from 84.8 to 145.9 [m.sup.2] [g.sup.-1] (mean ([+ or -] standard deviation (s.d.)) 111.5[+ or -] 18.6 [m.sup.2] [g.sup.-1] Table 2).

The dominance of micropores in allophane results in soils that contain allophane having a high SSA and consequently high adsorption capacities (Paterson 1977). Non-volcanic ash soils, unless smcctitic, typically have much lower values than those found across the estate. Because no smectitic minerals were found in any of the soils, it is concluded that the high SSA is a function of the allophane present in the soil that was indicated by the XRD and TEM results. Hughes et al. (2009) studied a variety of soil clays dominated by kaolin minerals and several reference kaolins and found that the SSA of the soil kaolins (without allophane impurities) varied from approximately 40 to 90 [m.sup.2] [g.sup.-1], with the highest values in clays from Indonesian soils formed on volcanic tuff dominated (over 90%) by short, tubular halloysite. Given these results and using the lower allophane SSA value of 400 [m.sup.2] [g.sup.-1] reported by Lowe (1995), Table 3 was constructed.

From Table 3, it can be seen that only 5% allophane can constitute approximately 22% of the total SSA. A similar calculation carried out by Lowe (1995) showed that only 1% of allophane or ferrihydrite accounted for 85% of the total soil mineral surface area in his hypothetical soil.

Implications of the mineralogy for other soil properties and plant growth

The location of the sugar estate has affected the properties of the soils and their potential effects on plant growth. The nearness of two volcanoes, which have supplied repeated amounts of volcanic material to the area, and the relatively low-lying topography have combined so that deposition of ash-derived material, especially via alluvium, has varied across the estate, both spatially and temporally. Superimposed on such differences have been variations in subsequent weathering of the minerals, especially as a result of local soil drainage. Although not the focus of the present study, some possible effects of the soil mineralogy can be suggested.

The sand and silt fractions of the soil revealed an overwhelming dominance of potash feldspars, especially sanidine. Weathering of these minerals (within the local environment initially mainly to halloysite, as indicated by XRD and TEM results) releases K that is able to replenish the soil solution and enable continuing plant supply of this element (Sadusky et al. 1987; Wood and Schrocder 1991; Poss et al. 1996; Sparks 2000; Miles and Farina 2014; Ramos et al. 2015). The presence of calcitc in some of the coarse fractions of the soils and calcrete observed in the field, coupled with the low Ca and Mg contents of the local groundwater (Smedley 2004), suggest that plant availability of these elements may be poor.

Although all soils were dominated by halloysite and small kaolinite particles, it is likely that the small amounts of nanocrystalline minerals have the major effect on the physical and chemical properties of the soils because of their large reactive surface area. Among the properties known to be influenced are soil water retention and transmission, ion exchange and aggregation (Shoji et al. 1993; McDaniel et al. 2012).

Compared with other areas of the estate, the lower-lying areas in the south and west contain the highest proportions of nanocrystallinc material (ferrihydrite and allophane), which is likely to result in a greater amount of adsorbed water being retained in the soil and a lower availability of P to the plant due to the high sorption of P by the nanocrystallinc minerals (Nanzyo 2002; McDaniel et al. 2012). In addition, the greater quantities of allophane are likely to enhance capillary rise, bringing salts from the groundwater table to the soil surface (Jorcnush and Sepaskhah 2003). It is these same areas that are irrigated with the highest-salinity water (Meyer et al. 2009) and thus these soils combine very highly reactive mineral surfaces with potentially damaging amounts of salts. Where these soils have sodic properties, the aggregating effect of allophane may not be able to offset the dispersion of clay particles, and consequently infiltration rate and hydraulic conductivity will be low because of the development of surface crusts and blockage of soil pores by dispersed clay particles. The observed poor crop growth in some of these areas of the estate likely testify to this combination of particular mineralogy combined with poor soil physical and chemical properties.


The mineralogy of the soils across the sugar estate consists predominately of potassium feldspars in the sand and silt fractions and halloysite and high-defect kaolinite, with small quantities of nanocrystallinc minerals in the clay fraction. The poor drainage over the estate would allow sufficient silicon to remain in soil solution for the direct formation of halloysite. The higher quantities of ferrihydrite and allophane and a higher ([Al.sub.o] + 1/2 Fen) percentage in the BH Ash indicate that buried ash layers that have been either alluvially or aeolian deposited play a major role in the amount of nanocrystallinc material remaining in the soil profiles, and may explain the variability in the percentage of nanocrystalline material within the various areas and between fields. Furthermore, more ferrihydrite and a higher ([Al.sub.o] + 1/2 [Fe.sub.o]) percentage in the south and west areas suggest greater alluvial deposition of nanocrystallinc minerals in these more lower-lying areas of the estate. Although all the soils on the estate are dominated by halloysite and small kaolinite particles, it is likely that the presence of only small amounts of allophane with high SSA has a strong effect on the physical and chemical properties of the soils. Among these, soil water retention and transmission and aggregation are likely to be of major importance under the current irrigated cropping system. The presence of allophane may encourage capillary rise of salt-enriched groundwater and, if combined with irrigation using low-quality water (as in the south and west areas), the aggregating effect of the nanocrystallinc minerals may be ineffective, allowing clay dispersion and reduction in infiltration and hydraulic conductivity.


The authors are grateful to Shirley Mackellar (Centre for Electron Microscopy, University of KwaZulu-Natal) who helped with the TEM work and Hanlie Botha (Department of Process Engineering, University of Stellenbosch) who provided the SSA data. Special thanks to Jan Meyer for getting us started, to Pierre Noel and Jean Robert Lincoln for their support and input into the project and to TPC Ltd for partial funding of the project and providing a postgraduate student bursary for T. S. Taylor.


Adamo P, Violante P, Wilson M.I (2001) Tubular and spheroidal halloysite in pyroclastic deposits in the area of Roccamonfina volcano (southern Italy). Geoderma 99, 295-316. doi: 10.1016/S0016-7061 (00)00076-8

Armas-Espinel S. Hemandez-Moreno JM, Munoz-Carpena R. Regalado CM (2003) Physical properties of 'sorriba'-cultivated volcanic soils from Tenerife in relation to andic diagnostic parameters. Geoderma 117, 297 311. doi : 10.1016/S0016-7061 (03)00130-7

Bates T (1960) Halloysite and gibbsite formation in Hawaii. Clays and Clay Minerals 9, 315-328. doi:10.1346/CCMN.1960.0090119

Brogna D (2004) Mapping of salinity and sodicity problems using remote sensing and GIS techniques at TPC Ltd., a sugar cane estate in Tanzania. MSc Thesis, Cranfield University, Silsoe, UK.

Broquen P, Lobartini JC, Candan F, Falbo G (2005) Allophane, aluminium, and organic matter accumulation across a bioclimatic sequence of volcanic ash soils of Argentina. Geoderma 129, 167-177. doi: 10.1016/ j.geodcrma.2004.12.041

Buhmann C, Fey MV, de Villiers JM (1985) Aspects of the X-ray identification of swelling clay minerals in soils and sediments. South African Journal of Science 81, 505-509.

Carter DL, Mortland MM, Kemper WD (1986) Specific surface. In 'Methods of soil analysis. Part 1, 2nd edition. Number 9 (part 1 in the series): agronomy'. (Ed. A Klute) pp. 413-423. (American Society of Agronomy: Madison, WI)

Drouza S, Georgoulias FA, Moustakas NK (2007) Investigation of soils developed on volcanic materials in Nisyros Island, Greece. Catena 70, 340 349.doi:10.1016/j.catena.2006.11.001

Fontes JC, Goncalves MC, Pereira LS (2004) Andosols of Terceira, Azores: measurement and significance of soil hydraulic properties. Catena 56, 145-154. doi: 10.1016/j.catena.2003.10.008

Funakawa S, Yoshida H, Watanabe T, Sugihara S, Kilasara M, Kosaki T (2012) Soil fertility status and its determining factors in Tanzania. In 'Soil health and land use management'. (F.d. MC Flcmandez Soriano) pp. 3 16. (InTech: Rijeka, Croatia and Shanghai, China). Available at soil-fertility-status-and-its-determining-factors-in-tanzania [verified 3 April 2016],

Gama-Castro JE, Solleiro-Rebolledo E, Vallejo-Gomez E (2000) Weathered pumice influence on selected alluvial soil properties in west Nayarit, Mexico. Soil & Tillage Research 55, 143 165. doi: 10.1016/S0167-1987 (00)00114-8

Flarsh J, Chorover J, Nizcyimana E (2002) Allophane and Imogolite. In 'Soil mineralogy with environmental applications'. (Eds JB Dixon. DG Schulze) pp. 291 -322. (Soil Science Society of America: Madison. Wl)

Hart RD, Gilkes RJ, Siradz S, Singh B (2002) The nature of soil kaolins from Indonesia and Western Australia. Clays and Clay Minerals 50. 198-207. doi: 10.1346/000986002760832793

Hart RD. Wiriyakitnateekul W, Gilkes RJ (2003) Properties of soil kaolins from Thailand. Clay Minerals 38, 71-94. doi: 10.1180/0009855033 810080

Hepper EN, Buschiazzo DE, Hevia GG, Urioste A, Anton L (2006) Clay mineralogy, cation exchange capacity and specific surface area of loess soils with different volcanic ash contents. Geoderma 135, 216-223. doi: 10.1016/j.geoderma.2005.12.005

Hughes JC, Gilkes RJ, Hart RD (2009) Intercalation of reference and soil kaolins in relation to physico-chemical and structural properties. Applied Clay Science 45, 24-35. doi:10.1016/j.clay.2009.04.006

1USS Working Group WRB (2014) World reference base for soil resources 2014. International soil classification system for naming soils and creating legends for soil maps. World Soil Resources Report No. 106, F AO, Rome.

Jager TJ (1982) Soils of the Serengeti Woodlands, Tanzania. Agricultural Research Reports 912, Centre for Agricultural Publishing and Documentation, Wagcningen.

Jorenush MH, Sepaskhah AR (2003) Modelling capillary rise and soil salinity for shallow saline water table under irrigated and nonirrigated conditions. Agricultural Water Management 61, 125-141. doi: 10.1016/S0378-3774(02)00176-2

Kanket W, Suddhiprakam A, Kheoruenromne I, Gilkes RJ (2005) Chemical and crystallographic properties of kaolin from Ultisols in Thailand. Clays and Clay Minerals 53, 478-489. doi:10.1346/CCMN. 2005.0530505

Kauthold S, Ufer K, Kaufhold A, Stucki JW, Anastacio AS, Jhan R, Dohrmann R (2010) Quantification of allophane from Ecuador. Clays and Clay Minerals 58, 707-716. doi:10.1346/CCMN.2010.0580509

Lowe DJ (1995) Teaching clays: from ashes to allophane. In 'Clays: controlling the environment. Proceedings of the 10th International Clay Conference', 18-23 July 1993 Adelaide, Australia. (Eds GJ Churchman, RW Filzpatrick, RA Egglcton) pp. 19-23. (CSIRO Publishing: Melbourne)

McDaniel PA, Lowe DJ, Amalds O, Ping C-L (2012) Andisols. In 'Handbook of soil sciences, 2nd edition, Vol. 1, properties and processes'. (Eds PM Huang, Y Li, ME Sumner) pp. 33.29-33.48. (CRC Press: Boca Raton, FL)

McKeague JA (1967) An evaluation of 0.1 M pyrophosphate and pyrophosphate-dithionite in comparison with oxalate as extractants of the accumulation products in Podzols and some other soils. Canadian Journal of Soil Science 47, 95 99. doi: 10.4141/cjss67-017

McKeague JA, Day JH (1966) Dithionite and oxalate-extractable Fe and Al as aids in differentiating various classes of soils. Canadian Journal of Soil Science 46, 13 22. doi:10.4141/cjss66-003

Mehra OP, Jackson ML (1958) Iron oxide removal from soils and clays by a dithionite citrate system buffered with sodium bicarbonate. Clay Minerals 7, 317-327. doi:10.1346/CCMN.1958.0070122

Meyer JH, Heathman W, Turner P, Betjak N (2009) Final report of a detailed soil survey at TPC estate, Tanzania. MWH Global, Framework Contract Beneficiaries Lot No. 6, Request for Services No. 2007/141038, Mt. Edgecombe, South Africa.

Miles N, Farina MPW (2014) Towards the more efficient use of fertiliser potassium: prediction of 'slowly-available' potassium reserves in soils. Proceedings of the South African Sugar Technologists' Association 87, 330-333.

Nanzyo M (2002) Unique properties of volcanic ash soils. Global Environmental Research 6, 99-1127.

Ndayiragije S, Delvaux B (2003) Coexsistence of allophane, gibbsite, kaolinite and hydroxyl Al-interlayered 2 :1 clay minerals in a perudic Andosol. Geoderma 117, 203 214. doi: 10.1016/S0016-7061(03) 00123-X

Nieuwenhuyse A, Verburg PSJ, Jongmans AG (2000) Mineralogy of a soil chronosequencc on an andesitic lava in humid tropical Costa Rica. Geoderma 98, 61-82. doi : 10.1016/S0016-7061 (00)00052-5

Parfitt RL, Wilson AD (1985) Estimation of allophane and halloysitc in three sequences of volcanic soils. New Zealand. Catena (Suppl 7), 1 8.

Parfitt RL, Childs CW, Eden DN (1988) Ferrihydrite and allophane for Andepts from Hawaii and implications for their classification. Geoderma 41, 223 241. doi: 10.1016/0016-7061(88)90062-6

Paterson E (1977) Specific surface area and pore structure of allophanic clay soils. Clay Minerals 12, 1-9. doi: 10.1180/claymin. 1977.012.1.01

Poss R, Fardeau JC, Saragoni H (1996) Sustainable agriculture in the tropics: the case of potassium under maize cropping in Togo. Nutrient Cycling in Agroecosystems 46, 205-213. doi: 10.1007/ BF00420555

Ramos CG, Querol X, Oliveira MLS, Pires K, Kautzmann RM, Oliveira LFS (2015) A preliminary evaluation of volcanic rock powder for application in agriculture as a remineralizer. The Science of the Total Environment 512-513, 371-380. doi:10.1016/j.scitotenv.2014.12.070

Sadusky MC, Sparks DL, Noll MR, Hendricks GJ (1987) Kinetics and mechanisms of potassium release from sandy Middle Atlantic Coastal Plain soils. Soil Science Society of America Journal 51, 1460-1465. doi: 10.2136/sssaj 1987.03615995005100060011 x

Sci J, Morato F, Kra G, Staunton S, Quiquampoix H, Jumas JC, OlivierFourcade J (2006) Mineralogical, crystallographic and morphological characteristics of natural kaolins from the Ivory Coast (West Africa). Journal of African Earth Sciences 46, 245-252. doi: 10.1016/j.jafrearsci. 2006.06.002

Sellitto VM, Palumbo G, Colombo C, Terribile F, Schulze DG (2010) Formation of volcanic ash soils in the Matese Mountains of southern Italy. In '19th World Congress of Soil Science: soil solutions for a changing world', 1-6 August 2010, Brisbane, Australia. (Eds R Gilkes, N Prakongkep) pp. 25-28 (Curran Associates, Inc.: New York, NY)

Shoji S, Nanzyo M, Dahlgren RA (1993) "Volcanic ash soils: genesis, properties and utilization.' Developments in soil science 21. (Elsevier Science Publishers: Amsterdam)

Singer A, Zarei M, Lange FM, Stahr K (2004) Halloysite characteristics and formation in northern Golan Heights. Geoderma 123, 279-295. doi: 10.1016/j.geoderma.2004.02.012

Singh B, Gilkes RJ (1992) Properties of soil kaolinites from south-western Australia. Journal of Soil Science 43, 645 667. doi: 10.1111/j.13652389.1992.tb00165.x

Singleton PL, Mcleod M, Percival HJ (1989) Allophane and halloysite content and soil solution silicon in soils from rhyolitic volcanic material. New Zealand. Australian Journal of Soil Research 27, 67 77. doi: 10.1071 /SR9890067

Smedley PL (2004) Groundwater quality: Tanzania. Available at https:// [verified 3 April 2016].

Soil Classification Working Group (1991) 'Soil classification: a taxonomic system for South Africa.' (Department of Agricultural Development: Pretoria)

Soil Survey Staff (2014) 'Keys to soil taxonomy.' 12th edn. (US Department of Agriculture. Natural Resources Conservation Service: Washington, DC)

Sparks DL (2000) Bioavailability of soil potassium. In 'Handbook of soil science'. (Ed. ME Sumner) pp. D-38-D-53. (CRC Press: Boca Raton, FL)

Tsai CC, Chen ZS, Kao Cl, Ottner F, Kao SJ, Zehetner F (2010) Pedogcnic development of volcanic ash soils along a climosequence in Northern Taiwan. Geoderma 156, 48 59. doi:10.1016/j.geoderma.2010.01.007

Vacca A, Adama M, Pigna M, Violante P (2003) Genesis of tephra-derived soils from the Roccamonfina volcano, south central Italy. Soil Science Society of America Journal 67, 198-207. doi: 10.2136/sssaj2003.1980

Watanabe T, Funakawa S, Kosaki T (2006) Clay mineralogy and its relationship to soil solution composition in soils from different weathering environments of humid Asia: Japan, Thailand and Indonesia. Geoderma 136. 51-63. doi:10.1016/j.geoderma.2006.02.001

Wood RA, Schroeder BL (1991) Release of non-exchangeable potassium reserves from a range of sugar industry soils. Proceedings of the South African Sugar Technologists' Association 65, 47-52.

T. S. Taylor (A,C), J. C. Hughes (A), and L. W. Titshall (A,B)

(A) Soil Science, School of Agricultural, Earth and Environmental Sciences, University of KwaZulu-Natal, Private Bag X01, Scottsville 3209, South Africa.

(B) Present address: Institute for Commercial Forestry Research, PO Box 100281, Scottsville 3209, South Africa.

(C) Corresponding author. Email:

Table 1. Range and mean [+ or -] s.d. (n = 23) of Fe, Al and Si
extracted by acid ammonium oxalate, sodium citrate-bicarbonatc-
dithionite and sodium pyrophosphate for the A and B horizons of the
sampled soils from the sugar estate in Tanzania n.d., not determined

Element   Horizon            Acid oxalate

                      Range     Mean [+ or -] s.d.

Fe           A      0.28-1.08   0.56 [+ or -] 0.2
             B      0.20 1.00   0.61 [+ or -] 0.2
Al           A      0.16 1.07   0.50 [+ or -] 0.3
             B      0.16-1.27   0.54 [+ or -] 0.3
Si           A      0.07-0.96   0.41 [+ or -] 0.3
             B      0.09 0.88   0.40 [+ or -] 0.3

Element   Horizon   Citrate-bicarbonate-dithionite

                      Range     Mean [+ or -] s.d.

Fe           A      1.12-2.90   1.74 [+ or -] 0.5
             B      0.93-3.34   1.90 [+ or -] 0.7
Al           A      0.19-3.27   1.21 [+ or -] 0.8
             B      0.46 3.13   1.45 [+ or -] 0.9
Si           A        n.d.             n.d.
             B        n.d.             n.d.

Element   Horizon           Pyrophosphate

                      Range     Mean [+ or -] s.d.

Fe           A      0.00-0.08   0.02 [+ or -] 0.02
             B      0.00-0.10   0.02 [+ or -] 0.02
Al           A      0.03-0.17   0.08 [+ or -] 0.03
             B      0.05-0.29   0.11 [+ or -] 0.05
Si           A        n.d.             n.d.
             B        n.d.             n.d.

Table 2. Specific surface area (SSA) of clay (<2 [micro]m) samples
from selected A and B horizons on the sugar estate

Sample           Area    SSA ([m.sup.2] [g.sup.-1])

  A horizon      North              97.3
  B horizon      North             113.1
  A horizon      West               84.8
  B horizon      West              144.5
  A horizon      South              91.0
  B horizon      South              93.8
  A horizon      South             110.8
  B horizon      South             116.4
B3 A horizon     East               99.1
E6 B horizon     East              102.2
  A horizon      Kahe              126.2
  B horizon      Kahe              145.9
KH29 A horizon   Kahe              120.4
KH15 B horizon   Kahe              115.6

Table 3. Hypothetical calculation for the Brunauer-Emmett-Teller
(BET) specific surface area (SSA) of clay minerals in the soils on
the sugar estate

Mineral                            Assumed              Assumed %
                                   BF.T SSA          mineral in clay
                            ([m.sup.2] [g.sup.-1])    fraction (A)

Allophane                          400 (B)                  5
Halloysite (tubular)                80 (C)                 80
Kaolinite (<0.2 [micro]m)           55 (C)                 15


Mineral                             Proportion
                                     of clay
                            SSA ([m.sup.2] [g.sup.-1])

Allophane                               20
Halloysite (tubular)                    64
Kaolinite (<0.2 [micro]m)              8.25

Total                                 92.25

(A) Allophane calculated from [Si.sub.o] data (present study);
halloysite and kaolinite estimated from X-ray diffraction and
transmission electron microscopy results in the present study.

(B) From Lowe (1995).

(C) From Hughes et al (2009).


Please note: Some tables or figures were omitted from this article.
COPYRIGHT 2016 CSIRO Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2016 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Taylor, T.S.; Hughes, J.C.; Titshall, L.W.
Publication:Soil Research
Article Type:Report
Geographic Code:6TANZ
Date:Nov 1, 2016
Previous Article:Multifractal analysis of soil hydraulic properties in arid areas.
Next Article:Effects of vegetation cover on sediment particle size distribution and transport processes in natural rainfall conditions on post-fire hillslope...

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters