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


     

Published online 8 June 2007
Published in Soil Sci Soc Am J 71:1119-1127 (2007)
DOI: 10.2136/sssaj2006.0322
© 2007 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nierop, K. G. J.
Right arrow Articles by Verstraten, J. M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Nierop, K. G. J.
Right arrow Articles by Verstraten, J. M.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Nierop, K. G. J.
Right arrow Articles by Verstraten, J. M.
Related Collections
Right arrow Water Quality
Right arrow Dryland Soils
Right arrow Carbon Sequestration

SOIL CHEMISTRY

Organic Matter in Volcanic Ash Soils under Forest and Páramo along an Ecuadorian Altitudinal Transect

Klaas G. J. Nierop*, Femke H. Tonneijck, Boris Jansen and Jacobus M. Verstraten

Institute for Biodiversity and Ecosystem Dynamics (IBED), Earth Surface Processes and Materials (ESPM), Universiteit van Amsterdam, Nieuwe Achtergracht 166, 1018 WV Amsterdam, the Netherlands

* Corresponding author (k.g.j.nierop{at}science.uva.nl).


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The volcanic ash soils along an altitudinal transect in Guandera Biological Station in northern Ecuador have been developed under varying vegetation around the upper forest line. Generally, the soils currently covered by forest are composed of Fulvic Andosols (melanic index >1.7) while those under páramo (tropical alpine grasslands) have developed into Melanic Andosols. This vegetation effect on soil formation is believed to be associated with differences in organic matter composition. In this study, we examined whether Fulvic Andosols differed from Melanic Andosols in organic matter composition. Using analytical pyrolysis techniques, we found hardly any differences in the organic matter characteristics related to vegetation cover, not even between soils that supposedly have been covered by forest and páramo for millennia. Small differences were found within the lipid compounds, while the polysaccharides and lignin were virtually absent from the (deeper) mineral soil horizons. The low abundance of polysaccharides in soils that have undergone severe organic matter decomposition is not unusual for most soils, but is uncommon in other Andosols studied with the same pyrolysis techniques.

Abbreviations: OM, organic matter • SOM, soil organic matter • THM, thermally assisted hydrolysis and methylation • UFL, upper forest line


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Andosols (FAO, 2006) or Andisols (Soil Survey Staff, 2006) are developed in volcanic deposits of variable age and exist in a large variety of climate zones. Few Andosols have formed on other weatherable rocks, and Andosols may eventually change into other soil types, such as Ferralsols or Podzols (Dahlgren et al., 2004). Volcanic ash soils cover about 0.84% of the world's land area (Dahlgren et al., 2004) and are dominated by amorphous Al silicates (allophane and imogolite), by ferrihydrite or by Al and Fe–soil organic matter (SOM) complexes (Parfitt et al., 1980). Andosols are dominated by either amorphous Al silicates (allophanic Andosols) or by Al and Fe–SOM complexes (nonallophanic Andosols). At relatively low soil pH (<4.5), the formation of Al(Fe)–SOM complexes inhibits Al hydroxide polymerization and thus the formation of amorphous Al silicates (Dahlgren et al., 1993). When SOM contents are lower and the soil pH is in the range of 5 to 7, coprecipitation of Al and Si to form allophane and imogolite is promoted (Nanzyo et al., 1993; Ugolini and Dahlgren, 2002).

In andic subsoils, 14C ages range from 5000 to 30000 yr BCE. These mean residence times are much longer than those observed in any other soil type (Wada and Aomine, 1973; Torn et al., 1997). Among mineral soil orders, Andosols contain the largest amounts of SOM equaling about 5% of the global soil C (Dahlgren et al., 2004; Eswaran et al., 1993), which means they play a distinct role in the global C cycle. Despite the large body of literature on Andosols, very little is known about the composition of their SOM. The studies focusing on organic matter have dealt mainly with the humic acid fractions and most of them are from Japan. They used absorption spectra to classify among humic acid types A, B, P, and Rp (Nanzyo et al., 1993, and references cited therein). According to these studies, vegetation has a profound effect on soil formation through various SOM transformation pathways. In general, under grasslands, Melanic Andosols are formed, which are dominated by humic acid type A, whereas under forest, Fulvic Andosols with humic acid types B and P develop, which are supposedly less humified than the A type (Honna et al., 1988). Furthermore, 13C-NMR was applied to Italian and Japanese volcanic soils (Conte et al., 1997a, 1997b, 2003; Golchin et al., 1997a, 1997b), while others have focused on lipid fractions (Naafs and van Bergen, 2002; Naafs et al., 2004, 2005). Only a few studies have examined a detailed molecular SOM characterization (Naafs, 2004; Nierop et al., 2005; Nierop and Buurman, 2006). In contrast to many non-volcanic-ash soils, the SOM of Andosols was dominated by polysaccharides and protein-derived compounds, while lignin and lipids were heavily depleted. Moreover, differences in vegetation coverage did not affect the SOM composition at the molecular level (Nierop et al., 2005).

Andosols in the Ecuadorian Andes develop at high altitudes (>3200 m) under a cool and relatively wet, tropical alpine climate. Under forest, thick ectorganic layers are formed on top of the mineral soils, while the soils under páramo (grassland) are devoid of such a layer. This vegetation-induced effect on soil profile development affects also the classification of the soils: the mineral soils under forest have a melanic index >1.7 and andic properties, but due to the thick ectorganic layers have to be classified as Cambisols or Histosols according to the World Reference Base classification (FAO, 2006). In contrast, the soils under páramo (grassland) are classified as Melanic Andosols (FAO, 2006). We studied the SOM from Ecuadorian Andosols from the Guandera Biological Station along an altitudinal transect intersecting the upper forest line (UFL) using pyrolysis and thermally assisted hydrolysis and methylation (THM). The aims of the present study were to examine (i) whether the SOM composition in these wet and cool tropical Andosols differs from SOM from warmer and drier Andosols in temperate regions, and (ii) to what extent the vegetation-induced differences in soil profile development and classification, i.e., fulvic–melanic distinction, are also reflected in SOM differentiation at the molecular scale.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Soil Profiles and Classification
The study sites are located within the Guandera Biological Station in northern Ecuador, within the province of Cajas, near the border with Colombia (Fig. 1). Guandera lies in the eastern Cordillera (mountain range) and preserves a (semi-)natural upper forest line at approximately 3650 m above sea level (Fig. 2, Table 1). The transition from forest to páramo is abrupt, indicating some human influence (Laegaard, 1992), but the forest still contains trees with a diameter at breast height of 70 cm. Some forest patches occur above the current UFL. The dominant species within the forest are Clusia flaviflora Engl., Weinmannia cochensis Hieron. and Ilex colombiana Cuatrec. and the páramo is characterized by Calamagrostis effusa (Kunth) Steud. (bunchgrass) and Espeletia pycnophylla Cuatrec. Guandera receives, on average, 1900 mm of precipitation annually and mean annual temperatures range from 10°C at 3000 m above sea level to 4°C at 4000 m above sea level. Soil pits of approximately 2-m2 surface area were excavated down to a depth of 1.5 to 2 m, depending on the soil profile, and soil profiles were described according to the FAO guidelines (FAO, 1990) and classified according to the World Reference Base (FAO, 2006). One paleosol (buried soil) was generally observed beneath the current soil, indicating at least two main events of tephra deposition (Tonneijck et al., 2006). We selected a forest patch above the current UFL [G5a, dominated by W. cochensis, Miconia trinifolia, I. colombiana, Blechnum schomburgkii (Klotzsch) C. Chr. and Myrsine andina (Mer.) Piploy], the volcanic soil of which has been developed into an Andic Cambisol with a fulvic horizon and which is a transition between a Histosol and an Andosol. Also, a profile in the páramo 20 m next to it was chosen, formed into a Melanic Andosol (G5b), to determine whether the effect of the current vegetation did not lead to a difference in soil genesis only, but was reflected in the SOM (trans)formation as well. In addition, we selected a site with a Histosol overlying a mineral soil that has a melanic index >1.7 and andic properties (G1, dominated by C. flaviflori), which presumably has been continuous forest during the last 3000 yr, and a site with a Melanic Andosol (G7) that most probably has always been páramo during that period (Table 1).


Figure 1
View larger version (24K):
[in this window]
[in a new window]

  		Fig. 1. Map of Ecuador: the square indicates the region where Guandera Biological Station is located.

 

Figure 2
View larger version (35K):
[in this window]
[in a new window]

  		Fig. 2. Map of the region where Guandera Biological Station is located (square in Fig. 1) showing Guandera Biological Station ({blacktriangleup}) and nearby towns or villages ({blacksquare}).

 

View this table:
[in this window]
[in a new window]

  		Table 1. Site descriptions.

 
Soil Sampling and Sample Pretreatment
The main focus was on the current soil profile. We took bulk soil samples for chemical analyses (approximately 1-kg field-moist material) from each horizon, and within thick horizons at regular depths. Each bulk soil sample was obtained from a vertical interval of approximately 5 cm, the middle of which was noted as the depth of the sample.

The field-moist bulk samples were homogenized manually and roots and coarse material (>2 mm) were removed. Part of the homogenized soil sample was dried at 30°C, sieved over a 2-mm mesh, and ground for analysis of total C and N, and for pyrolysis. Soil samples were stored at 2°C before analyses. Forest litter, including woody material, was collected from the soil surface. Páramo litter was gathered from the soil surface and from within the grass tussocks where the litter was predominantly located. Litter was dried at 30°C as soon as possible after collection (within 3 wk) and then stored at 2°C. Before analysis, litter was dried again at 30°C and then chopped to pieces <10 mm. Thereafter a part of the chopped litter was ground for analysis of total C and N, and for pyrolysis.

Chemical Analyses
Soil pH was measured with a glass electrode in water, a 0.01 M CaCl2 solution, and a 1.0 M KCl solution (w/v 1:5 for mineral and w/v 1:10 for organic samples). As an indicator of amorphous material, the soil pH was also measured after 2 min stirring with a 1 M NaF solution (w/v 1:50) according to Fieldes and Perott (1966). Total organic C and N of ground mineral and organic samples were determined by an Elementar VarioEL CNS analyzer (Elementar, Hanau, Germany) after drying at 105°C (mineral samples) or 70°C (organic samples) (Table 2). Pyrophosphate-extractable Al and Fe and acid oxalate extractable Al, Fe, and Si (Alp, Fep, and Alo, Feo, and Sio, respectively) were determined on field-moist mineral soil samples using the standard procedure described by Van Reeuwijk (2002). Allophane content of the samples was calculated from Alo, Alp and Sio according to Parfitt and Wilson (1985) as modified by Mizota and van Reeuwijk (1989) (Table 3). As recommended by Mizota and van Reeuwijk (1989), we applied a maximum (Alo – Alp)/Sio atomic ratio of 2.5, assuming that an excess of Alo should be allocated to hydroxy-Al (e.g., from Al – Alo or poorly ordered gibbsite).


View this table:
[in this window]
[in a new window]

  		Table 2. Carbon and nitrogen content and ratio, pH in water, CaCl2, and KCL, and soil organic matter (SOM) age of Profiles G1, G5a, G5b, and G7.

 

View this table:
[in this window]
[in a new window]

  		Table 3. Inorganic chemical characteristics of Profiles G1, G5a, G5b, and G7.

 
Pyrolysis and Thermally Assisted Hydrolysis and Methylation
Pyrolysis was performed on a Horizon Instruments Curie-Point pyrolyzer. Samples were heated for 5 s at 600°C. The pyrolysis unit was connected to a ThermoQuest Trace GC 2000 gas chromatograph (Thermo Scientific, Waltham, MA) and the products were separated by a fused silica column (J&W, 30 m, 0.32-mm i.d.) coated with DB-1 (film thickness 0.50 µm). Helium was used as carrier gas. The oven was initially kept at 40°C for 1 min, then heated at a rate of 7°C min–1 to 320°C and maintained at that temperature for 15 min. The column was coupled to a Finnigan Trace MS mass spectrometer (mass range m/z 45–600, ionization energy 70 eV, cycle time 1 s; Thermo Scientific).

Before THM, the residues after lipid extraction were pressed onto Curie-point wires, after which a droplet of a 25% solution of tetramethylammonium hydroxide in water was added to the samples, which were subsequently dried by a 100-W halogen lamp. Analysis of the THM products by gas chromatography (GC)–mass spectrometery was performed similar to the conditions used for pyrolysis. With THM, hydrolyzable bonds are cleaved and the resulting carboxylic acid and hydroxyl groups are transformed in situ into their corresponding methyl esters and methyl ethers. Identification of the compounds was performed by their mass spectra using a National Institutes of Standards and Technology library or by interpretation of the spectra, by their GC retention times, or by comparison with literature data.


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Organic C contents of the soils studied decreased with depth, but were always >8% (Table 2). The calculated allophane contents were <6% in the current soils (Table 3). Similar to Poulenard et al. (2003) and Buytaert et al. (2006), Alp/Alo ratios were very close to 1 in the surface horizons, but decreased with depth, and accordingly, the soils are considered as nonallophanic Andosols. As a consequence, retention of products by allophane or other noncrystalline clay materials on pyrolysis was of minor importance and the abundance of pyrolysis products released was high enough to obtain a representative picture of the SOM composition. As mentioned above, both mineral soils under a current forest cover have a melanic index >1.7 (fulvic) and are not Andosols according to World Reference Base classification (FAO, 2006), but for brevity we use the term Fulvic Andosols for these soils under study, while those under páramo are assigned as Melanic Andosols.

Profile G5a (Forest Patch in Páramo)
The pyrolysate of the litter (O1 horizon) of the forest patch (G5a) was dominated by products from lignin (guaiacol [2-methoxyphenol] and syringol [2,6-dimethoxyphenol] and their derivatives) and polysaccharides (2-furaldehyde, 5-methylfuraldhyde, 4-hydroxy-5,6-dihydro-(2H)-pyran-2-one, levoglucosenone, and levoglucosan) (Ralph and Hatfield, 1991), while a relatively small amount of lipids was encountered and represented by n-alkanes (C27, C29, and C31, dominated by C29) (Fig. 3). In the O2 horizon, lignin was still abundant, but alteration took place in the form of side-chain oxidation, e.g., the 4-propen-2-ylguaiacol and -syringol isomers decreased relatively, while further degradation was reflected by the relative decline of syringols with respect to guaiacols. In the mineral soil (Ah and Bw horizons), lignin-derived compounds were hardly present and polysaccharides were abundant in the top of the Ah horizon, and of which levoglucosan decreased compared with the overlying horizons. Furthermore, lipid products consisting of homologous series of n-alkenes and n-alkanes (C12–C31), n-alkanoic acids (C14, C16:1, C16, C18), and an n-alkanol (C22) became enriched. The relatively high abundance of the n-alkenes with chain lengths C18, C20, C22, and C24 probably reflects pyrolysis products of the corresponding n-alkanols either as free alkanols and subsequent dehydration induced by the mineral surfaces (Nierop and van Bergen, 2002) or esterified alkanols (Van Smeerdijk and Boon, 1987; Nierop et al., 2001). These alkanols were identified as methylated alkanols on THM (data not shown). The presence of 2- and other mid-chain alkenes (not indicated in Fig. 3) implies interaction of alkanols and esters with the mineral particles on pyrolysis (Nierop and van Bergen, 2002). The THM also revealed the presence of cutin and suberin by 10,16-dihydroxyhexadecanoic acid, 9,10,18-trihydroxyocttdecanoic acid, and {omega}-hydroxyalkanoic acids (C22 > C16). As C22 > C16, suberin seems to be a greater contributor to SOM than cutin (Kolattukudy, 2001; Naafs et al., 2005). This suberin predominance is not unexpected, because roots, which contain suberin, are the main source of SOM in deeper subsoils of Andosols (Naafs et al., 2005). Also relatively more abundant were the n-alkanes with chain lengths C27, C29, and C31 and they are likely to be evaporation products from, e.g., epicuticular waxes rather than true pyrolysis products. The homologous series of n-alkenes and n-alkanes has been attributed to cutan and suberan (Tegelaar et al., 1989; Nierop, 1998; Naafs, 2004) Given the higher suberin contribution mentioned above, in our case suberan is likely to be a more important source than cutan. Nevertheless, cutan is considered to be present in drought-adapted plants (Boom et al., 2005) such as the Clusia (Lüttge, 1999), which occur in these ecosystems. Toward the deeper Bw horizons, polysaccharides became depleted, while aliphatic compounds turned into the dominant fraction.


Figure 3
View larger version (25K):
[in this window]
[in a new window]

  		Fig. 3. Gas chromatograms of pyrolysates from Profile G5a: P = phenol, C = catechol, G = guaiacol (2-methoxyphenol), and S = syringol (2,6-dimethoxyphenol). Indicated side-chains are at the 4-position of P, G, or S: n-alkene (x); n-alkane (+); alkanoic acid ({diamondsuit}); alkanol (•); Cx indicates chain length.

 
Profile G5b (Páramo Near Forest Patch)
The litter (O horizon) of the páramo (G5b) clearly differed from that of the forest patch (G5a) and the pyrolysate displayed, not unexpectedly, a typical grass signature, i.e., lignin-derived compounds dominated by far by 4-vinylguaiacol (van Bergen et al., 1997; Nierop, 2001) (Fig. 4). Other abundant products were derived from polysaccharides, alkanoic acids, and C31 alkane. In the mineral soil, and similar to the forest patch, lignin was hardly present (only a small peak of 4-vinylguaiacol), and polysaccharides and particularly lipids dominated the Ah horizon. Further down, the pyrolysate of the Bw horizon consisted mainly of n-alkenes and n-alkanes with the same dominance of C18, C20, C22, and C24 n-alkenes as in the mineral soil under the forest patch. Similar to G5a, THM produced methylated alkanols with the same chain lengths and in the same distribution as these n-alkenes (data not shown). The resemblance between the subsoils of both G5a and G5b may suggest a similar origin of the SOM, and therefore, possibly a similar vegetation history as reflected by the alkanols and the slightly higher concentration of C29 alkane compared with the C31 analog. Therefore, we cannot conclude that the current differences in vegetation and subsequent soil development had important effects on the SOM composition.


Figure 4
View larger version (24K):
[in this window]
[in a new window]

  		Fig. 4. Gas chromatograms of pyrolysates from Profile G5b: P = phenol, C = catechol, G = guaiacol (2-methoxyphenol), and S = syringol (2,6-dimethoxyphenol). Indicated side-chains are at the 4-position of P, G, or S: n-alkene (x); n-alkane (+); alkanoic acid ({diamondsuit}); alkanol (•); Cx indicates chain length.

 
Profile G1 (Cloud Forest)
The sites G5a and G5b did not differ greatly in SOM formation, which is probably due to a too short a period of time in which the vegetation differed between the sites. Therefore, we also chose to study two profiles (G1 and G7) that presumably always have been under forest and páramo, respectively, during SOM transformation. The Clusia cloud forest profile (G1) is considered to be present at an altitude below the historical UFL since the forest was established in this area more than 3000 yr ago, and therefore it functions as a reference of forest-derived SOM. The main vegetation is Clusia, the woody litter (O1 horizon) of which appeared to have a relatively small amount of lignin (Fig. 5). Relatively large amounts of phenol, methylphenol, and catechol were observed on pyrolysis, probably derived from tannins (Galletti and Reeves, 1992). The most abundant products were C27 and particularly C29 n-alkane, lanosterols, and a carotene derivative. This latter compound was the most abundant compound in the O2 and O3 horizons, but it disappeared from the mineral horizons. Both lignin and tannin were still identifiable in the O2 and O3 horizons, and similar to G5a, degradation of lignin occurred as indicated by selective removal of syringyl moieties and side-chain oxidation. Polysaccharides were identified in the O1, O2, and O3 horizons, but were almost completely absent from the mineral horizons. At the top of the Ah horizon, THM revealed the presence of {omega}-hydroxyalkanoic acids, with C22 and C24 as the dominant chain lengths, implying again a predominant suberin, and thus root-derived SOM. The pyrolysates of Ah and Bw horizons mainly consisted of n-alkenes and n-alkanes (C10–C31). The C29 n-alkane remained the most abundant alkane in the mineral horizons. The relatively higher abundance of C18, C20, C22, and C24 n-alkenes in comparison with the other alkenes was also found in this profile, suggesting a rather stabilized SOM in terms of composition, and possibly a similar vegetative origin as profiles G5a and G5b.


Figure 5
View larger version (20K):
[in this window]
[in a new window]

  		Fig. 5. Gas chromatograms of pyrolysates from Profile G1: P = phenol, C = catechol, G = guaiacol (2-methoxyphenol), and S = syringol (2,6-dimethoxyphenol). Indicated side-chains are at the 4-position of P, G, or S: n-alkene (x); n-alkane (+); alkanoic acid ({diamondsuit}); alkanol (•); Cx indicates chain length.

 
Profile G7 (Páramo)
The litter (O horizon) of Profile G7 was rather similar to that of G5b, albeit that the lipids were relatively more abundant (Fig. 6). Also similar to G5b, lignin was hardly present in the Ah1 horizon and absent altogether from the Ah2 horizon. Both horizons contained some polysaccharide-derived compounds in their pyrolysates, coinciding with an increasing contribution of n-alkenes and n-alkanes. Once again, and similar to the other profiles, THM revealed that C22 {omega}-hydroxyalkanoic acid was much more abundant than the most prominent cutin constituent, C16 {omega}-hydroxyalkanoic acid, suggesting mainly a root-derived SOM. In contrast to G1, G5a, and G5b, only C22 alkene was more abundant than the other chain lengths in the pyrolysates of the mineral soil profiles of G7, which may point to a different origin of SOM than that found in the other profiles.


Figure 6
View larger version (25K):
[in this window]
[in a new window]

  		Fig. 6. Gas chromatograms of pyrolysates from Profile G7: P = phenol, C = catechol, G = guaiacol (2-methoxyphenol), and S = syringol (2,6-dimethoxyphenol). Indicated side-chains are at the 4-position of P, G, or S: n-alkene (x); n-alkane (+); alkanoic acid ({diamondsuit}); alkanol (•); Cx indicates chain length.

 
Effect of Vegetation on Soil Organic Matter Transformations in Volcanic Ash Soils
In contrast to the organic geochemistry of volcanic ash soils previously studied (Naafs, 2004; Nierop et al., 2005), neither chitin markers nor the overall predominance of polysaccharides were observed in the soils examined here. In addition, the relative enrichment of aliphatic compounds, consisting of low-molecular-weight lipids, cutin/suberin, and cutan/suberan did not exist in the other volcanic ash soils. Nevertheless, small traces of polysaccharides were presen t in the deeper Bw horizons of our soils where lignin was not identified. Similar to the soils from Madeira and Azores, only in the top of the mineral soil, i.e., the Ah horizons, were minor signals from lignin noticed. The Bw horizons examined contained 3.9% (forest) and 5.6% (páramo) calculated allophane, respectively, which is "moderate", but deeper down the buried profiles contained up to 13.4% allophane (Tonneijck, unpublished data, 2006). Allophane has been suggested as a possible sorbent to stabilize polysaccharides in Podzols (Parfitt et al., 1999) and this could be true as well for Andosols. This mechanism may explain the better preservation of polysaccharides compared with lignin and their smaller accumulation with respect to the allophane-richer soils from Azores Islands (13.5–25.6% allophane; Nierop et al., 2005). Lipids have been found to accumulate with decreasing soil pH, which could clarify the clear presence of aliphatic compounds in the soils studied (pH in KCl 4.0–4.4) compared with those from the Azores (pH in KCl 5.0–5.7; Nierop et al., 2005). While the SOM composition of the soils from Madeira and Azores resembled each other, the mineralogical composition was rather different (Naafs, 2004; Nierop et al., 2005). Therefore, the large differences observed in SOM composition and, apparently, stabilization processes of the Guandera soils compared with those from Azores and Madeira are probably due to environmental factors such as temperature rather than the mineralogical composition.

In general, all profiles studied exhibited more of a root- than a leaf-derived SOM type in the mineral horizons, as evidenced by the distribution of the {omega}-hydroxyalkanoic acids (two to six times greater concentration of C22 vs. C16). The n-alkanes are 3 to 55 times more abundant in leaves than in roots, however, whereas extractable n-alkanols yield 4 to 260 times greater concentrations from leaves than from roots for the 19 dominant species responsible for the main biomass input in the Guandera area (Jansen et al., 2006). Therefore, we conclude that the n-alkanes and n-alkanols identified in the mineral soils in this study are derived to a larger extent from aboveground plant tissues rather than from belowground ones.

Not unexpectedly, the n-alkanes from both forest litters (G1 and G5a) were dominated by C29, while those from páramo litter by C31. In the mineral horizons of G5a, G5b, and G7, both alkanes were identified in rather similar concentrations, suggesting a mixture of both forest and páramo as SOM sources through time, possibly supported by bioturbation. By contrast, the n-alkanes of G1 were always dominated by C29 along with some C27 and much smaller amounts of C31, implying (mainly) a forest origin.

The n-alkanol distribution in the soils studied is less straightforwardly explained: in G1 the presence of relatively abundant C18, C20, C22, and C24 n-alkenes in pyrolysates and identified as methylated alkanols on THM is indicative of forest-derived organic matter (OM), assuming that G1 was always covered by forest. A prominent occurrence of C26 alkanol, typical of grasses, was not found in the páramo litter of G5b and G7. Instead of that, the mineral horizons of G7 were dominated by C22 alkanol, while G5a and G5b were also dominated by C22 alkanol in the top, but, in contrast to the top, and particularly in the Bw horizons, the alkanols identified were C18 to C24, with a remarkable large C18 contribution. This latter alkanol distribution resembles that of G1 more than G7, suggesting a strong forest-derived OM for the Bw horizons of G5a and G5b. By contrast, the Ah horizon of G5b, with a very dominant C22 alkanol, point to páramo-derived OM, while the O2 horizons of G5a had a prominent C18 alkanol. This would suggest that C22 n-alkanol is typical of páramo while C18 n-alkanol, together with C20, C22, and C24 alkanols is indicative of a forest. The páramo vegetation has indeed a prominent C22 contribution, mainly from roots, particularly Espeletia pycnophylla (Jansen et al., 2006). The C18 alkanol is not a typical biomarker for the species, although many plant roots contain this alkanol in combination with longer chain analogs (Jansen et al., 2006). As mentioned above, the distribution of cutin- and suberin-derived {omega}-hydroxyalkanoic acids strongly point to root-derived OM and only a minor leaf source. Hence, the observed alkanols could also be derived from suberin in which they occur as ester-bound moieties (Kolattukudy, 2001). As cutin and suberin are less species specific than, e.g., extractable lipids, it is difficult to assess whether the observed differences actually reflect forest- vs. páramo-derived SOM. Future assessment of extractable lipids together with traditional proxies, such as pollen analysis, could shed more light on this issue.

As mentioned above, our aim was to determine whether the vegetation-related differences in terms of the various types of humic acids was reflected in SOM composition. The most abundant macromolecule present in plants that could distinguish between forest and páramo, i.e., lignin, is transformed in such a way that there is not much left. Also, polysaccharides and other biopolymers or their transformation products were virtually absent from the mineral horizons. As a consequence, the differences in SOM formation between the melanic volcanic ash soils under páramo grassland and the fulvic ones from the forested sites were very small, and were only reflected in the distribution of the aliphatic components. Lipids are generally not found in extractable humus fractions including humic acids, however, but remain in the humin fractions (Nierop et al., 1999), and can therefore not explain the melanic index. Although the páramo-covered Andosols from our study sites possess melanic horizons and those under forest contain fulvic characteristics, we did not find the differences often observed by others that related to vegetation in general (Honna et al., 1988; Drijber and Lowe, 1990), or related to lignin and other phenolics in particular (Drijber and Lowe, 1991). In the case of these latter compounds, it has become clear that they are decomposed very quickly in the Andosols studied here and in those from the Azores and Madeira (Naafs, 2004; Nierop et al., 2005).


   CONCLUSIONS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
The Guanderean Andosols developed under forest are distinctly different from those formed under grassland with respect to the occurrence of an ectorganic layer and the melanic index. These vegetation-induced differences, however, were not reflected by differences in the SOM composition of a number of Andosols investigated. All mineral soil horizons studied were characterized by a high degree of aliphatic compounds as assessed by analytical pyrolysis techniques. Polysaccharides, and lignin in particular, were almost completely absent from the Ah and Bw horizons. Only a few small differences in the lipid composition reflect the diversity in vegetation, but they cannot account for the variations in the melanic index, as aliphatic compounds are not involved in the determination of this soil classification qualifier. The question arises as to whether the distinction between Melanic and Fulvic Andosols is useful, since it clearly does not describe SOM composition or transformations.


   ACKNOWLEDGMENTS
 
We gratefully acknowledge the Ecuadorian Ministerio del Ambiente for issuing the necessary permits for conducting our research in northern Ecuador. Jatun Sacha and Grupo Randi Randi are thanked for allowing us to undertake our study in Guandera Biological Station and for helping us in the field. Ecopar is acknowledged for facilitating our research in the form of office assistance. Furthermore, we wish to thank Leo Hoitinga, Leen de Lange, and Ton van Wijk for their assistance in the laboratory. This research was supported by the Earth and Life Science and Research Council (ALW) (K.G.J. Nierop) and the Netherlands Foundation for the Advancement of Tropical Research (WOTRO) (F.H. Tonneijck and B. Jansen; WAN 75-405) with financial aid from the Netherlands Organisation for Scientific Research (NWO) and generously sponsored by Fjällraven in the form of clothing and gear.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.

Received for publication September 12, 2006.


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




This article has been cited by other articles:


Home page
 Clay MineralsHome page
R. L. Parfitt
Allophane and imogolite: role in soil biogeochemical processes
Clay Minerals, March 1, 2009; 44(1): 135 - 155.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nierop, K. G. J.
Right arrow Articles by Verstraten, J. M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Nierop, K. G. J.
Right arrow Articles by Verstraten, J. M.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Nierop, K. G. J.
Right arrow Articles by Verstraten, J. M.
Related Collections
Right arrow Water Quality
Right arrow Dryland Soils
Right arrow Carbon Sequestration

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