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

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (28) Citing Articles via Google Scholar Google Scholar Articles by Solomon, D. Articles by Zech, W. Search for Related Content PubMed Articles by Solomon, D. Articles by Zech, W. GeoRef GeoRef Citation Agricola Articles by Solomon, D. Articles by Zech, W. Related Collections Soil Fertility and Productivity Soil Biochemistry Soil Chemistry

DIVISION S-2 - SOIL CHEMISTRY

Soil Organic Matter Composition in the Subhumid Ethiopian Highlands as Influenced by Deforestation and Agricultural Management

^a Institute of Soil Science and Soil Geography, Univ. of Bayreuth, Universitätsstr.30, D-95440 Bayreuth, Germany
^b Ethiopian Agricultural Research Organization, Debre Zeit Agricultural Research Center, P.O. Box 32, Debre Zeit, Ethiopia

* Corresponding author (ds278{at}cornell.edu)

	ABSTRACT

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

 
Physical fractionation, degradative wet-chemical analysis and liquid-state ¹³C nuclear magnetic resonance (NMR) spectroscopy were used to assess the impact of land use changes on the amount and structural composition of soil organic matter (SOM) in bulk soils and size separates in the subhumid highlands of southern Ethiopia. Soil samples (0–10 cm) were collected from natural forest, tea plantations, 25-yr cultivated fields at Wushwush (Paleudalf), Podocarpus-dominated natural forest, Cupressus plantations, and 30-yr cultivated fields at Munesa (Palehumults) sites. Forest clearing and continuous cultivation led to significant depletion (P < 0.05) of total soil organic C (SOC) (55% and 63%) and N (52% and 60%) in the surface soils, respectively. Compared with the cultivated fields, lower proportions of SOC (51 and 27%) and N (49 and 13%) were lost from the tea and Cupressus plantations, respectively. The largest depletion occurred from the labile SOM associated with the sand separates concurrent with higher oxidation states of lignin. However, substantial amounts of these organic substrates were also lost from the stable SOM fraction. Particularly, SOM, associated with the silt-size separates, decreased suggesting that the SOM in silt was quite susceptible to land use changes and represents a moderately labile SOM pool in the soils under study. Solution ¹³C NMR spectra revealed larger proportions of protonated and C- and O-substituted aryl-C in the silt than in clay-size separates. In contrast, O-alkyl-C structures were more prominent in the clay than in silt-size separates, coinciding with the lignin distribution obtained by wet-chemical analysis. Deforestation and subsequent agricultural management not only resulted in SOM depletion but also markedly altered the chemical composition of SOM in the subhumid highland ecosystems.

Abbreviations: (Ac/Al)_V, mass ratios of acid to aldehyde for the vanillyl • (Ac/Al)_S, mass ratios of acid to aldehyde for syringyl units • ANOVA, analysis of variance • CEC, cation-exchange capacity • (Fuc + Rha)/(Ara + Xyl), the ratio of fucose + rhamnose to arabinose + xylose • (Gal + Man)/(Ara + Xyl), the ratio of galactose + mannose to arabinose + xylose • GlcN/GalN, the ratio of glucosamine to galactosamine • GlcN/MurA, the ratio of glucosamine to muramic acid • LSD, least significant difference • NMR, nuclear magnetic resonance • SOC, soil organic C • SOM, soil organic matter • S/V, the ratio of syringyl to vanillyl • VSC, the sum of vanillyl, syringyl and cinnamyl units

	INTRODUCTION

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

 
SOIL ORGANIC MATTER is a physically and chemically heterogeneous mixture of organic compounds of plant, animal, and microbial origin, and has components at different stages of decomposition. The total mass of organic C stored in soils has been estimated at 1576 x 10¹⁵ g C, of which about one third is found in the tropics (Mahieu et al., 1999). Thus, the SOM in tropical soils represents an important pool of the terrestrial C reserves and its transformation is an important element in the global C cycle.

Soil organic matter plays a major role in soil fertility by affecting physical and chemical properties, and also controls soil microbial activity by serving as a source of mineralizable C and N. In mature and undisturbed tropical ecosystems, a balance exists between the organic C input and output of the soil because of mineralization and leaching of dissolved organic matter (Zech et al., 1997a). Changes in land use and soil management can have a marked effect on the SOM stock as a result of the interactions between detrital input and mineralization mediated by soil microorganisms (Tate, 1987). Several studies in the past have shown that deforestation and cultivation of native tropical soils often lead to accelerated SOM turnover, and thereby to depletion of nutrients (N, P, and S) present as part of complex organic polymers. Therefore, understanding processes that govern SOM dynamics in tropical soils is essential from the viewpoint of long-term sustainability of agriculture and influence on atmospheric CO₂ concentration and greenhouse effect.

Despite the importance of SOM in subhumid and humid tropical soils, very little information is available on the complex biological, chemical, and physical processes involved during decomposition and humification of organic substrates in physically separated-size fractions and on the factors influencing these processes. Information about the impact of land use changes on the amount and composition of organic substrates such as lignin, carbohydrates, and amino sugars in particle-size separates can eventually improve the knowledge of organic matter dynamics in tropical ecosystems.

Johansson (1986) indicated that a considerable proportion of the terrestrial C is present in the form of lignin and carbohydrates and that SOM is predominantly derived from their decomposition. Lignin decomposes slowly and represents a recalcitrant fraction in soil and litter (Guggenberger et al., 1994). During decomposition of lignin, intramolecular bonds between phenylpropanoid components of the lignin are severed and oxidized, and phenolic derivatives are released. As biodegradation progresses, the fractions derived from lignin in soil become increasingly acidic (carboxylic) (Haider, 1992; Kögel-Knabner, 1993). Thus, acid to aldehyde ratio of phenolic moieties can be used to estimate the extent of lignin biotransformation (Zech and Kögel-Knabner, 1994; Solomon et al., 2000a).

The primary C input into forest-derived soils is, however, holocellulose (cellulose and hemicellulose), which makes up approximately half of the dry weight of leaves (Haider, 1992). Plant-derived sugars, especially pentose polymers (arabinose and xylose) serve as the major source of energy and C for soil microorganisms. In turn, the microorganisms synthesize primarily hexose polymers (e.g., galactose, mannose, fucose, rhamnose) and release them into the soil (Oades, 1984; Murayama, 1984). Both the concentrations and compositions of saccharides can, therefore, be used to assess plant–microbe relationship in SOM dynamics. However, they cannot provide information about the microbial origin of organic substrates in the soil (Zhang et al., 1997). On the other hand, amino sugars are derived from different groups of soil microorganisms (Sowden and Ivarson, 1974; Parson, 1981; Zhang et al., 1998). Hence, the amount of hexosamines and muramic acid and their ratios (glucosamine to galactosamine [GlcN/GalN] and glucosamine to muramic acid [GlcN/MurA]) can serve as useful indicators for bacterial and fungal contribution to SOM (Zhang et al., 1998; Amelung et al., 1999a).

Specific wet-chemical degradation procedures and subsequent chromatographic separation and detection of individual components are suitable to elucidate the nature of primary and secondary resources of SOM. However, they usually fail to characterize the complex structure of the humified stable compounds of SOM (Kögel et al., 1988). In contrast, ¹³C NMR spectroscopy can provide unparalleled structural information on the humic substances from many different sources (Preston, 1987) and can supplement the results obtained by degradative wet-chemical methods (Kögel-Knabner, 1997).

Therefore, the objectives of the present study were (i) to characterize phenolic, carbohydrate and amino sugar signatures in bulk soils and size separates, and (ii) to investigate the influence of land use changes on the chemical and structural composition of SOM in the subhumid highland ecosystems of southern Ethiopia. The results from the wet-chemical methods were supplemented by solution ¹³C NMR spectroscopy.

	MATERIALS AND METHODS

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

 
Site Description
The present research was conducted at the southwestern highlands (Wushwush) and the southeastern Rift Valley escarpment (Munesa) of Ethiopia. Wushwush is located at 7° 19' N lat. and 36° 07' E long. The altitude of the area is 1900 m above sea level. Mean annual temperature is 18°C with an average annual precipitation of 1800 mm. Geologically, the area is associated with Jimma Volcanics with abundant rhyolites and trachybasalts. The soils of the area are classified as Plinthaquic Paleudalfs (Soil Survey Staff, 1999), with clayey texture and dark reddish brown color. The Wushwush natural forest is mainly composed of Olea africana (Mill.), Syzygium guineense (Guill. ex Perr. Gmel.), Cordia africana (Lam.), Croton macrostachys (Hochst. ex Rich.), and Ficus vasta (Forsk.) The Munesa site is located at 7° 35' N lat. and 38° 45' E long. Mean annual temperature is 19°C with an average annual precipitation of 1250 mm. Parent materials of the Munesa area are of volcanic origin, principally trachytes and basalts with ignimbrites and pumices at the rift valley floor. The escarpment extends from 2100 to 3200 m and the plain descends gradually to the Rift Valley lakes at 1600 m above sea level. The soils of the area are classified as Typic Palehumults (Soil Survey Staff, 1999), with clayey texture and very dark reddish brown color. The natural vegetation of the Munesa site ranges from the Arundinaria alpina (K. Schum.), Hagenia abyssinica (Bruce ex J.F. Gmel.), Croton macrostachys (Hochst. ex Rich.), Podocarpus falcatus (Thunb. ex Mirb.), Olea hochstetteri (Baker.) dominated forest on the escarpment to Acacia woodlands (Acacia tortilis (Forsk ex Hayne), Acacia abyssinica (Hochst. ex Benth.) and Acacia seyal (Delile) in the semi-arid lowlands.

The land-use systems studied at the Wushwush site were natural forest, tea plantations (35 yr old) and fields cultivated for 25 yr, while at the Munesa site Podocarpus natural forest, Cupressus plantations (25 yr old) and fields cultivated for 30 yr were investigated. There is no well established fertilization program in the tea plantations. Depending on fertilizer availability, up to 100 kg ha^-1 Urea or NPK (50:10:20) were applied every 3 to 5 yr as a side dressing. In the cultivated fields of both sites, maize (Zea mays L.) was grown without fertilizer inputs. However, during the intermittent dry periods, sorghum (Sorghum bicolor L. Moench) was grown at Munesa. The plowing depth both at the Wushwush and Munesa sites varies between 10 to 12 cm. Crop residues that remain on the cultivated fields are normally collected and used as animal feed at both sites.

After considering the depth of cultivation in the region, we used a core sampler (200-cm³ core volume at each subsite) and collected soil samples in three replicates from the upper 10 cm of the different land-use systems in April 1998. This helps to minimize differences, which may arise because of the dilution of SOM because of mixing of the surface soil with that of the subsoil through cultivation. We selected three representative fields from each land-use system, and from each site collected nine subsamples in a radial sampling scheme (Wilding, 1985). A composite sample was then prepared from the subsamples. The samples were air-dried and sieved (<2 mm) prior to fractionation and chemical analysis.

Particle-Size Fractionation
Particle-size fractionation was done on <2-mm material (bulk soil) according to Amelung et al. (1998). After removing visible root remnants, 30 g of soil was ultrasonically treated with an energy input of 60 J mL^-1 using a probe type sonicator (Branson Sonifier W-450) in a soil/water ratio of 1:5 (w/v). The coarse-sand size separates (250–2000 µm) were isolated by wet sieving. To completely disperse the remaining material in the <250-µm suspension, ultrasound was again applied with an energy input of 440 J mL^-1 in a soil/water ratio of 1:10 (w/v). The clay-size separates (<2 µm) were separated from the silt (2–20 µm) and fine sand (20–250 µm) size separates by repeated centrifugation. The silt-size separates were separated from the fine-sand size separates by wet sieving. Coarse- and fine-sand separates were combined and finally all size separates were dried at 40°C before grinding them for chemical analysis. The recovery of size separates after ultrasonic dispersion, wet sieving, and centrifugation ranged from 97 to 98% of the initial soil mass.

Chemical Analysis
Carbon and N contents of bulk soils and particle-size separates were analyzed by dry combustion with a C/H/N/S-analyzer (Elementar Vario EL Hanau, Germany). The pH-H₂O and pH-KCl were determined in 1:2.5 soil/water (w/v) suspension using a glass electrode. Cation-exchange capacity (CEC) was determined with 1 M NH₄OAc (pH = 7.0) according to Avery and Bascomb (1974). Dithionite-citrate-bicarbonate extractable Al and Fe (Al_d, Fe_d) were determined after double extractions at 70°C for 15 min as described by Mehra and Jackson (1960). Oxalate-extractable Al and Fe (Al_o, Fe_o) were determined using atomic absorption spectrometer (Varian AAS-400, Varien Techtron, Mulgrave, Victoria, Australia) after extraction for 2 h with 0.2 M ammonium oxalate at pH = 3 in darkness (Blume and Schwertmann, 1969). Selected soil physical and chemical characteristics are shown in Table 1.

Carbohydrate-Carbon Constituents
Determination of neutral sugars and uronic acids was conducted according to Amelung et al. (1996). Individual neutral sugars and acidic sugars were liberated from noncellulosic soil carbohydrates with 4 M trifluoroacetic acid (TFA) at 105°C for 4 h. The hydrolysates were filtered through glass fiber filters and dried using a rotary evaporator. The samples were purified by percolation through XAD-7 and Dowex 50 resin columns (Fluka chemic GmbH, Neu-Ulm, Germany). Myoinositol was added as the first internal standard before hydrolysis and 3-O-methylglucose was added as a second internal standard prior to derivatization to determine the recovery of the myo-inositol. The resulting hexoses, pentoses, and uronic acids were converted to O-methyloxime trimethylsilyl derivatives and separated by gas-liquid chromatography.

Amino Sugar-Carbon Constituents
Amino sugars, i.e., glucosamine, galactosamine, mannosamine, and muramic acid were determined according to the method of Zhang and Amelung (1996). Samples from bulk soils and particle-size fractions were hydrolyzed with 6 M HCl at 105°C for 8 h. Before hydrolysis, myoinositol was added as the first internal standard. The final organic phase was dried with N gas at 45°C and finally dissolved in 300 µL of ethyl acetate-hexane (1:1). Identification and quantification of amino sugars was conducted by capillary gas-liquid chromatography. As a recovery standard, 3-O-methylglucose was used.

Nuclear Magnetic Resonance Spectroscopy
For liquid-state ¹³C NMR measurements, the soils were extracted three times with 0.1 M NaOH and 0.4 M NaF at a ratio of soil/extraction solution of 1:5 (w/v). The sample size was adjusted to ensure a minimum yield of 200 mg of organic C in the extracts. The extraction procedure followed the outline of Schnitzer (1982), as modified by Sumann et al. (1998). Replacement of 0.1 M Na₄P₂O₇ by 0.1 M NaOH-0.4 M NaF mixture does not affect the ¹³C-NMR spectra of alkali-extracts, rather it improves extraction yield compared with extractions using only 0.1 M Na₄P₂O₇ (Sumann et al., 1998). The combined extracts were dialyzed (molecular weight cutoff, MWCO, 12000–14000 Da) and freeze-dried. Liquid-state ¹³C NMR spectra were obtained on a Bruker Avance DRX 500 NMR spectrometer (Bruker Instruments, Billerica, MA). About 150 mg of the freeze-dried material was dissolved in 3 mL of 0.5 M NaOD in a 10-mm NMR tube. Conditions for ¹³C NMR were: spectrometer frequency, 125.77 MHz; inverse-gated decoupling; acquisition time, 0.33 s; delay time, 1.67 s; and line-broadening factor, 100 Hz. This factor was used to enhance the characteristic broad signals of extracted organic matter. Chemical shifts were given in parts per million relative to the resonance of an external standard of TSP (3-[trimethylsilyl] propionic acid). The signal areas were calculated by electronic integration and signal assignments were made according to literature data (Kögel-Knabner, 1997).

Statistics
Statistical analysis of the data was carried out on the replicates by one-way ANOVA. If the main effects were significant at P < 0.05, a post hoc separation of means was done by univariate least significant difference (LSD) test. Statistical analysis was conducted with STATISTICA 5.1 for Windows (StatSoft Inc., Tulsa, OK).

	RESULTS AND DISCUSSION

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

 
Soil Organic Carbon and Nitrogen Contents
The content of total SOC in the surface soils (0–10 cm) varied from 38 to 85 g kg^-1 soil at Wushwush and 38 to 103 g kg^-1 soil at the Munesa sites, whereas the amount of total soil N ranged from 3.7 to 7.8 g kg^-1 soil and 3.3 to 8.2 g kg^-1 soil at the two sites, respectively (Table 2). The amount of SOM in the surface soils of these subhumid tropical highland ecosystems is larger than the amounts reported for lowland humid tropical forest soils (Veldkamp, 1994; Van Noordwijk et al., 1997). However, they compare positively with the results of Möller et al. (2000) for humid tropical mountain ecosystems of northern Thailand. The large amount of SOM in the surface soils of these subhumid tropical highland forest ecosystems could possibly be attributed to the relatively slower decomposition of SOM because of the cool mountainous climate compared with the humid tropical lowland ecosystems with higher annual precipitation and temperature and to differences in litter quality.

Examination of physically fractionated-size separates showed that only 4 to 13% of the total SOC and 2 to 9% of the total N were found in the sand-size separates. The SOM associated with the silt contained 29 to 52% of SOC and 25 to 40% of soil N, while most of the SOM was bound to the clay-size separates (40 to 63% of the total SOC and 56 to 71% of the total soil N) (Fig. 1) . Besides the reactivity and specific charge characteristics of clay minerals and associated oxides and hydroxides, the higher active surface area controls the enrichment of SOM in the finer-size separates (Zech et al., 1997a). The C/N ratios of the size separates of both soils decreased with decreasing particle-size from 22.4 in the sand to 8.8 in the clay-size separates, reflecting a progressive SOM humification with decreasing diameter (Fig. 1). This was also revealed by the electron microscopy of organic matter associated with representative size separates of the soils under investigation. The scanning electron microscope micrographs showed that the SOM associated with the sand-size separates was composed of undecomposed and macromorphologically identifiable particulate plant residues (coarse sand) as well as partially decomposed organic debris (fine sand). The SOM associated with the silt- and clay-size separates, however, was composed of completely decomposed organic structures and microbial residues (Fig. 2) .

View larger version (34K): [in this window] [in a new window]   Fig. 1. Soil organic carbon (SOC) and total N contents and C/N ratios in size separates of the different land-use systems at the Wushwush and Munesa sites. Letters above the bars indicate significant differences between mean values of the same size separates of the different land use systems at P < 0.05; N. forest, natural forest; T. plantation, Tea plantation; C. plantation, Cupressus plantation.

View larger version (101K): [in this window] [in a new window]   Fig. 2. Scanning electron microscope micrographs (acceleration voltage = 15 kV) of organic matter associated with (a) coarse sand (2000–250 µm), (b) fine sand (250–20 µm), (c) silt (20–2 µm) and (b) clay (<2 µm) size separates of soils from the natural forest of the Munesa site.

Phenolic Signature
The concentration of lignin-derived C (g VSC kg^-1 SOC) and patterns of the total phenolic-C moieties in the surface soils (0–10 cm) differed significantly (P < 0.05) under the different land use systems. As shown in Table 2, soils of the natural forests at the Wushwush and Munesa sites had the largest concentrations of lignin-derived phenols followed by the plantations and cultivated fields. Alterations of the chemical structure of the vanillyl and syringyl moieties during lignin decomposition often results in increased acid to aldehyde ratios of vanillyl [(Ac/Al)_V] and syringyl [(Ac/Al)_S] units and decreased S/V quotient (Kögel et al., 1988). In the present study, the (Ac/Al)_V,S ratios generally decreased in the order: cultivation > plantation > natural forest. The S/V ratio in the bulk soils, however, increased in the order: cultivation < tea plantation < natural forest at the Wushwush and cultivation < natural forest < Cupressus plantation at the Munesa sites. These results indicated that the VSC-lignin in the natural forest and plantations was less oxidized than the lignin in the cultivated fields. The lowest S/V ratio found in bulk soil and size separates of the Cupressus plantations may be attributed to the inherently lower proportion of syringyl units in the gymnosperm lignin (Sarkanen and Ludwig, 1971).

The highest average proportion of VSC-lignin in particle-size separates was found in the particulate SOM associated with the sand (53 and 57%) followed by the SOM in the silt (34 and 33%) and clay (13 and 10%) size separates at the Wushwush and Munesa sites, respectively (Fig. 3) . The (Ac/Al)_V,S ratios increased with decreasing particle-size at both sites, whereas the opposite was true for the ratio of S/V units, suggesting a progressive lignin degradation and selective loss of syringyl units in the order: sand < silt < clay, independent of the land use systems. The relative amount and patterns of phenolic-C moieties in these soils are comparable with those obtained in temperate (Guggenberger et al., 1994; Glaser, et al., 2000) and tropical soils (Zech et al., 1997b; Guggenberger et al., 1999; Solomon et al., 2000a).

View larger version (39K): [in this window] [in a new window]   Fig. 3. Average concentrations of CuO-oxidation products (VSC) and ratios of lignin-derived phenols in size separates of the different land use system at the Wushwush and Munesa sites. VSC, vanillyl+syringyl+cinnamyl; (Ac/Al)V,S, acid to aldehyde ratio of vanillyl and syringyl; S/V, syringyl to vanillyl ratio; letters above the bars indicate significant differences between mean values of the same size separates of the different land use systems at P < 0.05; bars without letters are not significant; N. forest, natural forest; T. plantation, Tea plantation; C. plantation, Cupressus plantation.

Carbohydrate Signature
The total concentrations of noncellulosic carbohydrates (g kg^-1 SOC) significantly decreased in the order: natural forest > plantation > cultivation at the Wushwush and Munesa sites (Table 2). Microorganisms synthesize little if any pentoses, while xylose and arabinose are ubiquitous constituents of plant cells (Oades, 1984). Therefore, individual quotients of sugar monomers can be used to estimate the relative contribution of microbial-derived saccharides to the soil carbohydrate composition. Murayama (1984) and Oades (1984) used the ratios of hexoses to pentoses [galactose (Gal) + mannose (Man)]/[arabinose (Ara) + xylose (Xyl)] and deoxyhexoses to pentoses [fucose (Fuc) + rhamnose (Rha)]/[arabinose (Ara) + xylose (Xyl)] to estimate the contribution of plant-derived and microbially synthesized sugars to the soil monosaccharide composition. The (Gal + Man)/(Ara + Xyl) and (Fuc + Rha)/(Ara + Xyl) ratios in the bulk soils of the natural forests at the Wushwush and Munesa sites were higher than those from corresponding plantations and cultivated fields (Table 2). These results indicated the presence of higher proportions of microbial-derived carbohydrates in SOM of the natural forests compared with the SOM in the plantations and cultivated fields. Higher amounts of microbial-derived saccharides in forest than arable soils were also reported by Guggenberger et al. (1994) in Germany, Guggenberger et al. (1999) in Costa Rica, and Solomon et al. (2000a) in Tanzania. However, these authors reported higher concentration of TFA-hydrolyzable saccharides in the cultivated than in the forest soils. The lower carbohydrate concentrations in the arable soils of our sites may be attributed to the lower biomass input as a result of competitive use of crop residues (e.g., for animal feed, construction material, or cooking) coupled with the rapid depletion of monosaccharides because of accelerated mineralization of the original (forest-derived) SOM following clear-cutting and long-term continuous cultivation under the subhumid tropical highland environment. This suggestion is supported by our natural ¹³C abundance studies, where only about 25 and 55% of the SOC in cultivated soils of the Wushwush and Munesa sites was derived from maize (C₄) crop residues, while the remaining SOC at both sites was of forest (C₃) origin (Solomon et al., 2000, unpublished data).

The highest percentage of total TFA-hydrolyzable soil saccharides was found in the clay-bound SOM (50 and 49%) at the Wushwush and Munesa sites, respectively (Fig. 4) . The carbohydrates in the silt-size separates comprised on average 30 and 32%, whereas carbohydrates associated with the sand-size separates contained 20 and 19% of the total soil monosaccharides at the two sites, respectively. The ratios of hexoses to pentoses and deoxyhexoses to pentoses in physically separated-size separates increased in the order: sand < silt < clay. These results indicated greater contribution of microbial-derived sugars (galactose, mannose, fucose, and rhamnose) to saccharide spectrum of the clay, while plant-derived saccharides (arabinose and xylose) dominated the monosaccharide composition of the sand-size separates. Our results are inline with the results of Zech et al. (1997b) for soils of Argentina, Amelung et al. (1999c) for soils of North American prairie, Guggenberger et al. (1999) for soils of Costa Rica, and Glaser, et al. (2000) for the mountain soils of Khyrgyzia.

View larger version (43K): [in this window] [in a new window]   Fig. 4. Average concentrations of noncellulosic carbohydrates and ratios of sugar monomers in size separates of the different land-use systems at the Wushwush and Munesa sites. (Gal + Man)/(Ara + Xyl), galactose + mannose to arabinose + xylose; (Fuc + Rha/Ara + Xyl), fucose + rhamnose to arabinose + xylose; Letters above the bars indicate significant differences between mean values of the same size separates of the different land use systems at P < 0.05; N. forest, natural forest; T. plantation, Tea plantation; C. plantation, Cupressus plantation.

Amino Sugar Signature
Significantly higher (P < 0.05) total amino sugar concentrations (g kg^-1 SOC) were found in bulk soils of the natural forests compared with the plantations and cultivated fields (Table 2). The apparent depletion of amino sugars may be attributed to accelerated mineralization of hexosamines and muramic acid upon forest clearing and subsequent low-input agricultural management under the subhumid tropical highland environment. Glucosamine is an important constituent of the fungal cell wall (Chantigny et al., 1997; Zhang et al., 1998, 1999), while in terrestrial ecosystems muramic acid uniquely originates from bacteria (Kenne and Lindburg, 1983; Amelung et al., 1999a). Sowden and Ivarson (1974) demonstrated that little if any galactosamine is synthesized by fungi during fungi-inoculated incubation experiments. Hence, the ratios of both glucosamine to galactosamine (GlcN/GalN) and glucosamine to muramic acid (GlcN/MurA) were used to assess the fate of bacterial- and fungal-derived SOM in these tropical soils. According to Table 2, the patterns of amino sugars in the soils under investigation were considerably influenced by land use changes. The ratios of both GlcN/GalN and GlcN/MurA increased in the order: natural forest < plantation < cultivation at the Wushwush and Munesa sites. The higher GlcN/GalN and GlcN/MurA ratios in the continuously cultivated fields and plantations compared with the corresponding natural forests indicated higher depletions of galactosamine and muramic acid than glucosamine because of land-use changes. Rapid losses of galactosamine and muramic acid compared with glucosamine because of agricultural management of native soils were also reported by Zhang et al. (1997)(1999) in North America and by Solomon et al. (2000b) in northern Tanzania. The results of the present investigation further supports the suggestion by these authors that bacterial-derived amino sugars (galactosamine and muramic acid) are less stable compared with fungal-derived glucosamine in arable soils. The relative stability of glucosamine compared with galactosamine and muramic acid may be ascribed to the fact that it is the constituent of the recalcitrant glomaline, a glycoprotein produced in the soil by arbuscular mycorrhizal fungi (Wright and Upaghyaya, 1996; Zhang et al., 1998). The higher concentration of glucosamine might be partly explained by the enhanced growth of fungi in a slightly more acidic environment under the investigated arable soils than under the native forest vegetation.

The distribution of amino sugars in size separates of the soils under study was similar to that observed for the total SOC. Accordingly, the SOM in the clay (56 and 60%) was enriched in amino sugars compared with the silt (32 and 29%) and sand (12 and 11%) size separates (Fig. 5) . These results show that microbial contribution to the total organic matter accumulation increased with decreasing particle-size. Amino sugar ratios were used to evaluate the pattern of bacterial- and fungal-derived amino sugars among particle-size separates of the studied tropical soils. According to Fig. 5, the GlcN/GalN ratio increased in the order: sand < silt < clay, whereas the GlcN/MurA ratio decreased with decreasing particle-size, the highest being in the sand-size separates. The amounts and patterns of amino sugars in these soils are consistent with the results of Zhang et al. (1997)( 1998), Amelung et al. (1999a) for North American soils, and with our results for soils from northern Tanzania (Solomon et al., 2000b). The relative enrichment of muramic acid in the clay compared with the silt- and sand-size separates may be attributed to the increased stabilization of this bacterial-derived amino sugar with decreasing size separates. However, galactosamine was enriched in the sand compared with the silt- and clay-size separates, which might suggest that this amino sugar is most likely synthesized at an early stage of SOM decomposition (Zhang et al., 1998).

View larger version (38K): [in this window] [in a new window]   Fig. 5. Total amino sugar concentrations and ratios of glucosamine to galactosamine (GlcNl + GalN) and glucosamine to muramic acid (GluN + MurA) in size separates of the different land use systems at the Wushwush and Munesa sites. Letters above the bars indicate significant differences between mean values of the same size separates of the different land use systems at P < 0.05; N. forest, natural forest; T. plantation, Tea plantation; C. plantation, Cupressus plantation.

Carbon-13 Nuclear Magnetic Resonance Spectroscopy
Liquid-state ¹³C NMR spectra of alkali-extracts in bulk soils of the Wushwush and Munesa sites are shown in Fig. 6 and 7 . The spectra revealed multiple peaks between = 0 to 50 ppm, thus indicating a highly heterogeneous composition of alkyl-C species. The resonances between = 50 and 60 ppm may be assigned to methoxyl C in lignin as well as to C in amino acids (Baldock et al., 1990). The O-alkyl-C region ( = 60 to 110 ppm) was dominated by carbohydrates which generally produce signals around = 61 ppm (C₆), = 70 to 90 ppm (C₂ to C₅) and at = 102 ppm (anomeric C). Broad signals between 110 to 145 ppm were because of protonated and C-substituted aryl-C, while resonances between = 145 to 160 ppm arose from O-substituted aryl (phenolic) C and are indicative of lignin-derived compounds (Kögel-Knabner, 1997). The carbonyl-C ( = 160 to 200 ppm) was dominated by resonances because of carboxylic-C around = 175 ppm. The NaOH-NaF solution extracted 26 to 32% of the total SOC in bulk soils and particle-size separates of these tropical soils.

View larger version (18K): [in this window] [in a new window]   Fig. 6. Solution 13C NMR spectra of alkali-extractable organic matter in bulk soil and size separates of the different land use systems at the Wushwush site.

View larger version (18K): [in this window] [in a new window]   Fig. 7. Solution 13C NMR spectra of alkali-extractable organic matter in bulk soil and size separates of the different land use systems at the Munesa site.

The stacked solution ¹³C NMR spectra (Fig. 6 and 7) and relative proportions of different functional groups (Table 3 and 4) in particle-size separates indicated the presence of relatively higher proportions of protonated and C-substituted aryl-C and O-substituted aryl-C in the silt than in the clay-size separates of both sites. Thus, SOM associated with the silt-size separates showed higher aromaticity (0.38 and 0.37) compared with SOM in the clay-size separates (0.27 and 0.21) of the Wushwush and Munesa soils, respectively. In addition, the average ratio of O-aryl-C to aromatic-C in the silt appears to be higher than that in the clay-size separates of the two sites, respectively. These results are inline with results previously reported by Randall et al. (1995) and Guggenberger et al. (1995). According to Guggenberger et al. (1995), the decline of O-aryl-C to aromatic-C ratio with progressive lignin degradation from coarser- to finer-size separates reflects decreased input of lignin to the total aromatic C region in the course of SOM humification. Liquid-state ¹³C NMR spectra of the clay-size separates were enriched (37 and 36%) in O-alkyl-C structures compared with the silt (26 and 25%) size separates at the Wushwush and Munesa sites, respectively. This supports our results from the wet-chemical analysis, where the highest relative enrichment of microbial-derived saccharides (hexoses and deoxyhexoses) was obtained in the clay-size separates. Our results also concur positively with the results of Baldock et al. (1990), who demonstrated using incubation experiments of a uniformly labeled ¹³C glucose that utilization of glucose by soil microorganisms resulted in the synthesis and accumulation of predominantly O-alkyl-C, alkyl-C, and carboxylic-C in clay-size separates. However, the differences in the average proportions of alkyl-C and carbonyl-C structures in the silt- and clay-size separates of our sites were small and variable.

Comparisons of signal intensities and relative distribution of functional groups in alkali-extractable SOM associated with the silt- and clay-size separates exhibited only minor differences among the different land use systems (Fig. 6 and 7 and Table 3 and 4). The silt-size separates of natural forests showed slightly higher ratios of O-aryl-C/aromatic-C than those from the cultivated fields, suggesting a relative enrichment of refractory, C-substituted aryl-C in the silt-size separates of the cultivated soils. In addition, the solution ¹³C NMR spectra of the alkali extracts from the size separates of the cultivated fields and tea plantations exhibited distinct signals at 135 and 177 ppm compared with the corresponding natural forests. The presence of intense signals in the above mentioned regions have also been reported by Möller et al. (2000) in alkali extracts of forest soils from northern Thailand. These authors used 10 Hz line-broadening and mellitic acid as a standard and found out that these sharp peaks originate from mellitic (benzenehexacarboxylic) acid. Glaser et al. (1998) pointed out that mellitic acid is not produced by oxidation of organic materials obtained by common humification reactions but it rather is the product of the oxidation of charred plant materials. Spectra recorded with a line-broadening of 10 Hz from the alkali-extracts of size separates of the soils under investigation also indicated the presence of small but distinct mellitic acid signals in silt and clay separates of the cultivated fields and tea plantations (spectra not presented). These results may suggest the contribution of pyrogenic-C to the total SOC of the cultivated and plantation sites.

	CONCLUSIONS

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

 
Physical fractionation of the soils under investigation into sand-, silt-, and clay-size separates, coupled with wet-chemical analysis indicated the presence of three distinct SOM pools which are markedly different in the amount, composition and turnover of the organic matter associated with them. Thus, this study demonstrated the value of combining physical fractionation with degradative wet-chemical analysis in evaluating land use effects on the SOM properties of tropical soils. It also highlighted the importance and limitations of integrating solution ¹³C NMR spectroscopy to such studies.

Forest clearing and subsequent low-input agricultural management led to significant depletion of total SOC and N in bulk soils and size separates of these tropical soils. Compared with the continuously cultivated fields, lower depletion of SOM occurred from plantations. The phenolic-C signature indicated that VSC-lignin in the continuously cultivated soils was at a more advanced stage of oxidation than the lignin in the plantations and natural forests. The carbohydrate and amino sugar signatures indicated pronounced depletions of both the TFA-hydrolysable monosaccharides and hexosamines and muramic acid following land use changes. The largest depletion occurred from the labile SOM associated with sand. However, substantial amounts of these organic substrates were also lost from the stable SOM fractions. Particularly SOM associated with the silt was quite susceptible to land use changes, and that at least in the soils under study represents a moderately labile SOM pool. The results of liquid-state ¹³C NMR spectroscopy showed pronounced difference in C-functional groups among the silt- and clay-size separates. On the contrary, differences between land-use systems appeared to be very small. Protonated and C-substituted aryl-C and O-substituted aryl-C were more prominent in the silt than in the clay separates. In contrast, O-alkyl-C groups were dominant C structures in the clay compared with the silt-size separates, corroborating with results of the degradative wet-chemical analysis of these tropical soils.

Based on our results, it is possible to conclude that deforestation of remnant natural forests and subsequent agricultural management not only influenced the amount but also the chemical composition of SOM in these tropical soils. These have a negative impact on soil fertility, and thereby endanger sustainability of agriculture in the fragile ecosystems of southern Ethiopian highlands. Therefore, there is a need to develop appropriate soil and agronomic practices to overcome the current depletion of SOM and to ensure sustained agricultural production without environmental degradation in the region.

	ACKNOWLEDGMENTS

 
The authors thank the German Academic Exchange Service (DAAD) for providing fellowship for D. Solomon. The authors are also grateful to Dr. L. Haumaier for recording the ¹³C NMR spectra and to the Wushwush Tea Plantation and the Munesa Forest Enterprise, for allowing us to take samples from their sites.

	NOTES

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

 
D. Solomon, current address: College of Agriculture and Life Sciences, Dep. of Crop and Soil Sciences, Cornell Univ., Bradfield and Emerson Halls, Ithaca, NY 14853.

Received for publication December 18, 2000.

	REFERENCES

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

 

• Amelung, W., Z. Xudong, and G. Guggenberger. 1996. Determination of neutral and acidic sugars in soils by capillary gas-liquid chromatography after trifluoroacetic acid hydrolysis. Soil Biol. Biochem. 28:1631–1639.
• Amelung, W., W. Zech, X. Zhang, R.F. Follett, H. Tiessen, E. Knox, and K.-W. Flach. 1998. Carbon, nitrogen, and sulfur pools in particle-size fractions as influenced by climate. Soil Sci. Soc. Am. J. 62: 172–181.[Abstract/Free Full Text]
• Amelung, W., X. Zhang, K.-W. Flach, and W. Zech. 1999a. Amino sugars in native grassland soils along a climosequence in North America. Soil Sci. Soc. Am. J. 63:86–92.[Abstract/Free Full Text]
• Amelung, W., X. Zhang, K.-W. Flach, and W. Zech. 1999b. Lignin in particle-size fractions of native grassland soils as influenced by climate. Soil Sci. Soc. Am. J. 63:1222–1228.[Abstract/Free Full Text]
• Amelung, W., X. Zhang, K.-W. Flach, and W. Zech. 1999c. Neutral and acidic sugars in particle-size fractions as influenced by climate. Soil Sci. Soc. Am. J. 63:865–873.[Abstract/Free Full Text]
• Avery, B.W., and C.L. Bascomb. 1974. Soil survey laboratory methods. p. 19–25. In Soil survey laboratory methods. Tech. monogr., No. 6. Rothamsted Exp. Stn., Harpenden, UK.
• Baldock, J.A., J.M. Oades, A.M. Vassallo, and M.A. Wilson. 1990. Solid-state CP/MAS ¹³C N.M.R. analysis of particle size and density fractions of a soil incubated with uniformly labelled ¹³C-glucose. Aust. J. Soil Res. 28:193–212.
• Blume, H.P., and U. Schwertmann. 1969. Genetic evaluation of profile distribution of aluminum, iron, and manganese oxides. Soil Sci. Soc. Am. Proc. 33:438–444.
• Chantigny, M.H., D.A. Angers, D. Prevost, L.P. Vezina, and F.P. Chalifour. 1997. Soil aggregation and fungal and bacterial biomass under annual and perennial cropping systems. Soil Sci. Soc. Am. J. 61:262–267.[Abstract/Free Full Text]
• Glaser, B., L. Haumaier, G. Guggenberger, and W. Zech. 1998. Black carbon in soils: The use of benzenecarboxylic acids as specific markers. Org. Geochem. 29:811–819.
• Glaser, B., M.-B. Turrión, D. Solomon, A. Ni, and W. Zech. 2000. Soil organic matter quantity and quality in mountain soils of the Alay Range, Khyrgyzia, affected by land use change. Biol. Fertil. Soils 31:407–413.
• Guggenberger, G., B.T. Christensen, and W. Zech. 1994. Land-use effects on the composition of organic matter in particle-size separates of soils: I. Lignin and carbohydrate signature. Eur. J. Soil Sci. 45:449–458.
• Guggenberger, G., W. Zech, L. Haumaier, and B.T. Christensen. 1995. Land-use effects on the composition of organic matter in particle-size separates of soils: II. CPMAS and solution ¹³C NMR analysis. Eur. J. Soil Sci. 46:147–158.
• Guggenberger, G., and W. Zech. 1999. Soil organic matter composition under primary forest, pasture and secondary forest succession, Región Huetar Norte, Costa Rica. For. Ecol. Manage. 124:93–104.
• Haider, K. 1992. Problems related to the humification processes in soils of temperate climates. p. 55–94. In J.M. Bollag and G. Stotzky (ed.) Soil biochemistry. Vol. 7. Marcel Dekker, New York.
• Hedges, J.I., and J.R. Ertel. 1982. Characterization of lignin by gas capillary chromatography of cupric oxide oxidation products. Anal. Chem. 54:174–178.
• Johansson, M.B. 1986. The chemical composition of needle and leaf litter from Scots pine, Norway spruce and White birch in Scandinavian forests. Forestry 68:49–61.
• Kenne, L.K., and B. Lindburg. 1983. Bacterial polysaccharides. p. 287–365. In G.O. Aspinall (ed.) The polysaccharides. Academic Press, New York.
• Kögel, I., R. Hempfling, W. Zech, P.G. Hatcher, and H.R. Schulten. 1988. Chemical composition of the organic matter in forest soils: 1. Forest litter. Soil Sci. 146:124–136.
• Kögel-Knabner, I. 1993. Biodegradation and humification processes in forest soils. Soil Biochem. 8:101–135.
• Kögel-Knabner, I. 1997. ¹³C and ¹⁵N-NMR spectroscopy as a tool in soil organic matter studies. Geoderma 80:243–270.
• Mahieu, N., D.S. Powlson, and E.W. Randall. 1999. Statistical analysis of the published carbon-13 CPMAS NMR spectra of soil organic matter. Soil Sci. Soc. Am. J. 63:307–319.[Abstract/Free Full Text]
• Mehra, O.P., and M.L. Jackson. 1960. Iron oxide removal from soils and clays by a dithionite-citrate system buffered with sodium bicarbonate. Clays Clay Min. 7:317–327.
• Möller, A., K. Kaiser, W. Amelung, C. Niamskul, S. Udomsri, M. Puthawong, L. Haumaier, and W. Zech. 2000. Forms of organic C and P extracted from tropical soils as assessed by liquid-state ¹³C- and ³¹P-NMR spectroscopy. Aust. J. Soil Res. 38:1017–1035.
• Murayama, S. 1984. Changes in monosaccaride composition during the decomposition of straws under field conditions. Soil Sci. Plant Nutr. 30:367–381.
• Oades, J.M. 1984. The retention of organic matter in soils. Biogeochemistry 5:35–70.
• Parson, J.W. 1981. Chemistry and distribution of amino sugars. p. 197–227. In E.A. Paul and J.N. Ladd (ed.) Soil biochemistry. Vol. 5. Marcel Dekker, New York.
• Preston, C.M. 1987. Review of solution NMR of humic substances. p. 3–32. In R.L. Wershaw and M.A. Mikita (ed.) NMR of humic substances and coal. Lewis Publishers, Boca Raton, FL.
• Randall, E.W., N. Mahieu, D.S. Powlson, and B.T. Christensen. 1995. Fertilization effects on organic matter in physically fractionated soils as studied by ¹³C NMR: Results from two long-term field experiments. Eur. J. Soil Sci. 46:557–565.
• Sarkanen, K.V., and C.H. Ludwig. 1971. Lignins. Wiley-Interscience, New York.
• Schnitzer, M. 1982. Organic matter characterization. p. 581–594. In A.L. Page et al. (ed.) Methods of soil analysis. Part 2. 2nd ed. ASA, Madison, WI.
• Skjemstad, O.J., R.C. Dalal, and P.F. Barron. 1986. Spectroscopic investigations of cultivation effects on organic matter of Vertisols. Soil Sci. Soc. Am. J. 50:354–359.[Abstract/Free Full Text]
• Soil Survey Staff. 1999. Keys to soil taxonomy, 8th ed. Pocahontas Press, Blacksburg, VA.
• Solomon, D., J. Lehmann, and W. Zech. 2000a. Land use effects on soil organic matter properties of chromic Luvisols in the semiarid tropics: Carbon, nitrogen, lignin and carbohydrates. Agric. Eco. Environ. 78:203–213.
• Solomon, D., J. Lehmann, and W. Zech. 2000b. Land use effects on amino sugar signature of chromic Luvisols in the semi-arid part northern Tanzania. Biol. Fertil. Soils 33:33–40.
• Sowden, F.J., and K.C. Ivarson. 1974. Effects of temperature on changes in the nitrogen constituents of mixed forest litters during decomposition after inoculation with various microbial cultures. Can. J. Soil Sci. 54:387–394.
• Sumann, M., W. Amelung, L. Haumaier, and W. Zech. 1998. Climatic effects on soil organic phosphorus in the North American Great Plains identified by phosphorus-31 nuclear magnetic resonance. Soil Sci. Soc. Am. J. 62:1580–1586.[Abstract/Free Full Text]
• Tanaka, H., M. Homma, K.-I. Sakagami, and R. Hamada. 1992. Acid hydrolysis kinetics of soil carbohydrates. Sci. Tot. Environ. 117/118:145–153.
• Tate, R.L. 1987. Soil organic matter: Biological and ecological effects. John Wiley, New York.
• Van Noordwijk, M., C. Cerri, P.L. Woomer, K. Nugroho, and M. Bernoux. 1997. Soil carbon dynamics in the humid tropical forest zone. Geoderma 79:187–225.
• Veldkamp, E. 1994. Organic carbon turnover in three tropical soils under pasture after deforestation. Soil Sci. Soc. Am. J. 58:175–180.[Abstract/Free Full Text]
• Wilding, L.G. 1985. Spatial variability: Its documentation, accommodation and implication to soil survey. p. 166–187. In D.R. Nielsen and J. Bouma (ed.) Soil spatial variability. PUDOC, Wageningen, The Netherlands.
• Wright, S.F., and A. Upaghyaya. 1996. Extraction of an abundant and unusual protein from soil and comparison with hyphal protein of arbuscular mycorrhizal fungi. Soil Sci. 161:575–586.
• Zech, W., and I. Kögel-Knabner. 1994. Patterns and regulation of organic matter transformation in soils: Litter decomposition and humification. p. 303–334. In E.-D. Schulze (ed.) Flux control in biological systems: From enzymes to populations and ecosystems. Academic Press, San Diego, CA.
• Zech, W., N. Senesi, G. Guggenberger, K. Kaizer, J. Lehmann, T.M. Miano, A. Miltner, and G. Schroth. 1997a. Factors controlling humification and mineralization of soil organic matter in the tropics. Geoderma. 79:117–161.[ISI]
• Zech, W., G. Guggenberger, P. Zalba, and N. Peinemann. 1997b. Soil organic matter transformation in Argentinean Hapludolls. Z. Pflanzenernähr. Bodenk. 160:563–571.
• Zhang, X., and W. Amelung. 1996. Gas chromatographic determination of muramic acid, glucosamine, galactosamine and mannosamine in soils. Soil Biol. Biochem. 28:1201–1206.
• Zhang, X., W. Amelung, Y. Yuan, and W. Zech. 1997. Amino sugars in soils of north American cultivated prairie. Z. Pflanzenernähr. Bodenk. 160:533–538.
• Zhang, X., W. Amelung, Y. Yuan, and W. Zech. 1998. Amino sugar signatures of particle size fractions in soils of the native prairie as affected by climate. Soil Sci. 163:220–229.
• Zhang, X., W. Amelung, Y. Yuan, and W. Zech. 1999. Land-use effects on amino sugars in particle size fractions of an Argiudoll. Appl. Soil Ecol. 11:271–275.

This article has been cited by other articles:

				D. Solomon, J. Lehmann, J. Kinyangi, B. Liang, and T. Schafer Carbon K-Edge NEXAFS and FTIR-ATR Spectroscopic Investigation of Organic Carbon Speciation in Soils Soil Sci. Soc. Am. J., January 1, 2005; 69(1): 107 - 119. [Abstract] [Full Text] [PDF]

				D. Solomon, J. Lehmann, and C. E. Martinez Sulfur K-edge XANES Spectroscopy as a Tool for Understanding Sulfur Dynamics in Soil Organic Matter Soil Sci. Soc. Am. J., November 1, 2003; 67(6): 1721 - 1731. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited