Soil organic matter loss and fertility degradation under different agricultural
land uses in sandy soils of Northeast Thailand and the use of organic materials
of different qualities as a possible restoration measure

Vityakon, P.¹

Keywords: agricultural land uses, heavy fraction, humic substance, light fraction, microbial biomass, nitrogen mineralization, qualities of organic residues, soil organic matter pools

Introduction

Soil organic matter (SOM) can be used as an indicator of land degradation arising from agricultural land use. Different pools of SOM, notably labile and stable pools, as well as total SOM can give different kinds of useful information regarding soil fertility as well as soil degradation. Total SOM or total carbon (measured conventionally by wet or dry combustion) is not sensitive to impact of various ecological and management factors compared to the labile pool of SOM, i.e. microbial biomass and light fraction of particulate SOM.

Although SOM in sandy soils of Northeast Thailand is intrinsically low, the conversion of natural forest to agricultural land use that occurred in this region in the course of the past hundred years has resulted in significant further lowering of the SOM content. Some of the factors bringing about lowering of SOM in an agro-ecosystem as compared to the natural forest ecosystem include decreased amount of OM recycling, frequent and heavy soil disturbance and reduced biodiversity. To restore the SOM, input of organic materials is required with an emphasis on different qualities of organic materials to ensure the improvement of different pools of SOM.

The objectives of this paper are to review and synthesize results of several past studies in Northeast Thailand on (1) assessment of degradation of sandy soils under agricultural land use compared with original forest land use employing different pools of SOM as a soil degradation indicator, and (2) the use of organic materials of different qualities to restore SOM and fertility (nitrogen) in degraded soils.

Degradation of SOM and soil quality of sandy soils under agricultural land use

Soils of the undulating terrain of Northeast Thailand generally have coarse texture and low organic matter leading to their inherent low fertility. Different systems of land use and soil management have brought about soil degradation manifested by the decline in SOM and fertility. Vityakon (1991) compared the soil organic carbon (SOC) content of topsoils of Warin sandy loam (Oxic Paleustult) in three adjacent plots under forest, 10-year continuous cassava, and paddy rice. The SOC in the agricultural land was reduced by 40% (from 10.2 g kg^-1 under forest to 6.1 and 5.5 g kg^-1 under cassava and paddy rice, respectively). Saenjan et al. (1993) found that Korat sandy loam soil (Oxic Paleustult) under continuous pasture for 10 years had SOC content of 6.3 g kg^-1 which was reduced to 5.1 g kg^-1 after it was changed to an upland cash crop, kenaf, for 2 consecutive crops. Gibson (1988) found that in Korat sandy loam soil, organic C was lost at an average rate of 0.24 g kg^-1yr^-1 under continuous cassava (3 g kg^-1 soil organic C) as compared to the original organic C of 5.6 g kg^-1 of the virgin soil. Vityakon et al. (2000) found that without addition of organic residues, ongoing SOM oxidation reduced organic C at the rate of 855 kg C ha^-1 yr^-1 (or 0.39 g kg^-1yr^-1 @ 1.45 Mg kg^-1 soil bulk density) in a sandy upland soil of the northeast. These studies showed that SOM in these sandy soils tends to decrease after the original low disturbance land use is replaced with one involving more serious soil disturbance.

A recent study (Tangtrakarnpong and Vityakon, 2002; Tangtrakarnpong, 2002) compared different pools of SOM (labile and stable as well as total C) under forest and agricultural land use plots situated adjacent to each other along a gentle slope (with forest on the topslope) in typical undulating terrain of the northeast. The soils at the study site were typical sandy soils (Oxic Paleustults). The agricultural land uses included upland field crops of cassava and sugarcane and lowland paddy rice. All pools of SOM studied, i.e. total C, total N, soil litter, particulate organic matter, microbial biomass, and humic substance, were found to be lower in the upland cultivated fields than in the lowland paddy and the forest. SOM pools under forest land use were in most cases higher than those under agricultural land uses (Figure 1 a-f; Table 1).

Figure 1. Contents of organic carbon (a); microbial biomass carbon and nitrogen (b); total N (c); humic acid (d); soil litter (>2 mm) (e); and soil litter (1-2 mm) (f) of soils (0-15 cm depth) under different land-use systems. Means with similar letters are not significantly different at 95% level of probability (LSD). (Tangtrakarnpong and Vityakon, 2002)

Table 1. Carbon of particulate organic matter, i.e. litter (size >2 mm and 1-2 mm), light and heavy fractions and humic acid in sandy soils (0-15 cm depth) under different land-use systems

Factors contributing to high SOM in forest land include high inputs of organic residues and a low degree of soil disturbance. Dry dipterocarp forest at the study site produced 4.4 t ha^-1 of litter fall during the dry season (Dec 2000 – Aug 2001). Soil disturbance in the forest is wild fires occurring during the dry seasons, i.e. February and March. On the other hand, upland fields planted with cassava received a much lower residue input, i.e. 1.3 t ha^-1 in a cropping season, and suffered much more frequent and severe soil disturbance by various cultural practices including ploughing and weeding. In addition, a large part of the biomass produced by the crop was removed through harvest of the roots and removal of the leaves and stems from the field. Cultivation of sugarcane, however, did not lead to as low SOM pools as cassava. This was because sugarcane cultivation produces more organic residues than cassava. Leaf litter fall input in planted and ratoon crops were relatively high, i.e. 4.1 and 6.5 t ha^-1, respectively. In addition, extensive roots and stumps remained in the soil after harvesting. Overall, planted sugarcane produced higher SOM than the ratoon plot counterpart. In fact, it produced higher LF-C than and comparable HF-C fractions to the forest and paddy rice (Table 1), reflecting the high quantities of organic amendments that were received by this plot. As for paddy rice, it produced the highest SOM (with exception of LF and HF) among all the agricultural land uses due to high organic return from stubble which was incorporated back into the soil. In addition, the plots were located in a lowland area that received deposits of organic materials eroded from the uplands, especially in the rainy season. Furthermore, the paddy soil is slightly more clayey than the upland soils.

Soil structure was also degraded once forest land use was changed to agricultural land use. This same study (Tangtrakarnpong, 2002) showed that aggregate size and stability was lower in cultivated land use as compared to the forest (Figure 2). In addition, significant positive correlations were found between aggregate size and stability with the quantities of total C and humic acids (Table 2). Humic substances as well as partially humified particulate SOM play important roles in cementing soil particles into stable aggregates (Piccolo, 1996).

Figure 2. Mean weight diameter (a) and water stability of aggregates of soils (0-15 cm depth) under different land-use systems. Means with similar letters are not significantly different at 95% level of probability (LSD). (Tangtrakarnpong, 2002)

Table 2. Correlation coefficients of mean weight diameter (MWD) and aggregate stability in water with total C and quantity of humic acid under forest and agricultural (upland crops and paddy fields) land use systems

Restoration of soil organic matter in sandy soils employing organic residues of different qualities

Soil organic matter build-up

Applying continuous inputs of organic residues and minimization of soil disturbances are measures that have been shown to restore SOM in sandy soils in the Northeast. Input of organic residues can be done in situ through adoption of agricultural systems that provide organic input continuously. Such systems, that produce significant quantity of litter, both aboveground and belowground, to compensate for the harvestable parts taken out of the systems, include pasture, agroforestry systems with tree components that provide leaf and root litter, and rotational cropping systems incorporating green manure crops. The inputs of organic materials can also be brought in from outside the system, such as application of composts, animal manure and some organic industrial by-products. A 3-year field experiments-based modeling study in Northeast Thailand comparing continuous cassava (3-year cassava) with ley-arable (2-year ley + 1-year cassava) systems showed continuous decline at decelerating rates of organic C in continuous cassava during a 20 year period. On the other hand, a trend of organic C build-up was found under ley-arable farming system where 1-year cassava was alternated with 2-year ley pasture (Figure 3) (Wu et al., 1998). A long-term (10 year) SOM experiment on an upland sandy soil belonging to the Korat series (Oxic Paleustults) which is widespread in Northeast Thailand, where plant residues of different qualities are applied once a year, has shown accumulation of total C. In addition, it showed that frequent soil disturbance (the soil was ploughed every 2 months) lowered the total C below that of the soil deprived of additional organic materials (Figure 4) (P. Vityakon, unpublished data). These results serve to show that SOM build-up in sandy soils in the northeast is possible under a system where continuous input of organic materials is provided and soil disturbance is minimized.

Figure 3. Model-predicted changes in the contents of soil organic carbon under continuous cassava (solid lines) and ley-aerable (2-year ley + 1-year cassava) rotation (dashed lines) in the soils. (Wu et al., 1998)

The quality of organic residues applied is an important driving factor in the build-up of SOM in sandy soils. The quality of organic materials is determined by their chemical as well as physical characteristics that influence decomposition and nutrient transformation of the residues. Although chemical characteristics, notably residue contents of carbon, nitrogen, carbon to nitrogen ratio, lignin and polyphenols, have been more widely investigated than physical characteristics, i.e. size, toughness, cutin concentration etc., some investigations have produced a firm body of knowledge as to the influence of some physical properties of residues on their decomposition in soils (Heal et al., 1997).

Figure 4. Ten-year changes in soil organic carbon (0-15 cm soil depth) as influenced by organic residues of different quality and soil disturbance in a sandy soil. Vertical bars represent SED. (P. Vityakon, unpublished data)

It has been hypothesized that residues resistant to decomposition, i.e. those that contain large amount of recalcitrant C compounds, such as lignin and polyphenols, and possess high C/N ratios in general, can bring about greater SOM accumulation relative to easily decomposable residues, i.e. those with a low amount of recalcitrant C compounds and (usually) low C/N ratios. However, a still-in-progress long-term study of SOM accumulation in the northeast that compared various locally available organic residues of different qualities has found a more complex situation. The results showed that after the first cycle of one-year period of incorporation of these residues, the highest C build-up (total C in <1 mm particle size) of 33% of initial C applied was found in high quality residues, i.e. groundnut stover, whereas low quality residues, i.e. dipterocarp leaf litter, produced a C build-up of only 10% (Vityakon et al., 2000). The probable explanation for this unexpected outcome was that a large proportion of C derived from dipterocarp leaf litter remained in particulate form of >1 mm in size which was not included in total C (<1 mm) analysis. In addition, soil C build-up exhibited significant correlation with two indicators of the chemical quality of the residue, i.e. a positive correlation with %N (r = 0.66*) and a negative correlation with C/N ratios (r = -0.62*). The fact that the total C values were obtained from soil of particle size <1 mm reflects that the C accumulation taken into account was in those fractions only. The parallel results at the end of the second year of the experiment were similar to the first year. However, when the quantity of particulate organic matter larger than 2 mm size in the second year was analyzed, it exhibited a significant positive correlation with C (r = 0.66*), lignin (r = 0.68**), and polyphenols (r = 0.67*) contents (Vityakon, 2004). These latter results showed that low quality organic materials lead to accumulation of particulate macro organic matter and appear to support the original hypothesis.

Considering the effects of different quality plant residues on accumulation of the stable form of SOM, i.e. humic substance, an analysis of the results of an earlier study on decomposition of various types of leaf litter and associated humus formation (Adulprasertsuk et al., 1993) was conducted. It found that the leaf litter rate of decomposition (k) and their respective lignin contents (Table 3) was closely associated, i.e. negative correlation coefficients (r) of -0.84 was found. In addition, k was positively correlated with E₄/E₆ ratios of humic acid (r = 0.71). The E₄/E₆ ratios indicate the degree of development of humic acid, i.e. lower ratios indicate high degree of humic acid development (more condense aromatic structure). These results showed that the types of leaf litter that resist decomposition (possess high k and lignin) tend to bring about more developed humic acid than those that are more rapidly decomposed and have lower content of lignin.

Table 3. The E₄/E₆ ratios of humic acid in soils affected by litter from various trees of different quality as shown by their lignin and decomposition rates (k)

 Nitrogen transformation

Nitrogen mineralization and availability of mineral N in soils receiving organic amendments is an indicator of the effectiveness of the applied organic materials in restoring degraded soils. The chemical composition of organic materials (residue quality) has an important bearing on their decomposition and N mineralization. Studies on various green manure species have shown that N, lignin and polyphenols are the three most important limiting factors for N mineralization under aerobic conditions (Palm and Sanchez, 1991; Oglesby and Fownes, 1992; Constan-tinides and Fownes, 1994; Handayanto et al., 1994). A study in Northeast Thailand compared pattern of N mineralization of various on-farm available organic residues including groundnut stover, rice straw, and leaf litter from common farm trees, i.e. tamarind (Tamarindus indica) and dipterocarp (Dipterocarpus tuberculatus). It was shown that under upland aerobic conditions groundnut stover exhibited peak mineral N release 4 weeks after residue incorporation while the mixture of groundnut stover and rice straw delayed peak mineral N by 4 weeks (Figure 5). These results demonstrate that by mixing high C/N ratio residues (i.e. rice straw) with low C/N ratio residues (i.e. groundnut stover), we can manipulate the temporal pattern of N release to better suit the needs of the crop plants. This is the core idea of the ‘synchrony’ concept (Myers et al., 1997). On the other hand, rice straw and dipterocarp and tamarind leaves had an initial immobilization phase and did not contribute significantly to the mineral N pool even after the immobilization phase was over. The difference in N mineralization patterns among these residues is attributable to their chemical composition. Groundnut stover, which has high N and low C/N ratio, lignin and polyphenol contents, is considered to be a high quality residue, whereas rice straw and dipterocarp and tamarind leaves which have low N and/or high lignin and/or polyphenol contents (Table 4) are considered low quality residues. The influence of residue chemical compositions on N release is confirmed by the significant positive correlations between N release and N content of the residues, while negative correlations are found between N release and C/N ratio and polyphenol content of the residues during the first 4 weeks period after residues were incorporated (Table 5).

Figure 5. Net quantity of mineral nitrogen produced by different organic residues under upland conditions. Vertical bars represent SED. (Vityakon et al., 2000)

Nitrogen transformation processes (mineral­ization-immobilization of N in organic residues) are mediated by microorganisms. Parallel studies of patterns of microbial biomass N after residue incorporation (Figure 6) threw light on the resulting observed N mineralization-immobilization patterns especially that of groundnut stover + rice straw mixture which exhibited delayed N released compared with groundnut stover alone. It was found that microbial biomass N in the mixed treatment was the highest among various residues studied. This was the combined result of the large amount of available C (from both groundnut stover and rice straw) as well as the immobilization of N readily available from groundnut stover. This combined effect resulted in the delayed N mineralization observed as shown in Figure 5.

Agricultural land use in Northeast Thailand farming system is determined by its undulating terrain. This results in two major subsystems: upland fields for production of cash crops, principally cassava and sugarcane; and lowland paddy rice fields. The use of appropriate organic residues as N sources to improve soil fertility of the two subsystems is dictated by their differing soil water regimes. The upland subsystem is virtually continuously aerobic while the lowland subsystem is periodically submerged. It has been shown that N mineralization in anaerobic conditions is higher and faster than in aerobic conditions. A pot study (Vityakon and Dangthaisong, 2005) comparing the performance of various kinds of leaf litter (both leguminous and non-leguminous) from trees found in the northeast farming systems under submerged and aerobic conditions in sandy soils found that there was higher net N mineralization and lower immobilization under submerged than aerobic conditions. In addition, N mineralization was more rapid in submerged (first peak mineralization within 2-4 weeks) than aerobic conditions (Figure 7, 8 a, b). The higher and more rapid N mineralization in submerged than aerobic conditions is partly because the microflora in flooded soils (with a higher proportion of bacteria) have a lower N requirement stemming from their lower synthetic efficiency compared with the aerobic population (with more fungi and actinomycetes) (Broadbent, 1979; Ono, 1989 as cited by Vityakon and Dangthaisong, 2005). In addition, leaching and/or dilution of polyphenols may also lower efficiency of the compounds in binding proteins and consequently inhibit N release under submerged conditions.

Table 4. Chemical quality characteristics of plant residues

Table 5. Correlation coefficients relating amounts of soil mineral N (NH₄⁺ + NO₃^- -N) to chemical characteristics of plant residues under upland (n = 12) conditions

Figure 6. Dynamics of microbial biomass nitrogen as affected by different organic residues under upland conditions. Vertical bars represent SED. (Vityakon et al., 2000)

Based on chemical compositions alone, Palm et al. (2001) have classified organic residues into 4 types to guide the selection of residues that provide optimal N inputs for soil fertility management. The classification employs a minimum set of quality indica­tors, i.e. N, lignin and polyphenol content of organic materials (Figure 9). Vityakon and Dangthaisong (2005) have shown that this classification system may have limited applicability in a farming system that has both upland crops and paddy rice subsystems, such as is commonly found in Northeast Thailand, because organic residues (leaf litter from some trees) exhibited different N mineralization-immobilization patterns under submerged and aerobic conditions. For examples, under submerged conditions, leaf litter of Samanea saman and Xylia xylocarpa exhibited net N mineralization peak during the early phase (week 2) of decomposition followed by slight net N immobilization during the intermediate phase which preceded low net N mineralization in the final period (Figure 7a). However, under aerobic conditions they exhibited net N immobilization throughout the 16-week period after incorporation into the soil (Figure 8a). This has further shown that they should be managed differently under submerged and aerobic conditions. In submerged conditions, they showed some early low N release (Figure 7a) and, hence can be used as an immediate N source to supplement other higher release N sources. However, in aerobic conditions they did not release N during at least the 16-week study period (Figures 8a), therefore they may have to be managed in the way suggested for the lower categories of the classification of Palm et al. (2001), i.e. to be mixed with mineral fertilizers or high-quality organic materials to delay N release, or to be composted before soil application, or to be used as surface mulch for erosion and water control.

Figure 7. Net quantity of mineral nitrogen present in submerged soil of Northeast Thailand at various times after amendment with different organic residues: (a) legume residues, (b) non-legume residues. Vertical bars represent SED. (Vityakon and Dangthaisong, 2005)

Figure 8. Net quantity of mineral nitrogen present in aerobic soil of Northeast Thailand at various times after amendment with different organic residues: (a) legume residues, (b) non-legume residues. Vertical bars represent SED. (Vityakon and Dangthaisong, 2005)

Figure 9. Criteria for selecting organic residues as nitrogen sources for soil fertility management as proposed by Palm et al. (2001)

Conclusions

In comparison to natural forest, soil degradation has occurred in sandy soils under agricultural land uses in Northeast Thailand as indicated by declines in the various pools of organic matter. The degradation is due to reduced input or recycling of organic materials and higher disturbance of soil due to soil management practices. In upland fields, the degradation of SOM pools was more severe under cassava than sugarcane because sugarcane returns more organic residues to the soil than cassava. Paddy rice cultivation does not bring about significant SOM degradation because the paddy fields usually possess less sandy soil (higher clay content than upland fields), receive organic deposits eroded from upper lying land in undulating terrain, and recycle crop residues in the roots and stubble of the rice plants.

SOM levels in cultivated sandy soils can be built up by increasing organic inputs into the system in two ways. Firstly, by generating increased quantities of organic residues within the system itself, such as under pasture, rotation with green manure crops, or having trees that produce litter of desirable quality. It can also be achieved by bringing organic inputs from off-farm sources, such as industrial by-products, animal manure etc. Quality of organic residues also influence different pools of SOM. While high quality residues bring about short-term nutrient release, low quality (slowly decomposed) residues tend to enhance the stable pool (humic substance) of SOM. Another measure is to minimize soil disturbance, such as ploughing, since ploughing has been shown to reduce SOM content.

Restoration of soil fertility, notably nutrient N, can be achieved through the use of organic residues of different qualities. The high quality residues can release N within a short term, which can be useful when the crop requirement is immediate, otherwise the mineral N produced can be wasted. Mixing high and low quality organic residues at appropriate ratios can bring about the ability to manipulate the time of N release to coincide with crop demand. This is the basis of the ‘synchrony’ concept. Low quality residues are also useful as N sources when mixed with high quality residues or chemical fertilizers. N release of low-quality residues can be enhanced if applied in submerged conditions, i.e. in paddy rice conditions. Therefore, different kinds of organic residues are appropriate as N sources under submerged and aerobic conditions. In classifying organic residues for appropriate use as N sources, one needs to take into account environmental conditions affecting N mineralization, in addition to residue chemical compositions. In addition to non-harvestable organic residues, tree litter is another organic residue available in farming systems of Northeast Thailand that can potentially enhance SOM build-up and N fertility of the sandy soils.

Acknowledgments

I would like to thank A.T. Rambo for his editorial assistance and Supaporn Pongkhaek for her help in preparing the graphics.

References

Adulprasertsuk, M., Vityakon, P. , and Prachaiyo, B. 1997. Leaf litter of some native trees in Northeast Thailand: Their decomposition and derived soil organic matter characteristics. In: Proceedings of the International Conference on Tropical Forestry in the 21^st Century. Vol. 2. Global changes in the tropical contexts. Faculty of Forestry, Kasetsart University, Bangkok, 25-28.

Broadbent, F.E. 1979. Mineralization of organic nitrogen in paddy soils. In: Nitrogen and rice. IRRI, Los Banos, Philippines, 105-118.

Constantinides, M. and Fownes, J.S. 1994. Nitrogen mineralization from leaves and litter of tropical plants: Relationship to nitrogen, lignin and soluble polyphenol concentrations. Soil Biology and Biochemistry, 26, 49-55.

Gibson, T. 1988. The agronomic and soil fertility effects of leguminous ley pastures in sandy, upland soils of Northeast Thailand. Ph.D. dissertation, University of Queensland, 266 p.

Handayanto, E., Cadisch, G. and Giller, K.E. 1994. Nitrogen release from prunings of legume hedgerow tree in relations to quality of prunings and incubation method. Plant and Soil, 160, 237-248.

Heal, O.W., Anderson, J.M. and Swift, M.J. 1997. Plant litter quality and decomposition: An historical overview. In: Cadisch, G. and Giller, K.E. (eds.), Driven by nature: Plant litter quality and decomposition. CAB International, Wallingford, 3-32.

Myers, R.F.K, van Noordwijk, M., and Vityakon, P. 1997. Synchrony of nutrient release and plant demand: Plant litter quality, soil environment and farmer management options. In Cadisch, G., and Giller, K.E. (eds.), Driven by nature: Plant litter quality and decomposition. CABI, Wallingford, 215-229.

Oglesby, K. and Fownes, J.H. 1992. Effects of chemical composition on nitrogen mineralization from green manures of seven tropical leguminous trees. Plant and Soil, 143, 127-132.

Ono, S. 1989. Nitrogen mineralization from paddy and upland soils under flooded and nonflooded incubation. Soil Science and Plant Nutrition, 35, 417-426.

Palm, C.A. and Sanchez, P.A. 1991. Nitrogen release from the leaves of some tropical legumes as affected by their lignin and polyphenolic contents. Soil Biology and Biochemistry, 23, 83-88.

Palm, C.A., Gachengo, C.N., Delve, R.J., Cadisch, G. and Giller, K.E. 2001. Organic inputs for soil fertility management in tropical agroecosystems: application of an organic resource database. Agriculture, Ecosystem and Environment, 83, 27-42.

Piccolo, A. 1996. Humus and soil conservation. In: Piccolo, A., ed., Humic substances in terrestrial ecosystems. Elsevier Science, Amsterdam, 225-264.

Saenjan, P., Dejbhimon, K., and Paiboonsak, S. 1993. Relationships between organic carbon and optical characteristics of sandy soils receiving long-term organic matter. Thai Journal of Soils and Fertilizers, 15, 97-107. (in Thai with English abstract).

Tangtrakarnpong, S. 2002. Changes of different pools of soil organic matter under different land uses in undulating terrain of Northeast Thailand. A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Soil Science, Graduate School, Khon Kaen University, 171 p. (in Thai with English abstract).

Tangtrakarnpong, S., and Vityakon, P. 2002. Land use and soil organic matter in Northeast Thailand: microbial biomass, humic acid, and mineral N. Transactions of the 17^th World Congress of Soil Science, Bangkok, Symposium 5, Paper no. 461, CD ROM. International Union of Soil Science, 1-15.

Vityakon, P. 1991. Relationships between organic matter and some chemical properties of sandy soils with different land use and soil management. Thai Journal of Soils and Fertilizers, 13, 254-264. (In Thai with English abstract).

Vityakon, P., Meepetch, S., Cadisch, G and Toomsan, B. 2000. Soil organic matter and nitrogen transformation mediated by plant residues of different qualities in sandy acid upland and paddy soils. Netherlands Journal of Agricultural Science, 48, 75-90.

Vityakon, P. 2004. Long-term studies on organic matter build-up, nutrient transformation, and dynamics of fast and slow pools of organic matter in sandy soils as influenced by qualities of plant residues added. Research report of a project under the Thai government funding to KKU for FY 1999. Faculty of Agriculture, Khon Kaen University, 30 p.

Vityakon, P., and Dangthaisong, N. 2005. Environmental influences on nitrogen transformation of different quality tree litter under submerged and aerobic conditions. Agroforestry Systems, 63, 225-236.

Wu, J., O’Donnell, A.G., Syers, J.K., Adey, M.A., and Vityakon, P. 1998. Modelling soil organic matter changes in ley-aerable rotation in sandy soils of Northeast Thailand. European Journal of Soil Science, 49, 463-470.

¹ Department of Land Resources and Environment, Faculty of Agriculture, Khon Kaen University, Khon Kaen 40002, Thailand, Tel: 66 43 364639, Fax: 66 43 202361, e-mail: [email protected]