Session 5
“The role of organic matter and
biological activity”

Organic matter and biofunctioning in tropical sandy soils and implications
for its management

Blanchart, E.¹; A. Albrecht¹; M. Bernoux¹; A. Brauman¹; J.L. Chotte¹; C. Feller¹; F. Ganry³;
E. Hien³; R. Manlay^{1, 4}; D. Masse²; S. Sall² and C. Villenave¹

Keywords: soil microorganisms, soil fauna, crop management, soil carbon fractions

Introduction

Sandy soils are widely distributed in the tropics where they occupy most of arid and semi-arid areas. For instance, the total estimated extent of Arenosols is 900 million hectares, mainly in Western Australia, South America, South Africa, Sahel, and Arabia (WRB and FAO/Unesco soil map of the World). It is well known that these “problem soils” are characterized by a low soil organic carbon (SOC), a low cation exchange capacity (CEC), a high risk of nutrient leaching, a low structural stability, and a high sensitivity to erosion and to crusting. Both chemical fertility and physical stability are weak in these soils (Pieri, 1992; Sanchez & Logan, 1992). These characteristics are due to their sandy texture, the low reactivity of their clays, and to climatic conditions that often accompany tropical sandy soils. Due to their dominant mineralogy (generally: quartz, kaolinite, iron and aluminium oxides) and their sandy texture, the role of organic matter on the properties of these soils, on their potential of productivity and on the sustainability of agricultural systems is thus fundamental (Pieri, 1992; Feller et al., 1995b). The control of soil organic matter (SOM) on chemical (CEC, pH, some cations such as calcium and magnesium), and physical (porosity, structural stability) properties has often been demonstrated (Asadu et al., 1997). In sandy soils, it thus appears fundamental to manage all components that affect soil fertility: SOM, and soil biota. Biological processes are crucial to sustain the fertility of sandy soils as they control C and N fluxes (Menaut et al., 1985; Perry et al., 1989; Chotte et al., 1995; Lavelle, 1997). Like in other pedoclimatic zones, the assessment of C, N and P in agrosystems on tropical sandy soils is a useful tool to define sustainable intensification plans necessary to respond to population increases and global change issues (Manlay et al., 2002a, b, c).

In this paper, we will successively analyse (i) the specificity of sandy soils with regard to their organic status, (ii) the agronomic determinants of their organic status, (iii) the agronomic determinants of their biological activities and, (iv) the relationships between SOM and biota with regard to agro-ecosystem management. The three latter parts are based on case studies essentially from West Africa.

Organic status of tropical sandy soils

Relationships between soil organic matter and soil properties

SOM controls many chemical, physical and biological properties that affect the capacity of a soil to produce food, fibres and fuel. It is the main source of ecosystem energy, and also the main source and a temporary sink of nutrients for plants in the agrosystems.

SOM plays a major role in soil fertility through different functions (Feller, 1995a):

• The storage of nutrients (“mineral supply” function). Some minerals like P, Ca, K, Mg are associated in a non-exchangeable form to SOM. They are released during OM decomposition and their dynamics is thus dependent on that of OM. In soils like sandy soils naturally poor in these elements, OM constitutes a interesting reserve for them.
• The increase in CEC (“exchange and sorption” function). This function is linked to the surface properties of soil organic and organo-mineral components: cation and anion exchange capacity, physical and chemical adsorption and desorption properties. These properties define the availability of some nutrients, cation equilibrium and the efficiency of fertilizers and xenobiotic molecules.
• The improvement of soil structural stability (“aggregation” function). Structural stability determines many soil physical and biological properties.
• The stimulation of faunal, microbial and enzymatic activities (“mineralization and immobilization” = “biological” function) that determines carbon C, nitrogen N and phosphorus P and sulphur S cycles. These elements follow successions of mineralization and microbial immobilization. This controls the fluxes of these elements in the soil-plant system (storage or losses) or between different soil compartments.

Relationships between soil organic carbon stocks and texture

Many studies in West Africa showed that SOM content in soil surface horizons is dependent on soil texture (Jones, 1973; Boissezon, 1973; Feller et al., 1991). Feller (1995a) and Feller & Beare (1997) proposed to link linearly SOC content with fine soil particles 0-20 µm, i.e., clay + fine silt (C + FS) (Figure 1).

Figure 1. Relationships between soil organic content and clay + fine silt content in tropical 1:1 (low activity clay LAC) and 2:1 (high activity clay HAC) soils (adapted from Feller & Beare, 1997)

The relationship clearly indicates that the lower the clay + fine silt content, the lower the soil carbon content. Since this study, the same relationship has been observed in other tropical regions: in Senegal (Manlay et al., 2002b, c), in Martinique (West Indies) (Venkatapen et al., 2004). Tropical sandy soils are thus soils naturally poor in soil organic carbon.

Feller et al. (1991) showed that temperature was not a major determinant of SOC stocks differentiation in considered situations of West Africa (the effect of temperature is only expressed in altitude tropics with mean temperature below 18-20ºC). Feller et al. (1991) observed also only a weak effect of rainfall on SOC stocks with a slight increase in C stock in the most humid areas. Taking into account the effect of fine particles (C + FS) and rainfall R, the relationships between SOC content and these factors was:

SOC (gC.kg^-1 soil) = 0.47 (C+FS) + 0.002 (R) – 1.74 Thus, tropical sandy soils are naturally poor in SOC. When analysing the contribution of Arenosols (main sandy soils) to the total SOC stocks in all World soils, it can be calculated that although Arenosols represent 4.4% (ca. 6 millions km²) of total World soil area, these sandy soils contribute only to 0.6% (4.3 Pg C) of total SOC stock in the upper 30 cm (723 Pg C)¹.

Potential of carbon storage and sequestration in sandy soils

The potential of C storage can be assessed as the difference (∆C) between SOC in native or perennial vegetation and SOC in annual crops. Feller et al. (1991, 2002) observed that ∆SOC are more important for clayey soils than for sandy soils (Figure 2).

Figure 2. Effect of land use and soil texture on SOC content in different tropical soils (Feller, unpub. data)

Similar observations were made for other part of the tropics (Manlay et al., 2002c; Venkatapen et al., 2004). In sandy soils, SOC content in native or perennial vegetation, or in improved systems characterized by high organic inputs are not much higher than SOC content in annual crops. In West Africa, decreases in C content following the installation of crops represents 30 to 40% of non-cropped soils (Hien, 2004). Moreover, these variations appear more rapid for sandy soils (less than 5 years) than for clayey soils (5 to 10 years). As a consequence, sandy soils have a very low potential of carbon storage, compared to clayey soils. The role of tropical sandy soils in the mitigation of atmosphere greenhouse gaz (GHG) is thus very weak. Manlay et al. (2002c) hypothesized that the contribution of these soils to the global mitigation of GHG release does not necessarily require a local carbon sequestration. Settling people may be a means to limit deforestation and carbon release from more humid areas or more clayey soils. This can be achieved by a cropping intensification.

Organic compartments in tropical sandy soils

The morphologic observations at different scales (optical, electronic microscopy) of SOM associated with different particle size fractions in ferruginous and ferrallitic soils allowed Feller (1979) to gather SOM into 3 compartments:

• The fraction 20-2,000 µm is composed of plant debris at various stages of decomposition, associated with sand and coarse silt;
• The fraction 2-20 µm is made up of fungal and plant debris associated with fine silt and very stable organo-mineral aggregates;
• The fraction 0-2 µm consists of amorphous, colloidal OM, debris of plant and fungal walls associated with organo-mineral micro-aggregates.

Feller et al. (1991) and Feller (1995a) observed that these compartments vary with soil texture. In sandy to sandy-clayey soils of West Africa, the fractions 20-2,000 µm and 0-2 µm represent 30 and 36% of total soil carbon, respectively, while in clayey soils, they represent 17 and 58% of total soil carbon, respectively. Feller (1995a) studied the effect of soil texture on the variations in total SOC content and in organic compartment C content in soils: (i) in a succession deforestation-cropping, and (ii) in a succession cropping-fallowing.

In the former succession, the installation of crops after deforestation leads to decreases in SOC contents by 40%, 44%, and 55% in sandy, sandy-clayey, and clayey soils, respectively. Decreases in total C content are thus more important for clayey soils than for sandy soils. In the sandy soil, most of C is lost in the coarse organic fraction (20-2,000 µm) while in both other type of soils, total C loss is mainly due to losses in fine and medium-size fractions (0-2 and 2-20 µm) (Figure 3). For sandy soils, decrease in SOM is very rapid (3 years) for all fractions, even if the rate of decrease is lower for fine fractions than for coarse fractions.

Conversely, the installation of fallows after many years of cultivation leads to increase in SOC contents by 92%, 44%, and 36% in sandy, sandy-clayey, and clayey soils, respectively (Figure 4). In the sandy soil, most of C variation is linked to an increase in C of the coarse organic compartment (20-2,000 µm); in the sandy-clayey soil, the C content of the three organic compartments increase while in the clayey soil, total C increase was linked to increase both in coarse and fine fractions. Low soil C content due to 10 years of cropping is rebuilt more rapidly (after 8 years of fallow) for 20-2,000 µm fraction than for the other fractions (15 years of fallow).

Figure 3. Effect of installation of crops after deforestation on SOC contents in soil and in three organic compart­ments. Variation in C content (DC) between native vegetation and crops (adapted from Feller, 1995a)

Figure 4. Effect of installation of fallows after cultivation on SOC contents in soil and in three organic compart­ments. Variation in C content (DC) between crops and fallows (adapted from Feller, 1995a)

As a consequence, the renewal rate of C in organic compartments decreases from coarse to fine fractions. Mean residence time of the coarse fraction and in the medium + fine fractions in sandy soils has been estimated to as 12 and 30 years, respectively. When analysing the sandy soils only, the half time life was estimated to 8, 18 and 22 years for >50 µm, 2-50 µm and 0-2 µm fractions, respectively (Feller & Beare, 1997). This means that the coarse fraction (plant debris) in sandy soils plays a major role, in short- and medium-term SOM dynamics, on soil properties, and on soil-plant relationship.

In terms of agrosystem management, these results indicate that the restoration of SOM stock in sandy soils, which is linked to the dynamics of the coarse fraction, is possible in a medium-term (10 years). Conversely, SOM restoration in clayey soils is much longer and mostly concerns both fractions (Figure 5).

Figure 5. Variation in C content (DC) in two soil organic (fine and coarse) fractions as a function of soil texture in improved systems as compared with traditional systems (unpublished, adapted from Feller et at., 2002)

Functions of organic compartments

The three organic pools discussed above fulfil different functions in soils. As a whole, SOM is responsible for four main functions in soil: “mineral supply” function, “exchange and sorption” function, “aggregation” function, and “biological” function. The notion of functional compartment for SOM was discussed and quantified by Feller and co-authors (Feller 1995a; Feller et al., 2001). These authors demonstrated that in sandy tropical soils, the coarse organic compartment carries the biological function of the OM. This fraction plays an important energetic role as it represents more than 80% of easily decomposable C in sandy soils, but only 30% in clayey soils. On the other hand, medium (220 µm) and fine (0-2 µm) fractions are characterized in all soils by low C mineralization coefficients. Net N mineralization coefficients of coarse fractions are generally low especially when C/N ratio of the fraction is high. This coefficient increases from coarse to fine fractions; thus, in clayey soils, more than 85% of N mineralized comes from the fine fraction (0-2 µm) whereas in sandy soils, more than 50% of N mineralized comes from fractions larger than 20 µm. Studies in West Africa also showed that CEC increases with the decrease of organic fraction size. In these soils, C content, especially that of the fine fraction (0-2 µm) controls soil CEC (Guibert, 1999).

As a consequence, the way SOM improves soil properties depend on the compartments in which it is found (Feller et al., 2001). SOM in fine and medium fractions influences the capacity of a soil to store and exchange nutrients. In this respect, the application of a manure along with a N fertilizer is favourable for preferential storage of C in the fine fractions, thus showing the advantage of this practice in the stabilization of SOM (Hien, 2004). Conversely, SOM in coarse fraction has a rapid turnover and carries biological functions (mineralization of C, N, P in a short term). This fraction is specifically functional in soils with less than 10% clay (Feller, 1995a). Its function is biological: short-term mineralization of C, N and P, and storage ability for N or non-exchangeable calcium. The role of plant debris in the biogeochemical functioning of sandy soils appears fundamental. This is especially true for N, as in sandy soils, N initial reserve and storage potential are low and SOM turnover is rapid (Blondel, 1971a, b, c, d, e; Pieri, 1992; Ruiz et al., 1995). From an agronomic point of view, there is a need to favour agricultural practices that allow an important and constant restitution of plant or animal debris: composts, manures, successions of crops with strong root systems, short fallows, agroforestry, etc. (Ganry, 1991; Pieri, 1992; Feller, 1995a; Ganry et al., 2001; Manlay et al., 2002b, c).

Residue management, organic matter and organic compartments in sandy soils: case studies

It has often been demonstrated, for sandy soils of West Africa, that cropping systems that do not imply high levels of organic restitutions to the soil, either on a root form (fallows, pastures) or on organic amendments, lead to the decrease of plant productivity and/or to soil degradation (acidification, decrease in structural stability). This decrease is often linked to a decrease in SOC stocks (Feller et al., 1987; Pieri, 1992). The agricultural development of tropical sandy soils is often hindered by the fact that the decay of SOM is much more rapid than in clayey soils. This acceleration results not only from the low level of clay but also from the pattern of hydrometry throughout the year, both emphasizing the oxidation of SOM. The phenomenon is made all the more intense by the low soil protective colloids content.

The main question is: what is the relation between SOM and land productivity? Until the 1990s, the literature did not report a critical SOM content, assuming that the relation between SOM and productivity was more or less linear. Pieri (1992) studying Sudan-Sahel farming situations subjected to strong agro-environmental constraints showed that the strong relationship between the productivity of land and its organic richness were not rigorously linked. In Burkina Faso, Hien (2004) found a critical value of C in the soil, between 6 and 7 gC.g^-1 soil. The yields of sorghum decreased below 6 gC.g^-1 soil and stabilized above this value. Feller (1995a) established that the SOC threshold for the sustainability of agrosystems of Western Africa was 6.8 gC.g^-1 soil, this result being close to that of Hien (2004).

Here, we analyse the effect of different land uses on total SOM and distribution of C within the different organic fractions. Most of studies presented here come from West Africa (Senegal, Burkina Faso). Soils are sandy or sandy-clayey soils with sandy upper horizon; clay contents are always less than 15%.

Effect of annual crops and organic amendments

When natural vegetation is replaced with crops, one can observe a decrease in SOC stocks, and especially of C in the coarse fraction (>20 µm) (Feller et al., 1991). Manlay et al. (2002c) noticed that in crops (millet, maize, rice) in South Senegal (region of Sare Yorobana, soil with less than 10% clay), 90% of total C, 90% of total P and 95% of total N were found in the soil. As millet and maize received higher organic inputs and nutrients (manure, crop residues) than groundnut, their C and N contents were higher. In this region, the improvement of soil organic status under continuous crop can only be achieved in fields close to compounds where organic inputs are available.

Feller et al. (1987) and Feller (1995a) measured the effect of organic amendments on total C contents and SOC distribution in organic compartments, in a succession groundnut-millet in sandy soils of Senegal. In the first study (soil with 4% clay), C content was 2.0 gC.kg^-1 soil in the control and 2.4 gC.kg^-1 soil in the treatment with buried compost. All added carbon was found in the >50 µm fraction (Figure 6). In a second experiment (soil with 4% clay), C content in the control was 1.8 g.kg^-1 soil and it was 2.2 g.kg^-1 soil in the treatment with a straw mulch. In this case, all added C was found in the <50 µm fraction. In the third experiment (soil with 8% clay), the presence of a straw mulch leads to an increase of C content (4.3 g.kg^-1 soil) as compared to the control (3.1 g.kg^-1 soil). C increase was mainly in <50 µm fraction and also in >50 µm fraction.

Figure 6. Variations in C content in two soil organic fractions between control and treatments with organic amendments in three experiments (see text for details) (unpublished, adapted from Feller et al., 2002)

Organic transfers improve chemical properties in three ways: they are a net source of C and nutrients; they contribute to a gain in CEC and stimulate biological activity (Feller, 1995b; Asadu et al., 1997). Manlay et al. (2002c) observed also that organic practices in continuous crops had a more important effect on soil chemical status (P, Ca, K, CEC, S, pH) than fallowing.

Effect of cover crops

In Benin, the introduction of a cover crop (Mucuna pruriens var. utilis, Fabaceae) in maize crops, on a sandy soil (10% clay) lead to an increase in SOC content, and especially in C of the >50 µm organic fraction (Figure 7) (Azontonde et al., 1998; Bayer et al., 2001; Barthès et al., 2004). On the opposite, increase in SOC content is mostly linked to C increase in the <50 µm fraction in clayey soils.

Figure 7. Variation in C content (DC) in two organic fractions in systems with cover crops, as compared with traditional systems without cover crops. Effect of Mucuna pruriens in sandy soils in Benin, and effect of no-tillage with cover crops in clayey soils in Brazil (two situations PD1 and PD2) (Azontonde et al., 1999; Bayer et al., 2001)

Effect of fallows and agroforestry

If the important decrease in SOC contents after deforestation in the tropics is well established (Maass, 1995), the potential of fallows to increase C contents has also been demonstrated (Manlay et al., 2002b). But the effect depends on soil texture, tree species, management, etc. (Szott et al., 1999). In sandy soils of Senegal, Manlay et al. (2000) measured an increase of SOC content with the age of fallows (4.7 gC.kg^-1 soil in a 2-year old fallow, 9.0 gC.kg^-1 soil in a 26-year old fallow). In the same time, calcium, magnesium and CEC increased with the age of fallows. With ageing fallows, coarse root biomass increases while herbaceous biomass decreases. Thus, in sandy soils, SOC increase with the age of the fallows is linked to an increase in tree root biomass and to more important litter inputs (Asadu et al., 1997; Floret, 1998). In most of agrosystems, especially those that are frequently burnt, as in West African Savannas, roots represent the main SOC source (Menaut et al., 1985; Manlay et al., 2000). In South Senegal, the effect of fallowing on soil organic status was only noticeable in the upper 20 cm of soils, but there was no effect on soil physical properties (Manlay et al., 2002a). The installation of fallows rapidly led to increases in soil C content (by 30% in one year); this was due to a rapid develop­ment of trees. Then, SOC content increase was not so rapid (Figure 8), maybe because of a poor protection of SOM against oxidation by biological activities in sandy soils; thus the protection of SOM against mineralization, erosion and leaching is not very efficient (Feller & Beare, 1997). In fact, mesh-bag experiment showed that 40 to 60% of woody roots disappeared after 6 months of incubation (Manlay et al., 2004). Fallowing mostly affected the >50 µm organic fraction whose contribution to total C doubled after crop abandonment. It also allowed a rapid restoration of N and available P contents (Friesen et al., 1997; Manlay et al., 2004).

Figure 8. Effect of the age of fallows on C, Ca and Mg contents in South Senegal (Manlay et al., 2000)

In different sites of West Africa and West Indies, Feller (1995a) and Feller et al. (2001) obtained the same results as those obtained from South Senegal. Moreover, these authors demonstrated that in sandy soils, soil C increase observed in fallows (after crops) on sandy soils was mainly due to C increase in the >50 µm fraction, while in clayey soils, C increase in <50 µm fraction was mainly responsible for total soil C increase (Figure 9).

Figure 9. Variation in C content (DC) in two soil organic fractions in fallows, as compared with continuous crops, in soils with different clay contents (Feller, 1995a)

In Acacia plantations in Cameroon (soil with 5% clay), Harmand et al. (2000) measured after 4 years a SOC content increase as compared with continuous crops; this was mainly linked to an C increase in the >50 µm fraction. Agroforestry systems are often linked to a strong increase in the total SOC content of sandy soils (Figure 5).

As emphasized by Manlay et al. (2002c) C dynamics in fallows is a determining factor for following crops. A mineral fertilization without organic amendments leads to the mineralization of SOM and to a decrease in soil structure, pH and affects productivity (Pieri, 1992; Manlay et al., 2002c).

The biotic components in tropical sandy soils

As said above, SOM is the energetic source of soil biota and soil biota controls the dynamics of SOM, which is fundamental for the fertility and the properties of soils, and especially of sandy soils. Here we analyse some recent studies dealing with the relationships between land use, soil biota abundance and activity, SOM dynamics and plant productivity in sandy soils (West Africa).

Soil fauna

Soil fauna is known to influence soil chemical, physical and biological properties (Lavelle & Spain, 2001). Zoological groups more often studied with regard to plant productivity and soil properties are ecosystem engineers and nematodes. The former group gathers macroinvertebrates that modify soil physical organization through the production of biogenic structures, and modify the nature and availability of nutrients for other soil organisms (Jones et al., 1994; Lavelle, 1997). Main ecosystem engineers present in tropical sandy soils are termites and earthworms. Nematodes as they belong to different trophic categories affect soil microorganism communities (fungi and bacteria) and plants.

In sandy soils of the arid and semi-arid tropical areas, termites are generally the dominant group of soil macrofauna while earthworms are limited by low rainfall: below 800 mm of rainfall amount, earthworms become rare (Lavelle, 1983).

When present, the effect of earthworms on soil properties can be important. In the soil of the sub-humid savannas of Lamto, Ivory Coast (7% clay in the upper 20 cm of soil), communities are important (ca. 500 kg.ha^-1) and earthworms annually ingest up to 1,200 Mg soil.ha^-1 (Lavelle, 1978). As a consequence, the upper cm of soil is made up of earthworm casts that control physical and biological properties of soils (Blanchart, 1992; Martin & Marinissen, 1993; Blanchart et al., 1997). As showed in different field or laboratory experiments, earthworm activity tends to decrease C content of the coarse (>50 µm) organic fraction and to increase C content of the fine organic fraction in casts, as compared to non-ingested soil (Figure 10, adapted from Villenave et al., 1999). In these water-stable biogenic structures, SOM is physically protected against mineralization (Martin, 1991; Blanchart et al., 1993; Lavelle et al., 1998). The mutualistic interactions between earthworms and microorganisms, which start in earthworm gut and end in casts lead to a strong increase in microbial activities and a subsequent release of nutrients (N, P). The effect of earthworms on SOM dynamics is different according to the duration we consider: in the short term, earthworms stimulate microbial activity, decompose OM and release nutrients available for plants while in the long term, earthworms protect SOM against mineralization. Nevertheless, the presence of earthworms in cropped soil (with sandy upper horizons) does not seem to affect SOC stocks in a medium-term (Villenave et al., 1999).

Many studies have recently been dedicated to termite communities and activities in West Africa: effect on erosion and infiltration (Mando et al., 1996; Léonard et al., 2004; Valentin et al, 2004), on organic resource disappearance and nutrient release (Brown & Whitford, 2003; Rouland et al., 2003; Zaady et al., 2003; Ouédraogo et al., 2004), on soil microbial communities (Brauman, 2000; Fall et al., 2001, 2004; Ndiaye et al., 2003, 2004a; Jouquet et al., 2005), on nest properties (Fall et al., 2001; Mora et al., 2005). In a mesh-bag experiment in South Senegal, Manlay et al. (2004) measured a more rapid and important root disappearance in presence of fauna (mass loss 70% of initial root biomass after 12 months) than in absence of soil fauna (mass loss less than 50%). Termites and ants allowed the reallocation of OM and increased its availability for mineralization (in the presence of fauna, only a few fraction of C was stabilized in soil). In sandy soils, the important consumption of organic inputs by heterotrophic organisms is fundamental for the fertility of agrosystems.

Figure 10. Variation in C content (DC) in two organic fractions in soils (clay content from 8 to 22%) with inoculated earthworms as compared with soils without earthworms. Data come from different laboratory (incubation) and field (cultivation) experiments. Duration of the experiments is indicated (adapted by Feller, 2002, from Villenave et al., 1999)

Fallows (or agroforestry) allow the restoration of the biological control of ecosystem fertility (Manlay et al., 2002c). After crop abandonment, many studies show an increase of soil macrofauna (density, biomass, activity) (Fall, 1998; Manlay et al., 2000; Derouard, unpub. data). For instance, in South Senegal, the density of macrofauna was 3 times higher in a 10-year old fallow than in continuous crops (Fall, 1998). Some authors emphasize the importance of fallows in favouring ecosystem resilience and stability to climatic uncertainties, to poor nutrient status, and to poor physical stability thanks to the increase in soil diversity and density macrofauna and to root development (Menaut et al., 1985; Ewel, 1999; Manlay et al., 2002c).

The effect of termites on nematodes was studied in Senegal on a sandy soil and results showed that nematofauna structure in termite covers was comparable whatever the termite species, but it is different from that of the soil (0-10 cm). Many works show that plant parasitic nematode communities can be manipulated by managing vegetation, these nematodes being linked roots. Moreover, the pathogeny of nematodes depends on the structure of their community (Cadet & Spaull, 1998). For instance, it was demon­strated in Senegal that the presence of the species Helicotylenchus dihystera was associated with a reduction of the pathogeny of the whole nematode community because of the stimulation of root development (Villenave & Cadet, 2000). This species disappear with the establishment of crops after fallows; this may be due to the disappearance of woody roots. It thus seems necessary to preserve trees in agrosystems, and agroforestry could be a means to increase populations of D. dihystera and to reduce the impact of parasitic nematodes (Buresh & Tian, 1997).

Microorganisms

Microbial communities in soils are the actors of the decomposition of the organic matter. The complete decomposition of complex organic substrates such as organic residues relies on the succession of diverse microbial species characterized by different enzyme abilities (Swift et al., 1979; Zvyagintsev, 1994). In tropical sandy soils, very few investigations pointed out the importance of microbial community on decomposition processes.

Microbial status in fallows on tropical sandy soils

Organic and microbial status of soils (0-10 cm) under natural and improved fallows were studied in a Lixisol in two different field sites in Senegal (Sonkorong et Saré Yorobana) (Ndour et al., 1999; Ndour et al., 2001). At Sonkorong, soil organic matter and total microbial biomass were significantly higher in natural protected fallows than in non-protected ones and cultivated soils. No significant differences were recorded for non-protected situation, and cultivated soils. For managed situations, the duration of the fallow did not modify organic and microbial content of soils. Enzymes activities (ß-glucosidase, amylase, chitinase, xylanase) were investigated in these situations. Principal component analyses revealed a relationship between enzyme activities and the age (4, 11, and 21 year-old) and the management of fallows (fenced versus grazed), the vegetation (natural, Acacia holocericea, Andropogon gayanus). ß-glucosidase and amylase were significantly higher in the oldest natural fallows. The highest xylanase activity was recorded for the Andropogon gayanus improved fallows. This fallow showed also the highest chitinase, similar to that of the 21 year-old natural fenced fallow. Amongst the different management of the fallows, the introduction of Acacia holocericea depleted all the tested activities. In contrast comparisons between young and old fallows and crop plots at Saré Yorobana, did not show any significant differences. Coarser soil texture and higher frequency of land fires might explain these results.

Recent investigations on the impact fallow management on the diversity of the microbial community and the consequences of these modi­fications on soil organic decomposition function were carried out in a Lixisol (Senegal) (Sall et al., in press). Soil samples (010 cm) taken from a 21 year fallow and a plot that had been cultivated for 4 years after lying fallow for 17 years were incubated with or without the addition of Faidherbia albida litter under laboratory conditions (28ºC, 100% WHC) for 240 hours. Microbial diversity was assessed by molecular techniques (Denaturing Gel Gradient Electrophoresis) and in situ catabolic potential (ISCP) (Degens et al., 2000). In the non-amended soil, the activity of microorganisms was greater in the fallow soil, which had a greater microbial diversity than that in the cultivated soil. However, other soil properties (carbon and organic nitrogen content, total microbial biomass) may also explain this result. For the amended soil, only the first 8 hours of incubation showed a difference between the fallow and cultivated soil. During this period, the CO2 respiration in the fallow soil was higher than that recorded in the cultivated soil. This difference should be compared with the catabolic microbial diversity, which was higher in the fallow soil than in the cultivated soil. After this initial phase, the microbial community in the cultivated soil seemed to acquire similar functions to those in the fallow soil. These results show that the changes made to the microbial community by cultivation of a fallow over 4 years are not irreversible. The microbial community of this sandy soil very quickly recovers the same catabolic functions as those of the community in the fallow soil.

Effects of nematodes on microbial communities in tropical sandy soils

Nematodes can strongly affect microbial communities. In a microcosm experiment on a sandy soil (9.1%) from Senegal, the presence of bacterial feeding nematodes (Zeldia punctata or Acrobeloides nanus or Cephalobus pseudoparvus) led to a mean increase (+12%) in maize biomass compared to control soils and reduced concentrations of soil ammonium by the end of the experiment (50 days). Moreover bacterial feeding nematode activity led to a significant decrease in microbial biomass (-28%) and density of cultivable bacteria (-55%), however, nematodes stimulated bacterial activity (+18%) (Djigal et al., 2004).

Spatial distribution of biotic components

The distribution of organisms throughout the soil is controlled by the concentration in their substrates (Gray and Williams, 1971), soil water regime (Griffin, 1981), and soil structure (Elliott and Coleman, 1988; Hattori, 1988). Therefore any factors that modify these properties are likely to change the abundance and the activity of soil organisms.

Impact of termite biogenic structures on microbial abundance and diversity

In sub-sahelian sandy soils, termites are the only macrofauna actors during the dry season which last more than 7 months per year. Their activity translates mainly into the production of biogenic structures of various nature, size and constitution: mounds, soil sheeting, galleries and nest chambers. These soil translations are ecologically significant: in Senegal, 675 to 950 kg.ha^-1 of soil are moved on the surface in the form of sheetings and galleries (Lepage, 1974). In Kenya, soil translation exceeds 1,000 kg.ha^-1 (Kooyman & Onk, 1987). In the desert ecosystem of Chihuahua, about 2,600 kg.ha^-1 are transformed annually into sheetings (Mackay & Whitford, 1988). These foraging structures, aside from their quantitative importance, present physicochemical, enzymatic and microbiological characteristics, which do not only differ from the control soil but also reflect the diversity of the organisms that produced them (Seugé et al., 1999; Fall et al., 2001; Sall et al., 2002; Mora et al., 2003). Thus, in this ecological context characterized by a relative stability of edaphic factors (temperature, humidity, soil structure), termites represent one of the main factor governing the activity and diversity of the microbial community.

An experiment realized in Senegal (soil with 1,013% clay) demonstrated that the impact of termites on soil properties depends on their biotic affiliation (soil feeding vs fungus growing) (Fall et al., 2001; Sall et al., 2002) and the type of structure, i.e., soil sheeting or nest, produced (Ndiaye et al., 2004a, b). Soil sheeting produced by the two main fungus growing termite species in Senegal (Macrotermes subhyalanus and Odontotermes nilensis) are characterized by an increased in organic C and mineral N, resulting in an increased in soil respiration whereas the microbial biomass was unchanged (Ndiaye et al., 2004a) and the enzymatic activities were weaker than in soil (Brauman, 2002). Interestingly, these soil structures harbour a very different population of nematodes (Villenave and 2005, submitted) and fungi (Diouf et al., 2005), which demonstrates the role of termite as soil engineers. These properties did not depend on the quality of the organic substrate recover by the termite sheeting. Interestingly, these biogenic structures could be considered as a phenotypic characteristic of the species, as a multivariate analysis of the physi-cochemical, biochemical and microbiological of biogenic structures allows the separation of structures produced by different species of termites and earthworms (Seuge, PhD Thesis).

As underlined before, the termite nest of the soil-feeding termite has very different characteristics. Nests of Cubitermes niokoloensis with 5 times more C, 7 to 15 more N and 4 times more carbohydrates (Sall et al., 2002) could be seen as hot spots of organic matter and nutrients compared to the poor surrounding savannah soil. Moreover, the microbial community of these nests seems less diverse and heavily dominated by actinomycetes (Fall et al., 2004, Fall et al. submitted). Regarding N dynamics, the nests of soil-feeding termites present a decrease in potential denitrification and an inhibition of potential nitrification with the surrounding soil (Ndiaye et al., 2004). We could underline that the low or absence of the nitrification process seems a general feature of termite structures (sheeting and nest), showing a deep impact of termite on the global nitrogen cycle. Such modifications lead to important increases in NH4 and NO3 contents in biogenic structures (100 times more mineral N in nests of C. niokoloensis than in the soil). The absence of nitrification in termite nests despite high nitrate contents remains not completely understood. Brauman et al. (2002, 2003) hypothesised a termite or actinomycete origin (production of bactericide) or an inhibition by phenolic compounds presents in the nest.

In conclusion, termite mounds like earthworm constitute, in the context of the sandy tropical soil characterized by an intense mineralization rate, site of SOM preservations. The results reinforce the view of biogenic structures as earthworms cast and termite’s nest as true soil functional compartments like the rhizosphere.

Impact of soil structure

Soil is composed of an assemblage of solid particles and voids and represents the most complex habitat for organisms. Many authors have examined the effects of soil structure on the distribution and activities of the soil biota, including work on the distribution of soil microorganisms in particle-size fractions (Elliott, 1986; Gupta and Germida, 1988; Hattori, 1988; Kabir et al., 1994) and soil porosity (Killham et al., 1993). Much of the difficulty in studying the relationships between soil structure and soil microbial distribution and activity is based on our lack of knowledge of microorganims in undisturbed soil habitats. Therefore a gentle physical soil fractionation method based on a slaking procedure was developed and adapted for sandy soils (Chotte et al., 1993, 2002). This method has been used to describe the distribution of nematodes and microorganisms as part of a broader programme dealing with the impact of fallow shortening on soil fertility and biofunctioning.

Distribution of the nematode community within pores versus aggregates

Very few studies deal with the location in soil and activity of free living and plant parasitic nematodes. In the soil (14% clay) of Thyssé-Kaymor (Senegal), the repartition of nematodes in different soil fractions (aggregates >200 µm), inter-aggregates pores, fresh organic matter) vary according to their trophic behaviour (Figure 11) (Villenave, unpublished data).

Figure 11. Distribution of the different feeding groups of soil nematodes between soil fractions (in % of the total nematode number in the soil sample)

Bacterial-feeding nematodes were essentially localized in inter-aggregate pores (>50%) and an important proportion of these nematodes was localized in fresh organic matter (24%). A relatively similar distribution was observed for fungal-feeding nematodes.

The other trophic groups presented slightly different distributions: plant-feeders had more than 50% of their total number in aggregates >200 µm. Predators were essentially localized in inter-aggregate pores. The density of bacterial-feeding nematodes was 17 times higher in the outer part of soil aggregates (e.g. in inter-aggregate pores and in fresh organic matter per g dry soil) than in the inner part.

In a sandy soil (17% clay) nematode activity (at a density of about 10 bacterivorous Cephalobidae per gram of dry soil during 21 days) led to modi­fications of the structure of the microbial community of the outer part of the soil (macroporosity) whereas changes were not significant at the scale of the total soil. Nematodes mainly and directly affected bacteria present in their influence area. In a clayey soil, the proportion of bacteria physically protected from nematodes is higher than in a sandy soil; so the influence of these organisms on the whole microbial community might be lower than in sandy soil.

Distribution of microbial community within soil aggregates

The distribution of the microbial community within soil aggregates has been investigated in different fallow situations in order to test the impact i) of soil structure on microbial abundance and diversity, and ii) of fallow management. Theses studies have been carried out in a Lixisol (Senegal) (Chotte & Jocteur-Monrozier, 1999; Chotte et al., 2002). These investigations indicated that long-term fallow (19 y) under Pennisetum was found to stimulate aggregation, while all clay particles were easily dispersed from the 3 y fallow soil. Hot spots of potential N2 fixation (Acetylene Reduction Activity, ARA) were observed in coarse soil fractions (>50 µm), suggesting that these microhabitats were favourable to active N2 fixers. In contrast, more than 70% of the N2 fixing micro­organisms and 90% of the recovered Azospirillum were isolated from the dispersible clay fraction (0-2 µm). The reduction of the fallow period was responsible for the decrease of the amount of nitrogen potentially fixed by free-living bacteria. This was not due to the diminution of their abundance but to fact that environmental conditions favourable to their activity are not at their best in young fallow soil (lack of macro aggregates >2,000 µm). Diversity of Azospirillum species was assessed by hybridization with specific genetic probes on colonies within each fraction. This approach revealed the abundance of A. irakense in the 3 y fallow soil fractions only and a selective effect of fallow on A. brasilense/A. amazonense genomic species in the 19 y fallow soil. Similar works compared the distribution of cellulolytic bacteria. These bacteria, mostly represented by nonfilamentous cells, were mainly located within the organic residues (24% of the total number) and the silt-size aggregates (2-50 µm) (58%).

These studies clearly reveal that the changes of microbial communities as a result of modifications of land uses would have remain hidden if the investigation had been restricted to the non-fractionated soil. Current studies indicate that land management could have a deep impact of the functional diversity (denitrifier community) depending on the location in the different aggregate size fractions (Assigbetsé, personal com.). Further studies are needed to measure the consequence of the modifications in term of N20 fluxes, and the processes responsible for them.

Nitrogen mineralization in tropical sandy soils

In sandy soils, the evolution of mineral N during wet season can be divided into two main phases. The first phase is characterised by a significant net mineralization called nitrogen flush (on average 58 kg.ha^-1 on 1 m in Centre of Senegal); during this period (about 20 days) the net nitrification is also significant and it favours N losses by leaching originally largely of the acidification⁴. The second phase is characterised by a net mineralization and a very low to non-existent nitrifying activity (Blondel, 1971a, b, c). During this phase, the plant modifies the equilibrium by increasing mineralization when the mineral N contents of soil are low and promoting immobilization when these contents are high (Blondel, 1971d; Reydellet et al., 1997). The microbial biomass (BM-C) expressed as a percent of total soil organic C was higher than in temperate soil. The BM-C increased during rainy season. This might be a key factor in nitrogen flush at onset of rainy season in dry tropical areas, which is essential for installation of crop (Niane-Badiane et al., 1999).

Like for temperate agrosystems, plant N nutrition relies on soil organic stock, since most of N taken up by plants derives from N organic stock, even in fertilized plots (Niane-Badiane et al., 1999). Therefore, several studies have been targeted toward the manipulation of inorganic N fluxes through the management of organic resources at the field scale. The dynamics and the extent to which organic components decompose depend on soil characteristics and substrate quality. Quality of organic residues can be assessed by C to N ratio (Giller & Cadisch 1997), N content (Vigil & Kissel 1991), soluble-C content (Reinersten et al. 1984), lignin content (Berg 1986), lignin-to-N (Vigil & Kissel 1991), polyphenol-to-N (Palm and Sanchez 1991), and (polyphenol plus lignin)-to-N (Constantinides & Fownes 1994) ratios. Several studies have been carried out in semi-arid zones of West Africa (Senegal, Burkina Faso) to determine the impact of various litters on mineralization processes. Soil nitrogen mineralization patterns were investigated under field conditions in the presence of five leaf litters of different qualities, Faidherbia albida A. Chev., Azadirachta indica A. Juss., Andropogon gayanus.

Kunth., Casuarina equisetifolia forsk., and Eragrostis tremula Steud (Diallo et al., 2005). Any relationship could be drawn between litter quality (N content, cellulose, hemicellulose, lignin) and N mineralization during a mid-term field experimentation (12 months). In the presence of these litters, the concentration of inorganic N was higher than that in the control plot (without litter amendment). When comparing the inorganic N pattern in C. equisetifolia and F. albida amended soils, a higher inorganic N was measured in soil amended with C. equisetifolia despite the fact that F. albida had the lowest C to N ratio (21.4). The processes were then investigated during a 60 days laboratory incubation to compare the effect of Andropogon gayanus, Casuarina equisetifolia, Faidherbia albida on C and N dynamics in the presence or not of a source of inorganic N (Sall et al., 2003). The results indicated that during the first stage of incubation, CO2-C evolved was significantly correlated with the soluble C content of the litter. The pattern of soil inorganic N varied according to the litter quality. However, a similar immobilisation was obtained in soil amended with Andropogon gayanus and Casuarina equisetifolia, despite the fact that these materials have very different C:N ratios (51, and 35, respectively). The abundance of polyphenols in the Casuarina equisetifolia litter may explain this result. In fact, several studies have mentioned the negative effect of polyphenols on N mineralization processes (Palm and Sanchez 1991). The addition of inorganic N modified the patterns of CO2-C respiration and net N immobilization. The magnitude of these modifications varied according to the litter quality.

These studies indicated that the management of organic resources could be view as a means to modify N fluxes (and CO2) in sandy soils. However the definition of an accurate indicator to predict the decomposition of organic residues can not be based on a single parameter. It should take into account several litter characteristics (e.g. ratio of soluble C to phenol content, etc.). Moreover, the impact of the characteristics of the organic constituents on the gross CO2-C and inorganic N fluxes and on the diversity and function of soil microorganisms must be addressed.

Conclusion

Productivity of ecosystems characterized by sandy soils is generally low because of erratic rainfall pattern and soil texture; this results in a poor nutrient availability and unstable structure (Pieri, 1992). Studies on soil fertility, SOM dynamics and soil biofunctioning in sandy soils of West Africa, as presented above, emphasized the importance of coarse plant debris and soil biota in controlling most of physical, chemical and biological soil properties. The essential of the beneficial effect of organic management of soil fertility by fallows and manures is based on mineralization processes rather than on humification ones; this means that SOM content is a questionable indicator of the fertility of sandy soils and of the sustainability of agrosystems (Feller, 1995b). The response of biota to sandy soil constraints is a control of soil stability and porosity (perennial rooting systems, fauna, micro­organisms), a conservative management of inputs protected either in root biomass or in stable organic compounds (Menaut et al., 1985; Izac & Swift, 1994; Chotte et al., 1995; Giller et al., 1997). In sandy soils, biological mechanisms play a crucial role on processes driving plant nutrition. This has three implications:

• A particular attention should be given to SOM and biological processes. Studies confirm that in sandy soils, the coarse organic fraction (>50 µm) is the main relevant one (Feller et al., 2001).
• Soil biological components should be characterized and root component should be included in SOM total.
• Sandy soils should be seen as living and dynamic milieux.

As a consequence, SOC losses linked to biological activities (fauna and microorganisms) is the price to pay to maintain suitable soil organization and functioning (Perry et al., 1989; Manlay et al., 2002c).

As a consequence, cropping alternatives should take into account the traditional functions of fallows, i.e., biomass production and increase in biological diversity and activity (Feller et al., 1990; Pate, 1997; Manlay et al., 2000).

From a C sequestration point of view, although sandy soils have a poor potential of C storage, it seems possible to double C stocks in cropping systems through the integration of a tree component in culture and to use mineral fertilizers in order to stabilize SOM (Woomer et al., 1998). It seems also necessary to provide more important incomes to populations through intensified agrosystems; this would limit the need for other soils whose C storage potential is more important (more clayey soils, more humid zones). In West African savannas, as long term fallows are hard to achieve and as crop residues are often exported (fuel, building materials, cattle food), solutions could be rotations of crops with strong rooting systems, improved short-term fallows or agroforestry systems. Other practices such as hay-making, cover crops, slash-and-mulch, compost, no-till or integration of livestock could also be successful to increase or maintain C stocks and to make systems sustainable (Vierich & Stoop, 1990; Manlay et al., 2002c). Also, to increase SOC stocks, one can either increase C inputs or decrease SOM biodegradation processes. The first method consists in providing prehumified OM (composted manures), or to manage the quantity and quality of residues. The second method is to protect the soil with cover plants. The quality of SOM in an essential determinant of C storage (Feller & Ganry, 1982).

To limit fertility deterioration by acidification and allelopathy, appropriate cultural practices must be applied: varieties and agricultural profile (dense and deeply penetrating root system must be improved), crop rotation (monoculture is a poor practice; it neither improves the SOM balance nor sustains crop yield) and sowing date (early sowing date in semi-arid zone), and organic material applied (manure or compost, root residues possess the desirable quality).

References

Asadu, C.L.A.; Diels, J.; Vanlauwe, B. 1997. A comparison of the contributions of clay, silt, and organic matter to the effective CEC of soils of sub-Saharan Africa. Soil Science 162: 785-794.

Azontonde, A.; Feller, C.; Ganry, F.; Rémy, J.C. 1998. Le mucuna et la restauration des propriétés d’un sol ferrallitique au sud du Bénin. Agriculture et Développement 18: 55-62.

Barthès, B.; Azontonde, A.; Blanchart, E.; Girardin, C.; Villenave, C.; Lesaint, S.; Oliver, R.; Feller, C., 2004. Effect of a legume cover crop (Mucuna pruriens var. utilis) on soil carbon in an Ultisol under maize cultivation in Southern Benin. Soil Use and Management 20: 231-239.

Batjes, N.H. 1996. Total carbon and nitrogen in the soils of the world. European Journal of Soil Science 47: 151-163.

Bayer, C.; Martin-Neto, L.; Mielniczuk, J.; Pillon, C.N.; Sangoi, L. 2001. Changes in organic matter fractions under subtropical no-till cropping systems. Soil Science Society of America Journal 65: 14731478.

Berg, B. 1986. Nutrient release from litter and humus in coniferous forest soils – a mini review. Scandinavian Journal Forest Research 1: 359-369.

Blanchart, E. 1992. Role of earthworms in the restoration of the macroaggregate structure of a de-structured savanna soil under field conditions. Soil Biology and Biochemistry 24: 1587-1594.

Blanchart, E.; Bruand, A.; Lavelle, P. 1993. The physical structure of casts of Millsonia anomala (Oligochaeta: Megascolecidae) in shrub savanna soils (Côte d’Ivoire). Geoderma 56: 119-132.

Blanchart, E.; Lavelle, P.; Braudeau, E.; Le Bissonnais, Y.; Valentin, C. 1997. Regulation of soil structure by geophagous earthworm activities in humid savannas of Côte d’Ivoire. Soil Biology and Biochemistry 29: 431-439.

Blondel, D. 1971a. Contribution à la connaissance de la dynamique de l’azote minéral: en sol sableux au Sénégal. Agronomie Tropicale 26: 1303-1333.

Blondel, D. 1971b. Contribution à la connaissance de la dynamique de l’azote minéral: en sol ferrugineux tropical à Séfa. Agronomie Tropicale 26: 1334-1353.

Blondel, D. 1971c. Contribution à la connaissance de la dynamique de l’azote minéral: en sol ferrugineux tropical à Nioro du Rip. Agronomie Tropicale 26: 1354-1361.

Blondel, D. 1971d. Rôle de la plante dans l’orientation de la dynamique de l’azote minéral en sol sableux. Agronomie Tropicale 26: 1362-1371.

Blondel, D. 1971e. Rôle de la matière organique libre dans la minéralisation en sol sableux, relation avec l’alimentation azotée du mil. Agronomie Trop.icale 26: 1372-1377.

Boissezon, P. de 1973. Les matières organiques des sols ferrallitiques. In: Boissezon P. de, Moureaux C., Bocquel G. & Bachelier G. (eds.) Les sols ferralliti-ques, tome 4, IDT 21, ORSTOM, Paris, pp. 9-66.

Brauman, A. 2000. Effect of gut transit and mound deposit on soil organic matter transformations in the soil feeding termite: a review. European Journal of Soil Biology 36: 1-9.

Brauman, A.; Rouland, C.; Ndiaye, D.; Mora, P.; Villenave, C.; Chotte, J.L.; Duponnois, R.; Diouf, M.; Seuge, C.; Guisse, A.; Lepage, M. 2003. Typologie des structures biogéniques des termites, conséquence sur les flux d’azote dans les écosystèmes naturels et modifiés. In: Actes du colloque de l’Institut Français de la Biodiversité, Paris, 17-19 Mars 2003.

Brown, M.F.; Whitford, W.G. 2003. The effect of termites and straw mulch on soil nitrogen in a creosotebush (Larrea tridentate) dominated Chihuahuan desert ecosystem. Journal of Arid Environments 53: 15-20.

Buresh, R.J.; Tian, G. 1997. Soil improvement by trees in sub-Saharan Africa. Agroforestry Systems 38: 51-76.

Cadet, P. ; Spaull, V.W. 1998. Studies on the relationship between nematodes and sugarcane in South and West Africa: plant cane. Revue de Nématologie 8: 131-142.

Chotte, J.L.; Villemin, G.; Guilloré, G.; Jocteur-Monrozier, L. 1993. Morphological aspects of microorganism habitats in a vertisol. In: International Workshop on Soil Micromorphology (Townsville, Australia, 12-17 July 1992), Elsevier, Amsterdam, The Netherland, pp. 395-403.

Chotte, J.L.; Lavelle, P.; Blanchart, E. 1995. Gestion durable des terres en milieu tropical. Régulation biologique des processus de décomposition de la matière organique. In: Ganry F. & Campbell B. (eds.), Sustainable Land management in African Semi-Arid and Subhumid Regions. Proceedings of the SCOPE Workshop, Dakar, Senegal, CIRAD, pp. 89-97.

Chotte, J.L.; Jocteur-Monrozier, L. 1999. Les habitats microbiens des sols en jachère. Cas de sols ferrugineux tropicaux sableux du Sénégal. Séminaire “La Jachère en Afrique tropicale”, 13-16 avril 1999, Dakar.

Chotte, J.L.; Schwartzman, A.; Bailly, R.; Jocteur-Monrozier, L. 2002. Changes in bacterial communities and Azospirillum diversity in a tropical soil under 3 yr and 19 yr natural fallow assessed by soil fractionation. Soil Biology and Biochemistry 34: 1083-1092.

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

Degens, B.P.; Shipper, L.A.; Sparling, G.P.; Vojvodic-Vukovic, M. 2000. Decrease in organic C reserves in soils can reduce the catabolic diversity of soil microbial communities. Soil Biology and Biochemistry 32: 189-196.

Diallo, M.D.; Guissé, A.; Badian-Niane, A.; Sall, S. ; Chotte, J.L. 2005. In situ effect of some tropical litters on N mineralization. Arid Land Reasearch and Management 19: 1-9.

Diouf, M.; Brauman, A.; Miambi, E.; Rouland-Lefèvre C. 2005. Fungal communities of the foraging soil sheeting built by several fungus-growing termite species in a dry savanna. Sociobiology 45: 16-32.

Djigal, D.; Brauman, A.; Diop, A.; Chotte, J.L.; Villenave, C. 2004. Influence of some bacterial-feeding nematodes (Cephalobidae) on soil microbial community during maize growth. Soil Biology and Biochemistry 36: 323-331.

Elliott, E.T. 1986. Aggregate structure and carbon, nitrogen, and phosphorus in native and cultivated soils. Soil Science Society of America Journal 50: 627-633.

Elliott, E.T.; Coleman, D.C. 1988. Let the soil work for us. Ecological Bulletins 39: 23-32.

Ewel, J.J. 1999. Natural systems as models for the design of sustainable systems of land use. Agroforestry Systems 45: 1-21.

Fall, S. 1998. Impact de deux espèces de termites à régime alimentaire différencié sur la matière organique et le compartiment microbien de termitières. MSc thesis, UCAD, Dakar, 86 pp.

Fall, S.; Brauman, A.; Chotte, J.L. 2001. Comparative distribution of organic matter in particle and aggregate size fractions in the mounds of termites with different feeding habits in Senegal: Cubitermes niokoloensis and Macrotermes bellicosus. Applied Soil Ecology 17: 131-140.

Fall S.; Nazaret S.; Chotte J.L.; Brauman, A. 2004. Cell density and genetic structure of microbial community at the microenvironment level in a soil feeding (Cubitermes niokoloensis) termite’s mound as determined by enumeration and automated ribosomal intergenic spacer analysis fingerprints. Microbial Ecology 48: 191-199.

FAO-UNESCO 1995. Digital soil map of the World and derived properties. CD, version 3.5., November 1995. Rome, Italy.

Feller, C. 1979. Une méthode de fractionnement granulométrique de la matière organique des sols. Application aux sols tropicaux à textures grossières, très pauvres en humus. Cahiers ORSTOM, série Pédologie 17: 339-346.

Feller, C. 1995a. La matière organique dans les sols tropicaux à argile 1:1. Recherche de compartiments fonctionnels. Une approche granulométrique. Collection TDM, vol. 144, ORSTOM, Paris.

Feller, C. 1995b. La matière organique du sol: un indicateur de la fertilité. Application aux zones sahélienne et soudanienne. Agriculture et Développement 8: 35-41.

Feller, C.; Ganry, F. 1982. Décomposition et humification des résidus végétaux dans un agrosystème tropical. 3 – Effet du compostage et de l’enfouissement de divers résidus de récolte sur la répartition de la matière organique dans différents compartiments d’un sol sableux. Agronomie Tropicale 27: 262-269.

Feller, C.; Chopart, J.L.; Dancette, F. 1987. Effet de divers modes de restitution de pailles de mil sur le niveau et la nature du stock organique dans deux sols sableux tropicaux (Sénégal). Cahiers ORSTOM, Série Pédologie 23: 237-252.

Feller, C.; Albrecht, A.; Brossard, M.; Chotte, J.L.; Cadet, P. ; et al. 1990. Effets de différents systèmes de culture paysans sur quelques propriétés des sols et relations sol-plante dans la zone des Petites Antilles. In: SACAD (ed.) «Agricultures paysannes et développement: Caraïbe – Amérique Tropicale», Université des Antilles et de la Guyane, pp. 165-190.

Feller, C.; Fritsch, E.; Poss, R.; Valentin, C. 1991. Effet de la texture sur le stockage et la dynamique des matières organiques dans quelques sols ferrugineux et ferrallitiques (Afrique de l’Ouest en particulier). Cahiers ORSTOM, série Pédologie 26: 25-36.

Feller, C.; Beare, M.H. 1997. Physical control of soil organic matter dynamics in the tropics. Geoderma 79: 69-116.

Feller, C.; Balesdent, J.; Nicolardot, B.; Cerri, C.C. 2001. Approaching «functional» soil organic matter pools through particle-size fractionation. Examples for tropical soils. In Lal R., Kimble J.M., Follett R.F. & Stewart B.A. (eds) Assessment methods for soil carbon, Lewis Publishers, Boca Raton, pp. 53-67.

Feller, C.; Six, J.; Razafimbelo, T.; Chevallier, T.; De Luca, E.; Harmand, J.M. 2002. Relevance of organic matter forms associated to particle size fractions for studying efficiency of soil carbon sequestration. Examples for tropical agro-ecosystems. 17^th World Congress of Soil Science (Symposium 5), Bangkok, Thailand, 14-20 August, 2002 (unpublished results).

Floret, C. (Ed.) (1998) Raccourcissement du temps de jachère, biodiversité et développement durable en Afrique centrale (Cameroun) et en Afrique de l’Ouest (Mali, Sénégal). Final Report, European Community Commission (DG XII), Contract TS3CT93-0220, IRD, Paris, 245 pp.

Friesen, D.K.; Rao, I.M.; Thomas, R.J.; Oberson, A.; Sanz, J.I. 1997. Phosphorus acquisition and cycling in crop and pasture systems in low fertility tropical soils. Plant and Soil 196: 289-294.

Ganry, F. 1991.Valorisation des résidus de récolte à la ferme et maintien de la fertilité. Cas du Sud Sénégal. In: CIRAD (ed.), Savanes d’Afrique, Terres fertiles?, Montpellier, pp. 317-331.

Ganry, F.; Feller, C.; Harmand, J.M.; Guibert, H. 2001. The management of soil organic matter in semi-arid Africa for annual cropping systems. Nutrient Cycling in Agro-ecosystems 61: 103-118.

Giller, K.E.; Cadisch, G. 1997. Driven by nature: A sense of arrival or departure? In: Driven by nature. Plant Litter Quality and Decomposition. Cadisch G. & Giller K.E. (eds.) CAB International, Wallingford, UK, pp. 393-399.

Giller, K.E.; Beare, M.H.; Lavelle, P. ; Izac, A.M.; Swift, M.J. 1997. Agricultural intensification, soil biodiversity and agro-ecosystem function. Applied Soil Ecology 6: 3-16.

Gray, T.R.G.; Williams, S.T. 1971. Microbial productivity in soils. Genetics Microbiology Society Symposium, pp. 255-286.

Griffin, D.M. 1981. Water potential as a selective factors in the microbial ecology of soils. In: Parr J.F., Gardner W.R. & Elliott L.F. (eds) Water Potential Relations in Soil Microbiology, SSSA Special Publication Nº9, Madison, pp. 141-151.

Guibert, H.; Fallavier, P.; Romero, J.J. 1999. Carbon content in soil particle size and consequence on cation exchange capacity of Alfisols. Communications in Soil Science and Plant Analysis 30: 17-18.

Gupta, V.V.S.R.; Germida, J.J. 1988. Distribution of microbial biomass and its activity in different soil aggregate size classes as affected by cultivation. Soil Biology and Biochemistry 20: 777-786.

Harmand, J.M.; Njiti, C.F.; Bernard-Reversat, F.; Feller, C.; Oliver, R. 2000. Variations de stock de carbone dans le sol au cours du cycle jachère arboréeculture. Zone soudanienne du Cameroun. In: Floret C., Pontanier R. (eds.) «La jachère en Afrique tropicale», J. Libbey, Eurotext, Paris, pp. 706-713.

Hattori, T. 1988. Soil aggregates as microhabitats of microorganisms. Biology and Fertility of Soils 6: 189-203.

Hien, E. 2004. Dynamique du carbone dans un Acrisol ferrique du Centre Ouest Burkina: Influence des pratiques culturales sur le stock et la qualité de la matière organique. Thèse de doctorat de l’Ecole Nationale Supérieure d’Agronomie de Montpellier, 138 pages.

Izac, A.M.; Swift, M. 1994. On agricultural sustainability and its measurement in small-scale farming in sub-Saharan Africa. Ecological Economics 11: 105-125.

Jones, M.J. 1973. The organic matter content of the savanna soils of West Africa. Journal of Soil Science 24: 42-53.

Jones, C.G.; Lawton, J.H.; Shachak, M. 1994. Organisms as ecosystem engineers. Oikos 69: 373-386.

Jouquet, P.; Ranjard, L.; Lepage, M.; Lata, J.C. 2005. Incidence of fungus-growing termites (Isoptera, Macrotermitinae) on the structure of soil microbial communities. Soil Biology and Biochemistry 37: 1910-1917.

Kabir, M.D.M.; Chotte, J.L.; Rahman, M.; Bally, R.; Jocteur-Monrozier, L. 1994. Distribution of soil fractions and location fo soil bacteria in a vertisol under cultivation and perenial grass. Plant and Soil 163: 243-255.

Killham, K.; Amato, M.; Ladd, J.N. 1993. Effect of substrate location in soil and soil pore-water regime on carbon turnover. Soil Biology and Biochemistry 25: 57-62.

Kooyman, C.; Onck, R.F.M. 1987. Distribution of termite (Isoptera) species in Southwestern Kenya in relation to land use and the morphology of their galleries. Biology and Fertility of Soils 3: 69-73.

Lavelle, P. 1978. Les vers de terre de la savane de Lamto (Côte d’Ivoire): peuplements, populations et fonctions dans l’écosystème. Thèse de doctorat d’état, Université de Paris VI. Publication du laboratoire de Zoologie de l’ENS 12, 310 pp.

Lavelle, P. 1983. The soil fauna of tropical savannas. II. The earthworms. In: Bourlière F. (Ed.) Tropical savannas. Elsevier Scientific Publishing Co, Amsterdam, pp. 485-504.

Lavelle, P. 1997. Faunal activities and soil processes: adaptive strategies that determine ecosystem function. In: Begon M. & Fitter A.H. (eds.), Advances in Ecological Research 27: 93-132.

Lavelle, P. ; Pashanasi, B.; Charpentier, F.; Gilot, C.; Rossi, J.P.; Derouard, L.; André, J.; Ponge, J.F.; Bernier, N. 1998. Large-scale effects of earthworms on soil organic matter and nutrient dynamics. In: Edwards C.A. (ed.) Earthworm Ecology, St. Lucie Press, Columbus, Ohio, pp. 103-122.

Lavelle, P.; Spain, A.V. 2001. Soil Ecology. Kluwer Academic Publishers, Dordrecht, The Netherlands, 654 pages.

Léonard, J.; Perrier, E.; Rajot, J.L. 2004. Biological macro-pores effect on runoff and infiltration: a combined experimental and modelling approach. Agriculture, Ecosystems and Environment 104: 287-302.

Lepage, M. 1974. Les termites d’une savane sahélienne (Ferlo septentrional, Sénégal): peuplement, popula­tions, consommation, rôle dans l’écosystème. PhD thesis, Université de Dijon, 344 pp.

Maass, J.M.1995. Conversion of tropical dry forest to pasture and agriculture. In: Bullock S.H., Mooney H.A. & Medina E. (eds.) Seasonally dry tropical forests. Cambridge University Press, Cambridge, pp. 399-422.

Mackay, W.P.; Whitford, W.G. 1988. Spatial variability of termite gallery production in Chihuahuan Desert plant communities. Sociobiology 4: 281.289.

Mando, A.; Stroosnijder, L.; Brussaard L. 1996. Effects of termites on infiltration into crusted soil. Geoderma 74: 107-113.

Manlay, R.J.; Cadet, P. ; Thioulouse, J.; Chotte J.L. 2000. Reltionships between abiotic and biotic soil properties during fallow periods in the sudanian zone of Senegal. Applied Soil Ecology 14: 89-101.

Manlay, R.; Kaïré, M.; Masse, D.; Chotte, J.L.; Ciornei, G.; Floret, C. 2002a. Carbon, nitrogen and phosphorus allocation in agro-ecosystems of a West African savanna. I. The plant component under semi­permanent cultivation. Agriculture, Ecosystems and Environment 88: 215-232.

Manlay, R.J.; Masse, D.; Chotte, J.L.; Feller, C.; Kaïré, M.; Fardoux, J.; Pontanier, R. 2002b. Carbon, nitrogen and phosphorus allocation in agro-ecosystems of a West African savanna. II. The soil component under semi-permanent cultivation. Agriculture, Ecosystems and Environment 88: 233-248.

Manlay, R.J.; Chotte, J.L.; Masse, D.; Laurent, J.Y.; Feller, C. 2002c. Carbon, nitrogen and phosphorus allocation in agro-ecosystems of a West African savanna. III. Plant and soil components under continuous cultivation. Agriculture, Ecosystems and Environment 88: 249-269.

Manlay, R.J.; Masse, D.; Chevallier, T.; Russell-Smith, A.; Friot, D.; Feller, C. 2004. Post-fallow decomposition of woody roots in the Western African savanna. Plant and Soil 260: 123-136.

Martin, A. 1991. Short- and long-term effect of the endogeic earthworm Millsonia anomala (Omodeo) (Megascolecidae, Oligochaeta) of a tropical savanna, on soil organic matter. Biology and Fertility of Soils 11: 234-238.

Martin, A.; Marinissen, J.C.Y. 1993. Biological and physico-chemical processes in excrements of soil animals. Geoderma 56: 331-347.

Menaut, J.C.; Barbault, R.; Lavelle, P. ; Lepage, M. 1985. African savannas biological systems of humification and mineralization. In: Tothill J.C., Mott J.J. (eds.) Ecology and management of the World’s savannas. Australian Academic Science, Canberra, pp. 14-33.

Mora, P.; Seugé, C.; Chotte, J.L.; Rouland, C. 2003. Physico-chemical typology of the biogenic structures of termites and earthworms: a comparative analysis. Biology and Fertility of Soils 37: 245-249.

Mora, P.; Miambi, E.; Jimenez, J.J.; Decaëns, T.; Rouland, C. 2005. Functional complement of biogenic structures produces by earthworms, termites and ants in the neotropical savannas. Soil Biology and Biochemistry 37: 1043-1048.

Ndiaye, D.; Lepage, M.; Brauman, A.; Duponnois, R. 2003. Impact of a soil feeding termite, Cubitermes niokoloensis on the symbiotic microflora and plant parasitic nematodes associated with a fallow leguminous plant Crotalaria ochroleuca. Biology and Fertility of Soils 37: 313-318.

Ndiaye, D.; Lensi, R.; Lepage, M.; Brauman, A. 2004a. The effect of the soil-feeding termite Cubitermes niokoloensis on soil microbial activity in a semi-arid savanna in West Africa. Plant and Soil 259: 277-286.

Ndiaye, D.; Lepage, M.; Sall, C.; Brauman, A. 2004b. Nitrogen transformations associated with termite biogenic structures in a dry savanna ecosystem. Plant and Soil 265: 189-196.

Ndour, Y.N.; Chotte, J.L.; Fardoux, J. 1999. Statut organique et microbiologique de sols ferrugineux tropicaux en jachère naturelle au Sénégal. Séminaire “La Jachère en Afrique tropicale”, 1316 avril 1999, Dakar.

Ndour, Y.N.; Chotte, J.L.; Pate, E.; Masse, D.; Rouland, C. 2001. Use of soil enzyme activities to monitor soil quality of natural and improved fallows in semi-arid tropical regions. Applied Soil Ecology 18: 229-238.

Niane-Badiane, A.; Ganry, F.; Jacquin, F. 1999. Les variations de la biomasse microbienne d’un sol cultivé au champ: conséquences sur la reserve organique mobilisable (cas d’un sol ferrugineux tropical au Sénégal). Comptes-Rendus de l’Académie des Sciences 328: 45-50.

Ouédraogo, E.; Mando, A.; Brussaard, L. 2004. Soil macrofaunal-mediated organic resource disappearance in semi-arid West Africa. Applied Soil Ecology 27: 259-267.

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

Pate, E. 1997. Analyse spatio-temporelle des peuplements de nématodes phytoparasites dans les systèmes de culture à jachère du Sénégal. PhD thesis, University Lyon 1, 208 pages.

Perry, D.A.; Amaranthus, M.P.; Borchers, J.G.; Brainerd, R.E. 1989. Bootstrapping in ecosystems. Bioscience 39: 230-237.

Pieri, C. 1992. Fertility of soils: a future for farming in the West African Savannah. Springer Series in Physical Environment, Springer-Verlag, Berlin, 348 pages.

Reinersten, S.A.; Elliott, L.F.; Cochran, V.L.; Campbell, G.S. 1984. Role of available carbon and nitrogen in determining the rate of wheat straw decomposition. Soil Biology and Biochemistry 16: 459-464.

Reydellet, I.; Laurent, F.; Oliver, R.; Siband, P. ; Ganry, F. 1997. Quantification par méthode isotopique de l’effet de la rhizosphère sur la minéralisation de l’azote (cas d’un sol ferrugineux tropical). Comptes-Rendus de l’Académie des Sciences 320: 843-847.

Rouland, C.; Lepage, M.; Chotte, J.L.; Diouf, M.; Ndiaye, D.; Ndiaye, S.; Seugé, C.; Brauman, A., 2003. Experimental manipulation of termites (Isoptera: Macrotermitinae) foraging pattern in a sahelo-sudanese savanna: effect of litter quality. Insectes sociaux 50: 1-8.

Ruiz, L.; Ganry, F.; Waneukem, V.; Siband, P.; Oliver, R. 1995. Recherche d’indicateurs de fertilité azotée des terres. Agriculture et Développement 5: 38-46.

Sall, S.; Brauman, A.; Fall, S.; Rouland, C.; Miambi, E.; Chotte, J.L. 2002. Variation in the distribution of monosaccharides in soil fractions in the mound of termites with different feeding habits (Senegal). Biology and Fertility of Soils 36: 232-239.

Sall, S.N.; Masse, D.; Bernhard-Reversat, F.; Guissé, A.; Chotte, J.L. 2003. Microbial activities during the early stage of laboratory decomposition of tropical leaf litters: the effect of interactions between litter quality and exogenous inorganic nitrogen. Biology and Fertility of Soils 39: 103-111.

Sall, S.N.; Masse, D.; Ndour, N.Y.B.; Chotte, J.L. Does cropping modify the decomposition function and the diversity of the soil microbial community of tropical fallow soil? Applied Soil Ecology, in press.

Sanchez, P.A.; Logan, T.J. 1992. Myths and science about the chemistry and fertility of soils in the tropics. In: Lal R. & Sanchez P.A.(eds.) Myths and Science of Soils of the Tropics, SSSA, Madison, Wisconsin, SSSA Special Publication 29: 35-46.

Seugé, C.; Rouland, C.; Fall, S.; Brauman, A.; Mora, P. 1999. Importance of earthworms’ casts and sheetings of some termite species in different fallows (Kolda, Sénégal). In: IRD, CIRAD (eds.), La jachère en Afrique Tropicale, vol. II, pp. 141-149.

Swift, M.J.; Heal, O.W.; Anderson, J.M. 1979. Decom­position in terrestrial ecosystems. Studies in Ecology, vol. 5. Blackwell Scientific Publications, Oxford, 372 pages.

Szott, L.T.; Palm, C.A.; Buresh, R.J. 1999. Ecosystem fertility and fallow function in the humid and subhumid tropics. Agroforestry Systems 47: 163-196.

Valentin, C.; Rajot, J.L.; Mitja, D. 2004. Responses of soil crusting, runoff and erosion to fallowing in the sub-humid and semi-arid regions of West Africa. Agriculture, Ecosystems and Environment 104: 287-302.

Venkatapen, C.; Blanchart, E.; Bernoux, M.; Burac, M. 2004. Déterminants des stocks de carbone dans les sols et spatialisation à l’échelle de la Martinique. Les Cahiers du PRAM 4: 35-36.

Vierich, H.I.D.; Stoop,W.A. 1990. Changes in West African Savanna agriculture in response to growing population and continuing low rainfall. Agriculture, Ecosystems and Environment 31: 115-132.

Vigil, M.F.; Kissel, D.E. 1991. Equations for estimating the amount of nitrogen mineralised from crop residues. Soil Science Society of America Journal 57: 66-72.

Villenave, C.; Charpentier, F.; Lavelle, P.; Feller, C.; Brussaard, L.; Pashanasi, B.; Barois, I.; Albrecht, A.; Patron, J.C. 1999. Effects of earthworms on soil organic matter and nutrient dynamics following earthworm inoculation in field experimental situations. In: Lavelle P., Brussaard L. & Hendrix P. (eds.) Earthworm management in tropical agro-ecosystems. CABI Publishing, pp. 173-197.

Villenave, C.; Cadet, P. 2000. Rôle particulier de Helicotylenchus dihystera au sein des peuplements de nématodes phytoparasites (Sénégal). In: Floret C. & Pontanier R. (eds.) La jachère en Afrique tropicale. Rôles, aménagements, alternatives, John Libbey, Paris pp. 291-299.

Woomer, P.L.; Palm, C.A.; Qureshi, J.N.; Kotto-Same, J. 1998. carbon sequestration and organic resource management in African smallholder agriculture. In: Lal R., Kimble J.M., Follett R.F. & Stewart B.A. (eds.) Management of carbon sequestration in soil, CRC Press, Boca Raton, pp. 153-173.

Zaady, E.; Groffman, P.M.; Shachak, M.; Wilby, A. 2003. Consumption and release of nitrogen by the harvester termite Anacanthotermes ubachi navas in the Northern Negev desert, Israel. Soil Biology and Biochemistry 35: 1299-1303.

Zvyagintsev, D.G. 1994. Vertical distribution of microbial communities in soils. In: Ritz K., Dighton J. & Giller K.E. (Eds.) Beyond the biomass. Blackwell Scientific Publications, Oxford, United Kingdom, pp. 29-37.

¹ IRD, UR 179, 911 Avenue Agropolis, BP 64501, 34394 Montpellier cedex 5, France, [email protected]
² IRD, UR 179, Route des Hydrocarbures, BP 1386, Dakar, Senegal
³ CIRAD, avenue Agropolis, TA 40/01, 34398 Montpellier cedex 5, France
⁴ ENGREF, BP 44494, 34093 Montpellier Cedex 5, France

On-farm assessment of long term effects of organic matter management on soil characteristics of paddy fields threatened by salinity in Northeast Thailand

Clermont-Dauphin, C.¹; C. Hartmann¹; J.L. Maeght¹; E. Beriaux² and C. Sagnansupyakorn³

  Keywords: Organic matter management, paddy, fertilizer, farmer management

Keywords: Organic matter management (OMM); On-farm research; interactions; total C; salinity; sandy soils

Introduction

The possibility of rapidly forecasting long-term effects of cropping systems on soil fertility is one of the main concerns of sustainable agriculture (Hansen, 1996). In the rainfed paddy systems of Northeastern Thailand, this question is very important since many farmers, in order to reduce the cost of land preparation, burn the rice straw remaining on the fields after harvest. Conversely, others go to much trouble to incorporate it, or even to add animal compost, being convinced that these practices will improve the fertility and prevent salinization of the soil (Grunberger, and Hartmann, 2004). A decrease in organic matter content of soils in the longterm could be very harmful since many farmers of this region are too poor to invest in mineral fertilizers. Hence, they often rely on the indigenous soil N supply to satisfy most of the rice crop’s demand (Olk et al., 2000; Powlson and Olk, 2000).

One of the difficulties of the assessment of these OMM in Northeast Thailand is the need to take into account the interactions of various cropping practices existing over the region (Olk et al., 2000; Powlson and Olk, 2000). Indeed, due to the higher yields associated with the transplanting method of establishing rice compared to direct seeding (Pandey et al., 2002), a higher input of carbon to soils via the residues can be expected. Therefore, effects of burning the straw on soil carbon evolution might be greater in fields where transplanting is practiced than in direct-seeded fields. Likewise, the use of inorganic fertilizer could interact with OMM effects, not only via the increase of residue inputs to soils but also through the rate of carbon mineralization (Whitbread et al., 2003, Shirato et al., 2005). As different fields elevation could be associated with varied water dynamics and salinity levels (Arunin, 1984, Bolomey 2002, Grunberger and Hartmann, 2004), the effects of this factor on inputs of carbon as well as loss of carbon by erosion and mineralization could also be expected. Considering all these interactions, and the lack of a carbon turnover model validated for these tropical cropping systems (Parton and Stanford, 1989), the assessment of OMM over the region could be investigated using an on-farm approach rather than by a long term “in station” experiment. This on-farm approach, based on comparison of fertility criteria over a network of farmers’ fields, requires that the current cropping systems have been practiced for at least five years, so that differences in the soil fertility components can be attributed to differences in the current cropping systems (Hasewaga et al., 2005). With such a network of fields, derivation of the parameters of a generic model of organic matter evolution could be carried out. However, as many fields have to be explored, basic or summary models using few parameters could be more successful than complex models.

Few data are available on the long-term effects of the OMM practiced by the farmers in Northeast Thailand. Experimental data sets exist for the cassava system developed in the uplands of the region, but not for the paddy system (Shirato et al., 2005). Tangtrakarnpong and Vityakon (2002), regarding the paddy system as relatively homogenous in comparison with other land use types existing in the region (forests, cassava, sugarcane), showed that carbon pools (labile and stable) of the paddy system ranked second behind forest systems, as total soil organic matter decreased from 5.5 in the forest to 4.2 g kg^-1 in the paddy system, and microbial biomass, which corresponds to the pools of higher turnover rate, decreased from 116 in the forest to 78 mg kg^-1 in the paddy system. Whitbread et al. (2003) examined the effects of the removal or otherwise of straw, and of leaf litter application on soil carbon in experimental plots. However, because it was mainly focused on the testing of new organic matter management, this study did not consider the possible interactions of the various current cropping practices and topography of fields at the level of the region. Moreover, the burning of straw was not assessed. Other studies pointed out the possible positive effect of organic matter on soil pH, due to increase of Fe reduction in the flooded soils (Maegth, 2003; Quantin et al., submitted). In accordance with this hypothesis, Enet (2003), by comparing two fields annually receiving (or not) animal compost amendments, showed positive effects of organic matter application on soil pH, measured during the cropping period. It was however likely that this effect of organic matter could be a short term effect, not noticeable during the following dry period.

The purpose of this paper was to assess briefly, using surveys and observations over a network of farmers’ fields, long-term effects on soil carbon, nutrient content, soil pH, salinity and bulk density of various OMM strategies encountered in the paddy systems of Northeast Thailand. Comparison of annual inputs of carbon as crop residues and manure with the measured post-harvest carbon content of soils was used to formulate hypotheses for the carbon dynamics of these sandy soils threatened by salinity.

Materials and Methods

Study area

The study was carried out during the dry season of December to April 2005 in Khon Kaen Province (16ºN 102ºE) in the Northeast of Thailand. The study area was chosen as representative of the natural conditions of the region. The landscape is a gently undulating plateau containing the villages of Ban Daeng, Ban Non Bo, Ban Kraduang and Ban Non Khlong (Figure 1). The soils are classified according to the USDA Soil Taxonomy mostly as Paleaquults, and sometimes Paleustult in the central area (Craig and Pisone, 1988). They are sandy loams with a low clay fraction, mostly kaolinitic, and a low pH due to considerable leaching. Because of the presence of shallow, saline groundwater, some parts of the area can have an electrical conductivity of the saturated extract in the soil surface as high as 7 mS cm^-1. The mean annual rainfall in this region in the 1990s was around 1,000 mm, with 90% of this falling between May and the end of October. Daily mean temperatures vary from 20 to 28ºC during the year. Temperature and rainfall recorded at the experimental site of IRD at Ban Daeng during the 2004 cropping season are representative of the region’s annual mean (Figure 2).

A preliminary study in this region showed that the farmers mainly grow rice. The rice is transplanted once a year during the rainy season, with nothing grown in the dry season. The soil preparation, consist of two plowings to 15 cm depth using a powered cultivator. The rice is either transplanted as month-old seedlings or directly sown by broadcasting just before the second plow. Direct seeding is not associated with the adoption of zero tillage in this region, probably because herbicides are not used. It is likely that the main difference between soil preparation for direct seeding and transplanting is the soil moisture during the second plow. In case of transplanting, soil moisture should be very high so that the soil structure will look like a uniform mud after plowing. The most common rice variety (RD6) is glutinous, photoperiodic sensitive with a flowering period in mid-October and harvest in November. The sowing date depends on the rainfall and on the method of planting. Whereas transplanting has to be done into flooded soils, direct seeding can be done at the beginning of the rainy season when the rainfall is still light. Whatever the method of sowing, the rates of chemical fertilizers used are often less than 60, 25, 25 kg ha^-1 of N P and K, respectively. Three main methods of organic matter management can be distinguished: 1) straw burning (SB), 2) straw incorporation (SI) or 3) straw incorporation and animal compost application (SI+C) before soil preparation. Application of animal compost is generally less than 1 t ha^-1.Potential yields in this area, estimated at 4 t ha^-1 (Suzuki et al., 2003), are often not reached.

Figure 1. Map of the study area. Localisation of the paddy fields selected

Figure 2. Monthly rainfall (bars) and average daily temperature (º) recorded in IRD site in Ban Daeng in 2004

Surveys

A network of 53 plots, annually cropped with rice, were selected to represent the three main OMM methods, which were factorially combined with 2 topographic positions (lowlands, corresponding to altitudes from 170 m to 190 m, and uplands – altitudes between 191 m and 210 m), two methods of sowing (transplanting and direct seeding) and two fertilizer levels (low for applications of less than 20 kg N ha^-1, and high for applications from 20 to 60 kg N ha^-1). Applications of P and K are often combined with those of N. The structure of the network is presented in Table 1. Each plot corresponded to a set of practices stable over the last five years. The soil total N, C, cations content, pH and salinity level were considered in this study as output variables whose variations within the network could be explained by the various OMM methods interacting with various cropping practices. Within each plot, the following measure­ments and observations were made:

Table 1. Distribution of the fields according to organic matter management and altitude. Values in parenthesis indicate percentage of those fields using transplanting method for rice cropping

1. Soil analyses at the end of February (post-harvest period). At this time, the rice straw was still standing, as only the panicles were harvested by hand. No organic matter application or straw burning had yet been done. This period is probably the most appropriate to bring out long term effects of practices, as the short term effects of the crop management interventions are the least likely. Soil samples were made up of 5 cores collected from the 0-15 cm layer. Total organic carbon (C) was determined by the wet digestion method of Walkley and Black. Total N was determined by the Kjeldahl method. pH was determined in distilled water using 1:1 soil:solution ratio. Cation exchange capacity (CEC) was determined after a first exchange with 1M ammonium acetate at pH = 7, and a second exchange with 1 M NH₄Cl. Exchangeable (exch) Ca, Mg and K extracted with 1 M ammonium acetate at pH7 were determined by atomic absorption spectrophotometry and flame photometry. Available P was analysed using the Bray P nº2 method. Electrical conductivity (EC) was measured using a 1/5 soil water ratio.
2. Measurements of the amount of rice straw remaining on the fields after harvest (three replicates of 4 m²).
3. Measurements of the amount of animal compost spread on the plots during the dry season, and records of farmers’ estimates of the amount spread during the last five years
4. Bulk density of the 0-15 cm ploughed soil layer (three replications using a 100 cm³ cylinder).

Data Analysis

Effects of position of the field in the landscape, OMM, method of sowing and N fertilizer level on soil contents of C, exchangeable cations and on soil pH, EC and bulk density were analysed by four-way analysis of variance using the GLM procedure and type III sum of squares of SAS 9.1 (2003). Interactions between cropping practices were also analysed statistically. The Student-Newman-Keuls test was used to compare the average values of the recorded variables.

The mineralization rate of the organic matter (K2), which is supposed to be very variable according to soil and climatic conditions and cropping systems (Mary and Guerif, 1994; Olk et al., 2000), was deduced according to the basic equation of Jenny (1941), which considers the soil carbon content as homogenous as regards its decomposition rate:

Assuming the soil carbon content was at equilibrium:

where m is the amount of annual organic dry matter supplied (mainly as straw residues), c_r the carbon content of the residues, and K1 the iso-humic coefficient which generally varies according to the composition of the amendment (Mary and Guerif, 1994). C₀ is the soil C content in the first year of the OMM application. C₁ is the total C content given by the soil analysis at the end of February 2004. As it has to be expressed in t ha^-1 rather than mass (kg kg^-1), the data recorded for soil bulk density are used. The annual C supplied by straw residue humification (m* c_r *K1) was calculated assuming values of 15% for water content. K1 of rice straw was fixed at 0.15 (Mary and Guerif, 1994), and carbon content of the humus at 50%. According to the observation that the most lignin rich parts of the residues are not affected by the burning, this practice was assumed to reduce the annual C input of straw by only 50%. A specific contribution of the rice roots to the total input of carbon to soils was ignored. Carbon supplied by the compost was considered as negligible, as the fresh weights of these amendments were less than 1 t ha^-1.

Results

Relationships between different soil criteria

The mean values and standard deviations of each of the studied variables and their mutual correlations are presented in Table 2. Carbon content was significantly positively correlated with soil CEC. As well as CEC, carbon content was significant correlated with exch. K, Ca, and Mg. Total C and N were highly correlated, with a C/N ratio of 11 that was quite stable among fields (not shown). No significant correlations between carbon content and pH, bulk density or EC were found. EC was significantly correlated with Na and Mg.

Changes in soil characteristics according to practices

Whereas altitude had a highly significant effect on soil exch. K and Na, EC and pH, OMM had no significant long-term effect on any of the measured variables. Significant effects of method of sowing appeared on available P and bulk density. Available P was lower in direct sowing as opposed to transplanting. Exch. K and bulk density were lower and soil pH higher in fertilized than in unfertilized fields (Table 3).

Calculation of K2 assuming organic matter equilibrium

The dry weight of the straw incorporated in the SI and SI + C fields varied from 2 to 7 t ha^-1. This variation is significantly associated with the method of sowing, but not to the elevation of the fields or to the use of fertilizer (not shown). The dry weight of residues in the direct-seeded fields was 0.70 t ha^-1 less than in the transplanted ones. The mean value of K2 allowing equality between annual incoming carbon and mineralized carbon was 3% in the SI fields. The same value might not apply to the SB fields under the equilibrium hypothesis: if it is assumed that about 50% of the carbon of the straw is lost due to burning, the mean K2 value of these situations should not be more than 2% (Figure 3).

Discussion

This on-farm survey was conducted to compare long-term effects of the various OMM of paddy fields existing in a region. These OMM are developed on sandy soils with a very low carbon content of 0.5% on average (Table 2). The C/N ratio of 11 suggests that the organic matter mineralizes readily in all these situations. About 20% of the CEC is due to the humus. Exch. K and Ca were very low, consistent with the low CEC. EC values were, at the time of the study, mostly below the level considered to pose a salinity problem for rice production (Dobermann and Fairhurst, 2000); however, these EC are significantly correlated with the Na content of the soils, revealing that they are due to the upward movement of saline groundwater. Due to the effects of farmers’ practices, variations in soil bulk density and pH were not significantly related to EC levels.

This study revealed that the observed differences in OMM did not result in differences in the soil C content, nor in pH, EC, exch. cations or bulk density. This non-significant effect on total C is consistent with results of Whitbread et al. (2003) who compared the removal and non-removal of rice straw during a six-year experiment. Unless the rice straw was returned in combination with the application of at least 50-14-14 kg N-P-K ha^-1, they did not find any significant difference in total C of soils among the treatments. Non-significant effects of the various OMM on exch. K and available P are in accord with the fact that most of the K and P contained in the fresh straw is still present in the ash. The non-significant effect of OMM on soil pH agrees the non-significant effect of OMM on the soil carbon content and the soil redox status, as the process of increasing pH found by Quantin et al., (submitted) and Maeght, (2003), was related to the increase of soil carbon content and the reducing conditions of flooded soils.

Table 2. Average values, standard deviations of measured variables and correlations between them. Bold values mean that the correlation is significant at the 0,05 probability level

Table 3. Soil chemical properties of 52 rainfed paddy fields of Khon Kaen as affected by altitude, organic matter management, crop establishment method and N fertilizer level

Figure 3. Values of annual output of carbon assuming mineralization rate of K2 = 3% for total soil C content

Comparison of the carbon inputs with the calculated outputs using the basic equation of Jenny (1941) at equilibrium (Figure 3), suggested that the non-significant effect of burning straw is due to a very low mineralization rate (K2) of the organic matter remaining in the soil after burning. Hence, the quality of this organic matter could be different from that of fields where the straw is incorporated every year. The average K2 of 3% is deduced for these situations, whereas the highest values mentioned in the literature for sandy soils in temperate zones reach 2% (Boiffin et al., 1986). Using the Jenny equation, this K2 value refers to the total organic matter content of the soils, with no distinction between more or less recalcitrant, or old or fresh organic materials. Therefore, it is certainly much lower than the mineralization rate that would be expected in such a tropical area for the incoming residues only (Shirato et al., 2005). It’s clear that it would have been more accurate to distinguish two different pools in the total C amount: labile carbon, and less quickly mineralized carbon. The pool of labile carbon could be approached either by the microbial biomass (Alvarez et al., 1998), or following Blair et al. (1995), by the KMnO4 extractable C fraction. Although Jenny’s very basic model of carbon turnover is not the appropriate tool to accurately predict the qualitative evolution of organic status of soils, it had for this study the advantage of referring to an easily measurable carbon pool and to allow preliminary comparisons between different cropping practices.

The variation in exch. Na, EC and pH over the network of farmers’ fields were mainly related to the position of the fields in the landscape. The higher values recorded for these variables in the lower lands were due to the proximity of the saline water table. The non-significant effect of topographic position on soil C content could mean that carbon movement by erosion is negligible. The effect of the method of sowing on available P (Table 3) could be due to a higher uptake of P by the recently harvested rice in the direct seeding method, whereas with transplanting most of the P would be fixed by the ferrous iron concentrated in the submerged soils during the cropping period. The lower soil bulk density in transplanted fields with fertilizer application could be due to the higher root volume developed by the recently harvested rice, or perhaps to a higher level of labile carbon in these soils. The lower exch. K in fertilized plots (Table 3) suggests that greater uptake of the initial soil K was possible by the rice crop. The higher pH recorded in fields receiving the higher fertilizer should be confirmed as acidification effect of urea application is more often observed.

Conclusion

Using a survey approach over a network of fields, it has been possible, within a short time, to examine some long-term effects of practices developed in paddy fields of Northeastern Thailand. More detailed data are in many cases still needed to draw definite conclusions. Future research should include assessment of labile carbon, as a more sensitive and early indicator of a change in the organic status of soils. Input of carbon by the roots, and leaching by drainage should be assessed for more accurate estimation of carbon budgets. Loss of carbon by the light burning practiced by the farmers should be measured. Estimation of nutrient uptake and root density of the rice with the various practices would allow our hypothesis to be used to explain variation in P and K availability and bulk density over the network. Besides the effects of farmers’ practices on the soil chemical and physical attributes, changes in incidence of weeds, diseases and insects on the rice crop should be studied. The present preliminary results call into question farmers’ assertions about the decrease in the extent of saline patches in their fields over time due to straw incorporation. Although the practice of burning straw should be avoided owing to air pollution, as regards soil fertility it does not seem to differ from the incorporation of straw. Therefore, positive short-term effects of straw incorporation on rice grain yield need to be large enough to persuade the farmer to accept its higher cost compared with burning.

Bibliography

Alvarez, C.R., Alvarez, R., Grigera M.S., Lavado R.S., 1998. Associations between organic matter fractions and the active soil microbial biomass Soil Biology and Biochemistry, Volume 30, Issue 6, 767-773.

Arunin, S., 1984. Characteristics and management of salt affected soils in the Northeast of Thailand. In “Ecology and management of problem soils in Asia ”. Food and Fertilizer Technology Center for the Asian and Pacific Region. Taiwan, Republic of China pp. 336-350.

Blair, G.J., Lefroy, RDR., Lisle, L., 1995. Soil carbon fraction based on their degree of oxidation and the development of a carbon management index for agricultural systems. Australian Journal of Agricultural Research, 46, 1456-1466.

Boiffin, J., Keli, J., Sebillotte, M., 1986. Systèmes de culture et statuts organiques des sols dans le Noyonnais: application du modèle de Henin-Dupuis. Agronomie, 6, 437-46.

Bolomey, S., 2002. The seasonal dynamics of salinity in a small rainfed rice-cropped watershed in Northeast Thailand (Isan). Mémoire d’ingénieur de l’Ecole Polytechnique Fédérale de Lausanne, IRD, LDD, 61 p.

Craig, I.A. and Pisone, U. 1985. Overview of rainfed agriculture in Northeast Thailand. In: Proceedings of a workshop on Soil Water and Crop Management Systems for Rainfed Agriculture in Northeast Thailand. Khon Kaen University, 24 Feb - 1 Mar 1985, Khon Kaen Thailande, 24-37.

Dobermann, A., and Fairhurst, T., 2000. Rice: nutrient disorders and nutrient management - handbook and CD, Philippines IRRI, 191 p.

Enet, Y., 2003. Influence des pratiques culturales sur les caractéristiques du sol et sur la répartition de la salinité au sein de deux propriétés rizicoles du Nord-est de la Thailande. Mémoire de DESS, Université Paris XII Val de Marne, IRD, LDD, 72 p.

Grunberger, O., Hartmann C, 2004. Improving the Management of Salt Affected soils. Case of Saline patches in Rainfed Paddy fields in Northeast Thailand. Activity Report. 77 p.

Hansen J.W., 1996. Is agricultural sustainability a useful concept? Agricultural Systems, Volume 50, Issue 2, 117-143.

Hasegawa, H., Furukawa, Y. and Kimura, S.D., 2005. On-farm assessment of organic amendments effects on nutrient status and nutrient use efficiency of organic rice fields in Northeastern Japan. Agriculture, Ecosystems & Environment, Volume 108, Issue 4, 10, 350-362.

Jenny, H., 1941. Factors of soil formation. New York, Mc Graw Hill, 281 p.

Mary, B., and Guérif, J., 1994. Interêts et limites des modèles de prévision de levolution des matières organiques et de l’azote dans le sol. Cahiers Agricultures, 3, 247-57.

Maeght, J.L., 2003. Evolution geochimique d’un sol rizicole sale de Thailande. DEA National de Science du Sol. Ecole Nationale superieure Agronomique de Montpellier, France.

Olk, D.C., Van Kessel, C, Bronson, KF., 2000. Managing soil organic matter in rice and non rice soils: agronomic questions. In: Carbon and Nitrogen dynamic in Flooded soils. Kirk, GJD., and Olk, DC editors. Philippines, IRRI, p. 27-47.

Pandey, S., Velasco, L.E., Suphanchaimat N., 2002. Economics of direct seeding in Northeast Thailand. In: Direct seeding: research strategies and opportunities. Pandey, S., Mortimer, M., Wade, L., Tuong, T.P, Lopez, K., and Hardy, B. editors. Philippines, IRRI, p. 139-160.

Parton, W.J., R.L. Sanford, Jr., PA. Sanchez and J.W.B. Stewart. 1989. Modeling soil organic matter dynamics in tropical soils. Pp. In D.C. Coleman, J.M. Oades, and G. Uehara (eds.), Dynamics of Soil Organic Matter in Tropical Ecosystems. University of Hawaii Press, Honolulu.

Powlson, DS and Olk, D; 2000. Long term soil organic matter dynamics. In: Carbon and Nitrogen dynamic in Flooded soils. Kirk, G.J.D and Olk D.C (eds.) Philippines, IRRI, p. 49-63.

Quantin, C., Grunberger, O., Suvannang, N., Bourdon, E., (Submitted) Impact of agricultural practices on the biogeochemical functioning of sandy salt affected paddy soils in Northeastern Thailand. In: International symposium on management of tropical sandy soil for sustainable agriculture, 28 November – 2 December, Khon Kaen, Thailand.

Shirato, Y. , Paisancharoen, P., Sangtong, P., Nakviro, C., Yokozawa, M., Matsumoto, N., 2005. Testing the Rothamsted Carbon Model against data from long-term experiments on upland soils in Thailand. European Journal of soil Science, 56, 179-188.

Suzuki, K., Goto, A., Mizutani, M., Sriboonlue, V. , 2003. Simulation model of rainfed rice production on sloping land in Northeast Thailand. Paddy Water Environment, 1, 91-97.

Tangtrakarnpong, S., and Vityakon P., 2002. Land use and soil organic matter in Northeast Thailand: microbial biomass, humic acid and mineral N. Department of Land Resources and Environment, Faculty of Agriculture, Khon Kaen University. 17^th WCSS, 14-21 August 2002 Thailand.

Whitbread, A., Blair, G., Konboon, Y., Lefroy, R., Naklang, K., 2003. Managing crop residues, fertilizers and leaf litters to improve soil C, nutrient balances, and the grain yield of rice and wheat cropping systems in Thailand and Australia. Agriculture, Ecosystems & Environment, Volume 100, Issues 2-3, 251-263.

¹ Institut de Recherches pour le Développement (IRD), Land Development Department, Office of Science for Land Development.
² Université Catholique de Louvain-La-Neuve, Faculté d’ingénierie biologique, agronomique et environnementale, Belgique.
³ Land Development Department (LDD), Office of Science for Land Development.