Short-term effect of organic residue incorporation on soil aggregate stability along gradient in salinity in the lower cheliff plain ( Algeria )

A field experiment was performed to examine the effectiveness of adding two kinds of organic materials (wheat Straw and cattle manure) on the soil aggregate stability along a gradient in salinity in the lower cheliff plain (western Algeria). The experiment was set up in a factorial design by complete randomized blocks with three replicates. The treatments were applied at a rate of 4 g C kg -1 soil. One year after organic amendments, mean weight diameter (MWD) was measured using the Le Bissonais method (1996). Other soil properties involved in saline soil aggregation were also measured, namely, electrical conductivity (EC), exchangeable sodium percentage (ESP), pH, cation exchangeable capacity (CEC), porosity, saturated hydraulic conductivity (Ksat), microbial biomass carbon (MBC) and hot water extractable polysaccharide (HWEP). The results demonstrated a significant effect of organic inputs in improving aggregate stability; the effectiveness was related to the type of aggregate stability test that has been used, organic matter type and the level of soil salinity. The wheat straw proved to be more efficient in soil aggregation than cattle manure. Most of the considered soil properties were significantly influenced by organic inputs. Overall, correlation analysis revealed that enhanced microbial biomass was the most important factor in stabilizing soil aggregates (r = 0.72 with MWDMB and r = 0.64 with MWDFW). Therefore, MWD influences soil porosity (r = 0.67 with MWDMB and r = 0.56 with MWDFW), which in turn resulted in increased hydraulic conductivity (r = 0.77 with MWDMB and r = 0.87 with MWDFW).


INTRODUCTION
Stalinization is a process of soil degradation that is becoming a serious matter in the lower cheliff plain in the western of Algeria.The excessive salt amounts adversely affect soil physical, chemical and microbiological properties (Rengasamy, 2006).Soluble salts and exchangeable sodium affect soil structure mainly by slaking aggregates and clay dispersion, which results in water logging and surface crust formation and leads to poor aeration and decline in water infiltration and drainage (Barzegar et al., 1997;Tejada and Gonzalez, 2005), this consequently causes high runoff and soil erosion (Oster and Shainberg, 2001;Mandal et al., 2008).These conditions may further compromise the activity of microorganism and plant growing (Wong et al., 2008;Raj Setia and Marschner, 2013).Moreover, in saline soil, improving structural stability may be achieved by increasing soil electrolyte concentration or/and decreasing solidity (Abu-Sharar, 1987).In addition, structural stability of aggregates depends on the interaction among other factors which might be intrinsic or extrinsic to soil, by which organic matter occupies a central position.
Organic matter plays an important role in maintaining structural stability in most agricultural soils (Albiach et al., 2001;Ferreras et al., 2006;Le Guillou et al., 2011).In saline soils, the effects of organic matter on improving structural stability have been the subject of much discussion (Barzegar et al., 1997;Tejada and Gonzalez, 2005).Adding organic matter improves aggregate stability and soil porosity which in turn promotes water infiltration, enhances salt leaching, decreases the exchangeable sodium percentage and electrical conductivity and increases the soil microbiological activities (Tejada et al., 2006).
Unfortunately, most soils of semi-arid regions are poor in organic matter (Tejada and Gonzalez, 2008).Therefore, organic amendment has been widely used to increase the soil organic matter content (Bastida et al., 2008;Fernández et al., 2009).The effectiveness of organic inputs on improving soil structural stability is not only dependent on the quantity but also on the quality of adding organic materials specially their rate of decomposability and their capacity to induce soil microbial activity (Nelson and Oades, 1998;De Gryze et al., 2005;Abiven et al., 2009).
In our region, livestock production has decreased, therefore, the amount of manure as traditional organic product remains insufficiency.For this reason, adding organic waste such as crop residues would appear to be an alternative source of exogenous organic matter and a good strategy for soil saline remediation.But in Algeria, adding plant residues like wheat straw, as an organic amendment, is not a common practice and its effects on soil structural stability specially in saline soil has not been a subject to any investigation up to now.Although, a few studies about this topic have been carried out under controlled conditions need to be validated by studies under field conditions.
In this context, we have developed a field experiment in order to evaluate the comparative effectiveness of adding two contrast organic matters (wheat straw and cattle manure) on soil aggregate stability along gradient in salinity and in combination with soil properties in the lower cheliff plain.

Study area
The experiment was conducted at the National Institute of Agronomic Research (INRA) experiment station of Hamadena (The lower Cheliff plain), which is situated in west of Algeria, 35°55' north, 00°44'east, 48 m altitude.The climate is semi-arid with hot summer and cold winter, the mean annual average rainfall is 329 mm and maximum temperature reaches 41°C in July, and the minimum temperature drops to 3°C in January.
The studied soils were pedologically young with alluvial parent material and fine calcareous texture often with saline character.The mineralogy is complex, consisting of mainly a mixture of clay minerals of smectite-illite, kaolinite and chlorite (Douaoui et al., 2004).

Experimental plot design
The study included four experimental soil plots (A, B, R and I) forming a gradient in salinity.All soils have a clay loam texture (Table 1).According to the classification of the USSL Staff (1954), soil A was considered as non-saline, while the soils B and R are slightly saline, and soil I is saline.Basic pH indicates that these soils are alkaline.They are moderately calcareous with low levels of organic carbon and nitrogen.All ESP values were lower than 5%, this threshold was considered by Shainberg and Letey (1984) and Abu-Sharar et al. (1987) as level separating sodic from non sodic soils.
The experiment was established according to a factorial randomized block design including two treatments corresponding to an amendment by two types of organic matter with different C/N ratio (wheat straw and cattle manure) (Table 2) and a control (unamended soil).Each treatment is replicated three times.Each micro plot covers an area of 9 m² (3 × 3 m) with 2 m strip left to separate the neighboring micro plots to avoid edge effects.
The amount of the added organic matter corresponds to two or three times of the amount normally added by farmers, which is at rate 4 g C kg -1 soil, equivalent to 20 t/ha of dry matter.The organic products were applied to the soil's surface on the beginning of the tilling season (November month) and incorporated to 15 cm depth by rotary tiller (rotovator).The unamended control microplots were tilled similarly to the treated micro plots.The plots were kept bare (using chemical weeding) and without tillage during the whole experimental period.
Cattle manure is the main organic product used as organic amendment by the farmers in the study region.It constitutes a traditional organic supplement and a reference organic product especially in vegetable crops.While, wheat straw amendment may be realized automatically by the return of crop residues in the soil at the end of each harvest.

Soil sampling and analysis
Over a period of one year, samples were taken from each microplot (0 to 15 cm), at field moisture capacity conditions, using a spade.One soil sample was consisting of five soil cores; the latter were mixed and sieved at 3 to 5 mm to prepare one composite sample.Physic-chemical analyses were performed on air dried samples, while microbial and biochemical analyses were carried out on fresh or 4°C stored soil samples.
The aggregate stability was determined according to the method proposed by Le Bissonnais (1996).This method describes the soil materials' behavior subjected to the action of water and separates the different mechanisms of disaggregation of each other.It consists of applying three tests: fast wetting, mechanical breakdown, and slow wetting.Each treatment reflects a particular behavior (Total burst, disaggregation by mechanical energy and partial bursting or micro cracking).In this study, the slow wetting test has not been used because it is not appropriate the lower chellif plain's soils conditions and remains insignificant according to Saidi et al. (1999) and Douaoui (2004).
For the fast wetting test, 5 g of dry aggregates calibrated between 3 to 5 mm diameters were rapidly immersed in 50 ml deionized water for 10 min.These aggregates were then be sieved in ethanol, after removing the excess water by pipetting.In the mechanical breakdown test, the aggregates were gently immersed  in ethanol, which removed the air inside the pores to minimize the disaggregation burst.Then, they were subject to manual agitation in 200 ml deionized water by performing 10 times fast and-over-end movements.The solution was adjusted to 250 ml and was left for sedimentation after eliminating the water.
The sample was transferred to 50 µm sieve in the same way as the fast wetting test.The remaining aggregates on the sieve for the two tests were dried at 40°C for 24 h then sieved by using a column of 06 sieves of decreasing size.The remaining mass soil on each sieve was measured and expressed as a function of the initial mass.The results were expressed by calculating the Mean Weight Diameter (MWD) in mm for each treatment: MWD = Σ Wi.Xi Where i correspond to each fraction collected, W i is the dry weight of the fraction collected relative to the total soil used, Xi is the mean diameter of the fraction collected.
The soils' properties were determined using standard analytical methods.Salinity was measured by the electrical conductivity of the soil saturated paste extract (CEe) at 25°C (USSL Staff, 1954).The pH was measured in a 1:2.5 soil to water solution by using electrometric method.The granulometric analysis was carried out by Robinson's pipette method after oxidation of the organic matter with H2O2 and stirring in a sodium-hexametaphosphate solution.Total calcium carbonate is obtained by a volumetric calcimeter of Bernard.Organic carbon wasanalyzed by dichromate oxidation method, and the total N was estimated by Kjeldahl method.
Cation exchangeable capacity (CEC) was determined by the soil percolation with ammonium acetate solution at pH 7 (Metson, 1956).Exchangeable cations were measured by atomic absorption spectrometer (AAS), and used to calculate ESP (exchangeable sodium percentage) from the ratio of Na + to CEC (USSL Staff, 1954).The bulk density was used to calculate the total porosity.Saturated hydraulic conductivity (Ksat) was measured according to the method of (Hénin et al.,1958) based on Darcy's equation using a saturated soil column of 9 cm height and 2.5 cm in diameter.Soil Microbial Biomass (SMB) was estimated by the chloroform fumigation extraction method (Vance et al., 1987) on field moist samples.Microbial biomass was estimated as the difference in extracted amount of organic carbon between the fumigated and unfumigated samples.Hot-water extractable polysaccharide (HWEP) content was made by hydrolysis of sugar in hot water according to the method of Puget et al. (1999).The samples (1 g dried and ground soil sample <2 mm) were extracted using 20 ml deionized water at 80°C for 24 h.The supernatant was collected after centrifugation (20000 x g) and the polysaccharide content was measured by a phenol sulfuric colorimetric method at 490 nm according to Dubois et al. (1956).Glucose was used as a standard.

Statistical analysis
The data were submitted to two way analysis of variance (ANOVA).Pearson's correlation analysis between different soils parameters was used to establish possible statistical relationships.The significance of the correlation coefficient obtained is shown using *, ** and *** to indicate the 95, 99 and 99.99% probability levels, respectively.

Aggregate stability
The results of the analysis of variance (Table 3), showed the significant effect by adding organic product on the soil aggregates stability expressed as the mean weight diameter (MWD) for the two Le Bissonais tests (mechanical breakdown and fast wetting).This significant effect was observed for both types of organic materials and the degree of salinity.The interaction between the treatments remained without significant effect.
As presented on Figure 1, the fast wetting test is more disruptive than the mechanical breakdown test.The size of aggregates resulting from fast wetting test was still lower than the size of those resulting from mechanical breakdown test.Adding organic products increased aggregate stability for the all treated soils, but the increase was greater with the use of wheat straw rather than cattle manure.The MWD MB of straw amended soils increased by 0.46, 0.2, 0.18 and 0.15 mm comparing to the control of the Soil I, R, B and A, while it was 0.10, 0.18 and 0.12 mm for manure amended soils in Soil A, B and R, respectively.The MWD FW tended to have the   same increase but with a small magnitude.The organic amendment improving effect evolved following the soil salinity.Highly saline soils showed more resistance to the Le Bissonais test except for Soil A, in fast wetting test, which presented the lowest salinity and the highest resistance.

Physical and chemical soil properties
Table 4 displayed a significant effect of the adding organic materials for all the most measured parameters, viz.EC, pH, ESP, porosity and K sat , for both organic product and salinity degree.No significant effect was observed for the CEC and the interaction between treatments except for EC.
We observed an increase in soil EC after the addition of organic inputs, but the increases were more noticeable in more saline soils.Manure was more effective than straw, it increased soil EC by 0.07, 0.2, 2.34 and 2.54 dS/m compared to soil control for the Soil A, B, R and I, respectively.While it was only, 0.06, 0.08, 0.7 and 1.23 dS/m, respectively for same soils amended with straw.Nevertheless, we noticed a decrease in pH, ESP and CEC for all amended soils.
Soil pH and ESP were more sensitive to adding straw than manure.Both organic material type and salinity level influenced total soil porosities (Figure 2).The influence was significant at P<0.01 for organic matter type and at P<0.001 for the salinity soil.The effectiveness of organic matter on total porosity hence increases saturated hydraulic conductivity after applying organic products for all soils.

Soil microbiological properties
The changes in soil's physical and chemical properties as a result of organic amendment were often accompanied by soil microbiological properties changes, viz.soil microbial biomass (SMB) and hot water extractable polysaccharides (HWEP) (Table 5).Organic product type influenced significantly both the SMB and the HWEP at P<0.001.Soil salinity influences only the Microbial biomass, and the interaction between the two factors did not show any significant effect (Table 5).
As compared with the control, SMB and HWEP were higher in all amended soils.The increase in microbial biomass was a function of salinity level.However, the gradient in salinity was not reflected by the soil HWEP rates.The increase in microbial biomass in comparison to the non-amended control soil varied between 4.2 and 6.9 mg C.kg -1 soil in Soil A and R, with straw amendment, respectively (Figure 3).And 3.1 and 5.7 mg C.Kg -1 soil in Soil B and I, with manure amendment, respectively.Wheat straw had a stronger effect on microbial biomass than cattle manure.The HWEP contents increases did not follow any preferential trend.The Cattle manure, in this case, exhibited the largest effect.

DISCUSSION
The results indicated that applying wheat straw and cattle manure increased soil aggregate stability.This increase depended on the test used to evaluate aggregate stability, the soil salinity level and the type of organic product.In fact, soil's response against the aggregate stability tests is related to the nature and intensity of the disaggregated constraints and some soil properties (Saidi et al., 1999;Abiven et al., 2009).The two tests used to evaluate the aggregate stability allow the distinction between two elementary mechanisms of aggregate breakdown, viz, slaking caused by the compression of air entrapped inside aggregate during wetting (a rain of strong intensity and flooding) and mechanical breakdown depending on the applied shaking energy (Le Bissonais, 1996).Therefore, the fast wetting test intensity depends on the rate of wetting of the aggregates (Le Bissonnais, 1996).It is, therefore, a suitable test for our soils which have a large clay content (>45%).The mechanical breakdown corresponds to the process of particles' grabbing by the wet aggregates' surface due to raindrops' impact, where the kinetic energy is not absorbed but transformed into shearing force.This energy, expressed by a hand agitation, did not reach the sufficient levels to disaggregate these soils.This may be explained by the dominance of the macroaggregate rate (>200 mm) resulting from the mechanical breakdown test compared to the fast wetting test (Figure 4).This confirms and joints the research results found by Saidi et al. (1999) and Douaoui et al. (2004) in the low cheliff valley soils.Application of organic amendments increased soil EC as a result of organic compounds mineralization by microorganisms and the synthesis of soluble minerals which increased electrolytes concentration in the soil solution, this may help to improve the soil aggregate by promoting flocculation of clay minerals (Tejada et al., 2006;Lakhdar et al., 2010).A decrease in soil pH was attributed to microbial produced CO 2 , releasing organic acids and the displacement of exchangeable H + ions to the soil solution (Nelson et al., 1998).Our results underpin many studies which claim that soil organic amendment led to a decrease in pH (Ok et al., 2011;Bai et al., 2013).Moreover, wheat straw contains more cellulosic compounds than cattle manure and various organic acids can be released by cellulose degradation (Shan et al., 2008).
In our study, low ESP levels were not sufficient to affect soils' structure with sodicity (ESP < 5%).In addition, high release of calcium and magnesium compared to sodium and the preference of organic matter to fix and exchange the Ca and Mg ions than Na (Sumner, 1993), Thereby resulting an enhancement, the ability of Ca 2+ and Mg 2+ to replace Na + from the exchange complex which reduces the soil ESP.Furthermore, this result could be also associated to an increase in infiltration rate, enhancing soil washing and thus, sodium leach ability by water rainfall during the experimental period which also might possibly be the reason for decrease in soil CEC.
Generally speaking, organic matter developed aggregate resistance towards the two disaggregated mechanisms involved in Le Bissonais tests.This may be attributed to aggregate hydrophobicity, which slows down the rate of water penetration in aggregate porosity and decreases slaking on the one hand and the internal aggregate cohesion, which limits both slaking and aggregate breakdown on the other hand.However, the large positive effect of wheat straw on soil aggregate stability compared with cattle manure, reflects as well as the classification established by Monnier (1965) in his conceptual model, validated by Abiven et al. (2008), wherein this difference in intensity of aggregative effect may be due to the presence of pre-humic substances for wheat straw and humic substance for a decomposed manure.This indicates the importance of the intrinsic characteristics of the applied organic substrates and its decomposability for maintaining soil structure.The stimulation of soil microbial community leads to an increase of biological aggregating agents as well, such as fungal hyphae and microbial extracellular polysaccharides.However, it's known that soil salinity reduces the microbial activities as a result of high osmotic stress and ion toxicities.In our case, soil salinity did not cause these adverse effects.The microbial biomass increases might be due to the low initial soils' salinity levels (range between 1.5 and 5.2 dS.m -1 ).Indeed, the rapid adaption of the microorganisms to salinity conditions makes the environment less hostile.It may also be induced by the presence of labile organic compounds and the availability of high concentration of water soluble carbon considered as the main source to provide energy necessary in developing salt tolerance and stimulating microbial activity (Marschner et al., 2003).A significant correlation between microbial biomass and the mean weight diameter (r = 0.72 with MWD MB and r = 0.64 with MWD FW ) (Table 6) explained the crucial role of adding organic matter in stimulation microbial activity, particularly fungal biomass in stabilizing soil structure by increasing the cohesion through enmeshment of aggregates by filaments (Degens, 1997;Annabi et al., 2007).In our study, the increase in soil hot-water extractable polysaccharide as consequence of the organic amendment did not affect the soil structure.However, the HWEP contents did not show any relationship with MWD (r = 0.19 with MWD MB and r = 0.1 with MWD FW ).This may be due to the small values recorded in this study which remained lower than those found in literature especially in non-saline soils and considered to be insufficient to ensure resistance to disaggregative constraints resulting from the Le Bissonais tests.On the other hand, increasing aggregate stability enhances soil porosity (r = 0.67 with MWD MB and r = 0.61 with MWD FW ) and hence improves hydraulic conductivity (r = 0.79 with MWD MB and r = 0.81 with MWD ), which are key properties in improving soil and water management.These results indicate that adding organic products such as straw amendment may be a suitable practice to restore organic matter content in order to improve soil aggregate stability and sustain soil physical quality in the lower cheliff plain.

Conclusion
Adding organic materials increased soil aggregate stability in saline soils under field conditions.The intensity of aggregative effect depended on the stability method used to evaluate structural stability, organic matter inputs' quality and the soil salinity level.Wheat straw proved to be more efficient in developing a resistance to disaggregation constraints than cattle manure.The cattle manure stimulates further bacterial population and conversely, the wheat straw stimulates more fungal population.The enhanced microbial activity especially fungal biomass mainly improved aggregate stability by increasing the cohesion through enmeshment of aggregates by filaments.In the light of these results, using wheat straw as organic amendments may be a suitable practice to restore organic matter content in order to improve soil aggregate stability and sustain soil physical quality in the lower cheliff plain.Further works are required to study the evolution aggregate stability along with the time to confirm the positive long term effects of wheat straw as an organic amendment.

Figure 1 .
Figure 1.Mean weight diameter of aggregates (MWD) resulting from aggregate stability tests: Mechanical breakdown test and fast wetting test.Error bar indicates the standard error of the mean.

Figure 2 .
Figure 2. Soil physical and chemical properties.Error bar indicates the standard error of the mean.

Figure 3 .Figure 4 .
Figure 3. Soil microbiological properties.Error bar indicates the standard error of the mean.

Table 1 .
Properties of soils used in the study.
a ECe, Electrical conductivity of saturated extract; b ESP, exchangeable sodium percentage.

Table 2 .
Chemical properties of the organic products.

Table 3 .
Analysis of variance results for the Mean weighted diameter (MWD).

Table 4 .
Analysis of variance for the soils measured parameters.

Table 5 .
Analysis of variance results for soil microbial biomass and hot water extractable polysaccharides.

Table 6 .
Correlation coefficients between MWDMB and MWDHR and soils properties.