The aims of this study is in the first time to characterize the granulometric composition and analyze the phosphorus content in the soil of Maâmora forest in two sites (Bled Dendonn) and (Sidi Amira) depending on the depth and location. Second time using the principal component analysis method for determine the correlations between different parameters analyzed.

Keywords: phosphorus, soil, granulometric study, correlation, principal component analysis

Soil is an extremely complex medium which simultaneously takes place a large variety of physical, chemical and biological interactions.These interactions at the soil-root interface are little known, but it is at this level that the minerals and trace elements are assimilated by plants.^1,2 Among these elements include phosphorus.It is an essential nutrient for normal metabolic functioning of growth plants.^3‒5 It also represents a central role in all the process of photosynthesis, respiration and energy transfers.Various studies have shown the importance and efficiency of phosphate fertilizer on agricultural productivity and food security.⁶ The study of mobility of phosphorus in the soil has two advantages: from a chemical point of view to understand the availability of this element for plants; from an environmental point of view, it allows estimating the risk of groundwater pollution.Against this background we are analyzed the content of assimilable P₂O₅ in the soil of Maâmora forest in two sites; (Bled Dendonn) and (Sidi Amira) depending on depth and location.

Sites

Maâmora forest covers a little over a hundred thousands of hectares, very homogeneous appearance, consisting of variable thickness sands overlying a clay and sand composition of varying depth, called "red clay Maâmora". On the forest plan, Maâmora split into three homogeneous geographical units (Maâmora Western, Central and Eastern). These units in their turn divided into five districts called A, B, C, D and E in the west-east direction (Figure 1).

Figure 1 Localisation of the study area (a) in Morocco; (b) main tree species in the Maâmora Forest [7].

Figure 2 Variation of pH (water) as a samples function.

Figure 3 Variation of pH (KCl) as a samples function.

Soil characteristics studied

Analysis of total bases:

The most commonly total bases present on the exchange complex are: Ca²⁺, Mg ²⁺, K⁺, Na⁺.

The results presented in Table 1, it can be concluded several points:

1. The K⁺ content is almost same for both regions (Bled Dendonn) and (Sidi Amira).
2. A low content of Ca²⁺ to zero value in Sidi Amira site.
3. Mineral nutrition of plant is conditioned by entire absorbent complex; the exchangeable ions react with one another:
4. If K/Ca or Mg/Ca is greater than unity, the calcium nutrition may become deficient; so in soils low in magnesium, it is necessary that K/Mg is less than unity.
5. For all the sites studied; K/Ca and Mg/Ca is greater than unity, so there is a nutritional deficiency in calcium.

Carbon-nitrogen ratio (C/N):

The ratio (C/N) (total carbon to total nitrogen) can predict the importance of immobilization or mineralization in the soil incorporation of an organic substrate.

Depending on soil repository (1995),^7‒10 the norms of C/N ratio are:

1. 10<C/N<15 for active soil where mineralization of organic matter decomposes rapidly (high amount of nitrogen);
2. 15< C/N<25 need covered nitrogen to allow proper decomposition carbonaceous matter;
3. 25<C/N<40 for a very acidic soil, where the litter decomposition is very slow (incompletely decomposed humus).

In the soils studied the C/N ratio is less than 25 (Table 2 & Figures 4‒7), and the reorganization of released nitrogen dominate in soil mineralization.

Figure 4 Variation of N°/°°as a samples function.

Figure 5 Variation of C% as a samples function.

Figure 6 Variation of MO% as a samples function.

Figure 7 Variation of C/N% as a samples function.

From Table 2 & Figure 6 we see that the organic material MO represents a maximum value at 5.37 in BD surface site and a minimum value at 0.22 in a depth of 80cm of BD site, and this value decreases from the surface to the depth in both SA and BD collection sites. Concerning the variation of P₂O₅ and as shown in Figure 8 notice that the maximum value is recorded in the surface of SA sample at 9.47, and the values recorded in the samples collected at depths of 20cm and 80cm does not have a large variation in the two study sites.

Figure 8 Variation of P₂O₅as a samples function.

Physical method

Determination of pH:

According to Table 2 & Figures 2&3, we have the pH_H2O>pH_KCl with a small variation within the same site and using the classification of soil acidity we note that all samples are acidic (5<pH<6.5).Acidification is found mainly due to the implementation of resinous dead matter which is rich in acidifying compounds and the nature of the bedrock.

Granulometric analyzes:

For all soils studied the granulometric composition is homogeneous (Table 3), given the dominance of the sand fraction (2mm to 20µm).This sandy texture gives the soil a particulate structure and a lack of cohesion. As water reserves are low, they tend to dry out seasonally.The clay fraction takes the second position in the texture of the Maâmora soil followed by the silt.

Olsen method for assaying phosphorus in the Maâmora soil

Principle:

The extraction is carried out with an alkaline solution of sodium hydrogen carbonate at pH 8.5 for 1 hour at 20°C of temperature.

Calculations:

Depending on the sample and dilutions phosphorous, is expressed as P₂O₅°/°° as follows:

With: [a] = absolute content of P₂O₅ in γ/ml; read from the calibration curve. (γ=10^-3 mg =10^-6g=1µg).

V₁ = final volume in ml.

V₂ = collected volume for the spectrophotometer (in ml).

P = test sample of ground (g).

100 = coefficient to report the result to 1000 g of soil.

1/1000 000 = coefficient to spend γ in g.

Table 4 below shows measurements quantity of the assimilable phosphorus taken in Maâmora forest at two different sites (Sidi Amira and BladDendonn).

Interpretation:

Table 4 shows the results of taking measures in two different sites of the assimilablephosphorus in soil,We therefore find that the amount of phosphorus decreases with depth in addition to that amount is larger on the Sidi Amira site surface than BladDendonn and the amount of phosphorus in soil depends on the depth and location of soil.

Correlations study:

Table 5 shows the correlation matrix between physicochemical parameters determined. From the results of this matrix we can take the following conclusions:

The pH of water (pH_eau) is correlated positively with pH_KCl by correlation coefficient of 0.812.

Nitrogen positively correlates very significantly with carbon and organic matter (MO) by a correlation coefficient of 0.988.Nitrogen also correlates significatively with the ratio C/N (total carbon to total nitrogen) and P₂O₅ which the correlation coefficients are respectively 0.718 and 0.533. Carbon is correlated with the organic matter, ratio C/N and P₂O₅ which the correlation coefficients are respectively 1.000, 0.789 and 0.618. The organic matter is positively correlated with theratio C/N and P₂O₅which the correlation coefficients are respectively 0.790, 0.618. Theratio C/N is positively correlated with P₂O₅by the correlation coefficient of 0.653.

The principal component analysis:

According to the graphic projection of the principal component analysis we note:

For the studied parameters the total inertia of cloud axes is selected: Axis 1:74%, Axis 2: 12%, and as shown in Figure 9 which shows the correlation of the circle, this figure shows two groups. The first group formed by: pH_eau and pH_KCl, and the second formed by P₂O₅, C/N, MO, C and N.

Figure 9 Plot in the plane P₁xP₂ of principal component analysis of normalized factors studied.

From Table 6, which shows the correlation matrix of the soil granulometric composition studied we notes that: the clay positively correlates with limonf and limog whose correlation coefficients are respectively 0.544 and 0.488. As well limong is correlated significantly with sableg by correlation coefficient of 0.761. The granulometric composition of soils studied is homogeneous, two axes have been selected, and the respective contributions to the total inertia of cloud are: Axis 1:61%, Axis 2:27%. And as illustrated in Figure 10, according to this figure three groups that distinguish; the first is formed by limonf and clay, the second is formed by limong and sableg, and the third form by sablef.

Figure 10 Principal component analysis correlation circle of granulometric composition of studied soils.

We used the method of Olsen to follow the content of assimilable phosphorus depending on the depth for two different soils of Maâmora forest. For all soils studied the granulometric composition is homogeneous, given the dominance of the sand fraction (2mm to 20µm). This sandy texture gives the soil a particulate structure and a lack of cohesion. Such as water reserves are low, they tend to dry out seasonally. The clay fraction takes the second position in the texture of Maâmora soil followed by the silt. The quantity of phosphorus is analyzed decreases with depth in addition to this quantity is larger on Sidi Amira site surface than BladDendonn and this quantity of phosphorus depends on the depth and location of the soil. Using the principal components analysis statistical method (ACP) for the determination of different correlations between the studied parameters Maâmora soils.

None.

Authors declare no conflict of interest exists.

None.

1. Analysis method Determination of phosphorus in agricultural soils: colorimetric method after extraction with the Mehlich III method. Center of expertise in environmental analysis of Quebec, MA. 1010‒PC 1. 2003.
2. Leprun JC, Grouzis M, Randriambanona H. Post‒cropping change and dynamics in soil and vegetation properties after forest clearing: Example of the semi‒arid Mikea Region (southwestern Madagascar). Geoscience. 2009;341(7):526‒537.
3. Environmental Specifications N° 368 Sol. The phosphorus in soils state of the situation in Switzerland, Phosphorus in soils, fertilizers, crops and the environment. Federal Office for the Environment, Forests and landscape OFEFP Berne. 2004.
4. Breeuwsma A, Reijerink JGA. Phosphate saturated soils: a new environmental issue. Proc. European State‒of‒the‒art Conf. ì Delayed effects of chemicals on soil and sediments. 1992;1‒10.
5. Sharpley A, Daniel TC, Sims JT, et al. Determining environmentally sound soil phosphorus evels. J Soil Water Cons. 1996;51(2):160‒166.
6. Rabeharisoa L. Management and fertility and phosphatic fertilization of ferralitic soils Highlands Madgascar. state Thesis es Natural Sciences, University of Antannarivo, France. 2004.
7. Lahssini S, Lahlaoi H, Mharzi Alaoui H, et al. Predicting Cork Oak Suitability in Maâmora Forest Using Random Forest Algorithm. Journal of Geographic Information System. 2015;7:202‒210.
8. AFES. Pédologique Referential. In: Baize D, Girard MC, Editors. Main soil of Europe. Coord. INRA editions, Paris. 1992. 222 p.
9. AFES. Pédologique Referential. In: Baize D, Girard MC, Editors. Coord. INRA editions, Paris. 1995. 332 p.
10. AFES. A sound reference base for soils: the Pédologique Referential. Text in English. In: Hodgson JM, Eskenazi NR, Baize D, Editors. INRA Éditions, Paris. 1998. 322 p.