Carbon and nitrogen balances and CO₂ emission after exogenous organic matter application in arid soil

Abstract

Addition of organic amendments (OAs) could be a means to sequester carbon (C) in soils. However, the efficiency of C sequestration depends on how OAs evolve in the soil. A field study was installed in arid soil to which was added one of five OAs – olive husk-based compost (CM), palm leaf-based compost (CP), crushed olive pruning (GW), fresh olive mill wastewater (OMW) or fermented ovine manure (OM) – at an equivalent fixed level of 350 g of C/m². C and N mineralization were followed for 112 days by measuring the evolution of released CO₂ and mineral N evolution. The results showed that CM and CP did not disturb soil respiration, with a very low CO₂ emission or almost no respiration for the CM, while CO₂ release reached 7.6 g of CO₂/m² for GW and OMW. Soil organic carbon (SOC) was sustainably improved by 0.54% and 0.50%, respectively, for CM and CP. N mineralization showed no significant difference between amended and untreated soils. Based on these results, compost amendment was the most efficient for C sequestration to enhance soil fertility and consequently reduce the rate of CO₂ emission.

Keywords:

• soil respiration
• soil organic carbon
• carbon mineralization
• carbon storage

Introduction

Soils can represent a sink for atmospheric greenhouse gases (GHGs) such as CO₂, N₂O and CH₄. Exchanges of carbon (C) between soil and atmosphere are very important and can be positive (sequestration) or negative (CO₂ emission), related with soil management (Lal 2010). C sequestration in the terrestrial ecosystem mostly occurs by photosynthesis (Voronin 2006). The atmospheric C fixed by plants returns into the soil by natural incorporation of the plant residues (Paustian et al. 2000) or as organic amendment added by farmers directly or after transformation. The use of food industry organic wastes as organic amendment (OA), either fresh or through composting, is becoming a common practice (Flavel and Murphy 2006). Agricultural activities can reduce loss of C and enhance its storage in the soil (Hutchinson et al. 2007). Among the C conserving cultural practices, the most important is the application of no-tillage farming (Baker et al. 2007; Guo et al. 2016), cover crops (Lal 2004), agroforestry practices (Lal 2005) and OA (Ryals et al. 2014). Conversely, as a consequence of the intensity, farmlands that are intensively handled generate huge amounts of GHGs (Hutchinson et al. 2007), as well as changes in inland use that comprise deforestation and biomass burning, as well as the transformation of natural to agricultural ecosystems (Lal 2004).

Application of OAs is considered the most efficient strategy for the valorization of organic residues (Flavel and Murphy 2006; Thangarajan et al. 2013) by handling agricultural organic wastes such as crop wastes, agricultural by-products such as ovine manure, pig manure, olive mill wastewater (OMW), olive husk and green manure, and by reducing the problems caused by their accumulation such as their direct incorporation into the soil or after composting (Ferreras et al. 2006). Moreover, this alternative increases C sequestration in the soil. Cabrera et al. (2009) and Powlson et al. (2012) proposed that OA led to increased C storage in soils (Ryals et al. 2014). In addition, OA enhances soil fertility by improving soil physical and chemical properties, C content, and microbial activity (Bronick and Lalan 2005; Thangarajan et al. 2013). Consequently, plant growth and fruit yield are improved (Flavel and Murphy 2006).

Evolution of organic matter in the soil follows two paths, direct mineralization and humification (Lehmann et al. 2007). Labile organic C that contains easily degraded compounds such as sugars, proteins and carboxylates is directly mineralized (Fischer et al. 2010). This mineralization produces CO₂ and thus C loss due to microbial C decomposition (Bot and Benites 2005). Mineralization is a complex phenomenon that is linked to environmental conditions and to the properties of the organic matter. Mineralization is enhanced when external and intrinsic properties are suitable to develop microorganism activity (Xie et al. 2017). Indeed, soil temperature, weather temperature and humidity are key factors for microorganisms’ life and, therefore, high weather humidity and temperature enhance soil organic matter (SOM) degradation (Stone et al. 2014). Moreover, the higher the content of easily degraded components in the organic matter is, the higher its direct mineralization should be (Zech et al. 1997, Martín et al. 2012). N mineral availability in the soil and low immobilization of N-NO₃⁻ reduce C storage in the soil, and increase organic C mineralization and therefore greenhouse gas emissions (Eduardo et al. 2016).

The other path followed by the residual part of organic C after direct mineralization is humification. This process is achieved through successive decomposition and modification of dead organic matter that is converted to humic compounds or humus. These compounds are composed of more complex organic substances after biochemical processing. Humification results in stabilization of the organic matter achieved by its transformation into stable humic compounds including humic and fulvic acids and humine. Humic acids (10,000 and 100,000 Da) are complex aromatic macromolecules enriched with aliphatic compounds that bind to peptides and amino sugars (Stevenson 1982). Fulvic acids (1000 to 30,000 Da) contain aromatic and aliphatic structures strongly enriched by functional groups (Buffle 1988). Humine, with the highest molecular weight, is considered the most stable compound that is insoluble regardless of the pH (Tan 1994; Bot and Benites 2005).

However, humification is not the only method of organic matter stabilization. There are several mechanisms of stabilization of SOM, as described by Six et al. (2004). Physical protection is a mechanism of SOM stabilization, which is an occlusion of the organic matter within macro- and micro-aggregates or small pores. However, only the degradation of biochemically labile materials is delayed by physical protection mechanisms. Although the recalcitrant materials decompose slowly anyway, they are hardly affected by physical protection (Krull et al. (2003). SOM can also be stabilized by organo-mineral complexes through chemical protection and its interaction with mineral surfaces or with other organic molecules. These mechanisms are important to decrease the bioavailability and accessibility of SOM for soil microorganisms and soil enzymes (Sollins et al., 1996; Kelleher and Simpson, 2006; von Lützow et al., 2006; Bachmann et al., 2008). The selective preservation of certain recalcitrant organic compounds is also considered an important method of stabilization. Decomposition of organic compounds may be restricted due to their molecular-level characteristics such as elemental composition, presence of functional groups, and molecular conformation (Marschner et al. 2008).

Consequently, mineralization of exogenous organic matter (EOM) can be subdivided into primary mineralization that affects fresh litter, and secondary mineralization that represents a progressive degradation of stabilized organic matter (Guggenberger 2005). The C stored in soil is higher when the amount of stabilized SOM is increased. Consequently, reduction of atmospheric CO₂ and soil fertility conservation and improvement are enhanced when C sequestration in soils is improved. Thus, SOM stabilization should be enhanced versus mineralization. For this reason, more efficient organic matters are those containing more humic substances considered stable components, or more compounds suitable for the stabilization process such as lignin and the phenolic compounds that participate in humus formation (Nicolas et al. 2012). Additionally, environmental conditions affect the humification rate. Many studies have focused on organic C evolution and dynamic modelling of the organic C introduced (Al-Adamat et al. 2007; Falloon and Smith 2002; Saffih-Hdadi and Mary 2008).

Hence, different models were developed to pursue decomposition and stabilization of the EOM input: RothC-26.3, Century, Daisy, GEFSOC, AMG, etc. (RampazzoTodorovic et al. 2014). These models are important to anticipate variations in soil organic carbon (SOC) content under various environmental conditions and management practices. Moreover, many of these situations could improve understanding of the stabilization of C in soil (Ludwig et al. 2007).

However, to the best of the authors’ knowledge, no study has investigated the effects of characteristics of intrinsic organic matter on their evolution and their environmental impacts in arid soils, although it has been proposed that soils in semiarid and arid conditions with low C content have a large potential for C sequestration (Lal 2004). However, to the best of the authors’ knowledge, this has not been proven independently from organic amendment composition, nor has whether this potential for sequestration is important with all types of organic amendments. The aim of this study was to understand, based on the characteristics of exogenous organic matter, what path SOM evolution would follow and the capacity of arid soils to sequester C. The impact of soil C balance and C gain after EOM application were determined. Moreover, the mechanisms behind the observed impacts were assessed. Five organic amendments at different stages of maturity were used in field experiments. Soil respiration was followed, in addition to determining the part of stable C obtained, and the C balance and C gain or loss.

Materials and methods

Experimental study

A field study was undertaken in an arid climate from the end of February 2016 to the beginning of June 2016. This period was chosen because it is characterized by climatic conditions suitable for soil microbial activity, such as temperature between 20 °C and 30 °C and rainfall in the southern Mediterranean region. During this period, the highest and the lowest temperatures recorded were 35 °C and 8 °C, respectively. Daily soil temperature average was approximately between 17 and 18 during the first week and reached 28 at the end of the experiment. Rainfall was almost absent, reaching only 26.5 mm during the experimentation period of 112 days, much below the normal rainfall average (65 mm). Water was then added to allow microbial development. For this reason, experimental plots were irrigated weekly with 10 L/m² of water if it was not raining. Rainfall was 11 mm in April and 7 mm in May. Therefore, no irrigation was applied in these periods. Consequently, 130 mm total irrigation was added during the experiment. The total amount of water received during the whole experimental period was 148 mm.

The experimental field was non-cultivated arable land located in Sfax in central Tunisia (34°43’47”N, 10°44’12”E). The soil was classified as a sandy loam with 37.80% sand, 62.12% silt and 0.07% clay. It is an alkaline soil with a pH of 8.9, high electrical conductivity (EC = 15.09 mS/cm) and classified as saline soil according to the Food and Agriculture Organization of the United Nations (FAO) (Mathieu and Pietain 2003). This soil had a low amount of organic carbon, 0.6% equivalent to 600 g /m² soil and 0.025% of total N.

The field trial was installed based on a block design with six treatments and three replicates. Five EOMs were used. The supplied quantity of each EOM was calculated in order to achieve a supply of 350 g/m² of organic C or almost 0.35% of carbon added depending on EOM intrinsic C content. This amount of organic C corresponds to 3.5 T of C/ha equivalent to 10 T/ha at 10 cm depth of sheep manure, which is the common organic fertilizer used by farmers.

The organic amendments applied were manure, plant residues, manure/waste from fresh food industry, and manure/composts of plant residues including manure and olive husk (CM), a palm leaf-based compost (CP), olive pruning residues (GW), fresh OMW and fermented ovine manure (OM).

To apply 350 g/m² of organic C, the amounts of the different amendments applied were:

• Ovine manure: 1 kg/m^-2

• Olive mill wastewater: 9 L/m²

• Olive husk-based compost: 1.28 kg/m²

• Olive pruning residues: 0.79 kg/m²

• Palm-based compost: 1.17 kg/m²

The physical and chemical characteristics of these amendments are shown in Table 1.

Table 1. Physical and chemical characteristics of the five organic amendments.

	Olive husk compost (CM)	Palm leaf-based compost (CP)	Olive pruning waste (GW)	Ovine manure (OM)	Olive mill wastewater (OMW)
CO (%)	24 .96 ± 0.88	29.18 ± 0.62	43.98 ± 0.02	34.83 ± 0.13	42.71 ± 0.34
OM (%)	40.93 ± 2.14	51.85 ± 1.11	90.08 ± 0.26	66.45 ± 0.55	86.8 ± 0.28
N (%)	1.04 ± 0.03	1.47 ± 0.03	1.54 ± 0.03	1.77 ± 0.01	1.57 ± 0.01
C/N	24.00	19.85	28.55	19.76	27.20
pH	8 .89 ± 0.19	8.85 ± 0.55	7.89 ± 0.31	8.69 ± 0.22	4.56 ± 0.19
Ash (%)	50.06 ± 2.28	48.15 ± 1.6	9.91 ± 0.07	33.55 ± 0.35	13.20 ± 0.89
Na (%)	0.043 ± 0.0042	0.049 ± 0.0042	0.0119 ± 0.0004	0.036 ± 0.0056	0.0575 ± 0.0035
Ca (%)	0.175 ± 0.0212	0.41 ± 0.0989	0.17 ± 0.00	0.175 ± 0.0070	0.12 ± 0.0141
K2O (%)	0.17 ± 0.008	0.22 ± 0.0085	0.34 ± 0.0085	0.28 ± 0.0170	0.90 ± 0.2556

The amendments were incorporated into the superficial soil layer (0–10 cm) and were thoroughly mixed with soil by hand to ensure homogeneity. In addition to these treatments, a control plot without EOM addition was included. For all the plots other than OMW, the same amount of water as that brought by OMW addition was included.

During the first week, soil activity was important. For this reason, samples were taken daily during this period, then weekly until the end of the first month and every 14 days after that when soil activity seemed less important. The survey was followed until stabilization of respiration, 112 days after EOM addition.

For each plot, the soil sample was composed of three samples from three sampling points mixed together. Soil samples were taken from the upper layer corresponding to a depth of 0–20 cm.

Physical and chemical analyses of soil samples

Soil pH was measured in a soil–water mixture with a ratio of 1:2.5 (10 g of soil was mixed with 25 mL of water) using a glass electrode and a Neomet pH meter. The soil EC was determined using a conductimeter (inoLab) to measure the liquid extract of soil-saturated paste (Pauwels et al., 1992). Soil organic C was obtained using the Heanes method, which is a modified Walkley and Black method with external heating to help oxidize the total organic matter supplement to the heat from the addition of sulfuric acid to a potassium dichromate solution (Conyers et al., 2011). The total N content was determined by the Kjeldhal method (Pauwels et al. 1992). To determine mineral N content in soil, fresh soil samples were stirred with 50 mL of 1 N KCl, and after filtration, N-NH₄⁺ (ammonium) was oxidized by MgO to be transformed into gaseous NH₃ that was extracted by vapour distillation and caught in boric acid, and then the amount of N was evaluated by titration with 0.01 N hydrochloric acid (HCl). Then, N-NO₃⁻ (nitrate) N was transformed into NH₃ by reduction of aluminum alloy and caught in boric acid after vapour distillation, and the amount of N was determined by titration with 0.01 N HCl (Pauwels et al. 1992).

Soil respiration from the soil surface was measured in the field by the alkaline absorption method for a surface covered by sealed cylinders 15 cm high and 10 cm in diameter inserted to a depth of 2–3 cm, with three replicates for each plot. The CO₂ released from soil was measured for 24 hours before the sampling dates and determined by catching the CO₂ produced in 0.1 N NaOH, and the excess of NaOH was determined in laboratory by titration in the presence of phenolphthalein (Annabi 2009). Soil respiration was measured daily during the first week and then weekly, and the values obtained were expressed in mg/m².

Calculations and statistical analyses

To evaluate C mineralization induced by the different EOMs, the net CO₂ produced for each treatment (C-derived amendment) was calculated as the difference between the C-CO₂ produced by the amended soil and that produced over the same period of time by the control soil.

The gain or loss of SOC was estimated by calculating the difference between the SOC of the amended soil and the control soil at the end of the experiment, while SOC balance was obtained by subtracting the initial from the final SOC level (Gargouri et al. 2014).

Mean comparison analyses using analysis of variance (ANOVA) were achieved using SPSS 13.0 for Windows (SPSS Inc.). In addition, correlations and factorial analyses were performed in order to assess the mechanisms leading to the observed effects of different EOMs.

Results and discussion

Impact of EOM addition on soil pH and EC

Soil pH is an important parameter affecting the decomposition rate of SOM due to the positive correlation between soil pH and CO₂ emission from the soil (Reth et al. 2005). Soil pH was 7.65 for the control and OM, and 7.55, 7.48 and 7.45 for CM, CP and GW, respectively, whereas for OMW, the soil pH was 7.09, showing a significant difference from the other soils. This may be due to the acidic pH of the OMW (Figure 1).

Figure 1 Evolution of soil pH on day 0 and on day 112 for the control (S) and amended soils: green waste (GW), ovine manure (OM), compost of olive husk (CM), palm compost (CP) and olive mill wastewater (OWM); lower-case letters indicate significant differences between treatments on day 0 and upper-case letters indicate significant differences between treatments on day 112.

Soil pH increased slightly by the end of the experiment and the values ranged between 7.64 for GW and 7.89 for CP. Thus, the most important increase was observed for OMW, and all the soils had a quasi-neutral pH. The buffering capacity of the soil mitigated the effect of the low pH of OMW. These results agreed with those reported for OMW by López-Piñeiro et al. (2008). The observed increase in pH can be attributed to the ammonia production. Similar findings were described by Odlare et al. (2008), who observed an increase of soil pH after organic amendment. This was explained by the release of NH₃ with the mineralization of the organic N fraction. Thus, the differences between the treatments may be related to the different amounts of NH₃.

Some investigations showed that the application of organic waste as amendment in saline soil could decrease the soil pH to near neutrality (Oo et al. 2013; Diacono and Montemurro 2015).

The EC represents the soluble salt fraction of the soil. The soil studied was originally rich in soluble salts with a high EC, which at 15.06 mS/cm was considered saline soil, according to Richards (1954) who proposed that saline soil has an EC above 4 mS/cm.

The EC of the treated soils initially showed insignificant increases with the addition of the amendment, to 15.08 mS/cm for CP, 15.60 for GW and OMW, and 16.83 for CM; only with manure did the EC decrease, to 14 mS/cm (Figure 2). At the end, the values ​​of soil EC were 9.58, 9.86, 9.00, 8.74, 9.61 and 9.90 mS/cm for CM, CP, GW, OMW, OM and the control, respectively.

Figure 2 Evolution of electrical conductivity on day 0 and on day 112 for the control (S) and amended soils: green waste (GW), ovine manure (OM), compost of olive husk (CM), palm compost (CP) and olive mill wastewater (OWM).

The decrease in EC content can be explained either by the fixing of soluble salts in the organic material and in the resistant micro-aggregates or by the leaching of these salts from the surface with irrigation water poured weekly when the soil samples were mostly taken from the surface (0–20 cm).

Moreover, Oo et al. (2013) reported that after the application in saline soil of compost and vermicompost (compost transformed by various species of earthworms), EC was decreased. They interpreted their findings as caused by the increase of a salt-leaching mechanism. EOM addition to soils with high EC seems to reduce soluble salts within soil solution through their immobilization. However, these findings disagree with results reported by Franco-Otero et al. (2012), who concluded that with the addition of EOM, EC increased from 0.18 to 0.32 dS/m, due to EOM mineralization and the addition of the soluble salts.

Evolution of ammonium and nitrate

The contents of mineral N forms (N-NH₄⁺ and N-NO₃⁻) were investigated in the soil during the experimental period with regard to the important effect of mineral nitrogen on EOM decomposition. It is important to recall that the NH₄⁺ or NH₃ forms are derived from the mineralization of the organic N form by heterotrophic soil organisms (Ogbazghi et al. 2016; Guntiñasa et al. 2012). Thus, ammonium and ammonia undergo nitrification and are converted into nitrate if not consumed by the plants (Whalen and Sampedro 2010).

During the first day after EOM application, the NH₄⁺ content started out high, ranging between 9.78 and 11.68 mg/kg for CP and CM, respectively. Then, the NH₄⁺ content decreased between the second and the third day to between 3.83 for OM and 4.74 for CP (Figure 3a, and b).

Figure 3 Ammonium and nitrate content for the control (S) and amended soils: green waste (GW), ovine manure (OM), compost of olive husk (CM), palm compost (CP) and olive mill wastewater (OWM): (a) ammonium content during the experiment; (b) ammonium content during the first week; (c) nitrate content during the experiment; (d) nitrate during the first week.

Between the third and 84^th days, the ammonium content was almost constant and showed a decrease in trend at the end when the soil NH₄⁺content was 2.01, 2.79 and 3.2 mg/kg for the CP, OMW and GW treatments, respectively, and 1.33 mg/kg for OM and the control soil. Almost the total ammonium pool was lost by immobilization, leaching or nitrification, and also ammonium evolutions in all the treated soils were not significantly different from each other or from the control during the 112 days of the experiment.

From the beginning, the NO₃⁻ content was higher than the NH₄⁺content, ranging between 21.12 mg/kg for OMW and 36.71 mg/kg for the control (Figure 3c and d). Native nitrate content was higher than that in the amended soils with a significant difference in the OMW treatment during the first day. However, after addition of organic matter, no significant difference was observed between the second and fifth days. Nitrate content was almost stable after 84 days for CP and the control. Then, an increase was observed until the 112^th day: NO₃⁻ content reached 50.71 mg/kg and 65.17mg/kg for the control and CP, respectively. The increase of NO₃⁻ content in the soil can be explained by the lack of vegetation and the absence of plant roots to consume the N-NO₃⁻. Moreover, assimilation of the N-NO₃⁻ by the soil microbes required more energy than the amount used in NH₄⁺ assimilation. Thus, NH₄⁺ immobilization is more efficient than that of NO₃⁻ (Romero et al. 2015; Eduardo et al. 2016).

For the other treatments, CM, GW, OMW and OM, there was a decrease in soil NO₃⁻ content between the fifth and sixth days; it was not as significant as that between the 42nd and 56th days, but it is important to note, especially for OMW and GW. Moreover, nitrate content increased at the end of the experimentation period for all treatments.

As previously mentioned, NO₃⁻ in the soil was obtained through nitrification of ammonium, which explains the opposing trends for NH₄⁺ and NO₃⁻. The decrease of NH₄⁺ was accompanied by an increase of NO₃⁻ after the nitrification process.

This is based on the assumption that immobilization of NO₃⁻ by soil microbes will be faster in the presence of a labile C source used as a source of energy for soil microorganisms (Romero et al. 2015), and that the presence of sufficient labile C to fuel denitrification and immobilization in soil with low N content prevents nitrate accumulation in the soil (Quan et al. 2016).

Based on these results, and the deduction of Bernal et al. (1998) that fresh organic waste products have higher amounts of C easily degradable by microorganisms than compost, it can be assumed that the reduction in NO₃⁻ in soils treated with GM, OMW or OM was probably due to the organic C available in these fresh amendments.

Effect of EOM on total N

In this study, total N content in the soil was determined at the beginning and at the end of the experiment. A significant increase was observed in N content between the treated soils and the control. On day 0, the total N content in the treated soils increased by 88%, 66% and 49% for CP, CM and GW, respectively, while for the OM- and OMW-amended soils, the N gain represented 52% (Figure 4). After 112 days, the total N content in the control soil remained lower than with the other treatments. The final gain in total N was 42%, 57%, 29%, 25% and 24% for the CM, GW, OMW, OM and CP treatments, respectively, as compared to the control.

Figure 4 Total nitrogen content on day 0 and on day 112 for the control (S) and the amended soils: green waste (GW), ovine manure (OM), compost of olive husk (CM), palm compost (CP) and olive mill wastewater (OWM). Upper-case letters indicate significant differences between treatments on day 0 and lower-case letters indicate significant differences between treatments on day 112.

Organic and mineral fertilizers are the main sources of soil N (Flessa et al. 2000). Consequently, the observed increase of total N content in the treated soils at the beginning and the end of the experiment could be related to the many amendments comprising N.

The use of organic amendments should increase the total N concentration, with larger increments at higher application rates. It is apparent from these results that a higher amount of N was added to the soils with the CP application than with other EOM applications because it was added at the highest level to achieve the required amount of C. Nevertheless, with OM, the richest N amendment induced a low N content in the soil.

The soil N content was almost constant at the beginning and after 4 months for CM, OM and the control, but it increased in GW from 0.49 to 0.64 mg/g of soil and decreased for CP from 0.69 to 0.5 mg/g. Although CP had the highest NO₃⁻ content, there was a decrease in total N by the end of the 4 months.

The loss of N from the soil occurs via leaching or erosion, or by gas emissions into the atmosphere. The main cause of N loss is gas emission through NH₃ volatilization or NO, N₂O and dinitrogen (N₂) release through activation of nitrification and denitrification processes (FAO and IFA 2001). Therefore, the GW nitrate content maintained immobilization by microbes to synthesize proteins and retain N in an organic form. This hypothesis is supported by the work of Fortes et al. (2013) who showed that the same microorganisms take up the N mineral fraction to synthesize amino acids, proteins and other organic complexes. Therefore, the microbial biomass retains the N and then acts as a labile pool of N that is only rendered readily available to plants upon cell death, particularly if accompanied by nitrification.

Total N in the soil is constituted mostly by the organic N form which increased with EOM addition, and mineral forms are organic N mineralization products. The amendments used were rich in total nitrogen that induced many reactions and transformations in the soil via microbial activity, resulting in the mineralization of organic N and the production of NH₄⁺ followed by NH₃ and finally NO₃⁻.

Effect of EOM on SOM

The soil studied initially had a low content of organic matter. Obviously, organic amendment when added improved SOM content. Figure 5 reveals that the control soil had the lowest organic matter content at the beginning and at the end of the experiment. As for the amended soils, organic matter content was immediately increased with all the treatments. This significant increase was almost the same for all amendments because the same rate of organic C was added (e.g. 350 g/m²). The SOM content ranged between 1.83% for the OMW treatment and 2.04% for the CM treatment. For the other treatments, the SOM content was 1.95, 2.00 and 2.04% for CP, OM and GW, respectively. At the end, SOM was 1.91, 1.87, 1.53, 1.45, 1.40 and 0.99% for CM, CP, GW, OMW, OM and the control, respectively. Then, soils amended with composts showed the highest SOM contents, although the same amount of C had been added at the beginning.

Figure 5 Organic matter content on day 0 and on day 112 for the control (S) and the amended soils: green waste (GW), ovine manure (OM), compost of olive husk (CM), palm compost (CP) and olive mill wastewater (OWM); lower-case letters indicate significant differences between treatments on day 0 and upper-case letters indicate significant differences between treatments on day 112.

This significant difference in SOM content between treatments observed at the end of the experiment may be related to differences in the chemical composition of the amendments used. Numerous studies have shown that a higher C/N ratio may be associated with a lower decomposition rate of organic material (Majumder et al. 2008). This finding is in conflict with the present results. Indeed, initially, the C/N ratios for CP and CM were 19.8 and 24, respectively, lower than those for GW and OMW which were 28.5 and 27.2 (Table 1), although the highest amount of C stored in the soil was observed for CM followed by CP. Thus, in this study, C evolution in the soil was not controlled by the C/N ratio. Therefore, it can be estimated that the EOM applied here contained different amounts of resistant components such as lignin and in the soil (Bernal et al. 1998). Hence, the composted materials used in this experiment, which are CM and CP, may contain high concentrations of these organic constituents, improving the formation of the stable SOM that could be more resistant than those in the other treatments. This is based on the suggestion of Bruun et al. (2005) and Gul et al. (2014), who consider that residues with high lignin contents are more recalcitrant, are degraded more slowly and persist longer in soils than residues with low lignin contents. Moreover, the input of lignin increases SOM through its participation in humus formation.

Furthermore, the composting process produces stabilized organic matter enriched in humic substances. This process allows these substances to remain.

Soil respiration

Since the experiment was performed on clear soil without vegetation, soil respiration represents the CO₂ produced by departing organisms from the soil when degrading organic matter (Schlesinger and Andrew 2000). For this reason, the EOM leading to the highest respiration rate is more degraded than the others. In this case, the departing population of the soil, especially microorganisms, increases dramatically (Sollins et al. 1996).

Soil respiration was important at the beginning of the experiment, especially for GW, which reached 1.2 g/m²/day on the second day, while composts CP and CM had the lowest respiration values: 0.6 g/m²/day and 0 g/m²/day for CP and CM, respectively (Figure 6a and b). C mineralization decreased on the third day, reaching a secondary peak on the seventh day. The quick soil respiration decreased on the third day, and its immediate recovery was also observed by other authors (Bernal et al. 1998; Martín et al. 2012). These authors explained the high CO₂ emission between the first and the third days by the presence of a high concentration of easily degradable carbon, which led to a large growth of the microbial population in the soil (Bernal et al. 1998), or by the favorable conditions of humidity and temperature in the incubation (Martín et al. 2012) . On the other hand, the second peak that occurred in the same experiment may be attributed to the great variety of compounds contained in the samples and their different degrees of degradability (Bernal et al. 1998).

Figure 6 Net C mineralization for each treatment: green waste (GW), ovine manure (OM), compost of olive husk (CM), palm compost (CP) and olive mill waste water (OWM) during the experiment. (a) daily respiration during the first week; (b) daily respiration during the experimentation; (c) cumulative respiration.

It has been demonstrated that the mineralization process is controlled by the type of organic matter introduced and the labile C content (Smith et al. 2015; Yazdanpanah 2016).Therefore, the immediate high soil respiration for GW can be explained by the fact that the addition of olive residues with high concentrations of easily degradable components could be used as energy by the decomposer microbial community, resulting in an increase of C mineralization and CO₂ emission (Mahmoodabadi and Heydarpour 2014). Likewise, López-López et al. (2012) stated that compost amendment composition and the degree of organic matter stabilization could be responsible for differences in soil respiration.

C-CO₂ produced by OMW, GM and OM increased during the first week and was stabilized during the second week. The rate of CO₂ production observed with these amendments was very important, while for the two types of compost, CM and CP, respiration was almost absent.

The total amount of CO₂ produced during the112 days reached 7.6 g/m² for olive residues and OMW, 6.5 g/m² for manure, 1 g/m² for palm compost and only 50 mg/m² for olive husk compost (Figure 6c).

The large initial CO₂ emission during the first week was due to the rapid mineralization of the easily degradable fraction of the organic materials, according to Martín et al. (2012) who concluded that after the depletion of the most labile fraction, a constant and continuous C release from the most resistant fraction takes place. On the other hand, soil temperature increased at the end while the soil respiration was almost constant from the second week. These results confirmed the finding that soil temperature did not affect soil respiration, opposite to what was described by Ajwa and Tabatabai (1994).

Soils needed 2 weeks to recover their equilibrium with regard to CO₂ emission after EOM addition when using industrial wastes from fresh food, plant residues and manure. By using compost, the soil balance had not been disturbed. This was observed as a result of the partial mineralization of the compost during its production, leading to C losses throughout microbial decomposition of the labile organic matter (Bernal et al. 2009). Hence, during composting, stabilization of the organic matter occurred through humification, leading to the formation of stable organic complexes that are resistant to microbial decomposition (Mahmoodabadi and Heydarpour 2014). Therefore, the two types of compost, especially the olive husk-based compost, showed the slowest decomposition as well as the lowest mineralization rate (Cely et al. 2014).

Schlesinger and Andrews (2000) proposed that a small variation in the magnitude of soil respiration could strongly influence the atmospheric CO₂ concentration. Therefore, compost amendments should be the most efficient for C sequestration in soil and to enhance soil fertility and consequently reduce the rate of CO₂ emission, as a greenhouse gas, into the atmosphere. This is due to the fact that the compost is a fermented organic matter that had sustained the humification process (Mustin 1999).

SOC gain and SOC balance

Net C mineralization was the difference between CO₂-C produced by the amended soil and that produced over the same period of time by the control soil. Gain in SOC was evaluated by comparing the SOC content of treatments to that in the control, and C balance was obtained by comparing the amount of SOC at the end of the experiment to the initial amount or to every treatment (Table 2).

Table 2. Evolution of SOC for all the treatments, and evaluation of SOC gain and SOC balance.

	SOC 0	SOC 112	SOCbalance	SOCgain
S	0.61^a	0.58^a	− 0.03^bc	–
CP	1.09^b	1.09^cd	− 0.04^c	0.50^bc
GW	1.19^b	0.89^bc	− 0.30^a	0.31a^b
OM	1.17^b	0.82^b	− 0.35^a	0.24^a
CM	1.19^b	1.11^d	− 0.07^bc	0.54^c
OMW	1.06^b	0.84^b	− 0.22^ab	0.27^a

SOC_balance: The difference between the final and the initial SOC levels.

SOC_gain: The difference between the SOC of the amended soil and the control at the end of the experiment.

Letters indicate significant differences for columns (between treatments).

S: soil without amendement, CP: soil with palm leaves based compost, GW: soil with crushed olive pruning, OM: soil with fermented ovine manure, CM: soil with olive husk based compost, OMW: soil with olive mill wastewater, SOC0: initial soil oarganic C content, Soc112: final soil organic C content.

Organic C brought by EOMs was indirectly qualified and analyzed as SOC. Nevertheless, a part of this new SOC was lost during the first few days after its incorporation. The residual part of this SOC is very important to evaluate the efficiency of every amendment for C sequestration.

After 112 days, the C balance was negative for all the treatments. The lowest SOC loss as CO₂ was 0.07% for husk olive-based compost and 0.04% for palm leaf-based compost, while 0.22, 0.30 and 0.35% were the percentages of C loss during this period for OMW, GW and OM, respectively. These losses were strongly associated with respiration. Indeed, the treatments that showed the most important amount of CO₂ emission had the most important C loss. C balances confirmed that composts were more stable and contain more recalcitrant organic components against microbial degradation.

When evaluating the C gain as compared to the control evolving in the same conditions, the SOC contents were improved in the amended soils. Indeed, SOM increased by 0.24, 0.26 and 0.31% for OM, OMW and GW, respectively, while the CP and CM treatments showed an important SOC gain of 0.50 and 0.53%, respectively.

This amount of C stored in the soil at the end of the experimentation period could be considered the sustainable gain of C. CM, with a low C loss and the lowest respiration, induced the most important sustainable gain of C. This relation suggests that, as for C balances, the C gain after 112 days in the soil was also related to net CO₂ emission (Figure 7). The observed relation confirmed that the gain was strongly and inversely associated with the amount of CO₂ released from the amended soil. Moreover, the amount of initial C in the soil seemed not to affect the final C gain. Thus, the EOMs with high labile compounds that induced rapid SOM degradation resulted in weak carbon storage.

Figure 7 Projection of the different studied variables on the factorial plan. Factors 1 and 2 explained 76.42% of total observed variability.

The applied organic amendments led to an increase of SOC content. This result was consistent with those reported by González et al. (2010) and Yazdanpanah (2016), who observed an increase in SOC content after applying organic matter. However, the different types of EOMs did not induce the same SOC gain. Thus, not all the organic amendments could improve soil fertility and C sequestration with the same efficiency. Indeed, the type of material added to the soil had an effect. Sustainable SOC gain was greater for the composts than for other treatments although the same amount of C was added. Moreover, manure and OMW resulted in C gain corresponding to half of that obtained when using composts. Therefore, these amendments could be considered organic fertilizers with low efficiency.

This can be attributed partly to the stability of the compost against microbial degradation; it is rich with humified stable compounds as compared to other organic matter amendments and requires more time to be degraded.

Conclusion

Our results confirm that the evolution of EOM in the soil depends on the composition and maturity of the organic materials. Therefore, fresh organic material (i.e. green waste) seems to have high amounts of labile C and mineralized faster than the other treatments, whose high CO₂ emission resulted in C loss. The OWM and OM amendments followed the same trend and were also easily degradable, although for the two types of compost, the stable EOM showed the most important gain in sustainable SOC accompanied by little CO₂ emission. Therefore, these amendments were considered the best amendments, compared with the others, for remediation of soil fertility in terms of improving crop yields and reducing environmental problems, such as by minimizing CO₂ emission.

Effectively, the two composts were more stable and mineralized slowly. This property allows the gradual release of plant nutrients and more sustainable C sequestration in the soil. Therefore, compost helps to achieve a double objective, agronomic and environmental.

Thus, in the short term, organic amendments can cause temporary disturbances of soil proprieties, but C and N balances of the treated soil were then quickly recovered.

Hence, this alternative path may improve the sustainable SOC gain of arable soil in arid regions, increasing soil C sequestration and soil fertility through the addition of stable EOM. However, the ordinary amendments used by farmers, such as manure, had low efficacy. The use of composts in agricultural lands should be favored over fresh animal and vegetable residues.

Acknowledgements

This work was carried out in the Olive Tree Institute. The facilities and services of the Olive Institute are gratefully acknowledged. The study was performed within the frame of the GEET and APOQP laboratories. It was supported by the Ministry of Agriculture and Water Resources and the Ministry of Higher Education and Scientific Research. Special acknowledgments are extended to Mr. Nabil Soua and Mrs. Mouna Khlifi for their technical help and advice. The authors acknowledge Anne Lise Haenni for her help in improving the quality of this paper.

Disclosure statement

No potential conflict of interest was reported by the authors.