1. Introduction
The agricultural practices and soil management model implemented in developed countries since the second half of the 20th century are causing the accelerated degradation of this natural resource and the ecosystems affected. For this reason, international organizations such as the FAO, the European Union, and the International Union of Soil Sciences (IUSS), among others, are promoting more environmentally friendly agricultural models and developing regulations for their effective application. In this sense, many studies have shown that intensive agricultural activities, oriented exclusively towards obtaining higher economic yields, have led to the progressive degradation of the environment, with the consequent loss of productive capacity [
1,
2,
3]. This degradation is reflected in the soil in many ways, which includes episodes of physical degradation such as erosion [
4,
5], decreasing the soil porosity and hindering the exchange of greenhouse gases [
6,
7], as well as reducing the water retention capacity [
8].
These intensive agriculture models also originate chemical degradation processes, as is the case for soil salinization and contamination, especially in semiarid regions similar to the area where this study was conducted [
9], as well as to the appearance of nutritional deficiencies [
10]. Therefore, the development and implementation of agricultural models that incorporate the objectives of the economic profitability of farms and others related to the preservation of natural resources, especially soil, would be highly recommended; incorporating sustainable agriculture techniques oriented towards a circular economy model [
11], where most of the waste generated on farms is used [
12,
13,
14], is one such example. In this sense, the promotion of the circular economy is one of the priority objectives to protect our planet and alleviate the greenhouse effect that has been gradually increasing in recent years [
15]. One of the most widely used practices in sustainable agriculture is the use of organic fertilizers as a substitute for synthetic mineral fertilizers. Organic matter (OM) is a scarce component in the soils of semiarid climates, such as those of Southern Spain, and yet it is probably the most influential component of the properties of these soils. Thus, the use of organic amendments increases their quality and fertility [
16,
17] and has positive effects on production by providing essential macro- and micronutrients for crop development [
18], thereby increasing the yield [
19]. It also improves physical properties, such as the structure [
20,
21], as well as the chemical [
22] and biological [
23] properties of the soil.
In addition, it has been shown that the use of organic amendments favors the soil’s role as a carbon sink [
24,
25]; although, when animal manure is used the impact on soil SOC is highly variable and depends on factors that are not yet well understood. In this regard, international agreements, particularly the 2006 Intergovernmental Panel on Climate Change (IPCC) guidelines, recommend that countries report changes in SOC stocks as a result of manure application for national greenhouse gas inventories.
On the other hand, when a reduction in the OM content occurs it causes a deterioration of the soil’s physicochemical properties and, consequently, loss of productivity in the medium and long term [
26,
27]. The implementation of sustainable agricultural practices is accompanied by changes in most properties, which can affect soil fertility and alter the availability of nutrients for crops, all as a consequence of physical and chemical changes in rhizospheric environments [
28]. As a result, it is necessary to monitor the soil properties that are most sensitive to soil management. These properties can be chemical quality indicators, such as organic carbon (OC), total nitrogen (TN), Olsen phosphorus (P), electrical conductivity (ECext), pH, cation exchange capacity (CEC) and exchangeable bases, and assimilable elements, to help ensure that the transition of management change is conducted properly, ensuring the soil quality and productivity. Some of these quality indicators are rapid response indicators [
29], since the interpretation of their results allows for the prediction of soil quality evolution, because they are sensitive to soil management and agronomic management.
To carry out this study, a 2.5 km
2 farm was selected, located in the NE of the province of Granada (Southern Spain), where several projects on the sustainable management of its water, soil, and livestock resources were developed [
29]. The farm had a plot of almond trees, which occupied most of the farm, and a plot of cereals, both rainfed. There was also a plot with water available for irrigation, dedicated to the cultivation of vegetables. The farm was completed with a flock of sheep of the Segureña breed.
In this part of the farm, where irrigation water was available, an experimental design was implemented to validate the agronomic and environmental impact of the application of sustainable agricultural techniques, including the use of organic amendments, whose production and crop quality results were published by Sanchez-Navarro et al. [
30]; the environmental results are the subject of this article. Due to the marked continental character of the climate of the area, with winters in which minimum temperatures can drop below 0 °C for several months and very hot and dry summers, a farm management system was designed that combined a rotation of rainfed arable crops between October and May, based on cereals and legumes, with a cycle of outdoor vegetables (celery), taking advantage of the summer period.
In the starting hypothesis we considered that the use of organic amendments, as the basis of nutritional inputs, should have a positive impact on soil quality indicators compared to conventional inorganic fertilization and, hence, on some functions of the soil, such as food production and its role as a carbon sink. In addition to the use of other agricultural practices, such as crop rotation and extensive grazing, a model for the integrated and sustainable management of natural resources was implemented which, in economically depressed and depopulated areas such as the one where this study has been carried out, can represent a tool that optimizes the economic viability of farms.
The objective of this study was to determine the influence of the application of two types of organic amendments on sustainably managed crop soils versus conventional fertilization based on chemical fertilizers and intensive soil management on some soil quality indicators, their evolution in the study period, and the effect on soil functions.
3. Results
3.1. pH, CaCO3, Organic Carbon, Total Nitrogen, C/N Ratio, and Olsen P
Regarding the general soil parameters, it should be noted that for the pH values, both in water and KCl, the differences recorded were not statistically significant (p-value = 0.342 and 0.949, respectively), obtaining mean values of 8 and 7.5 for pHw and pHKCl, respectively. The CaCO3 also showed no differences between treatments, neither was any change observed over the 3 years of the experiment, obtaining mean values of 550 g kg−1.
As can be seen in
Table 5, OC presented significantly higher values in the LSM than in the other two treatments (W = 538,
p < 0.001), with a significant increase in this constituent during the trial period, while in the I and COA it remained unchanged. In the COA, the OC content was intermediate between I and LSM and remained approximately constant throughout the experiment. Finally, it should be noted that, although the OC values were not as high as in the LSM, significant differences were observed with I (W = 57,
p < 0.001).
As for total nitrogen (TN), higher values were recorded during the three years in the LSM and COA (
Table 5), such differences being significant between LSM and I (W = 459,
p-value = 0.0004) and between I and COA (W = 132,
p-value = 0.0013). There was no clear trend in this variable throughout the trial period, as it decreased between the first and second years and increased from the second to the third years in all treatments.
As for the C/N ratio, significant differences were also detected between the LSM and COA (W = 408, p-value = 0.013), with the highest values corresponding to the LSM, with an average of 8, while the lowest values were found in the COA plot, where the average value throughout the experiment was 6.9. The evolution of the C/N ratio over the three years did not show a fixed trend either, increasing between the first and second years and decreasing between the second and third, although the value at the end was statistically higher than at the beginning.
The Olsen phosphorus (P) showed that the LSM treatment presented significant differences with respect to I (W = 529, p-value = 7.05 × 10−7) and COA (W = 408, p-value = 6.59 × 10−6) such that the average of this macronutrient was higher in the LSM. As for its evolution during the experiment, a statistically significant increase was observed.
3.2. Cation Exchange Capacity and Exchangeable Bases
Regarding the CEC,
Figure 2A shows the evolution of this property in each of the trials over the three years of the experiment. As can be seen, the highest values were obtained in the COA and the lowest in the I, while the LSM had intermediate values. Likewise, the statistical analysis showed significant differences between all treatments (LSM-I:W = 497.5,
p-value = 1.449 × 10
−5; I-COA:W = 67.5,
p-value = 5.511 × 10
−6). As for its evolution throughout the trial, no significant differences were observed between the different years.
As for the bases of change, Na (
Figure 2B) did not show significant differences throughout the study between treatments, although a tendency to decrease throughout the trial was observed. The potassium content, however, was significantly different, at 99%, between the treatments (LSM > I > COA), with the statistical significance of these differences being as follows: between the LSM and I, W = 501.5,
p-value = 1.115 × 10
−5; LSM and COA, W = 518,
p-value = 2.203 × 10
−5; and no statistical significance between the I and COA. As for the evolution of the experiment over the years, an inverse trend was observed between the LSM and the rest, since in the former the potassium increased from 0.77 in year 1 to 1.09 g kg
−1 in year 3, while in the I the contents decreased from 0.56 to 0.29 g kg
−1, respectively, and in the COA they remained more or less stable with a mean value of 0.49 g kg
−1, as shown in
Figure 2C.
The assimilable Mg values also showed differences between treatments (
Figure 2D); thus, between the LSM and COA a statistical significance of W = 394 with a
p-value = 0.02948 was obtained, while between the I and LSM there was W = 186.5 with a
p-value = 0.0372, and in the I-COA pair the values were W = 99 with a
p-value = 0.0001006. Finally, over the three years the tendency was to remain constant, except in the treatments I and COA, where a decrease was observed between the second and third years.
3.3. Electrical Conductivity of the Soil Extract Saturation and Ion Content in the Solution
The ECext showed different trends according to the treatment applied, increasing slightly in the LSM throughout the experiment (
Figure 3A), from 4.5 in year 1 to 4.8 dS m
−1 in year 2, with the mean value in this treatment being the highest of the three (4.5 dS m
−1). In trial I, the ECext decreased, especially in the second year, with a mean value of 3.5 dS m
−1. The mean value in the COA was intermediate between the other two trials (4.2 dS m
−1), although the tendency in this trial was to increase throughout the experiment. The differences between the three treatments, with two degrees of freedom (df = 2), were significant with a
p-value of 0.003894 (Χ
2 = 11.0966, df = 2,
p-value = 0.003894), specifically between the LSM and I (W = 118.5,
p-value = 0.001556) with a 99% degree of significance, as well as between I and COA (W = 1075,
p-value = 0.02958), while between the LSM and COA trials there were no differences.
As for the composition of the soil solution, NH
4+, PO
43−, and CO
32− were not found, predominating instead as anions NO
3−, Cl
−, and SO
42−, while in the group of cations Na
+, K
+, Ca
2+, and Mg
2+ stand out. Thus, the mean NO
3− content was higher than 20 mmol
(−)L
−1 in the three trials, and there were no statistically significant differences among them (Χ
2 = 0.517, df = 2,
p-value = 0.7722). Finally, a very different trend was observed between the time evolution of the LSM and COA (
Figure 3B), since while the former tended to decrease throughout the experiment, the COA increased significantly between the first and second years, remaining stable from the second to the third year. The concentration of NO
3− in the I remained more stable, and only a slight decrease was observed between years 1 and 2. As for Cl
−, in view of
Figure 3C, it can be said that the large deviation in the results prevented establishing a defined behavior of this parameter; in spite of this, there were significant differences between the three treatments at the end of the experiment, with a degree of significance of 95% (Χ
2 = 19.0197, df = 2,
p-value = 7.412 × 10
−5), highlighting those between the treatments LSM and I, with a significance of 99%. Likewise, a significant increase in the concentration throughout the cycle was observed in the three treatments. Finally, SO
4= had a behavior over time very similar to that of Cl
− (
Figure 3D) such that it increased in the LSM and COA throughout the experiment, while in the control (I) it decreased from the second year onwards. The differences between the treatments were not as contrasted as in the case of Cl
− (Χ
2 = 7.7486; df = 2;
p-value = 0.02077), but they were significant between the treatments I and COA (W = 599,
p-value = 0.01149), where the minimum and maximum values were found, respectively.
As for the cations in the soil solution, Na
+ did not show a definite trend during the test period (
Figure 4A), with an increase in the LSM and COA treatments between the second and third years and a decrease in I starting from the first year. Among the different treatments, at a significance level of 0.05, there were no statistically significant differences (Χ
2 = 4.8915, df = 2,
p-value = 0.08666). K
+ showed significant differences among the three treatments (Χ
2 = 17.6592, df = 2,
p-value = 0.0001463) (
Figure 4B), with the sheep manure trial having the highest mean value of the three (5.6 mmol
(+) L
−1). Regarding the evolution over the three years of the experiment, an increase in the LSM and COA was observed between the second and third years, while the opposite was observed in the I. The Ca
2+ concentration in the soil solution was the maximum in the first year of the trial in the LSM and I, after which a decrease was observed, which stabilized in the third year; however, in the COA it remained stable over the 3 years (
Figure 4C). On the other hand, no differences were obtained with a statistical significance of 95% between the treatments (Χ
2 = 4.2368, df = 2,
p-value = 0.1202; df = 2,
p-value = 0.102).
The Mg
2+ concentration in the soil solution (
Figure 4D) increased in the three treatments throughout the study period, especially in the LSM and COA where a statistically significant increase was observed over the three years, while in the I it only appeared between the first and second years and remained constant between this and the third year. Despite these trends, the values were very similar in the three treatments and, at a significance level of 0.05, lacked statistical significance (Χ
2 = 0.9594, df = 2,
p-value = 0.619).
3.4. Assimilable Fe, Cu, Mn, and Zn
Of the assimilable elements, Mn and Fe were the most abundant, followed by Zn and Cu, as shown in
Table 6. When compared among the three treatments carried out, the LSM showed the highest Fe content, while the I and COA had the lowest value. The differences were, therefore, significant between the LSM and I (W = 440.5,
p-value = 0.001719) and the LSM and COA (W = 411,
p-value = 0.01152). In the case of Cu, the treatment I had the lowest mean value, showing significant differences with respect to the treatments LSM (W = 392,
p-value = 0.03278) and COA (W = 170.5,
p-value = 0.01576). Except in the first year, Mn had a similar concentration in the three treatments such that no significant differences were observed in the accumulated values. Finally, Zn had the highest average value in the LSM, representing twice the concentration found in the trials I and COA. The statistical analysis confirms that the differences found were significant between the LSM and trial I at 99% (W = 428.5,
p-value = 0.003885) and especially with COA (W = 459,
p-value = 0.0004363).
Considering the evolution over the three years of the study, in all three treatments there was, in general, a considerable decrease in the concentrations of Mn and Zn, while Fe showed an uneven trend in the three treatments and the Cu concentration increased to a statistically significant extent.
3.5. Organic Carbon Capture
As can be seen in
Table 7, the LSM treatment based on composted sheep manure, showed an increase in organic carbon capture (OCc) throughout the three years of the experiment, while in the COA an increase was observed only in the first year and a decrease in the rest, as in the case of I. Because of this, it can be said that the addition of organic amendments to the soil led to a significant increase in its capacity as a carbon sink, especially when comparing the LSM and I treatments where the greatest difference was obtained at the end of the trial, with 37.08% more OC per hectare than in I, as well as in the COA-I comparison where the efficiency ranged between 10 and 16%.
5. Conclusions
This study has shown that the use of organic amendments as fertilizers instead of inorganic fertilizers produced favorable changes in some soil chemical properties. Thus, during the study period, there was a clear increase in the OC content in the organically managed soils, while in the control plot (I) it remained fairly stable, with a slight tendency to decrease. Because of this, the addition of organic amendments to the soil, together with rotations with cereals and legumes in the cold seasons and the use of extensive grazing, is a cultivation technique that enhances the soil’s function as a carbon sink. In addition, fertility indicators also improved, as is the case for the macronutrients N, P, K, and Mg and micronutrients; this is especially true for Fe and Cu, which, given their sensitivity to agronomic management, can be considered as rapid response indicators. This contributed to the improvement in the soil’s ability to produce food, but with an additional added value—the preservation of the environment. Positive changes were also observed in the CEC and, therefore, in the soil’s capacity to adsorb ions and act as a substrate capable of fixing and partially immobilizing pollutants, such as heavy metals and other positively charged metabolites.
On the other hand, the ECext was also affected by the agricultural activities carried out over the three years of the experiment, especially as a consequence of the increase in the concentration of Na+ and Cl−; they reached higher values in the organic treatments LSM and COA, than in I, an aspect that can trigger a salinization process and that makes it necessary to review the doses of fertilizers applied, as well as the crop rotations that can act as phytoremediators.
In summary, it can be stated that the agronomic management of farms based on organic amendments, crop rotation, and extensive grazing appears to be a model of sustainable agriculture that preserves the quality of the soil as a source of nutrients and promotes its ecological function as a C sink.