Agroecology and Precision Agriculture as Combined Approaches to Increase Field-Scale Crop Resilience and Sustainability

Abstract

This study coupled precision agriculture with agroecology to improve the agricultural systems’ sustainability in a climate variability context, characterized by fewer rainy days and more extreme events. A three-year comparative analysis was carried out in a durum wheat rotation, divided into two plots of 2.5 ha each, one managed with conventional methods (CP, sunflower as intermediate crop) and another managed with an agroecological approach (AE, field bean as green manure crop), featuring prescription maps for site-specific mineral fertilization. The statistical analysis of durum wheat parameters, soil characteristics, and economic variables was conducted alongside the examination of climatic data. In AE soil, the exchangeable calcium was statistically different from CP soil (6044 mg kg⁻¹ and 5660 mg kg⁻¹, respectively). Cation exchange capacity was significantly higher in AE (32.7 meq 100 g⁻¹), compared to CP (30.9 meq 100 g⁻¹). In AE, wheat yield (2.36 t ha⁻¹) was higher than in CP (2.07 t ha⁻¹), despite extreme rainfall causing flooding in some parts of the AE plot. The economic balance was only 6% in favor of CP (EUR + 2157), confirming the AE approach’s resilience (EUR + 2027), despite the higher costs of cover cropping and site-specific fertilization. The novelty of integration between “smartish” precision agriculture and agroecology allows for sustainable management.

1. Introduction

The global population is projected to reach 9 billion by 2050, creating an urgent need for agriculture to meet the growing demand for food while addressing environmental challenges. However, factors such as biodiversity loss, reduced carbon sequestration in soils and vegetation, soil degradation, pollution, desertification, and the depletion of water and energy resources, compounded by climate change, are diminishing agriculture’s ability to meet these needs [1,2].

A primary contributor to these environmental challenges is the emission of greenhouse gases (GHGs), particularly carbon dioxide (CO₂) from the combustion of fossil fuels, along with methane (CH₄) and nitrous oxide (N₂O), which are generated through agricultural practices. Greenhouse gas concentrations in the atmosphere have reached levels not seen in the last 800,000 years [3]. Agriculture is responsible for a significant portion of these emissions, accounting for an estimated 21% of global GHG emissions in 2010 [4]. Furthermore, intensive farming practices that rely on monoculture systems have been shown to be particularly vulnerable to the effects of climate variability, especially in terms of abiotic stress, and have limited resilience. Intensive agriculture also contributes to 75% of global deforestation [5], further highlighting why conventional agricultural models are no longer sustainable for ensuring food security while preserving ecosystems. The alterations in the natural carbon cycle due to these practices contribute to climate change, which in turn affects agriculture. This includes changes in crop yield and water availability, particularly in regions like the Mediterranean, where a gradual reduction in rainfall and rising temperatures are increasingly evident [6,7]. The variability in rainfall patterns poses significant challenges for crop planning, including cultivar selection and sowing timing [8]. Moreover, irregular rainfall—whether early, late, excessive, or insufficient—can foster the spread of pathogens, pests, and diseases that negatively impact agricultural productivity [9]. The role of crop breeding in identifying resilient, stress-tolerant, and pathogen-resistant genotypes is critical in addressing these challenges [9].

Soil represents a significant carbon sink, capable of sequestering CO₂ and reducing atmospheric greenhouse gases. However, unsustainable agricultural practices, coupled with climate change, are accelerating soil degradation, diminishing its capacity to sequester carbon, particularly in arid and semi-arid regions. In this context, sustainable agriculture presents a potential solution, offering the ability to adapt to climate variations while ensuring food security and environmental protection [10]. Various international research programs have been developed, such as the 4 per 1000 for SOC, with the aim of increasing global soil organic matter reserves by 0.4% per year as compensation for global greenhouse gas emissions from anthropogenic sources [11]. Furthermore, within the regulations of the CAP (Common Agricultural Policy) that integrate the rules for the management of ‘greening’, there are areas of ecological interest (EFA). EFAs have a direct impact on biodiversity, the mitigation of greenhouse gas emissions from agriculture, and the improvement of the environment and landscape through fallow land, landscape features, terraces, buffer strips, afforested and agroforestry areas, and areas where there is a reduced use of on-farm inputs such as areas covered with catch crops and ground cover in the winter season [12].

Sustainable agricultural models, including organic farming, conservation agriculture, precision farming, and agroecology, are gaining prominence. These approaches aim to develop resilient agricultural systems that reduce reliance to external inputs such as fertilizers, herbicides, water, and energy, while enhancing resistance to climate-induced challenges [13]. Agroecological practices, in particular, focus on maintaining consistent yields without exacerbating environmental impacts by improving soil health, organic carbon content, plant health, crop productivity, and systemic resilience.

Increasing the sustainability of agricultural models through innovation is essential for ensuring high-quality food production while maintaining agro-biodiversity. Precision agriculture plays a key role in optimizing crop yields and minimizing environmental impact. By using advanced technologies to monitor and manage the variability inherent in agricultural production, precision farming aims to apply inputs more efficiently, benefiting both the environment and farmers’ economic well-being [14,15]. The present study seeks to evaluate the yield responses of wheat under agro-ecological and conventional farming systems in Central Italy, on a volcanic clayey soil classified as Typic Argixeroll, while also assessing the profitability of both systems.

This research aimed to improve the resilience of an agricultural system by merging sustainable agriculture practices, an integrative contribution still lacking in the current literature. The expected results could have relevance regarding the current European regulation context. The main limitation of the present research is that the case study refers to a single area in the Mediterranean region; it is therefore suitable that similar research be carried out in other environments as well.

Two integrated approaches based on sustainable agriculture models were utilized, as follows: (a) precision agriculture, which employs advanced technologies to monitor and manage agricultural variables with the aim of optimizing production, reducing costs, and minimizing environmental impact; (b) agroecology, which promotes the sustainable use of natural resources and reduces dependence on chemical inputs through practices such as crop rotation, cover cropping, and the use of green manure. The novelty of this study is the integration of “smartish” precision technologies with agroecological practices to enhance awareness of sustainable farming methods and increase resilience in the face of climate variability. The evaluation of wheat yield and profitability was conducted using crop morpho-physiological parameters, while the economic and environmental sustainability of the two management systems was assessed through profitability analysis and soil chemical and physical evaluations.

2. Materials and Methods

2.1. Site of Study

In 2020 an experiment was set up in a 5.0 ha field, located in the north-eastern outskirts of Rome, central Italy, latitude 42.103° N, 12.628° E (Figure 1), within the experimental farm of the Consiglio per la Ricerca in Agricoltura e l’Analisi dell’Economia Agraria—Centro di Ricerca per le Trasformazioni agroalimentari (CREA-IT), used for conducting research.

The geo-pedological description of the soil under investigation and the pedoclimatic characteristics of the area can be found in [16,17]. The soil was characterized by high clay and active limestone content, a low amount of organic matter, and a high level of degradation due to the intensive use of chemical fertilization, weeding and deep tillage. The physico-chemical characterization of the soil for the objectives of this study was carried out by collecting 2 kg of sample in 20 random sampling points (Figure 1) at a depth of approximately 0–20 cm.

The main soil parameters compared between AE (agroecological) and CP (conventional) samples in October 2019 and September 2021 are related to exchangeable calcium and cation exchange capacity. In 2020, sampling was carried out in AE to verify the effectiveness of phosphate fertilization.

Soil characterization was performed according to the official Italian method of analysis [18] by a UNI CEI EN ISO/IEC 17025: 2005-certified laboratory [19]. Soil analyses indicate an easily degradable soil with a low organic matter content and a high silt and clay content. In addition, the phosphorus concentration is low due to the presence of active limestone and a zeolitic mineral (analcime) that prevent the mobilization of phosphorus needed for plant nutrition.

2.2. Set-Up of the Experimental Test

The study area was managed following two approaches: conventional (CP), complying with the local techniques, and agroecological (AE), conducted by applying agroecology recommendations and some tools of precision agriculture. Each area was 2.5 ha. In February 2020, field beans were sown as a green manure crop in AE with harvest in June, while in April 2020, sunflower was sown in the conventional field with harvest in September 2020. In November 2020, the durum wheat variety Platone was sown in the two 2.5 ha fields.

Soil fertilization, assessed via the phosphorus concentration, was varied along the area showing the presence of at least three zones (Figure 2). Therefore, a map was created for the supply of phosphate in agro-ecological treatment, following the current regulations enacted by the correct regional administration [16].

On the basis of the P map, three areas of the AE plot were fertilized with mineral superphosphate at pre-sowing (September 2020)—157 kg ha⁻¹, 170 kg ha⁻¹, and 183 kg ha⁻¹. Phosphorous fertilization in AE was managed using a “smartish” VRT technology (by varying the speed of the tractor and setting the fertilizer spreader), while fertilizer distribution in CP was carried out evenly with a fertilizer spreader over the entire area (Figure 3).

Nitrogen fertilization at different dosages was carried out on both AE and CP. According to agroecology principles, in AE the field bean green manure supplied a further quantity of nitrogen through bio-fixation and the burying of the crop residues. Thus, the nitrogen requirements in the AE plots were calculated by an elemental balance, and 72 kg ha⁻¹ of 18–46 (NP fertilizer) was distributed. In CP plots, 150 kg ha⁻¹ of the same fertilizer was distributed as common farm practice. Both interventions were carried out using a centrifugal spreader (Lely). A cover fertilization on wheat was applied at the beginning of stem elongation, applying 52 kg ha⁻¹ of ammonium nitrate on the AE plots and 150 kg ha⁻¹ of both ammonium nitrate and urea on the CP plots (

Table 1. Nitrogen inputs (expressed in q ha⁻¹ and kg ha⁻¹) during cover fertilization in the two fields: CP = conventional and AE = agroecological.

Fertiliser	UF	q 2.5 ha−1		kg ha−1
		AE	CP	AE	CP
Diammonium-phosphate	26	1.3	3.75	52	150
Urea	46		3.75		150

UF = units of fertilization.

).

The crop rotations in the two different cultivation methods were as follows: field bean for green manure (February 2020)–durum wheat (November 2020) in AE; sunflower (April 2020)–durum wheat (November 2020) in CP.

Collection of Plant and Soil Samples

Samples of durum wheat plants were collected in July 2021. A total of 54 plants were collected from each sampling point (20, corresponding to the sampling for soil analysis) in both the agro-ecological treatment and the conventional field. The samples were characterized in the laboratory for biometric and productive traits. The soil samples were taken at two different periods: in October 2019 to establish phosphate and nitrogen fertilization schedules in AE and CP, and in September 2021 to compare the change in soil physico-chemical parameters between AE and CP after wheat harvest.

2.3. Analyses Carried out

2.3.1. Analysis of Chlorophyll Content (SPAD) and Foliar Nitrogen Content Carried out in the Field

During the growing season, once at physiological maturity, measurements of the chlorophyll content of the plants were taken using the Konika Minolta (Tokyo, Japan) Chlorophyll Meter SPAD-502. The amount of chlorophyll together with leaf nitrogen content was measured by considering the flag leaf, mid leaf and basal leaf on each plant sample. To calculate leaf nitrogen, Kjeldahl’s formula (Xiong et al., 2015) [20] relating SPAD and nitrogen was used.

2.3.2. Determining the Physiological Traits of Crop

The structural parameters of the field are listed in

Table 2. Structural parameters determined within the two treatments CP = conventional and AE = agroecological.

Physiological Traits of Crop	Unit of Measurement
Fresh weight	g
Dry weight	g
Straw weight	g
Water content	%
Roots partitioning coefficient	%
Stems partitioning coefficient	%
Ears partitioning coefficient	%
Leaves partitioning coefficient	%
Leaf area	cm2
Specific leaf area (SLA)	cm2 g−1
Leaf area index (LAI)	m2 m−2

. The definitions and calculation methods of each parameter are described within the Supplementary Materials.

2.3.3. Determination of Production Parameters: Grain Yield and Harvest Index

All the plants (54) collected for each sampling point were used to estimate the seed yield (in g) per unit area (in m²) through Equation (S13) in the Supplementary Materials and then converted to t ha⁻¹.

Harvest index expressed in % represents the ratio between the dry matter production of the tissues with economic value (the seeds in case of the wheat) and the dry matter production of the entire plant, excluding roots (Equation (S14) in Supplementary Materials). The HI valuation is mainly applied to wheat, barley, rice and leguminous crops, and the values lie between 0.50 and 0.60.

2.3.4. Determination of Soil Chemical and Physical Parameters

The main soil parameters compared in the two AE and CP treatments refer to the exchange calcium and cation exchange capacity. The comparison was carried out between the following:

• Conventional (CP1) and agroecological (AE1) treatment on soil samples taken in October 2019 (before wheat sowing). The parameters compared by statistical analysis, between the two theses, are listed in Table S3 of the Supplementary Materials;
• Conventional (CP harvest) and agroecological (AE harvest) treatment on soil samples taken in September 2021 (after the wheat harvest). The parameters compared in the statistical analysis are described in Table S4 of the Supplementary Materials.

2.4. Analysis of Climate Data and De Martonne Aridity Index

Climate data such as average temperature (°C) and average precipitation (mm) during the years 2018, 2019, 2020 and 2021 were analyzed. The weather data were obtained from the hydrometeorological station of the National Research Council (CNR, Monterotondo) (Lat, 42.1; Lon, 12.6; Altitude, 92 m asl in the vicinity of the experimental area).

Using climatic data, it was possible to determine the average temperature (monthly and annual) and the accumulated rainfall during the four years in order to generate climatic diagrams to summarize the temperature and rainfall trends within the experimental area over the year, to compare the climate, and to understand the annual distribution of temperatures and especially rainfall, thus identifying dry and arid periods. On this basis, the De Martonne aridity index was calculated with the following formula (Equation (S17) in Supplementary Materials) [21]:

$$\mathit{I}\mathit{A}={\displaystyle \frac{\mathit{P}}{\left(\mathbf{10}+\mathit{T}\right)}}$$

where the following pertains:

• IA is the aridity index;
• P is the average annual rainfall (mm);
• T is the mean annual temperature (°C).

2.5. Statistical Analysis and Software

The software Jamovi 2.2.5, (a graphical interface of R version 1.6, https://jamovi.org) was used for the statistical analysis and processing of the graphs. Statistical analysis was carried out for structural parameters, production parameters and soil chemical–physical parameters. After an initial exploratory analysis of the data and the calculation of descriptive statistics, we proceeded with the ANOVA test. In figures representing boxplots, the lower and upper hinges correspond to the first and third quartiles (the 25th and 75th percentiles). Whiskers extend from the hinges to the largest or smallest value no further than 1.5 × IQR from the hinge (where IQR is the inter-quartile range). Data beyond the end of the whiskers are plotted individually.

In the same study, the one-way ANOVA test was used to test the difference between the averages of the above-mentioned parameters in the two types of conduction, agroecological and conventional, in order to assess how the treatment might have affected production yield.

In view of the observed variability within the data, Welch’s ANOVA was used to test the homogeneity of variances [22].

3. Results

The results of the ANOVA of the chlorophyll and leaf nitrogen content data, together with the most relevant and statistically significant structural and productivity parameters of the treatments, are described in

Table 3. One-way ANOVA (Welch’s) of chlorophyll content, leaf nitrogen content, field structural parameters and production parameters in the two treatments; AE = agroecological and CP = conventional. Notes * p < 0.05, ** p < 0. 01, *** p < 0.001.

Parameters	F	df1	df2	p-Value
SPAD (SPAD units)	29.5	1	178	<0.001 ***
Leaf nitrogen (mmol/m2)	29.5	1	178	<0.001 ***
Total dry weight (g)	53.9	1	113	<0.001 ***
Dry weight roots (g)	11.6	1	145	<0.001 ***
Dry weight stem (g)	13.3	1	156	<0.001 ***
Total moisture (%)	6.0	1	175	0.015 **
Moisture content (%)	5.39	1	178	0.021 *
Leaves moisture (%)	10.0	1	174	0.002 **
Stem height (cm)	17.6	1	143	<0.001 ***
Number of ears	8.68	1	170	0.004 **
Leaf partitioning coefficient (%)	42.2	1	176	<0.001 ***
Ears partitioning coefficient (%)	0.39	1	178	0.534
Roots partitioning coefficient (%)	2.91	1	176	0.090
Stem partitioning coefficient (%)	7.07	1	176	0.009 **
Leaf area (cm2)	3.83	1	164	0.052
Leaf area index LAI	42.9	1	159	<0.001 ***
Specific leaf area (m2 kg−1)	8.27	1	52.2	0.006 **
Harvest index (%)	2.44	1	178	0.120
Seed yield (t ha−1)	1.26	1	17.7	0.277

. The differences between the mean values of each parameter compared in the two treatments are described in tabular form in the Supplementary Materials (Table S2).

3.1. Analysis of Chlorophyll (SPAD) and Foliar Nitrogen Content

An average value of 49.8 SPAD units was observed in the agroecological treatment, indicating a higher chlorophyll content than the 45.4 SPAD units in the conventional treatment (Figure 4). This difference was found to be statistically significant (p-value < 0.001). With regard to the results of the foliar nitrogen analysis, the difference is also statistically significant (p-value < 0.001), with the AE thesis showing a higher nitrogen concentration in the leaves (153 mmol m⁻²) compared to the CP (137 mmol m⁻²) (Table S2 of the Supplementary Materials).

3.2. Analysis of Plant Parameters

3.2.1. Weight of Dry Biomass

The total dry biomass production was statistically significant (p-value < 0.001) and higher in the conventional treatment (26.7 g), while in the agroecological thesis it was lower (15.7 g, Table S2 of the Supplementary Materials). In the AE treatment, there was a lower level of dry root biomass production (Figure 5) of 3.30 g compared to the CP treatment, which had a higher dry biomass value of 4.24 g. This difference was statistically significant between the two theses, with a p-value < 0.001. The dry biomass, in terms of stems (Figure 5), was higher in the conventional trial (6.69 g) than in the AE trial (5.47 g). Again, this was statistically significant (p-value < 0.001).

3.2.2. Moisture Content

The results of the % total moisture lost in the samples after the oven-drying process were statistically significant (p-value 0.021). The data show that the conventional treatment lost more water, about 46.4% on average, compared to 44.8% for the agro-ecological treatment, which is slightly lower (Figure 5; Table S2 of Supplementary Materials). The stems of the conventional treatment lost 57.8% moisture compared to the stems of the agroecological treatment, which lost less moisture—about 55.7% (Figure 5). This difference was statistically significant with a p-value < 001. The agroecological treatment shows that the percentage of moisture lost in the leaves (Table S2 of Supplementary Materials) is lower (29.6%); in contrast, the leaves of the conventional field lost more water (34.2%, Figure 5; Table S2 of Supplementary Materials), with a statistically significant difference of p-value 0.002.

3.2.3. Partition Coefficient: Leaves, Roots, Stems and Spikes

The values obtained from the statistical analysis of the foliar partition coefficient (Figure 5) between the two theses are 0.168% for AE compared to 0.137% for CP, with a statistically significant p-value (<0.001). Comparing the distribution rates within the ears (Figure 6), it can be seen from Table 3 that the difference is not statistically significant between the two conductions (p-value = 0.534). Nevertheless, the field managed with the agroecological technique showed a higher allocation of photosynthates within the ears (0.271%) than the field managed in the conventional way, which showed a lower value of allocation within the ears (0.265%, Figure 5). The root-allocated photosynthate content (Figure 5) in the AE treatment had a value of 0.212% compared to the slightly higher CP, which had a value of 0.230%, so the difference was not statistically significant (p-value 0.090). In the conventional treatment, the allocation of photosynthates within the stems was higher (0.368%) than under the agroecological treatment (0.350%) (Figure 5), with a statistically significant difference (p-value 0.009).

3.2.4. Height of Stems and Number of Ears

The result is that the agroecological field showed a higher stem height (67.1 cm) than the conventional field, with a lower value (62.6 cm), which was statistically significant (p-value < 0.001). The number of spikes was also statistically significant (p-value 0.004). Treatment CP produced a higher number of ears (6.7) than AE (5.9), as shown in Figure 6.

3.2.5. Leaf Area, Specific Leaf Area (SLA) and Leaf Area Index (LAI)

The conventional treatment showed a larger leaf area (Figure 7), with an average of 14.6 cm² compared to an average of 12.8 cm² in the agroecological treatment; however, the difference was not statistically significant. The highest value in terms of SLA (Figure 7) was 36.5 m² kg⁻¹ for the conventional thesis compared to the agroecological thesis, which showed a much lower SLA value (31.3 m² kg⁻¹), this being statistically significant, with a p-value 0.006. A higher LAI value (0.688) was observed under conventional farming compared to the value of 0.537 in the agroecological approach (Figure 7), with a statistically significant difference (p-value < 0.001).

3.3. Analysis of Production Parameters: Harvest Index and Seed Yield

A higher harvest index value was observed under agroecological management (0.355%) and lower one under conventional management, with a value of 0.326% (Figure 8). Statistical analysis did not reveal statistically significant differences (p value = 0.277), but a higher grain yield was observed in the agroecological treatment (2.36 t ha⁻¹) (Figure 8) compared to 2.07 t ha⁻¹ produced in the conventional treatment (Table S2 of Supplementary Materials). It must also be considered that in the agro-ecological field there was an excess of rain that caused some plots of land to flood, and many plants became lodged.

3.4. Profit and Loss Statement

The schedules of cultivation operations on the two plots of land are listed and described in

Table 4. Main operations conducted in the two treatments of AE (agroecological) and conventional (CP). The costs of mechanized operations include expenses for fuel, lubricant, depreciation of mechanical means and labor.

Period	Agroecological		Conventional
	Operation	Income/Expense (EUR)	Operation	Income/Expense (EUR)
September 2019	Soil minimum tillage by means of 2 harrowing operations	−90
February 2020	Purchase of field bean seed	−255
February 2020	Field bean sowing	−70
March 2020			Soil preparation (ploughing + 2 harrowing)	−180
March 2020			Purchase of basal dressing fertilizer—urea	−84
March 2020			Basal dressing fertilization	−80
April 2020			Purchase of sunflower seed	−219
April 2020			Sunflower sowing	−70
May 2020	Field bean green manuring	−120	Purchase of top dressing fertilizer—DAP 18–46	−96
May 2020			Top dressing	−70
August 2020			Sunflower harvesting	−440
August 2020			Sunflower selling	+1680
September 2020	Purchase of basal dressing fertilizer—DAP 18–46	−92	Purchase of basal dressing fertilizer—urea	−164
September 2020	VRT basal dressing fertilization	−120	Basal dressing fertilization	−80
September 2020	Soil ploughing	−90	Soil ploughing	−90
October 2020	Soil refiniment	−90	Soil refiniment	−90
November 2020	Purchase of durum wheat seed	−348	Purchase of durum wheat seed	−348
November 2020	Wheat sowing	−70	Wheat sowing	−70
January 2021	Common agricultural policy direct payment	+863	Common agricultural policy direct payment	+863
March 2021	Purchase of top dressing fertilizer—ammonium nitrate	−48	Purchase of top dressing fertilizer—ammonium nitrate	−120
March 2021	Top dressing	−70	Top dressing	−70
March 2021			Purchase of top dressing fertilizer—simple superphosphate	−585
March 2021			Top dressing	−70
July 2021	Wheat harvesting	−440	Wheat harvesting	−440
July 2021	Wheat selling	+2204	Wheat selling	+2117
January 2022	Common agricultural policy direct payment	+863	Common agricultural policy direct payment	+863
Net income		+2027		+2157

, which also shows the costs and revenues resulting from the various mechanized operations carried out in order to establish the economic balance of the cultivation.

3.5. Analysis of Chemical and Physical Soil Parameters

The soil parameters were compared between both the two experimental treatments and the two different time periods of 2019 and 2021. The results of ANOVA regarding the exchange calcium and cation exchange capacity (CEC) of the two theses in the pre-seeding period are reported in

Table 5. One-way ANOVA (Welch’s) for calcium exchange (ppm) and cation exchange capacity (meq 100 g⁻¹) in the two treatments; AE = agroecological and CP = conventional, before wheat sowing, on soil samples taken in October 2019. Note * p < 0.05.

Soil Parameters	F	df1	df2	p-Value
Exchange calcium (ppm)	6.05	1	17.5	0.024 *
CSC (meq 100 g−1)	5.07	1	17.8	0.037 *

.

For AE, the calcium exchange value was 6044 ppm, while in C it was 5660 ppm (Table S3 in Supplementary Materials). As far as cation exchange capacity is concerned, a clear difference was observed; it was higher for AE, with 32.7 meq 100 g⁻¹ versus 30.9 meq 100 g⁻¹ for CP.

The results of the statistically significant calcium exchange and cation exchange capacity (CEC) parameters found within the two theses during the harvest period are described in

Table 6. One-way ANOVA (Welch’s) for calcium exchange (ppm) and cation exchange capacity (meq 100 g⁻¹) in the two treatments; AE harvested = agroecological and CP harvested = conventional, after wheat harvest, on soil samples taken in September 2021. Notes * p < 0.05, ** p < 0.01.

Soil Parameters	F	df1	df2	p-Value
Exchange calcium (ppm)	8.91	1	17.2	0.008 **
CSC (meq 100 g−1)	8.40	1	17.6	0.010 *

.

As shown in Table S4 of the Supplementary Materials, the level of exchange calcium was higher in the agro-ecological field (5030 ppm) than in the conventional field (4690 ppm), and the difference between the AE harvested and CP harvested approaches was statistically significant (Table 6). Regarding the average difference in cation exchange capacity, it was highest in the agroecological field (27.8 meq 100 g⁻¹) and had a lower value in the conventional field (26.1 meq 100 g⁻¹).

3.6. Results of Climate Data and De Martonne Aridity Index

The thermo-pluviometric diagrams have been constructed and analyzed for the years 2018, 2019, 2020 and 2021 (Figure 9), together with the determination of the De Martonne index, in order to observe the trends of variations in the climate.

For the year 2018, the average annual temperature was 15.8 °C. The average rainfall for the year 2018 was 108.5 mm, with an extreme precipitation event occurring in July (

Table 7. Climate parameters; mean temperature (°C), mean precipitation (mm), De Martonne aridity index, climate types and extreme events, observed in the years 2018, 2019, 2020 and 2021.

Year	Average		Aridity Index	Climate Type	Extreme Events
	Temperature (°C)	Rainfall (mm)
2018	15.8	108.5	50.4	Wet	1
2019	16.1	60.4	25.4	Sub-humid
2020	16.3	40.9	18.7	Semi-arid Mediterranean	1
2021	16.1	47.7	21.9	Sub-humid	1

). The calculation of the De Martonne index measuring aridity produced a result of 50.4, which corresponds to the “wet” climate type (Table S1 of the Supplementary Materials).

In the year 2019 (Table 7), the average annual temperature was 16.1 °C, and the average annual precipitation was 60.4 mm, considering that no data were available in November (a month in which significant precipitation is observed in the Mediterranean region). The De Martonne index calculated for the year in question was 25.4, indicating a “sub-humid” climate type.

The year 2020 recorded an average temperature of 16.3 °C. The average rainfall over this year was around 40.9 mm, suggesting a reduction in precipitation compared to previous years (Table 7). Here, the De Martonne’s aridity index was 18.7, which corresponds to a “semi-arid Mediterranean-type” climate. For the year 2021, the average temperature recorded was 16.1 °C. The average rainfall for the entire year was 47.7 mm higher than in the previous year (Table 7). Based on the calculation of the De Martonne index, with a result of 21.9, the climate can be classified as “sub-humid”.

4. Discussion

4.1. Chlorophyll (SPAD) and Foliar Nitrogen Content

The agroecological treatment (AE) showed a higher chlorophyll content than the conventional treatment (CP) due to the improved N/P ratio induced by the optimized phosphorous and nitrogen fertilization, which induced a better nutritional status in the plant [23]. The higher foliar nitrogen content in agroecological treatment (AE) plants, in line with the results for chlorophyll content, indicates a higher photosynthetic and productive efficiency of the plants.

4.2. Structural and Productive Parameters

Higher total dry biomass was observed under CP treatment, especially in roots and stems, a consequence of the excess nitrogen fertilization, which induces higher vegetative efficiency [24]. The dry biomass levels of leaves and ears were similar in both AE and CP systems, indicating comparable reproductive efficiency, despite the higher vegetative growth in the CP treatment. Dry biomass and straw weight showed a behavior similar to that of fresh biomass: a higher value was observed in the plants of the CP system, again due to the greater availability of N in the soil, which induced vegetative growth.

This was also evidenced by a higher water content within the biomass observed in the different organs of the CP plants, especially in leaves and stems. The higher water content of plants under CP can be traced back to the excess nitrogen in the soil, which leads to a moisture content and vegetative vigor that increase with increasing N availability.

Considering the stem height, the AE treatment yielded taller stems than the CP treatment, which should result in the better capture of solar radiation, something that the lower plant density in the agroecological treatment (140 plants m⁻²) should also have contributed to. One study [25] shows how solar radiation interception and utilization are improved in low-density conditions.

The number of ears was higher in the conventional field (CP) due to the high vegetative vigor caused by over-fertilization with nitrogen. However, this does not automatically mean that there is a higher crop yield. In fact, it is reproductive efficiency, and not vegetative efficiency, that positively influences yield. In the case of the conventional field, reproductive efficiency was lower.

A higher photosynthetic partitioning was observed in AE leaves and ears, indicating good nutrient assimilation efficiency by the plant and an optimal N/P ratio in the soil during cultivation due to optimized fertilization. The high photosynthetic efficiency in this treatment results in improved reproductive efficiency, despite the reduced level of nitrogen fertilization in the field. On the other hand, a higher partitioning coefficient in roots and stems was observed in the conventional field (CP), confirming the higher vegetative vigor induced by the excessive doses of nitrogen supplied to the soil, with more biomass in the organs with a vegetative function (roots and stems) and less in those with a reproductive function (leaves and ears). Under the conventional treatment, the plant invests energy in vegetating to the detriment of reproduction (seed production).

Higher values in terms of leaf area, leaf area index and specific leaf area were observed for plants in the CP field, which, despite having a higher leaf biomass and more leaf area, had low chlorophyll and leaf nitrogen values, denoting a lower photosynthetic efficiency than agroecological. A greater leaf area increases the interception of solar radiation [26], but, in the case of a greater LAI (more leaf layers), less radiation reaches the leaves at the bottom of the plant. Therefore, the leaves at the top, intercepting most of the radiation, achieved light saturation, and some of the radiation could not be used for photosynthesis. Furthermore, the leaves of the lower layers, which were poorly lit, contributed a little less to the photosynthesis process, resulting in low photosynthetic efficiency, with a reduction in potential biomass production fueled by the obtainable radiation levels. The increases in LAI and SLA are related to the higher density of plants in this treatment (157 plants m⁻²), which led to a decrease in photosynthesis. The higher density of plants in this treatment induced a lower uptake of light radiation due to the greater reciprocal shading. Since SLA is also a measure of the amount of leaf area that captures light per unit of dry matter invested, this highlights how the higher plant density also negatively influenced the interception of light by plants in the CP.

The harvest index [27] was not statistically significant, although the value was higher in the agroecological field (AE) than in the conventional field (CP). This difference is explained by the fact that in treatment AE, there was more grain than straw production, while conversely, in treatment CP, more straw than grain was produced. This is attributable to the massive and indiscriminate amount of nitrogen fertilization applied in the CP field, which induced plants to vegetate rather than reproduce.

Finally, the highest productivity in terms of seed yield, although not statistically significant, was observed for the agro-ecological (AE) treatment. The higher yield in agroecological conduction was mainly due to phosphate and nitrogen fertilization. Phosphorus is involved in photosynthesis and in seed formation. Nitrogen fertilization favors the vegetative phase; the reduced inputs in AE compared to conventional conduction increased the production yield in this type of conduction.

The comparison between the two management systems showed that AE improved the photosynthetic efficiency, with a consequential increase in the production parameters of the crop, while the CP treatment showed an excess of vegetation, due to the unmethodical nitrogen fertilization that caused both a nutritional imbalance and an inefficient use of energy, which affected the productivity of the plants.

4.3. Income Statement

The results for the unit net profitability of the two crop rotation systems show a net difference of about EUR 130 ha⁻¹ year⁻¹ in favor of the conventional system, confirming the sustainability of agroecological management. In fact, in the two-year period under consideration, the agroecological system, characterized by the lack of a second cash crop in addition to the conventional one, the sunflower, achieved a rewarding economic result. In addition, there was an environmental benefit due to the replacing of chemical fertilization with green manure to produce a leguminous crop.

4.4. Soil Parameters: Calcium Exchange and Cation Exchange Capacity

A higher calcium exchange content was found in the results of the AE treatment (Table 5), in both the pre-sowing and harvest periods of wheat (Tables S3 and S4 of Supplementary Materials), leading to an improvement in soil structure. The higher calcium exchange content in AE resulted in a structure-improving advantage that reduced the negative effects of active limestone [28]. The higher cation exchange capacity (CEC) observed in this treatment (AE) was also induced by the higher calcium content [29].

4.5. Climatic Parameters

In analyzing the climatic variation over the time span from 2018 to 2021, we saw that the De Martonne index [21], calculated for each year, evidenced a transition from a humid climate (year 2018) to a sub-humid climate (year 2019), and towards a semi-arid climate (year 2020), before returning to a sub-humid climate in 2021, due to the higher rainfall in the latter year. Extreme rainfall events were also observed in 2018 (July), 2020 (December) and 2021 (January). The excess rainfall in December 2020 and January 2021 (the years in which the experimental trial was carried out) produced flooding on the lower slope of the field, especially at the lowest elevation points (due to the topographical depression), and this was most evident in the agro-ecological treatment (AE). In the latter, problems of lodging and plant death occurred due to the probable effect of the increased amount of active limestone in the soil, and the substitution of calcium (Ca) and magnesium (Mg) in the exchange complex with sodium (Na) and potassium (K), as well as root asphyxia during flooding.

4.6. General Considerations

As already observed regarding the photosynthetic, morphological and productive parameters described in Section 4.1 and Section 4.2, the AE system compared to CP showed several beneficial effects on crop sustainability. The improvement in soil physical–chemical fertility, due in particular to the variation in exchangeable Ca and CEC, the better adaptation to extreme climatic events that supported crop productivity, and the maintenance of a satisfactory economic level indicate the greater resilience of management with AE, compared to CP, which seems to be unbalanced and unmethodical.

Although few research results on the effects of the integration of agroecology and precision agriculture on durum wheat have been found in the literature, similar specific results to those of our research have been obtained by other authors.

The benefits and opportunities of adopting agroecological practices in the North African region have been described by Boutagayout et al. [30]. In particular, on wheat cultivation, AE showed the potential to replace chemical means.

Dargie et al. [31] studied the responses of wheat to fertilization with balanced rates of nitrogen, phosphorus, potassium, and sulfur on different soil types and agroecologies in Ethiopia. The agronomic efficiency of wheat decreased with increasing rates of N and P on all investigated soil types, as we observed in the CP treatment.

Site-specific nitrogen management maximized profit, with consequential increases in the net returns of the wheat belt in the USA [32].

The adoption of organic agriculture techniques reduced chemical inputs from the agroecological environment, replacing synthetic inputs with cover crops used as green manure [33].

5. Conclusions

In the face of the current climate change scenario, wherein agriculture is both a victim and offender, and in anticipation of an increasing demand for food, the adoption of agriculture strategies based on sustainability is an opportunity, but also an urgent need. The use of agroecological practices such as green manure, crop rotation combined with minimal tillage, and precision agriculture to manage site-specific fertilization has been confirmed as a powerful approach to achieve the following:

• Agriculture that is more environmentally friendly (both ecological and agroecological) by virtue of improving soil quality, reducing fertilizer excesses, and choosing varieties that are tolerant to new climatic conditions;
• Agriculture that benefits in terms of yield of output (equal to or greater than conventional agriculture, driven by productivity and profit alone).

The variable-rate fertilization adopted as described in the agroecological treatment had various advantages, which can be described from the following points of view: economic, with the reduction in fertilizer purchase costs; energy, due to the lower use of agricultural equipment; and environmental, due to the reduction in nitrogen fertilization, with the consequent reduction in the risk of nitrate leaching into groundwater.

Considering that the limitation of the study is related to its location in a single geographical area of the Mediterranean, it is appropriate that similar research be carried out in other environments in order to confirm the results of this research.