Leveraging Precision Agriculture Principles for Eco-Efficiency: Performance of Common Bean Production Across Irrigation Levels and Sowing Periods

Abstract

Optimizing irrigation and sowing schedules is critical for enhancing crop performance and resource efficiency, especially in water-limited environments. However, the balancing the trade-offs between crop yield, energy use, and environmental impacts remains a complex challenge. This study investigates the eco-efficiency of common bean (Phaseolus vulgaris L.) cultivation in Vojvodina region (Serbia) under three irrigation regimes (100%, 80%, and 60% of crop evapotranspiration—ET_c) and three sowing periods (mid-April, late May/early June, and late June/early July). A combined energy analysis and cradle-to-farm gate Life Cycle Assessment (LCA) was employed to assess sustainability trade-offs. Results show that early sowing with full irrigation achieved the highest crop yields, energy use efficiency, and net energy gain while minimizing specific energy input. However, this strategy also incurred the greatest environmental burden due to elevated water and fertilizer inputs. In contrast, late sowing and deficit irrigation reduced environmental impacts at the expense of productivity and energy performance. The most balanced outcome—combining acceptable yield with lower environmental pressure—was observed under early sowing (mid-April) and moderate deficit irrigation (60% of ET_c). Importantly, the study reveals discrepancies between energy and environmental assessments; energy analysis favors high-yield, high-input systems, whereas LCA emphasizes environmental burdens per unit area, often favoring low-input strategies. These findings underscore the need for integrated, site-specific management approaches that optimize both agronomic performance and environmental sustainability, particularly under growing climate and resource constraints.

1. Introduction

Common beans (Phaseolus vulgaris) are considered the most important legume for human consumption worldwide [1]. They are an important source of protein and fiber [2,3,4] and are regarded as a cost-effective food. Approximately 26 million tons of this crop are produced annually [5]. In Serbia, Phaseolus vulgaris has been a significant component of the traditional diet and life for centuries [6]. Today, they are cultivated on approximately 9100 hectares, with an average yield of around 1000 kg ha⁻¹ [7], primarily under rainfed (non-irrigated) conditions [8].

Climate change is increasingly undermining the viability of traditional agricultural systems, including bean (Phaseolus vulgaris) cultivation [9]. Observational studies [10,11] of climate parameters indicate a trend of increasing air temperatures, longer dry periods, more significant precipitation variability, and more frequent extreme weather events. Climate projections for Europe to the end of the century also indicate an increased likelihood of risks (heat and cold waves, river and coastal floods, droughts, wildfires, and windstorms), especially in the Western Balkans [12]. Specifically, by 2050, Serbia is projected to experience a summer temperature increase of +1.89 °C, a decrease in annual precipitation by 38.5 ± 22.2 mm, and an increase in evaporative demand of 65.7 ± 7.9 mm per year [13]. These changes may shorten crop cycles, shift irrigation needs, and expand the suitability for perennial crops in northern regions. To adapt, farmers must implement climate-smart strategies such as optimized irrigation, adjusted planting dates, soil conservation, drought-tolerant cultivars, and precision agriculture. While such approaches aim to sustain productivity and improve resource efficiency, some interventions, if not context-specific and well-managed, may inadvertently increase environmental pressures or resource competition [9].

Low-input cropping systems were introduced in Western Europe as a response to the environmental degradation associated with intensive agriculture. While these systems aim to mitigate environmental impacts, their benefits are often partially offset by reduced yields [14]. The sustainability trade-offs between low- and high-input systems are echoed in broader discussions about the environmental consequences of different intensification pathways [15]. These include comparisons between rain-fed and irrigated systems [16], organic versus conventional farming [17], diversified low-input versus conventional short-rotation systems [18], controlled-environment agriculture versus open-field production [19], and short versus long food supply chains [20]. This ongoing discourse highlights the pressing need for site-specific crop management strategies that promote climate adaptation, enhance energy efficiency, and reduce environmental impacts. Developing context-sensitive approaches is essential for ensuring a sustainable transition in food production systems and for promoting responsible resource use in increasingly variable agro-ecological conditions [21].

Energy and environmental assessments are becoming essential tools for guiding decision-making toward sustainability. Among these, the Life Cycle Assessment (LCA) methodology has gained prominence as a robust framework for quantifying both direct and indirect environmental impacts associated with crop production processes [21]. Numerous studies have been conducted on the LCA, especially on crops grown for biofuel production targeting a reduction in GHG [22]; field crops grown for oils [23]; fruits such as apples [24], citrus [25,26], and avocado [27]; and major field crops such as wheat, maize, and barley [28,29]. Studies have been performed on vegetable LCA, mainly on lettuce, onion, artichoke, and cabbage [30,31].

LCA studies on bean production have addressed a range of objectives across different countries. In Sweden, Karlsson et al. [32] investigated the climate impact and energy use of faba bean biorefinery processing. In Greece, Abeliotis et al. [33] assessed different farming methods and bean varieties, showing that environmental outcomes vary depending on the functional unit: high-input systems performed better per unit of output, while low-input systems showed advantages per unit of land. Key environmental burdens stemmed from the electricity required for irrigation and sheep manure use. In Canada, Bamber et al. [34] quantified the resource and environmental impacts of dry bean and faba bean production across four provinces, identifying fertilizer production and field-level emissions as dominant impact drivers in all bean types. In Italy, Ilari et al. [35] in Italy demonstrated that input choices at the field level—such as herbicide reduction in green bean production—can influence environmental outcomes, even when processing is the primary impact contributor. In Turkey, Demir [36] evaluated the performance of dry bean production and revealed that optimizing irrigation and reducing fertilizer use could enhance energy efficiency. In Iran, Kazemi et al. [37] concluded that improving agricultural practices and reducing reliance on chemical fertilizers could enhance energy efficiency and reduce emissions. A study in Spain, by Romero-Gámez et al. [38], found that bean cultivation in a greenhouse causes lower environmental impacts compared to open-field cultivation. A 2024 study by Reina Pérez et al. [39] evaluated the environmental impacts of a small-scale organic PGI white bean production site in northern Spain, identifying electricity consumption as the main contributor to most impact categories. Additionally, Del Borghi et al. [40] assessed common bean cultivation under future climate scenarios and found that optimizing sowing time and genotype selection could substantially reduce environmental impacts. Their results showed that advancing sowing by 15 days and using improved genotypes (ideotypes) led to over 20% reductions in carbon footprint, cumulative energy demand, water footprint, and ecological footprint, highlighting the potential of climate-adapted agronomic strategies in enhancing the sustainability of legume production. In tropical regions, Araujo et al. [41] showed that replacing mineral nitrogen fertilization with rhizobial inoculation in common bean reduced environmental burdens by ~19% per hectare and 21% per ton, with minimal yield gain (2%). However, irrigation and phytosanitary treatments remained major impact drivers, underscoring the need for broader crop management improvements. In Italy, Moresi and Cimini [42] performed a cradle-to-grave LCA on malted beans, finding a carbon footprint of around 3.0 kg CO₂-eq/kg. Irrigation during the agricultural phase accounted for over 50% of the overall environmental impact, underscoring the importance of water-efficient practices and drought-tolerant bean varieties for improving sustainability.

Despite the growing body of research, irrigation regimes and sowing periods are rarely taken into account as key variables in energy balance and LCA studies evaluating the sustainability of legume production. This omission is particularly relevant for semi-arid regions like the Western Balkans, where common bean (Phaseolus vulgaris) is traditionally cultivated. While northern latitudes often rely on rain-fed systems, the Mediterranean and Balkan climates require supplemental irrigation, making both water and energy use central to the overall environmental impact [29,32,43]. While legumes are widely recognized for their role in sustainable agriculture, there is a lack of integrated studies that simultaneously assess the energy and environmental impacts of legume production under varied irrigation and sowing regimes, particularly in Serbia. This knowledge gap limits the development and adoption of climate-smart agronomic strategies, such as adjusted sowing schedules and optimized irrigation regimes, which are increasingly necessary to sustain eco-efficiency under climate stress. In Serbia, the absence of such assessments hinders both farm-level adaptation and policy-level planning, restricting efforts to align cropping practices with local agro-climatic realities. As emphasized by Todorovic et al. [44], LCA-based agronomic insights are essential for supporting sustainable farm development and guiding context-specific management decisions.

This study aims to identify sustainable crop management strategies that optimize both water and energy use in common bean (Phaseolus vulgaris) cultivation under varying irrigation levels and sowing periods in Serbia. By integrating energy analysis with a multi-indicator Life Cycle Assessment (LCA), the research evaluates the environmental performance and resource efficiency of different agronomic scenarios. The objective is to generate actionable insights into how irrigation and planting time influence the eco-efficiency of legume systems in semi-arid environments. Ultimately, the study seeks to support climate-resilient decision-making in agriculture, enabling improved adaptation to climate change while minimizing environmental burdens and resource competition.

2. Materials and Methods

2.1. Assessment Framework and Evaluation of Crop Cultivation Strategies

This study combines energy analysis and LCA, allowing us to identify efficient, site-specific bean cultivation strategies. A total of nine strategies were assessed, derived from the combination of three irrigation treatments with the three sowing periods.

The irrigation treatments included the following:

• Treatment FI—full irrigation at 100% crop evapotranspiration (ET_c);
• Treatment RI80—moderate deficit irrigation at 80% of ET_c;
• Treatment SI60—severe deficit irrigation at 60% of ET_c.

The sowing periods were as follows:

• Standard sowing period (SPI) in mid-April (typical for Serbia’s climate);
• Late sowing period (SPII) at the end of May/beginning of June;
• Very late sowing period (SPIII) in the third decade of June/early July.

Table 1. Description of the nine agronomic treatment combinations used in the study, based on three sowing periods and three irrigation regimes in common bean (Phaseolus vulgaris) cultivation.

Code	Sowing Period	Irrigation Level
SPI-FI	Standard (mid-April)	Full irrigation (100% ETc)
SPII-FI	Late (end May/early June)	Full irrigation (100% ETc)
SPIII-FI	Very late (late June/early July)	Full irrigation (100% ETc)
SPI-RI80	Standard (mid-April)	Moderate deficit (80% ETc)
SPII-RI80	Late (end May/early June)	Moderate deficit (80% ETc)
SPIII-RI80	Very late (late June/early July)	Moderate deficit (80% ETc)
SPI-SI60	Standard (mid-April)	Severe deficit (60% ETc)
SPII-SI60	Late (end May/early June)	Severe deficit (60% ETc)
SPIII-SI60	Very late (late June/early July)	Severe deficit (60% ETc)

shows the summary of the nine treatment combinations based on the three sowing periods and three irrigation levels.

The experiment design, treatments (irrigation and sowing periods), and agronomic measures are detailed in Lipovac et al. [8].

2.2. Energy Analysis Method

Understanding the energy dynamics in crop production relies heavily on the energy equivalents of various inputs used in agricultural practices.

Table 2. Energy equivalent values of inputs (categorized by type and energy source) and outputs in common bean production.

Parameter	Energy Equivalents	Unit	Type of Energy	Source of Energy	Reference
Human labor	1.96	MJ h−1	Direct	Renewable	[24,29,43]
Seeds bean	14.7	MJ kg−1	Indirect	Renewable	[45]
Pesticide, unspecified	193	MJ kg−1	Indirect	Non-renewable	[24,29,43]
Herbicide	238	MJ kg−1	Indirect	Non-renewable	[24,29,43]
Diesel fuel	56.31	MJ l−1	Direct	Non-renewable	[24,29,43]
Nitrogen (N)	66.14	MJ kg−1	Indirect	Non-renewable	[24,29,43]
Phosphorus (P)	12.44	MJ kg−1	Indirect	Non-renewable	[24,29,43]
Potassium (K)	11.15	MJ kg−1	Indirect	Non-renewable	[24,29,43]
Tractor machinery	62.7	MJ kg−1	Indirect	Non-renewable	[24,29,43]
Water, irrigation	1.03	MJ m−3	Direct	Renewable	[24,29,43]
Bean, yield	20	MJ kg−1	-	-	[45]

presents the energy equivalent values for the inputs consumed and the energy embodied in the common bean produced, organized by type and source of energy. Human labor, seeds, and irrigation water were classified as renewable energy sources, indicating their sustainability and potential for continuous use without depletion. Inputs such as pesticides, herbicides, diesel fuel, nitrogen, phosphorus, potassium, and tractor machinery are categorized as non-renewable due to their reliance on finite resources or fossil fuels.

To evaluate the energy performance of different cultivation strategies, four key energy indicators were calculated based on energy input, energy output, and yield data: energy use efficiency (EUE), energy productivity (EP), specific energy (SE), and net energy gain (NEG). These indicators provide insights into how effectively energy is utilized in agricultural practices. The EUE measures how effectively energy is converted into useful outputs, such as crops or food products. EP evaluates the system’s efficiency in using energy to meet production objectives, while SE assesses how well energy is transformed into valuable crop output. NEG determines whether the production process results in a net energy gain or loss.

$$\mathrm{E}\mathrm{n}\mathrm{e}\mathrm{r}\mathrm{g}\mathrm{y} \mathrm{u}\mathrm{s}\mathrm{e} \mathrm{e}\mathrm{f}\mathrm{f}\mathrm{i}\mathrm{c}\mathrm{i}\mathrm{e}\mathrm{n}\mathrm{c}\mathrm{y} \left(\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\right)={\displaystyle \frac{\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l} \mathrm{e}\mathrm{n}\mathrm{e}\mathrm{r}\mathrm{g}\mathrm{y} \mathrm{o}\mathrm{u}\mathrm{t}\mathrm{p}\mathrm{u}\mathrm{t} \left({\mathrm{M}\mathrm{J} \mathrm{h}\mathrm{a}}^{-1}\right)}{\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l} \mathrm{e}\mathrm{n}\mathrm{e}\mathrm{r}\mathrm{g}\mathrm{y} \mathrm{i}\mathrm{n}\mathrm{p}\mathrm{u}\mathrm{t} \left({\mathrm{M}\mathrm{J} \mathrm{h}\mathrm{a}}^{-1}\right)}}$$

(1)

$$\mathrm{E}\mathrm{n}\mathrm{e}\mathrm{r}\mathrm{g}\mathrm{y} \mathrm{p}\mathrm{r}\mathrm{o}\mathrm{d}\mathrm{u}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{v}\mathrm{i}\mathrm{t}\mathrm{y} ({\mathrm{k}\mathrm{g} \mathrm{M}\mathrm{J}}^{-1})={\displaystyle \frac{\mathrm{C}\mathrm{r}\mathrm{o}\mathrm{p} \mathrm{o}\mathrm{u}\mathrm{t}\mathrm{p}\mathrm{u}\mathrm{t} \left({\mathrm{k}\mathrm{g} \mathrm{h}\mathrm{a}}^{-1}\right)}{\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l} \mathrm{e}\mathrm{n}\mathrm{e}\mathrm{r}\mathrm{g}\mathrm{y} \mathrm{i}\mathrm{n}\mathrm{p}\mathrm{u}\mathrm{t} \left({\mathrm{M}\mathrm{J} \mathrm{h}\mathrm{a}}^{-1}\right)}}$$

(2)

$$\mathrm{S}\mathrm{p}\mathrm{e}\mathrm{c}\mathrm{i}\mathrm{f}\mathrm{i}\mathrm{c} \mathrm{e}\mathrm{n}\mathrm{e}\mathrm{r}\mathrm{g}\mathrm{y} \left({\mathrm{M}\mathrm{J} \mathrm{k}\mathrm{g}}^{-1}\right)={\displaystyle \frac{\mathrm{E}\mathrm{n}\mathrm{e}\mathrm{r}\mathrm{g}\mathrm{y} \mathrm{i}\mathrm{n}\mathrm{p}\mathrm{u}\mathrm{t} \left({\mathrm{M}\mathrm{J} \mathrm{h}\mathrm{a}}^{-1}\right)}{\mathrm{C}\mathrm{r}\mathrm{o}\mathrm{p} \mathrm{o}\mathrm{u}\mathrm{t}\mathrm{p}\mathrm{u}\mathrm{t} \left({\mathrm{k}\mathrm{g} \mathrm{h}\mathrm{a}}^{-1}\right)}}$$

(3)

$$\mathrm{N}\mathrm{e}\mathrm{t} \mathrm{e}\mathrm{n}\mathrm{e}\mathrm{r}\mathrm{g}\mathrm{y} \mathrm{g}\mathrm{a}\mathrm{i}\mathrm{n} \left({\mathrm{M}\mathrm{J} \mathrm{h}\mathrm{a}}^{-1}\right)=\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l} \mathrm{e}\mathrm{n}\mathrm{e}\mathrm{r}\mathrm{g}\mathrm{y} \mathrm{o}\mathrm{u}\mathrm{t}\mathrm{p}\mathrm{u}\mathrm{t} \left({\mathrm{M}\mathrm{J} \mathrm{h}\mathrm{a}}^{-1}\right)- \mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l} \mathrm{e}\mathrm{n}\mathrm{e}\mathrm{r}\mathrm{g}\mathrm{y} \mathrm{i}\mathrm{n}\mathrm{p}\mathrm{u}\mathrm{t} \left({\mathrm{M}\mathrm{J} \mathrm{h}\mathrm{a}}^{-1}\right)$$

(4)

2.3. Life Cycle Assessment (LCA) Method

This method has four steps, which include goal and domain definition, inventory analysis, impact assessment, and interpretation of results.

In the first phase, the purpose and functional unit were defined. A “cradle to grave” cradle-to-farm gate (Figure 1) perspective was applied to analyze the environmental loads of the investigated strategies. Two functional units were used: mass-based yield (expressed as 1 kg or 1 ton of harvested bean grain) to represent production intensity, and 1 hectare of cultivated land to reflect input and land use efficiency. As the farming system serves a single purpose—food production—no allocation procedures were necessary, and all inputs and associated environmental impacts were fully attributed to common bean cultivation.

In the life cycle inventory, a table of lists of inputs (materials and energy) and outputs (emissions to the environment) per functional unit was produced. Input data included seeds, human labor, fuel, water, electricity, chemical fertilizers, pesticides, and machinery use, while the output was bean yield. These data were sourced from previous research [46], and a scaling model based on regional agronomic norms was applied to adapt the experimental field data to reflect typical local farming practices. This approach aligns with standard LCA methodology when site-specific inventory data are not directly available and introduces a realistic representation of average farm-level operations (

Table 3. Data input for common beans averaged over typical farming practices.

Parameter		Seeds	Irrigation	Electric Pump	Diesel	Tractor	Plant Protection	Human Labor	Crop Yield
Unit		kg ha−1	m3 ha−1	MJ ha−1	l ha−1	h ha−1	kg ha−1	h ha−1	kg ha−1
Treatments	SPI-FI	150	1770	31.87	80.33	10.33	0.80	60.00	4783.3
	SPII-FI	158	1330	39.30	84.67	12.00	1.80	40.00	4480.0
	SPIII-FI	157	1880	56.37	85.33	12.33	2.60	46.67	3463.3
	SPI-RI80	150	1365	24.57	80.33	10.33	0.80	60.00	4530.0
	SPII-RI80	158	930	27.73	84.67	12.00	1.80	40.00	4090.0
	SPIII-RI80	157	1380	41.33	85.33	12.33	2.60	46.67	3416.7
	SPI-SI60	150	1095	19.70	80.33	10.33	0.80	60.00	4333.3
	SPII-SI60	158	770	22.70	84.67	12.00	1.80	40.00	3626.7
	SPIII-SI60	157	1040	30.97	85.33	12.33	2.60	46.67	2793.3

). The irrigation infrastructure (pump and on-farm system) was modeled using generic data retrieved from AUSLCI [47] and the Ecoinvent database [48].

These indicators collectively offer a comprehensive view of the energy intensity and sustainability of each cultivation system. Emissions resulting from fertilizer application were calculated according to the rules defined by the Intergovernmental Panel on Climate Change [49] and additional sources [50]. Nutrient emissions were assessed in terms of ammonia (NH₃) and nitrogen oxides (NO_x) emissions to the air, nitrate leaching to groundwater, phosphorus emissions to water, and gaseous emissions such as nitrous oxide (N₂O) to the air—primarily resulting from nitrogen fertilizer application. Both direct and indirect N₂O emissions were accounted for; indirect emissions included N₂O released from atmospheric nitrogen deposition on soils and water surfaces, as well as from nitrogen leaching and runoff. Data from the EcoInvent database [48] were used to calculate the emission rate of indirect environmental pollutants due to seeds, chemical fertilizers, diesel fuel, and plant protection products.

Evaluated potential environmental impacts (e.g., carbon footprint, water use, toxicity) were based on the inventory data. The IMPACT World+ method was applied to quantify environmental impacts, using the latest characterization factors from Agez et al. [51]. The Ecoinvent version 3.1 was used for background system and emission data. The software used to perform the calculation was openLCA v.2.0.4 (GreenDelta, Berlin, Germany) [52], an advanced open-source software for LCA of products and services.

All steps culminated in the interpretation of results, intending to identify limitations, recommendations, and key findings to inform decision-makers about the environmental impacts and resource efficiency of different irrigation strategies and sowing periods for common bean cultivation.

3. Results and Discussion

3.1. Energy Balance and Performance Indicators

Table 4. Energy balance of Serbian bean grown under different irrigation (FI, RI80, and SI60) and sowing periods (SPI, SPII, and SPIII).

Strategy/Treatment	Energy Input (MJ ha−1)	Energy Output (MJ ha−1)	Net Energy Gain (MJ ha−1)
SPI-FI	18,167.70	95,666.60	77,498.90
SPII-FI	17,424.30	89,600.00	72,175.60
SPIII-FI	18,868.10	69,266.60	50,398.50
SPI-RI80	17,149.20	90,600.00	73,450.70
SPII-RI80	16,418.40	81,800.00	65,381.50
SPIII-RI80	17,610.70	68,333.30	50,722.50
SPI-SI60	16,470.10	86,666.60	70,196.40
SPII-SI60	16,016.00	72,533.30	56,517.20
SPIII-SI60	16,755.70	55,866.60	39,110.90

presents the energy balance across the different treatment combinations. Total energy input shows a slight variation among treatments, with the highest recorded in the latest sowing period (SPIII) under full irrigation at 18,868.10 MJ ha⁻¹, and the lowest in SPII with 80% of ET_c at 16,418.40 MJ ha⁻¹. Full irrigation (FI) consistently leads to the highest energy input for all sowing periods due to increased water application. In general, energy input increased with both delayed sowing and higher irrigation levels. These results align with Perrin et al. [53], who reported energy inputs for vegetable production ranging from 8500 MJ ha⁻¹ to 16,000 MJ ha⁻¹, slightly lower than the values observed in this study. Their research emphasizes the significant impact of environmental conditions on crop yield and energy efficiency. Similarly, Romero-Gámez et al. [38] reported energy inputs of approximately 12,000 MJ ha⁻¹ for open-field green bean cultivation, which are lower than our full irrigation treatments but comparable to those under moderate (RI80) and severe (SI60) deficit irrigation. Demir [36] reported 16,717.86 MJ ha⁻¹ in the Central district of Kırklareli province of Turkey, while Kazemi et al. [37] calculated a total input energy of 13,833.7 MJ ha⁻¹ in bean systems in Iran. In Lithuania, Šarauskis et al. [54] found input energy values ranging from 11,419 to 12,613 MJ ha⁻¹, depending on tillage intensity.

The higher energy input values observed in our study, particularly under full irrigation and delayed sowing, reflect both increased irrigation requirements and seasonal inefficiencies. However, part of the difference also arises from methodological variations between studies, including differences in input accounting, boundary definitions, and energy coefficients. These methodological factors can significantly influence the estimated energy balance, underscoring the importance of consistent frameworks in cross-study comparisons. Nevertheless, our results confirm that optimizing water use and planting dates play a crucial role in reducing energy inputs in bean production systems.

The highest energy output was observed under the standard sowing period (SPI) with full irrigation (FI), reaching 95,666.67 MJ·ha⁻¹, while the lowest output occurred under SP-II with moderate deficit irrigation (RI80) at 68,333.30 MJ·ha⁻¹. This highlights the clear productivity advantage of combining early sowing with full irrigation. Moderate deficit irrigation (RI80) led to energy output reductions of 5.30%, 8.70%, and 1.35% for SPI, SPII, and SPIII, respectively. Severe deficit irrigation (SI60) caused even greater reductions—9.41%, 19.06%, and 19.34%—compared to full irrigation in the respective sowing periods. These trends indicate that energy output declines with both delayed sowing and reduced water supply. The energy output levels reported by Romero-Gámez et al. [38], approximately 50,000 MJ·ha⁻¹ for green beans, are comparable to those recorded under deficit irrigation in this study. Demir [36] reported 34,440 MJ·ha⁻¹ for bean production in Turkey, Kazemi et al. [37] calculated 65,060.44 MJ·ha⁻¹ in Iran, while Šarauskis et al. [54] reported values ranging from 77,200 to 88,133 MJ·ha⁻¹ in Lithuania depending on tillage intensity. Compared to these, our full irrigation treatments show substantially higher energy outputs, suggesting that the yield benefits of combining early sowing with full irrigation are notable. The variability in energy output among treatments and across studies reflects differences in agro-environmental conditions and farm management practices. As noted by Bamber et al. [34] and Ilari et al. [35], such variability underscores the significant influence of irrigation strategies on the energy efficiency and productivity of legume cropping systems.

Net energy, calculated as the difference between energy output and input, ranged from 39,110.91 MJ·ha⁻¹ to 77,498.93 MJ·ha⁻¹ across treatments. The SPI_FI treatment achieved the highest net energy (77,498.90 MJ·ha⁻¹), indicating the most favorable balance between input and output. This finding aligns with Abeliotis et al. [33], who reported high energy efficiency in common bean production under optimal irrigation in Greece. Similarly, Bamber et al. [34] observed that full irrigation in Canadian faba and dry bean production resulted in higher net energy and yields. Net energy consistently decreased with later sowing dates and more severe irrigation deficits, with particularly sharp declines from SPII to SPIII under 60% ET_c. These results support Ghasemi-Mobtaker et al. [28], who concluded that reduced water availability directly limits yield and energy performance.

Figure 2 illustrates the distribution of energy inputs across treatments, differentiated by sowing period (SPI, SPII, SPIII) and irrigation level (FI, RI80, SI60). Among all input categories, fertilizers emerged as the most energy-intensive component, accounting for 28.7% to 34.8% of total energy use, followed by mechanization (27.52% to 32.12%). Irrigation energy demand contributed a smaller but notable share, ranging from 11.8% to 25.1%. Treatments under full irrigation exhibited higher electricity consumption, up to 18%, due to greater water pumping and distribution needs. Later sowing periods generally require more energy for irrigation, particularly under full irrigation. These results also emphasize the system’s heavy reliance on mechanized operations, consistent with findings by Bamber et al. [34], who reported significant diesel consumption in Canadian bean production systems.

Figure 3 illustrates the patterns of energy consumption in bean production across different irrigation regimes (FI, RI80, SI60) and sowing periods (SPI, SPII, SPIII). In the full irrigation (FI) treatment, direct energy use accounted for 48.85%, decreasing to 45.84% in moderate deficit (RI80) and 43.92% in severe deficit (SI60). This shift indicates a greater share of indirect energy use under deficit irrigation, primarily due to reduced reliance on electricity for water pumping. Overall, the application of deficit irrigation shifts the energy structure toward more indirect energy inputs, such as fertilizers and chemicals. Renewable energy contributed between 27.10% and 28.29%, while non-renewable sources (e.g., diesel, machinery, plant protection products) remained dominant, accounting for 71.71% to 72.90% of total energy use.

In terms of sowing periods, SPI showed a direct/indirect energy split of 44.72%/55.28%, while SPII slightly increased direct use to 45.19%. SPIII recorded the highest direct energy use at 48.72%, reflecting its greater demand for electricity and water under less favorable climatic conditions. Renewable energy contributions remained consistent across sowing periods (27.44 to 28.15%), while non-renewable energy use ranged between 71.85% and 72.25%, closely mirroring the irrigation trends.

These findings align with Romero-Gámez et al. [38] and Demir [36], who reported high environmental impacts in green bean production due to intensive non-renewable energy use. Similarly, Bamber et al. [34] and Ilari et al. [35] observed that non-renewable inputs dominate energy profiles in bean systems, particularly in high-input scenarios like our SI60 treatment, where non-renewable energy reached 72.90%. However, our results differ slightly from Abeliotis et al. [33] and Kazemi et al. [37], who emphasized energy inefficiencies in bean production systems. Notably, our FI treatment demonstrated a relatively balanced energy profile (48.85% direct, 51.15% indirect), suggesting potential for improved energy use efficiency under optimal irrigation.

Moreover, our findings are in line with Del Borghi et al. [21], who identified similar energy distribution trends in legume production. This is particularly evident in SPII, where energy use appeared more balanced across sources. Overall, the study reinforces the predominant role of non-renewable energy in bean cultivation and echoes sustainability challenges identified across the literature. These patterns highlight the need for targeted interventions to enhance renewable energy integration and input efficiency in future legume production systems.

3.2. Energy Performance Indicators

Figure 4 presents the calculated energy performance indicators across different sowing periods and irrigation treatments. Energy use efficiency (EUE) was highest under the standard sowing period (SPI) across all irrigation regimes, ranging from 5.73 (SPI-FI) to 5.57 (SPI-S60). EUE consistently declined with later sowing periods, reaching a minimum of 3.33 in SPIII under full irrigation (SPIII-FI). This indicates that delayed sowing, even when combined with optimal water supply, cannot compensate for the reduction in yield and increased energy inputs. The trend was consistent across all irrigation treatments (FI, RI80, and SI60), suggesting that earlier sowing is inherently more energy-efficient and reinforcing the strong influence of irrigation level on overall energy performance (Figure 4A).

Specific energy—defined as the energy input required per kilogram of output—was lowest under SPI with full irrigation (SPI-FI) at 3.49 MJ·kg⁻¹, signifying the highest efficiency. As sowing was delayed, specific energy values increased, reflecting higher energy demands per unit of yield. This trend was evident across all irrigation treatments, emphasizing the penalty of late sowing in terms of input–output energy balance (Figure 4B). These findings are consistent with Romero-Gámez et al. [38], who reported that intensive irrigation can reduce specific energy values in vegetable crops by improving yield relative to input energy.

Regarding energy productivity, the highest value was again observed in SPI with full irrigation (SPI-FI) at 0.29 kg·MJ⁻¹, highlighting its superior performance in converting energy into crop yield (Figure 4C). Notably, deficit irrigation treatments under SPI also performed well, with both SPI-RI80 and SPI-SI60 achieving 0.28 kg·MJ⁻¹, indicating that moderate and even severe water limitations can still support high energy productivity when combined with optimal sowing time. In contrast, productivity declined sharply with delayed sowing, particularly under severe deficit irrigation, where SPIII-SI60 (0.18 kg·MJ⁻¹) and SPIII-FI (0.17 kg·MJ⁻¹) exhibited the lowest energy productivity due to poor yield performance despite relatively high input energy.

In conclusion, the standard sowing period (SPI) consistently demonstrated the highest performance across all energy-related indicators—energy use efficiency, specific energy, productivity, and profitability—particularly under full irrigation (100% ET_c). This combination effectively optimizes energy use while maximizing yields, offering the best return on energy investment. However, as sowing is delayed (SPII and SPIII), both energy efficiency and productivity decline significantly, especially under deficit irrigation. The least efficient strategy was observed under SPIII combined with severe water stress (SI60), where increased energy inputs yielded minimal returns, resulting in low overall efficiency and profitability.

While full irrigation generally led to better outcomes across all sowing periods, its advantages diminished with delayed sowing dates. In contrast, deficit irrigation treatments (RI80 and SI60) consistently underperformed, particularly when paired with late sowing, indicating that both timing and water availability are critical for optimizing resource use. These findings align with those of Abeliotis et al. [33] and Kazemi et al. [37], who emphasized the role of adequate water supply and timely planting in maximizing energy efficiency in legume systems. Similarly, Perrin et al. [53] reported substantial variability in energy inputs and outputs across different cropping systems, a trend also evident in this study. Moreover, the observations are supported by Los et al. [4], who highlighted the complex chemical composition and high energy demands of Phaseolus vulgaris L., underscoring the importance of precise management of both sowing dates and irrigation strategies to achieve energy-efficient and high-yielding common bean production.

3.3. Analysis of Life Cycle Environmental Impacts

Table 5. Weighted environmental indices for bean production under different irrigation (FI, RI80, and SI60) and sowing periods (SPI, SPII, and SPIII).

Treatment	LCA Single Score €/ha	LCA Single Score €/ton
SPI-FI	99,262.2	20,751.7
SPII-FI	90,163.0	20,125.7
SPIII-FI	102,081.1	29,474.8
SPI-RI80	90,837.5	20,052.4
SPII-RI80	81,842.3	20,010.3
SPIII-RI80	91,680.2	26,833.2
SPI-SI60	85,219.6	19,666.1
SPII-SI60	78,514.0	21,649.1
SPIII-SI60	84,607.5	30,289.1

presents the weighted environmental impact scores for common bean production under nine sowing–irrigation treatment combinations. The Life Cycle Assessment (LCA) single score—expressed in EUR 2003 per hectare (ha⁻¹) and per ton (t⁻¹) of beans—captures cumulative environmental burdens from emissions, resource use, and ecosystem damage. Results reveal significant variability among treatments, driven by both sowing period and irrigation intensity.

Per hectare, the lowest impact was recorded in SPII-SI60 (late sowing, severe deficit irrigation) at EUR 78,513.97 ha⁻¹, followed by SPII-RI80 (EUR 81,842.28 ha⁻¹). These treatments benefited from reduced water and energy inputs. In contrast, the highest per-hectare impact was found in SPIII-FI (very late sowing, full irrigation), reaching EUR 102,081.12 ha⁻¹, due to high input demands during less favorable climatic conditions. SP-I–FI (EUR 99,262.22 ha⁻¹) and SPIII-RI80 (EUR 91,680.18 ha⁻¹) also showed elevated impacts, indicating that later sowing—even with reduced irrigation—can significantly increase environmental pressure.

From a per-ton perspective, SP-I–SI60 (early sowing, severe deficit irrigation) achieved the lowest impact at EUR 19,666.06 t⁻¹, followed by SPII-RI80 (EUR 20,010.34 t⁻¹) and SPII-FI (EUR 20,125.68 t⁻¹). These combinations balanced modest inputs with stable yields. Conversely, SPIII-SI60 produced the highest impact per ton (EUR 30,289.09 t⁻¹), highlighting poor eco-efficiency under delayed sowing and severe water stress. SP-III-FI (EUR 29,474.82 t⁻¹) and SPIII-RI80 (EUR 26,833.22 t⁻¹) followed similar patterns, where reduced yields failed to offset input reductions.

These findings highlight key trade-offs. While full irrigation improves yields, it also increases environmental burdens. However, higher yields—such as those in SP-I–FI—tend to dilute environmental impacts across a greater output, resulting in lower per-ton scores. This is consistent with findings from Abeliotis et al. [33] and Ilari et al. [35], who emphasized that high-input, high-yield systems may achieve better resource efficiency per unit of product. On the other hand, low-input strategies like RI80 and SI60 are not automatically more sustainable; for instance, SP-I–SI60, despite reduced inputs, had more than triple the per-ton impact of SP-I–FI.

Overall, delayed sowing consistently worsened environmental performance, particularly under deficit irrigation. Early sowing, especially when paired with optimized irrigation levels, supports more efficient resource use and lower environmental impacts. These results confirm conclusions from Kazemi et al. [37] and Romero-Gámez et al. [38], stressing that yield stability is critical for achieving sustainable outcomes. Targeted alignment of sowing and irrigation remains a key strategy for minimizing the environmental footprint of common bean production in semi-arid systems.

Irrigation and fertilizer use emerged as the main contributors to environmental impacts across all treatments (Figure 5), but particularly in SPIII, where irrigation requirements were the greatest. This is consistent with findings by Bamber et al. [34], Demir [36], Kazemi et al. [37], and Moresi and Cimini [42], who similarly identified irrigation and fertilizer inputs as dominant environmental hotspots in bean production systems. Their studies underscore the critical need to optimize water and nutrient management, especially under late-season conditions, where resource demands intensify.

Table 6. Estimated midpoint and endpoint environmental impacts of bean production in Vojvodina region (northern Serbia).

Name	Unit	SPI-FI	SPII-FI	SPIII-FI	SPI-RI80	SPII-R80	SPIII-R80	SPI-S60	SPII-S60	SPIII-S60
Climate change, long term	kg CO2 eq (long)	4798.63	4037.73	5016.32	4082.93	3330.87	4132.74	3605.79	3048.12	3531.91
Climate change, short term	kg CO2 eq (short)	4990.43	4217.01	5211.97	4262.96	3498.51	4313.85	3777.95	3211.11	3703.12
Fossil and nuclear energy use	MJ deprived	62,813.43	50,962.62	66,138.73	51,727.4	40,013.5	52,452.3	44,336.6	35,633.9	43,145.6
Freshwater acidification	kg SO2 eq	27.17	22.57	28.44	22.89	18.34	23.15	20.03	16.64	19.55
Freshwater ecotoxicity	CTUe	271,201.0	238,274.6	282,007.3	239,705.	207,167	243,123	218,703	194,724	216,682
Freshwater eutrophication	kg PO4 P-lim eq	6.64	6.55	6.66	6.56	6.47	6.56	6.50	6.44	6.49
Human toxicity cancer	CTUh	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00
Human toxicity non cancer	CTUh	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00
Ionizing radiations	Bq C-14 eq	36,343.33	30,656.79	38,001.92	31,007.4	25,386.7	31,414.4	27,449.9	23,278.7	26,934.8
Land occupation, biodiversity	m2 arable land eq·yr	561.21	581.59	581.20	559.53	579.93	579.12	558.41	579.26	577.71
Land transformation, biodiversity	m2 arable land eq	39.14	38.94	39.20	38.95	38.76	38.97	38.83	38.69	38.81
Marine eutrophication	kg N N-lim eq	3.86	3.57	3.93	3.60	3.32	3.61	3.43	3.22	3.39
Mineral resources use	kg deprived	70.53	60.40	74.35	60.43	50.42	61.88	53.70	46.43	53.40
Ozone layer depletion	kg CFC-11 eq	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00
Particulate matter formation	kg PM2.5 eq	5.76	4.78	6.04	4.84	3.88	4.90	4.23	3.51	4.13
Photochemical oxidant formation	kg NMVOC eq	50.16	39.74	53.10	40.36	30.07	41.00	33.83	26.20	32.78
Terrestrial acidification	kg SO2 eq	48.67	43.49	50.09	43.85	38.73	44.14	40.63	36.83	40.09
Water scarcity	m3 world-eq	138,781.5	118,783.1	144,486.1	120,331	100,560	121,707	108,028.	93,271.3	106,218
Human health	DALY	1.33	1.21	1.37	1.22	1.10	1.23	1.14	1.05	1.13
Ecosystem	PDF·m2·yr	6314.85	5662.57	6496.36	5702.37	5057.65	5740.21	5294.05	4815.69	5226.03

presents the midpoint and endpoint environmental impact scores for common bean production under nine sowing–irrigation treatment combinations. Results per hectare show that the highest environmental burdens were observed under very late sowing with full irrigation (SPIII-FI), especially for climate change long-term (5016.32 kg CO₂-eq), fossil and nuclear energy use (66,138.73 MJ deprived), and water scarcity (144,486.17 m³ world-eq). These elevated impacts reflect the increased resource inputs needed to sustain yields during late-season cultivation, which aligns with previous studies such as Perrin et al. [53] and Ilari et al. [35], who highlighted that full irrigation and high fertilizer application substantially raise greenhouse gas emissions and acidification potential. In contrast, late sowing with severe deficit irrigation (SPII-SI60) showed the lowest per-hectare impacts in most categories, including climate change (3048.12 kg CO₂-eq) and energy use (35,633.92 MJ deprived), confirming that reduced input intensity helps lower environmental burdens—though with potential trade-offs in productivity.

When viewed per ton of beans produced, a different pattern emerges. Early sowing with full irrigation (SP-I–FI) delivered the best eco-efficiency, with the lowest climate change impact (1003.2 kg CO₂-eq) and fossil energy use (13,131.73 MJ deprived), owing to its high yield, which dilutes environmental burdens across more output. Conversely, very late sowing with severe deficit irrigation (SPIII-SI60) recorded the highest impacts per ton in nearly all categories, e.g., 1264.4 kg CO₂-eq for climate change and 15,445.94 MJ deprived for fossil energy use—reflecting both yield penalties and persistent baseline input demands. This pattern supports findings by Kazemi et al. [37] and Romero-Gámez et al. [38], who noted that while deficit irrigation reduces absolute inputs, reduced productivity can offset these benefits when assessed per unit of output. Overall, the results confirm that full irrigation (FI) treatments lead to higher environmental impacts across most indicators, particularly under delayed sowing scenarios. However, moderate (RI80) and severe (SI60) deficit irrigation strategies, while effective in lowering resource use, may not always yield environmental benefits per ton due to associated yield losses. These findings highlight the need to evaluate both per-hectare and per-ton impacts to capture sustainability trade-offs and reinforce the value of context-specific management strategies for enhancing eco-efficiency in water-scarce regions.

4. Discussion

Irrigated legume production is gaining strategic importance in semi-arid regions; however, integrated assessments of how irrigation levels and sowing periods affect yield, energy performance, and environmental impacts remain scarce, particularly within the framework of sustainable intensification (SI). While low-input and rainfed systems are often promoted for their reduced environmental burdens per hectare due to lower external input use [33,55], their typically lower yields can lead to higher environmental impacts per unit of output, challenging their overall sustainability in resource-limited contexts [14,44]. This trade-off presents a key challenge in resource-constrained systems, where both productivity and environmental performance must be optimized. As several studies have noted [16,29,56], environmental burdens tend to rise with increasing input intensity, though higher input use does not necessarily lead to improved eco-efficiency, defined as economic returns relative to environmental costs. This is confirmed by Todorovic et al. [44], who found that moderate irrigation and nitrogen input levels optimized eco-efficiency in wheat systems, while excessive inputs substantially increased environmental impacts without proportional yield benefits. Similarly, Kulak et al. [14] emphasized that optimal—not minimal—input levels are key to achieving high eco-efficiency, depending on system baselines and selected indicators. More recently, Weltin and Hüttel [57] demonstrated that realizing the benefits of SI depends not only on adopting sustainable practices but also on how effectively they are implemented. As noted by Baum and Bieńkowski [55], variability in water, energy, and agrochemical inputs explains much of the performance differences in both productivity and environmental outcomes. These insights collectively support the growing role of trade-off analysis in evaluating complex system-level interactions and targeting interventions across multifunctional agricultural landscapes.

Our findings reinforce this perspective, showing that both sowing period and irrigation level significantly influence yield, energy use, and environmental outcomes. Standard sowing (SPI) combined with full irrigation (FI) consistently produced the highest yields and energy use efficiency. These outcomes align with prior studies emphasizing the agronomic value of early planting and adequate soil moisture during key crop stages [33,34,37,44]. However, full irrigation also led to significantly higher water and energy consumption, illustrating the challenge of maximizing productivity while minimizing environmental burdens [36,37,38,44]. In contrast, delayed sowing and deficit irrigation strategies—particularly SPIII combined with severe water stress (SI60)—lowered yields and energy productivity but also reduced environmental impacts. This finding highlights the context-specific nature of sustainable cropping strategies. As noted by Abeliotis et al. [33] and Bamber et al. [34], environmental performance is highly sensitive to input levels and site-specific conditions, limiting the value of generalized recommendations. Our findings confirm the need for site-specific management strategies that account for local agroclimatic conditions, input availability, and production goals [33,55]. Furthermore, Del Borghi et al. [40] emphasized that adjusting sowing time, in their case by advancing sowing 15 days under future climate scenarios (RCP8.5), can maintain or even improve environmental performance. When paired with improved genotypes, they achieved over 20% reductions in key impact categories. While our study found delayed sowing combined with water-saving strategies to be effective in reducing impacts, both approaches support the broader insight that sowing time is a critical variable for enhancing the eco-efficiency of bean production under climate stress, whether through early planting or strategic delays, depending on the management context.

A clear trade-off emerged between crop productivity and environmental impact. While full irrigation improved yields, it also significantly increased greenhouse gas emissions, fossil fuel depletion, and acidification, similar to trends reported by other authors [34,35,36]. Conversely, deficit irrigation strategies reduced environmental burdens but came at the cost of productivity, particularly under delayed sowing. This duality reinforces the need to balance agronomic and environmental objectives. Our findings suggest that moderate deficit irrigation combined with early sowing may offer a viable compromise, providing acceptable yields while lowering input use and environmental pressure. Such trade-offs should guide management decisions, particularly in semi-arid regions facing increasing water scarcity. As Malki et al. [58] note, both natural and anthropogenic factors can degrade water quality, further underscoring the need for more efficient water management in agriculture.

Our hotspot analysis confirmed that fertilizer use and irrigation water application are the main contributors to environmental impacts, a finding consistent with previous LCA studies on legume and field crop systems [35,36,37]. Treatments using full irrigation (SPIII-FI) demonstrated the highest impacts, particularly in terms of climate change and freshwater acidification, consistent with Romero-Gámez et al. [38]. The results also highlight the water–energy nexus, where irrigation significantly influences not just water use, but also energy consumption and associated emissions. Efficient irrigation scheduling and optimized nutrient management are thus critical levers for reducing the environmental footprint of legume production systems.

While the study contributes to broader insights on sustainable legume cultivation, the specific agroecological context of Serbia must be considered when interpreting the findings. Data from the Vojvodina region were used, where fertile chernozem soils, a temperate continental climate, and relatively moderate rainfall provide favorable conditions for bean production. These characteristics support high yields and efficient resource use when management is optimized. However, the generalizability of the results may be limited in other areas of the Western Balkans or Mediterranean basin, where less fertile soils, greater water stress, or climatic variability may prevail. Furthermore, differences in irrigation infrastructure, farm size, and access to inputs may influence the applicability of these strategies. As such, while early sowing and moderate irrigation were most efficient in this setting, their performance should be carefully evaluated under different biophysical and socio-economic conditions before broader adoption.

From a practical perspective, this study provides useful guidance for farmers and advisors seeking to improve the eco-efficiency of common bean cultivation. Our results show that early sowing combined with moderate irrigation offers the best compromise between yield and sustainability. As noted in many LCA studies, functional units influence the interpretation of environmental impacts. When performance is evaluated per unit of land, low-input systems may appear preferable, but when analyzed per unit of product, high-input, high-yield systems often show better resource use efficiency. This duality highlights the need to clearly define sustainability goals, whether focused on conservation, productivity, or resilience.

While this study provides valuable insights, several limitations should be acknowledged. The analysis focused on a single crop and a specific regional context, which may limit the generalizability of the findings. Economic factors, social dynamics, and post-harvest stages were not included, despite their relevance in shaping environmental outcomes. Although a scaling model was used to reflect typical local practices, actual farm-level variability in resource access, technology, and management may influence real-world performance. Future research should include multi-year, multi-location trials and incorporate economic and social dimensions to better assess sustainability trade-offs. While the life cycle approach effectively captures key environmental indicators, Del Borghi et al. [40] note that it cannot fully predict the behavior of complex agricultural systems. Instead, LCA should be used as a decision support tool to identify strategic leverage points, rather than as a standalone optimization model. This underscores the need for context-specific agronomic data and localized strategies to enhance resource efficiency and sustainability in legume production systems.

5. Conclusions

Optimizing irrigation and sowing schedules is essential for enhancing crop performance in legume-based systems, yet the trade-offs between yield, resource use, and environmental impacts remain insufficiently explored. This study offers one of the first multi-indicator assessments of common bean production in Serbia, integrating energy analysis and Life Cycle Assessment (LCA) to evaluate how different irrigation levels and sowing periods influence sustainability outcomes. The findings are relevant not only for the Western Balkans but also for other semi-arid regions confronting increasing climatic stress and resource scarcity.

Results show that standard sowing (mid-April) combined with full irrigation (100% ET_c) consistently delivered the highest yields, energy use efficiency, and productivity. However, this scenario also resulted in elevated environmental burdens due to high water and energy inputs. In contrast, deficit irrigation strategies reduced environmental impacts but compromised yield and energy performance, particularly under delayed sowing conditions.

Notably, a discrepancy emerged between the energy analysis and the LCA results. While energy analysis highlights scenarios with higher output as more efficient—favoring high-input, high-yield strategies—LCA reveals that such approaches can substantially increase environmental burdens per hectare. Conversely, treatments with lower energy inputs (e.g., deficit irrigation) often appear more environmentally favorable in LCA, despite being less efficient in energetic terms. This divergence underscores the importance of using complementary methods to capture the full spectrum of sustainability performance. The findings reinforce the critical role of sowing time and irrigation management in balancing productivity with environmental responsibility. Early sowing consistently improved both energy and environmental outcomes compared to later sowing periods. Future management strategies should therefore be context-specific, aligning agronomic practices with local agro-climatic conditions and resource constraints, while accounting for trade-offs revealed by both energy and environmental assessments.