Biostimulant-based nutrient management for energy-efficient and low-carbon mustard cultivation: a life cycle assessment approach for sustainable development

Abstract

Agriculture is an energy-intensive sector contributing significantly to greenhousegas emissions (GHGE). Implementing sustainable agronomic practices, such as biostimulant, can offer a promising pathway to enhance energy efficiency and reduce environmental burdens. This study explores the novel formulation of a biostimulant derived from stinging nettle and weed biomass, evaluating its efficacy in improving energy efficiency, carbon footprint reduction, and economic viability in organic and conventional mustard (Brassica juncea) cultivation. A two-year field experiment (2020–21 and 2021–22) was conducted at Pantnagar, India, using a factorial randomized block design. Three formulations, stinging nettle-based (KJ1), common weed-based (KJ2), and a 50:50 blend of both (KJ3), applied at four different rates (500, 1000, 1500, and 2000 L ha^⁻1). A treatment with recommended doses of fertilizer (RDF) as control. Among treatments, KJ1D3 demonstrated superior agronomic and environmental benefits, yielding the highest grain (1572.4 kg ha^⁻1) and stalk biomass (4700.9 kg ha⁻1), comparable to RDF. KJ1D3 also exhibited optimal energy metrics, with the highest net energy gain (89,605 MJ ha^⁻1), energy efficiency (9.19), and lowest energy input per unit yield (1.68 MJ kg^⁻1). Similarly, lower climate impact intensity metrics (CIIM_energy, CIIM_economics)was observed with biostimulant treatments. Life cycle assessment showed the lowest carbon footprint under KJ1D3 (0.059 kg CO₂-eq kg^⁻1 grain), alongside the highest sustainability index (6.51). Economically, it yielded the highest net profit (USD 481.2 ha⁻1) and benefit-cost ratio (1.89). These results underscore the potential of biostimulant in promoting sustainable and energy-efficient agriculture, aligning with SDGs 2, 7, 12, and 13. Additionally, the study offers new insights into the use of indigenous plant extracts as nutrient supplements within organic farming systems.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-48219-6.

Introduction

Agriculture is an energy-intensive sector that plays a vital role in meeting global food and socio-economic demands¹. In India, agriculture contributes approximately 18% of total greenhouse gas emissions (GHGE), largely due to the increasing use of synthetic fertilizers and pesticides over the past five decades². The adoption of green revolution technologies transformed India from a food-deficient to a food-secure nation. However, the introduction of high-yielding crops led to intensive chemical fertilizer application, irrigation mismanagement, and the declining use of organic amendments³. This shift resulted in declining soil fertility, micronutrient deficiencies, and structural degradation, making soils increasingly vulnerable to erosion. Given the ongoing depletion of natural resources and energy crises, integrating energy-efficient and sustainable agricultural technologies is now imperative^4,5. Energy inputs such as mechanical labour, fertilizers, pesticides, and fuel-based operations are central to modern farming. However, mechanization has increased dependence on fossil fuels, while urban migration has reduced the availability of animate energy sources like human and animal labour^6,7. This has necessitated the exploration of renewable, low-emission alternatives, especially in developing regions where agriculture remains a major energy consumer^8,9.

While innovations such as precision agriculture and mechanisation have significantly increased agricultural productivity, they have also escalated energy consumption and environmental pressures, including soil degradation and greenhouse gas emissions (GHGE)¹⁰. In response, organic farming systems have gained attention as a potential alternative due to their reduced reliance on synthetic inputs, offering lower energy use and environmental impacts per hectare. However, evidence on their effectiveness in mitigating GHGE remains mixed^11–13. Some studies report marginal differences between organic and conventional systems in terms of GHGE per unit of product, while others suggest that organic practices may perform worse due to lower yields^14–17.

To advance sustainable agricultural practices, recent research has increasingly explored energy use efficiency and emission reduction through innovative technologies. For example, artificial intelligence models have been applied to analyse energy consumption and GHGE in livestock systems¹⁸, while advanced machine learning techniques such as ANFIS-PSO and other hybrid algorithms have been developed to predict mercury and CO₂ emissions in industrial settings¹⁹. Similarly, lifecycle assessments in cropping systems have provided insights into the cost-emission balance of agricultural systems.

At the global scale, agriculture must simultaneously enhance food production, reduce fossil energy dependence, and mitigate greenhouse gas emissions under increasing climatic uncertainty. Input intensive production systems based on synthetic fertilizers contribute substantially to energy consumption and environmental degradation across both developed and developing regions. Biostimulant based, locally sourced alternatives that improve resource-use efficiency while lowering carbon intensity are therefore essential for achieving climate-smart and energy resilient agricultural systems worldwide.

Against this backdrop, the present study introduces biostimulant derived from renewable plant biomass, specifically, stinging nettle and common weeds, as a sustainable, low-impact alternative to synthetic agricultural inputs. These bio-based formulations aim to enhance plant growth, nutrient uptake, and resilience to abiotic stress while substantially reducing energy inputs and carbon emissions^20–23. Despite their ecological promise, there is a notable lack of field-scale, empirical data on the energy use efficiency, GHGE mitigation potential, and economic viability of biostimulant in major cropping systems²⁴. This study addresses that gap by evaluating the energy efficiency, carbon footprint, and cost-effectiveness of novel biostimulant formulations under both conventional and organic mustard (Brassica juncea) cultivation systems. Such a study contributes new empirical evidence to the evolving discourse on sustainable intensification and offers practical insights for integrating bio-based technologies into environmentally responsible farming models.

In this study, we hypothesize that targeted biostimulant application can match or surpass the agronomic benefits of conventional fertilizers while simultaneously enhancing energy productivity and environmental sustainability. Unlike prior studies that primarily emphasize yield or plant physiology, our approach adopts a comprehensive systems-based assessment, integrating agronomic, energetic, environmental, and economic outcomes to inform climate-smart and energy-resilient farming strategies.

The objectives of the present study were to:

• i.To evaluate the effect of novel plant-based biostimulant formulations and application rates on mustard (Brassica juncea) productivity and energy-use efficiency under field conditions.
• ii.To quantify the carbon footprint, carbon efficiency, and climate impact intensity metrics of biostimulant-based nutrient management in comparison with conventional fertilizer practice using a life cycle assessment approach.
• iii.To assess the economic viability and sustainability of biostimulant-based nutrient management systems in mustard cultivation.

Materials and methods

Experimental site

The experimental study was conducted at the D7 Block of the Norman E. Borlaug Crop Research Center (NEBCRC), Govind Ballabh Pant University of Agriculture and Technology, Pantnagar, Uttarakhand, India (29°01´ N, 79°48´ E and 243.83 m above mean sea level) during the rabi seasons of 2020-21 and 2021-22. The soil at the experimental location is classified under the Mollisol order, sub-order Udolls, great group Hapludoll, subgroup Aquic Hapludoll, and the Haldi soil series, with a loam texture²⁵. This site is part of the Tarai region, a narrow strip at the foothills of the Kumaon Himalayas, which gently slopes southward with a gradient of up to 2%²⁶. The region is situated in the ‘Trans-Gangetic agro-climatic zone and experiences a humid subtropical climate marked by extreme seasonal variations. Winters, particularly in December and January, can be quite harsh with temperatures dropping to around 4 °C, while summers during May and June can see temperatures soaring up to 40°C. The area receives an average annual rainfall of approximately 1450 mm, the majority of which occurs during the southwest monsoon season, typically spanning from mid-June to the end of September. The initial physico-chemical properties of the soil revealed a bulk density (BD) of 1.48 Mg m^⁻3, a mean weight diameter (MWD) of 0.4 mm, and water-stable aggregates (WSA) comprising 16.2%, soil organic carbon (SOC) content was 0.69%. Available macronutrient concentrations included nitrogen (N) with 175.6 kg ha⁻¹, phosphorus (P) 22.0 kg ha⁻¹, potassium (K) 111.4 kg ha⁻¹, and sulphur (S) 18.1 kg ha⁻¹.

Treatment details

The study was carried out using factorial randomized block design (FRBD) with additional treatment and replicated thrice²⁷. The experimental design included two factors, i.e., Factor A, which included three novel biostimulant formulations: stinging nettle-based (KJ₁), general weed-based (KJ₂), and a 50:50 blend of both (KJ₃). Factor B comprised four soil application rates of these biostimulant: 500 L ha^{− 1} (D₁), 1000 L ha^{− 1} (D₂), 1500 L ha^{− 1} (D₃), and 2000 L ha^{− 1} (D₄), applied to each Factor A treatment. Preparation of biostimulant is subjected under the section “Preparation and application of novel organic biostimulant”. The application doses (500–2000 L ha^{− 1}) were selected based on previously published literature on plant-based biostimulants, which are commonly applied as dilute aqueous or fermented slurries in relatively high volumes due to their low concentration of active constituents. Previous studies have reported application volumes of ~ 300 L ha^{− 1} per spray, often repeated multiple times during the season, resulting in cumulative application levels approaching 900–1500 L ha^{− 1}. Given the lack of standardized dose recommendations for crude botanical biostimulants, a graded dose range (500–2000 L ha^{− 1}) was intentionally adopted to capture sub-optimal to supra-optimal biological responses and establish a dose-response relationship, which is a common experimental strategy in plant extract-based biostimulant research^28–30. Additionally, a 100% recommended dose of fertilizer (RDF) was included as a control. RDF included 120 kg ha^{− 1} nitrogen (N), 20 kg ha^{− 1} phosphorus (P), 60 kg ha^{− 1} potassium (K), and 20 kg ha^{− 1} sulphur (S) applied through chemical fertilizers. The treatment combinations are as follows: KJ₁D₁, KJ₁D₂, KJ₁D₃, KJ₁D₄, KJ₂D₁, KJ₂D₂, KJ₂D₃, KJ₂D₄, KJ₃D₁, KJ₃D₂, KJ₃D3, KJ₃D₄, and RDF.

Crop establishment techniques and agronomic interventions

The Kranti variety, released by GBPUAT, Pantnagrar in 1988 at national level, was selected for field experimentation owing to its wide adaptability and popularity in the study region. Sowing was carried out in late November, with harvesting occurring in the second fortnight of March in both years. The crop was established at a seed rate of 4 kg ha^− 1, maintaining a row spacing of 30 cm and a plant-to-plant spacing of 10 cm. Before sowing, seeds were treated with three distinct biostimulant formulations for 10 min and subsequently air-dried overnight for the treatments KJ₁D₁ to KJ₃D₄. In conventional agricultural practice (RDF), seeds were treated with Imidacloprid at 120 mL per 100 kg of seed to prevent seed-borne diseases.

Nutrient management was executed by the designated treatment. In biostimulant-treated plots (KJ₁D₁ to KJ₃D₄), soil application was conducted before sowing, followed by pre-irrigation. Additionally, soil drenching was performed twice at 10-day intervals per treatment. For this, a 10% foliar spray of biostimulant formulations was applied using a knapsack sprayer at 10-day intervals, commencing from the 30th day after sowing until the complete pod development stage, i.e., up to 70 days after sowing. The fertilization schedule for the RDF treatment involved the application of 100-20-60–20 kg N-P-K-S ha^− 1. Nitrogen was applied in two split doses, whereas phosphorus, potassium, and sulphur were applied in full at the time of sowing.

Weed management was carried out manually, while thinning was performed on the 15th day after sowing, wherever necessary, to maintain optimal plant density. Irrigation scheduling was adapted to prevailing climatic conditions, with two irrigations provided in the first year and one in the second year to ensure adequate soil moisture. Pest management involved the application of biopesticides at the flowering stage in biostimulant-treated plots, whereas Monocrotophos 36 SL was used in the control treatment (RDF) to suppress aphid infestations.

Preparation and application of novel organic biostimulant

Stinging nettle-based biostimulant formulation (KJ1)

The stinging nettle-based biostimulant (KJ1) was prepared using a 100 L plastic container. Initially, 10 kg of fresh cow dung and 10 L of cow urine were combined. Separately, a premix containing 1 kg of mustard cake, 1 kg of sprouted blackgram seeds, and 1 kg of crushed jaggery was dissolved in 20 L of water. To this solution, 10 kg of freshly chopped Urtica dioica (stinging nettle) biomass was incorporated. Two days prior, paddy husk was boiled for 15–20 min, filtered, and reserved for inclusion. Additionally, 0.5 L each of fresh milk and buttermilk were added. All components were mixed thoroughly, and the volume was topped up with water. The mixture was stirred twice daily for 10 min using a wooden paddle. Fermentation was monitored through bubble formation on the surface, and the concoction was deemed ready after 20–25 days, once bubbling ceased. The final product was filtered to remove solid residues and stored for application. The prepared biostimulant had a pH of 5.66 and an electrical conductivity of 7.59 dS m⁻¹.

General weed-based biostimulant formulation (KJ2)

The KJ2 formulation followed the same protocol as KJ1, substituting stinging nettle with a mixture of locally available weed species. A total of 10 kg of freshly chopped foliage (1 kg from each species) was used, comprising Azadirachta indica (neem), Clerodendrum spp. (wild jasmine), Aegle marmelos (bael), Datura spp., Lantana camara, Mangifera indica (mango), Psidium guajava (guava), Calotropis spp., Ricinus communis (castor), and Ageratum conyzoides (billy goat weed). All selected plants were approximately five years old, except A. conyzoides, which was included at its flowering stage. Only leaf tissue was used; stems and roots were excluded. The prepared KJ2 biostimulant had a pH of 5.19 and an electrical conductivity of 6.93 dS m⁻¹.

50:50 blend of both (KJ3)

The composite formulation (KJ3) was designed by integrating equal proportions of stinging nettle and the aforementioned common weeds. Specifically, 5 kg of chopped U. dioica and 0.5 kg each of the ten selected weed species (totaling another 5 kg) were used. The remaining components and preparation method were identical to those described for KJ1. The prepared KJ3 biostimulant had a pH of 5.27 and an electrical conductivity of 6.26 dS m⁻¹.

Field application protocol

All biostimulant formulations were soil-applied as per treatment specifications prior to sowing, followed by pre-sowing irrigation. Subsequent soil drenching was performed twice at 10-day intervals. In addition, a 10% foliar spray was administered using a knapsack sprayer at 10-day intervals, commencing 30 days after sowing and continuing until the pod development stage was complete (approximately 70 days post-sowing). This multi-modal application strategy was intended to enhance nutrient uptake and plant vigour throughout critical growth stages. The preparation and utilization of the novel plant based biostimulant were carried out using locally available biomass, guided by traditional principles described in classical agronomic texts such as Vrishayurveda. Their application in this study was scientifically justified by the biochemical properties of the constituent materials, which formed the basis for subsequent field study. All plant materials were handled in accordance with accepted agronomic practices and institutional research norms, and no endangered or protected species were involved.

Calculation framework for total input and output energy

The datasets for inputs and outputs were quantified in energy terms using specific energy coefficients to ensure accuracy and standardization (Table 1). This conversion was performed by applying the corresponding embodied energy equivalents for each component and expressing the results in Mega Joules (MJ). The sources of direct energy (DE) and indirect energy (IE) associated with each agricultural input were systematically identified within an on-farm system boundary. DE encompasses energy sources that directly contribute to agricultural operations, including fuel (diesel) and human labour. Conversely, IE comprises embodied energy within inputs such as agricultural machinery, chemicals, biostimulant, and seeds. The energy output from the harvested products (grain and stover) was determined by multiplying the total yield by its respective energy equivalent, ensuring an accurate estimation of the total energy generated. The energy use indices were computed following the methodological^31,32, ensuring a standardized and systematic approach to evaluating energy efficiency parameters.

Table 1.

Embodied energy coefficients of agricultural inputs and harvested outputs.

S.No.	Particulars	Unit	Energy equivalent (MJ)	References
A.	Input
2.	Seed	Kg	24.0	31
3.	Electric motor	kg	64.8	32
4.	Farm machinery	kg	62.7	32
5.	Diesel	litre	56.31	32
6.	Men labour	hrs	1.96	32
7.	Women labour	hrs	1.57	32
8.	Irrigation (Diesel motor 5 Hp)	hrs	64.8	32
9.	Sprouted Urd	kg	14.7	32
10.	Biostimulant preparation (raw material)	kg	18.0	32
11.	Nitrogen	kg	60.6	32
12	Phosphorus	kg	11.1	32
13.	Potassium	kg	6.7	32
14.	ZnSO4	kg	20.9	32
15.	Milk	litre	5.3	33
16.	Cow dung	kg	17.5	34
17.	Mustard cake	kg	9.6	35
18.	Jaggery	kg	14.025	36
19.	Paddy husk	kg	15.0	37
20.	Monocrotophos 36 SL	litre	120.0	38
B.	Output
1.	Mustard grain	kg	25.0	38
2.	Mustard stalk	kg	12.5	38

Computational approach for carbon auditing

GHGE throughout the cropping period were systematically quantified by applying CO₂ emission coefficients specific to various agricultural inputs including agricultural machinery, diesel, chemical fertilizers, biostimulant, pesticides and irrigation, etc. within an on-farm system boundary. The total carbon input and output were quantified by aggregating the carbon equivalents of all agricultural inputs and outputs involved in on farm crop production.

*Plant biomass contains on an average 44% carbon content³⁹.

Climate impact intensity metrics (CIIM)

To provide an integrated evaluation of environmental performance and resource use efficiency, climate impact intensity metrics (CIIM) were employed. CIIM quantifies the environmental burden in terms of GHGE relative to key agricultural outputs such as energy generation, and economic return. These metrics serve as composite indicators of climate-smart agriculture by accounting not only for absolute GHGE but also for the efficiency with which agricultural systems convert inputs into energy, or economic outcomes^38–40.

Specifically, CIIM enables the comparison of biostimulant based and conventional nutrient management practices in terms of their climate efficiency, providing a finer-scale understanding of trade-offs and synergies between environmental sustainability and farm productivity. This multi-dimensional approach is particularly relevant for informing low-carbon agricultural strategies and aligns with sustainability goals such as SDG 13 (Climate Action) and SDG 12 (Responsible Consumption and Production).

Financial implications of input utilization

The Financial implications were assessed by considering only variable production costs, including labour expenses, agriculture machinery, recurring costs (seeds and chemicals, etc.), harvesting and threshing costs, while excluding land valuation. To evaluate profitability, net returns (NR) were computed as the difference between gross returns (GR) and total cost (TC) (NR = GR – TC). Concurrently, the benefit-cost (B: C) ratio was determined by dividing gross returns by total costs (B: C = GR/TC), ensuring a precise assessment of the economic feasibility of different treatments.

Statistical analysis

Statistical analysis was performed using the R software package, applying analysis of variance (ANOVA) to assess the effects of treatments on various parameters. Treatment means were compared using Duncan’s multiple range test (DMRT) at a 5% probability level (P < 0.05) to determine significant differences among treatments. Additionally, all graphical representations were generated using R software to ensure accurate and effective data visualization.

Results

Crop productivity

The pooled yield analysis over two consecutive growing seasons revealed significant treatment effects on both grain and stalk yield (Table 2). As the effects of year × treatment interactions were non-significant (p > 0.05), data from both years were pooled for further analysis. The highest grain yield was achieved in KJ1D3, followed by KJ1D4. In contrast, the lowest grain yield was observed in KJ2D2. Similarly, stalk yield was maximized in KJ1D4, followed by KJ1D3. Conversely, the lowest stalk yield was observed in KJ2D1.

The comparative yield analysis between the optimized biostimulant treatment (KJ1D3) and the conventional recommended dose of fertilizers (RDF) demonstrated no statistically significant differences in both grain and stalk yields at a p > 0.05 level of significance (Figs. 1 and 2).

Fig. 1.

Pooled analysis of best biostimulant treatment vs. conventional practice on grain yield. ^*X-axis Treatment; Y-axis Grain yield (kg ha⁻¹). KJ1D3 Stinging nettle–based biostimulant applied at 1500 L ha⁻¹; RDF Recommended dose of fertilizer. Bars represent mean values, and error bars indicate standard error of the mean (SEM). Columns sharing the same letter are not significantly different at p ≤ 0.05.

Fig. 2.

Pooled analysis of best biostimulant treatment vs. conventional practice on stalk yield. ^*X-axis Treatment; Y-axis stalk yield (kg ha⁻¹). KJ1D3 Stinging nettle–based biostimulant applied at 1500 L ha⁻¹; RDF Recommended dose of fertilizer. Bars represent mean values, and error bars indicate standard error of the mean (SEM). Columns sharing the same letter are not significantly different at p ≤ 0.05.

Energy flow dynamics and utilization efficiency

The energy input analysis revealed that chemical inputs constituted the highest share, amounting to 7342 MJ, followed by organic inputs at 1445.9 MJ and machinery usage at 594.8 MJ. The seed input contributed the lowest energy input of 96 MJ. In operational energy consumption, fuel usage accounted for 7686.3 MJ, representing the major energy input, while common labour contributed 1071.4 MJ (Figs. 3 and 4). The net energy (MJ ha^− 1) exhibited significant variation among treatments across both years (Table 3). Among the treatments, the highest net energy was recorded in KJ1D3 during 2020–2021 (89,605 ± 328 MJ ha^− 1), while KJ1D4 exhibited the highest value in 2021–2022 (94,929 ± 1,568 MJ ha^− 1). Conversely, the lowest net energy was observed in KJ2D2, with values of 58,871 ± 6,193 MJ ha^− 1 and 68,243 ± 1,900 MJ ha^− 1 in the respective years. The conventional RDF treatment recorded net energy levels of 74,185 ± 1,914 MJ ha^− 1 and 80,192 ± 4,934 MJ ha^− 1, which were lower than several biostimulant-based treatments, particularly KJ1D3 and KJ1D4.

Fig. 3.

Indirect energy distribution. ^*All the values are in MJ ha⁻¹.

Fig. 4.

Direct energy distribution. ^*All the values are in MJ ha⁻¹.

Table 3.

Comparative Bioenergetic assessment of various organic and conventional treatments across two growing seasons (2020–2021 and 2021–2022).

Treatment	Net energy (MJ ha− 1)		Energy Intensity (MJ kg− 1)		Energy Use efficiency		Energy Productivity (kg ha− 1)
	2020–2021	2021–2022	2020–2021	2021–2022	2020–2021	2021–2022	2020–2021	2021–2022
KJ1D1	62,055 ± 11,931cd (38670–85439)	73,803 ± 2286cde (69322–78283)	2.33 ± 0.41bcde (1.52–3.13)	2.02 ± 0.08bc (1.86–2.17)	6.71 ± 1.09cde (4.57–8.84)	7.79 ± 0.21 cde (7.37–8.20)	0.098 ± 0.016de (0.066–0.129)	0.128 ± 0.007bcd (0.114–0.141)
KJ1D2	76,569 ± 3439ab (69828–83309)	81,322 ± 2235bc (76941–85702)	1.96 ± 0.07efg (1.82–2.09)	1.86 ± 0.10c (1.66–2.05)	8.02 ± 0.31ab (7.41–8.62)	8.46 ± 0.2 bc (8.06–8.85)	0.131 ± 0.009a (0.113–0.148)	0.138 ± 0.019abc (0.100–0.175.100.175)
KJ1D3	89,605 ± 1328a (87002–92207)	84,658 ± 2591b (79579–89736)	1.68 ± 0.02g (1.64–1.71)	1.81 ± 0.06cd (1.69–1.92)	9.19 ± 0.02a (9.15–9.22)	8.74 ± 0.23b (8.28–9.19)	0.140 ± 0.005a (0.130–0.149)	0.146 ± 0.005ab (0.136–0.155)
KJ1D4	79,206 ± 8446ab (62651–95760)	94,929 ± 1568a (91855–98002)	1.89 ± 0.16fg (1.57–2.20)	1.62 ± 0.04d (1.54–1.69)	8.24 ± 0.77ab (6.73–9.74)	9.68 ± 0.14a (9.40–9.9)	0.128 ± 0.012ab (0.104–0.151)	0.156 ± 0.006a (0.144–0.167)
KJ2D1	59,970 ± 6401d (47424–72515)	68,628 ± 4983e (58861–78394)	2.37 ± 0.25bcd (1.88–2.86)	2.16 ± 0.15b (1.86–2.45)	6.52 ± 0.58def (5.38–7.65)	7.32 ± 0.45 e (6.43–8.20)	0.096 ± 0.004de (0.088–0.103)	0.121 ± 0.007bcde (0.107–0.134)
KJ2D2	58,871 ± 6193d (46732–71009)	68,243 ± 1900e (64519–71967)	2.40 ± 0.17bc (2.06–2.73)	2.11 ± 0.05b (2.01–2.20)	6.40 ± 0.56ef (5.30–7.49)	7.26 ± 0.17 e (6.92–7.59)	0.094 ± 0.014de (0.066–0.121)	0.106 ± 0.013de (0.080–0.131)
KJ2D3	62,512 ± 5391cd (51945–73078)	78,408 ± 3014bcd (72500–84315)	2.30 ± 0.17bcde (1.96–2.63)	1.90 ± 0.09c (1.72–2.07)	6.71 ± 0.49cde (5.74–7.67)	8.17 ± 0.27 bcd (7.64–8.69)	0.101 ± 0.005cde (0.091–0.110)	0.126 ± 0.005bcd (0.116–0.135)
KJ2D4	73,726 ± 15,899b(42563–104888)	79,929 ± 5018bcd (70093–89764)	2.05 ± 0.39cdefg (1.28–2.81)	1.89 ± 0.05c (1.79–1.98)	7.74 ± 1.45bcd (4.89–10.58)	8.31 ± 0.45 bcd (7.42–9.19)	0.119 ± 0.024abc (0.071–0.166)	0.133 ± 0.022abc (0.089–0.176)
KJ3D1	56,690 ± 1626d (53503–59876)	72,299 ± 8825de (55002–89596)	2.48 ± 0.05b (2.38–2.57)	2.03 ± 0.19bc (1.65–2.40)	6.22 ± 0.14ef (5.94–6.49)	7.66 ± 0.81 de (6.07–9.24)	0.094 ± 0.002de (0.090–0.097)	0.116 ± 0.018cde (0.080–0.151)
KJ3D2	69,476 ± 2293bcd (64981–73970)	82,035 ± 2188bc (77746–86323)	2.09 ± 0.06cdef (1.97–2.20)	1.84 ± 0.05c (1.74–1.93)	7.37 ± 0.21bcde (6.95–7.78)	8.52 ± 0.2bc (8.12–8.91)	0.110 ± 0.002bcd (0.106–0.113)	0.137 ± 0.021abc (0.095–0.178)
KJ3D3	74,096 ± 8112bc (58196–89995)	81,925 ± 6874bc (68451–95398)	2.01 ± 0.19defg (1.63–2.38)	1.84 ± 0.12c (1.60–2.07)	7.77 ± 0.74bc (6.31–9.22)	8.49 ± 0.62 bc (7.27–9.70)	0.120 ± 0.012abc (0.096–0.143)	0.134 ± 0.14abc (0.140–0.408)
KJ3D4	82,540 ± 3228ab (76213–88866)	84,254 ± 6531b (71453–97054)	1.82 ± 0.06fg (1.70–1.93)	1.81 ± 0.10cd (1.61–2.00.61.00)	8.54 ± 0.29ab (7.97–9.10)	8.70 ± 0.59 b (7.54–9.85)	0.133 ± 0.007a (0.119–0.146)	0.143 ± 0.015abc (0.113–0.172)
RDF	74,185 ± 1914bc (70433–77936)	80,192 ± 4934bcd (70521–89862)	2.84 ± 0.03a (2.78–2.89)	2.72 ± 0.19a (2.34–3.09)	5.46 ± 0.11f (5.24–5.67)	5.82 ± 0.29 f (5.25–6.38)	0.084 ± 0.006e (0.072–0.095)	0.096 ± 0.005e (0.086–0.105)

^*Values are expressed as mean ± standard deviation (SD). Values in parentheses represent the lower and upper confidence intervals. Means followed by the same superscript letter within a column are not significantly different according to Duncan’s multiple range test (DMRT) at p < 0.05.

KJ1D1 Stinging nettle–based biostimulant at 500 L ha⁻¹.

KJ1D2 Stinging nettle–based biostimulant at 1000 L ha⁻¹.

KJ1D3 Stinging nettle–based biostimulant at 1500 L ha⁻¹.

KJ1D4 Stinging nettle–based biostimulant at 2000 L ha⁻¹.

KJ2D1 General weed–based biostimulant at 500 L ha⁻¹.

KJ2D2 General weed–based biostimulant at 1000 L ha⁻¹.

KJ2D3 General weed–based biostimulant at 1500 L ha⁻¹.

KJ2D4 General weed–based biostimulant at 2000 L ha⁻¹.

KJ3D1 50:50 blend of stinging nettle– and general weed–based biostimulant at 500 L ha⁻¹.

KJ3D2 50:50 blend of stinging nettle– and general weed–based biostimulant at 1000 L ha⁻¹.

KJ3D3 50:50 blend of stinging nettle– and general weed–based biostimulant at 1500 L ha⁻¹.

KJ3D4 50:50 blend of stinging nettle– and general weed–based biostimulant at 2000 L ha⁻¹.

Energy intensity (MJ kg^− 1) exhibited a downward trend with biostimulant application, indicating greater energy efficiency. The lowest energy intensity was noted in KJ1D3, with values of 1.68 ± 0.02 MJ kg^− 1 in 2020–2021 and KJ1D4 (1.62 ± 0.04 MJ kg^− 1 in 2021–2022), demonstrating superior energy utilization efficiency. In contrast, the RDF treatment exhibited significantly higher energy intensity (2.84 ± 0.03 MJ kg^− 1 and 2.72 ± 0.19 MJ kg^− 1, respectively), implying lower energy conversion efficiency.

Energy use efficiency (EUE) was maximized in KJ1D3 (9.19 ± 0.02 and 8.74 ± 0.23) and KJ1D4 (8.24 ± 0.77 and 9.68 ± 0.14), signifying optimal energy transformation into yield. Conversely, RDF recorded the lowest EUE (5.46 ± 0.11 and 5.82 ± 0.29), highlighting its suboptimal energy performance compared to the best biostimulant treatments.

Energy productivity (MJ kg^− 1) exhibited a consistent trend across the study period, with the KJ1D3 treatment achieving the highest values (0.140 ± 0.005 MJ kg^− 1 in 2020–2021), whereas KJ1D4 demonstrated superior performance in 2021–2022 (0.156 ± 0.006 MJ kg^− 1), while RDF showed the lowest productivity (0.084 ± 0.006 MJ kg⁻¹and 0.096 ± 0.005 MJ kg^− 1), reinforcing the inefficiency of conventional nutrient management compared to biostimulant applications.

Carbon auditing

The analysis of carbon efficiency, carbon footprint, and carbon sustainability index (CSI) demonstrated substantial variation across treatments over both cropping years (Table 4). Among the treatments, KJ1D3 exhibited the highest carbon efficiency, recording 7.51 ± 0.09 in 2020–2021 and 6.97 ± 0.24 in 2021–2022, followed by KJ1D4 (6.70 ± 0.61 and 7.79 ± 0.20, respectively). In contrast, the RDF treatment had the lowest carbon efficiency, with values of 2.83 ± 0.03 in 2020–2021 and 2.96 ± 0.20 in 2021–2022, reflecting lower carbon-use efficiency. The carbon footprint was lowest in KJ1D3, with recorded values of 0.059 ± 0.001 in 2020–2021 and KJ1D4 (0.056 ± 0.001 in 2021–2022), indicating a reduced environmental impact. In contrast, the highest carbon footprint was observed in the RDF treatment, reaching 0.156 ± 0.001 in 2020–2021 and 0.149 ± 0.01 in 2021–2022, signifying greater carbon emissions per unit of yield. With regard to carbon sustainability, KJ1D3 maintained the highest CSI, attaining 6.51 ± 0.09 in 2020–2021 and KJ1D4 (6.79 ± 0.20 in 2021–2022), emphasizing its superior sustainability potential. Conversely, the RDF treatment exhibited the lowest CSI, with values of 1.83 ± 0.03 and 1.96 ± 0.20, respectively.

Table 4.

Carbon budgeting and sustainability metrics of various biostimulant treatments and conventional agricultural practice.

Treatment	Carbon efficiency		Carbon footprint (kg CE kg− 1 grain)		Carbon sustainability index
	2020–2021	2021–2022	2020–2021	2021–2022	2020–2021	2021–2022
KJ1D1	5.68 ± 0.93cd (3.85–7.50)	6.42 ± 0.27bc (5.89–6.94)	0.079 ± 0.013bcde (0.053–0.104)	0.069 ± 0.002bc (0.065–0.072)	4.68 ± 0.93cd (2.85–6.50)	5.42 ± 0.27bc (4.89–5.94)
KJ1D2	6.49 ± 0.26bc (5.98–6.99)	6.85 ± 0.36b (6.14–7.55)	0.068 ± 0.002defg (0.064–0.071)	0.064 ± 0.003cd (0.058–0.069)	5.49 ± 0.26bc (4.98–5.99)	5.85 ± 0.36b (5.14–6.55)
KJ1D3	7.51 ± 0.09a (7.33–7.68)	6.97 ± 0.24b (6.49–7.44)	0.059 ± 0.001g (0.057–0.060)	0.063 ± 0.002cd (0.059–0.066)	6.51 ± 0.09a (6.33–6.68)	5.97 ± 0.24b (5.49–6.44)
KJ1D4	6.70 ± 0.61ab (5.50–7.89)	7.79 ± 0.2a (7.39–8.18)	0.066 ± 0.005efg (0.056–0.075)	0.056 ± 0.001d (0.054–0.057)	5.70 ± 0.61ab (4.50–6.89)	6.79 ± 0.2a (6.39–7.18)
KJ2D1	5.49 ± 0.55cd (4.41–6.56)	5.98 ± 0.42c (5.15–6.80)	0.081 ± 0.008bcd (0.065–0.096)	0.074 ± 0.005b (0.064–0.083)	4.49 ± 0.55cd (3.41–5.56)	4.98 ± 0.42c (4.15–5.80)
KJ2D2	5.28 ± 0.39d (4.51–6.04)	6.00 ± 0.15c (5.70–6.29)	0.084 ± 0.006bc (0.072–0.095)	0.073 ± 0.001b (0.071–0.072)	4.28 ± 0.39d (3.51–5.04)	5.00 ± 0.15c (4.70–5.29)
KJ2D3	5.47 ± 0.42cd (4.64–6.29)	6.61 ± 0.34bc (5.94–7.27)	0.081 ± 0.006bcd (0.069–0.092)	0.067 ± 0.003bc (0.061–0.072)	4.47 ± 0.42cd (3.64–5.29)	5.61 ± 0.34bc (4.94–6.27)
KJ2D4	6.20 ± 1.14bcd (3.96–8.43)	6.59 ± 0.19bc (6.21–6.96)	0.073 ± 0.014bcdef (0.045–0.100.045.100)	0.067 ± 0.001bc (0.065–0.068)	5.20 ± 1.14bcd (2.96–7.43)	5.59 ± 0.19bc (5.21–5.96)
KJ3D1	5.21 ± 0.12d (4.97–5.44)	6.42 ± 0.61bc (5.22–7.61)	0.084 ± 0.002b (0.080–0.087)	0.069 ± 0.006bc (0.057–0.080)	4.21 ± 0.12d (3.97–4.44)	5.42 ± 0.61bc (4.22–6.61)
KJ3D2	6.08 ± 0.19bcd (5.70–6.45)	6.91 ± 0.2b (6.51–7.30)	0.072 ± 0.002bcdef (0.068–0.075)	0.064 ± 0.001cd (0.062–0.065)	5.08 ± 0.19bcd (4.70–5.45)	5.91 ± 0.20b (5.51–6.30)
KJ3D3	6.31 ± 0.59bc (5.15–7.46)	6.85 ± 0.47b (5.92–7.77)	0.070 ± 0.006cdefg (0.058–0.08)	0.064 ± 0.004cd (0.056–0.071)	5.31 ± 0.59bc (4.15–6.46)	5.85 ± 0.47b (4.92–6.77)
KJ3D4	6.85 ± 0.23ab (6.39–7.30)	6.88 ± 0.4b (6.09–7.66)	0.064 ± 0.002fg (0.060–0.067)	0.064 ± 0.003cd (0.058–0.069)	5.85 ± 0.23ab (5.39–6.30)	5.88 ± 0.4b (5.09–6.66)
RDF	2.83 ± 0.03e (2.77–2.88)	2.96 ± 0.2d (2.56–3.35)	0.156 ± 0.001a (0.154–0.157)	0.149 ± 0.01a (0.129–0.168)	1.83 ± 0.03e (1.77–1.88)	1.96 ± 0.2d (1.56–2.35)

^*Values are expressed as mean ± standard deviation (SD). Values in parentheses represent the lower and upper confidence intervals. Means followed by the same superscript letter within a column are not significantly different according to Duncan’s multiple range test (DMRT) at p < 0.05.

KJ1D1 Stinging nettle–based biostimulant at 500 L ha⁻¹.

KJ1D2 Stinging nettle–based biostimulant at 1000 L ha⁻¹.

KJ1D3 Stinging nettle–based biostimulant at 1500 L ha⁻¹.

KJ1D4 Stinging nettle–based biostimulant at 2000 L ha⁻¹.

KJ2D1 General weed–based biostimulant at 500 L ha⁻¹.

KJ2D2 General weed–based biostimulant at 1000 L ha⁻¹.

KJ2D3 General weed–based biostimulant at 1500 L ha⁻¹.

KJ2D4 General weed–based biostimulant at 2000 L ha⁻¹.

KJ3D1 50:50 blend of stinging nettle– and general weed–based biostimulant at 500 L ha⁻¹.

KJ3D2 50:50 blend of stinging nettle– and general weed–based biostimulant at 1000 L ha⁻¹.

KJ3D3 50:50 blend of stinging nettle– and general weed–based biostimulant at 1500 L ha⁻¹.

KJ3D4 50:50 blend of stinging nettle– and general weed–based biostimulant at 2000 L ha⁻¹.

Greenhouse gas efficiency index (GGEI)

The results of greenhouse gas efficiency index (or) climate impact intensity metrics showed variation among the treatments over both cropping seasons (Fig. 5a & b). In cropping season 2020–21, CIIM_energy values ranged from 0.051 to 0.081 kg CO₂-eq MJ ha⁻¹, with treatment KJ2D4 exhibiting the lowest emission intensity, statistically at par with KJ2D3 and KJ1D3. Conversely, KJ1D1 recorded the highest CIIM_energy, indicating elevated carbon costs per unit energy input. In 2021–22, emission intensities declined over previous year but remained significantly variable across the treatments with KJ2D1 and KJ2D2 showed lower CIIM_energy values (~ 0.061–0.063 kg CO₂-eq MJ ha⁻¹).

Fig. 5.

Climate impact efficiency metrics under different treatment combinations during 2020-21 and 2021-22 cropping season. a, b CIIM_energy expressed as kg CO2 –eq MJ ha⁻¹; c, d CIIM_economics expressed as kg CE–eq USD ha^-1 across two consecutive growing seasons (2020-21 and 2021-2022). Treatments are shown along the x-axis; error bars represent standard errors. Different letters indicate significant differences (p < 0.05) using DMRT tests.

The economics based CIIM assessment showed more variability across the treatments (Fig. 5, c&d). During 2020–21 cropping season, lowest CIIM_economics was observed with treatments KJ1D4 (9.8 kg CE kg⁻¹ USD) followed by KJ2D3. In contrast, KJ2D4, RDF, and KJ2D1 recorded the highest carbon emitting per USD values exceeding 20 kg CE kg⁻¹ USD. In the year 2021–22, KJ3D2 (6.3 kg CE kg⁻¹ USD) showed lowest CIIM_economics followed by KJ2D3. Conversely, KJ2D1 (14.5 kg CE kg⁻¹ USD) showed highest CIIM_economics values.

Economic analysis

The economic assessment indicated significant variability among the treatments in terms of gross return, net return, and benefit-cost (B: C) ratio (Table 5). Among the biostimulant formulations, KJ1 exhibited the highest gross return (USD 1016.7 ± 169.1 ha^− 1), net return (USD 481.2 ± 164.9 ha^− 1), and B: C ratio (1.89 ± 0.30), whereas KJ2 recorded the lowest economic returns. Regarding biostimulant application doses, the D4 treatment achieved the highest gross return (USD 1035.4 ± 165.3 ha^− 1), net return (USD 488.2 ± 165.3 ha^− 1), and an optimal B: C ratio (1.89 ± 0.30), respectively. Conversely, the D1 treatment exhibited the least economic returns. The conventional RDF treatment resulted in gross return, net return, and B: C ratio (USD 840.4 ha^− 1, USD 341.8 ha^− 1, and 1.68, respectively), which were comparatively lower than biostimulant-based treatments, particularly KJ1 and D4.

Table 5.

Economic returns and benefit-cost ratio as affected by biostimulant concoctions and dose levels (Two-year pooled data).

Treatment	Gross return (USD ha− 1)	Net return (USD ha− 1)	Benefit : cost ratio
Biostimulant Concoctions
KJ1	1016.7 ± 169.1a (685.2–1348.1.2.1)	481.2 ± 164.9a (157.9–804.4.9.4)	1.89 ± 0.30a (1.30–2.47)
KJ2	855.3 ± 163.0c (535.8–1174.7.8.7)	379.8 ± 159.9c (66.3–693.2.3.2)	1.59 ± 0.29c (1.02–2.15)
KJ3	941.2 ± 170.9b (606.2–1276.1.2.1)	405.6 ± 166.4b (79.45–731.7)	1.75 ± 0.30b (1.16–2.33)
Doses
D1	828.8 ± 155.7c (523.6–1133.9.6.9)	345.0 ± 155.7c (39.9–650.1.9.1)	1.58 ± 0.29c (1.01–2.14)
D2	909.1 ± 180.6b (555.1–1263.0)	377.4 ± 180.6b (23.4–731.3.4.3)	1.71 ± 0.33b (1.06–2.35)
D3	977.6 ± 152.3a (679.0–1276.1.0.1)	438.2 ± 152.3ab(139.6–736.7.6.7)	1.81 ± 0.28ab (1.26–2.35)
D4	1035.4 ± 165.3a (711.4–1359.3.4.3)	488.2 ± 165.3a (164.2–812.1.2.1)	1.89 ± 0.30a (1.30–2.47)
Additional treatment (For comparison)
RDF	840.4	341.8	1.68

^*Values are expressed as mean ± standard deviation (SD). Values in parentheses represent the lower and upper confidence intervals. Means followed by the same superscript letter within a column are not significantly different according to Duncan’s multiple range test (DMRT) at p < 0.05.

^*For calculation of biostimulant treatments 25% higher premium seed price was considered in gross returns.

KJ1 Stinging nettle-based, KJ2 general weed-based, KJ3 50:50 blend of both, D1 500 L ha^− 1, D2 1000 L ha^− 1, D3 1500 L ha^− 1, D4 2000 L ha^− 1, and RDF 100% Recommended dose of fertilizer.

Discussion

Crop productivity and profitability

The pooled results from the two-year field study indicate a significant improvement in mustard grain and stalk yields with the application of the stinging nettle-based biostimulant (KJ1), particularly at higher application rates (1500 and 2000 L ha^{− 1}). Specifically, grain yield increased by 43.7% and 42.4% under KJ1D3 and KJ1D4 treatments, respectively, compared to KJ2D2 (Table 2). Similarly, stalk yield was enhanced by 29.4% and 28.8% in KJ1D4 and KJ1D3, respectively, relative to KJ2D1. These results underscore the agronomic efficacy of the KJ1 formulation, suggesting its potential as a high-performing biostimulant for mustard cultivation. The observed yield improvements may be attributed to the superior nutrient-release dynamics associated with the stinging nettle-based formulation, which likely enhanced soil nutrient availability throughout the crop cycle. This is consistent with findings from previous studies, which have reported that biostimulant derived from plant biomass can significantly enhance soil nutrient mobilization and uptake due to their rich composition of humic substances, bioactive compounds, and micronutrients^41–43. The rapid mineralization of organic compounds present in KJ1 may have promoted early and sustained nutrient availability, aligning with reports by du Jardin (2015)²⁹ that plant-based biostimulant can stimulate root development and nutrient absorption through hormonal and biochemical signaling. Moreover, the enhanced nutrient assimilation facilitated by the nutrient content of biostimulant appears to have contributed to improved physiological performance of the mustard crop, as evidenced by increased biomass production. This finding echoes the conclusions of Rouphael and Colla (2020)⁴⁴, who demonstrated that biostimulant can significantly improve crop productivity by modulating plant metabolism and enhancing tolerance to abiotic stress. Beyond nutrient dynamics, the cumulative improvements in yield under KJ1 treatments are likely the result of a synergistic interplay between several soil and plant factors, including elevated soil organic carbon, enhanced microbial activity, intensified enzymatic processes, and the presence of growth-promoting compounds such as flavonoids and phenolic acids^45–48. This aligns with recent work by Savarese et al. (2022)⁴⁹, who highlighted the multifaceted mechanisms through which biostimulant contribute to soil health and plant performance.

Table 2.

Pooled yield dynamics of grain and stalk under different biostimulant treatments across two growing seasons.

Treatment	Grain Yield (kg ha− 1)	Stalk yield (kg ha− 1)
KJ1D1	1235.2 ± 217.6def (808.7–1661.6.7.6)	3832.9 ± 458.1de (2935.0–4730.8.0.8)
KJ1D2	1471.6 ± 153.7abc (1170.3–1772.8.3.8)	4244.5 ± 371.9bcd (3515.5–4973.4.5.4)
KJ1D3	1572.4 ± 66.8a (1441.4–1703.3.4.3)	4700.9 ± 342.9a (4028.8–5372.9.8.9)
KJ1D4	1558.5 ± 198.2ab (1170.0–1946.9.0.9)	4723.6 ± 454.1a (3833.5–5613.6.5.6)
KJ2D1	1182.0 ± 162.0ef (864.5–1499.5.5.5)	3648.8 ± 350.6e (2961.6–4335.9.6.9)
KJ2D2	1094.0 ± 156.2f (787.8–1400.1.8.1)	3768.6 ± 329.9e (3121.9–4415.2.9.2)
KJ2D3	1244.5 ± 160.7def (929.5–1559.4.5.4)	4022.9 ± 496.5cde (3049.7–5205.8.7.8)
KJ2D4	1386.7 ± 243.1bcd (910.2–1863.1.2.1)	4248.0 ± 488.7bcd (3290.1–5205.8.1.8)
KJ3D1	1144.6 ± 182.5f (786.9–1502.3.9.3)	3739.4 ± 479.7e (2799.1–4679.6.1.6)
KJ3D2	1349.6 ± 218.5cde (921.3–1777.9.3.9)	4233.4 ± 349.1bcd (3549.1–4917.6.1.6)
KJ3D3	1394.3 ± 156.4abcd (1087.7–1700.8.7.8)	4327.5 ± 354.8abc (3632.0–5022.9.0.9)
KJ3D4	1514.0 ± 134.0abc (1251.3–1776.6.3.6)	4519.0 ± 173.0ab (4179.9–4858.0)

^*Values are expressed as mean ± standard deviation (SD). Values in parentheses represent the lower and upper confidence intervals. Means followed by the same superscript letter within a column are not significantly different according to Duncan’s multiple range test (DMRT) at p < 0.05.

^*KJ1D1 Stinging nettle–based biostimulant at 500 L ha⁻¹, KJ1D2 Stinging nettle–based biostimulant at 1000 L ha⁻¹, KJ1D3 Stinging nettle–based biostimulant at 1500 L ha⁻¹, KJ1D4 Stinging nettle–based biostimulant at 2000 L ha⁻¹, KJ2D1 General weed–based biostimulant at 500 L ha⁻¹, KJ2D2 General weed–based biostimulant at 1000 L ha⁻¹, KJ2D3 General weed–based biostimulant at 1500 L ha⁻¹, KJ2D4 General weed–based biostimulant at 2000 L ha⁻¹, KJ3D1 50:50 blend of stinging nettle– and general weed–based biostimulant at 500 L ha⁻¹, KJ3D2 50:50 blend of stinging nettle– and general weed–based biostimulant at 1000 L ha⁻¹, KJ3D3 50:50 blend of stinging nettle– and general weed–based biostimulant at 1500 L ha⁻¹, KJ3D4 50:50 blend of stinging nettle– and general weed–based biostimulant at 2000 L ha⁻¹.

In terms of economic viability, the stinging nettle-based biostimulant also demonstrated superior profitability, driven by a combination of increased yields, a 25% organic market price premium, and reduced input costs compared to conventional fertilizer regimes. These findings are consistent with those of Deolu-Ajayi et al. (2025)^50,51, who noted that the use of biostimulant not only reduces dependency on costly synthetic inputs but also improves return on investment by enhancing crop quality and market value. Furthermore, the strong economic performance of biostimulant-based treatments in our study aligns with recent research emphasizing the role of bio-based inputs in improving farm-level resource use efficiency^52,53.

Taken together, these results support the notion that biostimulant technologies, particularly those based on locally sourced plant biomass like stinging nettle, offer a viable and sustainable alternative to conventional fertilizers. They not only deliver competitive agronomic performance but also align with broader sustainability goals, including reduced energy input, improved soil health, and economic resilience for farmers.

Energy budgeting: input–output analysis

The integration of biostimulant treatments resulted in a substantially lower energy footprint, with total energy inputs ranging from 10,863 to 10,941 MJ ha^{− 1}, which is 34.7% less than conventional agricultural practices (16,649 MJ ha^{− 1}) (Table 3). Previous studies have demonstrated that conventional systems are highly energy-intensive and inefficient in resource utilization⁵⁴. In the present study, the energy use efficiency (EUE) in conventional systems ranged from 5.46 to 5.82, notably lower than that observed in biostimulant-based treatments (6.71–9.68), reinforcing the resource-intensive nature of conventional agriculture. The elevated energy consumption in conventional systems is primarily driven by the production, transportation, and field application of synthetic fertilizers and pesticides, which contribute disproportionately to the total energy input, ~ 12.4 times higher than in biostimulant-based treatments. A typical trend observed in energy dynamics is that increasing energy inputs often lead to diminishing energy-use efficiency⁵⁵. Within this framework, biostimulant-based treatments emerged as sustainable alternatives to conventional management practices, exhibiting enhanced efficiency in optimizing the productive utilization of invested energy⁵⁶.

Carbon management

The results from this study clearly demonstrate that conventional agricultural practices are associated with a substantially higher carbon footprint, approximately 1.85 times greater than that of biostimulant-based treatments (Table 4). This elevated footprint is largely due to the intensive use of synthetic agrochemicals, which contributed 13.1% to the total input emissions. The life cycle of these chemical inputs, encompassing their production, transportation, storage, and application, is a major source of GHGE, as previously reported by Soni et al. (2013)⁵⁷. The environmental costs associated with synthetic fertilizer use are well-documented. Excessive and prolonged application can lead not only to soil nutrient imbalances and yield plateaus but also to increased nitrous oxide emissions, a potent GHGE. This raises significant concerns about the long-term sustainability of input-intensive farming systems, particularly in the context of climate change and soil health degradation.

In contrast, stinging nettle-based biostimulant treatments exhibited markedly superior carbon efficiency, a higher carbon sustainability index and lower CIIM_energy, CIIM_economics. These improvements are attributable to the reduced carbon intensity of input materials and enhanced carbon use efficiency in biomass production, with carbon efficiency and the carbon sustainability index reflecting short-term, biomass-associated carbon dynamics. Similar trends have been observed in studies by Hashemi et al. (2024)⁵⁸, who found that organic and low-input systems incorporating composts or bio-based amendments significantly lowered GHGE per hectare while maintaining competitive yields. Moreover, the observed reductions in carbon emissions in biostimulant treatments are consistent with the findings of Shabir et al. (2023)⁵⁹, who reported that biologically intensive farming systems generally have lower carbon footprints, particularly when evaluated on a per-hectare basis. The use of biostimulant can also improve soil organic carbon levels through enhanced microbial activity and root biomass, contributing to long-term carbon sequestration⁶⁰.

Importantly, the use of renewable, locally sourced biomass for biostimulant preparation further enhances the environmental sustainability of this approach. By minimizing dependence on energy-intensive industrial inputs, biostimulant-based agriculture contributes to climate change mitigation and aligns with global sustainability frameworks, including the Food and Agriculture Organization’s Sustainable Development Goals (FAO SDGs), particularly SDG 13 (Climate Action) and SDG 12 (Responsible Consumption and Production). These findings underscore the environmental advantages of transitioning toward biostimulant-based nutrient management strategies. They not only reduce the carbon intensity of agricultural production but also promote regenerative soil health and align with the principles of climate-smart agriculture.

While our study provides compelling evidence of the agronomic, economic, and environmental benefits of biostimulant-based farming, it is limited to a specific crop (mustard) and agro-climatic zone (Trans-Gangetic region). Variability in soil types, climatic conditions, and cropping systems may influence the generalizability of the findings. Additionally, long-term impacts on soil health, biodiversity, and nutrient cycling were not fully assessed. Future research should focus on multi-location trials across diverse agro-ecological zones and cropping systems. Integrating advanced monitoring tools, such as remote sensing and AI-based models, could further refine energy and carbon accounting. Investigating synergistic effects with other regenerative practices may also broaden the scope and scalability of biostimulant use in sustainable agriculture.

Limitations and challenges

Despite the comprehensive field based and systems level assessment, several limitations of this study should be acknowledged. The experiment was conducted at a single site and over a limited duration, which may constrain the generalizability of the results across contrasting agro-climatic conditions and soil types. The biostimulant formulations were derived from specific plant biomass sources, and variation in biomass composition and formulation stability could influence consistency of responses under broader field conditions. The economic assessment was based on prevailing market prices, which are subject to temporal and regional variability. Future multi-location and long-term studies with expanded life cycle boundaries are therefore required to confirm the robustness and scalability of biostimulant based nutrient management systems.

Conclusion

This study addressed the core research question of whether biostimulant-based nutrient management can serve as a viable, sustainable alternative to conventional fertilization in mustard (Brassica juncea) cultivation. The findings demonstrate that plant-based biostimulant, particularly the stinging nettle formulation at 1500 L ha^− 1 (KJ1D3), significantly enhanced energy-use efficiency, reduced carbon footprint, and improved net economic returns compared to the conventional RDF treatment. These results affirm the hypothesis that biostimulant can match or surpass the agronomic and environmental performance of synthetic inputs when applied strategically. Beyond yield improvement, the study highlights the potential of biostimulant to contribute to energy- and climate-resilient agriculture by optimizing resource use and lowering emissions. These insights carry broader implications for sustainable farming systems, particularly in regions facing soil degradation, energy constraints, and environmental stress. By leveraging locally sourced plant biomass, such biostimulant offer a scalable and ecologically sound pathway aligned with the Sustainable Development Goals (SDGs), notably SDGs 2, 12, and 13. Further research should validate the consistency of these responses across diverse agro-climatic zones and cropping systems, examine the long-term effects of repeated biostimulant applications on soil health, nutrient cycling, and carbon dynamics, and evaluate their integration with regenerative and precision agriculture practices to enhance scalability, resilience, and long-term.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Banavath Mahesh Naik : Conceptualization, Methodology, Investigation, Visualisation, Validation, Formal analysis, Writing - original draft. Sunita T. Pandey: Conceptualization, Investigation, Project administration, Fund acquisition, Validation, Visualisation. Amit Bhatnagar: Methodology, Formal analysis, Project administration, Validation, Visualisation, and Writing - review and editing. Subhash Chandra: Methodology, Project administration, Validation, Visualisation, and Writing - review and editing. Rajeew Kumar: Methodology, Project administration, Validation, Writing - review and editing. Karthika M: Formal analysis, and Writing - review and editing. K Srikanth Reddy: Formal analysis, Investigation, and Writing - original draft. Biswajit Pramanick: Methodology, Formal analysis, Writing - original draft, and Writing - review and editing.

Funding

No external funding was received for this study.

Data availability

The data supporting this study are available within the article and any additional information can be inquired by contacting the corresponding author(s).

Declarations

Competing interests

The authors declare no competing interests.