Bridging private and public benefits in sustainable weed management: insights from the OPER8 European network

Abstract

Introduction:

The transition toward herbicide-free weed management is central to the European Green Deal and the Farm-to-Fork Strategy. Despite the availability of mature non-chemical weed management (NCWM) techniques, their uptake across European farming systems remains limited. This study examines this adoption gap by integrating economic and socio-environmental evidence from twenty NCWM case studies in seven European countries within the OPER8 network.

Methods:

A mixed-method framework combined cost–benefit analysis (CBA) with a qualitative traffic-light indicator capturing social and environmental co-benefits through structured stakeholder scoring. Economic data were collected through harmonized farmer questionnaires and validated in national and European workshops involving more than 160 participants. Net Present Value per hectare (NPV/ha) was used as a comparable measure of private profitability, while qualitative scores were normalized and aggregated into a composite Socio-Environmental Impact Score (0–100). These results were integrated into a Dual-Axis Sustainability Matrix, positioning each cluster of practices along private economic and public socio-environmental dimensions of value creation.

Results:

Results reveal a continuum of configurations rather than a binary divide between viable and unviable options. Cover crops emerge as the clearest win–win cluster, combining positive private returns with high socio-environmental performance. Mechanical weeding is the closest to break-even among the negative-NPV clusters, while precision farming also occupies an intermediate position but remains negative in median terms. Mulching records the weakest private performance despite positive socio-environmental scores, whereas ecological management combines the highest socio-environmental performance with persistently negative private returns. Across the portfolio as a whole, socio-environmental scores remain medium to high, while positive private profitability is limited to a single cluster.

Discussion:

Temporal asymmetry between short-term private costs and delayed public benefits emerges as a central governance challenge. Wider NCWM uptake is likely to require hybrid incentive architectures blending result-based payments, risk-sharing instruments, and cooperative or service-based support arrangements. By linking micro-level farm economics with macro-level policy design, the OPER8 framework offers a transferable tool for evaluating agro-ecological transitions and supporting performance-based governance under the reformed Common Agricultural Policy (CAP).

1 Introduction

The reduction of chemical herbicides has become a cornerstone of the European Green Deal and the Farm to Fork Strategy, which aim to halve pesticide use and risk by 2030 (European Commission, 2020). Yet, despite decades of agronomic innovation and rising societal concern over environmental and health impacts, European agriculture remains structurally dependent on chemical weed control. This dependency reveals persistent technological, economic, and institutional barriers that constrain the diffusion of alternative practices (Kup and Saglam, 2014; Panagos et al., 2018; Gagliardi et al., 2023).

Non-chemical weed management (NCWM) encompasses a mature and diverse portfolio of agronomic and technological options—mechanical and thermal weeding, cover crops, mulching, precision systems, and ecological management—that can maintain yields while drastically reducing synthetic inputs (Gagliardi et al., 2026; Liebhard et al., 2024; DeVetter et al., 2015; Genger et al., 2018; Merfield, 2015; Rasmussen, 2004). Nevertheless, their uptake has been slow and uneven across Member States, underscoring the need to understand not only their technical feasibility but also their economic and governance dynamics (Lechenet et al., 2017; Riemens et al., 2021).

Existing studies have predominantly examined the technical and environmental performance of individual NCWM methods (Melander et al., 2017; Gagliardi et al., 2023), while parts of the adoption and agricultural innovation-systems literature have also addressed profitability, risk perception, knowledge exchange, and institutional barriers shaping uptake (Panagos et al., 2018; Klerkx et al., 2012). What remains more limited, however, is an integrated comparative perspective that jointly analyses private economic performance and socio-environmental implications across clusters of practices and governance settings.

A crucial step toward such systemic understanding lies in comparing groups of practices rather than isolated cases. Within the OPER8 project (European Thematic Network for unlocking the full potential of Operational Groups on alternative weed control), this principle was embedded in its structure as a seven European Thematic Network aimed at unlocking the full potential of Operational Groups (OGs) working on alternative weed control. Instead of analyzing single pilot initiatives, OPER8 connected networks of practitioners, researchers, and advisors across seven countries, transforming dispersed local experiences into a collective evidence base. Each national team assessed local practices—ranging from mechanical hoeing in arable crops to living mulches in orchards or inter-row cover crops in vineyards—through harmonized cost–benefit and socio-environmental frameworks. The cross-country synthesis distilled these heterogeneous insights into five functional clusters: mechanical weeding, cover crops, mulching, precision systems, and ecological management.

This bottom-up clustering was not a simple classification but a process of knowledge co-production reflecting both the technological configurations of weed control and the organizational logics underpinning their delivery. By grouping practices around shared mechanisms of action and governance requirements, the framework captures how innovation circulates through European Agricultural Knowledge and Innovation Systems (AKIS). It illustrates how networks—rather than individual farms—constitute the operative scale for experimentation, learning, and diffusion. The cluster perspective thus connects farm-level economics with the meso-level dynamics of advisory and cooperative systems, identifying the coordination and support mechanisms required for each type of non-chemical approach.

Analyzing performance at the cluster level enables the distinction between adaptive innovations, such as mechanical or precision weeding, which can be mainstreamed through service or market-based models, and regenerative transitions, such as ecological management or cover cropping, which depend on collective action and longer-term institutional support. This shift from case to cluster and from farm to network aligns with the analytical needs of the Common Agricultural Policy (CAP), which must design instruments addressing families of practices and knowledge systems with comparable cost structures, risks, and co-benefits rather than context-specific examples. By translating micro-level diversity into policy-relevant typologies rooted in real networks of practice, the OPER8 cluster approach bridges empirical evidence and the operational categories required to scale up agro-ecological innovation within Europe's performance-based policy architecture.

NCWM therefore provides a critical testing ground for the CAP's broader transition toward ecosystem-service-oriented payments (OECD., 2021). By reducing synthetic pesticide dependence, NCWM contributes directly to the Sustainable Use of Pesticides Regulation while simultaneously advancing multiple Green Deal objectives—soil health, biodiversity restoration, climate mitigation, and rural resilience. Realizing these synergies depends on aligning private incentives with public benefits through evidence that connects farm-level profitability to socio-environmental outcomes and supports the design of hybrid policy instruments combining investment facilitation, risk-sharing, and result-based rewards (Jacobs et al., 2016; Kenter et al., 2016).

The present study addresses this challenge by integrating economic and qualitative dimensions of evaluation within a comparative, cluster-based framework. Drawing on twenty case studies from seven EU countries, the analysis aggregates local data into five coherent clusters representing alternative pathways toward herbicide-free management. This structure allows both for comparability and for the identification of mechanisms that foster or hinder transition across regions and governance systems. The article therefore seeks to translate empirical evidence on non-chemical weed management into a governance-relevant framework to inform future CAP eco-schemes and result-based payments.

Three research questions guide the analysis:

• How do the main NCWM clusters perform in economic terms when assessed through comparable cost–benefit parameters?

• What types of social and environmental co-benefits do these clusters generate, and how can they be evaluated in an integrated manner?

• How can the joint interpretation of private and public outcomes inform the design of governance instruments for the agro-ecological transition?

The article adopts the cluster as the unit of analysis. Results are reported as descriptive distributions, cluster medians and dispersion ranges, and as cluster-level socio-environmental scores (0–100). Economic and socio-environmental evidence are then jointly interpreted through a Dual-Axis Sustainability Matrix, whose purpose is diagnostic and comparative: to reveal patterns that connect private profitability with public co-benefits, and to derive policy-relevant implications.

2 Methodology

2.1 Analytical framework and data sources

This study integrates a quantitative economic branch (cost–benefit analysis, CBA) with a qualitative socio-environmental branch within a multi-actor design. The empirical basis is a portfolio of twenty NCWM practices drawn from seven European countries—France, Italy, Greece, Spain, Latvia, the United Kingdom, and Sweden—originating from innovation networks and OGs that had already deployed herbicide-free solutions under commercial conditions. For the purposes of this article, evidence is aggregated into five functional clusters: Mechanical Weeding (MW: n = 5; France, Italy, Latvia, United Kingdom, Greece), Cover Crops (CC: n = 3; France, Greece, Italy), Mulching (MU: n = 3; France, Spain, United Kingdom), Precision Farming (PF: n = 6; Italy, Latvia, Greece, United Kingdom, Sweden), and Ecological Management (EM: n = 3; Greece, Spain). This distribution is consistent with typologies widely used in the IWM literature (Riemens et al., 2021) and with the categories emerging from the OPER8 workshops (Guilbert et al., 2023). Clustering enables comparative, policy-relevant analysis while preserving traceability to the full documentation in Vergamini et al. (2024) and Guilbert et al. (2023).

2.2 Participatory design, data collection, and harmonization

The empirical foundation of this study stems from a two-stage participatory process that defined, validated, and operationalized the portfolio of NCWM practices assessed. An initial European workshop (31 experts from 12 institutions) co-designed a set of evaluation criteria, covering agronomic feasibility, operationality, economic costs, and environmental impacts, to ensure both comparability and analytical rigor. These criteria were refined and tested during seven national workshops (161 participants including farmers, advisors, researchers, and policymakers), which identified and validated three representative NCWM practices per country.

The participatory phase described in Guilbert et al. (2023) thus produced the portfolio of validated cases, while the subsequent analytical phase organized the resulting dataset into five functional clusters. This clustering was performed ex post by the OPER8 analytical team, based on technical, economic, and organizational similarities emerging from the harmonized questionnaires and cost–benefit data rather than predefined typologies. Each cluster therefore represents a synthetic analytical unit capturing recurrent configurations of technology, investment intensity, and governance modes, allowing consistent comparison across contexts and meaningful policy interpretation (Table 1).

The empirical dataset included 20 purposively selected cases for which harmonized economic and socio-environmental information was available: 15 were based on direct farm-level surveys, while five were documented from project records and validated through partner inputs, workshops, and stakeholder consultation. The dataset should therefore be interpreted as a comparative portfolio of farm-tested and validated NCWM implementation contexts, rather than as a statistically representative sample of European farms or fields. Details on each practice, including country, farm area, data basis, cropping situation, and practice description are provided in Supplementary Table S1.

Consistent with this sampling logic, data collection focused on holdings and practice contexts that had previously implemented or tested herbicide-free weed control within OGs and related innovation networks. This ensured that the information reflected either the experience of actors who had directly managed these practices, or, for validated cases, the knowledge of project partners and stakeholders familiar with the practice and its territorial context. Farmers reached through the OG networks provided detailed quantitative and qualitative data via structured questionnaires, complemented by focus-group discussions to clarify technical and economic parameters. The harmonized template (see Guilbert et al. (2023), Appendix C; Vergamini et al., 2024; Annex 1) consisted of three sections: (a) Economic data—investment and operating costs (CAPEX, OPEX, transaction costs), revenues, yields, and labor time; (b) Social and behavioral aspects—workload perception, occupational safety, training, cooperation, and innovation attitude; (c) Environmental dimensions—soil erosion, biodiversity, water use, and greenhouse-gas emissions.

Pilot interviews tested the clarity of definitions and units of measurement, after which national partners adapted the template to crop and regional contexts. In most cases, questionnaires were completed jointly by farmers and researchers or advisors, ensuring on-site validation of costs and assumptions. Where necessary, complementary information was retrieved from national farm-accounting databases (e.g., FADN, ISTAT) to verify consistency and fill data gaps.

All quantitative variables were normalized per hectare to enable cross-country comparability, while qualitative responses were converted into numerical scores compatible with the socio-environmental indicators. After national validation, datasets were standardized and centrally reviewed for outliers or inconsistencies. A series of calibration meetings between country teams and the analytical unit harmonized the interpretation of cost categories and environmental indicators.

The resulting harmonized dataset was then aggregated by cluster to identify recurring economic structures and governance patterns across heterogeneous contexts. This procedure produced a unified, quality-controlled database that constitutes the empirical foundation for the comparative analysis presented in this paper.

2.3 Multi-source validation

Beyond the survey data, additional insights from focus groups and expert panels were used to validate and contextualize the findings. This triangulation between farmer experience, expert judgement, and quantitative information ensured that the dataset remained both empirically grounded and representative of real farm conditions. The integrated workflow consisted of two analytical branches: a quantitative branch, deriving the Net Present Value per hectare (NPV/ha) as a measure of private profitability; and a qualitative branch, translating socio-environmental performance into a structured score. The two branches were later merged for joint interpretation.

2.4 Operational detail of the CBA

The CBA provides the quantitative backbone of the study, translating heterogeneous technical and agronomic information into a common analytical framework applied at farm level and then aggregated by cluster. Each NCWM option is benchmarked against its conventional herbicide-based counterpart over a 20-year horizon, used as a harmonized comparative window across heterogeneous practices. This time frame was selected to capture both machinery depreciation and slower environmental or social effects that would not be fully visible over shorter periods. Within this horizon, replacement cycles were standardized using conservative economic service-life assumptions to preserve comparability across heterogeneous practices and countries. In particular, machinery investments were modeled over repeated 10-year cycles, while shorter-lived agricultural tools were assumed to require replacement approximately every 7 years. These values were used as harmonized replacement parameters for analytical purposes and should not be interpreted as maximum physical lifespans. The main harmonized assumptions and common input parameters applied in the CBA are summarized in Table 2. All flows are expressed in real terms and discounted at 4%, with one-way sensitivity between 2% and 6% reflecting typical ranges in European agriculture (Hanley and Barbier, 2009). Financial and physical inputs derive directly from farmer questionnaires validated in national focus groups, ensuring that calculations reflect real management conditions rather than experimental assumptions.

The analysis adopts the private (farm) perspective, distinguishing: CAPEX (machinery, equipment, infrastructure, installation), OPEX (labor, energy, maintenance, inputs), Transaction costs—TC associated with learning, coordination, and monitoring, included explicitly following the agri-environmental transaction-cost literature (Mettepenningen et al., 2009).

To preserve European comparability, two environmental modules are monetized but kept analytically separate from private returns (i.e., not folded into NPV to avoid double counting social vs. private benefits): Soil-erosion losses valued using cost ranges from Panagos et al. (2018); Carbon-emission changes valued using the European Investment Bank (2023) CO₂ shadow-price trajectory.

These modules are reported as harmonized add-ons for comparability across clusters. On the revenue side, benefits include yield changes, quality premia (where applicable), and input savings (herbicides, energy, irrigation). All quantities are normalized per hectare, making NPV per hectare (NPV/ha) the central indicator for cross-context comparison. The NPV is computed as:

where B_t and C_t are annual benefits and costs, r the discount rate, and T the time horizon. A positive NPV (> 0) indicates long-term financial viability, while a negative value signals that, under current prices and costs, the practice remains privately unprofitable.

For visual comparability across heterogeneous clusters, median NPV values were transformed into a normalized indicator (X_norm):

where NPV_max is the highest positive cluster median NPV, and NPV_min is the lowest negative cluster median NPV observed in the sample. This transformation ensures that X_norm = 0 corresponds to NPV = 0 (break-even), X_norm = +1 to the highest positive cluster median, and X_norm = −1 to the lowest negative cluster median. The transformation was used only for graphical representation in the Dual-Axis Sustainability Matrix; the underlying economic interpretation remains based on raw median NPV values.

Uncertainty analysis combined one-way sensitivity checks and Monte Carlo robustness analysis at case level. One-way sensitivity tested the discount rate at 2%, 4%, and 6%, while the Monte Carlo procedure was based on 1,000 random simulations. For each selected monetized variable included in the case-level CBA, uncertainty was represented through three-point discrete specifications (low, default, high) with associated probabilities. Assuming independence across variables, random joint draws were used to generate simulated distributions of NPV and BCR. In the present article, these case-level robustness outputs were used to support the comparative interpretation of cluster-level results. Computations are run at farm/case level; for the comparative reading used in this article, statistics are aggregated by cluster (e.g., medians and dispersion) to reflect recurrent cost structures and risk profiles without reporting individual cases. The core assumptions and common input parameters used in the present article are reported in Table 2, while full templates, case-level parameter definitions, and computational details are provided in Vergamini et al. (2024).

2.5 Qualitative socio-environmental assessment: traffic-light indicator and validation

Many of the impacts associated with NCWM practices lie beyond the reach of market valuation. To complement the CBA, a composite traffic-light indicator was developed to capture the social and environmental co-benefits of the five NCWM clusters. The assessment follows a semi-quantitative, multi-criteria approach (Munda, 2008; Gómez-Limón and Sánchez-Fernández, 2010), combining literature- informed weighting with participatory validation.

The construction of the indicator began with an extensive review of existing impact-assessment frameworks and iterative consultation among researchers and practitioners involved in the OPER8 national workshops. A total of 19 indicators were selected–10 social (e.g., occupational safety, labor conditions, learning and knowledge sharing, community cooperation, and perceived innovation capacity) and 9 environmental (e.g., soil fertility, erosion control, water use, biodiversity, and greenhouse-gas emissions). Each indicator was characterized by its causal proximity to the practice (direct or indirect) and its temporal horizon (short-term, medium-term, or long-term effect). This two-fold classification allowed the analysis to differentiate between immediate and systemic effects, assigning proportionally higher weights to direct, short-term impacts, while acknowledging greater uncertainty for distant or system-level outcomes (Turner et al., 2003; OECD., 2018). The full set of 19 socio-environmental indicators, together with their numerical weights, causal proximity, and temporal horizon, is reported in Table 3.

Farmers provided first-hand perceptions through questionnaires, whereas workshop-based assessments were obtained through national multi-actor participatory sessions. Both sources contributed ratings for each socio-environmental indicator on a five-point Likert scale (1–5, with 3 as neutral). Indicator scores were normalized to a common 0–1 scale, multiplied by the indicator-specific weights reported in Table 3, and aggregated into social and environmental sub-scores. Source-specific weighted scores were then combined through a source-weighting procedure inspired by inverse-variance weighting (Borenstein et al., 2009). The resulting aggregated scores were then rescaled to a 0–100 range for interpretation. Importantly, variance was not estimated empirically from the observed dispersion of Likert responses or from inter-rater agreement statistics. Instead, predefined variance classes were assigned ex ante as proxies for the expected degree of convergence associated with each assessment context. Workshop-based assessments, produced in heterogeneous groups including farmers, advisors, researchers, and policymakers, were treated as potentially more dispersed than case-level questionnaire responses. Lower assigned variance implied greater weight in the final weighted average. The procedure should therefore be interpreted as a convergence-based source-weighting approach rather than as a formal meta-analytic inverse-variance estimator. The workshop component was based on seven national workshops involving 161 participants overall, whereas survey-based inputs were collected at case-study level for the 20 NCWM cases included in the analysis.

For interpretability purposes, aggregated cluster scores were classified according to a three-level traffic-light scheme: green (high) for scores between 67 and 100, yellow (medium) for scores between 34 and 66, and red (low) for scores between 0 and 33. This visual coding allows rapid identification of clusters with high public-benefit potential even when private profitability is limited. The indicator was refined through several validation cycles, including the transnational workshop in Pisa (April 2023), to improve methodological transparency and usability by practitioners. Its final configuration provides a transparent and consistent framework for comparing social and environmental outcomes across heterogeneous contexts, offering a complementary qualitative perspective to the quantitative evidence produced by the CBA.

2.6 Integration of private and public outcomes: the dual-axis sustainability matrix

The economic and socio-environmental branches of assessment jointly form a mixed-methods framework that distinguishes private profitability from public-good generation while enabling their combined interpretation. The CBA captures the farm-level financial rationale for adoption, indicating whether an innovation is economically viable for an individual holding, while the traffic-light indicator measures the broader set of social and environmental benefits extending beyond the farm gate. Their integration offers a coherent view of how incentives, risks, and collective values interact in shaping agro-ecological transitions across Europe. The final stage of the analytical framework integrates these two branches into a single interpretative space. Rather than attempting full commensuration between monetary and non-monetary values, the goal is to visualize their interaction and reveal typological patterns across the five NCWM clusters. This step acts as a conceptual bridge between farm-level economics and public-good governance, illustrating how different forms of innovation distribute costs and benefits between private actors and society (Hanley and Barbier, 2009; OECD., 2018). Operationally, the integration is represented through a Dual-Axis Sustainability Matrix, a two-dimensional model inspired by portfolio-analysis approaches such as the Boston Matrix but conceptually aligned with sustainability-portfolio analysis in agricultural economics (Gómez-Limón and Sánchez-Fernández, 2010; van Grinsven et al., 2015). The NPV per hectare (NPV/ha) derived from the CBA occupies the horizontal axis, expressing the private economic dimension, while the Socio-Environmental Impact Score from the qualitative assessment forms the vertical axis, representing the collective dimension of value creation. Each cluster is thus positioned according to its median private profitability and public-value contribution, allowing a transparent and comparable reading of performance across diverse contexts (Mettepenningen et al., 2009; Panagos et al., 2018). Four stylized quadrants emerge from this configuration: (a) Win–win innovations, positive NPV and high socio-environmental scores, where technological and organizational change jointly enhance profitability and sustainability; (b) Trade-offs, positive private performance and below-high socio-environmental scores; (c) Public-good-dominant innovations, negative private performance and high socio-environmental scores, reflecting forms of public-value generation that are not fully matched by private returns under current conditions (Pascual et al., 2012; Gómez-Baggethun et al., 2010); (d) Non-viable or privately constrained options, negative private performance and below-high socio-environmental scores. In operational terms, the vertical split is defined by break-even private performance (NPV = 0), while the horizontal split corresponds to the threshold between medium and high socio-environmental performance in the traffic-light scheme (score = 67).

Although visually simple, the Dual-Axis Matrix encapsulates one of the core challenges of the agro-ecological transition: the temporal and spatial misalignment between those who bear the costs and those who capture the benefits—a coordination failure well documented in agri-environmental policy (Paavola and Adger, 2005; Mettepenningen et al., 2009; Hanley and Barbier, 2009). The matrix therefore acts not merely as a descriptive chart but as a diagnostic interface between empirical evidence and policy design. It enables researchers and policymakers to identify where intervention is most needed—through investment support, cooperative mechanisms, or result-based payments for ecosystem services—and to distinguish which innovations may scale autonomously through markets and which require collective or institutional mediation (Reed et al., 2018; Klerkx et al., 2012). Conceptually, this integration contributes to the broader literature on plural valuation and the governance of ecosystem services (Jacobs et al., 2016; Kenter et al., 2016). By juxtaposing monetary and non-monetary dimensions without reducing one to the other, the Dual-Axis Sustainability Matrix operationalises the coexistence of multiple value registers within sustainability assessment. Each axis preserves its analytical logic—financial performance on one side, socio-ecological contribution on the other—while their intersection reveals the spectrum of motivations, constraints, and interdependencies that shape farmer decision-making.

In this sense, the matrix translates a mixed-methods framework into an intuitive visual grammar connecting micro-economic reasoning with macro-level governance objectives under the European Green Deal and the evolving Common Agricultural Policy (CAP). The methodology does not aim to rank clusters prescriptively but to provide a comparative space for interpreting how economic and socio-environmental performances co-evolve under different conditions—recognizing that sustainable weed management represents not a single optimal pathway but a portfolio of viable trajectories, each requiring a specific combination of incentives, cooperation, and learning to achieve scale and resilience.

3 Results

The analysis addresses the three guiding questions defined in Section 1, linking the structure of costs and benefits across the five NCWM clusters (Q1), the generation of socio-environmental co-benefits (Q2), and their combined interpretation through the integrated CBA–traffic-light framework (Q3). Because the analysis aggregates 20 purposively selected cases into clusters of only 3 to 6 observations, the statistics reported below should be interpreted as descriptive and indicative rather than inferential. Cluster medians and dispersion ranges are used to summarize within-sample patterns, but they are not statistically representative of the wider European farming population and cannot be generalized without further validation.

Across the twenty case studies consolidated into five functional clusters (MW, CC, MU, PF, and EM), the results reveal a differentiated spectrum of economic and socio-ecological performances. The five clusters do not align along a single gradient of performance. Instead, they differ in the balance between total costs, expected benefits, private returns, socio-environmental scores, and robustness under uncertainty. At portfolio level, private profitability and socio-environmental performance do not move in strict opposition: socio-environmental scores remain medium to high across all clusters, whereas positive private performance is observed only in a limited part of the sample.

3.1 General cost–benefit patterns

Cluster-level CBA results show substantial variation in both the magnitude and composition of costs. As shown in Table 4, median total costs are lowest in MW (4,408 EUR/ha) and CC (6,864 EUR/ha), intermediate in PF (8,753 EUR/ha), and considerably higher in EM (27,583 EUR/ha) and MU (36,414 EUR/ha). Median benefits also vary widely, from 1,987 EUR/ha in MW to 18,491 EUR/ha in MU. These values show that higher benefits do not necessarily correspond to positive private returns, because the benefit profile of each cluster must be read together with its cost burden and dispersion range.

Median NPV values further differentiate the clusters. CC is the only cluster with a positive median NPV (+2,902 EUR/ha). All other clusters remain negative in median private terms: MW records −2,859 EUR/ha, precision farming −4,329 EUR/ha, EM −17,584 EUR/ha, and MU −20,690 EUR/ha. Dispersion is wide in several clusters, especially for MW and PF, indicating substantial heterogeneity across cases within the same cluster. This heterogeneity is visible not only in the spread of NPV values, but also in the ranges observed for total costs and benefits.

The cost structure further differentiates the five clusters (Figure 1). MW shows a predominantly capital-based profile, with CAPEX representing the largest share of costs. PF also shows a strong capital component, although with a more balanced distribution between CAPEX and OPEX. CC, MU, and especially EM are more strongly shaped by operating and recurring costs, with EM being almost entirely driven by OPEX in relative terms. In MU, high total costs are associated with a combination of material, labor, and logistics-related burdens.

Figure 1

The Monte Carlo summary confirms the same general ordering (Table 5). CC is the only cluster with a clearly positive median probability that NPV exceeds zero (0.88). For MW, MU, PF, and EM, the median probability remains zero. At the same time, the probability range extends from 0.00 to 1.00 in MW, PF, and EM, indicating that positive simulated outcomes occur in some cases but do not characterize the median cluster profile. MU is the most consistently negative cluster under simulation, with a probability range of 0.00–0.00 across cases.

Taken together, these results show that the observed differences between clusters arise from a combination of cost intensity, cost composition, and within-cluster variability. They also show that positive private performance is not widespread across the portfolio and remains concentrated in a limited subset of cases and clusters.

3.2 Cluster-specific results

The integrated comparison of private and socio-environmental performance is summarized in Table 6 and Figure 2. These results place the five clusters along a differentiated continuum rather than distributing them evenly across the analytical space.

Figure 2

Vergamini et al., 2024) and summarize aggregated cluster patterns; individual cases are not shown.

CC combines the highest normalized private performance with a high socio-environmental score, whereas EM combines the highest socio-environmental score with clearly negative private performance. MW, MU, and PF occupy intermediate positions characterized by medium socio-environmental scores and negative private returns.

MW, represented by five cases from France, Italy, Latvia, the United Kingdom, and Greece, combines a median NPV of −2,859 EUR/ha with a socio-environmental score of 61.9. Median total costs are relatively low compared with the other clusters, but the observed range is extremely broad, indicating strong case-to-case variation. The cluster also shows a broad probability range in the Monte Carlo summary, from 0.00 to 1.00, despite a median probability of zero. Within the negative-NPV clusters, MW is the closest to the break-even threshold in the absolute values reported in Table 4.

CC, represented by three cases from France, Greece, and Italy, is the only cluster with a positive median NPV (+2,902 EUR/ha) and combines this with a high socio-environmental score (70.6). Its median total costs remain moderate relative to the portfolio, while the Monte Carlo summary shows a median probability of positive NPV equal to 0.88, higher than in any other cluster. Within the integrated comparison, CC therefore stands out as the only cluster in which positive private performance and high socio-environmental performance coincide at the median level.

MU, represented by three cases from France, Spain, and the United Kingdom, records the highest median total costs in the portfolio (36,414 EUR/ha) and, despite the highest median benefits (18,491 EUR/ha), also the lowest median NPV (-20,690 EUR/ha). Its socio-environmental score remains in the medium range (61.5). The Monte Carlo results are uniformly unfavorable at cluster level, with a median probability of positive NPV equal to zero and a probability range of 0.00–0.00. This makes MU the most economically constrained cluster in the sample, even though its socio-environmental profile remains positive.

PF includes six cases from Italy, Latvia, Greece, the United Kingdom, and Sweden. It combines a negative median NPV (−4,329 EUR/ha) with the lowest socio-environmental score among the five clusters (55.5). Median total costs are lower than in MU and EM but higher than in MW and CC. The Monte Carlo summary indicates a median simulated NPV of −2,013.73 EUR/ha and a probability range of 0.00–1.00, while the median probability of positive NPV remains zero. This profile indicates heterogeneous outcomes and some positive cases under simulation, but without a positive central tendency at cluster level.

EM, represented by three cases from Greece and Spain, combines the highest socio-environmental score in the sample (71.8) with a strongly negative median NPV (−17,584 EUR/ha). Median total costs are high (27,583 EUR/ha), although lower than in MU, and benefits remain positive but insufficient to generate a positive median private return. The Monte Carlo summary shows a median probability of positive NPV equal to zero, with a range from 0.00 to 1.00, indicating that positive simulated outcomes occur in some cases but do not alter the central cluster profile. Within the integrated comparison, EM corresponds to the clearest case of high socio-environmental performance combined with weak private profitability.

These cluster-level results show that the five NCWM pathways differ not only in average performance but also in the balance between costs, benefits, and uncertainty.

3.3 Cross-cluster synthesis

Viewed as a whole, the integrated assessment of the five NCWM clusters reveals not a binary divide between viable and unviable practices, but a continuum of configurations across private and socio-environmental performance. CC occupies the most favorable position in this continuum, combining a positive median NPV (+2,902 EUR/ha), a high socio-environmental score (70.6), and the only clearly positive Monte Carlo profile at cluster level (median Prob. NPV > 0 = 0.88). At the opposite end, EM combines the highest socio-environmental score in the sample (71.8) with a strongly negative median NPV (−17,584 EUR/ha), while MU records the lowest median NPV (−20,690 EUR/ha) despite medium socio-environmental performance (61.5) and the highest median benefits (18,491 EUR/ha). MW and PF occupy intermediate positions, with negative median private performance, medium socio-environmental scores, and broader internal variability.

Across the portfolio as a whole, socio-environmental performance is more homogeneous than private profitability. All five clusters record medium to high socio-environmental scores, whereas only one cluster shows a positive median NPV. The Monte Carlo results reinforce the same pattern: isolated positive cases are present in several clusters, but only CC shows a clearly positive probability profile at the median level. Within the observed sample, the main empirical contrast therefore lies between clusters that generate socio-environmental value without positive private returns and the single cluster in which the two dimensions coincide.

4 Discussion

The results expose a central paradox in Europe's agro-ecological transition: credible alternatives to chemical herbicides exist and generate medium to high socio-environmental value across all five clusters, yet positive private profitability remains limited to only one cluster. The integrated CBA–traffic-light analysis shows that the adoption gap in non-chemical weed management cannot be explained by technical immaturity alone. Rather, it reflects a persistent misalignment between the distribution of private costs and the generation of broader social and environmental benefits. In this sense, the results indicate that the diffusion of NCWM depends on the interaction between economic viability, uncertainty, organizational arrangements, and the governance conditions under which practices are implemented.

4.1 Economic viability and organizational models

Economic viability constitutes the first threshold of diffusion. In median terms, CC is the only cluster that combines positive private profitability with high socio-environmental performance. MW is the closest of the negative-NPV clusters to the break-even threshold, while PF also occupies an intermediate position but remains negative in median terms. MU records the weakest private performance, whereas EM combines strongly negative private returns with the highest socio-environmental score. The weak private performance of MU can be explained by the specific cost structure of the mulching practices included in the cluster. In the analyzed cases, mulching systems involved surface-cover practices based on organic residues, compost, or biodegradable films. Although these practices generated relevant agronomic benefits, including reduced herbicide use, soil-moisture retention, and potential circular use of organic materials, their private economic performance was constrained by recurrent material, labor, transport, and application costs. Unlike capital-intensive options such as MW or PF, where fixed costs may be diluted through machinery sharing, contractor services, or higher utilization rates, the main cost components of MU tend to recur with each application cycle and often increase with the treated area. This explains why MU recorded the highest median benefits in the portfolio but also the highest median total costs and the lowest median NPV. Its unfavorable Monte Carlo profile further indicates that, within the observed sample, these recurring OPEX and logistics-related burdens were not sufficiently offset by monetized private benefits. Taken together, these patterns show that private viability depends not only on technical effectiveness, but also on how costs are structured and organized.

This distinction is especially clear when the cost profiles of the clusters are compared. MW and PF are more exposed to capital requirements, machinery use, and infrastructure-related constraints, whereas MU and EM are more strongly shaped by recurring operating, labor, and opportunity costs. The results therefore suggest that profitability depends less on the intrinsic technical principle alone than on the organizational setting through which innovation is delivered. This interpretation is consistent with the transaction-cost literature in agri-environmental schemes, which shows that coordination, learning, and monitoring costs can offset technical gains when farmers act individually (Mettepenningen et al., 2009; Klerkx et al., 2012).

The results also point to differentiated pathways of economic adjustment. In clusters such as MW and PF, cooperative or service-based models can potentially reduce the burden of fixed costs and improve utilization rates. By contrast, MU and EM remain weakly rewarded in private terms even when their socio-environmental contributions are substantial. CC stands apart because it shows that, under some conditions, positive private returns and high socio-environmental value can coincide. This suggests that the economic viability of NCWM is not a binary property of technologies themselves, but a relational outcome shaped by farm scale, organizational form, and institutional support.

4.2 Risk, uncertainty, and temporal asymmetry

A second key finding concerns the temporal asymmetry between private costs and public benefits. Across the portfolio, costs are incurred early and directly by farmers, whereas many socio-environmental gains materialize gradually and often beyond the farm boundary. This asymmetry is especially visible in EM and MU, but it also affects MW and PF, whose economic viability remains conditional and uneven. In this sense, the weak private incentives observed for most clusters are not simply a reflection of low technical performance, but of a timing mismatch between those who bear the costs and those who capture the benefits.

This pattern has two main implications. First, farmers face uncertainty not only in agronomic performance but also in the timing and recoverability of investments. Second, support mechanisms cannot rely on ex post compensation alone, but need to reduce exposure during the transition phase. The results are therefore consistent with the relevance of de-risking tools that complement conventional support measures, including risk-sharing arrangements, phased payments, cooperative investments, and advisory systems that reduce uncertainty through learning and demonstration. Performance-linked insurance, blended-finance instruments combining public and private capital, and carbon- or biodiversity-credit schemes may be particularly relevant in this respect, as they can help buffer early adopters against transition losses (Farber and Reichert, 2023). Overall, the evidence suggests that NCWM adoption should be understood as a transition problem under uncertainty rather than as a simple input-substitution choice.

4.3 Collective value creation and spatial coordination

The integrated results also highlight that the strongest socio-environmental contributions are not always captured within the farm gate. Several of the outcomes measured in the traffic-light assessment—soil protection, biodiversity enhancement, carbon-related effects, and landscape-level ecological functions—are likely to extend beyond individual farms and to accumulate through spatial coordination. This is particularly relevant for EM, whose socio-environmental scores remain relatively strong even when private returns are weak. The results therefore support the interpretation of many NCWM benefits as partly collective or public in character.

This dimension matters because it changes the scale at which performance should be evaluated and supported. If relevant benefits are diffuse, delayed, or jointly produced across neighboring farms, then individual adoption decisions alone may underprovide them. In this sense, the results are consistent with literature on collective agri-environmental action and coordinated contracts, which shows that group-based approaches can improve ecological coherence and reduce transaction costs in landscape-scale management (Carmona-Torres et al., 2011; Nguyen et al., 2022). The OPER8 cases suggest that clusters with stronger socio-environmental profiles may particularly benefit from organizational settings that enable coordination, knowledge exchange, and shared commitment over time.

This does not mean that all clusters require the same degree of collective organization. The results rather suggest differentiated needs. MW and PF appear more sensitive to access conditions, service arrangements, and infrastructure, whereas EM and MU appear more sensitive to the broader institutional recognition of non-market benefits. What emerges is not a single collective-action model, but a spectrum of governance needs, from shared machinery and advisory services to longer-term cooperative and landscape-scale arrangements.

4.4 Hybrid governance architectures and digital enablers

The differentiated profiles observed across clusters imply that no single policy instrument is likely to be effective across the full NCWM portfolio. This interpretation is consistent with the literature on result-based agri-environment schemes, which shows that payments linked to verifiable environmental outcomes may improve ecological effectiveness, but also raise issues of monitoring, implementation, and farmer exposure to performance risk (Burton and Schwarz, 2012; Herzon et al., 2017). CC, as the only cluster with jointly favorable private and socio-environmental performance, may benefit primarily from instruments that stabilize adoption and reward verified ecosystem outcomes. MW appears more responsive to cooperative machinery access, service-based provision, and investment facilitation. PF requires enabling conditions linked to infrastructure, training, advisory support, and data management. EM and MU, by contrast, remain more clearly dependent on longer-term public or collective support because their socio-environmental benefits are not matched by positive median private returns.

Within this context, digitalisation should not be treated as a marginal technical add-on. The results support its role as a potential enabler of result-based governance, particularly where verification and monitoring remain barriers to implementation. Remote sensing, in-field sensors, and digital field records can make it easier to track soil cover, vegetation dynamics, input use, or management practices, thereby supporting more credible payment designs and adaptive contracts. This matters especially for clusters whose justification rests partly on public-good provision rather than on private profitability alone.

More broadly, the results are consistent with governance mixes that combine result-based payments for verifiable ecosystem outcomes, risk-sharing instruments to buffer uncertainty, and knowledge-exchange systems that build adaptive capacity and trust among farmers, advisors, and policymakers (Klerkx et al., 2012; OECD., 2021).

4.5 Limitations and interpretive scope

The study also has limitations that define the scope of inference. First, the empirical base consists of 20 purposively selected cases organized into clusters of 3 to 6 observations. The reported statistics are therefore descriptive and indicative rather than inferential, and they should not be interpreted as representative estimates for the wider European farming population. Second, the cases were drawn from Operational Groups and related innovation networks that had already implemented or actively tested herbicide-free approaches. This likely favors contexts with stronger support conditions, greater motivation, or more mature experimentation, and may therefore overstate both economic and socio-environmental performance relative to broader farming populations.

A third limitation concerns aggregation. Clustering improves comparability and policy readability, but it also compresses internal heterogeneity. This is particularly visible in MW and PF, where wide cost, benefit, and NPV ranges indicate that outcomes vary substantially across cases. Finally, the socio-environmental indicator remains a structured but partly judgement-based composite, since it relies on weighting choices and participatory scoring. In addition, the concentration of cluster scores within the medium-to-high range may reflect both the generally positive profile of the selected innovation cases and the limited discriminating power of an aggregated composite indicator across heterogeneous practices.

These limitations do not invalidate the framework, but they suggest that it should be interpreted as a comparative diagnostic tool rather than as a population-level ranking device. Within those limits, the analysis still provides useful evidence on the differentiated alignment between private returns, public value, and governance conditions across NCWM pathways.

5 Conclusion and policy implication

This study set out to explain why non-chemical weed management remains only partially adopted in Europe despite the availability of mature technical alternatives. The results indicate that the central barrier is not the absence of workable technologies, but the persistent misalignment between private profitability and socio-environmental value. Across the five NCWM clusters, socio-environmental performance is consistently medium to high, whereas positive private profitability is observed only in one cluster. The main empirical contrast is therefore not between practices that do or do not generate public value, but between practices that do and do not translate that value into positive private returns under current conditions.

5.1 Key insights for agri-environmental governance

Three conclusions follow from this evidence. First, NCWM should be understood as a differentiated portfolio rather than as a single transition pathway.

Second, the results support differentiated policy responses rather than one-size-fits-all support. Clusters that are closer to private viability appear more likely to respond to cooperative investment, shared service provision, training, and infrastructure support. Clusters with stronger public-good profiles and persistently negative private returns are more likely to require longer-term result-based payments, collective contracts, or payment-for-ecosystem-services logics, consistent with the broader literature on conditional ecosystem-service incentives and their institutional design (Wunder et al., 2018; Salzman et al., 2018). The evidence is therefore consistent with hybrid incentive architectures that combine investment support, risk-sharing mechanisms, and performance-based remuneration according to the specific configuration of costs, returns, and co-benefits observed in each cluster.

Third, uncertainty and timing matter as much as average returns. This indicates that policy design should address not only average profitability but also the unevenness and fragility of returns during transition. In that respect, de-risking tools and adaptive support mechanisms are not secondary complements; they are part of the core governance architecture needed to make non-chemical pathways more credible to farmers.

5.2 From compliance to performance-based support

A central contribution of the paper lies in the mixed-method framework itself. By combining CBA, a structured socio-environmental traffic-light assessment, and a dual-axis matrix, the paper approach makes it possible to compare clusters without reducing all values to a single metric. This matters because transitions in weed management involve values that are only partly commensurable: private financial performance, environmental outcomes, social effects, and organizational requirements do not evolve on the same scale or over the same time horizon.

This makes the framework relevant not only for analysis, but also for policy design. In particular, it provides a structured basis for moving from practice-based support toward more performance-oriented governance. Where public benefits must be verified to justify differentiated payments, digital monitoring can play a strategic enabling role (Herzon et al., 2017). Remote sensing, in-field sensors, and digital field records can support the measurement of vegetation cover, input reduction, or management outcomes, thereby lowering some of the monitoring frictions that often limit result-based schemes. At the same time, these tools require advisory support, digital capacity, and clear data-governance arrangements if they are to remain inclusive and credible.

5.3 Transferability and future research

Beyond weed management, the framework proposed here is transferable to other agri-environmental transitions in which private and public value do not coincide—for example nutrient management, carbon farming, water-saving systems, or biodiversity-oriented practices. Its main strength lies in combining economic realism with participatory valuation, so that non-market effects remain visible without being forced into purely monetary terms.

Future research should extend the analysis in three directions. First, larger and more longitudinal datasets are needed to assess how private returns evolve beyond the case-study horizon and whether currently negative clusters improve over time. Second, cooperative and result-based arrangements should be tested more directly in field applications, especially for clusters with weak private incentives. Third, the socio-environmental framework could be expanded and refined through additional indicators and validation procedures, particularly for biodiversity, landscape functions, and digitally verifiable outcomes. These extensions would strengthen the empirical basis for performance-based governance while preserving the comparative logic developed in this study.

References

• 1

  BorensteinM.HedgesL. V.HigginsJ. P. T.RothsteinH. R. (2009). Introduction to meta-analysis.Chichester, UK: John Wiley and Sons. doi: 10.1002/9780470743386

  • CrossRef
  • Google Scholar
  • View reference in article

• 2

  BurtonR. J.SchwarzG. (2012). Result-oriented agri-environmental schemes in Europe and their potential for promoting behavioural change. Land Use Policy30, 628–641. doi: 10.1016/j.landusepol.2012.05.002

  • CrossRef
  • Google Scholar
  • View reference in article

• 3

  Carmona-TorresC.Parra-LópezC.GrootJ. C.RossingW. A. (2011). Collective action for multi-scale environmental management: achieving landscape policy objectives through cooperation of local resource managers. Landsc. Urban Plan.103, 24–33. doi: 10.1016/j.landurbplan.2011.05.009

  • CrossRef
  • Google Scholar
  • View reference in article

• 4

  DeVetterL. W.DilleyC. A.NonneckeG. R. (2015). Mulches reduce weeds, maintain yield, and promote soil quality in a continental-climate vineyard. Am. J. Enol. Vitic.66, 54–64. doi: 10.5344/ajev.2014.14064

  • CrossRef
  • Google Scholar
  • View reference in article

• 5

  European Commission (2020). Report from the Commission to the European Parliament and the Council on the Experience Gained by Member States on the Implementation of National Targets Established in their National Action Plans and on Progress in the Implementation of Directive 2009/128/EC on the Sustainable use of Pesticides (COM(2020) 204 Final).Brussels: European Commission. Available online at: https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A52020DC0204 (Accessed July 3, 2026).

  • Google Scholar
  • View reference in article

• 6

  European Investment Bank (2023). EIB project Carbon Footprint Methodology: Methodologies for the Assessment of Greenhouse Gas Emissions and Emission Variations. European Investment Bank. Available online at: https://www.eib.org/attachments/lucalli/eib_project_carbon_footprint_methodologies_2023_en.pdf (Accessed July 3, 2026).

  • Google Scholar
  • View reference in article

• 7

  FarberV.ReichertP. (2023). “Blended finance and the SDGs: using the spectrum of capital to de-risk business model transformation,” in Measuring sustainability and CSR: From Reporting to Decision-making, eds. S. Kacanski, J. Kabderian Dreyer, and K. J. Sund (Cham: Springer International Publishing),37–48. doi: 10.1007/978-3-031-26959-2_5

  • CrossRef
  • Google Scholar
  • View reference in article

• 8

  GagliardiL.FontanelliM.LuglioS. M.FrasconiC.PeruzziA.RaffaelliM.et al. (2023). Evaluation of sustainable strategies for mechanical under-row weed control in the vineyard. Agronomy13:3005. doi: 10.3390/agronomy13123005

  • CrossRef
  • Google Scholar
  • View reference in article

• 9

  GagliardiL.TramacereL. G.AntichiD.FrasconiC.SbranaM.SileoniG.et al. (2026). Alternative tactics to herbicides in integrated weed management: a Europe-centered systematic literature review. Agronomy16:220. doi: 10.3390/agronomy16020220

  • CrossRef
  • Google Scholar
  • View reference in article

• 10

  GengerR. K.RouseD. I.CharkowskiA. O. (2018). Straw mulch increases potato yield and suppresses weeds in an organic production system. Biol. Agric. Hortic.34, 53–69. doi: 10.1080/01448765.2017.1371077

  • CrossRef
  • Google Scholar
  • View reference in article

• 11

  Gómez-BaggethunE.De GrootR.LomasP. L.MontesC. (2010). The history of ecosystem services in economic theory and practice: from early notions to markets and payment schemes. Ecol. Econ.69, 1209–1218. doi: 10.1016/j.ecolecon.2009.11.007

  • CrossRef
  • Google Scholar
  • View reference in article

• 12

  Gómez-LimónJ. A.Sánchez-FernándezG. (2010). Empirical evaluation of agricultural sustainability using composite indicators. Ecol. Econ.69, 1062–1075. doi: 10.1016/j.ecolecon.2009.11.027

  • CrossRef
  • Google Scholar
  • View reference in article

• 13

  GuilbertC.PrezmanF.TatnellL.GodfreyK.SmeeM.DearloveE.et al. (2023). OPER8 project – deliverable 3.1: assessment of national action plans (WP3: assessment of solutions and development of the final inventory of best practices). Internal project deliverable, Horizon Europe Project OPER8.

  • Google Scholar
  • View reference in article

• 14

  HanleyN.BarbierE. B. (2009). Pricing nature: Cost–benefit Analysis and Environmental Policy.Cheltenham, UK: Edward Elgar Publishing.

  • Google Scholar
  • View reference in article

• 15

  HerzonI.BirgeT.AllenB.PovellatoA.VanniF.HartK.et al. (2017). Time to look for evidence: results-based approach to biodiversity conservation on farmland in Europe. Land Use Policy71, 347–354. doi: 10.1016/j.landusepol.2017.12.011

  • CrossRef
  • Google Scholar
  • View reference in article

• 16

  JacobsS.DendonckerN.Martín-LópezB.BartonD. N.Gómez-BaggethunE.BoeraeveF.et al. (2016). A new valuation school: integrating diverse values of nature in resource and land-use decisions. Ecosyst. Serv.22, 213–220. doi: 10.1016/j.ecoser.2016.11.007

  • CrossRef
  • Google Scholar
  • View reference in article

• 17

  KenterJ. O.BryceR.ChristieM.CooperN.HockleyN.IrvineK. N.et al. (2016). Shared values and deliberative valuation: future directions. Ecosyst. Serv.21, 358–371. doi: 10.1016/j.ecoser.2016.10.006

  • CrossRef
  • Google Scholar
  • View reference in article

• 18

  KlerkxL.Van MierloB.LeeuwisC. (2012). “Evolution of systems approaches to agricultural innovation: concepts, analysis and interventions,” in Farming Systems Research into the 21st Century: The New Dynamic, eds. I. Darnhofer, D. Gibbon, and B. Dedieu (Dordrecht: Springer), 457–483. doi: 10.1007/978-94-007-4503-2_20

  • CrossRef
  • Google Scholar
  • View reference in article

• 19

  KupF.SaglamR. (2014). Weed destruction in cotton fields using hot foam method and its comparison to certain other methods. ARPN J. Agric. Biol. Sci.9, 301–307.

  • Google Scholar
  • View reference in article

• 20

  LechenetM.DessaintF.PyG.et al. (2017). Reducing pesticide use while preserving crop productivity and profitability on arable farms. Nature Plants 3:17008. doi: 10.1038/nplants.2017.8

  • CrossRef
  • Google Scholar
  • View reference in article

• 21

  LiebhardG.GuzmánG.GómezJ. A.WinterS.ZallerJ. G.BauerT.et al. (2024). Vineyard cover crop management strategies and their effect on soil properties across Europe. Eur. J. Soil Sci.75:e13573. doi: 10.1111/ejss.13573

  • CrossRef
  • Google Scholar
  • View reference in article

• 22

  MelanderB.LiebmanM.DavisA. S.GallandtE. R.BàrberiP.MoonenA. C.et al. (2017). “Non-chemical weed management,” in Weed Research: Expanding Horizons, eds. P. E. Hatcher and R. J. Froud-Williams (Chichester: John Wiley & Sons), 245–270. doi: 10.1002/9781119380702.ch9

  • CrossRef
  • Google Scholar
  • View reference in article

• 23

  MerfieldC. N. (2015). False and Stale Seedbeds: The Most Effective Non-chemical Weed Management Tools for Cropping and Pasture Establishment.Lincoln: The BHU Future Farming Centre. Available online at: https://www.bhu.org.nz/wp-content/uploads/sites/155/ffc-files/weed-management/false-and-stale-seedbeds–the-most-effective-non-chemical-weed-management-tools-for-cropping-and-pasture-establishment-2015-ffc-merfield.pdf (Accessed July 3, 2026).

  • Google Scholar
  • View reference in article

• 24

  MettepenningenE.VerspechtA.Van HuylenbroeckG. (2009). Measuring private transaction costs of European agri-environmental schemes. J. Environ. Plan. Manag.52, 649–667. doi: 10.1080/09640560902958206

  • CrossRef
  • Google Scholar
  • View reference in article

• 25

  MundaG. (2008). Social Multi-criteria Evaluation for a Sustainable Economy.Berlin: Springer-Verlag. doi: 10.1007/978-3-540-73703-2

  • CrossRef
  • Google Scholar
  • View reference in article

• 26

  NguyenC.Latacz-LohmannU.HanleyN.SchilizziS.IftekharS. (2022). Spatial coordination incentives for landscape-scale environmental management: a systematic review. Land Use Policy114:105936. doi: 10.1016/j.landusepol.2021.105936

  • CrossRef
  • Google Scholar
  • View reference in article

• 27

  OECD. (2018). Cost-Benefit Analysis and the Environment: Further Developments and Policy Use.Paris: OECD Publishing. doi: 10.1787/9789264085169-en

  • CrossRef
  • Google Scholar
  • View reference in article

• 28

  OECD. (2021). Making Better Policies for Food Systems.Paris: OECD Publishing. doi: 10.1787/ddfba4de-en

  • CrossRef
  • Google Scholar
  • View reference in article

• 29

  PaavolaJ.AdgerW. N. (2005). Institutional ecological economics. Ecol. Econ.53, 353–368. doi: 10.1016/j.ecolecon.2004.09.017

  • CrossRef
  • Google Scholar
  • View reference in article

• 30

  PanagosP.StandardiG.BorrelliP.LugatoE.MontanarellaL.BoselloF.et al. (2018). Cost of agricultural productivity loss due to soil erosion in the European Union: from direct cost evaluation approaches to the use of macroeconomic models. Land Degrad. Dev.29, 471–484. doi: 10.1002/ldr.2879

  • CrossRef
  • Google Scholar
  • View reference in article

• 31

  PascualU.MuradianR.BranderL.Gómez-BaggethunE.Martín-LópezB.VermaM.et al. (2012). “The economics of valuing ecosystem services and biodiversity,” in The Economics of Ecosystems and Biodiversity: Ecological and Economic Foundations, ed. P. Kumar (London: CRC, Taylor and Francis), pp. 183–256. doi: 10.4324/9781849775489

  • CrossRef
  • Google Scholar
  • View reference in article

• 32

  RasmussenI. A. (2004). The effect of sowing date, stale seedbed, row width and mechanical weed control on weeds and yields of organic winter wheat. Weed Res.44, 12–20. doi: 10.1046/j.1365-3180.2003.00367.x

  • CrossRef
  • Google Scholar
  • View reference in article

• 33

  ReedM. S.VellaS.ChalliesE.VenteD. e.FrewerJ.Hohenwallner-RiesL.et al. (2018). A theory of participation: what makes stakeholder and public engagement in environmental management work?Restor Ecol.26, S7–S17. doi: 10.1111/rec.12541

  • CrossRef
  • Google Scholar
  • View reference in article

• 34

  RiemensM.SønderskovM.MoonenA.StorkeyJ.KudskP. (2021). An integrated weed management framework: a pan-European perspective. Eur. J. Agron.133:126443. doi: 10.1016/j.eja.2021.126443

  • CrossRef
  • Google Scholar
  • View reference in article

• 35

  SalzmanJ.BennettG.CarrollN.GoldsteinA.JenkinsM. (2018). The global status and trends of payments for ecosystem services. Nat. Sustain.1, 136–144. doi: 10.1038/s41893-018-0033-0

  • CrossRef
  • Google Scholar
  • View reference in article

• 36

  TurnerR. K.PaavolaJ.CooperP.FarberS.JessamyV.GeorgiouS.et al. (2003). Valuing nature: lessons learned and future research directions. Ecol. Econ.46, 493–510. doi: 10.1016/S0921-8009(03)00189-7

  • CrossRef
  • Google Scholar
  • View reference in article

• 37

  van GrinsvenH. J. M.BouwmanL.CassmanK. G.van EsH. M.McCrackinM. L.BeusenA. H.et al. (2015). Losses of ammonia and nitrate from agriculture and their effect on nitrogen recovery in the European Union and the United States between 1900 and 2050. J. Environ. Qual.44, 356–367. doi: 10.2134/jeq2014.03.0102

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 38

  VergaminiD.PaganoL.AntichiD.TramacereL. G. (2024). Assessment of Non-chemical Weed Control Solutions in Oper8 Project. Geneva: Zenodo. doi: 10.5281/zenodo.17609226

  • CrossRef
  • Google Scholar
  • View reference in article

• 39

  WunderS.BrouwerR.EngelS.Ezzine-De-BlasD.MuradianR.PascualU.et al. (2018). From principles to practice in paying for nature's services. Nat. Sustain.1, 145–150. doi: 10.1038/s41893-018-0036-x

  • CrossRef
  • Google Scholar
  • View reference in article