1 Introduction

Bioeconomy has great appeal as a potential means of achieving green growth and boosting competitiveness (Lühmann and Vogelpohl 2023). Bioeconomy plays a key role in driving economic growth in Europe and serves as a critical component in realizing a carbon–neutral future (Papadopoulou et al. 2023a, b, c). By creating new value chains, it enhances the competitiveness of businesses and strengthens the industrial base (Kleszcz 2021; Malorgio and Marangon 2021). The updated strategy underlines the need for EU organic businesses to diversify their product portfolios, business models and service offerings to create new high value-added products (European Commission 2018). Agriculture has a prominent place in this blueprint, as it is a crucial primary sector for biomass production. The bioeconomy paradigm is based on the principles of circularity and renewable biological resourcegols, such as plant and animal biomass (Papadopoulou et al. 2021). By embracing this model, we can provide innovative alternatives to address major socioeconomic and environmental challenges, such as increasing agricultural production to feed the expected global population of 10 billion by 2050, while reducing CO2 emissions and protecting the environment. Increased use of biomass and waste for energy could potentially reduce greenhouse gas emissions by 50% by 2050 (Kang et al. 2020). Notably, while the adoption and implementation of the bioeconomy has gained traction since 2015, the definition and approach are heterogeneous across countries pursuing different strategies (Devaney and Henchion 2018; Papadopoulou et al. 2023a, b, c).

The shift towards a bioeconomy framework has the potential to create new business opportunities in rural areas, especially in the vicinity of biorefineries that convert biomass into a range of biobased commodities such as food, feed, chemicals, bioenergy, biofuels, electricity, and heat (Fava et al. 2021). As low value feedstocks are expensive to transport, rural areas may have a comparative advantage that effectively counteracts any economies of scale associated with higher value-added processes (Ingrao et al. 2018). Similar to renewable energy, which is mostly produced in rural areas, there is no guarantee that the growth of the bioeconomy will lead to rural development due to several obstacles (Lainez et al. 2018). These challenges can include conflicting policies between countries or within the same country, uncertainty about environmental impacts, and a lack of attention to rural development concerns or objectives. In addition, in areas where fossil fuel-based economies are deeply rooted, investment and vested interests create significant dependencies that bioeconomy advocates need to address (Albrecht et al. 2021). However, in order to promote the development of the bioeconomy in rural areas and reap the benefits mentioned above, it is necessary to consider individual needs and implement appropriate measures (Leibensperger et al. 2021).

As mentioned above, there is a widespread belief in the potential benefits of the bioeconomy for achieving sustainable growth and competitiveness. However, some scientific publications have drawn attention to potential negative impacts associated with the bioeconomy. These include increased pressure on water resources and natural ecosystems, as well as doubts regarding its effectiveness in reducing emissions (Lago et al. 2019; Lazaridou et al. 2021; Priefer et al. 2017; Stegmann et al. 2020). These publications also raise concerns about competition for land, agricultural intensification, eutrophication, and the potential introduction of invasive species. While some argue that the bioeconomy has an inherent circularity, others caution against a linear business-as-usual approach that does not take into account circular economy principles. The concept of a circular bioeconomy was first introduced by the European Commission in 2015, but limited efforts have been made to comprehensively define and delineate its constituent elements (Kardung et al. 2021). The significance of the circular bioeconomy for agriculture served as another major motivation for this research. The circular bioeconomy, first and foremost, encourages sustainability in agricultural production by cutting down on wasteful and excessive resource use (Kalfas et al. 2023; Kalogiannidis et al. 2022a, b, c). Furthermore, it mitigates the environmental effects of agricultural practices and increases the adaptability of agricultural systems to climate change. All things considered, a key instrument for establishing sustainable production systems is the circular bioeconomy (Kalogiannidis et al. 2022a, b, c). Moreover, there is a lack of grassroots perspectives on the role of the circular bioeconomy within regional bioeconomy clusters, which play a pivotal role in the implementation of resource-efficient and circular bioeconomy strategies, such as integrated biorefineries and cascading use of biomass (Loizou et al. 2019; Paletto et al. 2022). As a result, the crucial contribution of regional clusters in driving the European bioeconomy is increasingly recognized (Cantner et al. 2018; Stegmann et al. 2020).

Greece is also making efforts to promote the bioeconomy, particularly in the Western Macedonia region. However, Greece lags behind other European countries in developing a bioeconomy strategy (Papadopoulou et al. 2022). It is important to emphasize that the Western Macedonia Region is undergoing a process of decarbonization and the bioeconomy can be a crucial factor in ensuring its economic sustainability, especially in the field of agriculture, which accounts for a significant part of the region's gross income (Papadopoulou et al. 2023a, b, c). The benefits of the bioeconomy in agriculture are manifold.

The integration of the bioeconomy into agriculture aims to optimize crop productivity in space and time through the implementation of advanced technologies and modernized production methods. Bioeconomy is built around specialized agrifood production systems that focus on either crop or livestock production with limited genetic diversity (Magrini et al. 2019). Agricultural intensification, characterized by heavy reliance on external inputs, especially energy and agrochemicals, has had negative environmental impacts such as soil degradation, decline in soil organic matter and quality, loss of agrobiodiversity, greenhouse gas emissions and nutrient losses that pollute waterways (Priefer et al. 2017; Trigkas et al. 2020).

Scientists agree that achieving agricultural sustainability will require significant changes to reconcile economic viability and social equity in food production with environmental goals (Bournaris et al. 2021; Movilla-Pateiro et al. 2021; Streimikis and Baležentis 2020). In the Mediterranean countries, highly specialized agricultural systems are predominantly based on intensive cereal-based cropping systems in monocultures or short rotations, such as wheat-summer irrigation or fallow, which result in high pest and disease incidence, loss of soil fertility and biodiversity (Vanino et al. 2022).

It is therefore imperative to introduce appropriate bioeconomic practices in crops and alternative management practices in typical intensive systems to promote the transformation of agricultural systems (Antar et al. 2021). This transition is a critical pathway to achieving the goal of securing resource availability through increased reliance on ecosystem services that minimize the use of external inputs and promote healthy agroecosystems (Matthews 2020).

Improving crop productivity and resource efficiency in arable cropping systems can be achieved by introducing temporal and spatial crop diversity that provides multiple ecosystem services through techniques such as crop rotation, integration of cover crops, green manuring and species mixes (Alem 2023; Rodriguez et al. 2021). Coupling agricultural diversification with bioeconomy practices such as conservation agriculture, organic farming and manure management can also lead to increased yields, profitability and resilience of farming systems in the long term (Sharma et al. 2021). However, the short-term economic costs of implementing bioeconomy strategies may outweigh their environmental benefits, necessitating the use of financial instruments to support combined bioeconomy approaches.

Although there is a general consensus on the potential agroecological and socioeconomic benefits of bioeconomy induced crop diversification, various technical, organizational and institutional barriers linked to dominant agrifood chains often hinder the adoption of crop diversification strategies (Kalogiannidis et al. 2022a, b, c; Mattas et al. 2022). New crops may face market exclusion or lack of technical knowledge and production skills, especially in the early stages of implementation (Borsellino et al. 2020). Farmers may also lack awareness of the benefits of crop rotation, the cost of machinery and new work organization, while market uncertainty remains high (Prestele et al. 2018). These and other forces for simplification encourage farmers to specialize in a few agricultural products in different regions (Kalogiannidis et al. 2023). Research and policy must therefore support agri-food production practices while ensuring environmental improvements.

To fill this gap, this study aims to delve into the perspectives of crop and livestock farmers regarding the advantages associated with implementing circular bioeconomy practices within their farming operations. Study area is heavily reliant on maintaining its agricultural endeavors by integrating aspects pertinent to this vital sector of the economy. Cluster analysis methodology was used to examine farmers' perceptions, with a questionnaire designed to explore their interest in circular bioeconomy practices and the barriers they face. The results were grouped into clusters with common beliefs and characteristics.

The paper is divided into five clearly defined sections. The second section covers the scope of the study and the methodology used. The third section presents the detailed results of the study. Future research and discussion are presented in the fourth section, while the fifth section highlights the limitations of the research and summarizes the conclusions drawn.

2 Methodology

2.1 Study area

Western Macedonia, as illustrated in Fig. 1, represents a topographically diverse region situated in the northern sector of Greece, sharing borders with Albania and North Macedonia (Western Macedonia Region 2021). This area boasts a captivating and idyllic natural panorama, adorned with a multitude of national parks and resplendent lakes. The economic foundation of Western Macedonia predominantly rests upon the pillars of agriculture, forestry, and mining, with a particular focus on the cultivation of tobacco, viticulture, and fruit production. Furthermore, the region is host to numerous hydroelectric power stations, which serve as the primary sources of electricity for the entire nation of Greece (Belke et al. 2019).

Fig. 1
figure 1

Regional units of the western Macedonia region Source: Author’s own creation

The region of Western Macedonia is divided into four distinct regional units, namely Kozani, Grevena, Kastoria and Florina, each of which is characterized by a unique economic profile, with different industries playing a central role in their respective economies (Kalogiannidis et al. 2022a, b, c). Kozani is known for its lignite mines, which have traditionally been a significant source of employment and income for the region. In addition, the region is home to several power plants that rely on lignite as their primary fuel source. Recently, efforts have been made to decarbonize the region and move towards cleaner forms of energy (Papadopoulou et al. 2023a, b, c). The Greek government has set a target to phase out the use of lignite by 2028, and several initiatives have been launched to promote renewable energy sources and improve energy efficiency in Western Macedonia. In addition, the region's agricultural sector is experiencing an upturn, with cereal, fruit and vegetable crops gaining in importance. Grevena, on the other hand, is characterized by its agricultural production, with crops such as wheat, maize and vegetables being the main agricultural products. Livestock farming, particularly sheep and goats, is also a notable aspect of Grevena's economic landscape. Kastoria, on the other hand, is renowned for its fur industry, which has historically been a major source of income for the local population. The region is also home to several textile and leather processing factories. In addition, Kastoria's tourism sector revolves around its picturesque lakes, historical sites and traditional architecture. Similarly, Florina is recognized for its agricultural productivity, with apple growing being of paramount importance (Kalfas et al. 2022). The region is also home to several food processing industries, including dairy and wine production. Florina's tourism industry is also noteworthy, attracting numerous visitors attracted by its scenic landscapes and ski resorts (Western Macedonia Region 2021).

2.2 Procedures and measurements

The aim of this study is to investigate the level of knowledge of farmers in Western Macedonia about the circular bioeconomy, the practices they follow and the barriers to adoption. To achieve this goal, k-means clustering was selected as a popular un-supervised machine learning algorithm used to cluster or group similar data points in a dataset (Bernhardt et al. 1996; Ghisellini et al. 2016; Papadopoulou et al. 2023a, b, c). For this reason, the study of its properties is of interest not only to the classification, data mining and machine learning communities, but also to the growing number of practitioners in marketing, research, bioinformatics, customer management, engineering and other application areas (Awasthi et al. 2021).

The k-means algorithm is one of the simplest unsupervised learning algorithms for solving the well-known clustering problem. The goal of k-means clustering is to partition a data set into k clusters, where each data point belongs to the cluster whose mean is closest to it. The algorithm works by iteratively assigning data points to their nearest cluster center and updating the cluster centers to be the mean of the assigned data points (Jain 2010; Reiff et al. 2018). This process continues until the cluster assignments stop changing or a maximum number of iterations is reached.

The basic steps of the k-means clustering algorithm:

  1. 1.

    Choose the number of clusters, k.

  2. 2.

    Initialize the k cluster centers randomly.

  3. 3.

    Assign each data point to the nearest cluster centers.

  4. 4

    Update the cluster centers to be the mean of the assigned data points

Repeat steps 3–4 until the cluster assignments no longer change or a maximum number of iterations is reached.

The Statistical Package for Social Sciences (SPSS) was used in this study to carry out the various statistical procedures.

The empirical research of this study is based on a quantitative survey dataset that collected responses from 412 farmers and livestock keepers in the Western Macedonia region. The following results provide an overview of the farmers, as the sample is small and slightly biased towards farmers who could be easily reached after a field survey, meaning that the results are indicative and cannot be directly generalized to the entire core population. This sampling method is called convenience sampling. Convenience sampling is a non-probability sampling method where subjects are selected based on their easy availability and accessibility to the researcher. Instead of employing random selection techniques, convenience sampling relies on choosing participants who are conveniently located or easily reachable (Etikan 2016; Rahi 2017). In 2022, when the survey was conducted, there were a total of 23,089 mixed, pure crop and pure livestock holdings in the region. Over ten years, the total number of agricultural holdings in the region of Western Macedonia de-creased by 768, while the size of holdings increased significantly, as reported by the Hellenic Statistical Authority.

The questionnaire design process began with a thorough review of previous studies on farm management and farmers' willingness to adopt new agricultural and live-stock practices. Researchers also reviewed surveys that examined farmers' knowledge and acceptance of the bioeconomy. Particular attention was paid to identifying factors that inhibit adoption of the bioeconomy. To ensure the accuracy and clarity of the questions (Forbes et al. 2018), the survey underwent several revisions and rounds of feedback from experts in agricultural economics, bioeconomy, regional development, agronomists and farmers. The final version of the questionnaire consisted of questions about the demographic characteristics of the participants, information about their farms and their attitudes towards the circular bioeconomy. The third section of the questionnaire (attitudes towards the circular bioeconomy) used a five-point Likert scale (1 = strongly dis-agree, 5 = strongly agree) (Joshi et al. 2015). The data collected facilitated the creation of participant groups based on their perceptions and provided insight into their profiles.

The researchers carried out fieldwork in the regional units of the Western Macedonia region for seven months, from February to August. They first confirmed that the participants were over 18 years old and informed them about the protection of their personal data before giving them the questionnaires to complete. The researchers were present to supervise the process and address any issues that arose. Out of a total of 420 completed questionnaires, 412 were considered suitable for inclusion in the study. Cluster analysis requires a minimum sample size of at least 20–30 participants for each cluster. The sample meets this requirement and is therefore considered satisfactory (MacKenzie and Podsakoff 2012). In general, a larger sample size is preferred to improve the accuracy and robustness of the results.

Between 15 and 20 January 2022, a pretest was carried out with 15 farmers to estimate, among other things, the time it would take to complete the questionnaire (possible ambiguities, difficulty in understanding questions) (Elias and Sreejesh 2022). It was found that it would take approximately 15 min to complete the questionnaire.

Prior to the cluster analysis, univariate statistics were calculated and one-way ANOVA tests were used to determine the differences between 7 socioeconomic demographic variables in relation to the 19 variables measuring the circular bioeconomy practices already adopted by farmers and the inhibiting factors for non-adoption.

The suitability of both the sample size and the number of variables for cluster analysis in this study was confirmed by a previous evaluation using Kayser–Meyer–Olkin statistics, which yielded a value of 0.648, exceeding the recommended threshold (Vareiro et al. 2013). If the KMO value is less than 0.5, it indicates that the sampling is not adequate and remedial action should be taken.

3 Results

3.1 Respondents’ profiles and general data

The demographic and economic characteristics of the farmers who took part in the survey are shown in Table 1.

Table 1 Frequencies of socio-economic characteristics of farmers

The survey included 364 male farmers (88.3%) and 48 female farmers (11.7%), which is to be expected given that agriculture is primarily a labour-intensive sector and typically male-dominated in rural areas. In terms of age, the largest proportion of respondents (38.3%) were in the 46–55 age group, with a significant number of farmers in the 36–45 (25.7%) and 56–65 (19.2%) age groups. The survey showed a lack of farmers aged 18–35 in the Western Macedonia region, which is consistent with regional census results indicating a decline in the population due to young people leaving for better job opportunities and living conditions in urban areas. More than 50% of the farmers in the survey are high school graduates, while 26.9% have only completed secondary school. University graduates and those with a Master's degree were represented in smaller numbers. Marital status was divided into three categories: 77.9% of the farmers were married, 11.7% were single and 4.9% were divorced. The majority of farmers (59%) live in villages, 31.6% in communes and only 9.5% in towns. In terms of income, 25.2% of respondents said they earned between €20,001 and €30,000, which is the average income for Greeks. Thirty per cent of farmers said they earned more than this amount, while 32% said they earned less than €20,000. The survey also revealed that 66% of farmers had been involved in agriculture for more than 15 years, while 34% had been involved for less than 15 years. This result is influenced by the age of the farmers interviewed. Finally, 254 farmers are engaged exclusively in pure crop farming, 15 farmers are engaged exclusively in animal husbandry and 143 farmers have mixed holdings.

Before presenting the results of the cluster analysis, it is important to highlight some details about the farmers and their familiarity with the circular bioeconomy. The farmers were asked to rate their familiarity with the concept of the circular bioeconomy on a five-point Likert scale, and 77.9% of them responded with agreement or complete agreement (Table 2). This optimistic response is in line with similar research conducted in Portugal and Germany (Berki-Kiss and Menrad 2022; Brandão and Santos 2022; Dallendörfer et al. 2022).

Table 2 Bioeconomy practices and disincentives

When asked if they support the implementation of the circular bioeconomy in agriculture, there is a 16.7% drop in the number of farmers who agree. It seems that although they are aware of the circular bioeconomy and its opportunities for the environment and their farms, they are not ready to change from their current production model to a bioeconomic one. Farmers expect others to make the transition, and once the resulting benefits, such as increased productivity and financial support, are confirmed, they will follow suit (Aare et al. 2021).

The results for the two variables mentioned above and 17 other variables related to circular bioeconomy practices that farmers implement on their farms and the barriers they face in adopting these practices are presented in Table 2. The mean and standard deviation based on farmers' preferences on a 5-point Likert scale, where 1 = strongly disagree to 5 = strongly agree, are also shown.

One-way ANOVA revealed notable differences between demographic characteristics and specific variables related to bioeconomy practices and barriers. Important conclusions were then drawn from these analyses.

According to Table 3, female farmers are more open to implementing the circular bioeconomy in agriculture than their male counterparts. In addition, women are more likely than men to adopt practices such as the rational use of pesticides, the use of indigenous plants and the burning of crop residues. They also recognize that the circular bioeconomy requires a high level of education. In contrast, male farmers prioritize the use of post-harvest pesticides more than women, and they see the circular bioeconomy as a temporary trend that will lose its potential.

Table 3 Mean scores and ANOVA tests by gender.

Subsequently, the survey results show that there are notable differences in the circular bioeconomy practices adopted, as well as in the age range of the farmers involved (Table 4). Younger farmers seem to have a better understanding of circular bioeconomy concepts, while older farmers have a longer history of implementing such practices, which are in fact traditional farming techniques that have been used for decades. Both age groups agree that farmers need financial support and education on circular bioeconomy issues.

Table 4 Mean scores and ANOVA tests for age.

Based on Table 5, there seems to be a consensus between primary school graduates and university graduates regarding circular bioeconomy practices and barriers to their implementation. This could be due to the fact that primary school graduates are likely to be older (as secondary education was not compulsory in Greece before 1989) and have a history of following circular bioeconomy practices. Conversely, university graduates have been exposed to good practices and recognize the benefits and prospects that such practices offer. In addition, high school graduates share similar views among themselves.

Table 5 Mean scores and ANOVA tests for education level.

No statistical differences were found between marital status and attitudes towards the circular bioeconomy. On the other hand, the place of residence is statistically different (Table 6). It can be observed that farmers living in villages may be less familiar with circular bioeconomy practices, but they show a greater willingness to implement them in their agricultural activities. In contrast, urban farmers seem to follow circular bioeconomy practices to a greater extent. However, there is a higher level of agreement and more consistent responses among livestock farmers, regardless of where they live. In addition, farmers living in villages are more likely to believe that the circular bioeconomy is a passing trend than their urban counterparts.

Table 6 Mean scores and ANOVA tests for residence.

Farmers tend to be more accepting of the circular bioeconomy and its practices when their income is higher. However, they do not support its mandatory implementation in agriculture and believe that the government should provide them with support and information (Table 7).

Table 7 Mean scores and ANOVA tests for annual income.

Farmers who have been involved in farming for a longer period of time are generally more receptive to circular bioeconomy practices than those with less experience, who appear to be less inclined to adopt these practices on their farms. It is generally recognized that information and training on the circular bioeconomy is essential for agricultural producers, and it is believed that the circular bioeconomy is a trend that will not become more widespread in the future (Table 8).

Table 8 Mean scores and ANOVA tests for years of involvement in agriculture-livestock farming.

ANOVA tests were not considered appropriate for comparing circular bioeconomy practices applied to different farm types. This is because farmers are assumed to follow specific practices depending on their farm type. For example, it is assumed that livestock farmers do not rotate crops, or that crop farmers do not graze their cattle. Consequently, the statistical results would not provide any new insights into the survey.

3.2 Farmers’ perceptions among clusters

A non-hierarchical k-means clustering method, designed to classify cases rather than variables, was used in line with previous studies (Aldino et al. 2021; Guevara-Viejó et al. 2021; Nagari and Inayati 2020). This method is more effective on larger datasets (n > 200) than the hierarchical technique. However, it requires a predetermined number of groups to be created (Niu et al. 2021).

Following the method used by Violán et al., (2018) and Murray and Grubesic (2013), a stepwise approach was used to create 2–5 groups based on the average score of 17 items measuring farmers' perceptions of circular bioeconomy practices and barriers to their adoption. Table 9 shows the distribution of sample percentages for each group and the different groupings (from two to five groups). The data show that the selection of four or five groups results in a minority group representing less than 10% of the total sample. In order to simplify the results and make them more understandable, it was decided to limit the clusters to three.

Table 9 Percentage sampled within each group.

To test the robustness of the clustering results obtained from the non-hierarchical k-means cluster analysis, an alternative clustering method was used. This involved the utilization of a two-stage cluster analysis (TSCA) that not only automatically determined the optimal number of clusters but also employed the same predictors as the non-hierarchical k-means approach. TSCA was chosen due to the qualitative nature of the variables derived from the research instrument (Anastasios et al. 2010; Loizou et al. 2013; Michailidis et al. 2011). The TSCA analysis unveiled an optimal solution comprising four clusters, which displayed similar sizes and characteristics to those of the non-hierarchical k-means cluster analysis (Papakosta et al. 2018). Specifically, the first, second, third, and fourth clusters encompassed 24%, 8%, 32%, and 11% of the sample participants, respectively. Additionally, the TSCA analysis revealed that the unclassified members from the fifth cluster in the non-hierarchical k-means analysis exhibited behavior akin to 15% of the sample population. The robustness of the non-hierarchical k-means cluster analysis was affirmed by the TSCA results, thereby warranting acceptance of the initial clustering outcome, which was subsequently integrated into the ensuing multivariate statistical methodology.

The study of the different groups included an analysis of the mean values for 17 items related to practices associated with the implementation of the bioeconomy and barriers to its adoption (Table 10), which helped to determine the degree of agreement or disagreement among the farmers regarding these figures for each group. Table 10 shows that most of the impacts contributed significantly to the identification of clusters. The issue of “lack of financial motivation” and “government support needed” appears to be a point of agreement among farmers and has been previously identified by other researchers, such as Lokhorst et al., (2011).

Table 10 Perceptions about bioeconomy practices and disincentives among clusters (percentage agreeing a and average scores b).

The demographic and social characteristics of the farmers belonging to the three groups are presented in Table 11, which shows the results derived from the cluster analysis.

Table 11 The demographic profile of the clusters of farmers.

4 Discussion

Based on the obtained results, the three clusters that were maintained can be described as the modernists, early adopters and latecomers (Dedehayir et al. 2017).

Cluster 1—Modernists: This cluster of respondents (30% of the total sample) seems to strongly agree with the rational and sustainable practices related to the circular bioeconomy in agriculture. The highest levels of agreement are related to the rational use of pesticides (92.6%) and reduced use of feed supplements (91.0%), closely followed by grazing livestock on pastures (91.8%) and increased feed storage (94.3%). However, there are some practices with lower levels of agreement, such as the use of indigenous plants (17.2%) and the burning of stubble, branches and crop residues (14.8%). Penalties on the financial aid applied for by farmers shall be imposed where it is found that the crop residues has been burnt without prior authorisation. This suggests that there may be some cultural or practical barriers to adopting these practices. In terms of barriers to adopting a circular bioeconomy, lack of financial motivation (88.5%) and lack of information and training (91.8%) are the main concerns, while the high level of education required (25.4%) and the belief that the circular bioeconomy is a fading trend (38.5%) are perceived as less relevant. Overall, this cluster ap-pears to be well informed and committed to sustainable agricultural practices and highlights the importance of financial incentives and support from government to drive the adoption of circular bioeconomy practices in agriculture.

Modernists are predominantly male (91.0%) and the majority are aged between 36 and 55 (75.4%). In terms of education, most respondents have completed secondary school (41.8%) or high school (37.7%), while a smaller proportion have completed university (12.3%). The majority of respondents live in a town (59.0%), while a smaller proportion live in a village (32.0%) or city (9.0%). In terms of annual household in-come, a quarter of respondents (26.2%) earn more than €40,000 per year, while a simi-lar proportion earn between €20,001 and €30,000 per year (22.1%). Most respondents have more than 20 years of farming experience (49.2%), followed by 16–20 years (30.3%), while the remaining respondents have less than 16 years of experience. The type of holding is predominantly mixed (95.1%), with only a small proportion having a pure crop holding (4.9%) and none having a pure livestock holding. Overall, this cluster seems to be comprised of experienced and well-established farmers, with a mix of educational backgrounds and income levels.

Cluster 2—Early adopters: This cluster has a larger sample size (n = 212) compared to Cluster 1 (n = 122) and is composed of individuals who generally agree with rational management practices in a circular bioeconomy, with high percentages agreeing with the rational management of pesticides (85.8%), appropriate irrigation systems (84.9%), and rational use of plant protection products (86.3%). However, there is a lower percentage of agreement with practices such as crop rotation (39.2%), use of native plants (12.7%), and livestock grazing on pastures (2.4%). In terms of barriers to adopting a circular bioeconomy, there is high agreement that lack of financial motivation (89.6%) and government support (96.7%) are major issues. There is also a lower level of agreement that a high level of education is required (19.3%), and that the circular bioeconomy is a fading trend (40.1%).

On the basis of the information provided, it can be seen that Cluster 1 has a higher proportion of female farmers than Cluster 2 (9% vs. 12.8%). In addition, Cluster 1 has a higher proportion of farmers with a lower level of education and more farming experience. With regard to the annual income of the household, cluster 1 has a more even distribution, while cluster 2 has a higher proportion of farmers in the middle of the in-come range. Concerning the type of holding, Cluster 2 has a higher proportion of arable farmers, while Cluster 1 has a higher proportion of mixed holdings.

Cluster 3—Latecomers: Based on the data provided, Cluster 3 represents 78 respondents or 19% of the total sample. This group appears to have lower levels of agreement with circular bioeconomy practices compared to the other clusters. They have relatively low average scores across most of the circular bioeconomy practices, with the highest score for livestock grazing on pastures at 24.4%. They also identified lack of financial motivation (76.9%) and government support needed (82.1%) as the primary barriers to adopting circular bioeconomy practices. In terms of demographics, the majority of respondents in this cluster are male (93.6%) and have farming experience of more than 20 years (41%). Most of them are also engaged in pure crop farming (94.9%).

Cluster 3 has the smallest number of farmers (19% of the total sample) and is characterized by a lower level of adoption of circular bioeconomy practices, with relatively low agreement scores across all practices. Farmers in this cluster identified lack of financial motivation and the need for government support as major barriers to adopting circular bioeconomy practices. This cluster is mainly composed of male farmers (87%), with a slightly older age distribution (36–55 years old) and lower levels of education (29% with high school education or below). Most farmers in this cluster reside in villages (83%), have an annual household income of 10,000–20,000 € (61%), and have between 11 and 20 years of farming experience (38%). The distribution of type of holding is relatively balanced across pure crop (56%), pure livestock (19%), and mixed holdings (24%).

The concept of the circular bioeconomy has recently attracted considerable interest for its ability to provide sustainable solutions to food production and waste management. It revolves around the use of biological resources, including crops and livestock, in a circular approach. This approach involves the production, use and subsequent reuse or recycling of these resources (Fraga-Corral et al. 2022; Sefeedpari et al. 2020; Venkata Mohan et al. 2019). The concept of the circular bioeconomy has gained significant attention in recent years for its potential to provide sustainable solutions for food production and waste management. Ranjbari et al. (2022) and Vea et al., (2018) have particularly highlighted the potential benefits of the circular bioeconomy. In particular, it can provide economic benefits to farmers and businesses by creating new revenue streams through the valorization of waste and by-products (Donner et al. 2022; Donner and de Vries 2021).

Farmers' perspectives on the circular bioeconomy can differ based on various factors, including their awareness levels, economic incentives, and the extent of circular bioeconomy implementation in their respective sectors (Awasthi et al. 2019; Khan and Ali 2022; Papadopoulou et al. 2023a, b, c). Some crop and livestock farmers view the circular bioeconomy as an opportunity to minimize waste and enhance resource efficiency (Kurnaz et al. 2022; Stegmann et al. 2020). They recognize the benefits of utilizing organic waste as manure or biofuel feedstock, which can improve soil health and reduce reliance on synthetic fertilizers (Antar et al. 2021). Livestock farmers, in particular, see the circular bioeconomy as a means to lower feed costs and enhance the environmental sustainability of their operations by utilizing manure as both fertilizer and an energy source (Chodkowska-Miszczuk et al. 2021; Liu et al. 2022).

However, some farmers may perceive the circular bioeconomy as a potential burden due to the expenses and efforts required to adopt new practices and technologies (Salvador et al. 2022). Concerns may also arise regarding the economic viability of circular bioeconomy practices, especially in the short term (Bottausci et al. 2022). Implementing circular bioeconomy practices can pose challenges, including the need for infrastructure and technological advancements. These challenges involve grasping the concept, accessing economic incentives, and assessing the feasibility of implementing circular practices within the sector.

In an attempt to explore farmers' perceptions of circular bioeconomy practices and identify ways to motivate and overcome barriers to implementation, this study focused on surveying farmers in the Western Macedonia region. Encouraging farmers to adopt circular bioeconomy practices may require training and awareness initiatives. These efforts could involve providing training sessions on circular bioeconomy practices, highlighting the benefits of sustainable agriculture, and showcasing successful case studies of farmers who have already embraced these practices. Governmental support, such as disseminating the latest research, promoting best practices, and offering financial incentives, can also play a crucial role in facilitating the adoption of circular bioeconomy practices among farmers.

A national or regional bioeconomy strategy can play a crucial role in promoting circular bioeconomy practices in agriculture. Such strategies can provide a framework for research and development, financing and investment, and policy and regulatory initiatives that can support the adoption of circular bioeconomy practices. National and regional strategies can also help to create a more supportive environment for circular bioeconomy initiatives by encouraging collaboration between different stakeholders, such as farmers, businesses and researchers.

5 Conclusions

Despite numerous international empirical studies on farmers' perceptions of the circular bioeconomy, the issue has received limited attention from Greek researchers. Moreover, none of the existing studies have explored the differentiation of farmers' perceptions based on their socio-economic characteristics or the geomorphological features of their residential areas. This study aimed to address this gap. Additionally, there are current agricultural subsidies linked to the implementation of the bioeconomy and associated practices, including renewable energy production.

Agriculture holds significant importance in the region's transition, being the second largest productive sector after energy. There are several reasons why adopting circular bioeconomy practices is crucial. Firstly, the agricultural sector is evolving with the introduction of new technologies and practices that require farmers to stay informed about the latest developments in order to benefit from them. Secondly, many farmers lack fundamental knowledge of agricultural practices, leading to lower productivity and reduced incomes. Thirdly, farmers face challenges such as limited water resources, increasing energy costs, and unpredictable climate conditions, necessitating a comprehensive understanding of modern management practices derived from the circular bioeconomy.

Incorporating cluster analysis can offer a targeted strategy for organizing bioeconomy training and awareness programs by categorizing farmers into distinct groups based on their shared perspectives on the bioeconomy's prospects. Research results indicate that farmers in the Western Macedonia region, despite differing views on the perception of the bioeconomy, demonstrate significant support for financial incentives associated with the bioeconomy. Given the nascent nature of the bioeconomy, it is reasonable to expect varying opinions among farmers regarding the benefits it can offer.

The present study has certain limitations, including its narrow focus on farmers from Western Macedonia. Although the survey's inclusion criteria were clearly defined, potential researcher bias in selecting farmers cannot be entirely eliminated. Additionally, the study assumes that the government will adopt a strategy to strengthen the bioeconomy and involve farmers as key stakeholders in this initiative. As farmers form an integral part of the bioeconomy, their inclusion in such a strategy is expected.