Sustainability and Brazilian Agricultural Production: A Bibliometric Analysis
[ad_1]
3.1. Descriptive Analysis of the Selected Sample
Following the methodology, the study identified 3139 documents, comprising 21,380 authors distributed among all documents (Figure 1b). The annual publication rate of documents was ~143, presenting a yearly growth rate of ~17%. The average age of documents is ~6 years, indicating that most documents were published in the last 6 years. It is noted that interest in the topic studied gained prominence from 2007 onwards, with exponential growth since then, with its peak being observed in 2021 (n = 546 publications), representing ~17% of the sample.
In this study, it is possible to suggest the growth of interest in research related to the topic following the boom in publications in 2007, since ~97% of documents were produced between 2007 and 2022. The growth of publications correlating sustainability and agricultural production in this period is corroborated by the progression of research production related to sustainable agriculture found by Sarkar et al. [9]. For Sarkar et al. [9], such growth may result from the prominence of sustainability in contemporary agro-economic perspectives. For Yu and Mu [16], sustainable development has been elevated to a national strategy and has migrated from the theoretical field of discussion to the practical field of implementation. As a result, its implementation and promotion of strategies for sustainable development have directly impacted the number of publications year after year. As an example, using the study by Abafe et al. [22], the author observed that, in recent years, aspects that correlate with the sustainable use of water in agriculture have become a severe concern of society, causing more research and strategies to be implemented for the rational use of the resource.
Consequently, it is possible to suggest that interest in more general aspects of sustainability and agricultural production follows the same trend, including in Brazil. In turn, the increase in interest positively impacts the creation of global research fields of interest to scientists and affected communities, which increases the number of publications on the topic. Furthermore, it is possible to observe the alignment of interest in the subject of study between Brazilian and foreign researchers since the interest boom occurs at similar times.
3.2. Identification of Institutions, Co-Authorships, and Collaborations
The results indicate that the University of São Paulo (USP, n = 1043), São Paulo State University of São Paulo (UNESP, n = 670), Federal University of Rio Grande do Sul (UFRGS, n = 349), and the Federal University of Viçosa (UFV, n = 262) have the highest numerical production of studies (Figure 2a). The four institutions represent ~67% of all document publications sampled. It is also possible to highlight the centrality of USP as well as its collaboration networks between national institutions (e.g., Universidade Estadual Paulista and Universidade Federal de Minas Gerais) and international (e.g., Ghent University, Belgium; and Universidad Nacional Autónoma de México, Mexico) (Figure 2b).
The prominence of the institutions listed in this study can be attributed to two factors: (i) the number of research programs with high-quality scores and (ii) the available financial support. Regarding the scores ranging, the research programs associated with Brazilian institutions periodically undergo governmental quality evaluations. In these evaluations, five criteria are considered: program proposal, teaching staff, student body, theses and dissertations, intellectual production, and social insertion. Ultimately, these programs receive scores ranging from 1 to 7. A score of 5 indicates that the course is considered ‘very good’, while scores of 6 and 7 are awarded to programs of excellence at the national and international levels, respectively. Among the 190 Brazilian research programs linked to the agricultural sciences field with scores between 5 and 7, approximately 57% (108 research programs) are affiliated with the top 10 institutions. Within this group, about 12% (23 research programs) are associated with USP, which boasts the highest number of research programs with international excellence (seven research programs). Most of these programs have over 50 years of experience, suggesting a high level of maturity (these data were obtained using public consultation in the Coordination of Superior Level Staff Improvement (Capes). Available at: https://sucupira-v2.capes.gov.br/sucupira4/programas?grande-area-conhecimento=5&search=&size=100&page=4 (accessed on 25 January 2024).
On the other hand, the financial support is closely related to these scores and their maintenance, for instance, specific financial resource programs to support academic excellence programs (Programa de Excelência Acadêmica (Proex/Capes/MEC)). According to the National Council for Scientific and Technological Development database (CNPq) (these data were obtained using public consultation to the website of CNPq Data Transparency (https://portaldatransparencia.gov.br/ (accessed on 25 January 2024)), other crucial federal research support agencies, and approximately 1350 research institutions received financial backing for research development between 2002 and 2022 (including research grants, financial project support, etc.). The top 10 institutions garnered 48% (BRL 1.7 billion) of the total financial resources available. UFV, USP, UNESP, and UFRGS collectively received 26% (BRL 941.5 million) of the overall financial resources. Notably, this figure pertains specifically to the financial support allocated for the agricultural sciences field. Thus, it is possible to suggest that the prominence of those institutions has been related to a higher number of research programs that present high-quality score ranging.
Figure 3 presents the authors’ co-occurrence network analysis. For Velasco-Muñoz et al. [13], the co-occurrence network analysis explores the social networks researchers form when collaborating on publications. In this way, connections between authors are formed when they produce a document jointly. Consequently, collaboration between authors can form a cooperative relationship between institutions [16].
It is observed that four researcher groups presented a collaboration network with more than four researchers. Among them, the groups formed by authors from USP (a) in purple), UNESP (b), UFRGS (c), and Embrapa (d) stand out. It can be observed that, although collaboration between researchers is more internally concentrated, there are collaborations between institutions, identified based on the flows in gray lines that depart from one group to another. It is worth mentioning that flows with greater thickness represent a greater cooperation density between authors. This finding indicates that, jointly, USP, UFRGS, and UNESP, can substantially influence the direction of future studies related to sustainability in agriculture based on multi-collaborative research, both between these institutions and among the others listed, promoting specialization, division of work [32], and promoting the dissemination of local studies to the scientific community as a whole.
3.3. Authors
The analysis of the authors’ activity is demonstrated in Figure 4. The figures comprised the 20 authors with the highest numerical productivity over time based on the h-, g-, and m-indexes.
Starting from the number of documents produced per author, the results suggest Faccio Carvalho, P. C.; Cherubin, M. R.; Cerri, C. E. P; Crusciol C. A. C.; and Cerri, C. C. as the five most numerically productive authors. Faccio Carvalho, P. C., has been a professor at the Department of Forage Plants and Agrometeorology at UFRGS since 1997, focusing on research and extension in the soil–plant–animal relationship. His studies encompass natural environments, cultivated lands, and integrated pasture grazing [33]. Cherubin, M. R., is a professor at the Department of Soil Science at ESALQ-USP. His studies center around quantifying and comprehending the impacts of land use, emphasizing the management and health of soil practices, carbon dynamics, and the provision of ecosystem services in both natural and agricultural ecosystems [33]. Cerri, C. E. P., is a professor at the Department of Soil Science at ESALQ-USP, with a focus on themes related to soil organic matter, global warming, climate changes, agriculture, and carbon credits, mathematical models, geostatistics, and geoprocessing [33]. Cerri, C. C., has been a professor at the Center of Nuclear Energy in Agriculture (CENA/USP) at USP since 1975. He specializes in researching soil organic matter dynamics in natural ecosystems and those modified by agricultural practices, livestock, and reforestation. His studies address greenhouse gas emissions, climate changes, and land-use changes associated with agriculture and livestock production practices [33]. Crusciol, C. A. C., is a professor at the Faculty of Agricultural Science at UNESP (FCA/UNESP). His expertise is concentrated on agricultural production systems, soil fertility management, plant nutrition, and applied vegetal physiology [33].
Faccio Carvalho, P. C.; Cherubin, M. R.; and Cerri, C. E. P. were responsible for ~2% of the total number of published studies (Figure 4a), indicating heterogeneity in document production. The five authors were responsible for publishing ~8% of the total sample. It is worth highlighting that both the production and the number of citations of Faccio Carvalho, P. C. and Cherubin, M. R. showed growth after 2013 (Figure 4b). However, although Faccio Carvalho, P. C. has presented studies since 2008, authors such as Bayer, C.; Cerri, C. E. P.; Cerri, C. C., and Hungary, M. shown oldest contributions, with publications before 2007. Like Faccio Carvalho, P. C., these authors presented a long sequence of contributions relating to sustainability and agricultural production. On the other hand, the results pointed to the concentration of publications and citations from 2015 onwards, which leads to the inference of an increase in the number of eminent researchers dedicated to studying sustainability in agriculture. With this, it is possible to suggest the importance and continuity of the area of knowledge.
According to the h-index, Faccio Carvalho, P. C. and Cherubin, M. R. present the highest value for the studied sample (n = 24), followed by Cerri, C. C. and Cerri, C. E. P., both with 22 (Figure 5a). The h-index is an index that attempts to level authors by calibrating their production to the impact of citations of their publications. The index comprises a set of the researcher’s most frequently referenced publications and how often those works were cited in other studies [22]. Considering the Faccio Carvalho, P. C. and Cherubin, M. R. results, although the h-index tends to increase with career time, the difference in the authors’ document production longevity was not a sufficient weight to distinguish them [34].
Notably, when considering the g-index, Faccio Carvalho, P. C. presents the best result (n = 46), followed by Cerri, C. E. P. (n = 38), with Cherubin, M. R. being third (Figure 5b). The g-index is a variant of the h-index. It seeks to emphasize densely cited studies, providing a greater impact of this study on the sample selected based on its citations within the time window. In the g-index, longer-lived authors present better results due to their longer careers, which may increase the number of citations and the relevance of their publications [35].
On the other hand, considering the m-index, Cherubin, M. R. presents the highest result (2.67), suggesting that this is a prosperous author in the study area (Figure 5c). The m-index is another variant of the h-index. Using years as a weight, this index considers the volume of works produced yearly since the author’s first publication and compared to authors such as Faccio Carvalho, P. C., Cerri, C. E. P., and Cerri, C. C., Cherubin, M. R. had a shorter contribution time, which infers a shorter career time in research development and can make the comparison among the authors in the sample unfair. However, since the m-index provides a fairer comparison between authors with different career times [36], it is possible to support the suggestion that Cherubin, M. R. is one of the most promising authors in the study area. Thus, authors with good numerical productivity at the beginning of their careers tend to present better results for the m-index.
It is important to highlight that the list of authors presented in the study was necessary solely and exclusively to present the methodology used to map the authors. Such results are restricted to the search methods and, consequently, to the sample in the database consulted. Furthermore, the intention of the present study was not to rank researchers as better or worse since each author, even those not listed, knows different areas of interest but of equal need, which could contribute to greater sustainability in agricultural production. In this sense, the main contribution of the methodology was to list potential promising authors for forming public policies that aim to propose solutions to the different problems that permeate the agricultural sector “from cradle to cradle”. With this, authors with longer and shorter careers could work in a multi-collaborative way to solve problems, combining practical experience accumulated in years of research with the most recently produced knowledge.
3.4. Citation Assessment
Regarding the citations, there are 122,759 citations and an average of 21.32 citations/document. Table 1 shows the top 10 most globally cited documents, with Lehmann et al. [37], Kattge et al. [38], and Poorter et al. [39] being the most cited documents. The top 10 most cited documents account for 5% of all citations, being published between 2003 and 2020. The document of Lehmann et al. [37] was the most prestigious among the others, presenting twice citations of the second and third most cited document. It is important to highlight that this study had the collaboration of Embrapa Amazônia Central. In general, Lehmann et al. [37] aimed: (a) to compare the fertility of an archaeological Anthrosol (i.e., Amazonian dark earth soils) to a typical upland Ferralsol of the central Amazon basin; (b) to assess how these Anthrosols maintain their high nutrient availability; and finally (c) to compare the effect of inorganic and organic amendments such as charcoal and manure on fertility and nutrient retention of a Ferralsol with those of an archaeological Anthrosol. For example, this document was used to highlight different features from anthropogenically made soils [40,41]; the application of some features observed in anthropogenically made soils as technologies (i.e., biochar fertilization) to the soil quality improvement (i.e., soil pH regulation and nutrient concentration improvement [42,43,44], water retention [45], etc.) and as potential tools for sustainable agriculture [44,46]. On the other hand, when considering the number of citations per year, the document of Kattge et al. [38] showed a higher number of citations. In general, this document introduces the latest version of a plant trait database. It discusses the importance of plant traits in understanding the functioning of ecosystems and the potential applications of the database in various research fields. This database provides a worldwide collection of plant trait data. Studies cited in this document highlight the initiative in data sharing and synthesis of plant trait data [47,48,49].
Considering the top 10 most locally cited documents (Table 2), De Moraes et al. [57] was the most influential document, showing more than twice the citations of the second and third most locally cited documents. This study presents the benefits of integrated crop–livestock systems for more sustainable agriculture production. Regarding soil–plant–animal aspects, the study highlights the greater environmental gains and less vulnerability, higher yields, and more financial gain of integrated crop–livestock systems compared to monocultures or non-integrated livestock farming. It is important to highlight the local citations, considering the number of times an author (or a piece of a document) in the sample has been referenced by other authors who are also authors in the studied sample [22].
Meanwhile, in terms of most locally cited references (Figure 6), De Moraes et al. [22] was the third most cited, after Alvares et al. [66] and R Core Team, the first and second most cited references, respectively. These references were generally cited as tool presentations used in experimental conditions in the documents to highlight statistical procedures and describe experimental conditions.
In summary, the sampled literature was mainly composed of experimental research interested in yield and production improvement (i.e., the use of biochar in vegetal production) and long-term studies perspectives interested in animal–vegetal integration. Also, the number of citations was generally high and provided different perspectives to explore sustainability in agriculture and livestock production systems.
3.5. Analysis of the Mapping of Keyword Co-Occurrence Networks
Keywords represent the focus of documents and their structure to a certain extent [16]. They illustrate the fundamental objective of the document and can be used to explain the core motivation of a study in detail [74]. The 10 most relevant terms occurring as the authors’ keywords were: “sustainability” (305 occurrences), “agriculture” (109 occurrences), “Brazil” (107 occurrences), “sustainable agriculture” (96 occurrences), “agribusiness” (76 occurrences), “soil” (63 occurrences), “soy” (57 occurrences), “livestock” (54 occurrences), “sustainable intensification” (54 occurrences), and “management” (52 occurrences; Figure 7).
The keywords “sustainability”, “sustainable”, “agriculture”, “agribusiness”, “livestock”, and “sustainable agriculture” partially mirrored the queries made in the database. This result suggests that these terms may be the most discussed areas in Brazil’s research on sustainability and agriculture [22]. Other terms such as “soil”, “soybean”, and “sustainable intensification”, among the others listed, may have direct or indirect applications for the sustainability of agricultural systems [22]. In turn, the word cloud more visibly presents the density and frequency with which the keywords occurred. It is possible to highlight words that, although with lower frequency and density (e.g., “food security”, “bioenergy”, “climate change”, and “ecosystem services”), emerge as potential applications in research related to the sustainability of agricultural systems (Figure 8). Furthermore, for Mulay et al. [75], words in smaller letters indicate potential study directions.
When the interrelations between the keywords are observed from the co-occurrence networks, it is noted: (i) the centrality of the term “sustainability” as well as (ii) its relationship with “agribusiness”, “Brazil”, and “agriculture”, which allowed us to infer a possible direction for the areas of agricultural studies towards the development of research aimed at more sustainable agricultural practices (Figure 9). As an example of practices, one can infer the reduction in deforestation in the Amazon and the Cerrado regions. We also note the emergence of less frequent relationships, which are of great relevance. As an example, we can mention “food security”, “climate change”, and “agroecology”, suggesting potential study directions for future themes.
Regarding the tree map results, the keywords Plus “management/management” (12%), followed by “yield” (8%) and “growth” (7%) stand out. Thus, it is possible to infer that zootechnical or phytotechnical performance has directed the main focuses of the research used as a reference to guide the writing of the sampled documents (Figure 10).
3.6. Suggestion for Identifying and Forming Multi-Collaborative Research Groups
Figure 11 presents a more precise overview of how the methodology used can suggest the formation of research groups correlating institutions and authors based on similar lines of research, understood based on the keywords. To this end, the three-field graph integrates information regarding institutions (AU_UN), authors (AU), and keywords (DE, ID). It is observed that the gray flows between the first two columns demonstrate that many studies are the result of collaboration between different institutions through the support of authors (Figure 11a). Likewise, the flows between the central and right columns suggest, based on the keywords, that many authors share the same area of knowledge in different institutions, which can provide information feedback [76,77] and new insights. Although they share similar themes, such as soybean production, the studies may differ in purpose, focusing on performance and yield or deforestation. The variation depends on the spatial frontier adopted or the specific link in the production chain assessed. The overarching idea is to extend the boundaries and connections within the production chain, facilitating the integration of results obtained at each production stage. This can be achieved through projects that complement one another and establish links across the entire chain. The figure also illustrates collaborations among multiple authors, indicating a common research direction with similar objectives. Based on the figure, the suggestion is to broaden the thematic scope of studies by incorporating diverse perspectives and exploring additional research themes. Furthermore, it is possible to suggest potential thematic areas in which authors could collaborate to form public policies and propose the formation of multi-collaborative research groups between institutions, promoting research based on themes of common interest (Figure 11b). As an example, we can suggest the collaboration between de Faccio Carvalho, P. C.; Cherubin, M. R.; Cerri, C. E. P.; Cerri, C. C.; Crusciol, C. A. C.; Martins, A. P.; Hungary, M.; Franzluebbers, A. J.; and Anghinoni, I.; for the formation of public policies aimed at issues related to carbon sequestration or/and carbon emissions in agriculture and livestock.
3.8. Future Vision of Brazilian Publications on Sustainability in Agriculture and Livestock
Figure 13 presents the vision for future publications using mapping analysis from the author’s keywords (a) and the author’s keywords on a thematic map (b). The clusters of the co-occurrence network are demonstrated as circles plotted graphically according to Callon’s centrality and density classification [82]. The larger the size of the circle reflects, the greater the occurrence of the term. The x-axis shows the centrality of the network cluster, or the degree of interaction with other groups in the graph, and measures the importance of a study topic. The y-axis represents the density that measures the internal strength of a network of clusters and the growth of the theme [83,84,85].
According to the results, most terms found occupied the upper right quadrant (Figure 13a). Terms occupying the upper right quadrant are suggested as prominent themes characterized by high density and centrality; therefore, these themes need to be developed and are important to study in future research [86]. In this quadrant, the suggestion of studies related to “sustainable intensification”, “crop–livestock integration”, and “agroforestry” stood out. More specific themes are concentrated in the upper left quadrant; however, there is high development, as indicated by themes with high density and low centrality [86]. Topics related to soil (i.e., “soil quality” and “soil compaction”) were highlighted in this quadrant.
Figure 13b demonstrates the formation of the quadrants, presenting them more intuitively. In this sense: (i) motor themes (top right quadrant), which shows the network of clusters with high density and centrality, position themes that are well-developed and crucial to structuring an area of knowledge; (ii) niche themes (top left quadrant), which presents themes with high density and low centrality, positioning themes of limited relevance; (iii) emerging or declining themes (lower left quadrant), which present themes with low density and centrality, implying minimally developed and marginal themes; and (iv) basic themes (lower right quadrant), with high centrality and low density, represent vital themes for the trans-disciplinarity of research [87]. Corroborating the results in Figure 13a, themes related to soil (i.e., “soil quality” and “no-tillage”) present themselves as well-developed, as well as “soy”. Themes such as “nitrogen” and “management”, as well as “sustainability” and “agriculture”, are themes presented on the border of the quadrant, referring to motor themes and niche and basic themes, respectively. This fact implies that such themes are on their way to becoming driving themes in the future with the increase in studies in the area of knowledge. Also, it can be inferred that “sustainability”, “agriculture”, and “Brazil” are themes of trans-disciplinary interest. Other themes such as “corn” (Zea Mays), “soil management”, “cover crops”, and “crop–livestock integration” were concentrated in the lower left quadrant (emerging or declining themes), which may indicate themes that are emerging from lines of research.
[ad_2]
